SECTION 1 INTRODUCTION
The profit split method (PSM) is one of the five transfer pricing methods delineated in Chapter II of the OECD Transfer Pricing Guidelines. These methods can be used to establish whether the conditions imposed on the commercial or financial relations between associated enterprises are consistent with the arm’s length principle.
The OECD guidelines of 1995 referred to the PSM as a method of “last resort”, to be used when other methods could not be reliably applied (para. 3.50). Yet, since the revision of the OECD Guidelines in 2010, the PSM is considered a pricing method to be applied in an equally reliable manner as the other methods in accordance with the “most appropriate method” criterion.
Due to the increased integration of multinational enterprises and the globalization of national economies and markets, the clarification of the PSM was one of the priorities identified in the action plan against Base Erosion and Profit Shifting (BEPS). Indeed, in order to develop rules that can prevent BEPS resulting from engaging in transactions which would not, or would only very rarely, occur between third parties, Action 10 called for clarification of the application of transfer pricing methods, in particular of the transactional profit split method, in the context of global value chains.
In June 2018 the OECD published a Report containing Revised Guidelines on the application of PSM that clarifies and significantly expands the Guidelines on when a profit split may be the most appropriate method. The Guidelines also note that the basic premise that the transactional profit split method is applicable where it is found to be the most appropriate method is unchanged.
In addition, the programme of work 2015-2019 of the Joint Transfer Pricing Forum (JTPF) referred to the PSM as one of the topics on which the JTPF should provide output in order to address the problems in the practical application of the PSM. A particular reference is made to the high degree of subjectivity encountered when stakeholders determine how to share the profit.
1.2. Aim of the work
The aim of the JTPF exercise was to take stock of how the PSM is applied within the EU and work towards a common approach to addressing the relevant challenges arising under the current OECD framework.
As pointed out in the OECD Guidelines, the main advantage of the PSM is that it can offer solutions in cases where all relevant parties make unique and valuable contributions and/or there is a high degree of integration. In such cases, it is frequent that the reliable information on comparables is insufficient for applying another transfer pricing method although as pointed out by the OECD, lack of external comparable per se should not lead to default use of PSM (para 2.128 and 2.143 of the OECD Guidelines). Secondly, when parties share the assumption of economically significant risks or assume closely related risks, its flexibility allows the determination of an arm’s length profit for the parties according to the actual assumption of the risks.
The Guidelines also point to the difficulties in the application of the PSM, including problems of measuring the relevant revenue and costs between all associated enterprises participating in the controlled transactions and the challenges of identifying appropriate profit splitting factors.
For these reasons greater clarification and standardisation, where possible, of the PSM could have significant benefits for both tax authorities and taxpayers as it would reduce compliance costs, simplify audits, and provide more predictability and certainty.
At the same time, it should be recognised that there is a tension between these desirable objectives and the need, as stressed in the Guidelines, for a method that is appropriate for the individual taxpayer. Indeed there are two different views: (i) that it should be possible to simplify the process by standardizing the application of the PSM for common business models as the aim is not complete precision but achieving a reasonable estimate in line with the arm’s length principle and (ii) that it is essential to emphasise the importance of the relevant facts and circumstances in each and every situation when applying the method, i.e. simplification is inappropriate as each case should be assessed individually regardless of the resultant complexities.
Indeed, under the traditional OECD framework, the accurate delineation of the transaction from the perspective of all parties involved should remain the starting point for any transfer pricing analysis.
The analysis of the economically relevant characteristics of the transaction and in particular, the functional analysis, supported by the information in the MNE group’s transfer pricing documentation, should reveal: (i) how value is generated by the group as a whole; (ii) the interdependencies between the functions performed by the associated enterprises; and (iii) the contribution that each of the associated enterprises makes to that value creation. In particular, the analysis of risks and the determination of which group entities take the key decisions related to control over risk as well as which of these entities have the financial capacity to assume the risk should help identify the most appropriate way of splitting the relevant profit from the transaction under review.
1.3. Delineation of the scope of this paper
In order to take stock and gather useful information on the application of the PSM, it was decided to carry out a survey on the experiences of the JTPF members in the practical application of the PSM (JTPF meeting of 8 March 2018).
The results of the Survey were presented at the JTPF Meeting of 26 June 2018. The Secretariat received 17 replies from Member States and 11 replies from Non-Government Members (NGMs) which provide a first insight of the current situation in the EU.
The survey did not show a direct correlation between the PSM and one specific industry. The PSM turned out to be applied in several different sectors such as the financial sector, industrial equipment, the automotive industry, the IT sector, trade in consumer goods, the pharmaceutical
industry, chemical industry, the food industry. However, even in these industries, the PSM has only been used to a limited extent so far.
The survey highlighted the choice of the appropriate splitting factors, their relative weights and the valuation of the contributions, especially heterogeneous contributions, as the main challenge in the application of the PSM (N.B. these featured in practically all replies as one of the main challenges in applying the PSM).
The lack of a common methodology on determining the profit splitting factors has been indicated as a cause that exacerbates the challenges and could undermine the prospects for applying the PSM even in circumstances where this may be the most appropriate method.
At the meeting of 24 October 2018, the JTPF considered 4 potential areas for further work on the PSM: (1) When to use the PSM? (2) How to split the profit? (3) Is there a need for simplification? (4) How to simplify the application of the PSM?
During the meeting, the members reached consensus on a 2-stage process whereby at the first stage, the focus should be on clarifying certain concepts related to the PSM and eventually at the second later stage on exploring ways for simplification .
This paper addresses the first stage and aims at clarifying certain concepts in applying the PSM: (i) when to use the PSM (i.e. in which circumstances it may be considered the most appropriate transfer pricing method) and (ii) how to split the profit based on the concepts described in the revised OECD Guidelines as well as by providing an inventory of recurrent splitting factors.
For the avoidance of doubt this report should be regarded as complementary to, and supportive of, the text of the OECD Revised Guidelines on the application of the Transactional Profit Split Method issued in June 2018.
The paper is structured as follows: section 2 provides a short description of the profit split method; section 3 seeks to clarify some key concepts related to the use of the profit split method also touching upon some challenging points; and section 4 describes a number of potential splitting factors listed in the Annex.
SECTION 2. DESCRIPTION OF THE PROFIT SPLIT METHOD
The PSM seeks to establish or test, in line with the arm’s length principle, an approximation of the results that independent enterprises would be expected to have achieved by engaging in transactions in comparable circumstances. In general, there are two mainly used approaches to splitting the profits: (i) the contribution analysis and (ii) the residual analysis.
Under the contribution analysis, the relevant profits from controlled transactions are allocated between the associated enterprises on the basis of the relative value of the functions performed, assets used and risk assumed by each of the associated enterprises engaged in these controlled transactions.
A residual analysis divides the relevant profits from controlled transactions into two types. The first type (the initial remuneration) consists of profits attributable to contributions for which a benchmark exists (typically less complex contributions for which comparables can be found). This is done by applying one of the traditional transactional methods or the transactional net margin method (TNMM). The second type (the residual) consists of profits (or losses) that relate to unique and valuable contributions, the shared assumption of economically significant risks (or the separate assumption of closely related risks) and/or a high level of business integration and remain after the first type. The method allocates this residual profit (or loss) among the participant parties in the controlled transaction based on the relative value of their contributions.
For the avoidance of doubt, the reference to “profits” to be split in this report shall mean profits and/or losses to be split.
SECTION 3. USE OF THE PROFIT SPLIT METHOD
The PSM, like any other transfer pricing method, should be chosen as the most appropriate method only after the accurate delineation of the transaction including the functional analysis. In addition, the PSM must be appropriate for the particular circumstances that it is aimed to be applied to.
The OECD Guidelines on the use of the PSM list the following indicators for determining whether the PSM may be considered the most appropriate transfer pricing method in a specific set of circumstances:
- the existence of a unique and valuable contribution by each party to the controlled transaction and/or
- a high level of integration regarding business transactions to which the transaction relates and/or
- The shared assumptions of economically significant risks or separate assumption of economically closely related risks by the parties to the transaction.
Besides the constraints already mentioned in the previous paragraph it is also important to indicate when it may not be appropriate to use the PSM:
- Where one of the parties to the transaction performs only simple functions and does not make unique and valuable contributions; and/or
- The accurately delineated transaction can be appropriately benchmarked (even when the accurately delineated transaction is quite complex), comparable transactions and functions can be identified;
The lack of comparables alone is insufficient for determining that the PSM should be selected as the most appropriate method; in such cases a pragmatic approach needs to be taken such as broadening the search criteria, without compromising the quality of the comparables, by looking at independent enterprises:
- with slightly different business strategies, business models or slightly different economic circumstances;
- situated in other geographical markets, but being active in the same industry or
- engaged in different industries, situated in the same geographical market.
The following section will address some of the challenging aspects of the indicators for determining whether a PSM can be used by way of a Q&A and provide examples in an attempt to clarify the circumstances where to use the PSM.
Some of the examples have been taken from the OECD Guidelines and/or doctrine articles. Due to time constraints, these examples have not been developed further.
3.1. Unique and valuable contributions
The OECD Guidelines mention that contributions are unique and valuable when they are not comparable to contributions made by uncontrolled parties in comparable circumstances and they represent a key source of actual or potential economic benefits.
3.1.1. Example of when the PSM is likely to be considered the most appropriate method due to unique and valuable contributions
Example 1 in annex II to chapter II of the OECD Guidelines present a case of a unique and valuable contribution that leads to the likely application of the PSM. It reads as follows:
Company A is the parent company of an MNE group in the pharmaceutical sector. Company A owns a patent for a new pharmaceutical formulation. Company A designed the clinical trials and performed the research and development functions during the early stages of the development of the product, leading to granting a patent.
Company A enters into a contract with Company S, a subsidiary of Company A, in accordance with which Company A licenses the patent rights relating to the potential pharmaceutical product to Company S. Based on the contract, Company S conducts the subsequent development of the product and performs important enhancement functions. Company S obtains the authorisation from the relevant regulatory body. The development of the product is successful and it is sold on various markets around the world.
The accurate delineation of the transaction indicates that the contributions made by both Company A and Company S are unique and valuable to the development of the pharmaceutical product.
Under these circumstances, the transactional profit split method is likely to be the most appropriate method for determining the compensation for the patent rights licensed by Company A to Company S.
3.1.2. Example of when the PSM is not likely to be considered the most appropriate method even if there are some valuable contributions
A small family owned group has two manufacturing companies in countries A and B. The founder and head of the family is resident in country C where he manages the holding company.
The holding company in country C has no financial assets apart from the participations of the manufacturing companies and has no employee apart from the head of family who acts as CEO of the entire group.
Each manufacturing company has developed its own patents and its own know -how for their own processes and sell directly to third parties. The manufacturing companies are in charge of their own raw material sources and production planning; they are characterized as fully-fledged manufacturing companies or entrepreneurs. There are no transactions between the two manufacturing companies as they run two different lines of business. However, there is a transfer pricing issue concerning charges from the parent company to the subsidiaries.
The marketing staff and the general managers of the manufacturing companies get together twice a year in Country C to discuss trends in market conditions and general strategies. Every time an entrepreneur wants to bring a new product to the market, the marketing strategy has to be approved by the holding, in other words, by the founder of the group.
Despite the fact that the decisions of the founder of the group are relevant to both family businesses (carried on by the two manufacturing companies), the simple fact of giving the final approval should not be confused with the concept of a unique and valuable contribution which would justify the application of the PSM. Instead, the contribution by the founder to the two manufacturing subsidiaries might qualify as a management service. Here one can take into consideration that the function performed by the founder may be one that could reasonably also be carried out by a CEO appointed based on a selection process.
3.1.3. How should unique and valuable intangibles be interpreted?
Intangibles within the concept of “intangibles that represent unique and valuable contributions” should be understood to encompass the definition of Chapter VI, paragraph 6.6. The definition of intangibles for transfer pricing purposes is broad and may go beyond legal or accounting definitions in national laws. It is intended to cover “something which is not a physical asset or a financial asset, which is capable of being owned or controlled for use in commercial activities, and whose use or transfer would be compensated had it occurred in a transaction between independent parties in comparable circumstances4.”
Whether a certain intangible is unique and valuable will depend on the facts and circumstances of the specific transaction and intangible.
3.1.4. Is the concept of “unique” separate from the concept of “valuable”?
The OECD Guidelines deal with terms “unique and valuable contribution” together. In this respect, par. 2.130 of the OECD Guidelines state that “The Contributions (for instance functions performed, or assets used or contributed) will be “unique and valuable” in cases where (i) they are not comparable to contributions made by uncontrolled parties in comparable circumstances, and (ii) they represent a key source of actual or potential economic benefits in the business operations. This means that when no comparables can be found, but the contribution does not represent a key source of economic benefits, it will usually be considered a unique but not valuable contribution. Hence, the PSM will most likely not be the most appropriate method and a solution in line with the OECD Guidelines (3.38) on limitation of available comparables should be sought. For the PSM to apply, it would be required that the contribution be considered both unique and valuable.
3.1.5 If a potential economic benefit were not realised, should the contribution still be considered “valuable”?
A unique and valuable contribution should represent a key source of actual or potential economic benefit in the business operations. Within this concept the important factor is not whether a (potential) economic benefit is realized or not, but if the success of the contribution is key for the success of the business operations. Therefore, a contribution could still be considered valuable to the transaction if a potential economic benefit is not realized.
3.1.6 Can a unique and valuable contribution be present and the PSM applied accordingly even if the parties of the transactions do not share the assumption of closely related risks?
The unique and valuable contribution of a party to a transaction will often be linked to the assumption of risk related to that contribution (control over risk) by the party, but will not necessarily lead to a joint assumption of the same risk.
The shared assumption of risk (or separate assumption of closely related risks) is a separate factor that may indicate that the PSM is the most appropriate method.
3.2. Highly integrated business operations
According to the OECD guidelines, the high degree of integration in relation to controlled transactions means that the functions performed, assets deployed and risks borne by parties to the transaction are so highly interlinked (for example, they may be involved in the same stage of the value chain) that they cannot be reliably evaluated on a separate basis using a one-sided method.
3.2.1. Example of when the PSM is likely to be considered the most appropriate method in relation to highly integrated transactions
One MNE group has two different main divisions, one for mobile construction cranes and one for tower cranes. Company A (resident in country A) is the original equipment manufacturer (OEM) of mobile construction cranes. Company B (resident in country B) is the OEM of tower cranes.
The two OEMs have decided to jointly enter into a common project and produce hybrid cranes, meaning that some parts and input for specific product functionalities of the end product stem from Company A and others from Company B. The profits realized with the sale of these hybrid construction cranes are directly linked to the innovative technological intellectual property (IP) of both OEMs.
Both OEMs purchase their relevant raw material and components for producing the hybrid cranes from unrelated third-party suppliers.
Company A first assembles all necessary components for the hybrid cranes and transports the semi- finished goods exclusively to Company B, which installs its core components. The still semi-finished cranes in turn need to be transported from Company B exclusively to associated Company A, which finalizes the hybrid crane with certain key parts so that the end-product can be sold to either associated distributors, which distribute the hybrid cranes to unrelated customers.
The functional analysis relating to the production of hybrid cranes and to their controlled transactions (i.e. transactions between the two associated OEMs) identifies that both manufacturers (as the parties to the controlled transactions) perform their respective R&D functions as well as most of the functions related to logistics and production.
The OEMs take their own important strategic decisions with regard to R&D as well as their use of intangibles, contribute their assets (both OEMs invest heavily in R&D) and assume the risks (including investment risk) for these activities. Both OEMs carry out economically significant key activities, contribute significant value to the transactions, take responsibility for the activities undertaken and assume the risks thereof. In terms of development, enhancement, maintenance, protection and exploitation (the so-called DEMPE functions) relating to the technology of hybrid cranes, the OEMs respectively develop, enhance and maintain intangibles. They also own unique assets and valuable intangibles, perform significant value driving functions and assume major risks. OEMs protect their IP (mostly technology-related IP) via the registering of respective patents.
In addition, an executive committee has been set up, in which leaders of both OEMs are represented and take decisions for the MNE group with regard to the hybrid crane construction market as a whole.
Taking into consideration all the facts and circumstances, both OEMs can be characterized as fully fledged manufacturers in relation to the production of the hybrid cranes. As both OEMs perform functions, use assets and assume risks related to the DEMPE functions, both enterprises should be remunerated on an arm’s length basis for their contributions.
Each OEM develops and owns unique and valuable intangibles in their respective production processes, but, as the accurate delineation of the transactions clearly highlights, their operations are highly integrated in the sense that the outcome of each OEM depends on the capacity of the other.
Additionally, one should take into account that there is a joint executive committee in which leaders of both OEMs are represented and take decisions for the MNE group hybrid crane construction market as a whole.
On the basis of the facts and circumstances, the interactions between the two OEMs can be characterized as highly integrated.
Therefore, the PSM is likely to be the most appropriate method for determining the profits for each OEM from the sale of hybrid construction cranes.
3.2.2. Example of when the PSM is not likely to be considered the most appropriate method in relation to transactions with a certain degree of integration.
One MNE group is active in the commercial vehicle industry with legal entities in all parts of the world.
The commercial vehicle industry is very capital intensive and its start-up costs are significant. It is therefore important to manage the value chain in an efficient manner in order to find synergies, control costs and drive profitability. The growth of the Group has been characterized by major acquisitions over the years and efforts to integrate the acquired business and create economies of scale and synergies by common development, platforms and IT systems. The Group HQ is located in country A and is the ultimate parent company and decision maker of the entire Group. Together with the parent companies for the respective business areas, also located in country A, the Group HQ has the overall responsibility for managing and driving the operations globally, including, but not limited to, the development, manufacturing and distribution of the Group’s products and services.
The Group HQ is the owner of the intangibles and use “contract research centres” located in 10 different countries, within and outside EU to implement R&D activities.
The Group HQ performs most of the economically significant functions involved in the research of the product development cycle, including prioritizing projects, budget aspects, measuring success, defining commercial parameters, assessing opportunities.
The Group HQ provides funds/capital and other economically significant assets, including intangibles for research or product development. The Group HQ bears the risks of the failure of the research.
The Group adopted a business model that has a high correlation with the Group’s centralized operational governance model, with guidelines, strategies and instructions ultimately decided by the management and board of Group HQ.
The contract research entities execute the R&D activities under supervision of the Group HQ which not only has capability to control or supervise but actually controls and supervises research or product development through its strategic decisions on how core functions are performed as well as monitors activities on a regular basis.
The contract research entities have no ownership rights on outcome of research (which vests with the Group HQ) and do not assume significant risks.
In that case, even though there is a degree of integration and interdependence (through the dependence of the contract research entities of the Group HQ decision-making), the assets, functions and risks of the Group HQ are separate and can reliably be evaluated in isolation from the functions, assets and risks of the contract research entities.
Although the centralisation does involve an element of integration of the activities, the contract research entities would generally be accurately delineated as service providers and be remunerated in accordance with the arm’s length principle.
3.2.3. How can circumstances of “highly integrated” be distinguished from the cases of “integrated” business operations?
The OECD Guidelines in the new para 2.133 recognise that most MNE groups are integrated to a certain extent. However, for the integration to be of a high degree, the assets, functions and risks of one party should be interlinked with, and it should not be possible to reliably evaluate them in isolation from, the functions, assets and risks of the other party. In that context, a need for clarification was also expressed with regard to the terms “interlinked” and “reliably”. Here again, there is a reference to the OECD Guidelines which consider that in such cases, benchmarking analysis is not feasible, since the interlinkages do not allow to reliably separate and evaluate the assets, functions and risks of each party.
In practical terms, the operations would be “highly integrated” when two or more legal entities share and/ or combine functions, assets and risks, while jointly achieve a common objective (e.g. a project whereby the parties jointly contribute to each part of the process towards a common outcome). “Highly integrated” does not necessarily mean using the same resources or service provider. Rather, there needs to exist a common objective and a combination of joint contributions. Additionally, it should be sustained that it would not be possible to materialise such a jointly achieved objective without the participation of the other party in the transaction.
3.2.4. Should the criterion of highly integrated business operations stand on its own or should it be considered in the context of the criterion on unique and valuable contributions?
The three main criteria for the application of the PSM described in sections C.2.2.1, C.2.2.2 and C.2.23 of the OECD Guidelines, while not necessarily being mutually exclusive (as indicated in para 2.126 of the OECD Guidelines), stand on their own.
The OECD Guidelines contain examples 6, 7 and 8 in Annex II to Chapter II, two of which feature a high level of integration but no unique and valuable contributions, and one where there is certain degree of integration, but the contribution of one party is not unique and valuable and can be benchmarked. The first two examples show that where there is a high degree of integration, the PSM is applicable even without unique and valuable contributions. The third example indicates that where there is only a limited degree of integration and only one party makes unique and valuable contributions, the PSM is unlikely to be appropriate. Taking these examples together, it is clear that where the criterion of highly integrated business operations is satisfied the PSM is likely to be applicable regardless of whether both parties also make unique and valuable contributions.
3.2.5. Should the presence of a cross-entity and cross-border business unit structure be an indication of highly integrated business operations?
Cross-border units are virtual organisations that run in parallel to the legal entity structure within MNE’s. They constitute a matrix organizational structure where an employee may have two reporting lines, one “hierarchical” (commonly, an employee within the same legal entity) and one “functional” (often an employee of the same business function/business unit who belongs to a different legal entity that can be across the border). In some cases, one business unit might comprise more than one legal entity and very often, there are no transactions among the legal entities of the same business unit.
The presence of a cross-entity and cross-border business unit structure should not per se be an indication of a highly integrated business operation for the purpose of applying the PSM.
The function of such a cross-entity and cross-border business unit structure would vary from one business to another. Therefore, an assessment would need to be performed on a case-by-case basis, taking into account the Guidelines on applying the arm’s length principle in Chapters I-III of the OECD Guidelines and in particular, the accurate delineation of the transaction including a functional analysis (function performed, risk assumed (control over risk and financial capacity to assume those risks, and assets deployed).
In certain cases, such a structure can have a result that decisions taken with regard to the control of risk are concentrated into that cross-entity business unit and no particular group entity is solely associated with that business unit. Therefore, from the perspective of the MNE legal structure, decisions relating to the control of risk would be spread out across different group entities. Decision- making and the related assets, functions and risks can be intertwined to such a high degree that it is not possible to separate them reliably and link separate assets, functions and risks to each group entity. Such a business structure can then give rise to highly integrated business operations, as well as to lead to a shared assumption of economically significant risks since the control of entrepreneurial risk is not concentrated into a single group entity.
In other cases, the decisions taken by the cross-entity business unit would be mostly related to activities within the group which aim to ensure that goals at the MNE level, by way of instructions, decisions, policies and other governance activities, are followed and interpreted by the group entities. In that case, even though the presence of a cross-entity business unit would indicate a degree of integration, the assets, functions and risks of those group entities could be more reliably separated and analysed. It follows that another method than the PSM could be found to be the most appropriate in such a setting.
3.3. Shared assumption of economically significant risks, separate assumption of closely related risks
The revised OECD Guidance relies heavily on the notion of the control over risk.
Where, according to an accurately delineated transaction, each party to the controlled transaction shares the assumption of one or more of the economically significant risks in relation to that transaction or the various economically significant risks are separately assumed by the parties but are closely interrelated, the PSM may be found to be the most appropriate method.
3.3.1. Should the hallmarks of “shared assumption of economically significant risks” and “separate assumption of closely related risks” be considered in isolation, or are these derivative criteria to the one on “highly integrated business operations”?
The OECD Guidelines point out in para 2.126 that the hallmarks of shared assumption of economically significant risks and separate assumption of closely related risks are not mutually exclusive with the one on highly integrated business operations, each concept should be considered and analysed on its own.
Example: two legal entities which are part of an MNE enter into an agreement to develop new technology. This new technology requires an important amount of investment that will be recovered if the development is successful. The two legal entities are not integrated and perform – in a separate manner – all the functions related to the project. Both legal entities share the risk of the investment and jointly achieve a common objective.
3.3.2. What is meant by risks being “closely related”?
In para 2.140, the OECD Guidelines clarify that risks are considered “closely related” where the playing out of the risks of each party cannot reliably be isolated. The Guidelines also refer to Example 10 in Annex II to Chapter II.
Annex 1 to this report contains a flow chart tree that simplifies the PSM selection process.
SECTION 4 HOW TO SPLIT THE PROFIT
The PSM seeks to split the relevant profit from the controlled transactions on an economically valid basis, in order to approximate the results that would have been achieved between independent enterprises in comparable circumstances.
The division of the relevant profits is generally achieved by using one or more profit splitting factors.
The OECD Guidelines state in paragraph 2.169 that the determination of the appropriate profit splitting factor(s) should reflect the key contributions to value in relation to the transaction. The key contributions to value in relation to the transaction should follow the part of the functional analysis dedicated to understanding how value is created by the MNE as a whole.
In para 2.166 of the OECD Guidelines, it is stated that profits should be split on an economically valid basis that reflects the relative contribution of the parties to the transaction and thus approximates the division of profits that would have obtained at arm’s length. As further specified, the splitting factors should be based on objective data and be verifiable and supported by comparables data, internal data, or both.
In particular the OECD Guidelines make it clear that no significant value should be attributed to mere legal ownership of rights to intangibles. What is important is control over risk and the key functions of development, enhancement, maintenance, protection, or exploitation of intangibles (see e.g. OECD Guidelines para. 6.54).This is because legal ownership can easily be attributed to an entity located anywhere, and which may lack substance. Similarly, in relation to the assumption of risks, the OECD Guidelines stress the need to control risk and have the financial capacity to assume the risk.
Annex 2 to this report contains a flow chart tree that shows, in a simplified way, a number of steps that may be considered when assessing when and how to apply the profit split method.
4.1. Inventory of profit splitting factors
At the JTPF meeting on 24 October 2018, it was decided that a non-prescriptive list of splitting factors with a brief analysis of the pros and cons of each splitting factor, circumstances and value drivers would be of benefit to both taxpayers and tax administrations in Member States.
With this aim, the splitting factors identified in the survey are listed in Annex 3 to this paper.
The inventory is neither meant to be exhaustive nor meant to create a hierarchy, but rather to focus on the most often encountered splitting factors. For the avoidance of doubt, under the current OECD Guidelines, the most appropriate method should always be sought and there is no presumption that these will be factors, or only the factors, in point. Taxpayers and tax administrations should not be expected to have to do additional work to explain why these factors should be disapplied in their particular case.
The splitting factors are grouped in the following broad categories:
A. People-based factors,
B. Sales/volume based factors,
C. Asset-based factors,
D. Cost-based factors,
E. Other factors
Many of these splitting factors are derived from financial or management accounting. In some cases, challenges are mentioned relating to the choice of whether to accumulate the indicators over time or take current year indicators, potential subjectivity in the selection of key employees, a potential issue over whether to adjust the indicators to purchasing power and the added complexity of doing so, an accounting valuation not always reflecting the fair value of the (intangible) assets.
A. When people functions are a key value driver to a specific business, either the factor of employee compensation and/or headcount may be used. There is, in any case, a judgment to be made with regard to which are the employees whose remuneration or headcount would be included in the factor. There is a view that the amount of employee compensation already reflects the relevant importance of each employee’s contribution. Nevertheless, this approach could be problematic concerning the inclusion of employees whose activities are not related to the unique and valuable contributions that determined the PSM as the most appropriate method in the first place, and on balance the key people performing important functions should be the ones included.
B. Sales and volume based splitting factors are most often used in combination with other splitting factors which reflect efforts in sales/distribution/marketing but also efforts in R&D/quality/etc. depending on the industry and group strategy.
C. The asset-based splitting factors may be used when the contributions are in the form of (intangible) assets. In theory, the asset value that is used as a splitting factor would have to be measured in accordance with the arm’s length principle rather than accounting value where there is likely to be a material difference between the two. In such a case, consideration of valuation methods could prove useful, although there is a danger of added complexity and subjectivity here. Another approach is to approximate the intangible asset contributions based on the royalty rates or franchise fees attributable to comparable intangibles.
There is also a view, although several members consider that this view is not in line with the OECD guidelines, that when it comes to asset-based splitting factors, solely tangible property should be taken into account since the mere legal ownership of intangible assets is now not considered to create value. Not only are intangibles problematic from a valuation perspective (due to the inherent complexity and subjectivity of valuation techniques), they may also be problematic from a locational perspective. Under this view, the “relative value” of intangibles can be better measured by using only the objective factors that reflect the various DEMPE functions relating to the intangibles. These are the people-based, tangible asset-based, and cost- based factors.
D. The cost-based splitting factors are often used in the joint performance of the value creating activities. The contribution in the form of a value creating activity is then reflected on the costs borne in the performance of that activity.
E. There is also a residual category of other factors that are often used as well. Those include contribution weightings assigned on the basis of functional analysis, external benchmarks and hedge fund financing.
SECTION 5 CONCLUDING REMARKS
The survey indicates that the PSM is not used very often at present and that when it is used, this mostly happens in the context of APA procedures. This may relate to the perceived high degree of subjectivity in the mechanism for profit allocation. Hence this method may not necessarily ensure tax certainty and may be exposed to risks of litigation.
That said, there is the possibility that the PSM may be applied more often in the future due to the emergence of new business models. Indeed, the European Commission’s proposal for a Council Directive laying down rules relating to the corporate taxation of a significant digital presence of 21 March 2018 explicitly refers to the PSM as a method for allocating profits in the digital economy.
Considering this, it would be useful to collect further data and monitor the practical application of the PSM in order to: (i) evaluate whether the mechanism could be further clarified and (ii) move onto the second stage mentioned in the Aim of the Work section (page 4) to explore ways of simplification.
|Indicative List of PSM Splitting Factors identified by JTPF Members|
Indicative Splitting Factors
|Description and Circumstances for Applying the Splitting Factor|| |
People based splitting factors
Remuneration of people who are key value drivers (e.g. executive and strategic management, employees related to DEMPE functions, traders)
|This splitting factor may be taken in consideration when the value creation is driven by the workforce and personnel knowledge and skills. In order to apply the factor it is important: A: to map the employees; B: to describe the functions carried out by the employees and identify those related to DEMPE functions; C: calculate the remuneration of the key employees.||Linkage to functions performed and risk assumption. Current year compensation expenses of employees is relatively easy to identify. – Implementation may be relatively simple with a homogeneous pool of key employees (e.g. traders in global trading business).||Difficulty to decide whether to use current or cumulative values and whether to use the remuneration or the headcount criteria.|
Selection of key employees is subjective and may be hard to verify. The total amount of the remuneration could be affected by efficiency issues cost of living and other market difference across jurisdictions. – Can be very sensitive to the movement of a small number of executives between entities.
Bonus and stock options are difficult to be considered.
Finally, considering the different cost of life among the various EU countries, adjustments should be considered on the basis of public data.
Headcount of people who are key value drivers (e.g. executive and strategic management, employees related to DEMPE functions, traders)
|Similar to the splitting factors based on the remuneration of the key empoyees, this splitting factor may be taken into consideration when the value creation is driven by the workforce and personnel knowledge and skills. In order to apply the factor it is important: A: to map the employees; B: to describe the functions carried out by the key employes and identify those related to DEMPE functions; C: to calculate the number of the key employees and (eventually) weigh it based on the importance of the functions performed or seniority. In general, the splitting factor based on remuneration is preferable compared to the headcount, since it takes into account how much the MNE values the contribution of various employees. Headcount may be preferable where differences in the cost of living and other factors make it difficult to use remuneration.||Link to functions performed and risk assumption. Current year headcounts are easy to identify.||Difficulty in deciding whether to use current or cumulative values; whether to use the remuneration or headcount criteria.|
The selection of key employees is subjective and may be hard to verify. The total number of the key employees could be affected by efficiency issues, the cost of living and other market difference across jurisdictions. It can be very sensitive to the movement of a small number of executives between entities.
Using only headcount numbers potentially ignores value contribution derived from relative employee experience or expertise.
Cost based splitting factors (other or broader than people costs)
Operating expenses – OPEX
|OPEX are expenses that a business incurs in carrying out an organization’s day-to-day activities but these are not directly associated with production.|
OPEX include rent, equipment, inventory costs, marketing, payroll, insurance and funds allocated for research and development. Although OPEX may be related to routine functions at times, they may be used as a splitting factor when the value creation is driven by the intensity of the activities performed by the entities.
|Costs are in general a verifiable indicator. Current year costs are relatively easy to obtain.||OPEX could be difficult to identify as they are usually spread through different budget lines. Organizational inefficiencies are not taken into consideration. Some costs (e.g. marketing and advertising) may have effects across multiple periods, whereas others only have current period effects. – Difficulty deciding whether to use current or cumulative values. – Need to consider whether only cash costs or also non-cash costs (e.g. depreciation and amortization) are included in the calculation. OPEX includes sometimes also passthrough costs (e.g., advertising paid by local LRD but totally|
managed by the principal).
Cost of Goods Sold – COGS
|COGS is the accumulated total of all|
costs used to create a product or service, which has been sold. COGS generally includes the cost of the materials used in creating the good, along with the direct labor costs used to produce the good or service. Although COGS may be related to routine functions at times, it may be used as splitting factor when the value creation is driven by the production activity.
|Costs are an verifiable indicator.|
Current year costs are relatively easy to obtain. Suitable in case of transactions between manufacturers.
|Organizational inefficiences are not taken|
into consideration. It does not take into account non-COGS related costs (e.g., R&D) that may impact on productivity, scale, etc. which would influence COGS and profit. Not applicable in case of manufacturer- distributor transactions.
Marketing Costs /Commercial Expenses
|The marketing costs are associated with|
the delivering of goods or services to customers. The marketing cost may include either the expenses associated with transferring the title of goods to a customer and the cost of promoting the goods or of the services being sold.
Marketing costs may be used as a splitting factor where the value creation is driven by the marketing activity.
|Costs are an verifiable indicator.|
Current year costs are relatively easy to obtain. Marketing costs are directly associated with a common intangible/ profit driver.
|Difficulty in deciding whether to use current|
or cumulative values. Potential need for distinguishing “routine” from “excess” marketing spending. – There may be a difference between where spending is incurred vs. where decision-making functions reside.
Brand Development Expenses
|The brand expenses are costs|
specifically linked to the brand. They may be used as a splitting factor when the value creation is driven by the brand.
|Costs are a verifiable indicator.|
Current year costs are relatively easy to obtain. Brand development costs are directly associated with a common intangible/ profit driver.
|Historical brand-related costs may be|
difficult to obtain. Difficulty in deciding which cumulative values to include, as brands are typically built over a number of years.
It may be difficult to separate brand building costs from ordinary marketing expenses.
There may be a difference between where expenses are incurred vs. where decision- making functions are taken.
|The R&D expenses are costs a company|
incurs in the process of developing new goods and services to best suit the company’s and consumers’ needs. They may be used as splitting factors especially where two or more entities concur to the development of an intangible.
|Costs are a verifiable indicator.|
Current year costs are relatively easy to obtain. R&D costs may be directly associated with a common intangible/ profit driver.
|Historical R&D costs may be difficult to|
obtain. Difficulty in deciding the amount of cumulative values for inclusion, as intangibles are typically built and maintained over a number of years. – Risk factors related to R&D spending may also vary over time with early stage spending being riskier than spending in later years.
It may be difficult to separate R&D costs from routine development, industrialization, and other ordinary expenses. There may be a difference between where expenses are incurred vs. where decision-making functions take place.
Sales or volume based splitting factors
|Turnover viewed as a market measure of value realized or created in a particular jurisdiction. It may be used where the level of sales is an indicator of contribution to the overall group profits.||This is objective if it is determined on the basis of uncontrolled transactions. It could potentially be applied in cases where soft intangibles are relevant, i.e. customer goodwill etc.||The turnover is not necessarily aligned with DEMPE functions. The turnover per se may not be a good measure for profit generation. Turnover may be affected by market/economic differences across countries (local inflation, exchange rates, etc.). It ignores contributions of the business (e.g. R&D) made outside of the market|
Volume of Trade
|Used to reflect efforts in sales/ distribution/ marketing||It is objective if determined on the basis of uncontrolled transactions.||This is similar to Turnover. Also, different|
trades may have very different market values.
Asset based splitting factors (tangible or intangible assets)
Value of Key Business Assets (incl. Intangibles)
Assets are predominantly the driver for the returns;
One should consider owned, leased or licensed assets.
|It can be a solution to complicated cases that involve unique intangibles||The book value of assets may not reflect their market value. In-house developed intangibles may not be included in the financial accounts. It may need to involve valuation techniques. Assets are not necessarily located in the same jurisdiction|
as where DEMPE functions take place.
Assets under Management
|Assets are predominantly the driver for the returns. Asset values are determined by external market transactions/ valuations as they refer to the market value of financial assets|
managed for investors.
|It can be a solution to complicated cases that involve financial assets.||It may have to be limited to the particular case of asset management|
|In circumstances where different intangibles contribute to the creation of value, the residual profit is split among the various types of IP using royalty rates as splitting factors. Alternatively, if one of the value drivers is IP for which CUPs may be available, a part of the residual profit can be attributed to the IP using the CUP.||It can be a solution to complicated cases that involve unique intangibles.|
No historical costs are needed.
|It may be difficult to find comparables for the royalty rates. If CUPs exist, it is arguable whether it be preferable to use the CUP method or a profit split.|
|The combined profit is split between the intangible developer/ owner and the intangible user, based on comparable splits from franchise agreements. In circumstances where different intangibles in combination with services contribute to the creation of value, the residual profit is split among the various types of combined IP and business services using franchise fees as splitting factors.||It can be a solution to complicated cases that involve unique intangibles and valuable services or business ideas/ models.|
Franchise agreements may be found between unrelated parties. In such cases, they can be readily (and objectively) used as a splitting factor based on market conditions (CUPs). No historical costs needed.
|It may be difficult to find comparables for the franchise fees as they capture unique combinations of IP and services and/or business methods/ideas. If CUPs exist, it is arguable whether it be preferable to use the CUP method or a profit split.|
Weights assigned on the value chain/ basis of functional analysis
|It breaks down the integrated business into its distinct, value-creating activities by way of a value chain/ functional|
|It is likely to better reflect the true value creation across the business.||It is perceived as more subjective because it is based on a qualitative analysis.|
|Provide external or market-based measures of value across the enterprise, by looking at external comparable companies that perform the different steps of the value chain.||Use of external comparable data. If there are no clear drivers for residual profit whilst there is residual profit after providing for returns on the basis of the external benchmarks, the external benchmarks may provide an objective way for splitting the residual profit.||The circumstances are usually highly case- specific, so external comparables might not be readily available. Although they are available on the market and accepted globally, external benchmarks are (mainly) used to remunerate limited risk functions. As the contributions should be unique and valuable or point out to a highly integrated business, it might be impossible to delineate the transactions in such a way as to come up with an arm’s length comparable.|
Hedge Fund Model
|Joint contribution of capital at risk and trading/ investment know-how.|
Business model where one party provides capital and strategic directions and the other party provides valuable investment/ trading know-how and day-
|Existence of third-party agreements with similar characteristics. Possibility of defining how to split the profit by using direct or indirect evidence from third- party agreements.||A hedge fund model may not be suitable for describing the particular situation of intra- group relations. It is highly industry/ market specific.|