INTM485025 - Transfer pricing operational guidance: Accurate delineation of the actual transaction: Risk

Scope of this guidance
Executive summary
The arm's length principle and accurate delineation
Step 1: Identify economically significant risks
Economic significance
Specificity
Step 2: Contractual assumption of risk
Step 3: Functional analysis to determine which entities control risk
Control of risk
Financial capacity
Steps 4 and 5: Consistency between contract and conduct and allocation of risk
Threshold for control
Allocation across entities
Step 6: Pricing the transaction, taking account of risk allocation
Risk control contributions
How to price contributions to control: when is it appropriate to use the TPSM?
Ex-ante or ex-post
Relevant profits
Application
Recognition of the accurately delineated transaction
Examples
Intangible assets
Captive insurance
Other financial instruments

Scope of this guidance

1.      This guidance addresses HMRC's application of the ‘6-step process for analysing risk’ (‘the 6-step process’) within Chapter I of the OECD Transfer Pricing Guidelines (‘TPG’).

2.      The significance of risk in a transfer pricing analysis has been a subject of debate amongst commentators following the 2017 and subsequent updates to the TPG.  In order to improve certainty in this area, HMRC is seeking to clarify its interpretation of (i) the 6-step process and (ii) its place within a transfer pricing analysis.

3.      Chapter I, including the risk framework, is focussed on economically relevant characteristics and the accurate delineation of controlled transactions. Risk assumption is only one of several aspects which are relevant to accurate delineation, and it is important to view this guidance in that context.  The TPG acknowledge that the guidance on risk within Chapter I is extensive, which is reflective of the complexity in this area as opposed to indicating an inflated importance over other factors.  It does as a matter of fact represent significantly more pages relative to the other aspects of a functional analysis required for accurate delineation, for example, functions or assets, but (per paragraph 1.59):

“The expanded guidance on risks reflects the practical difficulties presented by risks: risks in a transaction can be harder to identify than functions or assets, and determining which associated enterprise assumes a particular risk in a transaction can require careful analysis.”

4.      Though this guidance focusses on risk and specifically the 6-step process, it should not be interpreted that HMRC view the delineation of risk as having more significance than functions or assets, or the other economically relevant characteristics or circumstances of the parties to transactions.

5.      Further it is important to note that the primary focus of this guidance is delineation not pricing.  It is premature to form conclusions about pricing at the accurate delineation stage of a transfer pricing analysis, which is a matter for Chapter II onwards. The last step of the 6-step process does note, however, that a party may be compensated with a share of potential upside benefits and downside costs commensurate with its contribution to the control of an economically significant risk without it having contractually assumed or been allocated that risk, which can naturally lead to the consideration of pricing outcomes in later chapters.

6.      Because Step 6 has been a particular focus of disagreements of interpretation, this guidance does go on to consider, in general terms, what the TPG say about the transfer pricing methods which may be applied. In particular, this includes whether the Transactional Profit Split Method (‘TPSM’) might be appropriate and if so the question of which profits might be split. It explains HMRC’s general views but does not draw any definitive conclusion which may be applied to all cases. Each particular transaction must be considered in the light of its specific facts and circumstances.  For this reason, any examples within this guidance have no wider application to specific cases and only seek to illustrate particular points within this guidance.

7.      This guidance represents HMRC’s interpretation.  It highlights a number of circumstances where it is recommended that case teams discuss disputes with the Transfer Pricing Team in CS&TD Business, Asset & International.

Executive summary

8.      As noted, Chapter I, including the risk framework, is focussed on economically relevant characteristics and the accurate delineation of controlled transactions. The TPG (paragraph 1.33) are clear that this analysis in Chapter I is a prerequisite to establishing an arm’s length price for a controlled transaction, but guidance for pricing is in later chapters, primarily Chapters II & III.  This guidance focusses on risk and related functions, but this should not in practice be considered in isolation from the assets and related functions (paragraph 1.36), whether undertaken in the past or current periods.  Indeed whilst ‘DEMPE’ (development, enhancement, maintenance, protection, and exploitation) functions related to intangible assets and risk management functions should not be confused, there can be overlap (see paragraph 93, How to price contributions to control: when is it appropriate to use the TPSM?).

9.      The 6-step process identifies economically significant risks related to a controlled transaction with specificity and determines whether the contractual assumption of those risks aligns with the conduct of the parties, by reference to their capability and capacity to exert control over the risks and their financial capacity to assume them.  Where reallocation of risk is not warranted, some commentators have argued that (1) the other parties’ reward for contributions to control of risk should never include a share of the upside or downside of that risk; and (2) the TPSM will never be the appropriate TP method. HMRC do not agree with either of these positions, whether before or after the 2017 revision of the TPG. 

10.   With regard to (1), the TPG state that in such a case the accurate delineation respects the contractual risk assumption. However, whilst Chapter I doesn’t (as noted) specifically deal with how to price, it does provide in the last step of the 6-step process examples which illustrate where you might need to price contributions to risk control by companies not assuming (contractually or through allocation) that risk in a way that entails taking a share in risk outturn. 

11.   Moreover, the last step of the 6-step process explicitly provides for the possibility that parties may be compensated with a share of upside and downside commensurate with their contributions to risk control despite not contractually assuming or being allocated those risks.  

12.   Similarly, with regard to (2), HMRC does not agree with this interpretation of Step 6, which does not state any general rule as to whether a particular TP methodology might result in a share of upside or downside of a risk to a party following the allocation of that risk, only that this question should be dealt with in accordance with guidance in later chapters about the selection and use of such a TP methodology. HMRC consider that whilst this leaves open the possibility that a TPSM may be an appropriate methodology to establish a transfer price for a party allocated or contributing to the control of risk, it equally may not be appropriate, and users must be careful to not rule out other, potentially more reliable methods (i.e. by reference to paragraph 2.3, a comparable uncontrolled price (CUP) if identifiable). 

13.   As noted, it is not possible to provide comprehensive guidance that determines how to reward contributions to control of risk in all cases, but this guidance attempts to flag those instances where advice from the Transfer Pricing Team in CS&TD Business, Assets & International should be sought by case teams. Guidance around the selection and use of different pricing methodologies, which might inform reward for contributions to control, is found in Chapter II of the TPG and addressed in HMRC’s International Tax Manual (INTM421000- Transfer pricing: Methodologies: OECD Guidelines). 

The arm’s length principle and accurate delineation

14.   In the TPG, pricing in accordance with the arm’s length principle requires a comparison of the ‘conditions’ in a controlled transaction with those conditions that would have been made had the parties been independent and undertaking a comparable transaction in comparable circumstances (e.g. paragraph 1.33).

15.   This requires that the conditions and economically relevant circumstances identified for the controlled transaction are ‘accurately delineated’.  The accurate delineation of a controlled transaction requires analysis of its economically relevant characteristics, categorised (within paragraph 1.36) between:

• The contractual terms of the transaction;
• The functions performed by each of the parties to the transaction, taking into account assets used and risks assumed, including how those functions relate to the wider generation of value by the MNE group to which the parties belong;
• The characteristics of property transferred, or services provided;
• The economic circumstances of the parties and of the market in which the parties operate;
• The business strategies pursued by the parties.

16.   Focussing on the functional analysis, the TPG (in paragraph 1.56) consider this incomplete unless the material risks assumed by each party have been identified and considered since the actual assumption of risks would influence the price and other conditions of a transaction.

17.   The TPG go on (in paragraph 1.60) to explain the 6-step process for analysing risk:

1. Identify economically significant risks with specificity;
2. Determine how specific economically significant risks are contractually assumed by the associated parties under the terms of the transaction;
3. Determine through a functional analysis how the associated enterprises that are parties to the transaction operate in relation to assumption and management of the specific, economically significant risks, and in particular which enterprise or enterprises perform control functions and risk mitigation functions, which enterprise or enterprises encounter upside or downside consequences of risk outcomes, and which enterprise or enterprises have the financial capacity to assume the risk;
4. Consider whether the contractual assumption of risk is consistent to conduct of parties, specifically by reference to whether the party assuming a risk exercises control over the risk and has the financial capacity to assume the risk;
5. Where the party assuming risk does not control the risk, or doesn’t have the financial capacity to, reallocate the risk;
6. The actual transaction as accurately delineated should then be priced taking into account the financial and other consequences of risk assumption, as appropriately allocated, and appropriately compensating risk management functions.

18.   The application of steps 1-5 of the 6-step process, including the reallocation of risk, is therefore part of the accurate delineation, or identification, of the transaction to be priced.  It is not the non-recognition of a transaction and replacement with another, but the recognition of the substance of the actual transaction, beyond its legal form as explained in paragraph 1.140 of the TPG:

“Where the characteristics of the transaction that are economically significant are inconsistent with the written contract, then the actual transaction will have been delineated in accordance with the characteristics of the transaction reflected in the conduct of the parties. Contractual risk assumption and actual conduct with respect to risk assumption will have been examined taking into account control over the risk … and the financial capacity to assume risk …. and consequently, risks assumed under the contract may have been allocated in accordance with the conduct of the parties and the other facts on the basis of steps 4 and 5 of the process for analysing risk in a controlled transaction….”

Concluding:

“Therefore, the analysis will have set out the factual substance of the commercial or financial relations between the parties and accurately delineated the actual transaction.”

19.   Step 6 provides guidance as to how to price the accurately delineated transaction, but it is flagged (in paragraph 1.100), that this must be in accordance with the tools and methods as laid out in the following chapters of the TPG. The guidance for Step 6 limits itself to providing examples to illustrate how a company would be rewarded at arm’s length:

• For a risk they contractually assume and control (i.e., Company A in paragraph 1.101 / 1.83)
• For a risk they contractually assume but do not have the capability to control from an operational or financial perspective or have the financial capacity to assume (i.e., Company A in paragraph 1.103 / 1.85 which receives a risk-free return)
• For a risk they do not contractually assume which is allocated to them under Step 5 due to control (i.e. Company A in paragraph 1.102 / 1.84)
• For contributing to the control of a risk which it neither contractually assumes nor is allocated (i.e., paragraph 1.105).

Step 1: Identify economically significant risks

20.   Step 1 of the 6-step process provides guidance on the identification of economically significant risks, which must be identified with specificity.

Economic significance

21.   The TPG require that economically significant, i.e. not all, risks should be identified.

22.   Earlier, in paragraph 1.56, it is noted “in the open market, the assumption of increased risk would also be compensated by an increase in the expected return, although the actual return may or may not increase depending on the degree to which the risks are actually realised. The level and assumption of risk, therefore, are economically relevant characteristics that can be significant in determining the outcome of a transfer pricing analysis.”

23.   Further to this, and more explicitly within paragraph 1.71, it is highlighted that both the scale, and likelihood, of a realisation of a risk impacts whether it is an economically relevant characteristic to a transaction i.e. a risk which is economically significant.

24.   That analysis of scale and likelihood should be considered in absolute and relative terms in the context of that transaction.  Take the example in 1.74, where credit risk is highlighted "significant in the context of how value is created for the distribution function"; as in many transactions credit risk would not be considered economically significant but for the case of this MNE distributing heating oil it is.

25.   Paragraph 1.71 of the TPG notes there are many definitions of risk, but in a transfer pricing context it is appropriate to consider it as the effect of uncertainty on objectives of the business.  It emphasises close association between risk and profit potential:

“Risk is associated with opportunities, and does not have downside connotations alone; it is inherent in commercial activity, and companies choose which risks they wish to assume in order to have the opportunity to generate profits. No profit seeking business takes on risk associated with commercial opportunities without expecting a positive return. Downside impact of risk occurs when the anticipated favourable outcomes fail to materialise”

26.   Paragraph 1.71 is noting a correlation (i.e. a relationship) between risk and profit potential (or risk versus reward) not a causative link from the assumption of risk to the earning of profit. Put simply, any venture based on the earning of profits entails certain risks which cannot be avoided without removing that profit potential. As with any profit seeking venture, there will be some form of inherent uncertainty as to the actual profit (or loss) outcome.

27.   It is further noted at paragraph 1.71 that companies are likely to devote considerable attention to identifying and managing economically significant risks in order to maximise the positive returns from having pursued the opportunity in the face of risk.

28.   These risks which are necessarily assumed in the generation of residual profit are the group's economically significant risks, and these may be instructive as to the economically relevant circumstances of its group members.  As such they may be an economically relevant characteristic of any transaction those group members enter into (per paragraph 1.36).

29.   For example:

• If considering a transaction involving a ‘routine’ service provider, only those economically significant risks relevant to the service it provides should be considered in a comparability study because it is only these which are relevant to the reward that is attributable to the service.
• A group may offer different ranges of products / services in different jurisdictions involving the performance of different functions, utilisation of different assets and assumption of different risks. Therefore, the analysis may differ for a group service provider in one jurisdiction compared with another group in a different jurisdiction (per paragraph 1.133).

Economically significant risks cannot therefore be defined in isolation from the nature of the business of the parties to a transaction, seen in the context of the wider group.  

30.   In general, an MNE group will aim not to bear risk other than its economically significant risks or, if it must bear them, will seek to minimise the consequences of those risks crystallising – for example in some cases it may wholly or partially insure against risks that it cannot effectively control, or it may seek through its contractual arrangements with third parties not to bear such risks in the first place. By contrast it will bear and control its economically significant risks given the correlative relationship between those risks and profit potential.

31.   For example, take a generic manufacturing company:

• The business of the company is generation and fulfilment of orders for 'widgets', which it manufactures.
• It bears a risk that its factory might be destroyed, for example by some natural disaster.  This risk can be almost entirely mitigated, for example via property insurance, without significantly eroding the company's potential for profit in this business.
• It also bears the risk that it cannot generate / fulfil sufficient orders to achieve the necessary utilisation rate to meet its desired return on investment in the factory.  It can mitigate this utiliation risk, for example linked to the destruction risk through business interruption insurance, however it is unlikely to be possible to fully insure / guarantee against it.

This example illustrates HMRC's interpretation of the threshold for a risk to be considered 'economically significant' - it is one which cannot be substantially mitigated without significantly eroding a business's profit potential i.e. in the example above, the risk of destruction of the factory is not in itself an economically significant risk, but utilisation risk is likely to be.

32.   The TPG provide a list of sources of risk (in paragraph 1.72) as a framework to assist a transfer pricing analysis and considers the range of risks likely to arise from commercial and financial relations.  It is noted this list is not intended to:

• Be exhaustive;
• Suggest any hierarchy;
• Be rigid, as there may be overlap between categories; or
• Infer that externally driven risks (e.g. hazard) are less relevant than internally drive ones (e.g. infrastructure).

Specificity

33.   Paragraph 1.72 notes

“Risks which are vaguely described or undifferentiated will not serve the purposes of a transfer pricing analysis seeking to delineate the actual transaction and the actual allocation of risk between the parties”.

34.   Economically significant risks should be identified with specificity, and not bundled.  To do otherwise would potentially conflate / obscure the relevant control functions which are relevant to each specific economically significant risk and therefore the accurate delineation.  It is necessary to consider each specific risk, whether it is relevant to the transaction, the controls that are relevant to that risk, and the potential upside and downside consequences.

35.   This exercise requires balance and pragmatism. It is equally inappropriate to be too granular as it is to generalise. The actual commercial practice / control structures in a group can help inform the analysis to identify the economic significance of risk(s) with specificity i.e., two risks may be identified as economically significant and have structures in place for their control within a group – if those structures differ for each risk, they must be considered separately.   Where there is commonality of structure for the control of two distinct economically significant risks it may be possible to consider them as a bundle.  

36.   As illustrative examples (which will have no wider application, as each case must be judged on its own facts and circumstances):

• A group that sells branded goods may choose to control a marketing risk for a major brand separately from other marketing risks.
• A bank may employ different control structures for credit risk depending upon certain criteria, such as the size of a loan or the identity of the borrower.
• An oil major may subdivide production risk into many sub-risks associated with exploration, drilling, blow-out protection and so on.

These sub-risks may represent different specific economically significant risks and may be broken down further according to the group’s actual business practices.

37.   The purpose of accurate delineation is to observe the facts of a transaction. It does not necessitate a business to structure its normal commercial activity in some predefined way. We do not expect full control over all risks related to a transaction but are interested in the actual controls  in place which will inform a party’s view of its economically significant risk. A business’s attention will naturally be targeted at those risks which are most economically significant. Taking a ‘hazard’ risk like Covid 19– identifying it as an economically significant risk of a group is likely to be seeking to define a risk along different lines to the approach taken by the group.  Groups are unlikely to have had controls in place for a pandemic (hard to effectively control, high impact, but low likelihood), but its effects will have a bearing on the size of the potential upside / downside of economically significant risks which were controlled e.g. a market risk such as loss of sales through lack of differentiation from competitors, or operational risks such as supply chain disruption and staffing shortages.

38.   Alternatively, as paragraph 1.71 notes, a risk with limited economic consequences to a transaction is unlikely to be included as an economically significant risk, especially if the controls that are relevant to it are not distinct from controls that apply to a broader category of risk that includes the risk in question. In the example in paragraph 1.71 the controls relevant to the new flavour of ice cream are likely to be identical to those that apply to the product, or range of products more generally. Hence the requirement for specificity does not extend to this level of detail.

Step 2: Contractual assumption of risk

39.   The TPG (in paragraph 1.77) notes that the party or parties assuming risk related to a transaction may be set out in written contracts between them. 

40.   A contractual assumption of risk constitutes an ex-ante (anticipated, before the event) agreement to bear some or all of the potential costs associated with the ex-post (actual, after the event) materialisation of downside outcomes of risk in return for some or all of the potential benefit associated with the ex-post materialisation of positive outcomes.

41.   Understanding the contractual assumption of risk is a crucial first step in the analysis, which is necessary before proceeding to later steps. Whether contractual assumption is consistent with the conduct of the parties and their financial capacity to bear risk must then be assessed (per paragraph 1.81):

“It should not be concluded that the pricing arrangements adopted in the contractual arrangements alone determine which party assumes risk. One may not infer from the fact that the price paid between associated enterprises for goods or services is set at a particular level, or by reference to a particular margin, that risks are borne by those associated enterprises in a particular manner. For example, a manufacturer may claim to be protected from the risk of price fluctuation of raw material as a consequence of its being remunerated by another group company on a basis that takes account of its actual costs. The implication of the claim is that the other group company bears the risk. The form of remuneration cannot dictate inappropriate risk allocations. It is the determination of how the parties actually manage and control risks, as set out in the remaining steps of the process of analysing risk, which will determine the assumption of risks by the parties, and consequently dictate the selection of the most appropriate transfer pricing method.”

42.   Contractual terms may have inherent value, irrespective of underlying functions or control.  A possible scenario is exclusivity options acquired via an uncontrolled transaction between a group entity and independent party which subsequently impact an intra-group controlled transaction.

43.   As noted above, the accurate delineation of a controlled transaction includes not simply analysing risk as part of a functional analysis, but overlaying that on the contractual terms of the transaction and considering other characteristics inherent in the relevant property / services (including for example, exclusivity of rights to intangibles – see paragraph 6.118), wider economic circumstances and business strategies.  Further, the TPG state that parties to a transaction will consider the options realistically available to them.  It suggests no hierarchy of importance for the different economically relevant characteristics i.e., it doesn’t rule out contractual terms having more significance than control of risk, nor does it rule out economic circumstances having an impact on conduct against those contractual terms.

Step 3: Functional analysis to determine which entities control risk

Control of risk

44.   What constitutes ‘control’ is a question of fact in an individual case. The TPG provide some limited assistance in answering this question with reference to ‘risk management’ which it defines (in paragraph 1.61) as the:

• Capability to make decisions to take on, lay off, or decline a risk-bearing opportunity, together with the actual performance of that decision-making function;
• Capability to make decisions on whether and how to respond to the risks associated with the opportunity, together with the actual performance of that decision-making function; and
• Capability to mitigate risk, that is the capability to take measures that affect risk outcomes, together with the actual performance of such risk mitigation.

45.   Control over risk (per paragraph 1.65), requires only the first two elements of risk management, with mitigation capable of being outsourced.  However, 

“where these day-to-day mitigation activities are outsourced, control of the risk would require capability to determine the objectives of the outsourced activities, to decide to hire the provider of the risk mitigation functions, to assess whether the objectives are being adequately met, and, where necessary, to decide to adapt or terminate the contract with that provider, together with the performance of such assessment and decision-making.”

46.   The tests above are two-fold, requiring (i) capability to perform and (ii) actual performance.  Capability (per paragraph 1.66) covers:

• The competence and experience in the area of the particular risk;
• Possessing an understanding of the impact of their decision on the business; and
• Having access to the relevant information to support their decision making.

It is also considered that competence covers practical capability / capacity i.e., is it realistic to ascribe control or decision making over a group’s strategic development programmes to one director, if in practice this assumes they have made complex decisions with little time to gather and assess information?

47.   Control must be defined for each of the economically significant risks relevant to a transaction which, as noted above, will be informed by the specificity adopted in actual MNE group control structures. It is not an abstract question that may be satisfied by the presence of any element of risk control in the contract.  Different parties (including parties not named in the contract) might feed, perhaps over a period of time, into a decision-making process, and this would be relevant to Step 6.  That process is informative in identifying, but does not represent, the actual point (within that process) at which a decision is made, by which party and therefore where risk is managed and controlled. 

48.   Paragraph 1.76 discusses and compares:

• Where a board / executive committee of a group set the level of risk the group as a whole is prepared to accept in order to achieve commercial objectives, and establish a control framework for managing and reporting risk in its operations; and
• Line management in business segments, operational entities and functional departments which may identify and assess risk against commercial opportunities and put in place appropriate controls and processes to address risk and influence the risk outcomes arising from day-to-day operations.

In the example it then suggests that the wider policy setting in a group does not necessarily represent control over risk at the level of the transaction between two affiliated parties.

49.   Paragraph 1.76 has been cited by different commentators as suggesting decision making at line management is closer to the meaning of risk mitigation than control. 

50.   It is reasonable to assume that at a sufficiently broad level all risks will be controlled to some degree at the level of the board of a group, which takes ultimate responsibility for all decisions made throughout the group. 

51.   However, the requirement for specificity is relevant here. Whilst a board may set the guard rails within which its line management can take risk, the decision to take on risk may be made by line management within those parameters. There are various potential alternatives depending on the facts:

• Policy parameters may be wide, and therefore the decisions of line management within those parameters may still constitute control of a specific economically significant risks in their own right; 
• They may be narrow such that the establishment of the parameters by the board is the decision in point;
• They may entail a process of feedback and approval loops which constitutes continual monitoring and policy adaptation which may amount to control of specific economically significant risks, or may amount to mere ‘rubber stamping’.

This is also discussed within the Authorised OECD Approach (‘AOA’) on the Attribution of Profits to Permanent Establishments, which applies the TPG by analogy (for example in paragraph 76 onwards, Part III of the 2010 Report).

52.   Our view is the TPG show no bias towards any level of management, and determination for any transaction for within an MNE group must follow from a thorough understanding of the group’s commercial practices and how it makes decisions to take on, lay off or decline risk bearing opportunities, and whether and how to respond to risks associated with those opportunities.

53.   Therefore, in the example at paragraph 1.83 Company A’s ultimate accountability is not conclusive of who controls the specific risk in question. Unless Company A is the ultimate parent of the MNE group, it will have to obtain approval for a significant development opportunity from its board. This fact does not disturb the conclusion set out in the paragraph that development risk is controlled by Company A (and the contractual assumption of risk by Company A remains undisturbed by the six-step risk allocation process).

Financial capacity

54.   The TPG refer to the financial capacity to assume risk, in paragraph 1.64, as access to funding to take on or lay off the risk, to pay for risk mitigation functions and to bear the consequences of the risk if it materialises.

55.   It goes on to note that access to funding by the party assuming risk takes into account available assets and the options realistically available to access additional liquidity, if needed to cover the costs anticipated to arise should the risk materialise.  This will take into account the potentially beneficial impact of group membership i.e., implicit support as discussed in paragraph 10.76 onwards of Chapter X)

56.   As such, where an entity has the ability to control economically significant risks effectively and so enhance capital value, it is reasonable to suppose that capital could move to exploit that opportunity.  As noted in paragraph 1.64

“…exploitation of rights in an income-generating asset could open up funding possibilities for that party.” 

57.   That, as discussed in Step 6, is not delineating an intra-group funding arrangement with that party.  But, in practice this means that at arm’s length pricing will tend to correlate value with effective control of risk.  Again, as noted in the paragraph

“Where a party assuming risk receives intra-group funding to meet the funding demands in relation to the risk, the party providing the funding may assume financial risk but does not, merely as a consequence of providing funding, assume the specific risk that gives rise to the need for additional funding"

58.   This is the basis for the risk control framework in Chapter I, and the explicit emphasis on alignment with value creation in BEPS Actions 8-10. Although this principle was already well embedded in earlier versions of the guidelines, the 2017 version emphasises the importance of this relationship.

Steps 4 and 5: Consistency between contract and conduct and allocation of risk

Threshold for control

59.   Where, following Steps 1 to 3, it is determined that the party contractually assuming a risk, in its conduct, actually controls the risk and has the financial capacity to assume the risk, there are no further steps to allocate risk (per paragraph 1.87).  This does not mean Step 6 – which explicitly notes the pricing of contributions to control should be considered (See Step 6: Pricing the transaction, taking account of risk allocation) – becomes redundant i.e., the TPG state:

• In paragraph 1.87 (Step 4) that where contractual risk assumption is validated "there is no need to consider step 5, and the next step to consider is step 6".
• In paragraph 1.98 (Step 5) when risk has been allocated to the most appropriate party “other parties performing control activities should be remunerated appropriately, taking into account the importance of the control activities performed”.

60.   There does not appear to be any explicit de-minimis level of control envisaged within the TPG.  This is because, as described above, the delineation of where control of risk sits differs from the question of value inherent in contributions to control of that risk.  Those contributions might be considered part of a process towards which a decision is made, but where that decision is made should be a matter of fact. To reiterate, control should be:

• Related to the relevant economically significant risk i.e., if an entity assumes a bundle of risks, but only demonstrates control over one, the others should be subject to reallocation under Step 5; and
• Held by an entity with the capabilities, which it actually exercises, to manage those specific risks and the financial capacity to assume them.

61.   It should be remembered that risk control does not necessarily require the undertaking of risk mitigation, which may be outsourced.  In determining whether a party controls risk it contractually assumes, despite outsourcing or sharing elements of its management, the TPG ask whether comparable arrangements can be identified in uncontrolled transactions (paragraph 1.97).   Cases where significant control over key strategic assets, products, or services to a group is outsourced should be examined closely, to consider whether this is something which is observable between independent parties within similar economic circumstances.

62.   The need to consider risk with specificity may lead to an analysis which differentiates financial risk and its control from operational risk.  This is illustrated in Example 6, Annex I to Chapter VI, which references paragraph 6.63 and 6.64, highlighting circumstances where funding decisions overlap with development ones.  It is accepted by HMRC that the level and complexity of risk management within MNE groups for intra-group funding may not equate to that seen between independent lenders – what is important is establishing, in the context of the specific control environment of that specific group, whether the party to the controlled transaction had the capability to, and made, the decisions to take on financial risk.

63.   Below are a variety of examples designed to illustrate the consideration discussed above (as before, they have no wider application, as each case must be judged on its own facts and circumstances).  The bare facts of this hypothetical case are that it has been established that the development of IP owned by Company A is controlled by a committee of senior employees employed by affiliates located across an MNE. 

• If Company A’s only participant in that committee is a single employee, who is a senior marketing executive (i.e., with no experience relevant to IP development), they may not have the capability to exercise any form of control over the development risk (distinct from other economically significant risks such as risks inherent in commercialisation). In this sense, mere membership on a decision-making body does not automatically mean Company A is a decision maker.
• Committee decisions may be made unanimously or by majority agreement (or some other threshold).  If the decision is made by majority, it may be demonstrable that another company, Company B, controls the committee and Company A does not where Company B employs a quorum of committee members who represent a majority and have the ability to direct the decisions of that committee.
• The committee delegates day-to-day management of third-party contract research organisations to a project management team employed in Company B.  It may be that this project management is comparable to third party transactions and demonstrably an outsourceable risk mitigation function meriting only a routine reward.  The risk the committee manages would be that the contract research organisation performs its function competently. Alternatively, day-to-day management may exceed the type of mitigation functions commonly outsourced in such examples, and may itself amount to control of economically significant risks.
• The most difficult examples may be where Company A contributes a significant minority of participants to a committee, with one or more other companies providing the balance, but there isn’t a consistently applied policy for how agreement is reached.  Depending on the precise facts and circumstances it might be concluded that Company A retains control by reference to paragraph 1.94, although the ambiguity over the relative level of control may factor into the method applied to price the contributions of the other companies (see 'Risk control contributions' below).
• The above bullets consider, for simplicity, that financial risk is so highly interrelated to development risk that its control cannot be easily distinguished.  It might be possible that Company A’s participant, or more broadly the committee’s make-up, are senior C-suite / finance executives with sufficient industry experience to make decisions to take on financial risk associated with development, but decisions on how that funding is specifically applied to specific IP is taken elsewhere.  This again may factor into the method applied to price other contributions to control (again, see ‘ Risk control contributions’ below).

64.  In cases where there is dispute as to whether a party controls the risk that they contractually assume, the advice of Transfer Pricing Team at CS&TD Business, Assets & International should be sought by case teams.

Allocation across entities

65.   Step 5, in paragraph 1.98, states risk contractually assumed by an entity without the ability or financial capacity to control it, should be allocated to the entity which does, or if multiple candidates are identified, then to the entity (or group of entities) exercising the most control.  Any remaining entities performing control activities should (as noted) be remunerated as appropriate.

66.   Some commentators have questioned the practical difficulties of identifying relevant control, for example:

• Highlighting that certain individual decisions can be more important than others in managing the profit potential, depending on the economically significant risks e.g., the initial decision to invest in a large low risk capital project, as compared to investment in the development of an innovative product requiring more regular oversight / decisions;
• Citing difficulties in MNE groups e.g., matrix management structures, where decision making may arise at the end, or a point in, a collaborative process. This is compounded where globally mobile workforces are not confined to a set location.  As noted, (in paragraph 47, Control of risk) the actual point a decision is made has to be identified to determine where actual control is exercised.

67.   HMRC accepts that control over different economically significant risks will be exercised in different ways and frequency across different transactions and industries.   This is why it is important to consider the particular economically significant risks and control structures for the situation under consideration, and to apply the 6-step process to the actual facts and transaction under consideration. In the example given above, it may be that early, one off, decisions have greater comparative importance in the capital project than in the innovation product development.

68.   The focus of the TPG on identifying economically significant risks with specificity is intended to increase the practicality, by limiting the scope, of accurate delineation, as it follows that such risks will be the subject of focus by MNE groups (as mentioned above). 

69.   As with the assessment as to whether a party contractually assuming risk controls that risk, there is no fixed threshold for a contribution control in another party which does not contractually assume the risk. 

70.  Again, in cases where there is dispute as to which party controls the most risk for reallocation purposes, the advice of Transfer Pricing Team in CS&TD Business, Assets & International should be sought by case teams.

Step 6: Pricing the transaction, taking account of risk allocation

71.   The assumption of economically significant risks is relevant to pricing, but it is only part of a full accurate delineation. Pricing is determined only following application of later chapters of the TPG, and conclusions must be drawn by reference to all the economically relevant conditions of the controlled transaction and in the light of all of the guidelines.  For example, in some circumstances the assumption of risk will not prevent the transaction from being compared to uncontrolled transactions having similar risk characteristics as a basis for pricing.

72.   It is perhaps, prima facie, counterintuitive for a party assuming an economically significant risk to be rewarded using the transactional net margin method (‘TNMM’) as this can have the practical effect of ‘dampening’ the impact of risk outcomes within an arm’s length range determined by that party through benchmarking, but this is not ruled out in Chapter II.

Risk control contributions

73.   As highlighted above, the TPG limit commentary on Step 6 within Chapter I to stating the accurately delineated transaction (as established by Steps 1-5) should be priced in accordance with tools and methods set out in the following chapters.  It then goes on to highlight examples which illustrate where a company would be rewarded at an arm’s length:

• For the risk it contractually assumes and controls (i.e., Company A in paragraph 1.101 / 1.83) as verified by Steps 1-4;
• For the risk it assumes, not through contractual assumption but by allocation under Step 5, due to control (i.e., Company A in paragraph 1.102 / 1.84);
• For the risk it contractually assumes but do not have the capability to control from an operational or financial perspective (i.e., Company A in paragraph 1.103 / 1.85 which receives a risk-free return);
• For contributing to the control of risk which it neither contractually assumes nor is allocated (i.e., paragraph 1.105).

74.   It is therefore clear that all risk management functions relevant to an economically significant risk must be identified, regardless of whether the contractual allocation of risk is respected as the pricing of all contributions to control is required. Put simply, a contribution by one party to the control of a risk assumed by another is prima facie an economic relationship between the two which may be rewarded at arm’s length, and therefore it is necessary to consider what reward that contribution would earn.  

75.   Paragraph 1.105 addresses the reward for contributions to risk control by a party that is not assuming or allocated the risk [emphasis added]:

A party should always be appropriately compensated for its control functions in relation to risk. Usually, the compensation will derive from the consequences of being allocated risk and that party will be entitled to receive the upside and incur the downside…In circumstances where a party contributes to control of risk but does not assume the risk, compensation which takes the form of a sharing in the potential upside and downside, commensurate with the contribution to control, may be appropriate.

76.   This explicitly leaves open the possibility that reward for control functions outside of the contractual assumption or formal reallocation of risk (‘contributions to control’) could be appropriate. To read otherwise would be clearly inconsistent with paragraph 1.98 (Step 5) which states that if one party’s contractual risk assumption is respected then

“the other parties performing control activities should be remunerated appropriately, taking into account the importance of the control activities performed.”

77.    The guidance relating to Step 4 of the TPG does not explicitly state how other contributions to control should be priced, but it:

• Does require consideration of Step 6 (in paragraph 1.87);
• Doesn’t disapply any of the TPG, for example Chapter VII on intra-group services.

It would therefore be logically inconsistent with the guidance on Step 5 (and elsewhere in the TPG), and against the intention of the BEPS project, to consider that respecting a contractual arrangement means functional contributions from affiliated parties should not be priced in accordance with Chapter II onwards i.e., we read "allocated risk" in paragraph 1.105 as "contractually assumed or allocated risk".

How to price contributions to control: when is it appropriate to use the TPSM?

78.   HMRC accepts that, in most cases, it will be appropriate to price contributions to control of risk, without the assumption of risk, using a ‘one sided method’ e.g. cost plus, comparable uncontrolled price (‘CUP’) or TNMM.  It is important to consider the principles of Chapter II of the TPG to determine which method is appropriate in any specific case. The TPG do not however preclude the use of TPSM to price the reward for such activities. The following paragraphs discuss this possibility in more detail.

79.   Some have argued that:

• The 6-step process permits allocation of any share of residual profit to companies only if they are formally allocated associated risk in Step 5; and
• Pricing contributions to the control of risk using TPSM ‘reopens’ the delineation of risk (already addressed in Steps 4 and 5).

80.   Emphasis should be given to the words “compensation which takes the form of a sharing in the potential upside and downside, commensurate with the contribution to control, may be appropriate” in paragraph 1.105. 

81.  This does leave open the possibility that contributions to the control of risk could be rewarded by reference to a TPSM, but:

• it does not state that TPSM should be applied automatically; and
• this reward must be commensurate with the relative nature of that contribution i.e. in as of itself, control of risk may only merit a share of ex-ante profit in a TPSM (see Ex-ante or ex-post), unless it is accompanied by assumption of economically significant risks (per paragraph 2.159).

82.   The TPSM is not a method of delineation (i.e., assessment of the assumption of risk) but a method of pricing contributions / functions in a controlled transaction.  The fact that it may involve the sharing of profits and losses is a recognition of the imperfect nature of allocating an economically significant risk to one party to a transaction in a modern MNE where risk management functions may span both parties to the transaction and other members of that group.  The delineation, both the assumption of risk and identification of risk management services contributing to its control, provides a foundation from which an appropriate transfer pricing method / reward should be selected.

83.   The selection of the most appropriate transfer pricing method, and the nature and quantum of the reward will depend upon a range of factors, including:

• The pricing of routine contributions (which may include funding / capital);
• Whether it is possible to benchmark the return for risk control contributions against comparable uncontrolled transactions;
• Whether the risk control contribution is itself, or is part of, a unique and valuable contribution or is part of a highy integrated control framework, which would prevent reliable benchmarking;
• The upside and downside consequences related to that risk e.g., the need to consider risk with specificity may lead (as noted above) to an analysis which differentiates:
  • financial risk and its control from operational risk leading to different approaches to reward for contributions to the control of each
  • the relative level of autonomy of line management from boards in terms of risk control (as discussed in paragraph 51, Control of risk).

84.   Paragraph 1.100 notes pricing should be in accordance with methods laid out elsewhere in the TPG i.e., whether profit allocation is appropriate and the extent of any profit allocation (i.e. through a TPSM, or traditional transaction method) will depend on the factors noted above.

85.   There are three indicators of the potential appropriateness of the TPSM in Chapter II (paragraph 2.126):

• The existence of unique and valuable contributions which "represent a key source of actual or potential economic benefits in the business operation" (paragraph 2.130).
• A high level of integration of business operations such that "the way in which one party to the transaction performs functions, uses assets and assumes risks in interlinked with, ... the way in which another party to the transaction performs functions, uses assets and assumes risks" (paragraph 2.133).
• The shared assumption of economically significant risks (or separate assumption of closely related economically significant risks).

86.   A feature of both unique and valuable contributions, and highly integrated operations, is the difficulty identifying comparables:

• “they are not comparable to contributions made by uncontrolled parties in comparable circumstances” (paragraph 2.130)
• “one party to the transaction performs functions, uses assets and assumes risks is interlinked with, and cannot reliably be evaluated in isolation from, the way in which another party to the transaction performs functions, uses assets and assumes risks” (paragraph 2.133).

87.   Where there is a high degree of integration of business operations, and specifically control functions, it is more likely that contributions to control of risk don’t result in a reallocation of risk but require pricing using the TPSM.

88.   When considering contributions to control of risk by a party where there is no risk assumption, the third indicator – shared assumption of risk - will not apply, meaning disagreement as to whether the TPSM should be used is often centred on whether the contributions can be priced by reference to a comparable.

89.   Transactions between independent parties with different economically relevant characteristics and circumstances are not appropriate comparables where reliable comparability adjustments cannot be made (paras 2.16-2.17).  For example, it is unlikely that the structure of the reward to a hedge fund manager will be sufficiently comparable to assist in the pricing of contributions to control of risk by a service provider in another sector. That is not to say that where there is commonality within a sector in the structure of reward, even if there are no appropriate comparables and reliable comparability adjustments cannot be made, it might assist pricing e.g. it might be common in a sector for performance related service fees to be paid, where profits and losses are shared, but only via a clawback / reduced future fees.

90.   However, as the TPG note (paragraph 2.144), “the lack of comparables alone is insufficient to warrant the use of a transactional profit split”. The TPG note that “practical considerations dictate a more flexible approach to enable the CUP method to be used and to be supplemented as necessary by other appropriate methods, all of which should be evaluated according to their relative accuracy. Every effort should be made to adjust the data so that it may be used appropriately in a CUP method.”

91.   Ultimately a judgement must be made as to whether the use of the CUP method, with adjustments, provides a more reliable result than the use of a TPSM.

92.   This issue is most frequent in cases involving intangible assets. Paragraphs 6.56 – 6.58 in Chapter VI (Intangibles) identify ”certain important functions" which have special significance such as (but not restricted to):

• Design and control of research and marketing programme;
• Direction of and establishing priorities for creative undertakings including determining the course of “blue-sky” research;
• Control over strategic decisions regarding intangible development programmes; and
• Management and control of budgets.

Paragraph 6.57 explicitly acknowledges that:

“Because it may be difficult to find comparable transactions involving the outsourcing of such important functions, it may be necessary to utilise transfer pricing methods not directly based on comparables, including transactional profit split methods and ex ante valuation techniques, to appropriately reward the performance of those important functions…”

93.   The functions listed in 6.56 are DEMPE functions. Control of DEMPE and risk management functions should not be automatically equated, although there can be overlap e.g., control over decisions regarding IP programmes constitutes both risk control and a DEMPE function; whereas maintaining patent registration is a DEMPE function, but most likely characterised as risk mitigation.

94.   This is reflected in paragraph 6.70 which notes a company may be entitled to a share of profit if either ”it performs important functions as reflected in paragraph 6.56" or "contributes to control over economically significant risks as established in paragraph 1.105".  The use of either / or in paragraph 6.70 does not suggest any intended restriction of the profit split by reference to important functions, which would also be inconsistent with the rest of the TPG. It therefore clearly envisages TPSM could be the most reliable method in certain cases.

95.  In cases where there is dispute as to whether TPSM should be used to price contributions to control of risk, or the CUP method is applied with adjustment, the advice of Transfer Pricing Team in CS&TD Business, Assets & International should be sought by case teams.

Ex-ante or ex-post

96.   Once it has been concluded that TPSM is the most appropriate method, it must then be determined whether the profits to be split should be (i) anticipated (i.e., ex-ante); or (ii) actual (i.e. ex-post). Assumption of risk is only relevant in the latter, which includes variations from anticipated profits as affected by the playing out of economically significant risks.

97.   Paragraph 2.159 notes that the use of ex-post profits would only be appropriate where the accurate delineation of the transaction shows "that the parties either:

• share the assumption of the same economically significant risks associated with the business opportunity; or
• separately assume closely related, economically significant risks associated with the business opportunity

And consequently should share in the resulting profits or losses.”

98.   Paragraph 2.160 implies that if the TPSM is found to be appropriate only because either or both parties make unique and valuable contributions or are highly integrated, but one of the parties doesn’t share in the assumption of economically significant risks, the split of anticipated (or ‘ex-ante’) profits "would be more appropriate".

99.   Prima facie, it might be read that when contributions to control are made by a party not assuming risk, their reward must be by reference to ex-ante profits.

100. However, that would contradict other, explicit, statements in the TPG. For example, paragraph 2.137 of the transactional profit split guidance:

“Where a party contributes to the control of economically significant risk, but that risk is assumed by the other party to the transaction, this may, in some cases, demonstrate that it is appropriate for the first party to share in the potential upside and downside associated with that risk, commensurate with its contribution to control...”

And paragraph 6.70 of the guidelines:

“A party which is not allocated the risks that give rise to the deviation between the anticipated and actual outcomes under the principles of Sections D.1.2.1.4 to D.1.2.1.6 of Chapter I will not be entitled to the differences between actual and anticipated profits or required to bear losses that are caused by these differences if such risk materialises, unless these parties are performing the important functions as reflected in paragraph 6.56 or contributing to the control over the economically significant risks as established in paragraph 1.105, and it is determined that arm’s length remuneration of these functions would include a profit sharing element.”

101. It seems clear from this that parties contributing to the control of risk may merit reward based on ex-post profit where there is evidence that such a split would be agreed at arm’s length, e.g., where a comparable uncontrolled transaction is identified with ex-ante agreement to split actual profits / losses in a certain way.  Such arrangements may provide for variable pricing to address alternative ex-post scenarios (for example, but not limited to: best, base, and worst case forecasts for sales, or savings generated in procurement against a baseline).  Ex-ante arrangements designed to deliver consistent, or variable, ex-post shares of profit across the anticipated scenarios may also be present in the market.

102. It is worth reiterating at this point that the accurate delineation of the actual transaction is key.  As noted earlier in this Guidance economically significant risks should be identified with specificity, and where those risks are controlled is a matter of fact.  It follows that there may be cases where, prima facie, risk is controlled by the party contractually assuming it but upon further investigation that risk should be considered with more granularity and it is concluded that in fact that one risk represents several economically significant risks.  Because the 6-step process for each risk will differ it may be determined that some, but not all, should be reallocated, for example:

• Where boards / senior executives of the party contractually assuming risk delegate responsibilities to another party, and this responsibility entails sufficient risk as to be economically significant in its own right (see paragraph 51, Control of risk);
• Where financial risk is controlled by the party contractually assuming the risk, but that party is not capable of controlling and does not actually control operational risks (see paragraph 62, Threshold for control).

In such cases there would be shared assumption of economically significant risks, or separate assumption of closely related economically significant risks, which might merit a sharing of ex-post profits. 

103. In cases where there is dispute as to whether ex-ante or ex-post profits are relevant to a TPSM, the advice of Transfer Pricing Team in CS&TD Business, Assets & International should be sought by case teams.

Relevant profits

104. Various parts of the TPG refer to the sharing of upside and downside commensurate with that contribution of control of risk (for example, paragraph 1.105 & 2.137).

105. In many cases this might involve appropriate reward for the financial risk assumed by the party which contractually assumed the risk.  For example, in the cases of Company D and F in paragraph 1.119, Company F is entitled to a risk-adjusted rate of return to reflect its funding risk.  That rate of return would reflect the likelihood and scale of potential down-side (loss of amounts funded) coupled against the timing and variability of return.  This might account for a substantial amount of the ex-post profits otherwise attributed to Company D.

106.         Determining the reward for financial risk assumed may not always be so simple.  Paragraphs 6.63-6.64 note:

“The extent and form of the activities that will be necessary to exercise control over the financial risk attached to the provision of funding will depend on the riskiness of the investment for the funder, taking into account the amount of money at stake and the investment for which these funds are used…

…When funding is provided to a party for the development of an intangible, the relevant decisions relating to taking on, laying off or declining a risk bearing opportunity and the decisions on whether and how to respond to the risks associated with the opportunity, are the decisions related to the provision of funding and the conditions of the transaction. Depending on the facts and circumstances, such decisions may depend on an assessment of the creditworthiness of the party receiving the funds and an assessment of how the risks related to the development project may impact the expectations in relation to the returns on funding provided or additional funding required…

…The higher the development risk and the closer the financial risk is related to the development risk, the more the funder will need to have the capability to assess the progress of the development of the intangible and the consequences of this progress for achieving its expected funding return, and the more closely the funder may link the continued provision of funding to key operational developments that may impact its financial risk...”

107. In short, the greater the likelihood and scale of potential upsides and downsides of the business risk which funding is provided for, the harder it may be to distinguish control over that risk from the control required for the financial risk. 

108. As an example, there may be a Company D which funds development of IP, which is undertaken by another Company E.  Company E’s functions include risk management, it makes decisions to take on development programmes, but the question is whether the development risk is capable of distinction from the financial risk assumed by Company D:

• It may be that Company D employs executives with the capability to assess the IP development programmes undertaken by Company E, and makes decisions to release funds to these.   In such a case Company D will control its financial risk, and the question will be whether Company E:
  • Controls a discrete development risk i.e. the decisions it makes on how to spend funds from Company D constitute control of specific economically significant risks in their own right - as noted (in paragraph 102, Ex-ante or ex-post), this is an example where a TPSM splitting ex-post profits is appropriate, or
  • Makes contributions to control of the development risk which is controlled by Company D by virtue of its decisions – Company E might possibly be rewarded via a TPSM but this would be based on ex-ante profit (absent a comparable demonstrating ex-post profits are shared).

In either case Company D would be due a risk-adjusted return for its financial risk. 

• Alternatively, Company D may not have the capability to assess the IP development programmes and it may be determined that all economically significant risk it contractually assumed should be allocated to Company E (or elsewhere), along with any commensurate reward leaving it with no more than a risk-free return.

109. Further, as noted in 1.81, pricing arrangements adopted between group entities do not alone determine which party assumes risk.  Take another example of a Company (F) owning IP, contributing to its development, and funding another group Company (G) to provide development services, for which it is rewarded on a cost-plus basis.  It is possible after a review of the facts that the following is determined through accurate delineation:

• Company G owns its own IP, and employs and funds Company F to provide reciprocal development services;
• The control structure for financial and development risk is so highly integrated as to make it impossible to distinguish distinct independent control by either Company F or Company G over such risks related to their own IP.

It is therefore concluded that the transaction between Company F and Company G is not comparable to the contract R&D services on which the cost-plus reward was benchmarked, and that instead TPSM is an appropriate method due to both parties making unique and valuable contributions of IP, being highly integrated and separately assuming closely related risks. The transaction may be delineated as a co-development co-promotion agreement i.e. a type of profit splitting arrangement commonly seen in the sector in which the Companies operate.

110. In cases where there is dispute as to whether funding risk and business risk control are separable, the advice of Transfer Pricing Team in CS&TD Business, Assets & International should be sought by case teams.

Application

111. It should be reiterated that TPSM will not be applicable in all cases, and it has no primacy over other transfer pricing methods in Chapter II.  Where there are disputes it will often be the case that there are complicated fact patterns, and it will warrant careful consideration as to whether TPSM provides a more reliable result. 

112. Care should be taken in cases where a limited number of employees of an entity contribute to the control of a risk (i.e. a lack of ‘critical mass’) allocated to another entity. Whilst there is no materiality threshold for profit split, options realistically available to the entity allocated the risk may be wider in cases where it relies on a few individuals outside of its employ, but this will depend on the uniqueness of those employees’ contribution and the impact it has on the bargaining power of the entity employing them.

113. To revisit (but vary) our example Company A, its IP, and the MNE group-wide development committee (paragraph 63, Threshold for control - again noting it is designed to illustrate the consideration discussed above and has no wider application as each case must be judged on its own facts and circumstances): 

• Company A controls the development risk that is contractually allocated to it, but another Company B employs participants contributing to the control of development risk, for example the CEO and Deputy CEO, founders, and significant shareholders in the group, with relevant experience and capability, and whose approval is necessary to pursue R&D projects and at periodic intervals across the lifespan of a product’s development;
• If Company B is not making a unique and valuable contribution and is not highly integrated with the operations of Company A then whilst it might be difficult to benchmark their contribution, as independent groups wouldn’t outsource such decisions to third parties, every effort should be made to find comparables (including with reasonable adjustment).  As such it might be determined a benchmarkable reward (e.g. TNMM, cost plus, CUP, perhaps a justifying a gain/loss share) is appropriate 
• If, however, it is established that Company B’s contributions are unique and valuable, and / or highly integrated in nature (for example employing a significant minority of research department heads which sit on the development committee) this would merit the use of TPSM and a split of ex-ante profits (in the absence of an identified CUP which supports a split of ex-post profits).  The TPSM would generally have to take account of Company A’s contribution of funding, and the extent to which development and financial risk are closely linked (per 6.64).
• It is possible after further investigation:
  • It is determined that the risk contractually assumed by Company A should be considered with more granularity and it is concluded one or more economically significant risk should be reallocated to Company B.  In such a case a TPSM of ex-post profits might be appropriate (see paragraph 102, Ex-ante or ex-post); or
  • The wider delineation justifies Company A and B sharing all development expense and income from exploitation as it arises (e.g., as noted in paragraph 109, Relevant profits), as the transaction is in fact a collaboration, such that an ex-post TPSM may be appropriate.

114. Profit split calculations are not simple, and there is no one approach to identifying and splitting, relevant profits.  For example, in a case where there are board and line management contributions to control of risk (like that in paragraph 51, Control of risk), it might be possible to isolate relevant profits (risk upside and downside) resulting from line management contributions, or it might be necessary to use an allocation method to approximate the contribution of line management versus the board.

Recognition of the accurately delineated transaction

115. Every effort should be made to determine pricing for the actual transaction as accurately delineated (paragraph 1.141) i.e., including the allocation of risk under Step 5 and reward for any contributions to risk control by entities not assuming / allocated any risk. 

116. However, the TPG set out the criteria whereby a transaction can be disregarded for transfer pricing purposes.  This is, per paragraph 1.142:

…where the arrangements made in relation to the transaction, viewed in their totality, differ from those which would have been adopted by independent enterprises behaving in a commercially rational manner in comparable circumstances, thereby preventing determination of a price that would be acceptable to both of the parties taking into account their respective perspectives and the options realistically available to each of them at the time of entering into the transaction.

117. Non-recognition should not be used just because determining an arm’s length price is difficult, nor where (per paragraph 1.143) the accurately delineated transaction isn’t seen between independent parties (i.e., there are no identifiable comparable uncontrolled transactions).  These factors are insufficient justification unless coupled with a lack of economic rationality.

118. It follows that the reward for parties contributing to the control of risk, per paragraph 1.105, will not always be based on comparison with transactions between independent parties (the conclusion reached by some commentators) i.e., you can’t simply disregard the transaction if comparable transactions don’t exist.

Examples

119. The following examples are intended to illustrate the general principles set out above in some specific contexts.

Intangible assets

120. Chapter VI sets out the overarching requirement that contributions to IP value should always be appropriately rewarded. It engages the Chapter I risk control guidance with explicit link drawn to the DEMPE functions. Paragraph 6.56 emphasises the importance of applying this functional analysis by reference to the commercial context of the MNE group as a whole and the manner in which it creates value, including activities such as marketing and quality control activities across the group. Similarly, at paragraphs 6.65 to 6.68 the analysis of risk relating to intangibles is placed squarely in the context of the MNE group and how constituent risks are controlled in practice.

121. This guidance directly reflects the requirement in the Chapter I risk control framework for specificity in the identification of economically significant risk. When allocating residual profit arising from the exploitation of IP it is necessary to consider each economically significant risk that is associated with the DEMPE functions, and the control exercised over each risk when allocating that risk and the profit earning potential that is attributable to those functions.

122. This approach stands in stark contrast to transfer pricing policies sometimes seen in MNE groups. For example, contractual arrangements within an MNE group may allocate the whole of the residual profit arising from exploitation of IP to the legal owner of IP, even if that entity has limited functionality. This result may be defended on the grounds that the IP owner does exercise some control over IP risk, perhaps monitoring infringement risk globally, and it may be asserted that this is in accordance with the risk control framework in Chapter I and the assumption of all IP risk to that entity must be followed for the purpose of pricing. These conclusions are not justified, relying as they do on a strained and partial reading of the guidance. In particular, it is very unlikely in most MNE groups that holding IP will give rise to a single indivisible risk. It is likely that there will be a large number of specific economically significant risks when the commercial exploitation of IP is analysed correctly in accordance with the Chapter VI rules. These rules guide the application of the Chapter I risk control framework and so require risk to be allocated according to a specific analysis of the MNE.

123. It will not be sufficient to observe that there is some control of elements of IP risk to justify contractual terms that allocate all risks that relate to IP to the legal owner, because it is almost certainly wrong to identify IP as a single risk in the context of the MNE group. Rather than arriving at such a sweeping conclusion, a functional analysis is required to identify what the specific economically significant risks are and how they are controlled. This analysis proceeds by reference to the steps the MNE group necessarily takes in order to earn profit.

124. Identification of risk controls in such cases may not be straightforward, because of the highly integrated nature of most MNE groups. As noted in paragraph 6.53 of the TPG, it is possible that control functions might be exercised by an entity that neither owns the IP nor carries out most of the DEMPE functions.

125. Where an MNE group has many lines of business (for example many different brands or types of business activity) it may be found that actual control of risk takes place at a fairly granular level. There may be many economically significant risks that are subject to specific controls and so it will be necessary to identify decision makers with a matching level of specificity. In such circumstances approximate methods may be called for. The appropriate basis for approximation requires careful consideration of the circumstances of the MNE group, with a view to matching control of risk decisions with the upside and downside that is associated with that risk. In some cases, it may be necessary to proceed by way of head count of a class of decision makers within the MNE group to inform a profit split methodology that reflects a complex pattern of control of a large number of specific economically significant risks. This approach may be the basis for splitting either part or all of the residual profit, depending on the other circumstance of the group.  But as noted above, care should be taken with regard application of the TPSM based on the contribution of a limited number of employees contributing to risk control.

126. A methodology such as this is not undertaken to measure the relative contributions to the control of each specific risk but is instead an approximation that reflects the many risk controls in place. It implicitly assumes that the effort placed in control of economically significant risks is proportional to the profit that derives from their successful control, which may be a reasonable assumption.

Captive insurance

127. In order for a captive insurance premium to be a proper deduction from profits of the insured entity, the arrangement must be accurately delineated as one of insurance; and it must be commercially rational for the insured party to seek insurance of the insured risk.

128. Chapter X of the guidelines equates the accurate delineation with the question of whether the premium is received in the course of a genuine insurance business carried on by the insurer. A number of indicators are given to assist with this consideration, most especially at 10.199. Additionally, at 10.202 the guidelines consider whether the insurer has financial capacity to bear the risk, especially where in practice the insurer invests premiums with the insured MNE.

129. If a risk is one that the MNE group necessarily bears in order to make profit, it may be irrational for it to insure the risk, for several reasons:

• Because the MNE group has invested significantly in the control of such a risk, there would be a substantial information asymmetry between itself and any arm’s length insurer. Consequently, unless the insurer is able to mitigate the impact of the risk substantially through diversification, it is likely to require a greater premium than the real value of the risk to the MNE group
• An insurance premium that rewards the insurer for bearing economically significant risk that is operationally controlled by the insured entity is likely to remove most of the profit that results from successful control of risk. An outcome may be irrational if substantial investment of capital and expertise in effective risk control does not provide commensurate expectation of reward.
• Unless control functions are reflected in reduced premium (and would be so at arm’s length) it may be irrational for the MNE to both invest substantial resource in risk control and then insure the risk, so depriving itself of profit from its risk control functions.
• If the captive insurer invests premiums in the insured party, then any insurance claim may be reflected in a withdrawal of capital from the insured party, which may make the payments of premium irrational.

Other financial instruments

130. Other financial instruments may have much the same effect as captive insurance – for example a total return swap (‘TRS) will similarly separate profit from underlying profit generating functions. This is achieved by creation of a new risk borne by the issuer of the derivative that mirrors underlying operational risk of the party holding the TRS. Hence, risk exists in both parties and so a question arises as to the allocation of profit associated with that risk.

131. It is unlikely that in a non-financial group a premium payable in respect of a TRS should be a proper deduction, for one or both of the following reasons:

• It is likely that the arrangements are not accurately delineated as a TRS because the TRS is not issued in the course of a genuine derivative trading business;
• It is irrational for the holder of the derivative to both control risk and alienate associated profits.