BEPS Actions 8–10

June 13, 2019 | By Stephen Blough, KPMG in the United States; Brian Cody, KPMG in the United States; Prita Subramanian, KPMG in the United States; and Agata Uceda, KPMG in the Netherlands


This article represents the views of the authors only, and does not necessarily represent the views or professional advice of KPMG LLP.

Current state of play


The guiding principle for BEPS Actions 8-10 was that transfer pricing outcomes should be aligned with value creation. Tax authorities were concerned that some companies and tax authorities were applying existing transfer pricing rules in ways that were inconsistent with this principle.

The OECD identified three actions related to ensuring that transfer pricing outcomes are in line with value creation: Action 8 (Intangibles), Action 9 (Risks and Capital), and Action 10 (Other High-Risk Transactions). The OECD’s output on BEPS Actions 8-10 led to amendments of the Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations (OECD Guidelines). These amendments cover the following chapters of the OECD Guidelines:

  • Chapter I – new guidance on accurately delineating a transaction (with an emphasis on the role of risk), location savings and other local market features, assembled workforce and group synergies
  • Chapter II – new guidance on commodity transactions and revisions to the guidance on the transactional profit split method
  • Chapter VI – new guidance on intangibles
  • Chapter VII – revised guidance on services, with a new section on low value-adding intragroup services
  • Chapter VIII – revised guidance on cost contribution arrangements
  • Chapter IX – conforming amendments to guidance on business restructurings.

As evidenced by the list above, the OECD made extensive revisions to the OECD Guidelines, covering a wide range of transfer pricing issues. Of these, the most fundamental revisions relate to the guidance on risk in Chapter I and on intangibles in Chapter VI.


Risks and intangibles

Leading up to BEPS Actions 8-10, tax authorities had expressed concern that MNEs were artificially shifting income to low-tax jurisdictions by transferring their valuable intangibles and/or economically significant risks to group companies in those jurisdictions. The MNE often did not have much operational activity in the low-tax jurisdiction but earned significant profits there by virtue of its contractual assumption of risk or intangible property (IP) ownership. One approach the OECD took to addressing this particular concern was to include specific functional requirements for an entity to reap the economic benefits of intangible ownership or assumption of risk. According to the OECD, these requirements are consistent with arm’s-length behavior.

The OECD positioned the revised guidance on risk and intangibles as a strengthening and clarification of the arm’s-length principle. At the outset of the BEPS project, the OECD had noted that “special measures, either within or beyond the arm’s length principle, may be required with respect to IP, risk and over-capitalisation.” Through its work on Actions 8-10, the OECD concluded that special measures “beyond the arm’s length principle” were not required to meet the objectives of Actions 8-10 but that clarifying and strengthening the guidance on the arm’s-length principle were sufficient. The OECD’s assumption that the functional requirements included in the revised guidance on risk and intangibles mirror arm’s-length behavior was important to its conclusion that the arm’s-length principle remains the absolute standard for transfer pricing.


Control over risk and DEMPE

The starting premise of the OECD’s approach to assumption of risk and intangible ownership is that all members of an MNE, at arm’s length, need to receive appropriate compensation for the functions they perform, assets they use, and risks they assume. The approach presumes that important functions related to the control of risk and the development, enhancement, maintenance, protection, and exploitation (DEMPE) of an intangible contribute to the returns for the assumption of risk and the creation of the intangible’s value, respectively. The OECD’s approach, therefore, concludes that contractual assumption of risk alone or legal ownership of an intangible alone does not generate a right to the returns from risk assumption or the intangible. The legal owner of an intangible must provide a share of the returns from the intangible to the provider of DEMPE functions commensurate with the importance of those functions. A legal entity that assumes an economically significant risk must also control the risk.

For the OECD, control over risk involves the following two elements of risk management: (i) 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, and (ii) the 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. While the guidance also mentions financial capacity as a requisite for assuming risk, the focus of the OECD Guidelines is quite clearly on the control over risk.

Further, the OECD Guidelines note that risk management and DEMPE should not be thought of as necessarily encompassing different functions. A reasonable interpretation of the guidance in Chapter I on risks and Chapter VI on intangibles is that the two chapters prescribe the same principles for delineating the actual transaction between associated enterprises. A key conclusion of the two chapters is that only an entity performing key decision-making functions related to the key income generators for the MNE—be it risk assumption or the development of intangibles—could earn excess or residual income related to those key income generators.

In this way, the OECD’s guidance on control over risk and DEMPE is intended to prevent companies with no significant employees or minimal operational activity from earning significant risk-related or intangible-related returns.


Key implications of BEPS Actions 8-10

Potentially differing interpretations of the new guidance

According to the revised OECD Guidelines, contractual arrangements or funding alone do not entitle an entity to returns from intangibles or risk assumption. In order to earn returns from assuming risk or owning intangibles, an entity must have “substance” in the form of decision makers controlling the risks or performing important DEMPE functions. Cash boxes, i.e., entities funding intangible development or contractually assuming risks but with no significant people functions that were a major concern of tax authorities at the outset of the BEPS project, would not be entitled to the returns from economically significant risks and intangibles.

While the OECD Guidelines make it clear that legal ownership or contractual terms alone do not entitle an entity to returns, the guidance is less clear on degree, i.e., how much substance is required for an entity to earn the returns from risk assumption and intangible ownership. The new guidance is open to differing interpretation by tax authorities, leaving taxpayers with the challenge of determining the appropriate level of substance to satisfy the functional requirements for earning the rewards of risk assumption and intangible ownership—for example, how closely involved the decision-makers need to be in the making of decisions, how senior the decision makers need to be, how many decision makers, how often they need to meet, etc.

The role of contracts in determining returns is a related issue open to interpretation. While according to the OECD, contracts alone cannot determine who is entitled to returns from risks and intangibles, contracts are the starting point for determining who assumes risk and owns intangibles. Where multiple entities perform decision-making functions contractual assignments of rights to one or more of those decision-making entities will be respected under the OECD Guidelines. However, the language of the OECD guidance may be interpreted in different ways by tax authorities to support differing opinions on the role of contracts. For example, paragraph 1.94 notes that there may be more than one party to the transaction exercising control over a specific risk. Where the associated enterprise contractually assuming risk controls that risk and has the financial capacity to assume it “the fact that other associated enterprises also exercise control over the same risk does not affect the assumption of that risk by the first-mentioned enterprise….” Paragraph 1.105 goes on to say that “[i]n circumstances where a party contributes to the 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 that contribution to control, may be appropriate.” Some tax authorities may read paragraph 1.105 as implying that control over risk functions should be entitled to sharing in the returns from risk irrespective of contractual compensation terms in contradiction of the guidance in paragraph 1.94.

Notwithstanding the uncertainty introduced by the language in the OECD Guidelines, the Guidelines do not preclude the existence of contract risk control services or DEMPE services, e.g., contract R&D. Contract R&D services are still permissible. However, the important question will be whether and to what extent those contract R&D services entitle the service provider to returns from the intangible.“ [I]f the contractual arrangement between the associated enterprises is a contract R&D arrangement that is respected…remuneration for risk mitigation functions performed through the development activity would be incorporated into the arm’s-length services payment. Neither the intangible risk itself, nor the residual income associated with such risk, would be allocated to the service provider.”1 Thus, when R&D strategy and direction for an intangible is determined in one entity and contract R&D is performed in another, the OECD guidance would seem to say that the DEMPE (including risk-control) functions and therefore the intangible-related returns belong in the former entity.

An underlying issue is that the language in the OECD Guidelines is consensus language, i.e., no participant to the BEPS project objected to the revised guidelines. Therefore, tax authorities with potentially different points of view needed to agree on the language in the revised guidelines. The end result is that at least some of the language in the OECD Guidelines is amenable to differing interpretations, which is likely to be a source of increased controversy.


Continuing role of arm’s-length standard

A broader question raised by BEPS Actions 8-10 and subsequent discussions is the role of the arm’s-length standard in the system of international taxation. BEPS Actions 8-10 raised the spectre of a departure from the arm’s-length principle, which has been the foundation of transfer pricing rules for decades. Ultimately, the BEPS Actions 8-10 deliverables endorsed the arm’s-length standard, noting that there was no need for measures beyond the arm’s-length standard to realign transfer pricing outcomes with value creation. However, commentators have questioned whether prescribing functional requirements for intangible ownership and assumption of risk mark a weakening of the arm’s-length standard—in substance, if not in name. General economic theory indicates that economic activity requires both funding (or capital) and people who use that funding to create economic output. In particular for risky economic undertakings, such as the development of intangibles, funding is an important contributor to economic output. The OECD’s presumption that funding of the intangible in itself has little or no claim on the returns from the intangible if that funding is not provided by an entity having people controlling that funding and making important decisions related to the economic activity being funded prescribes arm’s-length behavior without empirically or theoretically considering other views of funding. In fact, examples where the provider of funding assumes substantial risk with minimal control can be found in uncontrolled market transactions, e.g., investors investing in hedge funds.

Another example of the potential for gradual erosion of the primacy of the arm’s-length standard is the BEPS Actions 8-10 guidance on the transactional profit split method (TPSM) and the practical implementation of that guidance by tax authorities. A key concern of taxpayers has been that the singling out of the TPSM for additional guidance could lead to the misuse or overuse of the TPSM. This overuse could lead to a move away from the arm’s-length standard and, in the extreme case of systematic, prescriptive use of the TPSM, could constitute formulary apportionment of profits.


Comparison to approach to taxation of the digital economy

Subsequent to the release of the final reports under the BEPS Action Plan, the OECD and several countries around the world have increased focus on their evaluation of the digital economy. While the BEPS Actions 8-10 work reaffirmed the arm’s-length standard and described the functional substance (in terms of people) required for earning a share of the MNE’s profits, paradoxically the new debate on the digital economy appears to be trying to attribute profits to jurisdictions without any, or only limited, functions (DEMPE or otherwise). With this new debate on the taxation of the digital economy many commentators believe that some countries are in effect backtracking on their commitment to the OECD’s earlier conclusion in BEPS Actions 8-10 that the arm’s-length principle would continue as the unchallenged standard for transfer pricing in all cases. In fact, some countries may be backing away from the guiding principle that transfer pricing outcomes should be aligned with value creation. It remains to be seen what the outcome of the digital economy work will mean for the BEPS Actions 8-10 conclusion on the primacy of the arm’s-length standard and the alignment of transfer pricing outcomes with value creation.


Regulatory and tax authority response to BEPS Actions 8-10

Like other OECD publications and recommendations, the OECD Guidelines do not have independent legal effect. Rather, the OECD Guidelines have effect through incorporation into bilateral tax treaties and jurisdictions’ domestic transfer pricing guidance. In addition, the OECD Guidelines can be influential as an interpretive guide to the arm's-length principle as set forth in the OECD’s Model Tax Convention on Income and on Capital.

Different countries have taken different approaches to the revisions to the OECD Guidelines. Some countries treat the OECD Guidelines as guidance on the implementation of the arm's-length standard not requiring a change in law—some of these countries treat the revised guidance as having retroactive effect while others treat them as being effective as of the date the BEPS Actions 8-10 final report. Some countries have adopted the revisions to the OECD Guidelines by legislation or rule making. Some countries whose tax rules have no formal connection to the OECD Guidelines have made changes to transfer pricing rules and administrative practice inspired by the OECD initiative. Yet other countries have not changed laws or regulations in response to the OECD Guidelines but consider existing rules to be consistent with the revised OECD Guidelines.

Regardless of the form of adoption of the OECD Guidelines, as a consequence of BEPS Actions 8-10, many tax authorities are more closely scrutinizing transfer pricing arrangements, focusing in particular on value chains and people functions in their jurisdictions. Similar to the variation in approaches to adoption of the revisions to the OECD Guidelines, tax authorities differ in their levels of scrutiny and interpretation of the revised guidance in the OECD Guidelines, resulting in greater transfer pricing controversy. As an example, compared to 2016, new transfer pricing cases were up by 25 percent and other cases by 50 percent in 2017 according to the Mutual Agreement Procedure Statistics published by the OECD on its website.2 In general, tax authorities around the world are working on increasing their transfer pricing capabilities. As more tax authorities develop transfer pricing capacity and adopt BEPS Actions 8-10, transfer pricing controversy is expected to rise.


Taxpayer response

Taxpayers have responded in various ways to the BEPS Actions 8-10 guidance, and some broad patterns in taxpayer responses to BEPS Actions 8-10 have emerged.

As discussed above, the most fundamental revisions to the OECD Guidelines relate to the guidance on risk and intangibles. The recommendations related to control over risk and DEMPE functions have been a significant catalyst for taxpayer response. Taxpayers are increasingly taking stock of their end-to-end value chains, identifying significant intangibles and economically significant risks, identifying key DEMPE and control functions, and determining where those functions are performed. Based on their detailed understanding of their value chains and international tax structures, taxpayers are identifying tax risks given the BEPS Actions 8-10 guidance and determining appropriate changes to their tax structures.

One trend that has emerged is that several of the MNEs that did hold significant intangibles in “cash boxes” have moved or are moving them to other entities with greater DEMPE and risk control functions. MNEs are also evaluating the appropriate level of substance in their principal companies or intangible-holding companies. Some MNEs are working on bolstering the DEMPE and risk control functions in these entities while some others are migrating their intangibles to other entities that do have sufficient decision-making functions.

The question of what level of decision-making functions is “sufficient” under the BEPS Actions 8-10 for risk assumption or intangible ownership has been a key one for many companies. Advisers have tried to resolve the lack of clarity in the OECD Guidelines using myriad tools: DEMPE checklists, RACI models3, value chain analysis, etc. Where not all DEMPE and risk control functions are located in the principal or IP-holding company, companies are working on clarifying contractual arrangements and ensuring that the principal companies do have sufficient decision-making authority related to important DEMPE and risk-related activities. In this context, MNEs are considering committees that make important decisions related to the key intangibles and risks (based on the facts of the particular industry and business of the MNE) in which employees of the principal company participate. As an added layer of decision making in the principal company, a board of directors with substantive decision-making authority in the principal company (a “smart board”) may be considered by MNEs. The smart board does not just act as a rubber-stamping authority but participates in decision making at a frequency appropriate to the particular business of the MNE.

Notwithstanding the amount of thought and effort companies are putting into designing their structures to be compliant with the revised OECD Guidelines, they are also actively evaluating risks related to their DEMPE and risk control functions since different tax authorities appear to have different interpretations of the concept. Further, companies are developing processes to continually monitor such risks. More generally, MNEs expect tax audits, litigation, and related controversy to surge as nations adopt BEPS Actions 8-10 and also other BEPS final actions, including country-by-country reporting. MNEs are preparing for the increased controversy —through added tax personnel, proactive engagement with tax authorities such as through advanced pricing agreements (APAs,) centralized and improved documentation, etc.


Views from around the world

United States

In public pronouncements, U.S. government officials have said that section 1.482 of the U.S. Treasury regulations4 are consistent with the revised OECD Guidelines. However, as noted above, the ambiguous language in the OECD Guidelines can mask differences in interpretation between the United States and other tax authorities.

For example, on the question of respect for contractual arrangements, the U.S. transfer pricing regulations respect contractual allocations of risk as long as they are consistent with the “economic substance” of the transaction.5 In considering the economic substance of the transaction, the following facts are relevant:

  • Whether the taxpayer's conduct over time is consistent with the purported allocation of risk
  • Whether the taxpayer has the financial capacity to fund losses
  • The extent to which each controlled taxpayer exercises managerial or operational control over the business activities that directly influence the amount of income or loss realized. In arm's-length dealings, parties ordinarily bear a greater share of those risks over which they have relatively more control.6

The U.S. rules on risk allocation, therefore, look similar to the OECD guidance in principle. However, unlike the OECD Guidelines, section 1.482 does not prescribe the meaning of “managerial or operational control.” In practice, the IRS and U.S. Tax Court have generally respected contractual allocations of risk by the taxpayer.

In the case of intangible development, section 1.482-7 rules on cost sharing arrangements permit intangible-related returns to cost sharing participants without regard to operational control over DEMPE functions. While the rules on economic substance in section 1.482-1 described above still apply to cost sharing arrangements, in the past the IRS has not in practice insisted on economic substance in a cost sharing participant in the form of managerial or operational control equivalent to the DEMPE functions or risk control functions of the OECD Guidelines in order for it to be entitled to returns from the intangible under development.

Notwithstanding past practice, there has been increasing awareness and interest within the IRS on the DEMPE and risk control concepts since the issuance of the BEPS Actions 8-10 guidance. The IRS is increasingly asking questions on the level of DEMPE and risk control functions in IP holding and principal entities. DEMPE and risk control have also become important discussion points between the IRS and other tax authorities in bilateral APAs and MAP discussions.


France has not issued specific guidance on DEMPE or risk control concepts so far. However, these concepts are now frequently used by the French Tax Authorities (FTA) in the context of tax audits, as well as in APAs. They notably put emphasis on the increase in value of intangibles resulting from their exploitation to challenge the remuneration of French distribution entities using intangibles (brands, patents) owned by foreign-related entities.

For instance, the FTA may consider a French distribution entity paying a royalty fee for the use of the brand to the group IP holder and incurring marketing or advertising expenses as making a significant contribution to the development of the brand in the French market. As a consequence, the FTA may conclude that the French entity should have received a higher remuneration, such as a higher operating margin or a share of the profits.

It is difficult to anticipate at this stage how the French Court would rule on the DEMPE concept, but many believe that the Court would probably follow the DEMPE analysis provided that the presentation of the case is supported with appropriate legal references and documentation.


Germany has not published anything with respect to the DEMPE and risk control concepts in BEPS Actions 8-10. New guidance on the application of the arm’s-length principle is expected in 2019, but it is unclear if and how the DEMPE and risk control concepts will be addressed.

Irrespective of the formal guidance, German tax auditors have already been using the DEMPE concept in German tax audits since even before 2015. For instance, the tax auditors may try to argue that some high value activities or even IP ownership should be attributed to the German entity based on the presence of one or more R&D or certain other types of personnel.

The highest tax court in Germany has ruled in some transfer pricing cases that Article 9 of the OECD Model Tax Convention overrules local legislation. As the DEMPE concept is part of the OECD Guidelines, the DEMPE concept is, thus, relevant. However, some uncertainty exists as to which date the DEMPE guidance is effective since the revised OECD Guidelines say that the changes brought about by BEPS Actions 8-10 are just a clarification of principles in older OECD Guidelines.

The Netherlands

The Netherlands has a new transfer pricing decree that was published May 11, 2018 (“new decree”). This new decree incorporates a selected number of topics from the revised OECD Guidelines (2017). In particular, the language from Chapter 1 of the OECD Guidelines on control over risk (Par. 1.60) has been copied almost verbatim.

The new decree includes a relatively elaborate section about transactions involving intangibles. This section specifically discusses options realistically available when transferring intangibles; transactions with low-tax jurisdictions; hard to value intangibles; the specific case of the acquisition or sale of shares in an independent company followed by a business restructuring; valuation methods and assumptions; and the use of databases for benchmarking intangibles.

The new decree includes a paragraph on DEMPE as well. Somehow surprisingly, the new decree specifies that Development and Enhancement “will be given more weight”—without giving specific examples about industry value drivers. The relevant paragraph reads as follows:

The OECD Guidelines often refer to the DEMPE functions in describing relevant functions regarding intangible fixed assets. These functions relate to Development, Enhancement, Maintenance, Protection and Exploitation. Depending on the facts and circumstances, an assessment must be carried out of the relative importance of the various DEMPE functions. In general, the Development and Enhancement functions will be given more weight in the assessment of the relative contribution to the value of the intangible asset in question.

In practice, the Dutch Tax Authorities have already been using the concept for many years, even before the DEMPE concept was introduced by the OECD. A discussion about DEMPE is now a key feature of many tax audits and APAs. There is, however, no case law yet that refers to DEMPE, neither specifically nor broadly.

United Kingdom

HM Revenue and Customs (HMRC) occasionally uses the DEMPE term in inquiries and correspondence, but it is not official government language and the HMRC has not issued anything specifically calling it out. Nevertheless, HMRC applies the DEMPE concept routinely and has done for many years. For example, it is not uncommon to receive questions about senior decision making, control of risk, and important decisions related to the development of the IP in tax audits.

The DEMPE or risk control concepts have not been tested in court, but given the common law nature of the English legal system and the emphasis on substance over form in language as well as legal concepts, any terminology used in litigation would potentially stand so long as its meaning was commonly understood. A court may well pick apart the pieces of the acronym to understand what they really mean in practice. Transfer pricing rarely goes to court in the United Kingdom, so there is no recent precedent to compare with.


Sweden has not amended its transfer pricing regulations following BEPS Actions 8-10. Having said that, Swedish legislation is very brief in relation to the arm’s-length principle, and instead, Swedish Tax Agency (STA) guidelines and case law point at OECD Guidelines as a sound and appropriate guideline to follow. The STA has also said in public meetings that the updated OECD Guidelines only confirm how the arm’s-length principle should always have been interpreted, and that they therefore can be applied retroactively.

The STA has used the updated Guidelines in several cases, and they often refer to DEMPE functions. For example, DEMPE functions were significant in a case involving a gaming company that did not have enough substance in the Maltese parent company.

It is likely that the DEMPE concept would also stand in court. However, it is very difficult to know how the court will define DEMPE functions. Likely it would depend on the overall picture and what functions are found in the other company. In many other cases the daily management has been important (Head of R&D, Head of IT, etc.). Board members and board decisions at site have not been sufficient.


Poland has not yet issued any specific guidance or decree referring to DEMPE specifically. The Polish tax authorities have not yet started applying the concept of DEMPE in audits or APAs. However, there is a growing awareness of the concept, which has led to more in-depth scrutiny of the legitimacy (and tax deductibility) of royalty payments. While not based on the DEMPE and risk control concepts, since 2018, royalties are subject to tax deductibility general restrictions (also intangible services), where any amounts in excess of a limit are not tax deductible. The limit is 3 million PLN plus 5 percent of the tax EBITDA. 

[1] Paragraph 1.62 of the OECD Guidelines.


[3] RACI is an acronym for responsible, accountable, consulted, informed. RACI models are used for identifying roles and responsibilities within an organization.

[4] Unless otherwise indicated, section references are to the Internal Revenue Code of 1986, as amended (the “Code”) or the applicable regulations promulgated pursuant to the Code (the “regulations”).

[5] Section 1.482-1(d)(3)(ii)(B).

[6] Section 1.482-1(d)(3)(iii)(B).