The OECD has released a Discussion Draft on Cost Contribution Arrangements (CCAs) as part of its ongoing activities related to eliminating Base Erosion and Profit Shifting (BEPS) by multinational enterprises.

This Discussion Draft, released at the end of April, seeks public comments by May 29, 2015 on its suggested changes to Chapter VIII of the OECD Transfer Pricing Guidelines relating to inter-company cross-payment arrangements for joint development or service activities.

US tax law has very detailed, prescriptive rules for US taxpayers engaging in research and development Cost Sharing Arrangements (CSAs), under US Treas. Reg. Sec. 1.482-7. In many cases, the Discussion Draft is consistent with the concepts found in the US rules. However, there are some notable exceptions, and it remains to be seen whether the US Treasury and Internal Revenue Service will propose any amendments to the US rules to conform to the ultimate OECD Guidelines, if and when they are formally adopted.

This note takes a brief look at several key features of CCAs in the Discussion Draft and their corresponding features in the US tax law and offers some observations.

Definition and types of CCAs

The Discussion Draft defines a CCA as a contractual arrangement among business enterprises to share contributions and risks involved in the joint development of tangible and intangible assets, or services, for the benefit of the participants. A CCA is not a separate legal entity per se, and it does not give rise to a fixed place of business of all the participants. The Discussion Draft indicates that there are two common types of CCAs: those established for the joint development of assets, and those for obtaining group services. The Discussion Draft notes several instances where the two types of CCAs might have slightly different rules, but generally treats both types the same.

The US regulations also can be viewed as endorsing the concept that there are two types of such contractual arrangements:  those relating to the development of intangible assets (Reg. Sec. 1.482-7, CSAs) and those relating to the provision of inter-company services (Reg. Sec. 1.482-9, Shared Services Arrangements, or SSAs). Like the Discussion Draft, under US rules, it is clear that a CSA is a contractual arrangement, and its existence does not by itself create a separate legal entity.

In practice, many US taxpayers use CSAs in connection with sophisticated international tax structures. These taxpayers generally adhere to the US requirements quite strictly since the consequences of having a purported CSA disregarded by the IRS can be enormous, in that the purported sharing of development costs may be replaced by an inter-company license of intellectual property, with a royalty much higher than the sharing of development costs. On the other hand, it seems few US taxpayers adopt formal CCAs for services, since the arrangement does not provide a materially different cross-charge that would exist absent that arrangement, assuming the legal terms between the parties are clear.

Determining participants

Each participant in a CCA must be assigned an interest in the intangibles, tangible assets or services that are the subject of the CCA. There is no mention in the Discussion Draft of whether the interest must be non-overlapping, exclusive, and perpetual, which is unlike the US rules for CSAs. In this regard, the Discussion Draft’s approach may be more flexible.

A more significant requirement noted in the Discussion Draft and further emphasized in Example 5 in the Appendix is that a CCA participant must have the capability and authority to make certain decisions regarding risk. The Discuss Draft flatly states that if an entity does not have such capabilities, it cannot be a CCA participant; however, it is not clear exactly how such a purported arrangement would then be re-characterized.

Under US regulations, there is no requirement of decision making authority for each CSA participant. In fact, some CSA participants are related party entities in low-tax jurisdictions which have few, if any, employees or daily activities, exploiting their rights instead by virtue of royalty-bearing licenses with affiliates. Here, the US rules are clearly more flexible. This difference is hardly inadvertent, however, since the primary target of the OECD BEPS initiatives are low-tax jurisdiction entities with little actual activities and significant profits.

Does cost mean cost – what is the arm’s length charge?

In applying the arm’s length principle, Section C of the Discussion Draft makes a leap beyond the title of the Discussion Draft itself:  except for low-value activities, charges between affiliates for a CCA must be made at an arm’s length value, not based upon costs. The charges must take into account the proportion share of anticipated benefits each participant is expected to receive. Each participant will then fund its relative share of the “contributions.”  Contributions is an expansive term:  it can take the form of provision of services, performance of development activities, or provision of pre-existing values. Contributions also include the ongoing balancing payments made between the parties. The Discussion Draft provides four examples of how the balancing payments are computed.

It is not entirely clear from the Discussion Draft how an ongoing developmental activity is to be valued. The examples in the Appendix give the appearance that value might be computed to a “costs plus” analysis, although the text of the Discussion Draft never states that. The examples compare one activity as being valued at 105 (compared to its cost of 100), and another activity with a value of 120 (compared to its cost of 100). The examples then describe the computation of the cross-payments, based on values. Since the cross-payments are computed on an annual basis, it appears that the values of the development activities are not dependent upon the future success or failure of the developmental efforts. However, the provisions on Balancing Payments are not clear on this point.

The US rules for sharing development costs under a CSA have never required a valuation of the intangible development activities provided. Indeed, the primary benefit of entering into a CSA is avoiding the challenge of ascertaining the value of the ongoing development activities, which may be currently quite speculative but potentially quite valuable in the future. Unlike the Discussion Draft, the US Administration has always endorsed the concept that a sharing of current development costs is arm’s length behavior, even if only one participant is providing all the development activity. And, while the IRS might make an adjustment to the relative shares of costs which have been shared in the past, the regulations are now clear that US taxpayers should not make any self-help efforts to “true-up” prior year shares based upon actual benefit experience.

In this area, the concepts between the two valuation regimes are fundamentally different. However, if a widely accepted “cost plus” valuation methodology is applied by the OECD, the differences might not be impossible to reconcile.

Buy-ins and buy-outs

The Discussion Draft makes it clear that changes in the membership of a CCA will generally trigger a reassessment of the proportionate shares, so that previous participants effectively transferring part of their respective interests must be compensated by the new entrants on an arm’s length value for such a buy-in payment. Similarly, a participant that leaves should be compensated with a buy-out payment from the remaining participants.

The US rules are similar. Any pre-existing assets or capabilities made available to a CSA participant, once called a buy-in payment but now known as a platform contribution transaction, must be compensated at its value, not cost. The determination of this value has been highly controversial over the last two decades between taxpayers and the IRS. The Treasury and IRS in 2009 adopted guidance in Reg. Sec. 1.482-7T (now Final), prescribing the transfer pricing methods and economic frameworks for determining PCTs. In particular, the regulations generally focus on the newly designed Income Method, and generally prohibit the prior residual profit split method, which was favored by taxpayers.

Perhaps unimpressed by the US experience, the OECD does not see a need to provide any new computational methods for determining buy-ins or buy-outs, nor need to rule out any existing transfer pricing methods.  Conversely, since the Discussion Draft adds no new economic thoughts to an area subject to much US tax history, it seems the US is unlikely to change its valuation methods, either. This could be an area where the US and its OECD partners with cross-border transactions will inevitably create competing assessments and disputes.

Structuring and documenting CCAs and CSAs

Section E of the Discussion Draft provides a list of conditions relevant to CCAs, including the qualifications of entities that may participate, the nature and extent of the arrangement, valuations, balancing payments, and buy-ins and buy-outs. It also provides a list of relevant information that the taxpayers should maintain, including a list of participants, scope of activities, duration of the arrangement, how expected benefits are measured, etc. Finally, Section E provides a list of documentation which is expected to be retained by the taxpayer for audit purposes.

The US rules for CSA terms, agreement features, and documentation are similar. In addition, the US rules provide for certain time frames for execution of documents, and submitting required notices to the IRS. In this area, the risk for not following the specified rules appears more harsh than the Discussion Draft’s suggestions.


While many provisions of CCAs and CSAs are similar, there are pronounced differences, particularly in relation to valuations of ongoing development activities and buy-ins, and in regard to the requirement that a CCA participant must have the capability to manage certain risks. Moreover, various items in the Discussion Draft are not precisely defined.

It is possible that the US Administration and US taxpayers will simply ignore this Discussion Draft, with the belief that the US rules, like other areas of transfer pricing, will not be changed by these OECD suggestions on eliminating BEPS.

Nonetheless, if the current Discussion Draft is unchanged in the final OECD Transfer Pricing Guidance, there will be significant mismatches between US practices and those of other OECD members, likely leading to contradictory assessments and disputes for taxpayers subject to both regimes.