The OECD recently released their draft paper titled Transfer Pricing Comparability Data and Developing Countries, which asks several important questions. As I read this document, my two questions were:
- Is the OECD’s apparent call for having the choice of transfer pricing methodology being dependent on the nature of available data putting the cart before the horse?
- Are its pleas for better data simply for the benefit of national tax authorities or could the same pleas assist multinationals in documenting whether their intercompany pricing policies are arm’s length?
Permit me to expand on my first question by posing three different fact patterns. If an affiliate located in a developing nation is either providing contract manufacturing activities for its parent corporation or if this affiliate is a mere distributor of the multinational’s products, then the Transactional Net Margin Method (TNMM) with the developing nation’s affiliate as the tested party is a natural way of evaluating the transfer pricing issue. It is this fact pattern that this OECD document envisions.
Paragraph 22, however, states:
If comparables are not available to evaluate a transaction from a domestic perspective, the question arises as to whether it may be appropriate and possible to evaluate the transaction by testing the return earned by the foreign counterparty. Testing the foreign counterparty may also mitigate the risk that the domestic party will only be allocated a routine return without considering whether it should share in any residual return that may arise in the transaction.
While reversing the selection of the tested party would likely be inappropriate if the domestic affiliate is a contract manufacturer or a mere distributor, my upcoming Journal of International Taxation paper on oil and mining transfer pricing suggests the foreign counterparty logically should be the tested party in the following situation. Consider an African mining affiliate of a multinational that utilizes a Swiss affiliate to sell its goods to European customers. If the mining affiliate takes the risks of variations in market prices, then this affiliate should be entitled to the economic rents from the commodity boom over the past decade. In this situation, the Swiss distribution affiliate would logically be the tested party if the analysis relied upon TNMM. This view is the position of the Canadian Revenue Agency in the upcoming Cameco litigation even though the facts suggests that the distribution affiliate may have entered into an off-take agreement with the mining affiliate.
The OECD document also suggests the use of other analyzes such as profit split approaches. Whether or not one should rely on profit split approaches over TNMM depends on whether the domestic affiliate owns a portion of the valuable intangibles and not the ability to locate domestic third party comparables. If the domestic affiliate does own a portion of the valuable intangibles, the facts would suggest the use of the Residual Profit Split Method which in part requires the estimation of the routine return for the domestic affiliate. As such, the analysis would still require comparable data for an appropriate evaluation of the routine returns in addition to the need to evaluate the relative contribution of intangible assets from each of the related entities involved in the intercompany transaction.
Paragraphs 9 and 10 note the lack of publicly traded companies in Africa, East Europe, and Latin America that may serve of potential comparables for TNMM analyzes noted in our first fact pattern. Transfer pricing practitioners evaluating the profitability of contract manufacturing affiliates and distribution affiliates in North America, East Asia, and Western Europe often turn to the breath and depth of financial information for publicly traded companies in those regions. If the tested party is located in Africa, East Europe, and Latin America, an alternative approach must be taken whether that approach be to use publicly traded companies in other regions or to turn to commercial databases of privately held companies. While the OECD notes that the national tax authorities in Africa, East Europe, and Latin America would benefit from being able to source financials if private companies in their regions, the representatives of multinationals would also benefit from such information when they are preparing their transfer pricing documentation reports.
The OECD also appears to be appealing to the vendors of such commercial databases of privately held companies, which include Bureau van Dijk and our firm. The OECD document notes several concerns with respect to the appropriate use of these commercial databases including the skill of the analyst for the national tax authority as well as the limited amount of financial data for any identified third party private company selected as a potential comparable. While a vast number of private companies exist and report at least limited financial information, a key question as to the quality of such financial information exists. Thomson Reuters welcomes the opportunity to discuss these matters with the tax authorities of developing nations.
Permit me to close these discussion by noting the experience of the New Zealand tax authorities noted in paragraphs 20 and 21. The New Zealand tax authorities – like experienced tax authorities in other nations – consider adjustments for comparability differences between the tested party and the third party companies selected as potential comparables. In our September 5, 2013 webcast entitled “Key Comparability Concerns for Third Party Financials”, we addressed similar questions with respect to both functional and regional comparability.