The BEPS project's new transfer-pricing notion of "cash box" appears to be inconsistent both with commercial reality and with longstanding Canadian outbound international tax policy.
In 1976, Canada adopted the policy of facilitating management (reduction) of the foreign tax obligations of Canada's multinational enterprises (MNEs) via group financing and licensing arrangements, carving those obligations out of the FAPI system. The primary mechanism of this carve-out was and remains subparagraph 95(2)(a)(ii), which converts passive inter-FA interest or royalty payments out of the payer FA's ABI into the recipient FA's ABI. This system survived an indirect challenge in 2007, when the anti-double-dipping rule in section 18.2 was enacted (but repealed in 2009, before its effective date), and it was strongly endorsed by the 2008 governmentappointed Advisory Panel on Canada's System of International Taxation.
The integrity and effectiveness of subparagraph 95(2)(a)(ii) may now be coming under another indirect attack, a result of the OECD's BEPS transfer-pricing theory for so-called cash boxes and the consequent taxation of income of a group financing or licensing company (or, more accurately, taxation of the sharing of the income among multinational group members). This article explains the BEPS cash-box theory, why it is invalid, and why it should therefore not apply in Canada.
The BEPS cash box disregards commercial reality. Actions 8-10 of the BEPS report, which make up the transfer-pricing approach, contain a cash-box notion: the stated purpose of that report is to align transfer-pricing outcomes with value creation while retaining and working within the confines of the arm'slength principle. The OECD defines cash boxes as "shell companies with few if any employees and little or no economic activity, which seek to take advantage of low or no-tax jurisdictions"; the cash box thus contemplates a group member that simply provides capital, such as funding or intangibles, for use by an operating company and that itself has limited activities. The OECD asserts that "[i]f the capital-rich member does not in fact control the financial risks associated with its funding, then it will be entitled to no more than a risk-free return, or less if, for example, the transaction is not commercially rational and therefore the guidance on non-recognition applies," and it similarly says, with respect to intangibles, that "[l]egal ownership of intangibles by an associated enterprise alone does not determine entitlement to returns from the exploitation of intangibles."
These OECD statements essentially suggest that economic returns in excess of a risk-free return are (or should be) attributable to labour and not to capital; furthermore, they suggest that excess economic returns are only allocable to an entity if its own employees perform the specific economic functions and it does not outsource functions to other group entities or external providers. The OECD statements are not supported by the arm's-length principle and are inconsistent with economic theory and practice.
According to the cash-box notion, the allocation of the risk from loans and licences between group members of MNEs— and hence the related profit—must be made on a factual basis and not a contractual basis. That process is said to have two actors: the party that has the "control over the risk" and the party that has the financial capacity to bear the risk of the loan. The first party's control appears to mean merely the decisionmaking authority to make a loan, but investors may routinely hire professionals to perform the task. Can the second party be other than the lender that puts up its money to make the loan? In any event, it is difficult to interpret those factors in such a way as to arrive at the OECD's conclusion that the lion's share of the profit from the intercompany loan or licence is allocable to group members as a whole and not to the group lender or licensor, which is allocated only "risk-free returns" (that is, three-month treasury bill rates). Thus, according to the OECD, if the lender or licensor does not employ persons to make and manage its investments but turns to group members to help out, it may not be entitled to the main portion of the income from the loan or licence.
Economic theory and practice arrive at a different conclusion. As a matter of practice, passive investors routinely engage financial advisers and asset managers, but they do not merely keep a risk-free return for themselves and pay the balance to the advisers. For example, the best-paid private equity fund managers receive no more than 20 percent of the income earned by the totally passive money-owning investors. The cash-box notion does not reflect that reality and is inconsistent with transfer-pricing policy, which is intended to mirror economic and business reality whenever possible.
In summary, the OECD's changes to its BEPS transfer-pricing guidelines seek to erode the arm's-length principle by promoting inconsistent exceptions, such as for cash boxes. The OECD's focus on labour appears to adopt a formulary apportionment that is based on salaries and wages. Therefore, it is not surprising that, last year, Justice Frank J. Pizzitelli of the TCC described the OECD's final package on BEPS as a step toward formulary profit attribution principles, and he said that the trend was likely to continue. In our view, the BEPS approach is also inconsistent with the Canadian policy reflected in subparagraph 95(2)(a)(ii).
In Canada's 2016 budget, Finance took the unusual step of saying that the CRA was then applying the revised OECD guidance on transfer pricing, which "provides an improved interpretation of the arm's-length principle." Significantly, however, Finance also said that the CRA would not adjust its administrative practices at that time in the two most controversial areas of the OECD's BEPS-related transfer-pricing work: the proposed simplified approach to low value-adding services and the treatment of cash boxes. Canada will decide on a course of action with regard to these measures after the OECD completes its followup work.
It should be reiterated that in the seminal transfer-pricing case of GlaxoSmithKline (2012 SCC 52), the SCC said that the OECD's transfer-pricing guidelines "are not controlling as if they were a Canadian statute and the test of any set of transactions or prices ultimately must be determined according to [Canada's transfer-pricing legislation] rather than any particular methodology or commentary set out in the Guidelines." The Canadian arm's-length principle is legislated in section 247 and thus any inconsistent OECD pronouncements—even those adopted as CRA administrative practice—are likely to fail in proceedings before a Canadian court.
The CRA and Finance should reject the notion of the cash box's relevance to Canadian tax law. However, comments made by Michelle Levac (transfer-pricing specialist at the CRA) at a transfer-pricing conference held in Toronto on August 30, 2016 appear to be inconclusive. Levac, who previously served as chair of the OECD's Working Party 6, said that the OECD transferpricing report recognizes the importance of funding "and the contributions of cash or capital." Levac went on to say, however, that there had been no "sea change" in the BEPS limitations attached to a cash box that funds intangible development but whose only economically relevant activities result in a risk-free return.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
