General Electric Capital Canada Inc. watershed transfer pricing case: THE 'GE CASE'
The "GE case" has far reaching implications for multinational companies operating in Canada. The decision in this case challenges the Organisation for Economic Co-operation and Development's (OECD's) interpretation of the arm's-length principle. The decision will impact many aspects of the transfer pricing practice in Canada — well beyond the narrow subject matter of financial guarantees between related entities. In this Tax Alert, we provide an overview of the GE case, as well as the implications and insights of this watershed case.Summary of facts and court decision
During the years under appeal, General Electric Capital Canada Inc. (the taxpayer) was a financial services company that carried on a number of businesses in Canada, including equipment, vehicle and real estate financing and technology management services. It was a wholly-owned, indirect subsidiary of General Electric Capital Corporation (GECUS), a US corporation. The taxpayer financed its operations by borrowing funds from capital markets through the issue of debt in the form of commercial paper and unsecured debentures. GECUS began guaranteeing the taxpayer's debt in 1988, but only started charging guarantee fees calculated at a rate of 1% per annum of the principal amount of debt outstanding, in 1995.The Tax Court decision
The Court allowed the taxpayer's appeal and vacated all of the Minister's assessments. The Court rejected the taxpayer's argument that the taxpayer's credit rating had to be determined on a stand-alone basis. In reaching this conclusion, the Court considered relevant authorities on the meaning of arm's length and concluded that the concept refers to "how independent parties negotiating with each other in the marketplace would behave for the purpose of achieving the best terms" in respect of the purchase or sale of goods and services (para. 196). In this context, the "implicit support" of the parent company could not be ignored.
In determining the arm's-length price, the Court considered significant expert advice. These experts articulated that there were three alternative models that could be utilized to determine the arm's-length guarantee fee that would apply to the tested transactions, namely (i) the yield approach, (ii) the insurance approach and (iii) the credit default swap approach.
The Court accepted that the "yield" approach put forward by the Crown to assess the guarantee's value was the most appropriate; however, it determined that the proper application of this approach, in consideration of the facts and circumstances of this case, favoured the taxpayer.
The Court concluded that the taxpayer's final credit rating without explicit support would be in the range of BBB-/BB+, significantly below AAA (the rating achieved with the guarantee in place), and one to two notches higher than the stand-alone credit rating (BB) excluding implicit support from GECUS. The interest cost savings to the taxpayer based on the differential between BBB-/BB+ and AAA was determined to be approximately 1.83% (based on one of the expert reports). Therefore, the 1% guarantee fee was found to be below an arm's length price in the circumstances.
Five major implications for transfer pricing:
1. Changes OECD definition of "arm's length"
2. Endorsement of yield approach for financial guarantee fees
3. Endorsement of "halo" effect for financial guarantee fees
4. Facts and circumstances paramount
5. Limitation of shareholder functionsChanges OECD definition of "arm's length"
The GE case decision challenges the OECD's definition of the "arm's-length principle" as treating related parties as "operating as separate entities rather than as inseparable parts of a single unified business." The Court describes the stand-alone model as a starting point, which then needs to be adjusted for certain effects of belonging to a multinational group.
Whether to take into account "implicit support" has been a key unanswered question for taxpayers trying to apply transfer pricing rules to financing transactions — should related party borrowers be regarded as stand-alone entities, or as subsidiaries of their parent companies?
The Court determined that implicit support of association with a parent company must be considered in determining an arm's-length price. The Court said that the question of how arm's-length parties would negotiate a price "becomes one of fact or, more precisely, one of identifying the economically relevant characteristics of the transaction that may influence the arm's-length parties in their negotiations." These include the implicit support due to affiliation with a parent company.
Unanswered is how far this concept of implicit support should be taken. When a royalty is being paid for a brand name, should some of the benefit carry a zero royalty if the parent company uses the same brand name? Does the decision mean for transfers of intellectual property that some of a subsidiary's intellectual property is economically "owned" by the parent company by virtue of the parent's implicit support? When bargaining power of two parties to a transaction is being measured, should the parent company's influence be factored in? The concept of implicit support could surely affect financial transactions that depend on credit worthiness, since the analogy to the present case would be evident.
What are the implications for transfer pricing purposes of "implicit" elements that are not supported or evidenced by formal legal obligations? In what other situations could the issue be raised of "implicit" elements impacting the determination of arm's length prices? Taxpayers should consider carefully how such a concept may apply to their particular facts and circumstances.Endorsement of yield approach
What approach should taxpayers consider when documenting a guarantee fee?
The Court decided that the most appropriate approach is to calculate the difference in yield rates between the rate at which the taxpayer could borrow without a guarantee, and the rate which it actually paid. The hypothetical rate at which the taxpayer could borrow without a guarantee must take into account the implicit support of the parent company.
The Court emphasized that this difference is not the guarantee fee — it is an upper bound for the guarantee fee, so the actual guarantee fee charged should be lower.
The details of the decision suggest a multi-step process:
1. Determine whether a guarantee was bona fide and necessary to the operations of the subsidiary. This step is stressed in the decision.
2. Determine the guaranteed party's credit rating without the guarantee.
3. Notch the credit rating up or down to reflect the effect of "implicit support."
4. Select the difference in yield rates corresponding to the credit ratings with and without the guarantee but including the adjustment for implicit support.
5. Adjust the corresponding difference for key differences between the transaction and arm's-length benchmarks.
6. The result is a maximum — select a lower guarantee fee.
How can a taxpayer show that a guarantee is necessary? The Court in this case pointed to a keep-well agreement that GECUS signed as evidence of the value of the guarantee. Evidence that arm's-length parties required such a guarantee could also show the guarantee was necessary. In the present case, the fact that the taxpayer would have been unable to procure standby letters of credit to support the commercial paper was also determined to be a relevant consideration. Taxpayers should examine the facts and circumstances of the transaction to identify evidence of the necessity of the guarantee.
The Court considered two other approaches to benchmarking the GE guarantee fee. First a credit default swap (CDS) approach was studied. The Court did not reject this approach, but said it was "analogous" to a yield approach. It is not clear from the decision what the Court meant by "analogous," but we would assert that the result of a CDS approach is very different in principle from the yield approach.
The Court also considered an insurance approach, which the judge rejected because it did not include a profit element. This is not to say that a "cost approach" — a lower bound on the guarantee fee — should not be considered. But taxpayers should tread carefully if they choose not to use the yield approach as one of their methods to benchmark a guarantee fee.Endorsement of "halo" effect for financial guarantee fees
Identifying the existence of implicit support is only the first step. The second step is to determine how much that implicit support is worth. In the present case, the judge decided that a third-party insurer would not attach much value to implicit support of a parent company.
To determine the adjustment required for implicit support, Standard & Poor's provides a methodology that characterizes the parent-subsidiary relationship within a spectrum that ranges from the parent viewing the subsidiary as a mere investment at one extreme to the subsidiary being highly integrated and relevant to the parent at the other extreme. This characterization determines the extent to which the parent would be incentivized to support its subsidiary in times of financial distress. Standard & Poor's lists 13 factors that help determine the characterization of the parent-subsidiary relationship. These factors include the following:
- The strategic importance of the subsidiary to the parent in terms of the subsidiary representing an important supplier to the parent or being an integral member of a highly integrated line of business
- Current and future ownership percentages
- Degree of management control and influence
- Common trade names and trademarks
- Do the parent and subsidiary reside in the same country?
- Are there common sources of capital between the parent and subsidiary?
- Does the parent have financial capacity to support its subsidiaries?
- The level and significance of the parent's investment in the subsidiary
- The parent's investment relative to the subsidiary's debt
- The nature of the parent's ownership — ranging from strategic to financial
- Management's stated position on the nature of the parent's ownership
- The parent's past behaviour with other subsidiaries
- The nature and significance of the potential risks
The significance of each of the above factors would depend on the facts and circumstances of the parent-subsidiary relationship and, as such, they would not necessarily be equally weighted.Facts and circumstances paramount
The Court specifically cautioned that the relevance of lessons from the present case for other situations was a matter of facts and circumstances. "… [O]ne must be aware that it would be dangerous for taxpayers to draw general inferences from this particular case." This underlines a key idea in the OECD Transfer Pricing Guidelines, one which is endorsed by the Canada Revenue Agency, that each situation must be considered relative to its own facts.Limitation of shareholder functions
In a comment potentially relevant to many transfer pricing situations, the judge confirmed that a company cannot be attributed functions performed by its subsidiary simply through the shareholder relationship. The Court stated, "the fundamental distinction remains that shareholders can appropriate the powers to appoint the officers, but not the powers of the officers to in fact manage the business." This statement is a confirmation that functions performed by a subsidiary cannot be said to be performed by the parent simply because of the parent's shareholding in the subsidiary.Final word
The GE case challenges the very notion of what "arm's length" means. An important lesson is that the economic circumstances of a subsidiary include the influence of the parent, which may have far-reaching implications for related-party transactions of many types. Measuring this influence must be pursued thoroughly by taxpayers.