1. Purpose of the blog
Intra-group guarantees are usually used by companies within the same multinational enterprise to obtain beneficial conditions for funding arrangements. For instance, lower interest rates due to the decreased level of the credit risk assumed by the lender (such as a bank) or/and extended borrowing capacity. In this blog, we will discuss the Transfer Pricing (TP) aspects related to intra-group financial guarantees based on the OECD 2017 TP Guidelines (OECD TP Guidelines)[1]OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations 2017, OECD Publishing, Paris (OECD TP Guidelines). and the OECD Discussion Draft on Financial Transactions (Discussion Draft) in June 2018.[2]BEPS Public Discussion Draft, BEPS Actions 8 – 10, Financial transactions, 3 July – 7 September 2018. The blog will focus mainly on downstream guarantees and will not deal with cross guarantees or upstream guarantees in detail.
2. Accurate delineation – Is an intra-group guarantee fee payable?
2.1. Introductory comments
There are two questions that need to be answered with respect to determining the arm’s length nature of the provision of a downstream guarantee i.e. (i) whether an intra-group service has been provided, and (ii) if so, whether the intra-group charge is in accordance with the arm’s length principle.[3]Para. 7.5 of the OECD TP Guidelines.
The answer to the first question depends on whether the activity (i.e. granting of a guarantee) provides the group member with economic value to enhance its commercial or financial position. This can be determined by considering whether an independent enterprise in comparable circumstances would have been willing to pay an unrelated party for the service.
The example below illustrates the analysis as to whether the provision of a guarantee amounts to an intra-group service by considering the guidance provided in the OECD TP Guidelines and the clarifications provided in the Discussion Draft. As to the basic facts, we have assumed that Co A is the parent company of the XYZ Group, which is a resident of State X. Its stand-alone credit rating is AAA. Co A owns a subsidiary, Co B, which is a resident of State Y. Co B needs funds to meet its working capital requirements, and approaches a bank resident in State Y for a loan of USD 10 million. The bank, after analyzing the financial position of Co B, may make various conclusions as per the cases considered below.
2.2. Passive association
As Co B is a part of the prestigious XYZ Group, its credit rating is adjusted upwards from a stand-alone BBB up to A. Due to its affiliation, we assume that the interest rate corresponding to A rating would be 8%. In this case the enhanced credit rating and the lower interest rate would be due to the so-called “passive association” due to Co B being part of XYZ Group. The benefit obtained purely from such association would not be treated as a provision of an intra group service, hence no fee would be payable.[4]Para. 7.13 & Para 1.167 of the OECD TP Guidelines; Paras 67-74 of the Discussion Draft.
2.3. Deliberate concerted action that does not provide a benefit
At the outset, it should be noted that comfort letters do not warrant the payment of a service fee.[5]Para. 142 of the Discussion Draft. This being said, arguably, a formal guarantee should generally provide a benefit to the borrower. However, this may not always be the case.
For the purpose of this situation, we will assume that Co A provides an explicit guarantee (i.e. a legally binding arrangement) with regards to the loan contracted by Co B. The effect of that deliberated action is to ensure that Co B is able to borrow debt which it could not have been able to obtain on a stand-alone basis. In this situation, it could be argued that such a guarantee should be treated as a ‘shareholder activity’ rather than a service.[6]Paras 7.6 and 7.9 of the OECD TP Guidelines. In other words, Co A has given the guarantee solely in his capacity as a shareholder. The Discussion Draft seems to confirm this position by stating that where the guarantee only increases the debt capacity of the borrower, it may be recharacterized as a loan to the guarantor followed by an equity contribution to the guaranteed entity. This said, it should also be noted that, in practice, it may be difficult to identify whether the extended guarantee only increases the borrowing capacity or acts both to allow the borrower to obtain a greater amount of debt and reduce the interest rate on the debt. The Discussion Draft suggests that in such a case the guarantee fee should be apportioned. A guarantee fee corresponding to the portion of the loan which is still respected, being made from the lender to the borrower, should be analyzed from the arm’s length perspective. The remainder of the fee corresponding to the loan re-characterized, being made from the lender to the guarantor, followed by an equity contribution from the guarantor to the borrower should be disallowed.[7]Para. 140 of the Discussion Draft.
It could also be argued that an explicit guarantee extended by Co A to Co B that does not provide a benefit beyond the credit rating enhancement attributable to implicit support would not confer any benefit to Co B. This would be the case when, for example, banking covenants include the default of another group member as an event that may cause the termination of a loan arrangement or other adverse consequences. In this case, the strategic interest of Co B for the XYZ Group would be so important that any default would lead to high costs, therefore due to being financially interdependent, the credit rating of Co B would approximate the group rating.[8]Para. 143 of the Discussion Draft. The factors which may be used to determine the status of an entity in this regard may include considerations such as legal obligations (including any guarantee commitments or regulatory requirements), strategic importance, operational integration and significance, shared name, potential reputational impacts, general statement of policy or intent, and any history of support.[9]Para. 69 of the Discussion Draft.
Likewise, a similar issue may occur in case two or more entities of the XYZ Group guarantee each other’s obligations (e.g. “cross-guarantees” issued in cash pooling transactions). In addition to the fact that it may be difficult to evaluate each guarantee in view of the number of the counterparties involved, it may also be difficult to determine the effect of the guarantee where the same risk is subject to multiple guarantees. Consequently, a conclusion may be drawn that the guarantees do not provide any benefit beyond passive association, while any support in the event of default from another group member should be regarded as a capital contribution.[10]Para. 144 of the Discussion Draft.
Last but not least, if a loan, guaranteed by Co A were obtained by Co B for the purpose of the acquisition of the participation in another company in State Y, the questions arises whether the provision of the guarantee could be treated as a ‘shareholder activity’. The OECD TP Guidelines state that the costs of raising funds for the acquisition of a participation in another company could be a shareholder activity. However, the Discussion Draft does not provide additional comments in this regard. Further clarification in this relation needs to be provided.
2.4. Deliberate concerted action that actually provides a benefit
For the purpose of this case, we assume that Co A provides an explicit guarantee (i.e. a legally binding arrangement) with respect to the loan contracted by Co B, and the applicable interest rate would be 6% instead of 8%, as the credit rating would be adjusted upwards from A to AAA due to the issued guarantee. Furthermore, unlike the above-mentioned case, the enhancement of Co B’s credit rating from A to AAA is attributable to a deliberate concerted action (i.e. benefit obtained beyond pure implicit support). Arguably, in this situation a guarantee fee should be payable.[11]OECD TP Guidelines, Section D.8 of Chapter I, Paras 1.158-1.159.
If Co A were to charge Co B a guarantee fee of 3%, Co B would be better off without receiving such guarantee since it completely offsets the benefit of Co B’s enhanced credit rating from A to AAA.[12]Paras 157-159 of the Discussion Draft. Therefore, the guarantee fee would need to be determined by allocating the benefit (i.e. the maximum spread of 2% being the difference between the interest rates of 8% and 6%) obtained by the guarantor and the borrower (also taking into account the impact of the implicit support) by resorting to the pricing approaches discussed below.[13]Paras 160-161 of the Discussion Draft.
3. Pricing the intra-group financial guarantees
The arm’s length level of the guarantee fee should generally be determined from the perspective of the guarantor and the guarantee recipient by establishing a range of fees that the guarantor would (at least) want to receive (typically covering all its costs and risks) and the fee that the guarantee recipient would (at most) be willing to pay. The analytical methods and approaches to pricing the guarantee fees, which are generally applied in practice, are outlined below.
3.1. Internal or External CUPs
The CUP method may be used when there are internal (i.e. where the borrower has other comparable independently guaranteed loans) or external comparables. Third party publicly available information on potentially comparable uncontrolled transactions (e.g. credit default swaps) may be considered as a benchmark provided there is a high degree of comparability with the intercompany guarantee. However, given that the information on unrelated party guarantees are not usually available in the public domain, the method may not be applicable.[14]Paras 146-148 of the Discussion Draft.
3.2. Yield approach
A typical approach traditionally adopted for calculating the arm’s length guarantee fee is the “yield approach”: a spread would be determined as a difference between the interest rate that the borrower would pay on a stand-alone basis vs. taking into account the explicit guarantee. However, the application of the yield approach was adjusted in the recent years in view of the court practice (General Electric case)[15]CA: General Electric Capital Canada, Inc. v. Her Majesty the Queen (2009) TCC 563; Her Majesty the Queen v. General Electric Capital Canada Inc. (2010) FCA 344. as well as the OECD TP Guidelines, to take into account the impact of implicit support. This means that only a benefit beyond the implicit support attributable to the explicit guarantee should be chargeable as a guarantee fee.
Transposing this statement to the above-mentioned example, such benefit would correspond to the difference between the borrowing terms obtained by the borrowing entity based on the credit rating with the guarantee (in the case at hand, 6%) and the credit rating as a member of the group (in the case at hand, 8%). The 2% spread constitutes the maximum that the borrower would be expected to pay. However, in order to ensure that the fee is arm’s-length (i.e., beneficial to both the guarantor and the borrower), the spread should be split between both Co A and Co B based on a combination of factors e.g. the negotiating position of each party, risk exposure for the guarantor, an incentive for the borrower to obtain the guarantee from the parent rather than an unrelated guarantor (taking into account any required adjustment as a third party can potentially provide a higher economic benefit compared to a related party). In practice, the application of this method may turn out be very subjective due to the lack of detailed guidance on which factors should be used.
3.3. Cost / valuation of expected loss approaches
These approaches consider the minimum fee that the guarantor will be willing to accept based on the valuation method (e.g. put option, credit default swap pricing models) by estimating the value of the expected loss or the probability of the guarantor of having to inject capital in case of default. Since the derived fee would not represent an arm’s length outcome of the bargain, it would need to be adjusted by considering the options realistically available for both the guarantor and the borrower.[16]Para. 153-154 of the Discussion Draft.
3.4. Capital support method
This method is based on determining the credit rating of the borrower without the guarantee but accounting for implicit support and identifying the amount of the additional notional capital required to adjust the borrower’s credit rating up to the credit rating of the guarantor. The guarantee fee is determined based on the expected return on this amount of capital, i.e. to the extent that it reflects only the results/consequences of the provision of the guarantee.[17]Para. 155 of the Discussion Draft.
4. Summary and conclusion
The subjectivity related to the pricing of intra-group financial transactions has significantly increased the tax risk of multinational groups in recent years. Once finalized, the additional guidance on pricing intra-group financial transactions provided in the OECD Discussion draft will require taxpayers and tax authorities to consider the issue of pricing guarantees taking into account the notions of implicit support, deliberate concerted actions, and other local tax considerations that may not provide a direct link to the OECD TP Guidance. For instance, in the 2010 US Tax Court decision in the Container Corporation case,[18]US: TC, Container Corporation v. Commissioner, 134 T.C. 122. the intercompany guarantee was characterized by the Court as a service for US tax reporting and withholding tax purposes as it was rendered from Mexico and the payment was treated as a foreign-source payment not subject to US withholding tax. It was in reaction to the decision in this case, that the US Congress changed the sourcing rule to treat certain guarantees of indebtedness as interest. Therefore, it may be suggested that the taxpayers would need to reassess the tax risks related to the intra-group financing arrangements, including financial guarantees, taking into account the applicable local regulations as well as possible interpretations of the new OECD Guidance in the jurisdictions of the borrower and the guarantor.
***
Vera Averyanova holds an LL.M. degree in international taxation from the University of Lausanne and based this article on the Master Thesis she submitted for the purpose of fulfilment of requirements of MASIT at the University of Lausanne (2015) which was updated during her research stay at the Swiss Institute of Comparative Law (Lausanne, August 2018).
Dr. Vikram Chand is the Executive Director of the Master of Advanced Studies in International Taxation and the Executive Program in Transfer Pricing at the Tax Policy Center, University of Lausanne, Switzerland.
The authors would like to thank Mr. Stefaan De Baets (Senior Counsel, Transfer Pricing, PwC) and Ms. Amanda Pletz for their valuable comments on the initial draft of the article.
________________________
To make sure you do not miss out on regular updates from the Kluwer International Tax Blog, please subscribe here.
References
