Can we hear a “woohoo”!? Yes, you read right, South Africa has finally had its first TP case and whaddaya know, the taxpayer won! In this case, the taxpayer, referred to very creatively as ABD, appealed against an increased assessment imposed on it by the South African Revenue Service (SARS). ABD is a South African telecommunications company with subsidiaries worldwide, to which it licences certain intellectual property (“IP”) in return for a royalty. And of course the pricing of this royalty was the issue at stake.
What is an arm’s length royalty?
Well that’s the million dollar question, and everybody agreed the arm’s length principle is the fundamental concept governing the valuation of intellectual property (IP) for TP purposes, everyone also seemed to agree that the OECD Guidelines is the key playbook to follow. IP is of course just intangible i.e. non-physical assets, the value of which can be derived from their potential to generate revenue. What IP was being paid for was contested back and forth, but drilled down to brand-related IP, including trademarks but not goodwill. Defining IP is super complex, and requires a lot of specialist analysis, so there were some meaty discussions by some meaty folks throughout the case.
Basically, ABD charged all of the Opcos an identical royalty rate of 1% and SARS was not happy with this and argued it was not arms-length.
Let’s talk pie
ABD’s counsel decided to paint the picture using a comparison of a pie. Their counsel said the first determination is about the size of the pie and the second is about how to divide the pie. One flavour of the pie that SARS and ABD agreed on, is that the arm’s length principle should be applied in determining the royalty i.e. what would it be if it was between two independent enterprises as opposed to a company taxed in one jurisdiction and its subsidiary taxed in another. So far, so obvious but then things got a bit spicier.
SARS went through a rigorous process to contest the 1%, and relied on many expert opinions, including an economist charmingly called Dr (Clean) Slate. But even SARS’ experts differed over how to calculate both the size and division of the pie. SARS settled on their economist’s approach, which surprise surprise supported a much larger share in the pie for the tax-man. This “flip- flop” approach by SARS does not seem to have been considered favourably.
So SARS, advised by Dr Slate, contended that ABD should have charged a variable royalty rate depending on the country and the year it was earned. The fluctuations created by adopting this approach are considerable. And we all thought tax authorities were adverse to varying royalty rates! Although maybe that´s just if the SA tax-payer is paying rather than receiving royalties…
ABD disagreed with Dr Slate’s calculation of the size of the pie but agreed with his division of the pie.
Summary of matters under dispute:
Ok there’s a lot to unpack in the various discussions around the appropriate arm’s length royalty and the key components are addressed below.
- Actual transaction
The rights licensed under the agreement was an issue of debate. SARS assumed the rights included all rights to a brand, including goodwill whereas ABD disputed this. The judge determined that the rights licensed under the agreements with the Opcos did not include goodwill. Sounds simple but what the IP included or didn’t include was fundamental to how it should be valued and how the arm’s length royalty should be derived.
- TP method
A few TP methods were thrown into the mix, each one resulting in a different financial outcome. ABD submitted a dual (or corroborative) approach in support of its position, by relying on two distinct methods, the Transactional Profit Split Method (“TPSM “) as well as a Comparable Uncontrolled Price Method (“CUP”) method to determine the arm’s length nature of the royalty, while SARS submitted the TPSM as the appropriate method. But then both sides had a different application of the TPSM! Ultimately the TPSM was not relied upon in the court’s decision (an argument for another future TP case). The Judge noted ABC did not need to succeed on both methods, one would suffice and the focus moved to the CUP.
- The Cyprus CUP
Hold onto your hats, this is where we get deep into TP territory. So a “CUP” in TP terms is the holy grail, it’s a comparable uncontrolled price i.e. a price charged to a third party for exactly the same transaction as is now being undertaken, as this is clear “proof” (if such a thing ever exists in TP) that the price is at arm’s length as you are charging a third party exactly the same as you are charging your connected party for exactly the same thing. Well in this case, there was a lot of talk about the “Cyprus CUP” – basically ABD had sold a Cypriot subsidiary to a third party after which ABD entered into an IP licensing agreement with the third party. This is called an “internal” CUP, and is generally considered more reliable than an external CUP, because it deals with a more comparable set of facts as it analyses a transaction entered into by ABD itself with a third party (as opposed to pure third party to third party transactions). Are you still awake? Which is preferable as the same IP and business factors are essentially being examined.
Well SARS argued that the Cyprus CUP was not comparable as no adjustments were made for the differences in the following three comparability factors:
- territories the opcos operate in compared to Cyprus;
- the exclusivity of the Cyprus agreement vs the non-exclusive right of the Opcos; and
- the short term duration of the Cyprus agreement compared to the medium to long term agreements with the Opcos.
SARS also questioned whether the Cyprus third party can be considered as an independent third party when it was historically owned by ABD. (Oh come on!)
The judge determined that adjustments to the CUP were not necessary to account for territory, exclusivity and duration differences and that the Cyprus third party is independent, considering its state at the time the contract was entered into. This is a relief as, in the world of TP, often a CUP even though almost impossible to find is the only bastion in a sea of uncertainty and if we couldn’t rely on that, we might have to throw in the towel.
Summary of the judgment
Ultimately it was determined that the Cyprus CUP does indeed serve as a reliable internal benchmark. As a result, the 1% royalty charged by ABD to other Opcos was considered reasonable under arm’s length principles. The judge also added that there was no factual basis for the Commissioner to have adjusted the royalty rate under section 31 of the Income Tax Act. A key point identified was that there appeared to be no incentive for ABD to undercharge its subsidiaries, as tax rates in the Opcos jurisdictions were no more attractive than SA. This is pretty important i.e. there was no tax mischief motive at play as even if the taxpayer were to shift profits to the other jurisdictions no tax would have been saved! In our view, a lot of money and time would have been saved had SARS applied some common sense before raising the assessment in the first place.
Takeaway
So SA has popped its cherry in terms of TP cases and no doubt there will be many more coming soon. There is much more to think about here and more steps yet to come. What is clear is that economic and expert analysis is critical to the defence of any TP policy, and its complexity cannot be under-estimated. Taxpayers and SARS are both willing to fight their corner when it comes to TP disputes and actually obligated to given the tax money at stake. If you’ve any TP concerns or would like to discuss whether this case could impact you, contact us today.
Meet the Authors
Today’s article was written by Deborah Alberts and Lerato Mahlafunya of our Transfer Pricing team, they can be contacted at dalberts@reganvanrooy.com and lmahlafunya@reganvanrooy.com
English 
French