There have been few transfer pricing-related tax cases in South Africa. They mainly dealt with technicalities of section 31 of the Income Tax Act (‘the ITA’). However, a recent tax court case deals with applying the arm’s length principle, which is at the heart of transfer pricing. The court considered the appropriateness of transfer pricing methods used by a multinational entity in the telecommunications industry.

Facts and dispute

ABD Ltd (‘ABD’) is a South African telecommunications company with global subsidiaries. From 2009 to 2012, it charged its subsidiaries royalties at a uniform rate of 1% for using its intellectual property (IP). 

SARS challenged this rate, arguing it did not reflect an arm’s length transaction. SARS based the additional assessments on an expert report. However, it later switched course by relying on a new report. This report proposed higher, variable royalty rates based on a “Willingness to Pay” (‘WTP’) survey. The taxpayer argued that the 1% rate was the Comparable Uncontrolled Price (‘CUP’). (The CUP method compares the price charged in a controlled transaction with that in a comparable uncontrolled transaction.)

Judgment

The court concluded that ABD’s 1% royalty rate reflected arm’s length terms and conditions, primarily based on the application of the CUP method. ABD used an internal comparable from an arrangement to make its IP available to an independent party in Cyprus on terms similar to its subsidiaries. The court ruled that this was a reliable and robust benchmark confirming the 1% royalty rate. When applicable, the CUP method is preferred under OECD guidelines, as it directly mirrors market conditions. This approach is straightforward and consistent with established transfer pricing principles.

The court sharply criticised SARS for abandoning its initial expert analysis without sufficient justification and relying on an untested WTP survey. It described this approach as speculative and unreliable, noting that it lacked legal and empirical validation in transfer pricing litigation. The criticisms against the WTP survey included that:

• It wrongly assumed that ABD licensed broader IP rights, including goodwill, to its subsidiaries.
• The survey questions were designed to lead respondents toward predetermined answers, undermining the reliability of the results.
• The survey attempted to gauge customer preferences from 2009-2012 in 2020. The court considered this to be inherently unreliable.

The court considered the WTP approach novel and untested in transfer pricing cases, raising concerns about its appropriateness.

Take-Home Message

Although this is only a tax court case, it provides a glimpse into how the South African courts may approach arm’s length principle-related disputes. It suggests a preference for established and recognised methods, such as those described by the OECD transfer pricing guidelines.

It arguably also demonstrates that the courts are likely to critically assess the appropriateness of the methods and assumptions. The type of criticisms raised by the Tax Court suggests that this statement by the Canadian courts in the case involving General Electric Capital, Canada, may well be a good indication as to how South African courts would approach these matters:

“[t]ransfer pricing is largely is question of facts and circumstances coupled with a high dose of common sense.”