Advance Pricing Agreements: Greater urgency needed

Published: 5 February 2021

The South African Revenue Service (SARS) recently issued a Discussion Paper on Advance Pricing Agreements (APAs).

APAs are agreements with revenue authorities concluded in respect of future transactions regarding the pricing of such transactions. The process of obtaining an APA is typically very onerous and can take several years. Therefore, it is only worth applying for APAs in respect of very significant transactions affecting a multinational enterprise (MNE). APAs are initiated by a taxpayer and require negotiations between the taxpayer, other group companies and one or more tax administrations. APAs are typically valid for an agreed period (e.g. threeyears).

APAs can be divided into unilateral, bilateral and multilateral APAs. Unilateral APAs would be concluded between the taxpayer and the revenue authority of a single country, whereas bilateral APAs and multilateral APAs would bind the revenue authorities of two or more countries respectively. Bilateral or multilateral APAs give a multinational enterprise (MNE) a much higher degree of certainty than unilateral ones since the agreement reduces the risk of a transaction being challenged in one country, despite having been accepted in another. However, since bilateral or multilateral APAs require the involvement of two or more countries in the process of negotiation, they are also correspondingly more difficult to negotiate and reach agreement on.

A related procedure is the so-called mutual agreement procedure (MAP), which is embedded in most (if not all) of the double tax agreements (DTAs) concluded by South Africa. MAPs are provided for in Article 25 of the Organisation for Economic Co-operation and Development (OECD) Model Convention. MAPs differ from APAs in that:

• A MAP process would typically be provided for in a DTA, whereas an APA process would not
• A MAP would be initiated only once a dispute has arisen between an MNE and a tax authority, whereas an APA process is designed to reduce the risk of such a dispute – by obtaining prior agreement regarding the pricing.

Because MAPs are provided for in the DTAs which SA has signed, SARS is obliged to engage in them. However, SARS has to date not implemented an APA programme.

The paper notes that SA has an advance tax ruling (ATR) programme. However, transfer pricing matters are expressly excluded from the ATR process.

The paper notes that the Davis Tax Committee (DTC) recommended the implementation of an APA programme in SA. However, the DTC noted the capacity constraints faced by SARS with regards to negotiating and concluding APAs. The DTC recommended some kind of specialist outsourcing arrangement. However, the paper stated on page 3 that “…no suggestions were made regarding how this could be funded or how conflicts of interest within the very small transfer pricing community in South Africa would be managed.”

The paper considers to what extent there are APA programmes in place in various countries around the world. Specifically with regards to Africa, it is noted that South Africa appears to have fallen behind its peers on the African continent in terms of putting such a programme in place. SARS states that this could adversely affect SA’s status as a leader on the African continent as well as its position as a gateway for investment into Africa.

SA should start with considering a bilateral APA programme. Only once that is in place can it be considered whether to extend it, to cater for unilateral and multilateral APAs or not.

The paper indicates that, while an APA programme is considered important, there are other priorities which need to happen first. These include curtailing base erosion, profit shifting and building transfer pricing capacity. It is estimated that it will take three to four years to implement an APA programme.

The paper is quite comprehensive and we agree with many of the points made. However, we believe that the conclusion (that there are other priorities and that it is likely to take three to four years to implement an APA programme) tends to overlook, or at least to underplay the significance, of the following key point made under heading 8 of the Conclusion: “Tax certainty is one of the fundamental requirements for foreign direct investment.” In the current economic climate attracting foreign direct investment will be crucial. The country does not have three to four years in order to implement a programme as important to making SA more attractive to foreign investors.

The following additional points are also worth considering:

• APAs are intended to avoid disputes. While APAs are time-consuming (and might be expensive if SARS is required to rely on external specialists), the time and expense of a dispute is even greater. Therefore, justifying the non-implementation of an APA programme by referring to capacity constraints is contradictory; those same constrained SARS resources are the ones required to initiate and conduct lengthy disputes.
• SARS envisages starting with bilateral APAs and then adding on unilateral and multilateral ones. Both bilateral and multilateral APAs require the involvement of other tax authorities, whereas unilateral ones would only involve SARS and the taxpayer. The simpler process of implementing a unilateral APA  would seem more logical as a starting point, followed by bilateral and multilateral ones.While ATRs are different from APAs, there will be some overlap (perhaps even a significant overlap) between the process of applying and obtaining an ATR as well as applying and obtaining a unilateral APA.
• as regards funding of external specialist consultants, SARS would presumably charge for APAs (as it currently does for ATRs). Therefore, this cost could be self-funded via the APA process itself, rather than requiring SARS to incur higher fixed costs by needing to employ additional skilled resources fulltime.
• the process of working together with skilled external consultants on APAs should achieve significant skills transfer to the SARS personnel involved.
• the perceived potential conflicts of interests applicable in the relatively small transfer pricing community in SA are also not necessarily as significant as SARS indicates.The same specialists are required to do different types of other transfer pricing work (including planning, documentation and assistance with dispute resolution) for a variety of clients and are accordingly used to having to manage potential conflict situations.
• retired professionals or professionals from outside SA could be used in response to the potential conflicts of interest concerns raised.
• the points made by the paper regarding the absence of an APA programme adversely affecting SA’s status as a leader on the African continent and the attractiveness of SA as a gateway for investment into Africa are also crucial. Again, it seems tardy to only commit to a three to four-year timeframe for remedying this shortcoming.
• the APA programme does not have to be perfect from the start. It would be preferable to implement a programme , and then to refine the programme on an ongoing basis, rather than to wait for the perfect conditions for such a programme (which might never exist).

In summary, while the paper represents a significant step forward, SARS should look to prioritise the implemention of a South African APA programme.