Malusi Gigaba stated in the Budget Review 2018 (Budget) that: "The deductibility of interest payments on debt acts as an incentive to use debt rather than equity funding, and can be used to strip profits from high tax countries." He goes further by stating that "[a] discussion document inviting comments will soon be published to facilitate public consultation".
This is a welcome development as it is approximately five years since the South African Revenue Service (SARS) issued its draft interpretation note (IN) into acceptable levels of debt. Taxpayers and foreign investors will therefore welcome the prospect of getting closer to greater clarification on what constitutes an arm's length acceptable level of debt and interest rate.
Inbound debt funding from connected persons currently falls within the ambit of transfer pricing rules in section 31 of the Income Tax Act 58 of 1962 (ITA). Historically, South Africa followed a simple formulaic approach in determining what an acceptable level of debt was. This was contained in Practice Note 2 (PN 2) which dates back to 1996. In 2012, despite a significant change in the transfer pricing rules, which had the effect of making the majority of PN2 redundant, the PN has not been officially withdrawn. SARS intended to replace the PN with its draft IN but as this remains in draft we are faced with two contrary pieces of guidance, neither having legislative teeth. The draft IN remains in circulation with no final interpretation or updated guidance from SARS on the interpretation of section 31 to inbound financial assistance. The result, increased uncertainty as to what would constitute arm's length levels of debt and interest rates.
The draft IN proposed a two test approach to applying transfer pricing rules to inbound debt. First, determining that the amount of debt funding is arm's length, i.e. would a third party lender be prepared to lend the borrower the same level of debt under the same terms and conditions? Secondly, is the pricing of the debt (the interest rate) also arm's length? As can be envisaged, this is no simple analysis.
The draft IN tried to provide some level of guidance by providing a so-called "risk harbour" whereby a borrower with debt to earnings before interest, tax, depreciation and amortisation (EBITDA) ratio of 3:1 was less likely to be subject to an audit. In addition, a similar approach was provided for the debt pricing whereby, interest rates of the weighted average JIBAR (Johannesburg Interbank Agreed Rate) (Rand denominated loans) plus 2% or weighted average of the relevant base rate (foreign currency denominated loans) plus 2% were considered to be of lower risk. Whilst giving some degree of comfort, neither of these guidance approaches could actually be relied on as they simply represented risk levels and not arm's length support. Thus to be certain, taxpayers actually need to undertake a detailed analysis as to what an arm's length level of debt would be based on the borrowing capabilities of the taxpayer and at what interest rate this would be lent. Such analyses are complex and costly, therefore it is unsurprising that this is an area that needs addressing.
What is perhaps more concerning is that since the issue of the draft IN, there has been a significant change in international precedent concerned with excessive debt deductions, both through the release of the final OECD BEPS reports and through international case law. Furthermore, South Africa has so many interest rate limitation sections in the ITA and there is a real need to align these.
The Davis Tax Committee (DTC) tasked with considering South Africa's tax policy and particularly the OECD Base Erosion and Profit Shifting (BEPS) reports in a South African context, has taken this on board.In its Second Interim Report on Action 4: "Limit base erosion involving interest deductions and other financial payments", the DTC took heed of the uncertainty arising from the current position, notably the lack of clarity provided by the conflicting guidance. The DTC recommended that a "safe harbour" with a fixed ratio be introduced in section 31 or as a minimum, the interpretation note be finalised to provide non-residents funding South African entities with more certainty as to acceptable levels of debt and debt pricing.
The DTC also recommended that South Africa follow as closely as possible the recommendations from the OECD and other international precedent. To this end, the OECD recommends an approach based on a fixed ratio rule which limits an entity's net deductions for interest to a percentage of EBITDA, with a range of possible ratios between 10% and 30%. The OECD recommendations also propose a group ratio rule alongside the fixed ratio rule. This would allow an entity with a net interest expense above a country's fixed ratio to deduct interest up to the level of net interest/EBITDA ratio of its worldwide group. This approach is not that different to the approach outlined in section 23M of the ITA.
The DTC also recommends that in considering the final approach to be adopted, cognisance should be taken of the outcome of international cases such as General Electric and Chevron cases. These cases basically endorsed the so called "halo effect" in terms of which the implicit support of the group resulting in a higher credit rating of a borrower should be considered in the factors determining the credit-worthiness of the borrower.
Finally, The DTC also recommends simplification of the rules and introducing ways to reduce the compliance costs for taxpayers with a low risk of BEPS through interest deductions. These could be by introducing a safe harbour with a fixed ratio or threshold based on a loan value or other measure. Interest rates acceptable for exchange control should also be aligned with acceptable levels of interest rates for transfer pricing purposes.
What is encouraging is the comment made by Minister Gigaba to the effect that "Government is striving for a balance between certainty, simplicity and adequate base protection to ensure a sustainable corporate tax base". It is therefore hoped that the discussion document will follow this and incorporate the OECD and DTC recommendations on transfer pricing rules applicable to allowable debt and interest deductions. Furthermore, it is hoped that after the problems following the release of the draft IN, the discussion document is circulated for comment and finalised sooner rather than later.