Editor’s Note:
In this fourth newsletter for 2024 we consider the following:
RSA Resident Trusts with non-resident beneficiaries revisited
Constitutional Court Tax Jurisdiction
The Tax Court – Status and Representation Determined (Guest Author)
Musings on SARS Advance Tax Rulings (ATR)
SARS Interpretation Notes (INs), Draft INs, BGRs and VAT Rulings Noter-Up

Tony Davey – Editor
Duncan McAllister – Co-Editor


Amendments to Section 25 B (1) and (2) effective 1 March 2024 (see Davey’s Locker 3.2023) have resulted in adverse consequences for DWT (Dividends), WHTI (Interest) and WHTR (Royalties) vis-à-vis non-resident beneficiaries of RSA trusts. In essence, the effect of the amendments is that revenue amounts received by a RSA trust and awarded to non-resident beneficiaries will nonetheless be taxed in the RSA trust thus negating the conduit principle for non-resident beneficiaries. This may at first glance appear innocuous vis-à-vis normal tax being trade (including rental) income as this aligns with the CGT treatment.

However, the adverse impact upon other taxes being interest, royalties and dividends (as well as negating provisions of Double Tax Agreements (DTAs)) is significant.

Resultant from the Section 25B amendments to tax the above types of income (termed ‘amounts’) in the RSA trusts, consequential amendments have been made to Sections 49D (Royalties) and 50D (Interest) to effectively remove such amounts from the operation of such withholding taxes.

Thus, the RSA trust will be subject to normal tax at the flat rate of 45% on interest and royalty income, instead of 15%, as possibly further reduced by a DTA. Dividend withholding tax imposed under Section 64E at the rate of 20% will be borne by the RSA trust and any reduction under a DTA is inapplicable, because the beneficial owner of the dividend is deemed to be the RSA trust and not the non-resident beneficiary.

In summary, non-resident trust beneficiaries are now disadvantaged as compared to other non-residents who hold RSA investments directly in their own name. Maybe it is time to cash out of a trust albeit that this may trigger a tax event.


We still await the respective judgements to be delivered in the Coronation and Thistle cases.

The Constitutional Court, according to its website ‘deals exclusively with constitutional matters – those questions that raise questions about the application or interpretation of the Constitution’. ‘The Supreme Court of Appeal (SCA) ……. is the highest court in respect of all other matters’.

A 2012 amendment, Section 167(3)(b)(ii) of the Constitution, extends the ambit of the Constitutional Court beyond constitutional matters if the disputed matter pertains to a point of law of general public importance. Prior to this amendment, the Constitutional Court heard few tax-related cases which involved, for example, the pay-now-argue later principle, the non-binding nature of SARS Interpretation Notes (INs). These were more clear-cut constitutional matters as they involve a persons (taxpayer’s) rights as per the Bill of Rights in chapter 2 of the Constitution.

In a post-amendment case, the Constitutional Court in Paulsen & Another v Slip Knot 2015(3)SA479 CC held that in determining its broader jurisdiction, the element of ‘general public importance’ must be present. The Constitutional Court after citing Kenyan and English Court decisions, stated as follows:
‘In sum, for a matter to be of general public importance, it must transcend the narrow interests of the litigants and implicate the interest of a significant part of the general public. It will serve a litigant well to identify in clear language what it is that makes the point of law one of general public importance’.

Given the above criteria required to qualify to be within the jurisdiction of the Constitutional Court, one wonders whether complex and specific tax legal points which would already have been traversed by the Tax Court, High Court and Supreme Court of Appeal, should ever be entertained by the Constitutional Court. Should the Constitutional Court be pronouncing upon the tax intricacies of Section 9D exemptions (Coronation case) or paragraph 80 of the Eighth Schedule (Thistle case)? Further to this the Judicial Services Commission’s explanation of criteria for judicial appointment, published 4 April 2022, states at paragraph 12.1 that ‘candidates for specialist courts would be expected to have particular experience and skill in the relevant area’. With respect, how many Constitutional Court judges have an H.Dip Tax or equivalent?



The recent High Court case of Poulter v CSARS (A88/2023) [2024] ZAWCHC 97 (2 April 2024) has provided some much-needed clarity on the question of representation in the Tax Court and the status of the Tax Court itself.

The genesis of the matter can be traced back to two other disputes involving the same taxpayer, the model Candice-Jean van der Merwe in her ongoing tax travails with SARS. (See ‘Davey’s Locker 6.2022).

The Poulter appeal to the full bench of the High Court

The principal point argued by the appellant taxpayer’s counsel, is that the Tax Court had misdirected itself in holding that the taxpayer’s father whose authority to do so was vouched by a power of attorney given by the taxpayer, was not entitled to appear on her behalf in that forum, being the Tax Court.

The taxpayer did not attend the hearing and relied upon the representation of her father to argue the case on her behalf. SARS relied upon Rule 44 (7) of the Tax Court Rules and argued that since her father was not a recognised legal practitioner (not being an attorney or advocate registered with the Legal Practice Counsel (LPC)) and in the absence of the taxpayer herself, the matter should proceed on a default basis. The Tax Court agreed with this line of argument and issued a default judgement.

On appeal, the full bench of the High Court in deciding that the South African common law did not permit lay person representation in courts of law traced the rule back to a practice established in the English courts in the 18th century.

Both counsel in the matter referred to section 33 of the Legal Practice Act 28 of 2014 (as amended) as authority that only a practising legal practitioner could appear in any court of law or before any board, tribunal, or similar institution in which only legal practitioners are entitled to appear. The High Court found that section 166 of the Constitution identifies the courts of law in our judicial system and the Tax Court is not a court referred to in section 166 (a) to (d) of the said provision.

The High Court went on to find that: ‘ …a tribunal intended to serve an administration purpose, even if it is labelled a ‘court’ by the legislation in terms of which it is established, and even if, in fulfilling its administrative role, it is required to act judicially’ is not a court unless it finds a place within the confines of section 166 of the Constitution.

The High Court compared the Tax Court to the position of the Council for Conciliation, Mediation and Arbitration (CCMA) in that both were creatures of statute. The CCMA derived its arbitration powers granted under the Labour Relations Act 66 of 1995 (as amended) (LRA) to fulfil the primary goal of the LRA which is to promote labour peace by the effective settlement of disputes.

By analogy, the Tax Court similarly functions as a body of ultimate assessment under the Tax Administration Act 28 of 2011 (TAA). The Court found that the Tax Court steps into the shoes of the Commissioner for the administrative function directed at achieving one of the important goals of the TAA, namely the correct assessment and recovery of taxes.

Other reasons for distinguishing the Tax Court from a Court of Law are that the sittings of the tax court are generally not open to the public whereas Courts of law, subject to exceptions in special cases or where expressly otherwise provided by statute, sit and determine cases in public. Further to this, Section 132 of the TAA provides that the judgements of the Tax Court must be published in a form that precludes identification of the appellant.

The High Court thus concluded after reviewing the available jurisprudence on the matter, that the Tax Court’s function is essentially that of an administrative tribunal.

The High Court then considered the question of whether laypersons are prohibited from representing taxpayers in the Tax Court. The High Court referred to numerous instances from the TAA and where the said Act authorised direct interaction with SARS and the taxpayer’s authorised representative regarding objections, appeals and ADR proceedings.

The Court pointed out that before a Court of Law, these are all the sort of actions that can be done only by a legal practitioner on behalf of a litigant. There is, however, no requirement in the TAA or Tax Court Rules, that the taxpayers authorised representative must be an admitted legal practitioner. The High Court highlighted that experience had shown that in the past taxpayers had often been represented in proceedings before a tax court by an accountant or a similarly qualified tax practitioner rather than a legal practitioner. The High Court thus upheld the taxpayer’s appeal and concluded that the Tax Court was misdirected in refusing to entertain the father’s representation of his daughter at the hearing of the appeal in the Tax Court. The Court directed that the matter be referred back to the Tax Court to be heard de novo together with an order for costs against SARS.

Closing comments

The Court has now declared that the Tax Court is not a Court of law but is an administrative tribunal similar to the CCMA. Right of representation by an authorised representative (given by written power of attorney) is a minimum requirement for the drafting of pleadings, appearance in the Tax Court and any and all interactions with SARS on behalf of the taxpayer client.

There is no limitation on who may represent the taxpayer subject to written authorisation by power of attorney. My suggestion is that such a representative should at least be a registered Tax Practitioner regulated under the TAA. This ensures standards are maintained re qualifications, experience, ethics and discipline.

Eds Note
We await whether SARS will appeal this High Court decision. Although we concur that sections 166(a) to (d) of the Constitution refer to general courts of Law, subparagraph (e), refers to ‘any other court established or recognised in terms of an Act of Parliament …..’. Is the TAA not an Act of Parliament? On a pragmatic note however, we welcome the decision.

Member, Legal Practice Council and SARS accredited Tax Practitioner


In the era of the late tax doyen Costa Divaris’ Newsletters which spawned this Newsletter, I commented favourably on the advantages of an ATR for purposes of achieving tax certainty on complex proposed transactions. My experiences with SARS ATR Division were both constructive and professional.

Upon perusing recently published Rulings (see updates at end of this Newsletter), I observed that although the factual background, description of the proposed transaction, sections of the tax legislation and the resultant SARS Ruling are set out, it would be useful if the rationale behind the Ruling was provided. For example, the recent BPR400 pertained to whether an issue of shares by a company to a BBEEE trust for no consideration constituted a donation. SARS ruled that on the facts in this circumstance there was no donation and hence no donations tax was payable. The result was given, but not the explicit rationale therefor.

I surmise that the rationale was that as the recipient trust was a PBO, the section 10(1)(cN) exemption applied, thus rendering the donation exempt under section 56(1)(h). Alternatively, the transaction may not have been a donation because the elements of ‘pure liberality’ and ‘disinterested benevolence’ (as per the Welch case 2004 ZASCA) were absent, as the company had a business motive. Another alternative is that the issue of shares by a company is not a disposal of property and hence not a donation.

Although SARS BPRs are not binding on third parties, a summary of the rationale by SARS would nevertheless guide taxpayers in their future transaction planning where the facts might not be exactly the same, but at least the principles would already be clarified vis-à-vis SARS stance on a matter.

Further to this, the workload on SARS ATR division would likely be reduced as this guidance may obviate many unnecessary Ruling applications.