The draft note commences by describing a missionary as a member of a religious mission, which consists of a group of people sent by a religious body to perform religious and social work, educational or hospital work.  Missionaries often operate under the banner of a society, agency, association, fellowship or some denominational body.

Typically, a missionary is not employed by a specific organisation, but relies on contributions to meet living expenses and operating costs of the mission work.  Sometimes these contributions come directly to the missionary, and sometimes via a missionary organisation that controls and administers donations and deploys them to missionaries in the field according to needs and availability of resources.  And here is the reason for claims that missionaries are not taxable on contributions to them; not employed, and dependent on charitable giving.

The draft note predictably starts with paragraph (c) of the definition of “gross income” in section 1(1) of the Income Tax Act.  This paragraph provides that an amount falls within paragraph (c) if it is received or accrued in respect of services rendered or to be rendered by the recipient.  The paragraph makes no exception where the amount is a voluntary award, or where it emanates from outside the Republic.

“In respect of” connotes a causal relationship between the amount and the services of the missionary.  The question is not: what prompted the donor to make the donation or, in the case of a controlling missionary organisation, to award the amount to the missionary?  Rather the question is: why was the payment made to the recipient?  This question was succinctly answered by Howie P in Stevens v CSARS [2006] 69 SATC 1 SCA.  The amounts were received by the recipients because they were employees who had received a benefit directly linked to their employment, and the payments were therefore taxable.

Lest it be argued that the missionary is not employed, the case cited in the meaning of “voluntary awards”, CSARS v Kotze [2002] 64 SATC 447 C, supports the “in respect of“ argument as well as and perhaps better than Stevens v CSARS, because the taxpayer in CSARS v Kotze was not an employee of the entity, the SA Police Service, that paid him the amount.  The service Kotze rendered was to alert SAPS as to what he suspected was an IDB transaction about to happen.  As a result, the illicit diamond dealers were caught in the act.  To reward Kotze for his public spirited action, the Commissioner of SAPS for the Western Cape awarded him with an amount maintained for this purpose.  Kotze won in the tax court, where the court relied on a decision of the UK tax court that such a payment was in the nature of an accolade and not for services rendered.  The fact that the recipient was Bobby Moore, captain of the only England team ever to win the FIFA World Cup, perhaps at least subconsciously influenced the court.  This argument did not succeed in CSARS v Kotze, where the court found that “services need not be rendered by virtue of any contract, nor need the amount received or accrued be by reason of any contract or obligation: it can be a purely voluntary payment”.

Next, the draft note states that an amount may be a donation in the hands of the payer, but nonetheless fall within paragraph (c).  Section 56(1)(k)(i) of the Act contemplates this type of situation by exempting from donations tax “any property which is disposed of under a donation as a voluntary award…the value of which is required to be included in the gross income of the done in terms of paragraph (c) of the definition of ‘gross income’ in section 1”.

Finally, the draft note makes the point that income received or accrued in respect of services has its source where the services are rendered.  And this is the Republic, where the missionaries are providing their services, and not the country from which the donations come.

Although the document is in draft form, it is difficult to envisage a different conclusion in the final version.

She would buy a residential property and live in it while she refurbished it and put it back on the market a few months later. Having sold it at a gain, she would buy another property and repeat the process. This went on to the point where she had bought, refurbished and sold five properties in four years, in each case using the primary residence exclusion to avoid paying tax on her gains. In each instance both the purchase price and selling price were below R2 million.

Her new tax consultant was concerned when she saw the taxpayer’s property dealing history and warned her that SARS could decline to allow her the primary residence exclusion on the grounds that her conduct amounted to a trade. The taxpayer was adamant that so long as she ticked the primary residence box on her annual tax return, which she was entitled to do because she at any particular time owned only one property, in which she resided, it was axiomatic that the exclusion applied. Therefore whenever she sold a property it was her primary residence and she was entitled to the exclusion.

The tax consultant said she would decline to tick the primary residence box on the taxpayer’s tax return, because she was concerned that this was an offence under the Tax Administration Act (TAA). It was possibly fraud, certainly misrepresentation, and non-disclosure of material facts. She was concerned about being seen to collude with the taxpayer in committing a tax offence. Not only would she be exposed to sanctions by SARS, but her Recognised Controlling Body (RCB) would be likely to take a dim view of her action. She warned the taxpayer that, should SARS at any time challenge her claim for the primary residence exclusion, SARS could reopen previous assessments for as far back as the taxpayer been conducting this activity. The three year prescription period for assessments does not apply where there has been fraud, misrepresentation, or non-disclosure of material facts.

The taxpayer poured scorn on the tax consultant’s opinion, which she described as off the wall, misguided and ill-informed. In her view, SARS was the tax expert, not the tax consultant, and if SARS accepted her claim for the exclusion it was no business of the tax consultant to second guess SARS. Needless to say, the taxpayer and the tax consultant parted ways.

Who is correct in this situation? Has the tax consultant acted correctly in terms of the tax legislation?

Gross income is defined in section 1 of the Income Tax Act as the total amount, in cash or otherwise, received by or accrued to or in favour of a person, excluding receipts or accruals of a capital nature. When we consider whether a receipt or accrual is capital in nature, and then potentially subject to the provisions of the Eighth Schedule, it is necessary to consider case law, and in particular the question of the intention of the taxpayer. For example, in CSARS v Capstone 556 (Pty) Ltd [2016] 78 SATC 231 ZASCA the Supreme Court of Appeal held that in order for a profit to be revenue in nature “the gain must be acquired by an operation of business in carrying out a scheme for profit-making”. In so finding, the court drew on a long line of decisions. A good example relevant to the taxpayer is Natal Estates Ltd v SIR [1975] 37 SATC 193, where the court found that in determining whether any particular case is one of realising a capital asset or carrying on a business of selling land for profit, the totality of the facts of the case must be considered in their relation to the ordinary commercial concept of carrying on a business or embarking upon a scheme for profit. Considerations will include, inter alia, the intention of the taxpayer both when acquiring and selling the land; the owner’s activities in relation to the land prior to decision to sell, and the light that these considerations throw on the owner’s statements of intention.

In the present matter the taxpayer set about using the primary residence to her advantage and proceeded to acquire, refurbish and sell her primary residences five times in four years. She argued, correctly, that at any particular time she owned one residence and used it as her primary residence. However, paragraph 45 of the Eighth Schedule, after providing in subparagraph (1)(b) that the gain on disposal of a primary residence is excluded from CGT if the proceeds do not exceed R2 million, then provides in subparagraph (4)(b) that subparagraph (1)(b) does not apply to the disposal of a primary residence where the taxpayer “used that residence or a part thereof for the purposes of carrying on a trade”. It is submitted that the tax consultant was correct to conclude that the taxpayer was conducting a trade in acquiring, refurbishing and selling residential properties. It did not avail the taxpayer that at each material time whichever property she then owned was her primary residence. Subparagraph (1)(b) prohibited the exclusion. And the tax consultant was wise to decline the appointment. SARS could report the tax consultant to her RCB under section 241(2) of the TAA, which provides that a senior SARS official may lodge a complaint with an RCB “if a registered tax practitioner has, in the opinion of the official-

• Without exercising due diligence prepared or assisted in the preparation, approval or submission of any return, affidavit or other document relating to matters affecting the application of a tax Act;
• Unreasonably delayed the finalisation of any matter before SARS;
• Given an opinion contrary to clear law, recklessly or through gross incompetence, with regard to any matter relating to a tax Act;
• Been grossly negligent with regard to any work performed by a registered tax practitioner;
• Knowingly given false or misleading information in connection with matters affecting the application of a tax Act or participated in such activity; or
• Directly or indirectly attempted to influence a SARS official with regard to any matter relating to a tax Act by the use of threats, false accusations, duress or coercion, or by offering gratification as defined in the Prevention and Combating of Corrupt Activities Act 2004 (Act 12 of 2004)”

The tax consultant would be vulnerable under all but the last of these if she permitted the taxpayer to claim the primary residence exclusion, on the facts before her. She has therefore acted correctly in declining to act for the taxpayer.

The taxpayer operates a loyalty programme known as Clicks ClubCard. Membership is voluntary and in no way limits the freedom of customers to shop wherever they choose. When a customer presents a membership card when making a purchase, one loyalty point is earned and recorded in the records of Clicks for every R10 spent. At quarterly intervals, a customer who has accumulated at least 100 points is awarded a voucher for R10 for each 100 points. Vouchers may not be exchanged for cash, but may be redeemed against the cost of subsequent purchases.

During the 2009 year of assessment, the taxpayer claimed an allowance of about R44 million to be deducted from its gross income, calculated on the basis of the cost of sales to the taxpayer in honouring vouchers that the taxpayer expected customers to redeem in the following tax year. SARS disallowed the claim and rejected the taxpayer’s objection to the disallowance.

The taxpayer succeeded on appeal to the tax court, for the following reasons:

• it was artificial and factually incorrect to regard the taxpayer’s expenditure as arising from a different contract from the first purchase and sale contract that had occasioned the customer’s acquisition of the points;
• the first purchase and sale agreement triggered the both the earning of income by Clicks and an obligation on the taxpayer to incur future expenditure;
• the obligation to incur future expenditure was therefore incurred under the same contract from which the income was earned and the requirements of section 24C were met.

Section 24C provided in 2009 that for the allowance for future expenditure to apply:

1. SARS must be satisfied that an amount of expenditure “will be incurred” after the end of the year;
2. in a manner that the amount will be deductible in a subsequent year of assessment; or in respect of the acquisition of an asset in respect of which any deduction will be admissible under the Act;
3. the income of a taxpayer in any year of assessment includes or consists of “an amount received by or accrued to him in terms of any contract”, and SARS is satisfied that all or part of the amount will be used to finance future expenditure which the taxpayer will incur “in the performance of his obligations under such contract”.
4. The allowance must be added back to income in the following year of assessment.

The most recent decision on section 24C was CSARS v Big G Restaurants (Pty) Ltd [2018] 81 SATC 185 SCA, in which the court held that the income and the expenditure must arise from the same contract. It does not avail the taxpayer if two contracts are “inextricably linked”. The operative concept is “contract”, not “scheme” or “transaction”.

SARS contended that there were at least three contracts: the ClubCard contract, which was issued free of charge and gave rise to no income in the taxpayer’s hands; the first contract of purchase and sale, when the customer bought merchandise from the taxpayer and triggered the award of points under the ClubCard contract; and the second contract of purchase and sale, when the customer bought merchandise and was entitled to redeem the voucher. The points awarded arose from the ClubCard contract. So the probable future expenditure arose from the points awarded under the ClubCard contract.

The taxpayer contended that the only issue for determination was whether or not the first contract of purchase imposed an obligation on the taxpayer, as the tax court had found. There was, according to the taxpayer a “direct and immediate connection” between each qualifying contract of sale and the obligation on the taxpayer to issue rewards to the customer. The ClubCard contract itself did not create or impose on the taxpayer any exigible obligation to grant any rewards on the taxpayer. The conclusion of a qualifying purchase not only brought into existence an exigible obligation on the taxpayer to issue rewards, but also determined the content of that obligation, with reference to the value of the qualifying purchase. It followed that the “same contract” requirement was met.

The court stated that the ClubCard contract establishes the right of the customer to receive points and then vouchers redeemable against subsequent purchases. This was how the taxpayer described the position in its reply to the SARS enquiry. When it came to the objection, however, the taxpayer shifted its ground. It continued to say that the expenditure was incurred in performing its obligations under the loyalty programme, but started to equivocate regarding the relationship between this and the contracts that generated the rewards. It stated that “there is no separate contract of purchase and sale relating to the goods purchased – the customer’s presentation of the ClubCard when paying at the till-point being inextricably interwoven with and an integral part of each purchase and sale of goods transaction entered into by the ClubCard customer”.

The phrase “inextricably linked” was the kiss of death for the taxpayer. The court referred to the decision in Big G, where the SCA had rejected this concept in relation to section 24C. The income from the first sale contract would be used to finance the acquisition of stock for future sales, thus enabling the taxpayer to meet its obligations under the second sale. The contract that created the right to income was the first sale. The contract that obliged the taxpayer to honour the vouchers was not the first sale, nor the second sale, but the ClubCard contract, a different contract from either of the sale contracts.

The taxpayer’s argument had as its object the reduction of the contractual relationship with a customer to a single qualifying contract of sale, which is both income-earning and obligation-imposing, because the taxpayer is obliged to award points to the customer. This argument ignored the reality of the arrangement, in which three contracts are operative under the loyalty plan. Consequently, the taxpayer did not have access to the section 24C allowance.

After the unanimous decision of the court, delivered by Dlodlo JA, Wallis JA, the author of the definitive judgment on interpretation in Natal Joint Municipal Pension Fund v Endumeni Municipality [2012] ZASCA 13, dealt with the decision in Big G, to show, according to the learned judge, why the outcome of Big G’s current appeal to the Constitutional Court would not affect the present judgment, and why the clear link between “a contract” and “such contract” in section 24C was fatal for the taxpayers in both Big G and the present matter. In doing so he, not surprisingly, applied the principles he had summarised in Natal Joint Municipal Pension Fund, by considering the reason for the introduction of section 24C in 1980. This was to provide relief to taxpayers who in the ordinary course of their operations would be required to make provision for replacement of machinery and equipment in order to keep their operations up to date. The learned judge referred to situations where construction companies receive upfront payments from clients to enable them to obtain the necessary equipment and materials to commence a contract.

In the case of businesses such as Big G, sensible management would in any event dictate that the external appearance in internal décor be regularly refurbished, regardless of whether or not the undertaking was operating under a franchise agreement. To allow a provision for such future expenditure would be to permit the deduction of expenses before they had been incurred, which would offer taxpayers a means of manipulating the timing of tax payments. The SCA had found in Big G that the sale of meals and the expenditure incurred on refurbishment arose from two different contracts. And this was the same argument that applied in the present matter. When a customer buys goods from the taxpayer and leaves the shop, that is the end of the transaction. It is only later, when the customer returns to the shop and makes another purchase, that the loyalty points awarded in terms of the ClubCard contract, and based on the value of the first transaction, come into operation. And the taxpayer’s obligation under the second transaction arises only from the need to acquire the goods necessary to conclude the second sale contract. If one views the matter from the perspective that the loyalty programme is merely an undertaking to offer the customer a discount on the next purchase, that hardly qualifies as expenditure contemplated under section 24C.

The essence of the argument was whether there was a sufficiently close connection between the expense and the taxpayer’s production of income. A secondary question, that of prescription, would arise should the court find against the taxpayer.

The facts were not in dispute. The taxpayer had implemented the share incentive scheme in 2004, the object being to afford employees the opportunity to participate in the scheme in order to promote the growth and profitability of the group. The selected employees were key managerial staff; the contribution was for the purposes of the scheme; the employees did benefit; the contribution was not capital in nature; the scheme was legitimate and its agreements, implementation and transactions not simulated. On 30 November 2004 the Trust was established. The sole beneficiary was Spur Holdco as to capital, shares and income. The Trust then acquired the share capital of a shelf company, Newco. On 7 December 2004 the taxpayer concluded a contribution agreement with the Trust in terms of which it made the R48,5 million contribution to the Trust. The trustees were obliged to use the contribution to subscribe for preference shares in Newco, redeemable only after five years and carrying a coupon rate equivalent to 75% of SA prime. Newco in turn used the subscription price to purchase 8,2 million shares in Spur Holdco, the listed company of the group. The participating employees were offered ordinary shares in Newco at par in proportions determined by Spur Holdco. They were not entitled to deal in their shares for at least seven years after issue, and the shares of any employee who left within that period were forfeited and used for later allocation to other employees.

Newco made no distributions during the five year term of the preference shares, so the participating employees became entitled to the accumulated growth in the Trust’s holding in Spur Holdco. By the end of the five year period, the trust was entitled to accumulated dividends of R22,5 million. This was settled by transferring to the Trust the equivalent value of Spur Holdco shares. Newco then redeemed the preference shares and disposed of the balance of its Spur Holdco shares. Out of the proceeds Newco paid dividends of R28,2 million and R635 000 to the participating shareholders in 2009 and 2011 respectively.

On 13 December 2010 the scheme was terminated and the participating employees became discretionary dividend beneficiaries of the Trust. The taxpayer’s R48,5 million, not being repayable to the taxpayer, vested in Spur Holdco, as did the preference share dividends of R22,5 million. It appears from the narration, although not expressly stated, that the trustees distributed these two amounts to the vested beneficiary. Thus the scheme came to an end.

In disallowing the deduction, SARS contended that the participating employees had not benefitted from the R48,5 million distribution. The only beneficiary was the sole vested beneficiary, Spur Holdco. Only if the participating employees had benefitted directly from the contribution could the expense qualify as a deduction under section 11(a) and thus as an expense incurred in the production of income.

The tax partner at Spur Group’s auditors had given evidence to the tax court and explained the rationale behind the scheme as devised. Under cross-examination by counsel for SARS as to why this scheme had been used and not the simpler provision of loans to the participating employees, he explained that loans can be a disincentive if, as often happens, the increase in value of the shares fails to keep pace with the capital amount of the loan and accumulated interest. The employee bore the risk of the share price falling to the point where the loan liability exceeded the amount the employee could obtain from selling the shares. Making the employees, in effect, dividend beneficiaries removed this risk. It seems that what the tax partner was getting at, without expressing it in these terms, was that the whole scheme had to be looked at holistically. The use of the Trust and Newco and Spur Holdco, and the preference share issue, and the acquisition of Spur Holdco shares, were all parts of a scheme designed to provide incentives to participating employees, to the benefit of the taxpayer and the employees, while mitigating the risk of loss to the employees.

The chief financial officer of the Group was one of the participating employees. She explained that, as a service oriented business Spur expected its employees to work irregular hours, not the usual 8 to 5 regime. They needed to be rewarded for this inconvenience. Spur was a very dividend rich company and in order to derive profits it needed an enthusiastic, committed and competent workforce. She affirmed that her participation in the scheme contributed significantly to her desire to remain employed at Spur.

The court proceeded to traverse the familiar principles from case law relating to deduction of expenditure, beginning with the locus classicus, Port Elizabeth Electric Tramway Co Ltd v CIR [1936] 8 SATC 13 CPD. The question is twofold: (a) whether the act to which the expenditure is attached is performed in the production of income; and (b) whether the expenditure is linked to it closely enough. The learned judge, Watermeyer J as he then was, went further to point out that the expenditure itself need not be necessary in order to earn income; the purpose of the act entailing the expenditure must be looked to. If it is performed for the purpose of earning income, then the attendant expenditure is deductible.  The learned judge concluded: “all expenses attached to the performance of a business operation bona fide performed for the purpose of earning income are deductible whether such expenses are necessary for its performance or attached to it by chance or are bona fide incurred for the more efficient performance of such operation”.

In CIR v Genn & Co (Pty) Ltd v CIR [1955] 20 SATC 113 AD, the Appellate Division referred with approval to Port Elizabeth Tramway and introduced “the closeness of the connection between the expenditure and the income earning operations” as a means of applying the test.

More recently, in CIR v Pick ‘n Pay Employee Share Purchase Trust [1992 65 SATC 346 AD the court stated that in a tax case one is not concerned with what possibilities the taxpayer foresaw and with which he reconciled himself. “One is concerned with his object, his aim, his actual purpose”.

This succinct summary was quoted with approval in Warner Lambert SA (Pty) Ltd v CSARS [2003] 65 SATC 271 SCA, a case that the tax court used extensively.in the present matter. Warner Lambert was the South African subsidiary of the US pharmaceutical giant. In terms of the Sullivan Code principles developed during the apartheid era, local operations of US companies had to permit no discrimination in the workplace and had to incur significant expenditure on social responsibility (SR) projects. The case revolved around whether the SR expenditure was deductible. The court found that it was incurred for the purposes of trade and for no other, because without access to the products and formulas of the US parent its income would have dried up. The SR expenditure had not added to the subsidiary’s income earning structure, which was complete. The SR expenditure had been incurred in order to protect its earnings. The SCA regarded these payments as similar to insurance premiums.

The tax court found that, on the evidence, the dominant purpose of the scheme “was to protect and enhance the business of the taxpayer and its income by motivating its key staff to be efficient and productive and remain in the taxpayer’s employ”. The taxpayer had incurred the expenditure for the purpose of earning income. The majority of the High Court fully aligned itself with these remarks. It acknowledged that the bulk of the benefit inured to the Spur Group, but that did not detract from the actual purpose of the expenditure as affirmed in the evidence. “It, in fact, is quite clear that maintaining a contented and motivated workforce forms part of the costs of performing the income producing operations and is crucial to the Spur Group’s commercial success and profitability”.

Perhaps unconsciously, the court was echoing a 1978 decision of the Swaziland Court of Appeal in COT v Swaziland Ranches Ltd [1978] 40 SATC 232 SwCA, where the court had to interpret the meaning of “buildings used in connection with farming operations” as provided for in the Swazi tax legislation. The majority of the court found that the expenditure incurred in erecting a school solely for the use of the children of employees and two beerhalls for employees was incurred to achieve a happy and contented workforce. As a result, the buildings were used in connection with farming operations.

There is thus strong argument that, on the correct facts, expenditure incurred to motivate and achieve a satisfied workforce can meet the general deduction formula test. The judgment also validates the use of employee incentive schemes, provided they are properly established and operated.

It is necessary to add a note of caution. Salie-Hlophe J issued a strong minority judgment, finding for SARS, fundamentally on the basis that the R48,5 million had not been incurred as contemplated in section 11(a) but had merely flowed through the Group as a vehicle through which to create the dividends, which in turn were the incentive. Because of her view, the learned judge was obliged to consider the prescription question. She stated that she would have found that there had been misrepresentation of material facts in the tax returns in that Spur had answered “no” to certain questions. The judgment does not indicate which these questions were. The fact that the financial statements had accompanied the tax return did not avail the taxpayer. Therefore, she would have found that the three year prescription period provided for in section 99 of the Tax Administration Act did not apply.

Given the strong minority judgment, and the amounts involved, SARS might well launch a further appeal to the Supreme Court of Appeal.

The taxpayer is a member of the Atlas Copco Group, with its parent company in Sweden. In terms of group policy, its main business is to sell or lease, and subsequently service, machinery and equipment, including spare parts and consumables, imported mainly from Sweden and used in the mining and related industries. Group policy, known as the Finance Controlling and Accounting Manual (FAM), required the taxpayer to write down the value of its trading stock by 50% if it had not sold in the preceding 12 months and by 100% if it had not sold in the preceding 24 months.

The taxpayer duly applied this policy and reflected the resulting values in its tax returns for the 2008 and 2009 years of assessment. SARS took the view that this policy did not comply with the provisions of section 22(1)(a) of the Act, which provides for the carrying value to be reduced to the extent that it had been diminished by reason of “damage, deterioration, change of fashion, decrease in market value or for any other reason satisfactory to the Commissioner”. In the first appeal, the tax court had identified what it called the crisp legal dispute between the parties as being “whether the nett realisable value (‘NRV’) of the Atlas Copco SA’s closing stock, calculated in accordance with IAS2, IFRS, South African Statements of Generally Accepted Accounting Practice (‘SA GAAP’) and the policy, may and should, where it is lower than the cost price of such trading stock, be accepted as representing the value of trading stock held and not disposed of at the end of the relevant years for purposes of section 22(1)(a) of the Income Tax Act”. The tax court found that the policy led to a “sensible and business-like result” and was a “just and reasonable basis” for valuing the taxpayer’s stock as contemplated in section 22(1)(a). In doing so it had followed the decision of the tax court in Volkswagen, but without the benefit of knowing that the SCA had subsequently reversed this decision.

On appeal by SARS, the SCA noted that section 22(1)(a) is concerned with the value of a taxpayer’s trading stock at year end. SARS has the power to allow a deduction in the four circumstances specified in the section: damage, deterioration, change of fashion, decrease in market value or for any other reason satisfactory to the Commissioner. As it had done in Volkwagen, the court observed that this provision is couched in the past tense; in other words, the diminution in value must have already occurred. The exercise was thus one of looking back at what had happened during the past year of assessment.

The taxpayer’s case was that the reference to market value in section 22(1)(a) is the same as NRV as employed in the accounting statements. However, the SCA traversed its decision in Volkswagen, summarising the five “important observations” made by Wallis JA:

• whilst annual financial statements prepared in accordance with a group’s accounting handbook serve a valuable purpose in providing a true picture of the company’s financial affairs, they are not necessarily equally applicable to the determination of the taxpayer’s tax liability;
• although there is some scope for overlap, not all the elements of accounting standards relate to the same matters as the section;
• the determination of NRV is based on an assessment of future market conditions. It is forward looking, which has the result that future expenditure is taken into account and becomes deductible a prior year;
• whether NRV reflects a diminution of value of trading stock for purposes of section 22(1)(a) depends, not on its acceptance as part of GAAP, but on its conformity to the requirements for diminution in value as determined on a proper interpretation of the section; and
• the fiscus is concerned with the valuation of trading stock, the question being whether trading stock as a whole had suffered a diminution in value. [This remark had raised some unease in tax circles, suggesting as it did that section 22(1)(a) does not call for an item by item consideration of the value of trading stock].

Accordingly, found the court, the tax court in the present matter had erred as had the tax court erred in Volkswagen.

Atlas Copco, by its own admission in evidence, had not considered whether there had been a diminution by reason of any of the criteria mentioned in section 22(1)(a). It had merely applied the group’s policy on aging without regard to any other factor. This purely time-based approach was not entirely consonant with the requirements of section 22(1)(a). It was an arbitrary, fixed and rigid company policy and “did not present the most reliable evidence available at the time in respect of any diminution in value”. The taxpayer’s auditor testified that they had identified only three product lines that had been sold below cost during the year, and these at between 24% and 26% below cost. As the court commented, this is a far cry from the application of 50% or 100% in terms of the policy. In fact, the auditor conceded that the group policy was “a very aggressive policy”.

As the court stated, this evidence indicated that the taxpayer’s approach to the valuation of its trading stock was flawed and ordinarily this would have been dispositive of the appeal. However, it was necessary to deal with each of the six categories that made up the taxpayer’s trading stock.

Slow moving and overstock categories. The taxpayer asserted that it operated in the mining sector and had to meet orders at relatively short notice. For this reason it had surplus stocks of various items from time to time. The FAM policy was applied to these items, not because they had deteriorated, but because they had been on hand for longer than the group’s 12-month or 24-month policy. There was no indication of any diminution in these items; it was “at best an unmotivated guesstimate” as to whether there would in future be demand for them.

Goods in transit. These were goods that for whatever reason had to be returned to the Swedish parent company. On the evidence of the taxpayer’s witness, the taxpayer relied on an unsubstantiated estimate of the value of these stocks.

Demonstration items. Although the taxpayer’s witness conceded that these ought to have been recorded in a fixed asset register, they were kept as part of inventory and valued for tax purposes at 50% of cost, save for items that could not be located; these were reduced by 100%. The evidence suggests a less than satisfactory treatment of these items.

Dynapac stock. This consisted of road construction heavy equipment and spare parts. This had been acquired in October 2008, a mere two and a half months before the financial year end. Under the “enormous pressure” of year end preparations, the taxpayer had simply followed Dynapac’s valuation policy and written assets off by 100%. There was no evidence that the diminution conditions of section 22(1)(a) had been considered, let alone applied.

Standard cost items. These comprised items acquired from the parent company, only to be notified by the latter later during the tax year of a price increase or decrease. The taxpayer was unable to tender any satisfactory evidence as to the value of the stock at year end.

It was apparent to the court that the taxpayer’s approach in respect of all the stock categories was that, because it held thousands of items of stock at year end, it was not feasible to value each item individually. The tax court had accepted this explanation in support of the proposition that the legislature could not have intended that a trader should assess every item of closing stock in these circumstances. The SCA found that this acceptance was misplaced. SARS had never contended that the taxpayer had to assess each item individually. SARS had accepted that the practice of sampling is a well-recognised method of dealing with high volume trading stock. However, this was not what the taxpayer had done in the present matter.

For all the reasons given, the decision of the tax court could not stand. The appeal was upheld with costs, including two counsel.

The taxpayer had also appealed against SARS’ imposition of interest under section 86quat of the Act for underestimating provisional tax based on the taxpayer’s estimate of taxable income. The court found no warrant for remitting this interest.

The SCA has now twice within a year confirmed that the application of accounting standards to valuing trading stock, without reference to the four circumstances in section 22(1)(a) of the Act, is not acceptable for tax purposes. Unfortunately, however, the court did not clarify the aspect of the Volkswagen judgment that has caused the unease in tax circles, and which appears in (e) above. In fact, all the court did was to refer without comment to the statement of Wallis JA at [46]: “However, I can see no reason for the Commissioner to accept that Volkswagen’s trading stock had diminished in value on the basis of a calculation where Volkswagen took advantage of the ‘swings’, where the NRV was lower than cost price, but disregarded the ‘roundabouts’, where the reverse was true. For tax purposes the question was whether Volkswagen’s trading stock as a whole had suffered a diminution in value” With the utmost respect, if by ‘roundabouts’ the learned judge was referring to increases in the NRV of some items of trading stock, these are irrelevant for purposes of section 22(1)(a). The section is clear: if the value of an item of trading stock has fallen below its cost price, in consequence of one or more of the circumstances contemplated in section 22(1)(a), its carrying value for tax purposes is the lower figure. If the NRV of the item has increased above cost, its carrying value is the cost price.

However, the court in Atlas Copco accepted SARS’ statement that sampling is a well-recognised, and by implication acceptable, method of dealing with high volume trading stock. The wider implication is that blanket valuation methods are not acceptable, if trading stock is not high-volume, an item by item, or at least by groups of stock of similar nature and value, approach is required.

[1] (834/2018) [2019] ZASCA 124

[2] (1028/2017 [2018] ZASCA 116

For the 2008, 2009 and 2010 years of assessment, VWSA had closing stock consisting of unsold vehicles.  Some of these were manufactured or, in the case of trucks and buses, assembled at its Uitenhage plant, while others were imported, and there were some second hand vehicles drawn from its own fleet.  In terms of section 22(1)(a) of the Income Tax Act, 1962 (Act), VWSA was required to determine the value of this stock for tax purposes.  It did this by determining the net realisable value (NRV) in accordance with IAS-2 and the IFRS Handbook, which yielded an amount less than the cost price of the trading stock.  VWSA it claimed a deduction from the cost price of the trading stock represented by the difference between that and NRV.

SARS conducted a lengthy audit of VWSA’s tax affairs for the years in question, at the end of which SARS rejected VWSA’s calculation of the value of closing stock.  This resulted in the issue of additional assessments reflecting substantial increases in the value of the stock, with corresponding increases in the taxable income.  It was against these additional assessments that VWSA had objected and successfully appealed in the tax court.

The SCA pointed out that the direct relationship between the value placed upon trading stock on hand and taxable income on the other enabled taxpayers to manipulate their taxable income upwards or downwards as it suited them.  For this reason section 22(1)(a) provided that a taxpayer who claimed that the NRV of an item of stock was less than its cost price would have to satisfy SARS that this claim was acceptable.  The criteria for so deciding were set out in the section, these being: damage, deterioration, change of fashion, decrease in the market value or for any other reasons satisfactory to SARS.

Section 22(3) provides that the taxpayer may add to the actual price paid for the goods, the costs incurred in getting them into their current condition and location, and any further costs required to be included in terms of any generally accepted accounting practice approved by the Commissioner.  The generally accepted accounting practice contended for by VWSA, namely IAS-2 read with the IFRS Handbook, was to take into account certain categories of costs, described generally as rework/refurbishment costs; outbound logistics; marine insurance; sales incentives; distribution fees; warranty costs, costs relating to the “Audi Freeway Plan” and the “Volkswagen AutoMotion Plan” and roadside assistance costs.  Did these expenses add to the costs of the trading stock?

Eksteen J, presiding in the tax court, had decided that the NRV determined in terms of IAS-2 was an appropriate method by which to determine the value of trading stock for purposes of section 22(1)(a).  He stated: ”In all the circumstances, whereas section 22(1) is silent as to the manner of valuation of trading stock at the conclusion of a year of assessment in order to determine whether a diminution in value has occurred the adoption of the NRV as a method of the assessment of value provides a sensible, businesslike result which accords, in my view, with the purpose of section 22(1) in the context of the Act and with the weight of authority.”

In the SCA, Wallis JA identified and then discussed the four circumstances that could lead to a diminution in the value of trading stock below cost price before proceeding to discuss each in turn:

Damage, deterioration, change of fashion or decrease in market value.  To these must be added other circumstances satisfactory to the Commissioner.  The court drew attention to what it described as an important aspect of the language used, namely that it was couched in the past tense: “an amount by which the value of the trading stock has been diminished”.  All the events that may have contributed to the diminution must have happened in the past.  This did not exclude entirely the element of futurity.  “An example might be knowledge that a glut had built up in the market for a perishable commodity, where that glut would ensure a marked, certain and unavoidable decline in the price of that commodity in the following year”.

However, expenses in making the goods marketable, such as rectifying minor damage incurred in transit, packaging costs, transportation expenses to the point of sale, fees and commissions to retailers, advertising costs and some allowance for post-sale remedying of defects were not relevant to the cost price of the goods.

Counsel on both sides agreed that the baseline for determining whether any diminution in value had occurred was the cost.  The question was whether some event or agency reduced the value to below cost.  An example was a catastrophic fall in the price of wool, as in CIR Jacobsohn 1923 CPD 221, where the taxpayer would never have been able to recoup his cost, let alone realise a profit, on sales of his stocks of wool after the price had plunged after he had acquired his stocks.  The court offered similar examples, such as a purveyor of swimming trunks or briefs (scants?) when “baggies” became the attire of choice.  Or the dramatic rise and decline in the popularity of Blackberry and Nokia phones, which may have resulted in retailers being left with stocks that consumers had rejected as they moved on to smartphones.

And it was here that tax law diverged from accounting standards.  The court accepted that the International Financial Report Standards issued by the International Accounting Standards Board served a valuable purpose in providing a fair picture to investors, shareholders and creditors of companies about their financial affairs.  In doing so, it was important that the picture was fair, in regard to both the past trading activities of the company and to its future prospects.  In fact, it might be more important to know that prospects for the year ahead were gloomy, than that the company had made substantial profits in the year past.  This was why annual financial statements contained many forward looking statements and why IAS-1 on the Presentation of Financial Accounts required management to assess the company’s ability to continue as a going concern.  The auditor was required to assess the appropriateness of management’s use of the going concern basis of accounting and to identify any material uncertainty that may cast significant doubt on the company’s ability to continue as a going concern.

IAS-2 required that financial statements reflect the estimated selling price of inventory in the ordinary course of business, less the estimated costs of completion and the estimated costs necessary to make the sale.  This was not what section 22(1) postulated.  As a result, the items in dispute, because they had not actually been incurred by the end of the year, did not fall to be taken into account to reduce the closing stock value of the trading stock.

VWSA had classified the items forming part of its NRV calculations as “distribution and selling costs”, consisting of rework and refurbishment costs, outbound logistics, marine insurance, distribution fees, sales incentives, warranty costs, costs relating to the Audi Freeway Plan and the Volkswagen AutoMotion Plan and roadside assistance costs.  It was apparent, in the view of the court, that nearly all these costs would have been incurred in the pursuit of or following sales.  The sole potential exception was damage to vehicles in stock having suffered damage during the year, which was a very minor contributor to the calculations, being a mere R525 per vehicle.  This basis of calculating the NRV of stock on hand accorded with IAS-2; the question was whether it accorded with income tax principles as contemplated in section 22(1).

The court found that the use of NRV was inconsistent with two basic principles underpinning the Income Tax Act.  The first was that taxable income is levied from year to year on the basis of events during each year.  Trading prospects for the ensuing year are not SARS’ concern.  This could be expressed by saying that tax is backward looking.  By contrast, NRV is forward looking.  The second inconsistency was that using NRV meant that expenses incurred in a future year in the production of income accruing in that year became deductible in a prior year.  The tax court had erred in failing to recognise that section 22(1) is not concerned with contrasting cost price with a value determined by “an appropriate method by which to determine the actual value of trading stock in the hands of the taxpayer at the end of the year of assessment”.  Whether using NRV was a sensible and businesslike manner of valuing trading stock was neither here nor there.  The question was whether the result of using NRV accurately reflected the diminution in value of trading stock contemplated in section 22(1).  The court found that only the very minor item of R525 per vehicle for refurbishment necessitated by damage could possibly qualify as diminution in value.

Accordingly, the appeal succeeded and the additional assessments for 2008, 2009 and 2010 were confirmed.

I conclude with one slightly disquieting remark of the court.  It seemed to suggest that trading stock need not be valued on a line by line basis.  With respect, this cannot be correct.

Subject to this reservation, this decision is, with respect, correct in terms of established income tax principles. Our courts have long held that accounting treatment is, at best, some indication of the taxpayer’s intention, but it is not determinative of the nature of a transaction.  Seventy years ago, in Sub-Nigel Ltd v CIR [1948] 15 SATC 381 AD, the Appellate Division, as it then was, made this point.  The same court reiterated this principle 17 years later in SIR v Eaton Hall (Pty) Ltd [1975] 37 SATC 343 AD.  Since then, and by coincidence, given the fact that the VWSA matter was about trading stock, at least two further judgments have made the same point in dealing with trading stock issues:  Richards Bay Iron & Titanium (Pty) Ltd & another v CIR [1995] 58 SATC 55 AD, a case who main claim to a place in the pantheon of celebrated tax cases is its analysis of the definition of “trading stock”; and CSARS v AA the Motorist Publications (Pty) Ltd [2001] 63 SATC 325 CPD.

Attempts to apply the provisions of IFRS in tax matters are understandable, given the inevitable complexity of the calculations required to arrive at values consistent with IFRS and perhaps understandable reluctance to do them all again for tax purposes.  Moreover, IFRS itself appears in the Income Tax Act, two examples being section 24J and 24JB.  One could comment on whether it is appropriate for the decisions of an unelected group of accounting specialists located far from South Africa to have any place in our income tax legislation; but that is a debate for another day.

A special trust established inter vivos, as was the case in the present matter, is one created for the benefit of one or more persons with a disability as defined in section 6B of the Income Tax Act, 1962 (Act).  This means a moderate to severe limitation of any person’s ability to function or perform daily activities as a result of a physical, sensory, communication, intellectual or mental impairment that has lasted or has a prognosis of more than a year.  Clearly, dementia qualifies under this definition.

In addition to the applicant, there were other beneficiaries of the trust.  The trust was discretionary as to all beneficiaries.  The fact that there were other beneficiaries did not affect the trust’s status as a special trust, because their discretionary right would come into operation only on the death of the applicant.  This is an important point for estate planning purposes; one may establish a special trust that continues seamlessly for the benefit of the remaining beneficiaries without the need to terminate the trust and create a new one for those beneficiaries by establishing two classes of beneficiary and only the disabled person having rights until that beneficiary’s death.

The purpose of the trust was to provide for the applicant’s care and maintenance when she was no longer able to do so.  SARS ruled that the distribution to the trust was not a donation as contemplated in sections 54 and 55 of the Act.

This decision is interesting, because it did not state what the distribution was if it wasn’t a donation.  It seems also that the drafter of ruling was confused, because its title was the same as this summary.  The reason for this conclusion would be illuminating – and helpful to estate planners.  Perhaps the reason is that there was a quid pro quo in that the trustees were obliged to care for the applicant when the time came. It is trite law that a payment to a person with a corresponding obligation is not a donation.  An alternative conclusion would have been that the distribution was a donation but that it was exempt from donations tax under section 56(2)(c), a bona fide contribution towards the maintenance of any person as the Commissioner considers to be reasonable.  There is no requirement that the contribution be made directly to the person for whose maintenance it is made.

The dispute between the parties involved both section 24C and section 11(d), the latter presumably relating to the familiar repairs v improvements debate in relation to the nature of the refurbishments.  By agreement between the parties, the current case dealt only with the section 24C aspect.

Section 24C is commendably brief.  Its requirements are that:

• an amount forming part of a taxpayer’s income must accrue to the taxpayer
• during a year of assessment
• in terms of any contract
• and the amount will be used in whole or in part to finance future expenditure
• to be incurred by the taxpayer in the performance of the taxpayer’s obligations under that contract. (Emphasis added)

Where these five conditions are present, the taxpayer is entitled to an allowance in respect of the future expenditure relating to the amount.

The taxpayer operated a number of restaurants in a franchise chain under various franchise agreements between it and the franchisor.  The agreements were identical in all material respects.  The provisions central to the present matter were:

• that the taxpayer actively operate the business, which implied the provision of meals to customers
• that failure to do so was a material breach
• a number of clauses dictating to the taxpayer the branded products it must use, how its restaurant must be constructed in accordance with the franchisor’s requirements, how its staff was to be trained, supervised and even replaced at the franchisor’s election, the prices that it could charge, the monthly franchise fee payable, that it must maintain the restaurant in such manner as determined by the franchisor, upgrade and/or refurbish it, play only pre-recorded music approved by the franchisor, only offer food and beverage items approved by the franchisor, operate during business hours specified by the franchisor, and contribute to the franchisor’s marketing fund in the manner specified
• and a host of other provisions from which it was clear that the franchisor exercised complete control over the operations of the taxpayer.

As Counsel for the taxpayer put it, it was no exaggeration to say that the franchisor exercised almost absolute control over the taxpayer via the franchise agreement.

Perhaps the most crucial provision in the matter was Clause L1.4, which provided that the franchisee agreed “…to upgrade and/or refurbish the Restaurant at reasonable intervals determined by the Franchisor to reflect changes in the image, design, format or operation of…[the franchisor’s restaurant chain]…from time to time and required of new…franchisees subject to approval by the franchisor of detailed plans and specifications for all construction, repair or refixturing in connection with such upgrading or remodelling…’”.

The court considered Interpretation Note 78, which states in relation to the obligation requirement in section 24C that there must be a “high degree of probability and inevitability that the expenditure will be incurred by the taxpayer.”

SARS argued that it is an absolute prerequisite that there be a single contract from which both income accrues or is received and an obligation is created to incur future expenditure.  There could be no quibble with this submission.  SARS then contended that there were in fact two contracts: the contract for the franchise, which included the obligation for future expenditure, and which created the right of the taxpayer to establish and operate the restaurants under the franchise licence and trademark of the franchisor in exchange for payments of franchise fees; and the day-to-day sales of meals to customers, who in turn paid for them and to which the franchisor was not a party.  It was largely out of the proceeds of the second contract, together with possible bank finance, that the franchisee carried out the obligatory improvements.  Consequently, according to SARS, the costs of the improvement obligation did not emanate from the franchise contract.  There were thus two separate contracts, legally independent and separate from each other; no income was received by or accrued to the taxpayer in terms of the franchise agreement; no future expenditure was incurred in terms of the sales transactions to customers; and therefore section 24C did not apply.  Perhaps clutching at straws, Counsel for SARS argued that, although the obligation was absolute, it was nonetheless dependent on the franchisor’s approval and thus conditional.

Counsel for the taxpayer argued that, while it was so that the taxpayer received income from its customers due to sales of meals, the fact of the matter was that it could not provide those meals, and thereby earn its income, without the franchise agreement.  The franchise agreement was the fons et origo of the taxpayer’s income, and here lay the clear, direct causal link.

After traversing a number of cases dealing with section 24C, and others addressing the meanings of key words and phrases, and applying the by now familiar rules of interpretation set out in Natal Joint Municipal Pension Fund v Edumeni Municipality [2012] (4) SA 593 SCA, the learned judge found herself persuaded that, given the language used in the franchise agreement; the context and apparent purpose to which section 24C was directed; Interpretation Note 78; and the authorities to which she had referred on key words and phrases, the franchise agreement and the sales of meals to customers were inextricably linked and not legally independent and separate.  The fact that the taxpayer was obliged under the franchise agreement to operate the restaurant and provide meals to customers, and failure to do so would have been a material breach, evidently weighed with the court.

Judgment thus went to the taxpayer.

As to costs, the taxpayer applied for an order against SARS, on the grounds that it was unreasonable for SARS to suggest that the allowances claimed did not qualify under section 24C.  However, the court did not consider that SARS had been unreasonable in adopting the approach that it did.  Each case in matters such as the present one was very fact specific and, from the decisions of the tax court to which the court had been referred, it seemed clear that there was generally no easy answer to this type of debate.  As a result, there was no order as to costs.

While franchisees of “chain” restaurants will applaud this decision and set about applying it to their situations, much would depend on the provisions of each franchise agreement.  As the court correctly observed, each situation is fact specific and there is no room for a one size fits all approach to the application of section 24C.

The circumstances hark back to the unlamented secondary tax on companies (STC) and in particular a loophole provided by section 44(9) of the Income Tax Act, 1962 (Act) that needed to be closed.

Pienaar Brothers (Pty) Ltd (not the taxpayer) was an operating company.  It needed to introduce a BEE element of ownership into the company.  To this end, acting on legal advice, section 44 of the Act was used.  The taxpayer was acquired in March 2007.  A condition of an amalgamation transaction as applied to the current matter was that the company, the “amalgamated company”, disposed of its assets to the newly acquired company, the “resultant company” in exchange for shares in the resultant company.  The amalgamated company then disposed of those shares, being its only asset, to its shareholders and its existence was terminated.

On 16 March 2007 the taxpayer, then bearing the name Sererebule Trading 15 (Pty) Ltd, concluded the amalgamation transaction, effective from 1 March 2007, in terms of which it acquired the assets of Pienaar Brothers (Pty) Ltd partly in exchange for shares in the taxpayer.  The value of the assets was of course treated as the subscription price for the shares, and credited, to the extent of their par value to the share capital of the taxpayer and the rest, amounting to R29.5 million, to its share premium account.   The reason for the transaction was to introduce a BEE component into shareholding.

On 3 May 2007 the taxpayer distributed the R29.5 million to its shareholders from its share premium account.  At that date, paragraph (f) of the former, and equally unlamented, definition of “dividend” in section 1 of the Act excluded from the definition any amount distributed from the share premium account of a company.  Section 44(9) confirmed that this exemption applied to distributions in terms of an amalgamation transaction.

On 7 May 2007 the BEE shareholders acquired 25.1% of the share capital of the taxpayer, now shorn of its substantial share premium account, for what was presumably a small, affordable amount, thus achieving the BEE objective of the transaction.

The problem that led to the present matter arose when SARS imposed STC on the taxpayer’s R29.5 million distribution to its shareholders.  In other words, SARS denied the exemption provided for in paragraph (f) of the definition of “dividend”.  The taxpayer objected on the grounds that on 3 May 2007 paragraph (f) still applied, as the amending legislation following the annual Budget proposals on 20 February 2007 had not yet been promulgated.

In his Budget speech on 20 February 2007 the Minister announced that certain retrospective amendments would be introduced to deal with certain anti-avoidance arrangements relating to STC.

On the following day SARS issued a press release announcing that the STC exemption for amalgamation transactions contained in Section 44(9) of the Act was to be withdrawn with immediate effect and replaced with section 44(9A), which closed the loophole with effect from that day, 21 February 2007.

The amending legislation was promulgated on 8 August 2007, deleting section 44(9) and in effect ensuring that distributions such as the one carried out by the taxpayer on 3 May 2007 would be subject to STC, with effect from 21 February 2007.  The taxpayer contended that it had never received any formal notice of this retrospective legislation, and it was unconstitutional for legislation to affect transactions already concluded under previous legislation.

The court proceeded with a long and interesting analysis of retrospective and retroactive legislation, which is beyond the scope of this article.  Perhaps the discussion may be very briefly summarised by saying that the court found that the retrospective legislation did not arbitrarily deprive the taxpayer of its property.  Parliament had used a well-accepted mechanism for amending legislation to protect the government’s ability and need for taxes in order to run the country.  Taxpayers had been warned, by the Minister on 20 February, and by SARS on 21 February, and by vigorous discussions between interested parties and SARS and National Treasury, which led to amendments to the original proposals, that the amending legislation was on the way and would be back dated to 21 February 2007.  Consequently, the backdating to 21 February could not be considered to be unconstitutional.

As the court, with respect correctly, pointed out, this is a familiar mechanism.  As far back as December 1988, for example, just as taxpayers were going off for their end of year break, the then Minister of Finance, Barend du Plessis, warned taxpayers that in the amending legislation following the 1989 Budget he was going to introduce two anti-avoidance measures, to take effect from the day of this announcement.  The resultant additions to the Act were section 8E, at that time called preference share schemes, and section 103(5), the interest/dividend swap schemes.  Taxpayers were thus put on notice that between then and the promulgation date of the 1989 amending legislation that they entered into preference share schemes and dividend/interest swaps at their peril.  In 1989 we did not yet have our current constitution, but it is submitted that retrospective legislation with advance warning is both constitutional and an understandable weapon in the hands of the fiscus.

Finally, most recently in their draft response dated 14 June 2017 to comments raised at the hearings of the Parliamentary Standing Committee on Finance, National Treasury (NT) and SARS referred to Pienaar Brothers in their response to complaints that the Budget announcement, published on 22 February 2017, that the rate of dividends tax was to be increased from 15% to 20% with effect from that date, was retroactive and made without a mandate from the legislature.  NT and SARS pointed out that the court in Pienaar Brothers had found that retrospective legislation was acceptable.  Dividends tax is imposed when a dividend is paid in the case of listed companies, and on the earlier of when it is paid or due and payable in the case of unlisted companies.  The response continued:

“South African law distinguishes between retroactive legislation and retrospective legislation. Retroactive legislation means legislation that changes the law with effect from a date in the past, in respect of events or transactions irrespective of whether they occurred before that date, typically where legislation provides that from a past date, the new law shall be deemed to have been in operation. On the other hand, retrospective legislation means legislation that affects an event that occurred prior to the date on which the legislation was promulgated but on or after the date on which the proposed change in the law was first announced.

Applying the above to the given circumstances, the proposed increase of the DT rate from 15 per cent to 20 per cent with effect from 22 February 2017 is not retroactive as it does not seek to tax dividends that were paid before 22 February 2017.  The proposal was effective from the date of the announcement, not from a date in the past.  However, the proposed increase can be viewed as retrospective as it has been implemented before the legislation has been promulgated.  Other proposals in the Draft Rates Bill, such as changes to personal income taxes, can be characterised in the same manner. In fact, most rates and threshold changes take place after the announcement on Budget Day, and begin to be implemented before the tax laws are enacted (normally around December, about ten months after the announcement).  Given the market sensitivity of tax announcement, this practice is the norm in order to ensure that taxpayers do not rush to restructure their tax affairs to lower or avoid paying the full amount of the expected tax.

All over the world, it is not uncommon for taxation measures to be enacted with retrospective operation and for those measures to commence from the date of the budget announcement, rather than the date of a transaction or enactment of legislation.  Generally, there is acceptance that amendments to tax legislation may apply retrospectively, where the Government has made an announcement of its intention to introduce legislation with sufficient detail of the proposal and subsequent legislation providing for commencement with effect from the date of announcement.  It is international practice for countries to accept that retrospective amendments may be appropriate where a retrospective provision (i) corrects an unintended consequence of a provision, (ii) addresses tax avoidance and (iii) might otherwise lead to a significant behavioural change that would create undesirable consequences.”

The dividends tax issue is now water under the bridge, but Pienaar Brothers has confirmed the acceptability of retrospective legislation in appropriate circumstances.

An NED is not defined in the Income Tax Act, 1962 (Act).  According to the King III report, an NED must provide objective judgment independent of management, must not be involved in the management of a company, and must be independent of management on issues such as strategy, performance, resources and diversity.  Put differently, the NED must not countenance undue influence and must show no bias.

BGR 40 takes these concepts and places them into the context of the Fourth Schedule to the Act, which provides the definitions of “remuneration”, “employee” and “employer”.  The definitions are interrelated: an employee is a person who receives remuneration; remuneration is something paid to an employee; and an employer is a person who pays remuneration to another person.

The Fourth Schedule recognises two tests for determining whether a person is an employee.

The first is the so-called common law test, which broadly determines that a person who earns a salary, wage, stipend, commission, fee, bonus or some similar reward for services as an employee.

The second consists of two statutory tests which, even though the recipient is carrying on an independent trade, determine whether the reward for services is remuneration for purposes of the Fourth Schedule.  These are the “premises” test, where the services must be performed mainly at the premises of the client; and the “control or supervision” test, where the client exercises control or supervision over the manner in which the duties are to be performed or the hours of work.  Both statutory tests must be satisfied in order to render the recipient an employee in receipt of remuneration.

A moment’s reflection should lead one to conclude that a genuine NED, one who meets the criteria set out in King III, for example, cannot be an employee but must be an independent contractor.  The prohibitions placed upon employees as to the deductions they may claim against remuneration will consequently not apply to NEDs, who may claim various expenses denied to employees.

Of course, if a person professes to be an NED but the facts indicate otherwise, not only is the company in breach of its governance obligations but it will also be in breach of its obligation to withhold PAYE from amounts paid to the “tame” NED.

Now to BGR 41.  If an NED is not an employee, the question arises as to the nature of the amounts paid to the NED.  BGR 41 deals with this.  The NED, being an independent person, is carrying on an enterprise as defined in the Value-Added Tax Act, 1991 (VAT Act).  Employment can never qualify as an enterprise for VAT purposes but the NED is not an employee.  The next question is whether the NED is required or chooses to register as a vendor under the Vat Act.  The crisp test here is whether the NED is continuously or regularly carrying on the enterprise of an NED.  It is submitted that infrequent or occasional services as an NED would not qualify as an enterprise.  However, an NED who is conducting that office correctly is likely to be conducting an enterprise because regularity and continuity are surely requirements of a genuine NED.

If the NED’s enterprise generates fees in excess of the compulsory registration threshold, currently R1 million, the NED must register.  An NED whose fees are less than the threshold but who nonetheless wishes to register as a vendor may do so provided the fee income has exceeded R50 000 in the preceding period of 12 months.

In summary, an NED whose conduct meets the criteria expected of an independent director is not an employee of the company.  The NED must submit invoices for services and, if necessary based on the monetary thresholds, register as a vendor and levy output tax on fees.

February 2017