1

IN THE HIGH COURT OF SOUTH AFRICA

(GAUTENG DIVISION, PRETORIA)

Case Number: 87760/2014

DELETE WHICHEVER IS NOT APPLICABLE

(1) REPORTABLE: YES/NO.

(2) OF INTEREST TO OTHER JUDGES: YES/NO.

(3) REVISED.

DATE SIGNATURE

In the matter between:

PIENAAR BROTHERS (PTY) LTD APPLICANT

And

COMMISSIONER FOR THE SOUTH

AFRICAN REVENUE SERVICE 1ST RESPONDENT
THE MINISTER OF FINANCE 2ND RESPONDENT

JUDGMENT
 2

Fabricius J,

1.

“Tax legislation is not a promise and a tax payer has no vested right in the Internal

Revenue Code”. (Justice Blackmun in US v Carlton, 512, US 26 (1994)

Does this dictum apply in South African law?

In this opposed application, the Applicant initially sought the following relief:

1. “Declaring Section 34 (2) of the Taxation Laws Amendment Act 8 of 2007

(“the Amending Act”) to be inconsistent with the Constitution and invalid to the

extent that it provides that Section 44 (9A) of the Income Tax Act 58 of 1962

(“the ITA”), which was introduced by virtue of Section 34 (1) (c) of the

Amending Act, shall be deemed to have come into operation on 21 February

2007 and to be applicable “to any reduction or redemption of the share capital

or share premium of a resultant company, including the acquisition by that

company of its shares in terms of Section 85 of the Companies Act, 1973 (Act

No 61 of 1973), on or after that date”;

2. In the alternative to paragraph 1 and in any event:

 3

2.1 Declaring that the provisions of Section 44 (9A) of the ITA did not apply to

the distribution by the Applicant on 3 May 2007, to its registered

shareholders at that date pro rata to their shareholding, of an amount of

R29 500 000 out of Appellant’s share premium account;

2.2 Declaring, in consequence thereof, that the First Respondent’s assessment

of the Applicant on 13 December 2011 to secondary tax on companies

(“STC”), in relation to a dividend cycle commencing on 23 September 2005

and ending on 3 May 2007, and to interest calculated from 1 July 2007 to

the “date payable” (given as 5 January 2012) (“the STC assessment”), was

invalid; and

2.3 To the extent necessary, setting aside the STC assessment;

3 Referring the order sought in paragraph 1 above to the Constitutional Court for

confirmation”.

At the hearing of the application, Mr M. Chaskalson SC suggested that I could insert

a full-stop after the word ”invalid” in the first sentence of the first prayer, and that the
 4

remainder of the prayer was not necessary. After the hearing, Applicant’s Counsel

provided me an amended Draft Order which read as follows:

“It is ordered that:

1. To the extent necessary for the purposes of order 2 below:

1.1 Section 34 (2) of the Taxation Laws Amendment Act 8 of 2007 (“the

Amending Act”) is declared to be inconsistent with the Constitution and

invalid; and

1.2 The order sought in paragraph 1.1 above is referred to the Constitutional

Court for confirmation.

2. It is declared that

2.1 The provisions of Section 44 (9A) of the ITA did not apply to the distribution

by the Applicant on 3 May 2007, to its registered shareholders at that date

pro rata to their shareholding, of an amount of R29 500 000 out of the

Appellant’s share premium account;

2.2 In consequence thereof, the First Respondent’s assessment of the Applicant

on 13 December 2011 to secondary tax on companies (“STC”), in relation to

 5

a dividend cycle commencing on 23 September 2005 and ending on 3 May

2007, and to interest calculated from 1 July 2007 to the “date payable”

(given as 5 January 2012) (“the STC assessment”), was invalid; and

3. The STC assessment is declared invalid and set aside”.

2.

First Respondent’s Counsel Mr W. Trengrove SC, was not in agreement that prayer

2.2 was a competent prayer and as a result thereof further brief Heads of Argument

in that context were requested, and supplied. With reference to Medox v

Commissioner for SARS, 2015 (6) SA 310 (SCA) at par. 13 to 15, it was

contended that no order by this Court should result in precedent whereby taxpayers

are permitted to bypass the specialist tribunal of the Tax Court in favour of a direct

approach to this Court. I agree.

 6

3.

After the Applicant filed its original Heads of Argument, it was apparently advised to

file a further Supplementary Affidavit alleging a new cause of action for the

unconstitutionality of Section 34 (2) of the Taxation Laws Amendment Act 8 of

2007 (“the Amendment Act”), on the basis of its inconsistency with Section 25 (1)

of the Constitution. (The “property clause”)

4.

It was contended that the challenge to the constitutionality of Section 34 (2) would

only arise in the event that the Applicant’s primary “interpretational argument” was

rejected, and that Section 34 (2) was held to have retroactive effect to the

Applicant’s completed transaction. The challenge to Section 34 (2) was based on

the fundamental right to property proceeds on the basis that the retroactive removal

of the exemption STC in paragraph (f) of the definition of “dividend” without

adequate notice, would have amounted to a deprivation of property that was both
 7

procedurally and substantively arbitrary, and thus inconsistent with Section 25 (1) of

the Constitution.

5.

In the Founding Affidavit, Applicant says that the primary purpose of this application

was to declare unconstitutional and invalid a retrospective amendment to the Income

Tax Act 58 of 1962 as amended (“the ITA”). On the basis of such invalidity, and in

any event, (i.e. even if invalidity is not established) on a proper interpretation of said

amendment, the Applicant sought an order the amendment did not apply to the

distribution by it to its shareholders, on 3 May 2007, of an amount of

R29 500 000, in respect of which First Respondent assessed it for secondary tax

on companies (“the STC assessment”). The Applicant accordingly also sought

setting aside of the STC assessment and ancillary relief.

 8

6.

Factual background: The Transaction:

Applicant gave a lengthy explanation of the factual background and the relevant

transaction in the Founding Affidavit. I was informed by Counsel for the parties that

for purposes of their argument, the motive of the Applicant to enter into the relevant

transaction was irrelevant. For present purposes therefore a summary of the facts

suffices and this summary is given by Applicant itself in a memorandum by its

Chartered Accountants to the First Respondent dated 22 September 2011:

1. “Serurubele Trading 15 (Pty) Ltd (“the Taxpayer”) entered into an

amalgamation transaction in terms of Section 44 of the Income Tax Act 58

of 1962 (“the Act”) in which it acquired all the assets of Pienaar Brothers

(Pty) Ltd on 16 March 2007, which acquisition had effect from 1 March

2007 in terms of the Sale of Business Agreement.

2. As part settlement of the purchase consideration the taxpayer issued shares

to Pienaar Brothers at the purchase price less the assumed liabilities,

 9

(“equity consideration”), which equity consideration less the par value of the

shares was credited to share premium account of the tax payer.

3. On 3 May 2007 the Board of Directors of the Taxpayer resolved, in terms of

Section 90 of Companies Act No. 61 of 1973 read with Article 21A of the

Tax Payers Articles of Association, to make a distribution to its

shareholders pro rata to their shareholding, of an amount of R29 500 000

out of the tax payer’s share premium account (“the Distribution”). The

applicable law on 3 May 2007 in the context of the definition of a “dividend”

in Section 1 of the Act meant that a “dividend” excluded from its ambit any

amount distributed out of the share premium account (not being profits

previously capitalised to the share premium account). It was Applicant’s

submission that as at 3 May 2007 when the distribution was made, the

Distribution did not constitute a “dividend” as defined in the Act and no STC

was therefore due and payable by the tax payer on the Distribution as the

Distribution was made out of the share premium account of the tax payer
 10

which share premium arose from the issue of ordinary shares at a premium

over the par value.

7.

In the Founding Affidavit Applicant explained the position as follows:

1. “Applicant was previously known as Serurubele Trading 15 (Pty) Ltd. On 7

August 2007, it changed its name to its current name following the

transactions that will be referred to;

2. Applicant operates a business supplying and distributing personal protective

clothing for use in various industries and it acquired this business on 16

March 2007 with effect from 1 March 2007 as a going concern from the

company then known as Pienaar Brothers (Pty) Ltd (“OLD CO”);

3. It was deemed necessary to introduce a Black Economic Empowerment

Equity partner into the business and accordingly “OLD CO” sought advice

from its Attorneys pertaining to the method to be used to implement a BEE

transaction. It was regarded as important that such BEE partner should be

 11

able to buy into a new company which would take over business from “OLD

CO”, so as to ensure that the new shareholders would not be exposed to any

unexpected historical liability in “OLD CO”. The price of the shares to be

acquired by the new partner also had to be affordable;

4. Applicant’s Attorneys advised “OLD CO” to apply the group restructure

provisions in Section 44 of the ITA (pertaining to so-called “amalgamation

transactions”), which will allow for the achievement of these commercial

objectives in a tax efficient manner;

5. In March 2006 the Applicant (then called Serurubele Trading 15 (Pty) Ltd)

was acquired with a view to bring a vehicle for the envisaged amalgamation

transaction;

6. Based on the advice of the Attorneys, it was envisaged that the Applicant

would acquire the business from “OLD CO” in an amalgamation transaction

and that the BEE partner would thereupon take up an equity’s taken the

Applicant;
 12

7. The relevant agreements were drawn, company resolutions were prepared

and the final Sale of Business Agreement between “OLD CO” and Applicant

was signed on 16 March 2007, with effect from 1 March 2007;

8. All relevant conditions precedent were fulfilled and on 1 April 2007 “OLD

CO” transferred the business to the Applicant. Applicant duly discharged the

purchase price;

9. In the language of Section 44 (1) of the ITA, in this “amalgamation

transaction” “OLD CO” was “the amalgamated company” and the Applicant

was the “resultant company”;

10. One of the requirements of an “amalgamation transaction” is that, as a result

thereof, the amalgamated company’s existence has to be terminated. To this

end, “OLD CO” distributed the consideration shares to its shareholders pro

rata to their shareholding where after “OLD CO” was liquidated;

11. At this point, Applicant was the owner of the business. On 3 May 2007,

Applicant’s directors resolved to distribute to its registered shareholders, pro

rata to their shareholding, an amount of R29 500 000 out of the

 13

Applicant’s share premium account (“the Distribution”). This Distribution was

duly implemented on 3 May 2007;

12. The BEE shareholder had then to be introduced into the Applicant and

accordingly the existing shareholders of the Applicant between them sold

25.1 of Applicant’s issued share capital to Naha Properties (Pty) Ltd;

13. The transfer of their shares to Naha Properties was confirmed by the

Applicant’s directors on 7 May 2007.

8.

The present dispute pertains to the liability of the Applicant to STC on the

Distribution that I have referred to.

9.

I will now follow the course of Applicant’s submissions and argument contained in

the Founding Affidavit read with the written Heads of Argument which were thorough

and detailed, gratefully received.

 14

10.

No STC liability at the time of Distribution:

At the time the Applicant’s directors resolved to make the Distribution, and when the

distribution was effected and finalized, it did not amount to a “dividend” for purposes

of the imposition of STC under the ITA. This was so by virtue of paragraph (f) of

the definition of “dividend” in Section 1 of the ITA in that the distribution represented

a reduction of the Applicant’s share premium count to which the first proviso of the

definition of “dividend” did not apply.

Had the distribution been a dividend for purposes of the ITA at the time it was

made, the Applicant would by virtue of Section 64B (7) of the ITA have been

required to pay STC on the amount of the distribution and to render the associated

STC return by no later than 30 June 2007, the distribution having been made on 3

May 2007.

Despite that, so Applicant says, it has been assessed for STC on the distribution by

virtue of a retrospective amendment of the ITA.

 15

11.

The amendment to Section 44 of the ITA:

The Applicant refers to the following relevant facts:

1. In the budget speech of 20 February 2007 the then Minister of Finance

made reference in general terms, to an intention to pass retrospective

legislation to deal with certain anti-avoidance arrangements relating to STC.

He provided no further detail as to what arrangements were to be addressed

or in what manner;

2. On 21 February 2007 First Respondent issued a press release in terms of

which inter alia the STC exemption for amalgamation transactions contained

in Section 44 (9) of the ITA was stated to be withdrawn with immediate

effect. The particular statement reads as follows: “21 February 2007:

The STC exemption for amalgamation transactions contained in Section 44

(9) of the Income Tax Act, 1962, is withdrawn. This exemption permits a

permanent loss of STC, rather than a deferral of tax, which is the intent of

the amalgamation provisions”. I may say at this stage that it is clear from this
 16

statement that the “exemption from amalgamation transactions” is the target

of the intended reform.

3. Applicant says that this exemption pertained exclusively to the exemption

from STC of the disposal of consideration shares by an amalgamated

company (such as “OLD CO”) to its shareholders. It did not relate in any

way to the exemption in paragraph (f) of the definition of “dividend”, and

therefore had no bearing on any conduct anticipated to be undertaken by the

Applicant;

4. On 27 February 2007, SARS and National Treasury released for public

comment a Draft Taxation Laws Amendment Bill of 2007;

5. In keeping with the press release, the Bill proposed the amendment of

Section 44 of the ITA by the deletions of Sections 44 (9) and (10) thereof,

which amendments would be deemed to have come into operation on 21

February 2007 and would apply in respect of any disposal of an equity

share, or any deemed declaration of a dividend, by an amalgamated

 17

company (i. e. on the present facts “OLD CO” not the Applicant) on or after

the date;

6. The amalgamation transaction, the distribution and introduction of the BEE

partner was completed in early May 2007 as I have pointed out;

7. On 7 June 2007, the Taxation Laws Amendment Bill was published

together with an explanatory memorandum. This Bill no longer proposed

deletion of Sections 44 (9) and (10), but instead proposed the insertion of

Section 44 (9A). The Bill also proposed that the amendment be

retrospective to 21 February 2007.

It is convenient at this stage to refer to the Explanatory Memorandum relating

to the insertion of Section 44 (9A): “As a theoretical matter S. 44

amalgamations should act as a deferral mechanism. All assets and tax

attributes would roll over from the target company to the acquiring company

with the acquiring company subsequently bearing these tax benefits and

burdens. This same theory holds for the Secondary Tax on companies

(STC). The distribution of acquiring company shares in an amalgamation is

 18

accordingly free from STC. However, the profits of the target company do not

roll over to the acquiring company. The net result is often a complete STC

exemption when the acquiring company makes a distribution of former target

company assets.

It has come to Government’s attention that certain private stakeholders are

attempting avoidance transactions that are specifically aimed at exploiting

this gap. In these transactions, a pre-existing target company with substantial

assets and profits is amalgamated into a newly formed company without

assets or profits. The newly formed company then distributes the former

target company assets, but this distribution is free from the STC due to the

lack of profits within the newly formed acquiring company. From the above

anomaly, the proposed amendment inserts Section 44 (9A) which deems

resultant company equity share capital (and share premium) arising from the

amalgamation to be profits not of a capital nature available for distribution to

shareholders to the extent of any profits distributed by the amalgamated

company in terms of subsection (9). The result is that the amalgamated

 19

company’s profits are effectively rolled over to the resultant company, so that

STC remains payable when the resultant company makes subsequent

distribution”.

8. Applicant says that this was the first indication of any amendment that would

impact upon the STC position of an entity in the position of Applicant;

9. On 8 August 2007 the Taxation Laws Amendment Act 8 of 2007 was

promulgated. Section 34 (1) (c) of the Amending Act inserted into Section

44 of the ITA a new Section 44 (9A). The effect of that appears from the

Explanatory Memorandum note that I have quoted. Section 34 (2) of the

Amending Act provided that Section 44 (9A) was deemed to have come

into operation on 21 February 2007 and would be applicable “to any

reduction or redemption of the share capital or share premium of the

resultant company, including the acquisition by that company of its shares in

terms of Section 85 of the Companies Act, 1973 (Act No 61 of 1973), upon

or after that date”.

 20

10. Applicant states that at no stage prior to the conclusion and implementation

of this actual amendment was Applicant placed on any guard by any public

statement by either of the Respondents to the effect that it might be exposed

to an STC liability in relation to the distribution of any amount from its share

premium account, whether with retrospective effect or otherwise.

12.

The audit of Applicant and the resultant assessment:

In January 2011, First Respondent commenced with an audit of the Applicant’s tax

affairs for the 2007 year. On 6 December 2011, First Respondent notified the

Applicant in an “assessment letter” that an adjustment would be made in respect of

STC, and more particularly that STC in the amount of R3 687 500 (12.5% of

R29 500 000) was to be levied on the Applicant. It was also stated that the

applicable dividend cycle for STC purposes was the period ending 3 May 2007. A

formal notice of assessment of STC was issued by First Respondent on 13

December 2011. This assessment reflected the applicable dividend cycle as

 21

commencing on 23 September 2005, ending on 3 May 2007. The assessment

included interest calculated from 1 July 2007 to the “date payable” given as 5

January 2012. The First Respondent relied upon Section 64B (9) of the ITA in

imposing the assessed interest. On 20 February 2012, Applicant delivered an

objection to the assessment. This objection was disallowed by the First Respondent

on 16 March 2012. The Applicant appealed against the assessment following an

unsuccessful attempt to resolve the dispute through Alternative Dispute Resolution

process, the parties filed pleadings in the Tax Court. The First Respondent filed a

statement of grounds of assessment and Applicant filed a statement of grounds of

appeal. One of the grounds of appeal raised by the Applicant was that Section 34

(2) of the Amending Act was invalid, because it infringed the constitutional principle

of legality, insofar as it purports to make Section 44 (9A) effective from a date

earlier than that on which it was enacted. The appeal of this dispute was due to be

heard during October 2014, but the parties agreed that this appeal would be

postponed sine die to enable the Applicant to launch the present proceedings in this

Court.
 22

13.

Applicant states that it emerged during the relevant correspondence that there was a

dispute between the parties as regards the legitimate ambit of the present

application. This dispute will arise only if I find that the retrospective amendment is

not invalid as being inconsistent with the Constitution. In that event, Applicant still

intends to argue that the provisions of Section 44(9A) do not apply retrospectively

to the distribution in the present circumstances. This was the Applicant’s so-called

“second basis”. First Respondent took the view that Applicant’s second basis should

not be entertained by me, or by this Court, but only by the Tax Court in due course.

Applicant’s view is that its second basis involves issues of law and the interpretation

application of statutes in relation to which did not anticipate any factual disputes

arising. Once this Court was seized with the matter pertaining to constitutional

validity of the amendment, so it was put, there was no reason either in law or based

on convenience, for me not to determine the residual issue between the parties. This

involved the interpretation application of statutes and there were numerous examples

of the High Court assuming jurisdiction over such disputes involving tax statutes in
 23

particular. It would be an exercise in futility and a waste of money, so it was put, to

require this dispute to be dealt with piece-meal.

14.

Applicant’s legal submissions:

It was said that the crux of the Applicant’s complaint was that it relates to the

constitutionality of the retrospective amendment in that such retrospective legislation,

which ex post facto deems the law at a particular time to be what it was not, offends

against the principle of legality and the rule of law which lie at the heart of our

constitutional dispensation. The prejudice to the subject of such legislation is only

heightened where, as here, it purports to attach adverse consequences to

transactions which have been completed and arising from which persons have

acquired vested rights before such promulgation. To the extent that the Respondents

would seek to rely upon the public statements referred to above which preceded the

amendment, it would be contended that these are of no relevance in assessing the

legality of retrospective amendment. In any event, on the present facts, no such

 24

public statements involved an amendment of the nature of Section 44 (9A). Even if

the retrospective amendment was not unconstitutional per se, the Applicant

contended, in accordance with its second basis, that it did not apply to the

distribution, either because that transaction was already completed at the time of the

amendment or because Section 44 (9A) was not capable of being applied in a

manner that is fair and practically effective in the context of the ITA as a whole.

15.

Applicant’s argument on the interpretation issue:

It was submitted that Applicant’s attack was not on the content of Section 44 (9A),

but on the purported retroactivity of the amendment. The prime relief sought was an

order of constitutional invalidity. The second order, couched as an alternative to the

first, was to the effect that the provisions of Section 44 (9A) of the ITA did not in

fact apply to the distribution when it was made. The second order was based on

statutory interpretation. Since I would strictly speaking not be required to decide the

constitutional issue if I were to find that the Amendment Act, on a proper

 25

construction, did not apply to the transactionary retrospectively, it would be

convenient to deal first with the interpretation issue.

In Zantsi v Council of State, Ciskei and Others 1995 (10) BCLR 1424 (CC), the

Court decided that where it was possible to dispose of a case without addressing

any potential constitutional issue, that course should be followed. It was also

submitted that whether Applicant succeeded on the constitutional or interpretation

issue, it would not ask this Court to actually set aside the disputed assessment. That

was for the Tax Court to do in accordance with its powers under Section 129 (2) of

the Tax Administration Act 28 of 201.

16.

In the context of retrospectivity of legislation, it was pointed out that South African

case law distinguishes between retrospectivity of legislation in the “weak” and

“strong” sense. A statutory provision is retrospective in the weak sense if it

prospectively effects, or changes the consequences for the future of, pre-existing

transactions and matters. An enactment is retrospective in the strong sense if the

 26

provision is deemed to have been in force from an earlier date than that on which it

was in fact enacted.

See: National Director of Public Prosecutions v Carolus and Others 2000 (1) SA

1127 at 1138 to 1139, par. [33] to [36].

Applicant is of the view that in the present case we are concerned with

retrospectivity in the strong sense, or retroactivity, inasmuch as Section 34 (2) of

the Amendment Act stated that a new Section 44 (9A) would be deemed to have

into operation on 21 February 2007, even though the Amendment Act was only

promulgated on 8 August of that year.

17.

The Court’s power to grant declaratory relief (in the context of the interpretation

argument):

It was submitted that there is ample authority to the effect that the High Court has a

power to determine tax cases pertaining to issues of law.

 27

See: Metcash Trading Ltd v C:SARS 2001 (1) SA 1109 (CC) at par. 44, where

Kriegler J stated: “Indeed, it has for many years been settled law that the Supreme

Court has jurisdiction to and determine income tax cases turning on legal issues…”.

The determination of the proper meaning and ambit of a statute is a question of law.

It was submitted in the present context, that the crucial question is whether the

statute that applies to the facts is “sufficiently clear”, and that is a question of law. It

was submitted that there is no material dispute of fact in relation to any issue that

could impact upon the interpretation question. The nature, content and timing of the

distribution is common cause. The date of promulgation of the Amendment Act is

clear. The legal and financial impact of the amendment on taxpayers in the position

of the Applicant and its holding company is also not in dispute. The Applicant’s BEE

motivation for the transaction is not relevant to the interpretation of the Amendment

Act. The submission was therefore that there is no impediment to this Court

determining the legal issue of interpretation. Relief sought in this context is limited to

a declaratory order, and once the ambit of the law has been established, the Tax

Court will be asked to address the merit of the assessment in that light. I would
 28

therefore not be impinging unjustifiably on the jurisdiction and powers of the Tax

Court.

On ordinary principles, a Court will always retain a discretion whether or not to

entertain an application for declaratory relief.

See: Section 21 (1) (c) of the Superior Courts Act 10 of 2013 and Herbstein and

Van Winsen, The Civil Practice Of The High Courts Of South Africa, 5TH Edition

at p. 1430.

18.

The merits of the interpretation issue:

It was submitted that the Amendment Act had to be interpreted in the same way as

any other statutory provision, and that the question was whether, on a proper

interpretation, the introduction of Section 44 (9A) actually had retroactive effect so

as to render the distribution subject to STC. Reference was made to National

Director of Public Prosecutions v Carolus supra, where the following was said:

“An important legal rule forming part of what may be described as our legal culture
 29

provides that no statute is to be construed as having retrospective operation (in the

sense of taking away or impairing a vested right acquired under existing laws)

unless the legislature clearly intended the statute to have effect: see: Peterson v

Cuthbert and Company Ltd 1945 AD 420 at 430”.

In Bellairs v Hodnett and Another 1978 (1) SA 1109 A, it was said that not only is

there a presumption against retrospective activity, but “even where a statutory

provision is expressly stated to be retrospective in its operation it is an accepted rule

that, in the absence of contrary intention appearing from the statute, it is not treated

as affecting completed transactions …” (at 1148 F – G).

The basis of this presumption was stated in Carolus (supra at par. 36) to be

elementary considerations of fairness which dictate that individuals should have an

opportunity to know what the law is and to conform their conduct accordingly.

Reference was also made to Du Toit v Minister of Safety and Security 2009 (1)

SA 176 SCA in par. 10 with reference to an English decision that “generally there is

a strong presumption that a legislature does not intend to impose a new liability in
 30

respect of something that has already happened, because generally it would not be

reasonable for a legislature to do that…”.

In Carolus (par. 42) the Court also referred to the position in English Law and in

particular to House of Lords in L’Office Cherifien Des Phosphates and Another v

Yamachita-Shinnihon Steamship Company Ltd: The Boucraa [1994] 1 AC 486

([1994] 1 ALL ER 20). In that case the main opinion was delivered by Lord Mustill

who … referred with approval to the following statement by Staughton LJ in

Secretary of State for Social Security and Another v Tunnicliffe [1991] 2 ALL ER

712 (CA) at 724 f to g: “In my judgment the true principle is that Parliament is

presumed not to have intended to alter the law applicable to past events and

transactions in a manner which is unfair to those concerned in them, unless a

contrary intention appears. It is not simply a question of classifying an enactment as

retrospective or not retrospective. Rather it may well be a matter of degree – the

greater the unfairness, the more it is to be expected that Parliament will make clear

if that is intended”.
 31

Lord Mustill continued (at 525 F to H (AC) and 30 e to g (ALL ER): “Precisely how

the single question of fairness will be answered in respect of a particular statute will

depend on the interaction of several factors, each of them capable of varying from

case to case. Thus, the degree to which the statute has retrospective effect is not a

constant. Nor is the value of the rights which the statute effects, or the extent to

which the value is diminished or extinguished by the retrospective effect of the

statute. Again, the unfairness of adversely affecting the rights, and hence the degree

of unlikelihood that this is what Parliament intended, will vary from case to case. So

also will the clarity of the language used by Parliament, in the light shed on it by

consideration of the circumstances in which the legislation was enacted. All these

factors must be weighed together to provide a direct answer to the question whether

the consequences of reading the statute with the suggested degree of retrospectivity

are so unfair that the words used by Parliament cannot have been intended to mean

what they might appear to say”. In that context reference was made to the dictum of

Wallis J in Natal Joint Municipal Pension Fund v Endumeni Municipality 2012 (4)

SA 593 SCA, where the Judge said the following in par. 18: “Interpretation is a
 32

process of attributing meaning to the words used in a document, be it legislation,

some other statutory instrument or contract, having regard to the context provided by

reading the particular provision or provisions in the light of the document as a whole

and the circumstances attendant upon its coming into existence. Whatever the

nature of the document, consideration must be given to the language used in the

light of the ordinary rules of grammar and syntax; the context in which the provision

appears; the apparent purpose to which it is directed and the material known to

those responsible for its production. Where more than one meaning is possible each

possibility must be weighed in the light of all these factors. The process is objective

not subjective. A sensible meaning is to be preferred to one that leads to insensible

or un-business-like results or undermines the apparent purpose of the document.

Judges must be alert to, and guard against, the temptation to substitute what they

regard as reasonable, sensible or business-like for the words actually used. Do so

in regard to a statute or statutory instrument is to cross the divide between

interpretation and legislation…”

 33

Thereafter, and in Bothma-Batho Transport (Pty) Ltd v S. Bothma en Seuns

(Edms) Bpk 2014 (2) SA 494 (SCA), the following was said (par. 2): “Whilst the

starting point remains the words of the document which are the only relevant

medium through which the parties have expressed their contractual intentions, the

process of interpretation does not stop at a perceived literal meaning of those words,

but considers them in the light of all relevant and admissible context, including the

circumstances in which the document came into being”.

19.

In that context it was submitted on behalf of Applicant that while Section 34 (2) of

the Amendment Act expressly makes Section 44 (9A) retrospective to 21 February

2007, it does not expressly state that it affects completed transactions. It is

common cause that all the elements of the amalgamation transaction, including the

distribution, were completed before the Amendment Act was passed. It was

therefore submitted that the mentioned presumptions should prevail, and that

Section 34 (2) should be interpreted as not applying Section 44 (9A) to this

 34

completed distribution. It was contended that I should also have regard to the wider

consequences of an interpretation that purports to render a Section 44 (9A)

retrospectively operative.

The very fact of imposing tax ex post facto tax on a completed transaction, is

prejudicial and unfair to the taxpayer, who has a well-established right to know what

the law is, and to conform his or her conduct accordingly, and to arrange his or her

affairs in a manner that attracts the least tax within the context of existing legislation.

In this case, the prejudice extended to the BEE shareholder who could not have

known that the Applicant had an STC liability when it agreed to purchase the shares.

The Courts have endorsed the mentioned principle and I was referred to CSARS v

NWK Ltd 2011 (2) SA 67 (SCA) par. 42.

It was further contended that a Court must pay particular close attention to the

practical impact of retrospectivity on the legislative scheme as a whole. The

legislator was most unlikely to have intended a provision to operate retrospectively

where, apart from purportedly altering the law in the particular respect, it does not at

the same time make the consequential modifications, and establish the practical
 35

legislative machinery, that will allow the retrospective rule to be practically, fairly and

sensibly implemented. The present amendment fails this test according to Applicant,

in that the provisions of Section 64B (which deals with STC) in various respects

could not sensibly accommodate the introduction of Section 44 (9A) on a

retrospective basis.

20.

Payment of STC:

Section 34 (2) of the Amendment Act had the effect that on 8 August 2007, a

distribution by a resultant company (the Applicant), would retrospectively be deemed

to have been a dividend attracting STC. Section 64 B (7) requires that STC must

be paid by not later than the last day of the month, following the month in which the

“dividend cycle” relevant to such dividend ends. Section 64 B (7) stipulates further

that the payment must be accompanied by a return in prescribed form. Failure to

comply with this obligation is a criminal offence in terms of Section 75 (1) (a). The

effect of Section 34 (2) in relation to a resultant company which was deemed by the
 36

new Section 44 (9A) to have declared a dividend on 3 May 2007 was that the

company’s dividend cycle would have ended on 3 May 2007. It would follow from

Section 64 B (7) that the resultant company was obliged by 30 June 2007 to pay

the STC and to file the prescribed return. However, as at 30 June 2007 there was

no law in force which imposed STC on the distribution and it would thus have been

impossible for the resultant company to comply with Section 64 B (7). It was

therefore submitted that it was clear that Section 64 B (7) could not apply to the

retrospective deemed dividend. This meant that it was necessary for the Amendment

Act to have contained provisions to deal with the date for payment and the rendering

of returns in respect of retrospective deemed dividends, but it did not. The effect of

this was, according to Applicant, that as at 8 August 2007, the resultant company

was immediately both in default of a statutory obligation to pay STC by 30 June

2007, and guilty of a criminal offence for not filing a return on time. This sort of

unfair and anomalous consequence could not have been intended and it supports

the contention that it was not intended to apply Section 44 (9A) to distributions

completed prior to the promulgation of the legislation.

 37

21.

Payment of interest:

The same concerns applied to the payment of interest as contemplated in Section

64 B (9). Section 64 B (9) refers to Section 64 B (7) and imposes interest if the

STC is not paid by the required date, i. e. 30 June 2007. Applicant submitted that

there could be no obligation to start paying interest as from 1 June 2007 in respect

of an STC liability which did not exist at that date. It was thus necessary for the

Amendment Act also to introduce provisions regarding interest on unpaid STC

arising from retrospective deemed dividends, but it did not contain such provisions.

22.

Corporate shareholders of the resultant company:

If a resultant company was deemed by Section 44 (9A) to have declared a dividend

on 3 May 2007, any corporate shareholder of that company that received the

dividend would be entitled to an STC credit for such dividend in terms of Section 64

B (3). At the time of receiving the dividend, the holding company would clearly not
 38

have recognized the STC credit, since the resultant company was not liable for STC.

If the holding company, having received a distribution from the resultant company on

3 May 2007, which was at not that time a dividend, then paid a dividend to its own

shareholders shortly thereafter, say on 31 May 2007, the holding company’s

dividend cycle for STC purposes would have ended on 31 May 2007. If the

amended Section 44 (9A) had actually been in force as from 21 February 2007,

the holding company would have been able to claim an STC credit in respect of the

dividend received by it. But, because no such law was enacted on 8 August 2007,

the holding company would have been required to pay STC by 30 June 2007,

without any STC credit for the amount received from the resultant company.

It would not be a solution to say that the deemed dividend could simply be claimed

by the holding company as an STC credit at a later time. First, the holding company

may in fact never be in a position to declare further dividends. Second, only

dividends accruing to the holding company during the relevant dividend cycle may

be claimed as an STC credit. It follows that if the holding company did not claim the
 39

STC credit in respect of the dividend cycle ending on 31 May 2007, it could not

claim such credit in respect of any later dividend cycle.

It was therefore contended that in these circumstances the anomalous and patently

inequitable situation arises that by virtue of the retroactive introduction of Section 44

(9A), both the resultant company and the holding company, would pay STC on their

full respective distributions. One could not suppose that the legislature intended such

an anomalous outcome. If it had, one would have expected the amending legislation

to contain provisions in its machinery to enable the holding company retroactively to

claim the STC credit in respect of the relevant dividend cycle.

23.

Capital gains tax consequences:

Under the law at the time of the distribution, the person receiving a distribution from

a company share premium account, would have been deemed to have received a

“capital distribution” under par. 76 of the Eighth Schedule to the ITA dealing with

capital gains tax (“CGT”). This would have amounted to a part-disposal of shares
 40

under par. 33 of the Eighth Schedule, activating the provisions of that Schedule

and leading to a capital gain or a capital loss, depending on the base cost of the

shares. Under the new Section 44 (9A) the distribution was deemed to be a

distribution of a profit not of a capital nature. It follows that it would not have

constituted a capital distribution for CGT purposes, and would not have given rise to

a capital loss or gain in the hands of the shareholder.

The Amendment Act provides no machinery to reverse the capital gain or loss that

accrued at the time of the original distribution. This meant that a shareholder who

was liable for CGT in respect of the distribution could not receive its capital gain, but

must at the same time accept the diminution in value in its shares as a result of the

resultant company having to pay STC retrospectively. It was therefore submitted that

the absence of such mechanism in the Amendment Act was another important

contextual factor in favour of an interpretation to the effect that the new Section 44

(9A) was not intended to apply to completed distributions.

 41

24.

It was submitted that these concerns were real, and went to the heart of the

practical functioning of the STC regime and arise from the plain language of the Act.

The consequences of the retrospectivity led to unfair and anomalous results, and it

can therefore not be accepted that Parliament intended the new provision to apply to

completed distributions. No such anomalies or difficulties arise if the new Section 44

(9A) applies only to transactions and distributions that occurred after its

promulgation. In the context of the interpretative challenge it was accordingly

submitted that Section 34 (2) should be interpreted to limit the retroactive

application of Section 44 (9A) to transactions or distributions that were not

complete before 8 August 2007.

It will be noticed from Applicant’s argument and certain authorities and dicta relied-

upon, that considerations of “fairness” and “unfairness” are sought to be introduced

onto the centre-stage.

 42

25.

First Respondent’s argument:

First Respondent gave a broader background and explained that in late 2006 and

early 2007, the Respondents became aware of a loop-hole in the Income Tax

Regime. Section 64 B of the Income Tax Act, levied a tax (STC) on the net

dividends declared by companies. Section 44 (9A) created a loop-hole in the STC

regime in that it allowed companies engaged in amalgamations to avoid paying STC.

This loop hole created a real risk that the National fiscus would suffer extensive and

permanent harm as it was put. Accordingly, in the budget speech on 21 February

2007, the Minister announced that legislation was being prepared to close the loop-

hole and that it would be made retrospective to that day. I have referred to the

process that followed thereafter.

 43

26.

The loop-hole:

STC was introduced by Section 64 B and 64 C of the Income Tax Act. It was the

tax on net dividends, that is, on a company’s distribution of its profits to its

shareholders. It was not meant to tax capital distributions. That was why par. (f) of

the definition of a “dividend” in Section 1 of the Act excluded any distribution that

represented “a reduction of a share premium account of a company”.

Section 44 of the Income Tax Act facilitates amalgamations. Section 44 (1) defines

an amalgamation as a transaction by which a company (the “amalgamated

company”) disposes of all of its assets to another company (the “resultant

company”) as a result of which the amalgamated company is terminated. Section 44

(9) catered for amalgamations, such as the Pienaar Brothers’ amalgamation, where

the resultant company (“NEW CO”) issued shares to the amalgamated company

(“OLD CO”) which the latter then distributed to its shareholders as a dividend in

specie. Such a dividend would ordinarily have attracted STC. Section 44 (9)
 44

however exempted it from STC by deeming the distribution not to be a dividend for

purposes of STC.

The purpose of this exemption was to render an amalgamation STC neutral by

exempting the distribution by the amalgamated company (“OLD CO”) of its shares in

the resultant company (“NEW CO”). Parliament assumed that the distributable

income previously held by the amalgamated company (“OLD CO”), would be rolled

over into the resultant company (“NEW CO”) and thus attract STC, as it would have

done in the amalgamated company (“OLD CO”), if and when distributed by way of

dividend declared by the resultant company (“NEW CO”).

The assumption however overlooked the fact that distributable income in the hands

of the amalgamated company (“OLD CO”) may change character and become share

premium in the resultant company (“NEW CO”) as happened in the Pier Brothers’

transaction. The parties to such a transaction may then avoid STC altogether.

The amalgamated company (“OLD CO”) surrenders its distributable income to the

resultant company (“NEW CO”) in return for new co shares. Its distribution of the
 45

new co shares to its own shareholders is a dividend but, exempt from STC by

Section 44 (9).

The resultant company (“NEW CO”) receives the assets of the amalgamated

company (“OLD CO”) but, in its hands, they represent share premium and not

distributable income. Any distribution to shareholders by the resultant company

(“NEW CO”), from its share premium, does also avoid STC because, it is a capital

distribution and not a “dividend” as defined.

The amalgamation accordingly allows the parties to avoid STC that would otherwise

have been payable by the amalgamated company (“OLD CO”) on its distributable

income. This is the loop-hole that gave rise to the amendment of the Act in 2007.

26.

The amendment:

In the light of the authorities referred to in par. 17 above, it is necessary and

important to look at the circumstances that existed at the time the topic of closing

the loop-hole arose, and the Second Respondent’s reasons for doing so.
 46

1. SARS was first alerted to the loop-hole by a proposed take-over by Brait of

Shoprite in late 2006. This proposed transaction did not materialize but, it

meant that SARS became aware of the loop-hole and realized that it could

be exploited to avoid STC on a massive scale. SARS accordingly issued a

public statement on 10 January 2001 and this, according to Mr W.

Trengrove SC on behalf of the First Respondent was the first in a series of

seven relevant notifications or warnings to interested taxpayers that the

Treasury was intent of closing the particular loop-hole. The public statement

said that certain corporate transactions were “structured in such a way that

they show complete and reckless disregard for tax morality and South African

Tax Law”. It gave notice that it intended “to carefully examine these

transactions in order to ensure that no impermissible tax loss occurs”, and let

it be known that “the architects of certain tax aggressive structures will not be

permitted to abuse South African tax provisions in ways clearly unintended

by the legislature”. The second notification occurred on 21 February 2007

when the Minister of Finance delivered his budget speech. He said that the
 47

Government would enact “standard anti-avoidance measures that will

commence on conclusion of this speech”. This intimation was repeated in the

budget review published later that same day. The next notification by way of

an elaboration occurred when a press statement was published on 21

February 2007. The statement made it reasonably clear that SARS planned

to seek the repeal of the exemption in Section 44 (9) with retrospective

effect from 21 February 2007. This would indeed have closed the particular

loop-hole. The fourth notification occurred when a Draft Bill was published

on 27 February 2007 that provided for the repeal of Section 44 (9) with

retrospective effect from 21 February 2007. SARS made presentations on

the Draft Bill to the Parliamentary Portfolio Committee on Finance on 28

February and 9 March 2007. This was at least the fifth relevant notification

to taxpayers. There then followed a process of public consultations and

representations on the Draft Bill. This was the sixth relevant notification in

this process.
 48

This was the process during the period March to May 2007 when the

Pienaar Brothers’ transactions were made and implemented. The seventh

relevant notification occurred on 7 June 2007 when the Amendment Bill

was tabled in Parliament. The Amendment no longer sought the repeal of

Section 44 (9), because SARS and the Minister had been persuaded by

representations on the Draft Bill to retain the exemption under Section 44

(9) but, to plug the loop-hole in a different way. This was done by enacting

a new Section 44 (9A) that rendered a distribution by the resultant company

(“NEW CO”) of its share premium that previously constituted divisible income

in the hands of the amalgamated company (“OLD CO”), subject to STC. The

motivation for this amendment was set out in the Explanatory Memorandum

that I referred to in par. 11.7 above.

27.

The Amendment Bill was adopted and then promulgated as the Amendment Act on

8 August 2007. Section 34 (1) (c) of this Act introduced a new Section 44 (9A)
 49

of the Income Tax Act. Section 34 (2) gave retrospective effect to the amendment

and read as follows:

1. “(c) shall be deemed to have come into operation on 21 February

2007 and applies to any reduction or redemption of the share capital

or the share premium of a resultant company including the acquisition

by that company of its shares in terms of Section 85 of the

Companies Act, 1973 (Act 61 of 1973) on or after that date”.

In my view this Section is clear and it applies to “any reduction…” (I underline).

On Applicant’s argument it must be read to mean that it does not apply to a

completed action.

28.

The proper interpretation of the amendment:

Mr W. Trengrove SC agreed that there was a strong presumption against the

retrospective operation of legislation, and referred me to a number of decisions in

this regard, one of them being Veldman v TPP, Witwatersrand Local Division 2007
 50

(3) SA 210 (CC) at par. 26. This was but a mere presumption and had to yield to

the clear intention of the legislature. In Du Toit v Minister of Safety and Security

supra, it was explained that this presumption “may be rebutted, either expressly or

by necessary implications, by provisions or indications to the contrary in the

enactment under consideration”. The same approach was adopted by the

Constitutional Court in Veldman supra at par. 48 and 68. It endorsed the position

of Innes CJ in Curtis v Johannesburg Municipality 1906 TS 308 at 311 where the

following was said: “The general rule is that, in the absence of express provision to

the contrary, statutes should be considered as affecting future matters only; and

more especially that there should if possible be so interpreted as not to take away

rights actually vested at the time of their promulgation. The legislature is virtually

omnipotent, but the Courts will not find that it intended so inequitable a result as to

the destruction of existing rights unless forced to do so by language so clear as to

admit of no other conclusion”.

In this case the relevant amendment was clearly retrospective and applied “to any

reduction … of the share capital or share premium of a resultant company … on or

 51

after that date” (21 February 2007). It was submitted that the clear language of this

amendment indicated expressly that it applied to completed transactions.

“Any” ordinarily means “no matter which”. It does in itself not suggest exceptions.

See: Shorter Oxford English Dictionary, 6th Edition.

29.

It was contended that in an effort to overcome this, the Applicant pointed to “un-

businesslike” and “impractical” consequences that would flow from a retrospective

application of the amendment. With reference to the argument that it was impossible

to file the relevant return and pay the STC by no later than 30 June 2007, it was

submitted that this argument overlooked that, even when a legislation operates

retrospectively, the Court “will not give greater retrospective operation to an Act than

its language renders necessary”.

See: Incledon (Welkom) v Qwa-Qwa Development Corporation 1990 (4) SA 798

(A) 804.
 52

In this case, Section 34 of the Amendment Act made it clear that he knew Section

44 (9A) must be deemed to have come into operation on 21 February 2007. It

was however silent on the knock-on effect of the amendment on the provisions of

Sections 64 B and 64 C. Parliament clearly did not intend to criminalise the failure

of a resultant company to render a return for, and pay the STC retrospectively

imposed on it before the Amendment Act came into force. I may at this stage

interpose and refer to S. 35 of the Constitution which states that an accused person

has the right not to be convicted of an act or omission that was not an offence at the

time of the commission or omission. The obvious answer would be to interpret the

amendment to mean that the tax payer has duty to file a return and pay the STC,

only arose on 8 August 2007 when the amendment came into force. This

interpretation would give effect to the purpose of the amendment while, at the same

time, observing the rule that the Court should not give greater retrospective

operation to the amendment than was necessary.

 53

30.

Such an interpretation was adopted by Innes CJ in Curtis v Johannesburg

Municipality supra.

31.

As far as the question of interest was concerned, it was submitted that this was

readily overcome by a similar interpretation that the tax payer’s duty to pay the STC

only came into effect on promulgation of the Amendment Act on 8 August 2007,

which meant that interest would only run from that date.

32.

As far as the debate around the STC credit was concerned, it was submitted that

this concern was not well-founded. Once it is understood that the amendment had

retrospective effect, it meant that, as a matter of law, there was never any capital

gain or loss. Moreover, the amendment came into operation well before the end of
 54

the tax year concerned, meaning no capital gain or loss would have been declared

or assessed.

In any event, it was contended that what all of these points have in common, is that

they argue against the retrospectivity of the amendment because parliament would

have addressed these issues if it intended to render the amendment retrospective.

But this overlooks the clear and express language of the amendment, as one

cannot, by a process of inference, argue that the legislature intended the

amendment to have meaning incompatible with its express and clear language.

33.

Applicant’s reliance on cases for “contextual” interpretations, citing Natal Joint

Municipal Pension Fund v Endumeni Municipality supra, paragraphs 18, 25 and

26, Bothma-Batho Transport (Edms) Bpk v S. Bothma en Seun Transport supra,

paragraphs 10 to 15, do not assist the Applicant’s argument. The relevant dicta

make it clear that important parts of the context of the statute are the “circumstances

intendant upon its coming into existent”, the “background to the preparation
 55

production of the document” and the “apparent purpose to which it is directed and

the material known to those responsible for its reduction”.

In this case, the history of the amendment makes it plain that the Minister, SARS

and Parliament were determined to close the loop hole with effect from the

Minister’s budget on 21 February 2007.

34.

I agree with these submissions by Mr Trengrove SC. It is clear that this was what

the Respondents intended to do. They obviously could not commit themselves on

the day of the budget speech, as to the exact manner in which this would be done,

inasmuch as a public participation process and a parliamentary process had to be

followed subsequently. At the date of the parliamentary speech, and shortly

thereafter neither of the Respondents could have given a promise or indeed had

made a promise that the loop-hole would be addressed in a certain manner only.

Such an approach would defeat the whole purpose of a parliamentary debate and all

the processes preceding it.

 56

35.

With reference to applicant’s complaint that the effect of the amendment is

“prejudicial and unfair” to taxpayers, it was said that this submission was unfounded

for the following reasons:

1. Section 44 (9) created a loop-hole that taxpayers could exploit to avoid

STC. That was unfair to the bulk of taxpayers and the public at large.

2. The effective collection of revenue was essential to the provision of services

to ordinary South Africans.

In Metcash Trading Ltd v Commissioner South African Revenue Service

and Another 2001 (1) SA 1109 (CC) par. 60, it was said that the fiscus

plays a vital role in the public interest of collecting taxes for the economic

well-being of the nation as a whole.

3. The amendment also did not place amalgamating companies in any worse

position than other companies or treat them unfairly. On the contrary, it

simply ensures that they – like all other companies – pay STC on the

distribution of income to their shareholders.

 57

36.

In a recent decision of the Supreme Court of Appeal of South Africa, in New

Adventure Shelf 122 (Pty) Ltd v The Commissioner of the South African Revenue

Service [2017] ZASCA 29 (28 March 2017), and with reference to the particular

facts before that Court, Leach JA said the following in par. [28]: “In any event, even

if in certain instances it may seem “unfair” for a taxpayer to pay a tax which is

payable under a statutory obligation to do so, there is nothing unjust about it.

Payment of tax is what the law prescribes, and tax laws are not always regarded as

“fair”. The tax statute must be applied even if in certain circumstances a taxpayer

may feel aggrieved at the outcome”. In that case the taxpayer was assessed for a

capital gains tax arising from the sale of immovable property where the sale was

cancelled more than three years after the assessment of capital gains tax, but

before the purchase price had been paid in full.

There are of course degrees of “unfairness” and such conceptual reasoning may

also become involved in statutes that are prospective in operation. Prospective

legislation may, and often does, affect vested rights as well. Such statutes may be
 58

tax statutes, they may relate to property, they may relate withholding tax on

dividends as well as to Customs and Excise statutes, by way of example only. Apart

from Bill of Rights considerations, I agree with what was said by the Supreme Court

of Canada in Gustavson Drilling (1964) Ltd v M.N.R. [1977] 1 S.C.R. 271 at 283,

where Dickson J said: “No one has a vested right to continuance of the law as it

stood in the past; in tax law it is imperative that legislation conform to changing

social needs and governmental policy. A taxpayer may plan his financial affairs in

reliance on the tax laws remaining the same; he takes the risk that the legislation

may be changed”.

In United States v Carlton supra, the United States Supreme Court, by way of

Justice Blackmun said the following, and I agree with due respect in the context of

the present debate that “Tax legislation is not a promise, and a tax payer has no

vested right in the Internal Revenue Code”. It was also pointed out, quite logically,

that prospective changes in law may be “unfair” to certain persons or institutions. In

the context of tax statutes specifically, rigidity is not to be expected and the fiscus

must be able to function effectively taking into account changing demands of society.
 59

It must act lawfully of course, but I do not hold that general considerations of

“fairness”, which is in any event a relative term, can be the over-riding consideration

in the present instance.

If, as in the present context, companies are treated equally in the context of having

to pay STC on the distribution of income to their shareholders, it is indeed difficult to

argue one particular company resulting from an amalgamation process is being

treated so unfairly that a Court would able to say that a retrospective tax statute

could not have been intended to be applied to its factual circumstances.

37.

In the light of all of the above mentioned considerations, I therefore hold that the

amendment is clear, its purpose is rational and that it applied to all transactions

including completed transactions.

The result is that prayer 2 of the Notice of Motion cannot be granted.

 60

38.

The constitutional challenge:

Applicant’s argument:

On the assumption that prayer 2 is refused and it is held that the legislature

intended the Applicant to pay STC ex post facto on the distribution, Applicant

submitted that the Amendment Act was invalid on the grounds of being inconsistent

(to the extent of its retrospectivity) with the foundational constitutional value, the rule

of law.

39.

Before dealing with this argument, a preliminary issue was referred to. Applicant

says the challenge was brought because it had been made subject to a retroactive

tax assessment subjecting it to a substantial and unexpected STC bill. Any other

taxpayer in the same position would have been equally prejudiced and this prejudice

is inherent in having to pay tax that was not payable at the time. The prejudice also

extends to the shareholders of the Applicant, including the BEE shareholders whose
 61

purchase price was based on the net asset value of the Applicant which did not take

account of the additional STC burden at the time.

40.

The Rule of Law and the Constitution:

I was referred to a number of academic writings in this context and it was quite

correctly submitted that not only must Government act in accordance with laws, but

also that the laws must have a certain essential quality, namely, in the present

context, that laws should be reasonably clear, accessible and prospective in their

operation.

See: B. Beinart, (1962) Acta Iuridica 99, where the following was said at 106:

“Several of the attributes or tendencies of law outlined above are not only its natural

consequences which explain why the Rule of Law that is preferable to the rule of

discretion; they at the same time indicate what are the desirable features in the

formulation of legal rules. To these attributes legal rules should therefore as far as

possible conform, and if they do not, then, even if they have been lawfully enacted,
 62

they would infringe the principles of the Rule of Law. Of such features the prime one

is that of certainty; which has rightly been addressed by modern authors … It is an

element of certainty which enables persons to know and predict their rights, and

therefore the Rule of Law postulates that law should not be formulated in wide

general term but should be reasonably clear and precise; otherwise a decision by

discretion is imported. It also postulates that laws should not be applied

retrospectively to the disadvantage of individuals, for retrospective laws give the

individual no fair chance of knowing and predicting his rights and of knowing how to

regulate his conduct unless he be remarkably prescient …”

Section 1 (c) of the Constitution and read with Section 2, provides that the Rule of

Law is one of the founding values of the Constitution and that any law of conduct,

inconsistent with the Constitution, is invalid. Given this status, all arms of

Government are bound by the fundamental value of the Rule of Law, so it was

correctly submitted.
 63

41.

With reference to Dawood and Another v Minister of Home Affairs 2000 (2) SA

936 (CC) at par. 47, where it was said that “an important principle of the Rule of

Law (is) that rules be stated in a clear and accessible manner”, it was submitted

that in the present context the fundamental premise is that persons subject to the

law must be able to find it, and understand it, so that they can regulate their affairs

accordingly. A tax payer is also entitled to organise his financial affairs in such a

way as to pay the least possible tax, and in this context I was referred to CSARS v

NWK Ltd 2011 (2) SA 67 (SCA) at par. [42]. As a general rule this is of course so,

although a Court will in appropriate instances carefully look at the form of a

particular transaction and examine its true nature and substance. It is clear that the

general principle that a taxpayer may organise his financial affairs in such a way as

to pay the least tax permissible is not cast in stone, and the power of a Court to

examine any particular transaction as to its true nature and substance does not

mean that this power would be in conflict with the mentioned principle. This was
 64

clearly formulated in Erf 3183/1 Ladysmith (Pty) Ltd and Another v Commissioner

of Inland Revenue 1996 (3) SA 942 (A) at 953 A – F.

I was also referred to President of the Republic of South Africa and Another v

Hugo 1997 (4) SA 1 (CC) at par. [102], the minority judgment of Mokgoro J. In the

same vein one can refer to Pharmaceutical Manufacturers Association of South

Africa: In re Ex parte President of the Republic of South Africa 2000 (3) SA 674

(CC) at par. [39]. In Veldman v Director of Public Prosecutions: Witwatersrand

Local Division 2007 (3) SA 210 (C), Mokgoro J, writing for the minority said the

following at par. [26], with reference to Calder v Bull 3 US 386 (1798) at 388 and

396: “Generally, legislation is not to be interpreted to extinguish existing rights and

obligations. This is so unless the statute provides otherwise or its language clearly

shows such a meaning. That legislation will affect only future matters and not take

away existing rights is basic to notions of fairness and justice which are integral to

the Rule of Law, a foundational principle of our Constitution. Also central to the Rule

of Law is the principle of legality which requires that law must be certain, clear and
 65

stable. Legislative enactments are intended to “give fair warning of their effect and

permit individuals to rely on their meaning until expressedly changed”.

As it stands, this exposition is generally accepted, but it must be said that context is

everything in law, and obviously one needs to examine the particular statute and all

the facts that gave rise to it. This principle applies expressly in Criminal Law. See:

Section 35 (3) (l) of the Constitution, but our Courts have yet to consider definitely

whether outside the Criminal Law context, retrospective legislative amendments can

be constitutionally valid.

It was therefore submitted in the light of the mentioned constitutional imperative, the

Courts must vindicate the Rule of Law by setting aside legislation which contravenes

that principle. No longer are the Courts limited to techniques of strict statutory

interpretation in the light of presumptions to express their disapproval of breaches of

the Rule of Law. Such legislation is contrary to the Constitution and therefore

invalid.
 66

42.

I was also referred to the provisions of Section 172 (1) (a) of the Constitution for

which obliges Courts in determining constitutional matters to declare that any law

which is inconsistent with the Constitution to be invalid to the extent of its

inconsistency.

43.

Foreign law:

It was submitted that foreign law could provide some guidance, although ultimately

the matter would have to be decided on the basis of our own constitutional

requirements. It was also not strictly required to take account of International Law in

terms of Section 39 (1) (b) of the Constitution as the Applicant herein did not rely

directly on a right in the Bill of Rights. (Apart from the fact that at a later stage a

further point was raised that the retrospective amendment unlawfully infringed the

Applicant’s rights to property.)

 67

I was then referred to the position in the United States, where retrospective

amendments to tax laws have been held to be constitutional permissible where they

meet a “due process” test, which is linked to the rationality of the legislative purpose

of the amendment.

See: United States v Carlton supra.

I have already referred to the Carlton decision and the fact that the “detrimental

reliance principle” is not limited to retroactive legislation, an observation which is in

my opinion entirely logical.

The rationality principle that must underlie a legislative purpose is of course also part

of our constitutional reasoning, and the Pharmaceutical Manufacturers Association

decision supra makes this abundantly clear.

In the context of the United Kingdom, it was submitted that those Courts have not

been required to decide a challenge to retroactive amendments on the basis of the

Rule of Law principle, which does not have formal constitutional status, but that

retroactive tax laws have instead been challenged on the basis that they infringe the

right of tax payers to peaceful enjoyment of their positions under the European
 68

Convention of Human Rights. It is true that the United Kingdom does not have a Bill

of Rights, which expressly, such as in the South African Constitution, states that the

Rule of Law is a foundational principle.

It is in my view abundantly clear that the Rule of Law principle applies in the United

Kingdom. See for instance: R (Evans) v Attorney General [2015] UKSC at par. 52.

In Halsbury’s Laws of England, 5TH Edition 2014, Vol. 20, reference is made to this

principle as being part of the legal system. The Constitutional Reform Act 2005,

also declared in S. 1 (a) that nothing in its provisions adversely affects the existing

constitutional principle of the Rule of Law.

In the context of the position in the United Kingdom, I was referred to National

Westminster Bank plc (Respondents) v Spectrum Plus Ltd and Others

(Appellants) [2005] UKHL 41.

In the present context it is worthwhile to refer to certain dicta of Lord Nicholls of

Birkenhead. It was amongst others argued that the decision in Siebe Gorman and

Co Ltd v Barclays Bank Ltd [1979] to Lloyds Rep 142 was wrong and should be

overruled. The Judges were in agreement with that contention, but it was similarly
 69

submitted to them that they should overrule that decision for future purposes only. In

that context, Lord Birkenhead made some relevant observations: “

4. The starting point is to note some basic, indeed elementary features of this

country’s judicial system. The first concerns the essential role of Courts of

law. In the ordinary course the function of a Court is adjudicative. Courts

decide the legal consequences of past happenings. Courts make findings on

disputed questions of fact, identify and apply the relevant law to the facts

agreed by the parties or found by the Court, and award appropriate

remedies.

5. The second feature concerns the wider effect of a Court decision on a point

of law. To promote a desirable degree of consistency and certainty about the

present state of “the law”, Courts in this country have all adopted the practice

of treating decisions on a point of law as precedents for the future. If the

same point of law rises in another case at a later date the Court will treat a

previous decision as binding or persuasive, depending on the well-known

hierarchical principles of “stare decisis”.

 70

6. The third feature is that from time to time Court decisions on points of law

represent a change in what until then the law in question was generally

thought to be. This happens most obviously when a Court departs from, or

an Appellate Court overrules, a previous decision on the same point of law.

The point of law may concern the interpretation of a statute or it may relate

to a principle of “a judge-made” law, that is, the common law (which for this

purpose includes equity). A change of this nature does not always involve

departing from or overruling a previous Court decision. Sometimes the Court

may give a statute, until then free from judicial interpretation, a different

meaning from that commonly held.

7. The fourth feature is a consequence of the second and third features. A court

ruling which changes the law from what it was previously thought to be

operates retrospectively as well as prospectively. The ruling will have a

retrospective effect so far as the parties to a particular dispute are

concerned, as occurred with the manufacturer of the ginger beer in Donahue

v Stevenson [1932] AC 562.

 71

8. People generally conduct their affairs on the basis of what they understand

the law to be. This ‘retrospective’ effect of a change in the law of this nature

can have disruptive and seemingly unfair consequences. ‘Prospective

overruling’, sometimes described as ‘non-retroactive overruling’, is a judicial

tool fashioned to mitigate these adverse consequences. It is a shorthand

description for Court rulings on point of law which, to greater or lesser extent,

are designed not to have the normal retrospective effect of the judicial

decisions.

9. Prospective overruling takes several different forms. In its simplest form

prospective overruling involves a Court giving a ruling of the character sought

by the bank in the present case. Overruling of this simple or ‘pure’ type has

the effect that the Court ruling has an exclusively prospective effect. This

ruling applies only to transactions or happenings occurring after the date of

the Court decision. All transactions entered into, or events occurring before

the date continue to be governed by the law as it was conceived to be before

the Court gave its ruling.

 72

10. Other forms of prospective overruling are more limited and ‘selective’ in their

departure from the normal effect of Court decisions. The ruling in its

operation may be prospective and, additionally, retrospective in its effect as

between the parties to the case in which the ruling is given. More the ruling

may be prospective and, additionally retrospective as between the parties in

the case in which the ruling was given and also between the parties in any

other cases already pending before the Courts. There are other variations on

the same theme”.

The process described by the learned Judge is almost the same in the South African

legal system. The South African Constitution gives this process a constitutional

blessing inasmuch as Section 172 (1) (b) states that, when a law is declared

inconsistent with the Constitution and thus invalid, a Court may make an order that

is just and equitable including an order limiting the retrospective effect of the

declaration of invalidity.

The learned Judge also referred to the general practice that had been adopted in the

United Kingdom with reference to the decision by Lord Reid in West Midland
 73

Baptist (Trust) Association Incorporated v Birmingham Corporation [1970] AC

874, 898 – 899, where the following was said: “We cannot say that the law was one

thing yesterday but is to be something different tomorrow. If we decide that [the

existing rule] is wrong we must decide that it always has been wrong, and that

would mean that in many completed transactions owners have received too little

compensation. But that often happens when an existing decision is reversed”.

44.

It is clear from this incisive summary that not only do certain statutes affect rights or

vested rights retrospectively, but that decisions of Courts do so in many cases. I am

not aware of any authority, and none has been provided to me that those results

would mean that any such statute or decision is unconstitutional per se, irrespective

of the reason for the adoption of the statute or the facts of a particular case before a

Court of law, and irrespective of its wording.

In Deutsche Morgan Genfell Group Plc v HMCIR [2006] UKHL 49, Lord Hoffmann

again pointed out (at par. 23) that it is a common occurrence that judicial decisions
 74

change the law retrospectively inasmuch as rulings will have a retrospective effect

as far as the parties to a particular dispute are concerned at the very least.

45.

In THE QUEEN on the application of R. HUITSON v HMRC [2010] EWHC 97,

(Admin) the debate concerned an employment arrangement that sought certain tax

advantages. It was accepted by the High Court that the particular arrangement had

no genuine commercial purpose and could be described as artificial. A particular

statute had the effect of changing fiscal legislation regarding double taxation relief

with a retrospective effect and the question before the Court was whether such

retrospective amendment struck a fair balance as required by Article 1 of the First

Protocol to the Convention for the Protection of Human Rights and Fundamental

Freedoms, with reference to the material jurisprudence of the European Court of

Human Rights. In the context of the particular legislation that was given retrospective

effect in the interest of public policy Parker J said the following at par. 33: “It seems

to me that these two features emphasized the importance Parliament attached to the
 75

public policy to which I have referred. In my view, these events sent out a clear

signal to taxpayers and their advisors that the legislature would be very likely to take

effective and decisive steps to count, even with retrospective measures, any

attempt, through artificial arrangements, to take advantage of a double taxation

arrangement …”

Dealing with Article 1 to the mentioned Protocol which read as follows: “Every

natural legal person is entitled to the peaceful enjoyment of his possessions. No one

shall be deprived of his possessions except in the public interest and subject to the

conditions provided for by law and by the general principles of International Law.

The preceding provision shall not, however, in any way impair the right of a State to

enforce such laws as it deems necessary to control the use of property in

accordance with the general interest or to secure the payment of taxes or other

contributions or penalties”, the learned Judge referred to certain general

propositions:

1. In securing the payment of taxes, a national authority must strike a “fair

balance” between the demands of the general interest of the community and
 76

their requirements of the protection of the individual’s fundamental rights,

including the right that a person enjoys to “the peaceful enjoyment of his

possessions”;

2. In framing and implementing policies in the area of taxation, the State will

enjoy “a wide margin of appreciation and the Court will respect the

legislature’s assessment in such matters unless it is devoid of reasonable

foundation”. The domestic analogue of the margin of appreciation is the

discretionary area of judgment and is especially wide in the field of social

and economic policy;

3. The more the subject matter of legislation concerns matters of broad social

policy, the less ready will be a Court to intervene;

4. Nonetheless the Court will carefully examine all the relevant circumstances

including the history of the challenged provisions, to determine whether a

fair balance has been struck;

5. These principles apply to tax legislation that is retrospective;

 77

6. With reference to Article 1 of Protocol No. 1: retrospective legislation is not

as such prohibited. The question to be answered is whether in the

Applicant’s specific circumstances, the retrospective application of the law

imposed an unreasonable burden on them and thereby failed to strike a fair

balance between the various interests involved;

7. The imposition of a tax is not devoid of reasonable foundation by reason only

that it may have some retrospective effect.

I also regard these propositions as being particularly instructive in the present

context, although the ‘fair-balance’ test is not the test in our constitutional

dispensation.

46.

I was provided with a very lengthy report on “Retroactivity of Tax Legislation by

Gribnaw JLM, and Pauwels, MRT (EDS.) (2013) (EATELP International Tax

Series; Number 9) Amsterdam, published in 2013. This report summarizes the

position in most European countries in the context of the national reports presented
 78

to the Congress in 2010. It refers to the position in Germany where the

Constitutional Court held in principle that there was a ban on retroactivity, but

subject to exceptions. It stated that the first exception is the situation in which a

reasonable taxpayer cannot claim trust or confidence in the still prevailing legal

situation, which is the case from the date of adoption of a Bill in Parliament, or in

the case of an evidentially unclear or unconstitutional legal situation. A second

exception would be the situation in which the confidence in the prevailing legal

situation has to be subordinated to the interest of the legislator to change the law

retroactively. This applies for instance if the disadvantage to the taxpayer is

negligible in the context of the principle of proportionality, and also where the

legislator can claim overriding, urgent or compelling public interest. In Chapter 2.2,

the observation is made that the principle of legal certainty is a fundamental principle

of law, but it is also beyond discussion that the legislator should be able to change

its legislation, including tax legislation. There are various justified reasons to change

tax legislation, such as a change of tax policy and social and technical

developments.
 79

The Congress document comprises more than 500 pages and further reference

thereto would be beyond the scope of this judgment, but one could easily write a

doctoral thesis on this topic. It was also pointed out that an important consideration

was whether there are constitutional limitations to retroactivity of these tax statutes.

Although one is reminded that tax statutes should be interpreted in the same way as

any other statute, it is important to take into account the policy behind the legislation

as well as the wording of the legislation itself. I have briefly dealt with these two

considerations.

47.

Application to South Africa:

It was submitted by Applicant that as in Germany, the Rule of Law compels a

conclusion that strongly retrospective tax statutes should be presumed to be

constitutionally invalid. It was however expressly not suggested that our

constitutional dispensation would never allow the legislature to expressly introduce

retrospective legislative amendments. There could well be exceptional cases where

 80

this could be done without attracting constitutional sanction. The touch-stone would

however always be whether the retrospectivity amounts, in the particular

circumstances of their case, to a contravention of the Rule of Law.

In that context I was referred to the Bozoleck J in Robertson v City of Cape Town

2004 (5) SA 412 (C) in par. [135]. It was said therein that “retrospective legislation

contravenes the Rule of Law where it unreasonably or unfairly impairs the ability of

those bound by the law to regulate their conduct in accordance therewith”. I may just

add at this stage there are obviously degrees of unfairness and not all laws are “fair”

and the real question would be whether a law is “unjust” i. e. whether it passes

constitutional muster, i. e. was the law, accepting its language is clear, passed for a

rational reason? I do not agree, if that was suggested, that “unfair impairment” is the

appropriate test in our constitutional dispensation.

48.

It was then submitted that given further facts to be discussed below it was

unnecessary for me to make far-reaching statements of principle as to when exactly

 81

exceptional circumstances will exist to justify retrospectivity, although I may choose

to do so. It was submitted that this case could be decided on a simple factual basis.

49.

It was contended that knowledge of proposed retrospective amendments to the law

is fundamental to the Rule of Law, and essential for taxpayers to be able to regulate

their conduct in accordance with those amendments. Hence, it was submitted that

unless there was adequate warning of the intention to implement the change

retrospectively, such that the taxpayer cannot be said to have been entitled to rely

on the law continuing to apply, a retroactive amendment could never pass

constitutional muster. No authority was given for this proposition and I will deal with

it when I consider the Respondents’ arguments in this particular context. One can at

this stage obviously ask what is the extent of the knowledge that one is entitled to

have, and what does amount to an adequate warning in each particular case? It

was argued that assuming that the Courts were to conclude that the issue of

adequate warning can justify a retroactive tax amendment, the precise level of
 82

warning that would meet the standard, and how it would have to be publicly

communicated, would remain an issue. The formal adoption of a Bill by Parliament

would be easier to defend than less formal public communications at an earlier

stage of the process, given the numerous changes which can arise in the course of

the pre-legislative and legislative processes. It was said that the requirements of

accessibility and forewarning are not achieved by speeches made in Parliament,

press statements and the like. Such documents may or may not come to the

attention of persons affected by the matters with which they deal. Furthermore, even

if a person learns of a proposed legislative change, he cannot be expected to

regulate his affairs in accordance with a position which may or may not be enacted

in the future, nor can he fairly be expected to refrain from economic activity until the

law is passed. This applies to the present facts, where the Applicant had a

compelling business-need to proceed with the BEE transaction (of which the

distribution was a necessary element) so as to position itself to acquire future

business. Indeed, in this case, it was said, as at 21 February 2007, being the date

of the retrospective operation, not even the Commissioner himself had

 83

conceptualized how Section 44 (9) of ITA would be amended, let alone

communicated to the public how it would ultimately operate.

50.

It was also submitted that to the extent that a “warning” of retrospective legislation

may be potentially relevant as justification for retrospectivity, that warning must self-

evidently pertain to the actual amendment that is implemented. It does not help to

warn of a specific change, and then to implement something different with

retrospective effect. This is however precisely what happened in this case and this is

fatal to the validity of the statutory provision. No authority was given that would

support this approach and I am certainly not aware of any. In any event, I do not

agree that a “precise” warning is required, if any at all. Economic circumstances

generally will demand a degree of fluidity. Rigidity does not belong to a modern

jurisprudence, and even less in tax legislation.

 84

51.

The Commissioner contended that the reasons for adopting the retrospective

approach would be relevant as to whether the legislation offended the Constitution.

It was argued that whether this is correct, and whether revenue interest would

qualify as public interest considerations, need not be decided here. I may interpose

to say that I do not agree and am of the view that this is indeed a relevant

consideration and I will deal with that topic. Applicant’s Counsel continued to argue

that before the State can attempt to rely on such an argument, it would be

incumbent upon it to demonstrate the relevant and compelling public interest

considerations that justify the departure from the underlying norm of the Rule of

Law. Such justification must relate not only to the decision to amend the law per se,

but also to why the change had to operate with retrospective effect. In the absence

of such justification the amendment cannot stand. It was then submitted that in the

present case, the Respondent had failed to provide any, let alone adequate,

justification for the retrospective insertion of Section 44 (9A).

 85

52.

The amendment differs from the “warnings”:

It was argued that for purposes of understanding the Applicant’s submissions, it was

necessary to be aware of two types of distribution to shareholders that can take

place in the context of Section 44 amalgamation transactions. The first would be a

distribution of the consideration shares which are received by the amalgamated

company. Distribution of such shares is a compulsory part of an amalgamation

transaction as the amalgamated company had to be liquidated in order to comply

with the provisions of Section 44. The second distribution is not an essential

requirement. Where the resultant company has paid for the assets with an issue of

par-value shares, the shares are usually issued at a premium which is reflected in

the share premium account. That constitutes capital available for distribution to

shareholders as a dividend by the resultant company. The two types of distribution

differ in at least the following fundamental respects: the first is a compulsory part of

a Section 44 transaction, whereas the second is not, and, the first is made by the

amalgamated company, whereas the second is made by the resultant company.

 86

53.

The facts pertaining to the process of an amendment of Section 44 and the public

statements made in this regard are common cause.

The starting point was a press statement released by the Commissioner on 21

February 2007. Shortly after the budget speech, Treasury released a Draft

Taxation Laws Amendment Bill for public comment. It was not formally introduced

as a Bill in parliament at that stage. It proposed that Sections 44 (9) and 44 (10)

be deleted. The proposal that Section 44 (9) be deleted had apparently been

identified by the Commissioner and National Treasury as a “simple solution” to the

concern that the amalgamation transactions involved potential tax free distributions.

The Minister had in fact testified that a legislative amendment was necessary in

order to align the wording of Section 44 (9) with what he alleged to be the intention

of the legislature. Thus, he identified as the “problem” the provisions of Section 44

(9), which was ultimately not amended, and not the definition of “dividend” which

was ultimately amended. Since this proposed amendment would have applied only

to the compulsory distribution of consideration shares by an amalgamated company,

 87

it had no bearing whatsoever on the tax position of a resultant company such as the

Applicant. The repeal of Section 44 (9) would not have caused the Applicant’s

distribution of share premium to be treated as a dividend.

The Draft Bill was introduced for public comment, and then followed a process of

public submissions within the Parliamentary Portfolio Committee. The Draft Bill was

not cast in stone and it did not emanate from the legislature, but the Executive which

does not make laws. In those circumstances, the public could not be expected to

view the Draft Bill as a firm indication of the likely intent of the legislature, so it was

argued.

The process of public hearings before the Portfolio Committee, and the receipt of

public comment, cannot amount to a firm statement of specific legislative intent of

which the public should have been aware. Moreover, there was substantial public

resistance to the proposed amendment to Section 44, with a number of

Respondents not accepting that there was any need for a change at all.
 88

54.

A revised Draft Legislation was then prepared and submitted by SARS to the

Treasury on 18 April 2007, but this is not before me.

55.

The communication by the Commissioner to Treasury of certain proposals does not

amount to a public “warning” as to the specific changes to the law so it was

contended.

56.

The matter then returned to the Portfolio Committee. There were further public

hearings on 4 May 2007. Representatives of SARS and Treasury presented a draft

response document which had not been approved by either the Commissioner or the

Minister. This addressed the resistance to the original proposal.

 89

57.

It was only after the 4 May 2007 meeting that approval was sought from the

Commissioner and the Minister (which was granted), to alter the original remedy

proposed. The Minister states that “after consulting with the relevant stakeholders

and the public in terms of the legislative process, the matter was reconsidered, and

a compromise was reached, taking into regard the concerns raised during the

consultative process. It was contended that this is hardly an unusual situation where

legislation is subject to public participation processes, and therefore subject to

significant change before the final path is set.

The formal introduction of the Bill into parliament occurred on 7 June 2007 and this

was the first public indication of a firm intention to introduce legislation to would

subject a distribution of share premium by a resultant company to tax, let alone that

that particular solution was intended to operate retrospectively. This occurred more

than a month after the Applicant’s distribution took place.

 90

58.

The Bill was then put before Parliament and accepted. It was argued that it is

entirely feasible that the legislature, exercising its constitutional powers, would

amend or alter the Bill, in accordance with the necessary parliamentary processes.

This was a further reason why even legislation proposed in a formal Bill in

parliament has limited significance when assessing the extent to which members of

the public should anticipate the possibility of the law being amended retrospectively.

59.

It was submitted that on no basis does the process followed by the Respondents

justify the conclusion that adequate warning of the legislative change that was

actually implemented was given, let alone that it would operate retrospectively. It

therefore could not be said that when the Applicant made its long-planned

distribution, taxpayers in South Africa knew, or should have known, that the law

affecting the “dividend” status of such a distribution might change with retroactive

effect.
 91

60.

Applicant then dealt with the argument of Respondents in this particular context,

namely that the amendment that was eventually introduced was sufficiently linked to

the original proposal to deprive tax payers of the right to rely on the existing law in

regard to Section 44 as a whole. The First Respondent said the following in his

Affidavit: “At no stage … was there any doubt that the Treasury and SARS regarded

the pre-amendment wording of Section 44 as fundamentally problematic, the

Treasury and SARS would continue to pursue a change to the existing legislation,

and that such legislation would be retrospective in operation”.

It was argued that on no basis was this good enough to justify the obvious

infringement of the Rule of Law caused by making an unannounced amendment to

legislation applicable to a taxpayer. The Commissioner suggested that as a result of

the proposal to do away with Section 44 (9) retrospectively, all taxpayers ought to

have lost any confidence in any part of Section 44 continuing in force. It seems to

me that this “confidence” argument applies equally in German Law to a large extent,

but the correct question may well be, and that is the one asked in the American
 92

context, namely whether tax laws do make a promise irrespective of the demands of

the fiscus? The answer there was a resounding “no”. The Carlton decision supra

made this clear, amongst others.

61.

It was submitted that I ought to adopt a rigorous approach towards infringements of

the Rule of Law. Just as the Courts have extended themselves to the maximum in

applying the presumption against retrospective legislation, so to should they require

a very high level of correlation between the changes to the law of which taxpayers

were notified, and the actual legislative amendments that follow, before being

satisfied that taxpayers must suffer the consequences of the change retrospectively.

The vague notion that taxpayers should reasonably anticipate some change to the

law, whether or not it is the one that is originally announced, is clearly incompatible

with the warranted rigorous approach. It was submitted as a result that Respondents

could not argue that the amendment gave effect to what was warned of.
 93

62.

With reference to the Minister’s suggestion that following the announcement of 21

February 2007, the Applicant and others would have been placed on guard to

arrange their tax affairs in a manner that does not produce the loss of the STC

which the intended legislation was to address, it was argued that what the intended

legislation was to address, was the loss of STC by virtue of Section 44 (9)

distribution, not a distribution of share premium by the resultant company. The

Applicant could not be expected to arrange its tax affairs in relation to intended

legislation which does not pertain to its situation at all. For these reasons, it was

submitted that a sufficient public advance warning of a retrospective change was

missing in this case. This would make it unnecessary for me to decide whether a

prior public warning would in fact suffice in principle to validate retrospective

legislation, or what the precise nature of an effective warning would be. In this case

there was, in effect, no warning at all.

 94

63.

For these reasons it was submitted that retrospective introduction of Section 44

(9A) without prior and precise warning, fatally infringes the Rule of Law and justifies

the constitutional relief sought.

I am not aware of any authority or legislative provision that provides that a fairly

precise warning need to be given before the legislature can pass retrospective

legislation, whether in general, or in the case of a tax statute. In the latter instance,

economic demands must be considered in the context of the purpose and effect of

an intended statute. If the tax statute is rationally connected to a legitimate purpose,

no precise warning is required, if one at all.

64.

Justification for retrospective addition of Section 44 (9A):

It was submitted that faced with a clear breach of the Rule of Law, the Respondents

presumably intend to argue that public interest considerations come to the rescue in

justifying the breach. The Commissioner stated in the Answering Affidavit that he
 95

accepts that there may be circumstances in which retrospective legislation will offend

the Constitution. Whether this is the case will depend on the nature of this

legislation concern, the effect on members of the public and the reasons for

adopting the retrospective approach. This is especially the case in the context of tax

legislation which is enacted for the benefit of the fiscus and thus the country and the

public as a whole. The Commissioner then gave detail as to why it was considered

that there was a real risk that the national fiscus would suffer extensive and

permanent harm if the particular legislation was not made retrospective. The

Commissioner then gave details as to why, during November 2008, SARS officials

were alerted to a legislative defect that had existed since November 2002, but that

was then at risk of coming to be exploited in the context of a possible Brait /

Shoprite transaction. If that transaction had proceeded the loss to the fiscus would

have been R 1.5 billion plus. It is not necessary to refer to any detail in this regard

as it was argued, at the level of principle, that it is not competent to describe

existing legislation as somehow being “defective”. The law as it stood at the time

was a due product of the legislative process. As with all tax legislation, it regulated
 96

the tax consequences of specific types of transaction and expressly permitted the

amalgamated company to distribute consideration shares STC-free. Similarly

questionable, at the level of principle, so it was put, is the contention that retaining

the law as it was created a risk of “extensive and permanent harm” to the fiscus.

Since “harm” presupposes the loss of something already held, this suggests that the

fiscus is in fact “entitled” to more tax than the law provides. This principle cannot be

applied to tax law. What the fiscus is entitled to is what the law provides. In any

event, even if “harm” were a legitimate consideration, the Applicant submitted that

no case has been made out to this effect inasmuch as there was no suggestion that

in the five years that Section 44 had been in force, any tax payer had “used” the

provisions of the Section, or that the fiscus actually suffered any “loss”. If the

Shoprite transaction was the primary “driver” for the amendment, once that

transaction was off the table, the legislature had no reason to fear any particular

prejudice to the fiscus if the amendments were not retrospective. There was also no

concrete evidence that any other amalgamation transaction was proposed to take

place, or actually did take place, in the period between February and August 2007.
 97

There was also no evidence to support the loose averment that there was a real

likelihood of a “flood of Section 44 amalgamations”. There was also no evidence

that the Commissioner would have been unable to apply the general anti-avoidance

provisions to such transactions if they were undertaken for the sole or main purpose

of obtaining a tax benefit.

It was also argued that the Respondents had put up no facts or explanations why

the ultimate solution had to be made retrospective. The fact that Government

identifies the need “to close any tax avoidance schemes in relation to the STC”

obviously does not mean that this must be done with retrospective effect.

65.

Remedy:

If I were to find that the Amendment Act was unconstitutional I am empowered by

the provisions of Section 172 (1) (b) of the Constitution, to make any order that is

just and equitable. It was submitted that in the present context a just and equitable

order would entail a remedy of severance.

 98

See: Coetzee v Government of the Republic of South Africa 1995 (4) SA 631

(CC) at par. 16.

In this case severance would merely entail the striking out of the words “shall be

deemed to have come into operation on 21 February 2007”, and the words “on or

after that date” from S. 34 (2) (c) of the Amendment Act. This would remedy the

unconstitutionality of the Amending Act, but leave all of its remaining provisions,

which embody “main objective of the Statute” intact and operate prospectively.

Severance in this case would not require any “special treatment” and there would be

no need for an order suspending the order of constitutional invalidity. Applicant also

asked for an order as to costs, including the costs of two Counsel.

66.

First Respondent’s argument:

The constitutional validity of the amendment:

Apart from the challenge relating to Section 25 of the Bill of Rights which was

introduced at a later stage, Applicant’s main challenge related to the provisions of

 99

Section 1 (c) of the Constitution – the Rule of Law, and did in this context not rely

on any provision of the Bill of Rights.

67.

It was submitted by Mr W. Trengrove SC on behalf of First Respondent, that the

Constitution does not in general out-law retrospective legislation, except in the

context of Criminal Law, i. e. Section 35 (3) (l). The question therefore was to

which extent the entrenchment of the Rule of Law inhibited or prohibited

retrospective legislation. It is correct that our Courts have not squarely considered

the issue and certain limited comments have all been in passing.

See: Pharmaceutical Manufacturers Association of South Africa: In re: Ex parte

President of the Republic of South Africa supra at par. 39;

Veldman v Director of Public Prosecutions supra, par. 2;

Robertson v City of Cape Town supra at par. 92, which was reversed on appeal to

the Constitutional Court in City of Cape Town v Robertson 2005 (2) SA 323 (CC),

although the Constitutional Court did not with the retrospectivity issue.
 100

68.

The Applicant also accepted that the Rule of Law does not preclude retrospective

legislation altogether. In the present case, Applicant conceded only a very narrow

exception, namely “when there was adequate warning of the intention to implement

the change retrospectively”. Therefore, in this case, it was Applicant’s contention

that the retrospective amendment was invalid, because there had not been adequate

notice for its enactment.

Mr Trengrove SC submitted that this approach is untenable for the following

reasons:

1. It is inconsistent with the approach in the foreign jurisdictions to which our

Courts have frequently look for guidance in such matters, such as Canada,

the United States, the European Union and the United Kingdom;

2. It is inconsistent with the approach the Constitutional Court has laid down in

relation to the constitutional scrutiny of legislation; and

3. The Applicant’s challenge must in any event fail, even on its own test.
 101

69.

Retrospective legislation, a necessary tool of modern Government:

It was submitted that retrospective legislation is a necessary tool of modern

Government throughout the world’s leading democratic societies. I will refer to some

examples hereunder.

There are also examples of such approach in South Africa. The following examples

were given:

1. The Income Tax Act always proceeded from the unspoken premise that a

Trust was a person in law, and thus subject to taxation. The High Court in

Friedman v CIR 1991 (2) SA 340 (W), held in October 1990, that a Trust

was not a person in law and was thus not subject to taxation. The judgment

of the High Court was upheld in the Appellate Division in CIR v Friedman

1993 (1) SA 353 (A).

The implications of the judgment would have been calamitous. Parliament

therefore moved quickly to undo it and did so by the Income Tax Act 129 of

1991, by which the main Act was amended by defining a “person” to include
 102

a Trust. The amendment only came into force in July 1991, but with

retrospective effect from March 1986, i. e. with retrospective effect of more

than five years. It needs scarcely be said that numerous vested rights would

have been affected by this amendment;

2. In Cohen Brothers Furniture v Minister of Finance 1998 (2) SA 1128

(SCA), the Court was concerned with a withholding tax on dividends levied

in the Ciskei. A judgment of the South African Special Income Tax Court,

handed down in March 1992, cast doubt on the validity of the Ciskei tax.

The Ciskei Government thereupon enacted new legislation for the imposition

of the withholding tax on dividends in 1993, with retrospective effect from

March 1985;

3. There have also been occasions when it has been necessary for Parliament

to make even more drastic retrospective amendments. When there was

doubt cast on the validity of the Exchange Control Regulations,

promulgated under the Currency and Exchanges Act 9 of 1933, Parliament

amended the main Act by Act 23 of 1987, and thereafter by Act 48 of 1988,
 103

on both occasions with retrospective effect from 1 December 1961. The

amendments were thus made with retrospective effect of some 26 and 27

years respectively;

4. Our Courts have always developed common law with retrospective effect,

despite the fact that the “development” may sometimes change the common

law altogether, or introduce a new rule. The observations of Lord Hoffmann

referred to in par. 43 above, make this abundantly clear.

See: K v Minister of Safety and Security 2005 (6) SA 419 (CC) par. 16,

Mukaddam v Pioneer Foods 2013 (5) SA 89 (CC) at par. 68.

The Constitutional Court recognised these implications of development of

the common law, and for the first time accepted in Masiya v Director of

Public Prosecutions, Pretoria 2007 (5) SA 30 (CC) at par. 51, that the

Court may, by exception, limit their development of the common law to

operate only with prospective effect. In other words, the retrospective

development of the common law remains the default rule and it thus

common place for the Courts to change the law with retrospective effect; I
 104

agree with that contention and it is a fact that occurs regularly in the High

Courts. The Westminster Bank decision supra made particularly apt

observations in this context.

70.

Also, our constitutional era introduced even greater fluidity in the state of the law.

Under the doctrine of objective constitutional validity, an unconstitutional law is

rendered invalid with full retrospective effect from the outset.

See: Cross-Border Road Transport Agency v Central African Road Services 2015

(5) SA 370 (CC) par. 13 to 20.

This default rule is however subject to the wide discretion vested in the Courts by

Section 172 (1) (b) of the Constitution, to limit or suspend the retrospective

operation of the declaration of invalidity of a law.

See: Executive Council, Western Cape Legislature v President of the Republic of

South Africa 1995 (4) SA 877 (CC) at par. 107.

 105

It is clear from the relevant dicta that a Court must in certain instances avoid or

control the consequences of a declaration of invalidity of post-constitutional

legislation, where the result of invalidating everything done under such legislation is

disproportional to the harm which would result from giving the legislation temporary

validity. The concept of proportionality therefore clearly plays a role in the context of

judging constitutionality vis-à-vis retrospectivity.

It was therefore contended that the capacity to change the law with retrospective

effect is a common and necessary rule of modern Government and modern

jurisprudence. It has even received constitutional recognition. This is incompatible

with any suggestion that there somehow lurks in Section 1 (c) of the Constitution

either a total prohibition of retrospective legislation or one so constrained as the

Applicant contended for.

 106

71.

Comparative law:

I have already referred to the 2013 treatise by Gribnow and Pauwels, on behalf of

the European Association of Tax Law Professors.

It is clear that three important conclusions were contained therein relevant for

present purposes:

1. Retroactive tax legislation is a commonly known phenomenon in the

countries referred to in this general report;

2. Only in countries like Poland, Portugal and Hungary, is there a mere

prohibition of such retroactive tax legislation;

3. The constitutional restrictions on retroactive tax legislation vary significantly.

At p. 66 of this treatise, the following appears: “In general, it can be observed that

in the various countries the standards that Courts impose for retroactivity on tax

legislation differ significantly. On the one side, there are countries in which the

Courts (almost) fully leave the issue of granting retroactive effect to tax legislation to

the discretion of legislature (or Parliament, as the case may be). On the other side,
 107

there is a group of countries in which an (almost) absolute prohibition of retroactive

taxes applies. Between these opposite positions, there are countries in which Courts

review whether legislators’ decision to grant the retroactive effect stays within certain

(formal and/or substantive) standards. These differences are at first sight

remarkable”.

The Applicant contended that retrospective tax legislation is presumptively

unconstitutional and that, without proper warnings to taxpayers, a retroactive

amendment can never pass constitutional muster. It was submitted that such a rule

would make of South Africa a rare exception, wholly at odds with the significant

democracies to which our Courts usually look for guidance.

72.

Canada:

In Canada, for example, there is no constitutional restriction on retrospective tax

legislation. It is worthwhile quoting Prof. P. W. Hogg, Constitutional Law of Canada

5TH Edition Supplemented Vol. 2, p. 51, where he says that the Canadian Charter
 108

permits retrospective legislation as “a proper tool for modern Government”. Prof.

Hogg has been frequently cited by the Supreme Court of Canada, and our

Constitutional Court. He said the following: “Apart from Section 11 (g) [which deals

with criminal offences] Canadian Constitutional Law contains no prohibition of

retroactive (or ex post facto) laws. Changes to the common law by judicial decision

are always retroactive. There is a presumption of statutory interpretation that a

statute should not be given retroactive effect, but, if the retroactive effect is clearly

expressed, then there is no room for interpretation and the statute is effective

according to its terms. Retroactive statutes are in fact common. For example, a

taxation law is often made retroactive to budget night, when the law was publicly

proposed; otherwise there might be room for avoidance action by tax payers during

the hiatus between the budget and the enactment of the law. Another common

example is a retroactive statute to change the law as it was declared to be in a

judicial decision: a law that has been interpreted in an unexpected way, or has been

held to be invalid on remediable grounds, may be amended or validated retroactively

to restore the legal position to what it had been believed to be prior to the judicial
 109

decision. The power to enact retroactive laws, were exercised with appropriate

restrain, is a proper tool of modern Government. Section 11 (g) diminishes this

power only by excluding the creation of retroactive criminal offences. Other kinds of

laws may still be made retroactive”.

It is my view that these views are of equal importance in the present proceedings.

73.

In British Columbia v Imperial Tobacco Canada [2005] 2 SCR 473 par. 69 and 70,

the Supreme Court cited Prof. Hogg with approval and reiterated at par. 71 that

retroactive tax is “not constitutionally barred”. It said the following: “The absence of a

general requirement of legislative prospectivity exists despite the fact that

retrospective and retroactive legislation can overturn settled expectations and is

sometimes perceived as unjust … Those who perceive it as such and perhaps take

comfort in the rules of statutory interpretation that require the legislature to indicate

clearly any desired retroactive or retrospective effects. Such rules ensure that the
 110

legislature has turned its mind to such effects and “determined that the benefits of

retroactivity (or retrospectivity) outweigh the potential for disruption or unfairness…”

The “unfairness test” is in my opinion not part of our law such as an isolated

concept.

74.

Applicant had submitted that the position in South African law is different than that of

Canada, because of the provisions of S. 1 (c) of our Constitution, which entrenches

the Rule of Law. Mr Trengrove SC however submitted that this did not actually

distinguish us from Canada, inasmuch as the Canadian Charter of Rights and

Freedoms also recognises in its preamble that Canada is founded on the Rule of

Law. In the Imperial Tobacco case supra, the Supreme Court of Canada also

considered whether the Rule of Law precluded retrospective legislation. It recognized

that the Rule of Law is “a fundamental postulate of our constitutional structure …

that lies at the root of our system of Government”. It accepted that “unwritten

constitutional principles” including the Rule of Law, are “capable of limiting

 111

Government actions”. Nevertheless, it rejected a submission that the Rule of Law

requires all legislation to be prospective (par. 63). It rejected the notion that the

Rule of Law precludes retrospective legislation, except in the case of Criminal Law.

Also in Singh v Canada (Minister of Citizenship of Immigration) [2006] 3 FCR 70

(9 December 2005), the Canadian Federal Court of Appeal reiterated that there was

no requirement of legislative prospectivity embodied in the Rule of Law, and that

Prof. Hogg accurately set out the state of Canadian Law in the paragraph that I have

quoted above.

75.

United States:

At a federal level, retrospective legislation, and retrospective taxation legislation in

particular, freely pass muster as long as it meets the threshold requirement of

rationality.

In United States v Carlton 512 US 26 (1994), Justice Blackmun cited the earlier

case of Welsh v Henry 305 US 134 (1938) 146 to 147, where it was said that no
 112

citizen enjoys immunity from the burden of paying costs of Government and

retrospective imposition does not necessarily infringe due process. In the context of

the rationality threshold requirement it was said that “provided that the retroactive

application of a statute is supported by a legitimate legislative purpose furthered by

rational means, judgments about the wisdom of such legislation remain within the

exclusive province of the legislative and executive branches …

To be sure, … retroactive legislation does have to meet a burden not faced by

legislation that has only future effects … the retroactive aspects of legislation, as

well as the prospective aspects, must meet the test of due process, and the

justifications for the latter may not suffer for the former … that that burden is met

simply by showing that the retroactive application of the legislation is itself justified

by a rational legislative purpose”.

Many years later, and certainly not coincidentally, the rationality of legislation was

deemed by the South African Constitutional Court to be the crucial question whether

legislation offended the Rule of Law, or a particular provision in the Bill of Rights.

See: Pharmaceutical Manufacturers of South Africa supra at p. 708 to 709.

 113

76.

The European Union:

In Hail v Rutgen [2001] QB 27 C (CA) par. 49, the English Court of Appeal

commented that “there is no general principle under the (European Convention on

Human Rights) that changes in Civil Law should not operate retrospectively”. As I

have said in paragraph above, the only basis upon which a retrospective taxation

statute can be challenged under the European Convention, appears to be Art. 1 of

The First Protocol to the ECHR, which is equivalent to our Section 25 (1) of the

Constitution. I have already referred to the decisions of Huitson in that context.

77.

England:

It was submitted that under English Law there has never been any bar to

retrospective tax legislation, and I have referred to a number of relevant decisions of

the Courts.
 114

In James v IRC [1977] STC 280 at 284, the following was said: “As the

Constitutional Law of England stands today Parliament has the power to enact by

statute any fiscal law, whether of a prospective or retrospective nature and whether

or not it may be thought by some persons to cause injustice to individual citizens”.

78.

Germany:

I have stated what the general position in Germany is, although it must be

remembered that the stare decisis doctrine plays no role in the German Courts, and

that they make decisions based on the particular facts of each individual case. It is

therefore not surprising that its Constitutional Court acts progressively, having regard

to the facts before it and the demands of society. What does seem clear however is

that the public interest element in each particular case plays an important role.

Proportionality is also considered in that balancing process.

Prof. Klaus Rennert, President of the German Administrative Court, presented a

paper on “The Protection of Legitimate Expectations under German Administrative

 115

Law”, to a seminar of the Association of Councils of State and Supreme

Administrative Jurisdictions of the European Union on 21 April 2016, in Vilnius,

Lithuania. On the topic of ‘The protection of legitimate expectations vis-à-vis the

legislature he said that if any restrictions are to be imposed on legislature, then –

leaving aside Union law – these restrictions may derive exclusively from

Constitutional Law. It was incumbent on the Constitutional Court to word such

restrictions. He then put the position in German law as follows:

“In its order of 11 May 1960, the Federal Constitutional Court introduced the

distinction between the so-called real (retroactive) and the so-called artificial

(retrospective) effects of laws which then went on to shape its further adjudication.

A retroactive effect was given, the Court held, “if the law intervenes with facts and

circumstances of the past that have already been completed,” by contrast, it held

that the retroactive effect was not an issue – or not in the same way – in cases in

which laws had only retrospective effect and “intervened only with facts and

circumstances or legal relationships given at present that have not yet been

completed and stipulate changes for the future.”

 116

a) According to the adjudication handed down by the Federal constitutional

Court, any retroactive effect will be in principle unconstitutional and

impermissible. As its reasons, the Court relied significantly on the concept of

protecting legitimate expectations, which it stated was a component of the

constitutional principle of the rule of law. In a state governed by the rule of

law, citizens are entitled to trust that their actions, taken in compliance with

applicable laws, will continue to be recognised by the legal order, along with

all of the legal consequences that originally had been tied to such actions. It

is not permissible for the state to retroactively strip its citizens of this legal

position or to devalue it. Nothing else applies if the lawmaker has enacted a

law for a limited period of time – such as a law providing for a tax benefit for

a certain number of fiscal years – even if this period has not yet expired and

the amendment of the law would concern only the remainder of that period.

However, there are a few exceptions to this principle, albeit within very

narrowly defined bounds. For the most part, the exceptions can be traced

back to the first of the three questions regarding the protection of legitimate
 117

expectations that we had posed at the outset, this being the question of

whether the citizen’s expectations actually merit protection. This is not the

case where a citizen had to count on the law’s being amended because a

corresponding announcement had been made in the political sphere, or

because the previous legal situation was unclear and confused and the

amendment served to obtain the necessary clarity. Likewise, the citizen’s

confidence does not merit protection where the amendment of the law or the

new provision of the law does not result in any disadvantage at all for the

citizen, such as when an invalid provision is replaced by a provision that

addresses the same substance and that is valid; the Court later expanded

this by stating that a citizen must accept as inescapable any petty

disadvantages. Finally, the Federal Constitutional Court also addressed the

second of our three initial questions and accepted the adoption of a

retroactive law, even in the face of contravening legitimate expectations

meriting protection, if the balance struck between the two shows that the

interest of the public in obtaining a new legal status outweighs the interests
 118

of the individual in seeing the previous status continue to apply. However,

the overriding nature of this public interest must be very clear; the Federal

Constitutional Court has stipulated “imperative grounds of general welfare” as

a pre-requisite, and it takes the meaning of the term “imperative” very

seriously.

b) While, according to what I have outlined above, the retroactive effect of law

is as a rule impermissible under constitutional law, the so-called artificial

retrospective effect is as a rule permissible. We need to keep in mind that

the Federal Constitutional Court uses this term to designate a law that

intervenes with processes that have commenced but have not yet been

concluded, and that does so with effect for the future. Such amendments of

the legal status are widespread and customary; prohibiting them would

paralyse all legislation. This is why the mere expectation that laws that

currently are valid will not change in future is not afforded any protection

under constitutional law. However, the principle of protecting legitimate

expectations has not lost its significance entirely in this context. On the
 119

contrary, it may force the lawmaker to create an appropriate transitional

provision that prevents too abrupt a modification of the legal situation and

absorbs its detrimental consequences for the parties (and their fundamental

rights) affected by way of complying with the principle of proportionality.”

79.

Although the German Basic Law (Grundgesetz) together with the Canadian Charter

of Rights formed the most important basis of the South African Bill of Rights, the

legal system of Germany as a whole, is vastly different to the South African

jurisprudential order. Our system, in the present context is more closely aligned, for

historical reasons mainly, to that of England, Canada and even the United States. (I

must emphasize: in the present context.)

The decisions on the present topic arrived at by the English Courts are in my view

particularly persuasive in this case where no direct decision on the constitutional

point raised by Applicant herein, has been as yet made by a South African Court.
 120

Applicant obviously prefers the “strict” approach of the German Constitutional Court

to completed transactions, in my view, and for the reasons stated, the less “rigorous:

or “strict” approach should be followed at least in the case of tax statutes where a

rational reason appears in the interest of the fiscus and where a broad warning was

given that a particular deficiency or “loop-hole” would be addressed by the

authorities.

80.

The test under our law:

It was submitted that the foreign law comparison makes it clear that retrospective

laws are permissible and indeed common place in countries based on the Rule of

Law. At the same time it was not suggested that Parliament may legislate with

retrospective effect as it pleases. The real question is what the standard is by which

the constitutional validity of retrospective legislation is judged. I agree with that

contention and intend to follow that approach.

 121

This question must be answered with reference to the standards of review laid down

by our Courts when the constitutional validity of a statute is challenged. There are

two main standards:

1. The first is the “rationality” test. This is the standard that applies to all

legislation under the Rule of Law entrenched in S. 1 (c) of the Constitution.

See: New National Party of South Africa v Government of the Republic of

South Africa and Others 1999 (3) SA 191 (CC) par. 19;

2. The second, and more exacting standard, is that of “reasonableness” or

“proportionality”, which applies when legislation limits a fundamental right in

the Bill of Rights. Section 36 (1) of the Constitution provides that such a

limitation is valid only if it is “reasonable and justifiable in an open and

democratic society”.

In Bobroff and Partners Inc v De La Guerre, South African Association of

Personal Injury Lawyers v Minister of Justice and Constitutional Development

2014 (3) SA 134 (CC) par. 6 to 8, the Constitutional Court explained the difference

between these two standards as follows: “The Constitution allows judicial review of
 122

legislation, but in a circumscribed manner. Underlying the caution is a recognition

that Courts should not unduly interfere with the formulation and implementation of

policy. Courts do not prescribe to the legislative arm of Government the subject-

matter on which it may make laws. But the principle of legality that underlies the

Constitution requires that, in general, the laws made by the legislature must pass a

legally defined test of ‘rationality’ …”, and

A rationality enquiry is not grounded or based on the infringement of fundamental

rights of the Constitution. It is a basic threshold enquiry, roughly to ensure that the

means chosen in legislation are rationally connected to the ends sought to be

achieved. It is a less stringent test than reasonableness, a standard that comes into

play when fundamental rights under the Bill of Rights are limited by legislation.

In those cases the Courts have a more active role in safeguarding rights. Once a

litigant has shown that legislation limits fundamental rights, the limitation may only

be justified under Section 36 of the Constitution. Section 36 expressly allows only

limitations that are “reasonable and justifiable in an open and democratic society

based on human dignity, equality and freedom””.

 123

81.

It was also submitted that there is an intermediate standard that applies in only one

context – when assessing whether a law violates the prohibition of arbitrary

deprivation property in Section 25 (1) of the Constitution. The Constitutional Court

held in First National Bank of South Africa Ltd T/A Wesbank v Commissioner for

the South African Revenue Service and Another 2002 (4) SA 768 (CC) par. 98 to

100, that a deprivation is arbitrary within the meaning of Section 25 if it is made

“without efficient reason”. This standard is more exacting than the rationality

standard in terms of Section 1 (c) of the Constitution, but less exact than the

reasonableness standard in terms of Section 36 (1) of the Constitution.

82.

The question then is which of these standards applies when a retrospective law is

enacted. It was submitted that the answer was quite clear: if the law limits a

fundamental right, the exacting “reasonableness” standard applies. If the law permits

a “deprivation of property” under Section 25 (1) of the Constitution, the intermediate

 124

standard of “sufficient reason” applies. If however, the law does not infringe upon the

Bill of Rights, then the question is merely whether it passes muster under Section 1

(c) of the Constitution, then the basic “rationality” standard applies. I agree with this

contention. Also, the rationality standard for a retrospective law to pass constitutional

muster happens to be the same that was applied by the United States Supreme

Court in the Carlton case supra. It is however a more exacting standard than that

set out under Canadian, English and European Law. I may add that it cannot be

seriously contended that none of these countries abide by the Rule of Law.

83.

Application to the facts:

It was submitted that the difficulty for Applicant was that once this rationality

standard applies, its case much inevitably fail.

There was an unintended loop-hole in the Income Tax Act created by S. 44 (9).

SARS considered on the basis of the possible Shoprite/Brait transaction, that a

“flood” of such transactions would occur and that there was a real risk the national
 125

fiscus would suffer extensive and permanent harm. This transaction -, which was

pending when the decision was made to close the loop-hole, would have resulted in

a loss of approximately R 1.5 billion in unpaid STC.

The Applicant complained that the evidence on flood of other transactions that was

anticipated, was vague and also hearsay. In answer thereto, it was contended that

this complaint misses the point. Parliament is not required to wait for iron-clad

evidence before it enacts to prevent damage to the fiscus. On the contrary, it is

perfectly appropriate for it to act proactively, whenever it fears such damage may

occur.

In Minister of Home Affairs v NICRO 2005 (3) SA 280 (CC) par. 35, the Courts

stressed in a Section 36 limitation context: “There may for instance be cases where

the concerns to which the legislation is addressed are subjective and not capable of

proof as objective facts. A legislative choice is not always subject to Courtroom fact-

finding and may be based on reasonable inferences unsupported by empirical data.

When policy is in issue it may not be possible to prove that a policy directed to a

particular concern will be effective. It does not necessarily flow from this however,
 126

that the policy is not reasonable and justifiable. If the concerns are of sufficient

importance, the risk associated with them sufficiently high, and there is sufficient

connection between means and ends, that may be enough to justify action taken to

address them”.

84.

It was therefore submitted that it was eminently rational to close the loop hole with

retrospective effect. A mere prospective amendment would have encouraged tax

payers to exploit the loop-hole in the last few months before the loop-hole was

closed and this was also the view of Prof. Hogg in the Canadian context, where he

made it clear in Constitutional Law of Canada 5TH Edition supra Vol. 2 p. 51, that

“A taxation law is often made retroactive to budget night, when the law was publicly

proposed; otherwise, there might be room for avoidance action by tax payers during

the hiatus between the budget and the enactment of the law”.
 127

85.

Mr Trengrove SC submitted that the Applicant might well feel that it was unfairly

treated as a consequence of the change, but in the absence of a breach of

fundamental rights, this is simply irrelevant.

In Law Society of South Africa and Others v Minister of Transport and Another

2011 (1) SA 400 (CC) par. 5, it was said that “The requirement of rationality is not

directed at testing whether a legislation is fair or reasonable or appropriate. Nor is it

aimed at deciding whether there are other or even better means that could have

been used. Its use is restricted to the threshold question whether the measure the

law giver has chosen is properly related to public good it seeks to realize …”

In the light of these considerations, all that remained is the Applicant’s complaint

that the manner in which parliament ultimately closed the loop-hole differed from the

manner in which the Minister had originally foreshadowed in his budget address. I

have already referred to the relevant facts in this context, and the process that was

followed.
 128

It was contended that the amendment ultimately was made less drastic than the one

originally proposed, because it maintained the exemption in Section 44 (9) and

merely qualified it by the new Section 44 (9A). It moreover closed the loop-hole in

line with the underlying purpose of Section 44 to ensure that amalgamations are

STC neutral. The Applicant complained that the distribution of its share premium

was not taxable when it was made on 3 May 2007. This is correct. But, the

Applicant had been notified at the time that SARS intended to plug the loop-hole by

the repeal of Section 44 (9). If SARS had implemented its plan, “OLD CO’s”

distribution of the consideration shares to its shareholders, would have been subject

to STC. The net effect of the change of plan was accordingly merely that “OLD CO”

was not liable for STC on its distribution of the consideration shares, but “NEW CO”

became liable for STC on its distribution of its newly acquired share premium. What

this means is that even on Applicant’s own test, it cannot succeed. It contended

there had been no adequate warning, but this not so in the context of the budget

speech of 21 February 2007 and the Draft Bill published on 27 February 2007. All

tax payers were thus given ample notice that, to put it at its lowest, they could not
 129

safely rely on Section 44 (9) of the Income Tax Act after 21 February 2007. The

question of whether the Applicant and its advisors actually made themselves aware

of the budget speech or Draft Bill is in the present context not relevant.

I agree with that approach. I am not aware of any provision in any of the

jurisdictions that I have referred to, or indeed in ours, to the effect that the warnings

given must relate to the exact same amendment that is ultimately made. To adopt

such an approach would undermine the parliamentary process and the public

participation process completely. It would also mean that parliament would be bound

by an announcement made by the executive. Applicant had already suggested that I

do not need to find how precise a warning in this context must be, inasmuch as in

the present proceedings, no warning at all had been given. I do not agree with this

contention, the facts show otherwise, and it loses sight of the fact that there may be

cases where no warning needs to be given at all. I am therefore not of the opinion

that a precise warning must be given in each and every case, nor that a warning, of

whatever ambit, needs to be given in all cases. In my view, a proper approach

would be to judge the legality of retrospective amendments on a case-by-case

 130

basis, having regard to the various considerations that I have referred to. The

Constitution itself certainly does not prohibit retrospective legislation in civil law.

86.

Mr Trengrove SC also contended that because Applicant brought this matter for

purely commercial gain, it should be directed to pay the costs of SARS including

costs of two Counsel, and in this context relied on Weare v Ndebele N. O. 2009 (1)

SA 600 (CC) par. 78.

87.

Second Respondent’s argument on the constitutional challenge:

Mr I. Semenya SC and his juniors also handed up comprehensive Heads of

Argument for which I thank them. In general it was Second Respondent’s position

that Parliament was competent to pass legislation, prospective or retrospective, in

accordance with procedure and within the limits of the Constitution, provided that
 131

such law was not arbitrary or capricious. Mr Semenya SC then dealt with the

argument under the following headings:

88.

The constitutional framework:

The Constitution obviously does not prohibit the passing of legislation in the civil

sphere that has retrospective effect, and this ought to be the starting point in any

argument. In the criminal sphere, the position is of course different and retrospective

operation is proscribed in the context of the provisions of S. 35 (3) (l) of the

Constitution.

89.

The Rule of Law:

I have referred to the relevant principles and it is also clear that as yet there has not

been any test on whether the retrospective operation of a statute is inconsistent with

the Rule of Law. Mr Semenya SC also referred me to the Imperial Tobacco

 132

Canada decision supra, as well as to the decision in Air Canada v British

Columbia [1989] 1 SCR 1161, where a majority of the Supreme Court of Canada

confirmed the constitutionality of a certain Gasoline Tax Act, which retroactively

taxed certain companies in the Airline Industry.

90.

It is clear from the Canadian decisions that their parliament has an unfettered

discretion in deciding the effective dates of new tax laws provided that the intended

retrospective effect must be clear and unequivocal.

In Gustavson Drilling (1974) Ltd v Minister of National Revenue [1977] 1 SCR

271, an oil company tried to deduct in 1964 drilling expenses incurred prior to

1960. Tax laws passed in 1962 had repealed the right to claim such deductions for

tax years following 1962. The majority of the Court upheld the legislation on the

basis that “no one has a vested right to continuance of the law as it stood in the

past” (at 282).

 133

In Attorney General of Quebec v Irvin Toy Ltd [1989] 1 SCR 927 (SCC), it was

amongst others held that S. 7 of the Canadian Charter does not guarantee a right

to the permanence of a statute. It was contended that similarly, and because our

Constitution was greatly inspired by the Canadian Constitution as it was put, the Bill

of Rights in the South African Constitution does not guarantee a right to the

permanence of a statute, nor does it protect economic rights.

The submission therefore was that there was no reason why South African Courts

should not follow the position in Canada and ultimately, where legislation expressly

states that it is to operate retrospectively, the only enquiry is whether the legislation

was passed in accordance with restrictions of the Constitution.

91.

United States:

The United States Supreme Court has to date dismissed all due process challenges

relating to retroactive income tax provisions. Reference to due process in this

context is the equivalent of the Rule of Law.

 134

In Pension Benefits Guarantee Corporation v R.A. Grey and Co 467 US 717

(1984), the Supreme Court held that if the retrospective statute in economic policy is

“supported by legitimate legislative purpose furthered by rational means, judgments

about the wisdom of such legislation remain within the exclusive province of the

executive and the legislative branches. The Court said that “strong deference”

should be accorded to economic legislation and this was no less applicable when

that legislation is applied retroactively. In the same case, it was also stated that it

would be doubtful that retrospective application of an Amendment Act would be

invalid under the due process clause even if it was suddenly enacted without any

period of deliberate consideration.

92.

I have already referred to the decision of Welsh v Henry supra, wherein it was

pointed out that taxation is not a penalty, but rather a distribution of the burden

which those who enjoy the benefit in a State must carry. It also pointed out (at 149)

that there may be a particular need for allowing income tax retroactivity inasmuch as
 135

“experience has shown the importance of reasonable opportunity for the legislative

body, in the revision of tax laws, to distribute increased costs of Government among

its tax payers in the light of present need for revenue and with knowledge of the

sources and amounts of the various classes of taxable income during the taxable

period preceding revision. Without that opportunity accommodation of the legislative

purpose to the need may be seriously obstructed if not defeated”. It also held (at

146) that no tax payer could “assert surprise, or complain of arbitrary action in the

retroactive apportionment of tax burdens to income at the first opportunity after

knowledge of the nature and amount of the income is available”.

In US v Carlton supra, it was also held that retrospective closing of tax loop-holes

is something that individuals might expect parliament to do upon discovering them.

Relying on loop-holes remaining open, or relying on closure being prospective only,

was not a rational expectation. As to the question of what actually constitutes

adequate notice, the Carlton majority asserted that “lack of notice is not dispositive”.
 136

93.

In the context of the Applicant’s assertion that knowledge of the amendment was

essential in order to regulate its affairs, reference can usefully be made to the

Carlton decision supra, the Benefits decision supra and also to W. David

Slawson in Constitutional And Legislative Considerations In Retroactive Law-

making, 48 CAL. Law Review 216 (1960) at 226, where the author said the

following: “Reliance on existing rules therefore must be sacrificed to some extent to

the need for change. It is this basic and simple conflict that is often overlooked in

writings on retroactivity. The conclusive distinction between valid and invalid

retroactive law is the element of “surprise” – whether it serves to give effect to or

defeat the bona fide and reasonable expectations of the person it affects. Such a

test is far too rigorous since it ruled out all but the most inconsequential legislative

changes”. I agree with that reasoning in the present context as well.

It is therefore clear from the mentioned authorities and if one follows the approach of

the United States, the impugned provision herein would not be inconsistent with the

Constitution. In the United States the crux really is whether the legislation was
 137

rationally related to a legitimate legislative purpose and that is of course also the test

in South Africa, as has been said, on numerous occasions.

94.

United Kingdom:

I have dealt with a number of relevant authorities on the present topic above.

In the matter at hand, the language is in my view clear and Applicant did not

contend that it was not clear. In cases of unclarity the position in English Law seems

to be that greater emphasis is then placed on the question of unfairness, and in

particular the degree thereof. In the present instance, unfairness is in my view

certainly not the decisive question. Many laws may be “unfair” in many particular

instances, but they are not unjust if they have been lawfully and constitutionally

passed by Parliament. Laws are in general aimed at the broader public and are in

the main not concerned whether or not a particular statute unfairly affects a

particular individual simply on the basis that it may not unfairly affect the majority of

the populace.
 138

95.

Mr Semenya SC submitted that in the present instance there was ample evidence in

the budget speech and the press release of 21 February 2007, that future

legislation was contemplated that would have retrospective operation, and that such

legislation would target amalgamation transactions which result in the STC lost to

the fiscus. Mr Semenya SC submitted that inasmuch as the UK decisions seem to

strike a fair balance between the taxpayer and the public or State interest in

securing payment of taxes, the Applicant in this case did not use it as a basis for the

challenge, and there was also no reason to import such a requirement which on its

terms limits the legislative power inconsistent with Section 44 of the Constitution.

96.

Australia:

In Deputy Commissioner of Taxation v Zammitt [2012] NSWDC 135 at par. 36, the

Court held that the Government had the power to enact retrospective legislation
 139

even when that legislation impacts on the results of proceedings currently pending in

the Courts.

In Bawn (Pty) Ltd v Metropolitan Meat Industry Board [1971] 92 WN (NSW) 823 at

842, the Court said: “Once it is accepted that the general principle of construction

recognises that a statute may operate retrospectively so as to disturb and alter

substantive rights which accrued before the commencement of the statute, provided

that the statutory intention … is manifested with sufficient clarity …it is not easy to

see why any different rule should be applied to the possible operation of a statute on

rights which have already accrued …”

The enquiry therefore is whether the statute is couched with “sufficient clarity” that it

is to have retrospective effect and if no doubt arises, the statute must be given effect

on its own terms.

See also: Attorney General of New South Wales v World Best Holdings Ltd and

Others [2005] NSW CA 261.

A word of warning was also given by Mason P in that decision (at par. 153). He

was troubled with the invocation of “unfairness” or a scale of unfairness or injustice,

 140

“Courts have no mandate to construe legislation by reference to perceptions of

morality that are not already firmly embedded in fundamental common doctrines or

the statute itself”. I would agree with that approach in general terms. In the present

case, Applicant in any event, never argued that its particular circumstances were so

grossly unfair and oppressively affected that the amendment could not be regarded

as rational as a result. Nor was the amendment aimed solely at itself.

It was submitted therefore that Australian jurisprudence therefore does support the

Applicant’s case, and I agree.

I have already referred to the decisions of the European Court of Human Rights, and

its approach also does not support the position of the Applicant herein.

97.

Mr Semenya SC, in the context of Applicant’s contention that knowledge of

proposed retrospective amendment to the law is fundamental to the Rule of Law,

pointed out that there does not appear to be any authority on which this proposition

rests. Certainly, our Constitution does not have this constraint. There was also no
 141

foreign law which required knowledge of the proposed retrospective amendment to

be there and more so, that such knowledge would be fundamental to the Rule of

Law.

Similarly, Applicant did not provide any authority for their contention that “knowledge”

or “adequate warning” is constitutionally required for tax legislation to pass

constitutional muster. In any event, if it were to be found that such “knowledge” or

“adequate warning” was essential, it was submitted that the process that was

followed, and I have given all relevant details, was sufficient and ought to have put

any taxpayer who was contemplating amalgamation transactions with a view to

derive STC exemption from such, would have been placed on full guard that

legislation was going to be amended to remove the particular exemption. Despite

this warning, Applicant went ahead with the amalgamation transaction. There was in

any event no authority for the proposition that a specific notice was required, and the

reason was obvious: it is the executive who initiates legislation and it is Parliament

which passes the legislation sometimes with an amendment and certainly after

public participation processes in any event.

 142

I have no hesitation in agreeing with these submissions for the reasons stated.

98.

Should the Notice mirror the actual amendment?:

It was argued that Applicant offered no authority for the proposition that “warning” of

retrospective legislation must “self-evidently pertain to the actual amendment that is

implemented”. That is of course correct. Applicant however also contended that I

need not decide this inasmuch as in the present sentence no “warning” was given at

all, and I have already said that I do not accept that on the facts. I have already

given my reasons why a notice need not mirror the actual amendment intended, if it

is given in advance of any public-participation process and parliamentary

proceedings.
 143

99.

Reasons for retrospectivity/justification:

There is no authority for the proposition that retrospective tax legislation would

survive constitutional scrutiny only if there were “good reasons” for it. It is not for a

Court to say what a good “reason” is. Foreign law also does not support such an

approach. The only question is whether a legitimate legislative purpose is indicated.

In the present case, the Government’s purpose was to remove the tax exemption in

amalgamated transactions. To do so retrospectively was also justified, because

there was loss of STC revenue rising from amalgamations which was previously

intended to be deferred and not permanently lost. More importantly, there was a

general announcement that the intended amendment would remove that loop-hole.

That was sufficient. I agree with Mr Semenya SC’s submission in this regard, and I

have given my reasons.

 144

100.

Complete or incomplete transactions:

The language of the present amendment is clear, it refers to “all” transactions. It was

also submitted that it was immaterial whether a transaction was completed or not if it

falls within the period of the retrospective operation of that legislation. All the foreign

judgments, to which reference had been made, were concerned with completed

transactions. It appears to me that the German Constitutional Court however, does

make a general distinction between completed and uncompleted transactions, but

even in those instances there are sensible exceptions. Not surprisingly I must add,

in that modern jurisprudence should in my view never be dogmatic, especially not in

the field of fiscal legislation, as economic considerations seem to be presently in a

constant state of fluidity, and not only in South Africa.

 145

101.

Exceptional circumstances and rigorous approach:

There is in my view no basis for holding that under the present Constitution,

Parliament can only pass retrospective legislation if “exceptional circumstances”

exist. I also do not agree that a Court is obliged to adopt a “rigorous approach,

which would require “a very high level of correlation” between the changes to the

law of which the taxpayer has been notified and the actual legislative amendment

that follows, as Applicant contended for. There is no authority for this proposition

and in any event, I do not agree with it, inasmuch as there is clearly no such

constraint on the powers of Parliament at all. Nor would such approach be practical

in the context of tax statutes.

102.

I agree with Mr Semenya SC’s conclusion that the constitutional attack on the

impugned provision must fail. There is nothing in our Constitution which prohibits

parliament from passing retroactive or retrospective legislation. There is nothing in

 146

other jurisdictions of similar constitutional structure that prohibits such passing. Also,

and more significantly, there is nothing internal in the Rule of Law which renders

retrospective legislation per se unconstitutional.

103.

Applicant’s property challenge: Section 25 (1) of the Constitution:

In a Supplementary Affidavit, Applicant sought to establish a further cause of action

based on Section 25 (1) of the Constitution.

This challenge would arise only in the event that the Applicant’s interpretational

argument was rejected, and Section 34 (2) of the Tax Amendment Act is held to

have retroactive effect to the completed transaction.

104.

This challenge is based on the fundamental right to property proceeds on the basis

that the retroactive removal of the exemption from STC in par. (f) of the definition

“dividend” without adequate notice would have amounted to a deprivation of property

 147

that was both procedurally and substantively arbitrary and thus inconsistent with

Section 25 (1) of the Constitution states:

1.) “No one may be deprived of property except in terms of law of general

application, and no law may permit arbitrary deprivation of property.

2.) Property may be expropriated only in terms of law of general application –

a) for a public purpose or in public interest; and

b) subject to compensation, the amount of which and the time and

manner of payment of which have either been agreed to by those

affected or decided or approved by a Court.

3.) The amount of the compensation and the time and manner of payment must

be just and equitable, …

4.) For the purposes of this Section –

a) …

b) property is not limited to land.”

 148

105.

It was submitted that the retroactive application of Section 34 (2) to completed

transactions would be that Applicant, and any similarly situated taxpayers, became

obliged to pay the relevant STC. This would then amount to a deprivation of property

for purposes of Section 25 (1) of the Constitution.

“Deprivation” of property was broadly defined in First National Bank of South Africa

Ltd T/A Wesbank v Commissioner for the South African Revenue Service and

Another 2002 (4) SA 768 (CC) par. 57, as being any interference with the use,

enjoyment or exploitation of private property.

Also, in National Credit Regulator v Opperman 2013 (2) SA 1 (CC) par. 66, where

it was stated that “interference significant enough to have a legally relevant impact

on the rights of the affected party amounts to deprivation”. It was said that reference

to this decision was appropriate to the creation of an unavoidable obligation to pay

money where none previously existed resulted in the deprivation of money by the

State.
 149

106.

In First National Bank of South Africa v Minister of Finance supra, Ackermann J

was of the view (par. 100), that a deprivation of property is “arbitrary” as meant by

Section 25 when the law referred to in Section 25 (1) does not provide sufficient

reason for the particular deprivation or is procedurally unfair.

This approach to Section 25 (1) was then confirmed in Mkontwana v Nelson

Mandela Metropolitan Municipality 2005 (1) SA 530 (CC) at par. 34.

107.

It was therefore submitted that in the present instance both the procedural and

substantive requirements of non-arbitraries have not been met. The retroactive

application of Section 34 (2) would be procedurally arbitrary, because it would

deprive the Applicant and others of property in circumstances where they were given

no fair opportunity to organize their affairs. Applicant however did not contend that

any retroactive application of taxation legislation violates Section 25 (1). Rather, its

complaint is confined to retroactive application of legislation in circumstances where

 150

the public was not previously given reasonable notice of the intended legislation and

thus a fair opportunity to organize their affairs.

I have already held that:

i) there is no over-riding duty to give notice irrespective of the facts;

ii) in the present instance there was sufficient notice of general impact;

iii) there is similarly no over-riding duty to give notice that states

precisely what the intended legislation will entail.

108.

On behalf of First Respondent it was argued that:

i) the impugned provisions do not give rise to a deprivation of property;

ii) even if they did, the deprivation was not arbitrary;

iii) Section 36 (1) of the Constitution permitted a limitation of rights.

Applicant’s argument seems to proceed from the premise that a person who incurs a

liability imposed by law, suffers a deprivation of property. Section 25 (1) is intended

to deal with situations where the law takes away or interferes with the use and
 151

enjoyment of assets. The fact that a law creates a civil liability does not in itself

deprive the taxpayer of property unlawfully. If it were otherwise, every tax, levy, fee,

fine and administrative charge would constitute deprivations for purposes of Section

25 (1).

109.

In Mkontwana supra, the majority held at par. 32 “Whether there has been a

deprivation depends on the extent of the interference with or limitation of use,

enjoyment or exploitation. It is not necessary in this case to determine precisely

what constitutes deprivation. No more need be said than that at the very least,

substantial interference or limitation that goes beyond the normal restrictions on

property use or enjoyment found in an open and democratic society would amount to

deprivation.”

See also: Offit Enterprises (Pty) Ltd v Coega Development Corporation (Pty) Ltd

2011 SA 293 (CC) at par. 39, where this dictum in Mkontwana supra was

confirmed, as it was in Shoprite Checkers (Pty) Ltd v MEC for Economic

 152

Development, Eastern Cape 2015 (6) SA (CC) at par. 14 and 73, Tshwane City v

Link Africa and Others 2015 (6) SA 440 (CC) at par. 58, 168 and 173, Agri SA v

Minister for Minerals and Energy 2013 (4) SA 1 (CC) at par. 67.

110.

It was therefore submitted that Applicant had to establish that the impugned

provisions give rise to a substantial interference with property rights that go beyond

the normal restrictions on property use or enjoyment in a democratic society. In my

view it cannot be argued that all taxes involve a “deprivation” of property, in the

context of Section 25 (1). A State cannot exist without taxes. Society receives

benefits from them. Taxes are not penalties. Neither can they be, without any

qualification, be regarded as unjust deprivation of property use. If it is Applicant’s

view that only retroactive taxation gives rise to such deprivation, then again, no

unjust deprivation occurred here. The State used a well-accepted mechanism to

close a loop hole in a statute. It did not solely target the Applicant. Its purpose was

rational. It gave ample warning of its intention. The retroactive amendment does in
 153

my view also not amount to illegitimate deprivation. Sufficient reason was

established and the process was fair in the present context, not that “fairness” is a

requirement.

See: Reflect-All 1025 CC v MEC for Public Transport, Roads and Works,

Gauteng Provincial Government 2009 (6) SA 391 (CC) at par. 33.

In my view it is clear, as I have said, all tax payers were given sufficient notice, to

put it at its lowest, that they could not safely rely on Section 44 (9) of the Income

Tax Act after 21 February 2007. Applicant sought to distribute its shares in reliance

on the Section 44 (9) loop-hole, even after it had been publicly announced that this

would be closed with retrospective effect. The fact that this was done in a manner

different to that which was initially contemplated does not render the process

procedurally unfair. Were I to hold otherwise, the democratic parliamentary and

public participation process would be seriously undermined. The executive does not

make laws.

See: Oriani-Ambrosini v Sisulu, Speaker of the National Assembly 2012 (6) SA

588 (CC) par 46 to 49.

 154

It can in my view never be seriously argued that when a Minister announces

proposed retroactive legislation to cure a problem, parliament is effectively precluded

from adopting a different method. In the present instance, various stakeholders

greatly resisted the Minister’s proposal and as a result, the original method to close

the loop-hole was altered. The amendment adopted by parliament was not arbitrary

and therefore not in breach of Section 25 (1) of the Constitution. Further, and in

any event, the amendment was reasonable and justifiable in terms of Section 36 (1)

of the Constitution.

111.

The property challenge can therefore not be upheld.

112.

Costs:

The Respondents sought a cost order against the Applicant. I have a discretion

which must be exercised judicially, having regard to all relevant circumstances. One
 155

such consideration is the general rule that in constitutional litigation an unsuccessful

litigant ought not to be ordered to pay costs. This is not an inflexible rule. The

litigation is neither vexatious nor frivolous. It involves a topic that has not yet

squarely been decided by a Court in South Africa. The Applicant was justified in

seeking clarity on the constitutional issues.

See: Affordable Medicines Trust v Minister of Health 2006 (3) SA 247 (CC) at

par. 296 to 297.

113.

The following order is therefore made:

1. The application is dismissed;

2. There is no order as to costs.

_____________________________
JUDGE H.J FABRICIUS
JUDGE OF THE HIGH COURT GAUTENG DIVISION, PRETORIA
 156

Case number: 87760/2014

Counsel for the Applicant: Adv M. Chaskalson SC

Adv M. W. Janisch SC

Adv P. Seseane

Instructed by: Edward Nathan Sonnenbergs Inc

Counsel for the 1st Respondent: Adv W. Trengrove SC

Adv S. Budlender

Instructed by: Van Zyl Le Roux Inc.

Counsel for the 2ND Respondent: Adv I. Semenya SC

Adv N. Mayet

Adv Z. Gumede

Instructed by: The State Attorney

Date of Hearing: 15 - 16 March 2017

Date of Judgment: 29 May 2017 at 10:00