Introduction

1. Detailed below are Pricewaterhouse Coopers’ written submissions on the themed sections of draft legislation, and related explanatory memoranda, issued on 30 September 2004, and intended to form the Revenue Laws Amendment Bill, 2004.
2. For ease of reference we have collated our comments in the same manner as the annexures of themed provisions issued by SARS.
3. We highlight that we have also taken this opportunity to raise certain other issues, which we consider are not currently addressed by the draft legislation.
1. To the extent we consider these other issues to be anomalies, or only minor in nature, we have included these within the relevant annexure, most notably in Annexure 13, which contains the textual and technical amendments (though we have sought in each such instance to identify the fact that the proposed draft legislation does not raise the particular point).
2. To the extent we consider the issues raised may require a greater degree of consideration from SARS and/ or National Treasury we have included them in a further Annexure (16).
4. Unless otherwise stated, all references to section numbers throughout this document are to sections in the Income Tax Act, 1962.
5. Again we thank both SARS and National Treasury for the ongoing opportunities afforded us to participate in the development of our country’s tax legislation. In addition, from an overall perspective, though we have identified various concerns, we would congratulate both SARS and National Treasury on draft legislation which is on the whole understandable, justifiable and well considered. The fact that no changes of the magnitude seen in recent years, in terms of complexity and the introducing of new concepts, is also welcomed.
6. The table on the following page highlights those annexures in respect of which we have raised comments at this time.

Yours faithfully

David Lermer James Aitchison

Tax Director Senior Tax Manager

Subsection (1)

We would highlight that the proposal to include the amounts considered in this subsection within "income" is not possible in terms of the current definitions in the Act.

"Income" as defined is any residual amount of gross income after deducting any amounts exempt under Part I of Chapter II. No provision is made in its definition for the inclusion of other items such as those considered in section 8B(1).

This can be contrasted to the definition of "taxable income" which includes, at (b) of the definition, specific provision for the inclusion of other amounts in terms of the Act.

Placing such amounts directly into taxable income is however not considered viable or desirable as this would preclude the deduction of any related costs of disposal.

Although it would be possible to amend, in section 1, paragraph (i) to the definition of gross income to include reference to section 8B in the same manner as it currently does for section 8A we suggest that it is more simple and effective to replace the reference in to "income" section 8B(1) with a reference to "gross income".

Likewise you will note from our comments in respect of Annexure 2 that we recommend a similar approach be adopted in respect of section 8A.

Subsection (2)(b)

We are concerned that the 90 per cent test as currently framed may deny the relief (to both employees and companies) in circumstances where this is not appropriate.

Specifically we envisage situations whereby a group of companies is structured such that one employment company is used to enter into the employment contracts with all personnel of a number of, or all of, the group’s subsidiaries, e.g. to reduce payroll and administration costs and ensure consistency in evaluation of performance criteria etc.

It may well be that broad-based employee share plans are to be implemented for only certain of the operating subsidiaries, and then only to the employees who, although employed by the employment company, carry on work in respect of such operating subsidiary.

In such instances the legal employer may well be different to the company issuing the shares and provision requires to be made for such eventuality.

By way of example, a holding company (A) currently holds all the shares in each of its five subsidiaries, B, C, D, E and F.

F is an employment company and the employment contracts for the group’s 410 personnel are with F. The employees relate to the group companies as follows; A (5), B (100), C (100), D (100), E (100) and F (5).

It is intended that broad-based employment share plans be implemented in respect of each of B, C and D (and not A or E), targeted in each instance to all the employees whose work relates to those companies.

However it would appear that as these plans would cater only for 300 out of F’s 410 total employee roll (some 73%) that none of the shares granted in any of B, C and D will qualify, the 90% test in section 8B(2)(b) not being met in respect of "that employer" (F).

As respectively 100% of the "employees" in each of B, C and D are in fact granted shares in those companies this appears to be an unfortunate consequence of a group structuring its employment aspects, and which in our view requires resolution.

Subsection 2(d) – preamble

The explanatory memorandum on Annexure 1 makes it clear that "one or more" of the restrictions envisaged in section 2(d) may apply. However we are concerned that the section as currently framed does not achieve this effect, referring as it does to "other than – (i), (ii) or (iii).

We therefore suggest that the wording of the preamble be amended to read

"……. equity shares, other than any or all of the following – "

We refer you also to our comments below in respect of paragraph 11A of the Fourth Schedule where we have identified a fourth restriction which we consider should be permitted.

Subsection (2)(d)(ii)

We recommend that the reference in subsection (2)(d)(ii) to "the employer or that group company" be amended to "the employer or any other company within the same group of companies" in order to avoid a group structure becoming complicated with cross holdings.

For example, assume a group consists of a holding company A, and its two directly owned subsidiaries, B and C.

B is the group’s employment company, employing all the group’s personnel for all subsidiaries. B, as employer, grants certain employees qualifying equity shares in C in terms of a broad-based employee share plan.

Should it transpire that these employees wish to dispose of their shares and there is a right of acquisition by the group, it may well be preferable for A to acquire those shares so as to retain all non-employee held shares within one company in the group.

The wording currently proposed permits only B (the employer) or C (the company in whom the employee holds the qualifying equity shares) to acquire those shares from the employee.

It is our view that such restriction is not necessary and an acquisition by A (as holder of the C shares in the group) should be permitted.

Subsection (2)(d)(iii)

Subsection (2)(d)(iii) requires to be made subject to subsection (d)(i) and (ii) so that the right of the employer’s group of companies to administer the shares or to acquire them is not limited to the timeframes in (d)(iii). We recommend that this be achieved through the inclusion of a proviso to (d)(iii) stating;

Location of deduction

Again our concern here focuses on the appropriate treatment in a group of companies, particularly where a group utilises an employment company which undertakes all the employee contracts.

Assume again a simplified group structure, a holding company A, holds all the shares in both an employment subsidiary B and an operating company C. B has the employment contracts for all group personnel, including all employees who undertake the work of C.

A broad based plan is entered into, with employees being granted shares in C, the operating company. The shares granted are sourced from the shares currently held by A.

Under the proposed section 11(lA) it would appear that any deduction would be granted to B as employer ("employee of that person").

In addition, the introduction of the clarity sought should prevent any opportunity for multiple deductions in different companies within a group of companies, in respect of the same shares granted, through the utilisation of split-employment contracts.

Limitation on any deductions

Whilst the intent behind the reference to "in lieu of any other deduction" is both clear and readily justifiable, as currently framed it could be said that no deduction will be available in respect of section 8B unless another deduction is first available, in respect of which the effective section 8B deduction could be said to be in lieu of.

Accordingly we recommend that ", if available," be inserted after "other deduction" in order to ensure the relief operates as intended.

Subparagraph (6)

It appears to us that the inclusion of the proposed requirement is at odds to the detail in the explanatory memorandum whereby it is recognised, in the second paragraph under the heading "Subsequent sales" that certain employees may be unaware of their tax obligations on disposal.

We would suggest that rather than placing the onus on the employee to inform the company, that an obligation is placed on the employee to provide details to the company upon request by the company. The company will in any event learn of the disposal when its shareholders register requires to be updated and can then write to the employee, advising the employee of such recommended obligation, requesting the information.

It is recognised that the changing of such reporting obligation in this manner will necessitate a further amendment to section 69 to cover also subsequent disposals.

This leaves the requirement to report disposals to SARS on the company, which should be in a better position to comply with such condition.

This would however require to be added as a further permitted restriction in terms of section 8B(2)(d), as the stipulation, in terms of the employee share plan, of an obligation to provide details of subsequent sales to the company on request could be said to be a restriction imposed by the employer as envisaged in section 8B(2)(d).

We consider that the "Result" detailed in respect of the "Example 2" provided could be misleading as it refers to the "Company only being required to report the transaction."

The relevant taxpayer, Mr A, will also be required to report the disposal (transaction) on his tax return form, if any, for the year, and we suggest this is included in the detail.

Stamp duty/ uncertificated securities tax

In order to provide full tax relief for these broad-based employee share plans it will be necessary to provide relief from stamp duty (or uncertificated securities tax ("UST") where appropriate) on both the issue of shares in a broad-based employee share plan or the transfer of shares under such a plan. This is particularly so where the shares granted to the employee are transferred from an existing group company, as opposed to sourced

from the issue of new shares, as in such instances the associated tax liability would otherwise fall on the employee as transferee/ recipient.

Although the shares are to be granted at nil consideration such that the associated stamp duty/ UST liability could also be nil, both the Stamp Duties Act, Act 77 of 1968, ("the Stamp Duties Act") and the UST Act, Act 31 of 1998 ("the UST Act"), provide that the in certain instances that market value will apply, see for example section 3(1) of the UST Act and paragraph (3)(h)(i) of Item 15 to Schedule 1 of the Stamp Duties Act.

In order to provide certainty therefore w recommend that exemptions be granted in terms of paragraphs 1 and 2 and 3 of Item 15 to Schedule 1 of the Stamp Duties Act, and Section 6(a) and (b) of the UST Act.

General

We are concerned that the draft Bill results in a potential exposure to double taxation from the possible interpretation of treating an option as a separate and distinct restricted equity instrument such that there is taxation both on the market value of the option terminating and then on the value of the underlying share.

This double taxation could also be said to arise in the case of an unrestricted option, where it appears that both the grant of such an option could be taxed and the acquisition of the underlying share.

In addition we point out that Schedule 7 requires to be updated to take account of the wider ambit of the proposed section 8A. In particular the current prohibitions on Schedule 7 ‘s application where there is an "exercise" within section 8A of any right to acquire any marketable security requires to be extended to cater for all other instances where a charge arises, e.g. the grant of an option.

Finally it is essential to ensure that instruments granted in respect of options which were granted under the current section 8A, prior to amendment, are excluded from the ambit of the revised section.

Subsection (1)(a)

We refer you to our comments above in respect of section 8B(1)(a) and recommend that the reference to "income" be replaced with a reference to "gross income" and that as a consequence of this the current reference to section 8A in paragraph (i) of the definition to gross income be deleted.

Subsection (3)(b)(iii)

We would recommend that, for purposes of certainty and clarity, the meaning of "disposes" be cross-referenced to the definition in paragraph 1 of the Eighth Schedule.

Subsection (5)(a)(i)

The reference to at arm’s length can give rise to uncertainty and we would suggest that in the context of section 8A the preferable terminology to apply is "market value".

An arm’s length price need not be the market value of an asset, and the oft used example is that of an antiques dealer who offers someone a low price for an item which they accept thereby resulting in an arm’s length transaction even though the dealer knows the true market value of the item is much higher.

Subsection (7)(a)

Paragraph (a) of the definition of ‘restricted equity instrument’ refers to "any restriction". Should this not limited to only those restrictions that will prevent the taxpayer from freely disposing of the equity instrument at market value by virtue of their employment?

Effective date

The proposed effective date states that the section will apply to equity instruments granted on or after the date of promulgation. However, unlike section 8B, section 8A does not contain a definition of "date of grant". We therefore recommend that in order to provide certainty that a cross-reference to the section 8B definition be inserted.

Subsection (1)(b)

We are concerned that the net has been cast too wide in terms of the proposed deduction from the initial amount and that no deduction is appropriate in terms of amounts paid for services in connection with the transaction.

Accordingly we recommend that the following be inserted after the final reference in the proposed wording to transaction, operation or scheme;

"…., other than payments for services which reflect an arm’s length rate, …"

Subsection (1)(c)

The reference to a lease and leaseback arrangement is confusing in that this is not a defined term within the Act, nor is any cross-reference to another section made. This contrasts totally to the immediately considered issue of sale and leaseback which refers to section 23G.

We suggest therefore that if this terminology is to be retained, that an appropriate definition, and cross-reference thereto, be included.

Whilst it is accepted that the approach now adopted in the proposed section 8F is already enshrined in section 8E we remain of the view that the measures are overly harsh and punitive.

In instances where either section applies, in the paying company’s hands the result is to recharacterise interest paid as dividends. This results in both a denial of tax deduction and the incurral of a STC cost on the associated deemed dividend.

The recipient both remains taxable on the interest received and receives no STC credit.

It is our view that this is a significant departure from the general anti-avoidance stance adopted in section 103, whereby the parties (i.e. both of them) would effectively be placed in the position they would have been were the transaction characterised as the Commissioner views it.

The approach in section 103 would thus treat the income of the recipient as tax-free dividend income and it is our view that sections 8E and 8F should be amended to this fairer basis.

The proposed section 24M seeks to postpone the accrual of any amounts that are not quantifiable in a particular year until such time as they do become quantifiable. In like vein, it also seeks to postpone the incurral of expenditure in respect of the acquisition of assets until such expenditure is quantifiable.

We are concerned that the wholesale exclusion of connected persons from these provisions is arbitrary. Whilst the need to prevent abuse of these provisions for anti-avoidance purposes is understandable we consider it would be misguided to take the view that all transactions between connected parties are aimed at that effect with the need to prevent any such transaction ever qualifying.

Although it is recognized that certain connected persons could rely on the inter-group relief, such relief will only apply where there is at least a 75% equity share holding.

In order to overcome the arbitrary exclusion in respect of connected persons we suggest that the exclusion of connected persons be qualified to apply only in instances such as those envisaged in paragraph 38 of the Eight Schedule, where the parties do not contract at arms length.

In the absence of these provisions applying in the context of a connected party transaction, does this mean that the common law rule that there must be a ‘amount’ before there can be an accrual applies?

The reference in section 20B(2) to ‘income’ is likely not correct. If the allowance in section 11(o) has not yet been claimed, nothing can be included in income, e.g. a recoupment. Should the reference perhaps be to ‘taxable income’ instead which will then likely refer to capital gains?

The amendment to paragraph 25 is difficult to understand without reading the explanatory memorandum, and even then it is not clear it achieves the desired effect.

It is understood that what it seeks to achieve is take into account proceeds that only arise subsequent to a disposal of a pre-valuation date asset in determining the ultimate capital gain or loss on disposal of the particular asset. The structure of the text is difficult to understand and it is suggested that it be reworded to be clear on what is intended.

Although the explanatory memorandum refers to claims which are unquantifiable as intended in section 24M, this provision will affect more than just scenarios where a claim is unquantifiable. Accrual is postponed not only by an amount being not quantified but also by conditions etc. Is it the intention to bring such scenarios also within the scope of paragraph 35A?

Section 24N caters only for the sale of equity shares, although in reality similar scenarios apply for the disposal of businesses.

Very often a business is sold on the same basis and creates similar problems. It is suggested that the provision also be made applicable in the instance of the sale of a business as a going concern and the proceeds depends on future profits.

Again there is an arbitrary exclusion of connected persons. We suggest that the exclusion be qualified with an arms length requirement as is recommended above for section 24M and as already exists in the legislation in paragraph 38.

The proposed provisions in Annexure 4 do not provide for the scenario where there may be a fixed amount of proceeds, but that such proceeds may be reduced by some future uncertain event (e.g. payments in respect of warranties and indemnities granted at the time of disposal which are to be treated as a return of the purchase price).

We suggest that provision be made to cater for a redetermination of capital gains/losses in instances where proceeds are later reduced.

In the alternative, is it envisaged that the full proceeds would be unquantified until such time as any period for the warranties or indemnities expires?

We disagree with a number of the examples provided in the explanatory memorandum and have included reworked versions of such instances below.

Revised example 7

In the example the annual proceeds accrue over the five years, but the total guaranteed amount of R100,000 is brought in as receipts in the first year (2005). We cannot see that the facts differ in any material respect from those provided for example 5.

In addition, as the base cost itself is only used in the calculation in the first year its repetition in each year in the explanatory memorandum example is confusing. Equally the cumulative receipts is irrelevant otherwise than as a note or check.

Revised example B

The example detailed in the explanatory memorandum reflects a scrapping allowance of R610,000. However, as the asset is not a depreciable asset, no scrapping allowance can be claimed. It must therefore rather be a reference to capital loss rather than scrapping allowance. Accordingly, as in our revised version below, a suspended capital loss should rather be accounted for.

Revised example C

A recoupment can only arise where the amounts received exceed the tax value of the asset. Accordingly there could be no recoupment in 2004 and 2005, as is reflected in the explanatory memorandum, as the total amounts received at that stage do not exceed the tax value of R500,000.

In the explanatory memorandum it was treated correctly in 2004, but in 2005 a total recoupment of R200,000 was incorrectly entered.

Revised example D

Although we agree with the taxable gain and inclusion in the ordinary taxable income, no suspended capital loss is shown for 2004 and 2005 and there can be no additional scrapping allowances in 2005 and 2006 as only additional proceeds was received in that periods.

Revised second example after example E (immediately preceding heading "D. Transfers of trading stock assets"

Although we agree with the taxable gain, no suspended scrapping allowance is shown for 2004 and 2005.

Last example

In the facts of the example no base cost is supplied and no figures for the 2009-year, although these figures is taken in account in the calculations.

Amendment proposed

We welcome both the combining of the current sections 10(1)(h) and 10(1)(hA) into a single, simplified, section 10(1)(h), and the retrospective application in the case of certain foreign funds which are tax-exempt in their home jurisdiction.

From the outset, we believe that the legislation proposed is very practical and we believe a fair way to address the issue. By prospectively applying the technical amendment, there would be no loss to the South African fiscus of taxes already collected and by retrospectively addressing retirement funds, there is now clarity on the matter and taxpayers would not have doubts on how to interpret the technical amendment.

Remaining concerns

Uncertainty remains however in respect of the Namibian long-term insurance and unit trust industries, which are not currently covered by the proposed amendment. In previous meetings between the relevant stakeholders and SARS and National Treasury the impact of the potential technical amendment was discussed, together with the possibility of not enforcing the application of the current law in certain instances (i.e. where for administrative reasons it was not possible to introduce the requested tax exemption with retrospective effect and for those taxpayers identified in this paragraph, SARS would not seek to collect the tax on SA source interest accrued prior to the amendment to sections 10(1)(h) and 10(1)(hA)).

Clearly if the South African tax legislation is not actively enforced in respect of these specific instances for prior years no further changes to the proposed legislation would be required. However, for such an approach to be successful, and in order to provide certainty both to the market and to the auditing firms who provide assurance on the accounts of the Namibian companies involved, we consider it would be necessary for the Commissioner to issue a press statement to this effect.

We understand that precedent for such approach does exist. In the early 1990’s regarding the introduction of section 10(1)(hA), we understand the South African authorities issued a public notice to the effect that no attempt would be made by SARS to collect the tax which would otherwise have been levied on foreign investors on SA source interest accrued prior to the introduction of section 10(1)(hA), though we have been unable to locate a copy of this notice.

Whilst we acknowledge this may conflict with the general premise stated in section 88B(1) that the Commissioner should not forego any taxes, duties, levies etc. it seems clear that in certain instances, such as those identified above, such action could be justified for a greater purpose.

In the event that the issuing of such a public notice is considered unacceptable, and understanding that there is limited time to affect the required amendments, we would like to propose the following pertaining to these industries.

Long-term insurers:

Section 29(14)(c), until replaced by section 29A for years of assessment commencing on or after 1 January 2000, provided an exemption for insurers on income derived in the Republic in respect of a business conducted in Namibia.

Section 29A applies only to life companies registered in South Africa and so the deductions it permits an insurer in respect of its taxable funds do not apply to Namibian insurers.

The result, when combined with the current Namibian/ South African double taxation agreement, is that Namibian life companies would be liable to South African tax on interest earned between 1 January 2000 and 31 December 2004. This appears to be an anomalous gap period and it is our understanding that there was no intention to deny such companies the relief previously afforded them by section 29(14)(c).

Accordingly we recommend that the proposed section 10(1)(h) also be made retrospective in respect of these companies as follows, with the existing clause (2)(ii) being renumbered appropriately;

Namibian unit trust funds:

In terms of the previous Namibia/ SA DTA, there was no specific inclusion of interest and consequently Namibian unit trust funds were liable to South African tax on SA source interest income only where such income was attributable to a permanent establishment in South Africa, which would not normally be the case.

The revised Namibia/ SA DTA (effective 1 January 2000) however provided for a right to tax interest in the territory of source at a rate not exceeding 10%. Unfortunately no prohibition of this right was effected in respect of tax-exempt entities.

Accordingly we recommend that the proposed section 10(1)(h) also be amended retrospectively in respect of these funds as follows, with the existing clause (2)(ii) being renumbered appropriately;

28. the receipts and accruals of which are exempt from tax in the country of which that fund is resident; and
29. that fund is resident in a country which has for the purposes of applying any regulation made under section 9 of the Currency and Exchanges Act, 1933 (Act No. 9 of 1933) has been included in the common monetary area

We would point out that the proposed deletions in the Exemptions from the duty under paragraph (1) or (2) of paragraphs (b) and (h) were effected last year under section 163(1)(c) of the Revenue Laws Amendment Act, Act 45 of 2003 and accordingly do not require to be enacted again.

Similarly the proposed deletion of the Exemptions from the duty under paragraph (3) of paragraph (p) was effected by section 163(1)(h) of the Revenue Laws Amendment Act, 45 of 2003.

It is noted from the explanatory memorandum that the proposed section 67A is intended only as an interim measure pending a separate Bill to be issued in 2005 following a further consultation period. Equally it is noted that this interim measure is intended to enable SARS to obtain valuable information on tax practitioners currently in practice.

In order to provide certainty we would suggest that the legislation include details of the information which the Commissioner may require in the course of the registration. As currently framed the legislation effectively gives the Commissioner carte blanche to request any details, including those which may not be relevant.

Instances where ruling applications may be rejected

Provision is included (section 76G(2)(a)) for the Commissioner to reject any application regarding the application of any general or specific anti-avoidance section. In our view this would be a key exclusion severely limiting the benefits offered by an advance ruling system as, without this aspect, the system would be incomplete.

The anti-avoidance ruling is a key factor which has driven the taxpaying community to seek the introduction of an advance ruling system. Taxpayers and their advisers are as capable of reading the tax legislation and applying their minds to the provisions and interpretation thereof as SARS. Whilst comfort is sometimes desired that an interpretation accords with SARS’ view, of far more importance is the obtaining of certainty that SARS do not consider the transaction to constitute tax avoidance, thereby triggering the relevant anti-avoidance provisions.

Time frames

No time frames have been detailed in which SARS must respond to a ruling request. We would suggest that if rulings cannot be given, at least on an "in principle" basis within a month, the system will contribute only a fraction of the potential benefits which it is aimed at delivering.

Throughout the Act time limits are placed on both taxpayers and the Commissioner as regards provisions of information and responses and we consider that this Part IA is a key element of the legislation where such requirements on the Commissioner are in fact essential.

Retrospective revocation of rulings

The proposed section 76N provides that the Commissioner may in certain circumstances withdraw a ruling with retrospective effect.

Whilst it is accepted that some attempt has been made to limit the extent to which this power be exercised we are concerned that this does not go far enough.

In addition it would appear that there is no right of objection or appeal for applicants which we would suggest, if this power is retained, is essential

The crux of our concern is that one of the core principles behind any ruling system is that it must provide certainty. The inclusion of section 76N severely detracts from the certainty that the proposed system offers, effectively laying taxpayers open to a second reviewer from SARS taking a different position and holding that the official who granted the ruling did so in error.

Scope

The discussion paper issued previously for consultation on a proposed system of advance rulings stated at paragraph 4 on page 7:

The related explanatory memorandum accompanying Annexure 11 states that the provisions relating to advance rulings will apply not only to the Income Tax Act, but also, mutatis mutandis, to Estate Duty, tax on retirement funds and Value Added Tax.

We have however not seen any provision in the draft legislation giving effect to this intent and it is obviously essential that this be included in the final version.

We would also point out that for many transactions certainty may also required in respect of Stamp Duty, Uncertificated Securities Tax and Transfer Duty and the ruling process should include these also.

Definitions of advance tax ruling, binding class ruling, binding general ruling & binding private ruling

These definitions appear to be limited to the Income Tax Act and we refer you to our above comments under the headings "General – Key issues", "Scope".

Again we refer you to our comments under the headings "General – Key issues", "Scope" as to the scope of the advance ruling process proposed, and the need for it to extend beyond the ambit of the Income Tax Act.

Subsection (2)(e)

We are concerned that the requirement proposed for the identification of subsequent transactions as currently drafted is cast too wide as it states;

We would suggest that the first "may" be replaced with

Subsection (2)(h)

We would welcome some clarification as to the types of reasons the Commissioner expects to see and accept under this subsection. In our view all that is required is a confirmation that, in the applicant’s view, all the information requirements in section 76E have been complied with and none of the excluded circumstances in section 76G apply.

Subsections (2)(f) and (2)(k)

In our view these subsections effectively duplicate each other and accordingly we recommend that subsection (f) be deleted.

Subsection (2)(l)

Rather than a description of information to be removed, we would suggest that, in accordance with the approach in subsection (k) that the applicant provide the sanitised version of the ruling to be published. This is subject to our comments below as regards the proposed section 76O.

Subsection (1)(a)(iv)

We are uncertain as to the value of this subsection as it is unclear as to how SARS will be able to police whether or not a transaction in respect of which an application is received is being seriously contemplated or not.

Subsection 1(b)

We recommend that certain separate provision be made in respect of the circumstance detailed in subsection (1)(b). It is our understanding that many taxpayers would like to seek rulings on this specific matter in order to avoid penalties should their position be incorrect.

We do not propose that any ruling in this respect by the Commissioner be binding in the same way as the other rulings provided for in the proposed Part IA, but rather that, if on the same facts available to the taxpayer, the Commissioner reaches the same decision, in the event of any later change resulting from fresh information being determined, then the Commissioner must waive any penalties applicable.

Subsection (1)(d)(i)

We are concerned that no rulings will be given on "vexatious" matters. Aside form the fact that what is vexatious is a very subjective matter, certain of the issues on which taxpayers may seek a ruling may well be those which vex the taxpayer in their interpretation and application and give rise to need for a ruling.

Subsection (1)(d)(iii)(bb)

An exclusion has been suggested where a transaction is the subject of draft legislation. We suggest that a restriction to this exclusion be applied such that it can only apply where the relevant draft legislation has already been published. As noted under the headings "General – Key issues", "Time frames" above, we consider that in order for an advance ruling system to work, "in principle" decisions must be given within at least a month. It would thus be impractical to withhold rulings pending legislation which may be commenced in February following a Ministerial Budget speech but which may not be issued to the public until the end of September, as was the case this year.

Subsection (2)(a)

We refer you to our comments above under the headings "General – Key issues", "Instances where ruling applications may be rejected"

Subsection (2)(b)(v)

We do not consider the proposed exclusion where rulings on similar matters have previously been issued to be justified in all instances. It must be borne in mind that any such previous rulings may have passed the period for which they had effect and that in any event would have only applied to the transaction in respect of which that particular pervious ruling application was made. Furthermore new legislation or legal precedent may have transpired since the original ruling.

It is our view that SARS must accept they may be required to issue similar rulings (for which they will of course receive further fees) and that this exclusion should be deleted.

Subsection (2)(c)

We are unclear as to the justification of the proposed exclusion of instances where the Commissioner considers the ruling would be unduly time-consuming or resource intensive. Should the taxpayer choose to implement a transaction on which a ruling application was denied on these grounds SARS will in any event require to undertake at least the same level of work, which would have been required in considering the ruling application, should they choose to later query or challenge the transaction.

Subsection (3)

As is clear from above there are already concerns that the current excluded circumstances may be too extensive as they stand. We therefore do not support the granting to the Commissioner the power to determine further such excluded instances without a right of appeal or consultation process.

Subsection (4)

We suggest that for clarity a time requirement be included in which the taxpayer must respond.

We recommend that a further subsection be introduced which deems all interpretation notes currently in force to be deemed to be binding general rulings.

The explanatory memorandum refers to binding general rulings as being similar to the interpretation notes being issued at present and the treatment recommend above would thus be preferable to that provided for in section 76(I)(5).

Subsection (1)(b)

We recommend that the words "in all material respects" follow "are the same" to prevent anomalous situations whereby a minor change such as the use of a different offshore group company in a transaction, with no impact on the South African tax position, could otherwise prevent the application of a ruling. The proposed section 76K(2) already makes use of the concept of materiality and so we do not consider its inclusion here should give rise to any concerns.

In addition we recommend that provision be made that should the figures or values involved change, without impact on the underlying tax principles, that this will not jeopardise any ruling granted. Again it would seem unfair, for example, were a ruling to be void in respect of a transaction involving the transfer of listed shares simply as a result of the value of those shares changing between the date of the ruling being sought and the date the transaction is effected.

The purpose of the ruling process is to provide certainty on the application of the tax legislation and its underlying principles, which (apart from limited instances such as thin capitalisation) should not be affected due to a change in figures.

Subsection (1)(d)

We are of the view that this restriction should apply only where the assumptions or conditions are detailed on the written ruling granted, and accordingly the following words require to be appended to the clause;

Retrospective effect

We recommend that the circumstances in this subsection be extended to cover also instances where the transaction has, following issue of a ruling, been published.

Binding private rulings

We are unable to support the publication of private binding rulings, even where the identity of the taxpayer concerned is withheld.

Taxpayers may have invested a huge amount in developing their transaction which gives them a competitive edge in the market place and in their marketing. To violate the confidentiality surrounding this development will result in taxpayers opting not to go for a ruling thereby negating the benefits of the system.

In addition the effect of publication is that taxpayers not only pay for the development of transactions but they effectively also pay for sign-off by SARS, all of which is then open to be piggy-backed on by their competitors.

If any part of such a ruling request concerns a specific question of law which is of general interest to taxpayers then it should be included in a general binding ruling, but in such format that the proprietary rights of the applicant is not prejudiced in any way.

The discussion paper issued for consultation on the introduction of an advance ruling system proposed (paragraph 4.5.1.1 of page 8) that taxpayers could suggest rulings be issued on a particular provision of a tax law. No provision appears to have been made for that in this section and we would suggest that this also be moved onto a formal basis in line with the approach for other ruling requests.

Subsections (1)(a) and (b)

As with our second comment above in respect of section 76J(1)(b) we consider that changes to the figures involved, as opposed to any other facts which influence the tax principles involved, should be excluded from the "material" test.

Subsection (1)(c)

As with section 76J(1)(d), we are of the view that this provision should apply only where any condition or assumption of the Commissioner is documented in a written response to the applicant.

Whilst we welcome the fact that the Commissioner must provide draft rulings to the applicant prior to publication we would suggest that this is meaningless unless there is an obligation placed upon the Commissioner to take cognisance of any reasonable comments raised by the applicant.

We would also recommend that a minimum time period, say 20 business days, be afforded the applicant during which time they can prepare their response, and prior to the expiry of which the ruling may not be published.

In addition it would appear that there is no sanction on the Commissioner should he fail to comply with these sections and we would suggest that, given the potential seriousness in cases where intellectual property of the applicant is disclosed which they would have requested been removed if given a chance for comment, that appropriate sanctions be introduced in this respect.

One possibility identified in this respect would be for the Commissioner to be prohibited from modifying the relevant ruling in terms of section 76N in cases where he fails to meet his obligations under these sections.

Should the reference to "this section" not be to "section 76E" as is the case for the cross-reference from section 76Q(1)?

Subsection (2)

We question whether the words "as part of any transaction, operation or scheme" are in fact required.

The acquisition by a company of an asset, or the issuing of shares by that company (or a connected person in relation to that company) will always be a "transaction" and accordingly we would suggest that the words quoted above are superfluous and should be removed.

We point out also that "debt instrument" is not yet a defined term in the Act and accordingly presume that this is intended to bear its common meaning, capturing, inter alia, debentures and loan notes.

We are concerned that no time limit has been proposed for the targeted anti-avoidance measure envisaged in this subsection and also that no exceptions to the denial of the treatment in subsection (1) have been granted for transactions effected for bona-fide commercial purposes.

Groups of companies undertake reorganisations on an ongoing basis, as business needs and the economic climate changes. It would seem unduly draconian were a subsequent transaction some years later to impact on the base cost analysis on a prior transaction.

In this regard we refer you to the 3^rd example provided in the explanatory memorandum under the heading "Exceptions" (though see also our comments below in respect of the explanatory memorandum) and highlight that if Parent were to subscribe cash for the issue of new shares in Sub 1 (e.g. to provide funds for further expansion) at present, but later (even say five or more years on) reorganised its group structure such that it transferred all of its shares in Sub 1 (both the shares it originally held and the ones now issued) to Sub 2 (e.g. to realign its divisional operations following a period of growth through acquisition) then it would be penalised through the stripping of its base cost. In the absence of relief under Part III this could result in a massive capital gain in Parent’s hands. In the event that transfer of the Sub 1 shares to Sub 2 was effected under a Part II transaction this potential exposure would simply be passed to Sub 2.

It appears that such problems arise as a result of the term "indirectly". Whilst we appreciate the need for such extended anti-avoidance measure we remain firmly of the view that it is not appropriate for these measures to apply in all circumstances and the proposed legislation should be amended to include some exceptions to the specifically described circumstances.

In this respect we recommend exceptions to the circumstances envisaged in subsection (2) in the following instances;

1. where the second leg of the transaction which would otherwise result in the whole being viewed as an indirect cross issue takes place after a given length of time, say 18months, being the time frame used in Part III transactions; and
2. where, as in our example above, there are bona fide commercial reasons to support both legs of the transaction.

It could be that the above instances are in fact cumulative, i.e. that the exclusion will only apply where both above circumstances are present. In addition it could be made a requirement that in such instances the taxpayer must apply for an advance ruling under the proposed Part IA of Chapter III in order that SARS can determine whether such a bona fide commercial purpose exists.

Aside from the wording as it stands potentially resulting in, what we consider to be, an unjust consequence it is not at all clear how either the taxpayer or SARS could set up systems to track such instances and the use of a time-limit as suggested could mitigate the extent of this potential problem.

Subsection (3)

We are also unable to support the extent to which the current proposed wording may have effect as we again consider it to have the potential to be unduly punitive in certain circumstances.

Take for example the case of a company having issued a debt instrument in exchange for the acquisition of shares issued to it. That company may subsequently (even years later where the debt instrument is a long term loan or debenture) transfer its liability in such respect to another party in return for another form of consideration (e.g. passing both the liability and an asset of equivalent value, even some or all of the shares acquired, to another party to facilitate the first company’s liquidation).

In such circumstances it will not be the company which issued the debt instrument, being "that company" as referred to in the subsection, which would ultimately pay any amounts due under the instrument, but the company assuming the liability. It seems unfair in such circumstances to penalise, permanently, the company which originally issued the debt instrument in this manner.

Again we suggest that certain exceptions from the circumstances detailed in subsection (3) be provided for. These exceptions could in fact be the same as those suggested for subsection (2) above.

In addition we would suggest that for clarity the reference to "the amounts are paid" be amended to "any amounts paid".

Subsection (4)

We would question whether subsection (4) is actually necessary. Section 41(2) of the Act makes it clear that the provisions of Part III, including both sections 42 and 43, effectively override other provisions of the Act, which would otherwise apply to the contrary, including this new section 24B.

If it is nonetheless still considered necessary for subsection (4) to be included, we would point out that it is then necessary to extend its ambit to cover also the acquisition of assets in terms of amalgamation transactions (section 44) and intra-group transactions (section 45), both of which can involve the acquisition of assets in exchange for the issue of shares.

Section 44(4)(a) specifically envisages such an issue of shares in return for assets, and the exemption under subparagraph (g) from the stamp duty otherwise payable under Item 15 (1) or (2) of Schedule 1 to the Stamp Duty Act, Act 77 of 1968 evidences that such transfer of assets in return for the issue of shares, by way of an intra-group transaction under section 45 of the Act, is also already envisaged by the legislature.

The issue of a loan note by a company in terms of an intra-group transaction would also not be an uncommon method of structuring a transaction and reference to section 45 would also be required to then nullify the effects of subsection (3).

In addition should subsection (4) be retained, it is necessary then to ensure that subsection (1) is subject not only to subsection (2) as is currently worded, but also subsection (4).

Effective date

Neither the Annexure published initially nor the draft of the consolidated draft Bill seen include a specified effective date for this section. We would suggest that, in order to provide certainty for capital gains purposes where we would suggest the law has not been clear in this respect, the date be structured such that subsection (1), in so far as it relates to capital assets it applies to any acquisition since 1 October 2001.

Alternatively, and in any event, the effective date for subsection (1) should at least be structured such that it applies in respect of all assets acquired before its implementation but which are not disposed of until after its effective date, rather than applying only to acquisitions after any effective date.

Present law

We highlight that we do not concur with the SARS view expressed in terms of the status under the current law, which view states that only in cases of transactions falling within Part III can a base cost be obtained for capital goods acquired in return for the issue of shares.

The references to paragraphs 11(1)(b) and 11(1)(d) of the Eighth Schedule, should be to paragraphs 11(2)(b) and 11(2)(d).

Proposal – General Rules – Shares for Assets

As noted above in respect of the proposed subsection (4), (should it remain in light of our above comments) the reference in the explanatory memorandum to the pre-existing exceptions being transactions under sections 42 and 43 should also refer to transactions under sections 44 and 45.

Examples dealing with transfer of real estate to a company

The second example involving the transfer of real estate to a company refers to paragraph 11(2)(c) of the Eighth Schedule. The correct reference is to paragraph 11(2)(b).

Examples dealing with exceptions (under heading "shares issued for shares or debt")

The third example dealing with direct and indirect cross issues of shares and/ or debt instruments (involving Parent, Sub 1 and Sub 2) refers in the result provided to Sub 1 having a zero base cost in the Sub 2 shares received. However in the facts provided Sub 1 does not hold any Sub 2 shares, only Parent does. However, from the facts provided it seems to us that this example should in any event not fall within the remit of the excepted circumstances as it involves the transfer of cash (Parent to Sub 1) and not the issue of a debt instrument and there is no matching cross-holding between Sub 1 and Sub 2 at any time.

The results provided for the fourth example appear to omit the result that Parent will have a zero base cost in the Sub shares acquired and we suggest for clarity this be included.

Examples dealing with exceptions (under heading "debt issued for shares or debt")

The result for the second example does not include what Sub’s position is in terms of its base cost in the promissory note and we recommend this also be included.

Subsection (3) - limitation on interest costs

Whilst we appreciate that a similar concept is currently included in section 11(bC), (to be repealed) we question why it is considered appropriate to reduce the interest cost attributable to financing of taxable foreign dividends by the amount of any potentially unrelated exempt foreign dividends received.

If a taxpayer holds shares in two foreign companies, A and B, the former of which generate taxable foreign dividends and the latter of which generate exempt foreign dividends no interest is deductible in respect of financing costs for the latter in any event due to the operation of subsection (1).

Accordingly we cannot see why any exempt dividends are considered relevant to the amount of interest incurred to obtain the former, taxable, foreign dividend income stream. We therefore are of the view that the following words should be deleted from the proposed section;

Effective dates

There appears to be a typing error in (a), with the letters "ruing" included after "… interest incurred". In addition (b) refers to "section 11C(4) and (4)", which we presume should be a reference to "section 11C(4) and (5)".

Inclusion in income

We refer you to our comments in respect of Annexures 1 and 2 on the proposed inclusion of amounts within income. It is our view that the current statutory form does not permit for such inclusions to be made. Accordingly the current references to "income" should be to "gross income".

Inclusion in income

We refer you to our comments in respect of Annexures 1 and 2 on the proposed inclusion of amounts within income. It is our view that the current statutory form does not permit for such inclusions to be made. Accordingly the current references to "income" should be to "gross income".

Extent of application

We have grave concerns regarding the proposed changes to this subsection as they will result in the conversion of both exempt amounts (e.g. foreign inheritances) or capital amounts (proceeds from the disposal of a capital asset) received by the trust into income of the beneficiary.

It is essential that the income of the beneficiary only be taxable if it would have been taxable had the trust been a resident.

Accordingly we recommend that subsection (2A)(a) be amended to read

We are concerned that a degree of ambiguity exists as to the interaction between section 31A and Part III of the Act, primarily due to the differing language adopted, being the use of "provision and "section".

For the purposes of certainty, it would be useful to have the wording in s41(2) clarified to ensure that s31A overrides the Part III rules only to the extent they are in conflict and not wholesale.

Our concern is that where s31A applies it could be seen to fully negate the reorganisation rules, including the relevant definitions, with a consequent potential flow through effect to the availability of the related transfer taxes reliefs.

We do not believe that was ever the intention and have discussed this previously with Johan de la Rey of SARS. Whilst he agreed with us both as regards the intent of the legislation and our position that a reading can already be taken that s31A does not fully override Part III we consider certainty in the wording remains preferable.

Accordingly we suggest the following amendment be made to section 41(2);

1. to the extent that any conflict arises, section 31A; and
2. section 103"

Definitions of "domestic financial instrument holding company" ("DFIHC") and foreign financial instrument holding company ("FFIHC") - Exclusion of trade debtors

Certain debts of a company are excluded where they were included within the "income" of the company (or controlled company in relation to that company) in respect of which the test is performed.

"Income" however is a defined term in the Act and has different connotations depending on whether a company is a resident or non-resident. This concern is already being addressed elsewhere in the proposed legislation contained in this draft Bill (see proposed amendments to sections 7(8) and 25B).

Although arguably in regarding the FIHC tests, no issue should arise as in the context it could be argued income bears its common meaning and not meaning ascribed to it in the section 1 definition, we are concerned that, in the case of non-resident subsidiaries of the company in respect of which the test is to be performed (both DFIHC and FFIHC tests), and in the case of the company itself (FFIHC test only), even if a debt arose from its trading activities and constituted part of its taxable results in the foreign jurisdiction in which that company is tax resident, that it will not have been included within "income" as defined for SA purposes.

Accordingly we suggest that the definitions be amended to read as follows;

DFIHC

FFIHC

Incorrect referencing

Section 43(7) requires to be updated through the removal of the cross reference to the FFIHC company definition in s9D, which is now contained in s41.

Subsection (4) - Degrouping clawback

Again we appreciate the need for certain of the changes proposed from an anti-avoidance perspective but consider that a number of issues arise from other of the changes made.

Whilst the current section 45(4) correctly triggers only the rolled over gain or loss it appears that the proposed wording results not only in any such rolled over gain or loss being realised, but also any additional profit or loss resulting from further changes to an asset’s value whilst held by the transferee company becoming subject to tax. This is a direct result from utilising the concept of a disposal to a connected party, which will be treated as occurring at market value on that date.

As we understand the anti-avoidance to be targeted specifically at preventing the combining of a second (or later) group company with the ultimate transferee such that should they leave the group together no degrouping charge will arise, the appropriate value at which the disposal should be deemed to take place would be the market value as at the date of the transfer preceding the transfer to the ultimate transferee (i.e. the date of acquisition by the transferor).

We therefore recommend the wording in subsection (4)(b) should be;

This would combine and trigger all rolled over gains and losses other than any from the date of acquisition by the final transferor in a chain to the date the ultimate transferee departs from the same group of companies as the initial transferor.

Parity would thus be provided both for

1. the current position whereby any gain or loss attributable to the period the asset is held by the transferee is not triggered simply by virtue of the fact that it leaves the transferor’s group of companies; and
2. other assets held by the transferee which were not acquired from another group company (or were so acquired without the benefit of Part III reliefs) and in respect of which no degrouping charge would apply

It should be noted however that the above is not a perfect solution as, where there are multiple transaction preceding the degrouping of the original transferor and ultimate transferee, there is potential for more of any rolled over gain or loss realised than is appropriate to be treated as a connected party gain or loss to the original transferor

when it a portion in fact may still relate to change in the asset’s value when held by a later transferor in the chain.

However it is considered the wording recommended above represents an improvement on that proposed in the draft legislation, as the latter is actually far more punitive than merely triggering any rolled over gain or loss together with any incremental change in value in the transferee’s hands. Although under the proposed wording the disposal takes place at market value (by virtue of it being deemed to be to a connected party) the reacquisition takes place only at the base cost immediately prior to the disposal, i.e. the same taxable result will arise again in future, resulting in either a double gain or a double loss.

Subsection (6)(a) - Permitted transfers of financial instruments (no amendment currently proposed in draft legislation)

Provided a company to be transferred is not a financial instrument holding company, subsection (6)(a)(iv) permits a transferor to transfer shares in a controlled group company in relation to that transferor to another group company within the relief mechanisms of section 45.

We do not however consider the restriction to only controlled group companies in relation to the transferor to be justified. A number of group’s have historically adopted cross holdings in their subsidiaries and this wording currently prevents them from unwinding such structures, to consolidate all such holdings within the group, on a tax-free basis.

We would request that the limitation on the shares which may be transferred be amended to "equity shares in a controlled group company which is not a DFIHC or a FFIHC"

If the current wording exists to combat some perceived anti-avoidance opportunity of which we are not aware then we would suggest in the alternative that a new subsection (6)(a)(v) be introduced which would allow the transfers of shares in controlled group companies (even where these are not controlled in relation to the transferor) only in circumstances where such transfer is to effect a consolidation within the group of different holdings in the company whose shares are being transferred.

Subsections (3) and (3A)

Profits of South African branches

The insertion of section 64B(3A) purports to capture the essence of certain details previously contained in Section 64B(3). However, whereas the proviso in section 64B(3), prior to the current amendment proposed, effectively provided for STC credits to be obtained in respect of any foreign dividend exempt under section 10(1)(k)(ii)(aa), the proposed section 64B(3A)(d) provides such STC credits only in certain of the circumstances previously catered for.

The current wording in force, pre the proposed amendment, ensures that STC credits are available in respect of foreign dividends which were originally sourced either from either

1. profits which had already been subject to South African tax, e.g. any South African branch profits of a foreign company [section 10(1)(k)(ii)(aa)(A)]; or
2. dividends from a South African company (where STC would already have been paid on the underlying profits) [section 10(1)(k)(ii)(aa)(B)]

The granting of STC credits to foreign dividends sourced from the former category was justifiable on the basis that a South African branch is subject to normal tax at the rate of 35% as compared to the 30% payable by a South African company. The additional 5% has itself previously been justified by the tax authorities as a charge in lieu of STC which branches are not subject to.

It is our view that were the proposed removal of the STC credit on such foreign dividends as are envisaged in (i) above enacted, then the rationale for levying a higher rate of tax on branches would be significantly diminished.

In addition this will adversely impact on a number of South African companies who hold investments in foreign companies (including those in the common Monetary Area in respect of which no exchange control restrictions apply) which in turn operate branches in South Africa.

Take for example a corporate South African shareholder which invests in a listed Namibian company. That Namibian company, either directly or through one if its local, Namibian subsidiaries, operates a South African branch.

Assume the South African branch makes profits of R200, it will then pay R70 tax in South Africa (35% tax rate attributable to branches) with the remaining R140 of profits being available for distribution by the Namibian company of which it is a branch. If the R140 is paid as a dividend to South Africa, under the proposed section 64B(3A) it carries no STC credits and so when those profits are paid out by the corporate South African shareholder STC of some R16 will be payable resulting in a total effective South African tax rate (ignoring Namibian taxes) of over 42%.

Accordingly we recommend that, unless there is an intention to ensure parity between South African tax resident companies and South African branches as regards the normal tax rate payable that section (3A)(d) be amended to cover all foreign dividends exempt under section 10(1)(k)(ii)(aa), in the same way as the proviso in the current section 64B(3) does.

We have suggested below a revised wording for section 64B(3A)(d).

International groups of companies

We are aware of instances where even the proposed wording of section 64B(3A)(d) does not cater for the granting of STC credits to profits returned to South Africa by way of a dividend from a foreign group company even where the profits from which that foreign group company paid the dividend are attributable to dividends received from another South African group.

In particular the proposed wording does not cater for instances where no equity shares are held in the offshore group company but profits are paid back to South Africa on preference shares which are held as a result of the imposition of regulatory requirements, such as those imposed by the South African Reserve Back.

Again it is our view that it is essential that the granting of STC credits to such dividends, as is available under the current legislation, not be removed as this will directly result in double taxation.

Whilst we appreciate that provision would require to be made to ensure that there is no duplication of credit relief given, we remain of the view that, within a group of companies, a certification system, whereby the original South African declaring company provides a certificate of STC paid and attributable to a dividend coming back into South Africa could be made to work. Provision could be made that such companies could only issue certificates to group companies and limited to the attributable portion of dividends paid out of South Africa.

Suggested wording

Resolution of the problems identified above could be achieved through the rewording of subsection (3A)(d) as follows;

"(d) any foreign dividends unless –

1. those dividends arose directly or indirectly from dividends declared by a company which is a resident, and –
27. current proposed subsection (3A)(d)(i); and
28. current proposed subsection (3A)(d)(ii);
2. the declaring company holds a certificate as envisaged in section [new section in Act required] evidencing that secondary tax on companies has been paid by a member of the same group of companies on the underlying profits from which such dividend is declared; or
3. those foreign dividends are exempt in terms of section 10(1)(k)(ii)(aa)(A)

Provided that any foreign dividends so received shall be deemed to have been declared first from dividends or profits received directly or indirectly from South Africa."

Alternatives to (ii) could be either

27. the taxpayer will have to prove to the Commissioner's satisfaction that the foreign dividend was sourced from previously taxed SA profits; or
28. the company declaring the foreign dividend will have to certify that the source of profits from which the foreign dividend is declared directly or indirectly comes from SA taxed profits, with a deeming provision in instances where that foreign company has both South African and non-South African shareholders that any portion attributable to SA taxed profits (carrying the STC credits) goes first to the SA shareholders (to avoid double tax) rather than being split on a pro-rata basis.

Subsection (f)

We would request that, for the purposes of clarity, the explanatory memorandum confirm that a shareholder will be regarded as subject to STC in circumstances where no STC is actually payable by that shareholder as a result of the use of STC credits from other dividends received.

Proviso to subsection (5)(f)

We would repeat our previous requests that, in order to provide certainty, some clarification be introduced as to what is considered to be covered by the phrase "formed solely by one company within that group of companies".

Our experience is that companies themselves rarely if ever directly form other companies, rather using legal advisers to incorporate new companies on their behalf or buying "off-the shelf" companies. Whilst off the shelf companies may be harder to cater for we would suggest the following wording be inserted after "within that group of companies" "or at the instigation of one company within that group of companies,".

An alternative to this would be for SARS to issue an interpretation note or a practice note providing guidance and confirmation as to what is covered. This could even form the basis of a Binding General Ruling.

Subsection (2)(g) (no related amendment currently included in draft legislation)

The Revenue Laws Amendment Act 2003 (Act 45 of 2003) saw the effective combing into the current section 64C(2) of the previous subsections (2) and (3). In addition to the instances when a deemed dividend was previously deemed to arise (old section 64C(3)(a) to (e)), that Act (45 of 2003) introduced two new events which would give rise to a deemed dividend, in the current section 64C(2)(f) and (g).

We are concerned that the latter of these, section 64C(2)(g), is too widely drafted and can, in certain circumstances, give rise to an unintended consequence.

There are instances where a South African company may support its offshore interests through the making of an interest free loan. Whilst the rationale for such loans can often be substantiated such that no transfer pricing adjustment is made under section 31 (and accordingly no deemed dividend arises under section 64C(2)(e)), section 64C(2)(g) will now result in the triggering of a deemed dividend. In addition, even though no transfer pricing adjustment is considered appropriate this deemed dividend seems not to be merely the benefit arising from the use of such loan (i.e. a notional interest rate or the return earned) but rather the face value of the loan granted.

The exemptions otherwise available in section 64C(4) will not always be available, either because, e.g. the debtor does not utilise the loans in the Republic (s64C(4)(g)) or is not a 75% group company (s64C(4)(k) and (l)).

Accordingly we would suggest that a proviso to section 64C(2)(g) be introduced, along the lines of the wording suggested below;

Subsection (4)(k)

A further proviso, equivalent to that existing in section 64B(5)(f) is required to cater for like circumstances in the case of deemed dividends.

We are concerned that the extent of coverage proposed by section 66(1A)(a) may in fact be too extensive.

The requirement for any persons who derive gross income of a South African source to render a South African tax return would capture, for example, foreign parent companies who advance funds on loan account to their South African subsidiaries and earn interest thereon. Although such interest would be exempt from South African tax under section 10(1)(h) (under the proposed new wording, previously such interest would have been exempt under section (10)(1)(hA)) it would nonetheless fall within the foreign parent companies "gross income" as defined.

There seems no point in extending the filing requirement to such instances as this will result only in additional costs for the taxpayer and additional administration and costs for SARS with no benefit to the fisc.

We are also concerned that placing such requirements on non-residents would be to the detriment of the desired attracting of portfolio capital referred to in Annexure 5, which deals with the exemption of South African interest for non-residents.

Liability of individuals for taxes withheld

Paragraph 16 of the 4^th Schedule was amended by the RLAA 2003 by the insertion of subparagraphs 2A to 2D to provide for personal liability of certain employers and individuals in respect of tax withheld but not accounted for to SARS.

However we are concerned that, in particular subparagraph 2(C) extends far beyond the remit stated in the explanatory memorandum as

Subparagraph 2(C) however states simply

We are concerned therefore that subparagraph 2(C) covers not only tax withheld and not paid over but all other aspects such as for example underpayments of tax due to differences of opinion on sub-contractor status. Clearly this was not the intent and we suggest that the language be amended to reflect the intent as stated in the explanatory memorandum.

Disposals by non-residents (no related amendment currently in draft legislation)

Paragraph 2 subjects to South African income tax capital gains realised by non-residents on the disposal of immovable property in South Africa or interests in such property. Such interests include the holding of shares in a company, where more than 80% of the net asset value of that company is attributable, directly or indirectly, to immovable property in South Africa.

A key exemption from the application of these CGT rules is where the immovable property is held by the relevant company as trading stock, i.e. if any profit arising on the sale of the immovable property would fall to be taxed as revenue income in the hands of the company, as opposed to capital, then the disposal of the shares by a non-resident will not fall within South Africa’s CGT net.

It appears clear that the purpose of this exemption is not to levy tax on capital gains of non-residents where the South African fiscus will ultimately benefit from the higher rate of income tax attaching to the disposal of the underlying asset on revenue account.

We are concerned however that an anomaly exists in this regard when considering mining companies. Whilst it may on occasion be that, due to the nature of the mining rights held, which rights very often constitute a large part of a mining company’s value, more than 80% of such companies value is indirectly attributable to immovable property in South Africa it is clear that the disposal of the assets eventually derived from the owning of those rights (being the minerals or ore extracted) will be subject to income tax as revenue income. Even though the mining right itself remains a capital asset, such minerals or ore will in due course constitute the trading stock of the mining company.

It is hoped that this impact on non-residents is simply an unintended consequence in the legislation due to a peculiarity in the mining industry, whereby a mining company may hold a capital asset related to immovable property (the mining right) at one point in time, but with a view to converting that to a revenue asset (extracted mineral or ore) over time. We consider this to be distinguishable from other industries whereby the capital asset is used to produce the trading stock, as opposed to effectively being converted into the trading stock.

Accordingly we would suggest that subparagraph 2(2) be amended by the addition of the following words at the end of the existing sub-paragraph,

1. any permit, authorization, lease, right or permission described in section 9(1)(fA)(i), (ii) or (iii);
2. any old order right as defined in section 1 of Schedule II to the Mineral and Petroleum Resources Development Act, 2002 (Act No. 28 of 2002); and
3. any such similar right."

It is not considered necessary to say that this deeming is only for the purposes of this subparagraph only as paragraph 2(2)'s application is already limited to paragraph 2(1)(b)(i) and the inclusion in paragraph 2(2) of the proviso proposed above limits its extent in the same manner.

Sale of the mining right itself by the company in which an interest is held would remain subject to SA CGT.

We welcome the proposed changes in respect of the deemed acquisition at market value of assets belonging to a person who commences to be a resident (the dropping the deemed disposal the day prior to becoming resident).

However we consider that some further minor amendment is required to avoid any lack of clarity.

The amended paragraph 13(1)(g)(ii) states the happening of an event in paragraph 12(4) is the date that the event occurs. However the wording of the revised paragraph 12(4) may now be said to consider two events, one being the commencement of residency and the other being the deemed acquisition of assets. Whilst it is accepted that the two events are intended to be simultaneous the wording does not make this explicitly clear.

Accordingly we recommend paragraph 13(1)(g)(ii) be further amended to read

In addition, to ensure certainty, we recommend paragraph 12(4) be amended as follows

"… cost equal to the market value of each of those assets at that time, …."

Consequent upon the amendments to paragraphs 12(4) and 13(1)(g)(ii), certain amendments will be required to paragraph 24.

For example the reference in subsection 24(1) to "expenditure allowable in terms of paragraph 20 incurred on or after that date …" should be suffixed by (other than expenditure treated as incurred by virtue of paragraph 12(4)".

Incorrect referencing

The cross reference to the section 41 definition of "foreign financial instrument holding company" should be to the definition in section 9D.

Paragraph 64B replaces the participation exemption previously located in section 9D(9)(h) which was based on the, more relaxed, definition in section 9D and we are unaware of any reason why this is not continued and suggest that the cross reference be amended accordingly.

Appropriateness of limitation

In the case of disposals by residents it appears that, under paragraph 64B(2)(b) the exemption will only apply where the acquirer is a non-resident.

We have a concern that this can only serve to discourage growth through the expansion of domestic SA groups of companies, as vendors will now prefer to find non-resident acquirers in order to avoid tax on the disposal.

In the event that a resident purchaser can be found the vendor may require a sale to a non-resident subsidiary of the purchaser in order to still avail themselves of the exemption but this in turn results in the creation of "loop" structures which is at odds to the policy of the Reserve Bank.

No such requirement concerning the purchaser’s tax residency status was present in the previous participation exemption in section 9D(9)(h).

In addition, given that it would be relatively straightforward to structure a taxpayer’s affairs such that any foreign companies are held by an intermediate offshore holding company (for whom the non-resident acquirer condition is not applicable) we do not consider that this limitation adds any benefit or protection to the fisc. Dividends from such intermediate holding company, including those from profits arising on the tax free disposal of an interest in a further foreign company, would benefit from the exemption in section 10(1)(k)(ii)(dd) and so be entirely tax free in the SA resident recipient’s hands in any event.

It would appear that, for a resident disposer, the result of the non-resident acquirer requirement may be to encourage intermediate offshore holdings rather than direct SA ownership, which we understand to be at odds to current national economic policy.

Accordingly we submit that the requirement that the acquirer be a non-resident be removed.

Operation post an unbundling transaction

The issue identified here is best illustrated through an example;

• A, B and C are South African incorporated and tax resident companies
• A owns 100% of B, who in turn owns 100% of C.
• The market value of B is 1,200, and A's base cost in those shares is 800
• The market value of C is 200, and B's base cost in those shares is 70
• The total inherent capital gains (assuming disposal of C by B, then disposal of B by A) is 530

B unbundles C to A under s46 of the Act (s46(6) provides that this amount is deemed first to come from share premium - for these purposes we assume it all does).

• The disposal by B is ignored for CGT purposes
• A now owns 100% of B and C
• The market value of B is 1,000, and A's base cost in those shares is 667 (being 1000/ 1200 * 800)
• The market value of C is 200, and A's base cost in those shares is 133 (being 200/ 1200 * 800)

The issue

• If A now disposes of B it realises a gain of 533 (proceeds = market value 1000, less base cost of 667 above plus additional proceeds under para 76(1)(b) of 200)
• If A disposes of C it realises a gain of 67
• The total inherent gain is now therefore 600

It seems clear that the intent cannot have been for additional tax to arise as a result of the operation of a relieving mechanism (s46) in the Act. This particular problem arises due to the operation of p76(1)(b) following on from the provisions of s46(6).

Accordingly we would suggest that paragraph 76(1)(b) not apply in respect of capital distributions received by a shareholder under an unbundling transaction effected under s46. In addition we recommend that this amendment be made retrospective to 1 October 2001.

Double counting of proceeds

We are also concerned that there are other instances where the application of paragraph 76(1)(b) can lead to substantially more than double taxation. One such scenario has been included by way of an example below.

Mr X subscribes R100 for shares in Co A. This is made up of R10 for 10 shares each of nominal value of R1, and the balance of R90 as share premium.

At a later stage Co A issues capitalisation shares (so effectively distributing the share premium account) so Mr X now holds 100 shares each of nominal value R1.

Mr X now sells all these shares to a third party for R100.

In respect of his original 10 shares (for which he paid R100) he makes a capital loss of R90. Proceeds of R10 (total proceeds of R100 x 10/100 shares) against base cost of R100.

In respect of the further 90 shares received as a capital distribution he makes a capital gain of R90. Proceeds of R10 (total proceeds of R100 x 90/100 shares) against a base cost of zero. The capitalisation shares have a nil base cost by virtue of paragraph 78.

At this stage Mr X is in a neutral position, he has a gain of R90 which is offset by a loss of R90, and logic says this is the correct result as he has in total received proceeds of R100 for an asset he paid R100 for.

However, paragraph 76(1)(b) also applies to deem Mr X to have further proceeds of R90, being the nominal value of capitalisation shares issued. As a result Mr X is treated as having realised a capital gain of R90 in respect of his disposal.

Clearly this is an anomaly which requires resolution.

Section 6 - exemptions

It appears that the equivalent relief to that granted by exemption (g) to the duty payable under Item 15 (1) or (2) of Schedule 1 to the Stamp Duties Act, being exemption from stamp duty on the issue of shares in the context of an intra-group transaction, has been omitted from the UST Act. Whilst we acknowledge that the application of this relief in the context of a listed company within a group of companies is likely to be limited, we are nonetheless of the view that for consistency an equivalent exemption is required under section 6(a) of the UST Act and we request that this now be included.

As highlighted in paragraph 3 of the covering note accompanying these schedules, PricewaterhouseCoopers have identified a number of other issues not raised in the draft legislation issued for comment.

Any such issues of a minor or technical nature are included within the relevant annexure, with the issues contained in this Annexure being those where PwC believe that SARS and/ or National Treasury would choose to consider in greater depth. The issues in this Annexure are covered under the following headings;

1. foreign provisions;
2. corporate reorganisations; and
3. capital gains

I Section 9D(1) and section 9D(9)(b)

Definition of business establishment and related exemption

Section 9D(9)(b) exempts certain income of a controlled foreign company ("CFC") provided that CFC first has a "business establishment", as defined, in a country outside South Africa.

Paragraph (a)(i) of the definition of "business establishment in section 9D(1) provides that the place of business is suitably equipped with, inter alia, such on-site management and employees as are appropriate for the purposes of conducting the business. In effect the paragraph requires that the place of business directly contracts with its own personnel.

It is our view that there are instances where such restrictive requirement is not appropriate.

Large groups may include a number of operating subsidiaries in a given offshore territory for a number of reasons, be it for purposes of limitation of liability on diverse business lines or due to differing investment partners into different businesses. In addition it is not uncommon in such cases for a group to also form an employment company in that territory to employ the personnel for all the investments there. This approach is commonly used to minimise payroll costs and administration, with the employment company simply making a recharge to the operating companies to recharge the salary costs borne initially by that company together with any related administration costs or fixed costs such as office rental.

Other issues raised by the use of employment companies are highlighted in our comments in respect of Annexure 1, dealing with the broad-based employee share plan initiative.

Accordingly we recommend that the following proviso to paragraph (a) be inserted into the legislation;

David – I realise that by limiting the above to groups of companies it is not as far as you want it to go, but maybe we have more chance with this as a first stage? – if you want to go all out, have you any suggestions as to alternative wording?

II Section 9D(2A) and section 9D(9)(b)(iii)

The purpose of section 9D of the Act is to impute the income of controlled foreign companies ("CFCs") in the hands of a SA resident, with the effect that the income of a CFC may become taxable in the hands of the resident. In this respect section 9D(2) is effectively a charging section, providing in essence that there shall be included in the income of a resident for any year of assessment such proportionate amount of the income of the CFC as relates to the participation rights held by that resident.

Section 9D(2A) then prescribes the manner in which the net income of a CFC to be attributed to residents must be calculated. In this regard it is provided that the net income must be determined as though the CFC had been a resident.

As the CFC imputation system could otherwise bring about a tax regime more onerous to SA taxpayers than most of its trading partners, certain relief measures and exemptions from the application of sections 9D(2) and 9D(2A) were provided for in section 9D(9). In particular, post RLAA 2003, section 9D(9)(b)(iii) provides that the provisions of section 9D(2) shall not apply to the extent that the net income of a CFC is attributable to any business establishment of that CFC in a country other than the Republic. Such exemption does not however extend to passive income (dividends, interest, royalties etc) where such income exceeds ten per cent of income and capital gains of the CFC.

In certain circumstances it is submitted that the language adopted can lead to an anomalous situation whereby a SA resident can have income imputed to them (and be taxable thereon) where a similar investment in a South African venture would have had no such impact.

Take for example the case of company X, a CFC held 100% by a SA company. Company X is incorporated, managed and controlled, and has a business establishment in, say, Botswana. It earns active income and interest. Company X is still in start-up phase and as such has incurred losses.

Its results comprise the following items;

Income from active operations R100

Interest income R15

Expenditure relating to active operations R150

In determining the net income of Company X that must be attributed to residents, it is arguable that the excess of interest income over 10 per cent of the total income (i.e. R3.5, being R15 less 10% of R115) must be imputed to SA resident shareholder.

Against this is the fact that Company X does not have any net income (it has made an actual economic loss of R35) and accordingly it could be argued that there is no net income attributable to residents.

This latter approach appears consistent with the notion that a resident should be taxed in respect of a CFC’s results only as if that CFC were a SA resident operation carried on directly by that SA resident shareholder.

There remains a degree of doubt however and accordingly we submit that the legislation requires amendment to clarify the position.

III Section 9D(9)(b)(iii)(bb)(A)

Requirement for FIHC test to be performed each time income is derived

Section 9D(9)(b) broadly exempts from the imputation to a SA shareholder income of a foreign company where that foreign company has a valid business establishment overseas.

An exception to this exemption is where that company earns passive income and at the time of so earning that income it would be considered a foreign financial instrument holding company.

The test in s9D(9)(b)(iii)(bb)(A) requires the determination each time the foreign company derives such passive income of its status in terms of the FFIHC test.

We submit that this is a overly burdensome requirement from an administrative point of view (both for the company or its SA shareholder or indeed for SARS) and a cost perspective (requiring market values to be obtained each time income is derived).

We suggest therefore that from a practical perspective that, purely for the purposes of the business establishment exemption, the reference to "at the time the amounts are so derived" be removed and replaced with a fixed point of time in order that the company

need only perform limited tests. One obvious solution would be to use the wording "at the commencement of that financial year", thereby treating the company on one basis throughout the year, though other options exist such as at the end of a financial year or the average of opening and closing positions.

Alternatively we would welcome the issue by SARS of a practice note detailing the approach which they adopt in relation to this issue.

IV Section 9D(9)(b)(iii)(bb)

Permitted form of otherwise "passive" income

Section 9D(9)(b)(iii)(bb) effectively provides that the business establishment exemption still applies to recipients of "passive" style income provided they fall within certain named categories, being banking, financial services, rental or insurance.

It appears to us that companies in receipt of royalty income, but who actively manage and maintain their underlying asset in respect of which the royalty is charged, and incur expenses in relation thereto, should also be included in this list.

We are of the view that it is quite possible to distinguish between companies who perform such an active role and those which receive truly passive royalty income such as on book sales, and that the former should still qualify for the business establishment exemption.

Such distinction between types of income is well established in the UK under the "Earl of Howe principle" set down in the early 1900’s, whereby pure profit income was distinguished from income against which expenses must be incurred in order to derive the income.

V Section 10(1)(k)(ii)(dd)

Extent of exemption from tax on foreign dividends

We should like to seek clarity as to whether the exemption provided for in s10(1)(k)(ii)(dd) extends only to dividends from investments where a greater than 25% equity stake is directly held, or also to those where a comparable stake is indirectly held. Although the current wording may be regarded as ambiguous, we are concerned that it could be applied to exempt only dividends from directly held investments.

We have detailed below an example of a situation where we consider a potential anomaly could occur.

SA Parent Co holds, indirectly, a stake of >25% in Overseas Co 2.

Overseas Co 1 is not a member of the same group of companies as SA Parent

• foreign regulatory requirements, e.g. foreign investors may be prevented from holding over a certain level of equity stake, we understand that this is seen in some territories like Greece in respect of certain industries, and in others such as the UAE across the board. In such circumstances where additional funding is required and SA Parent Co is the only party with resources or willingness to invest further funds it cannot do so through traditional equity stakes.

• foreign thin capitalisation rules, e.g, where further funds as envisaged above cannot be introduced by way of vanilla debt funding, or

• foreign withholding taxes, e.g. it may be more beneficial in terms of SA’s double taxation agreements to pay preference dividends direct to SA rather than through the intermediate holding company. This aspect is clearly to SA’s advantage from a currency flow/ foreign profit repatriation perspective.

Our understanding is that the intention would be for dividends on the preference shares, arising on an investment in which at least 25% of the equity shares are held, though indirectly, to be free from SA tax.

It appears anomalous that a dividend from Overseas Co 1 that is funded solely by a dividend from Overseas Co 2 is tax free in the hands of SA Parent Co, but a similar dividend paid directly from Overseas Co 2 to SA Parent Co may be subject to SA tax. This anomaly also appears to ignore the economic group unit and the premise that as far as possible, as a matter of preferred policy, our tax laws should seek to tax economic units.

We should be grateful if this matter could be clarified either by such changes to the language used in the legislation as are required to clearly reflect that intent, or through the issue by SARS of a practice note confirming the interpretation that would permit dividends from indirectly held investments satisfying the minimum holding requirement to be free of SA tax.

I Extension of relaxed DFIHC/ FFIHC definition in group scenarios

The Act promulgated on 22 December 2003 provided a relaxation of the DFIHC and FFIHC definitions in respect of certain group transactions. In particular the value of financial instruments where market value equals base cost was excluded from the calculation in the case of intra-group transactions and liquidation transactions by virtue of the provisos inserted to s45(6)(a)(iv) and s47(6)(b). The rationale behind this was that such assets did not give rise to the loss trafficking, which we understand to form the basis of SARS and NT’s justification for the inclusion of the financial instrument holding companies tests and the acceptance that groups should be able to move cash around tax-free.

However this concept was not extended to the other relief mechanisms, presumably because they can apply outside a group scenario.

We consider however that it would be relatively straightforward, and desirable to extend this principle to the other relief mechanisms to the extent that they apply in a group scenario through the use in sections 43, 44 and 46 of the following language;

Section 43(7)

Section 44(12)

Section 46(7)(a)

II Section 43 - Practical issues

In seeking to apply section 43 for a client in respect of the acquisition of minority shareholdings in a listed subsidiary a number of issues came to the fore;

• For minority shareholders to determine whether s43 can apply to the disposal of their shares they require to determine whether that company is a financial instrument holding company [s43(7)]. It is unclear how such minorities can hope to obtain such information.

• Although s43(2)(b) takes the practical approach in terms of the base cost of shares in the acquirer's hands, presumably for administrative simplicity as the alternative would be to take on the numerous different base costs of all shareholders, this ignores the fact that for the relief to apply in the first instance the acquirer would need to determine whether each shareholder is sitting at a break even position or an inherent gain [s43(1)(a)], whether the shareholder holds the asset as trading or capital

• As in the bullet point above, if an acquirer cannot determine whether in fact s43 applies, they will not know whether the various clawbacks [s43(4) & s43(5)] are of relevance

This problem is exacerbated by the mandatory nature of the provision in a non-group scenario

III Stamp duty and intra-group transactions

As regards certain intra-group transactions, exemptions from the stamp duty levied on the transfer of shares in terms of Item 15(3) of Schedule 1 to the Stamp Duties Act are provided in paragraph (x)(iv) and (x)(vii)(aa), respectively covering those transactions regulated by section 45 of the Income tax act and those transactions which would be so regulated were an election made for s45 treatment to apply.

However we are concerned that these particular exemptions do not go far enough.

There are at least two further scenarios we envisage where section 45 may not apply but where we consider that the stamp duty relief should still be afforded.

The first is where the asset being transferred is shares in a group company which is a financial instrument holding company. As noted previously the inclusion of the financial instrument holding company tests are aimed at prevention of loss trafficking and do not concern themselves with the stamp duties aspects. Where a transaction is still effected outside of the reliefs within s45 we are of the opinion that the fact a company may be a FIHC should not preclude the stamp duty relief from applying.

The second is where the transferee company is not a resident, although still a member of the same group of companies. Where this is particularly relevant is where the current

shareholder of a SA company is a non-resident. In the context of an international reorganisation such shareholder can often transfer their shares to another group company free of tax, including SA CGT, but the one remaining cost is SA stamp duty. It is our view that exemption should also be granted in respect of intra-group transfers between non-residents.

Accordingly we submit that the exemption from stamp duty in terms of Item 15(3) as regards intra-group transactions should be amended to apply to all transfers of shares in controlled group companies to another group company.

IV RSC levies

Whilst we appreciate the notion that companies should contribute to the infrastructure costs of the regions in which they operate we would suggest that an exemption be provided from the regional establishment levy for dividends paid from one group company in a region to a holding company within that same region.

The principle enshrined in our income tax law is that the same income should not be subject to tax twice, but the levying of RSC levies on dividends is just that, a taxation on the movement of the profits arising from one external source of income and we submit that this approach also be adopted for RSC levies, at least in so far as they relate to group activities.

It must be remembered that in many instances group structures exist for commercial reasons such as segregation of liability exposures and that it is open to the group to consolidate all their operations in one entity in which case the double levying of RSC levies would not apply. We submit that the commercial drivers which demand separate group companies should not give rise to such adverse RSC consequences.

It is accepted that it would be politically insensitive to extend this exemption to dividends paid from group companies in one region to a holding company in a second region as there would obviously be a loss of income to the second region which could be perceived as placing one region at a disadvantage as compared to another, but this should not prevent the relief currently being sought from being granted (exemption where group members are in same region) so as to avoid double taxation on the same profits.