Submissions on 29 September 2000
to the Portfolio Committee on Finance
in respect of the 3rd draft Bill (dated 22 September 2000)
introducing the world-wide basis of taxation

By : David Lermer for and on behalf of PricewaterhouseCoopers

We commend the SARS on their transparency during the drafting process and we express our appreciation for the opportunity to offer our input on the proposed legislation.

As with a previous submissions, our representations in respect of the proposed amendments have been split into four categories, viz provisions where :

  1. the rules could result in adverse economic implications for SA (from an inward investment perspective);
  2. the rules could result in adverse economic implications for SA (from a promotion-of-exports, outward investment, and ‘African Renaissance’ perspective);
  3. the rules are harsher than those of other jurisdictions – the UK or US has generally been used as a benchmark for this purpose; and
  4. the legislation results in anomalous situations or does not cater for certain scenarios.

Invariably some points can be seen to qualify as issues under more than one heading. However, before dealing with our comments relating to the draft legislation itself (under the four headings indicated above), our first submission addresses the question of timing.

  1. Timing : Extend the consultation period
    1. As previously noted, there is no doubt that the currently proposed rules are highly complex and require highly sophisticated legislation. In this respect, we have repeatedly commended the SARS on what we can only describe as impressive pieces of work – under the circumstances.
    2. At the end of the day, the tight time schedule is very ambitious and not only places the SARS under pressure but, more importantly, in our view, does not allow for sufficient public debate and consultation.
    3. We appear again to be continually running against the clock and running out of time. Surely there should be a sustained period of public consideration and debate of this landmark legislation, of months not days. Whilst there has been a wider representation of professionals at the discussion meetings with SARS, the legislation has only been a public document for less than two weeks and was refined and placed on the SARS web-site only last Friday, 22 September 2000. This has given little time for review prior to the submission deadline for written representations of 27 September 2000.
    4. We therefore repeat our submission and appeal for an extended consultation period. Grappling with legislation of this level of complexity and of such significance to the South African economy, we submit, requires months rather than weeks of public scrutiny.
    5. We accordingly set out below our initial comments and representations on the draft bill.

  1. Potential obstacles for foreign inward investment

  1. High net worth foreign nationals becoming ‘ordinarily resident’
    1. The new rules offer no relief to high net worth individuals who have traditionally viewed SA as a business-opportunities or retirement destination. This would apply to individuals spending more than 90 days a year in South Africa and so would also catch those who spend the summer here. Typically, such individuals have large sources of foreign income and capital accumulated over long periods before moving to SA.
    2. To tax such individuals on foreign income generated out of pre-residence activities, which foreign income is not remitted to or used in SA, will serve as a material disincentive for such high net worth individuals. In the light of SA’s current need for foreign capital and skills investment it is submitted that such individuals should not be discouraged from settling here.
    3. The currently proposed definition of ‘resident’ also has a retrospective test in that the residence test for the tax year to 28 February 2002 will be based on physical presence in the years to February 1999, 2000 and 2001.
    4. Currently, the taxation of foreign investment income applies to individuals who are ordinarily resident in South Africa. The interpretation of ordinary residence is a grey area of our legislation and most of the individuals in this category would have more than one residential property globally and ties to more than one country. In most cases, the individuals concerned would not be viewed as ordinarily resident in South Africa and so the proposed changes will have an immediate impact on 1 March 2001.
    5. It follows that individuals in this category have only two options, pay tax on their world-wide income from 1 March 2001 or leave. It is suggested that in seeking to remain outside the time-based test for ‘residence’, their immediate exodus from this country by the end of May 2001 (i.e. before 90 days of the 2002 tax year has elapsed) is assured. Alternatively, those unable to leave timeously – because of becoming aware of the detail of the legislation too late – could well cheat.
    6. It should also be noted that the top and middle categories of this class of individual are opinion makers and their marketing of SA for foreign investment should not be overlooked or discounted.
    7. From our research, it is interesting to note that this type of individual, in the main, is willing to pay for certainty. Generally, they would rather pay some local tax than pay professional fees to avoid it. The former would give certainty of tax treatment, would increase South Africa’s tax base and potentially accelerate tax collections.
    8. Several countries allow negotiation with the expatriate taxpayer (e.g. Switzerland) often agreeing a fixed amount of tax per annum, based on various factors. Of course, there are also many low tax jurisdictions that will offer permanent resident status for an annual fixed fee. For example, Malta allows this for a fixed tax charge per annum of approximately GBP10,000 on all non-Maltese income.
    9. It is therefore submitted that an equitable compromise would be to tax such individuals on all SA and foreign income arising out of new activities undertaken after becoming SA resident. There are three possible solutions:

    1. It is submitted that, at the very least, the residence test should only consider tax periods commencing on or after 1 January 2001.
    2. We have prepared schedules detailing the likely loss to the fiscus of introducing the world-wide residence basis of taxation without an appropriate exemption for foreign high net worth individuals. The schedules have been prepared from a case study of nine such families living in South Africa and are for illustrative purposes.
    3. The potential loss to the economy and tax to the fiscus is significant. On the basis that there are 1600 such families currently in South Africa, the potential differential between attracting this type of individual with his or her capital and expertise and losing this capital and skills base amounted to some R6 billion to R16 billion over a ten year period in recurring taxes foregone. This ignores once-off taxes, such as VAT or transfer duty on fixed property investment or the economic spends on which the taxes are based.
    4. It must be said that obtaining accurate statistics on this matter has been difficult and it is not clear how accurate the figure of 1600 is. We have no doubt however that the models attached accurately reflect the potential adverse trend that the current legislation would have on the South African economy, if the legislation is introduced without any of the suggested exemptions. The schedules similarly show the positive trend on the South African economy and increase in the tax base that result from a policy and related legislation, attracting such individuals.
    5. We are of the view that the potential loss of this category of individual will be felt at all levels of the economy, from the informal sector to formal providers of goods and services.
    6. The attached schedules illustrate the estimated benefits to the economy in terms of employment and indirect taxes for the fiscus, moneys injected into the economy in terms of tourism, employment, fixed property, business investment etc. Our request is that given the results of this initial research, that this matter be given serious consideration.
    7. We have tested our views with other accounting and legal firms and find that their views accord with our own.

  1. International headquarter companies of foreign multinationals
    1. As previously suggested (by the Katz Commission, and by us and others) special exemption and relief provisions should be introduced to promote a favourable regime in SA for foreign-owned regional / international holding companies. The basic principle is the taxation of local income, but exemption for what is effectively a flow-through of foreign income.
    2. Prior to the introduction of the CFE ("Controlled Foreign Entity") and taxation of foreign dividend legislation, and now the residence based system of taxation, South Africa had a very attractive system for foreign investors to place holding/headquarter companies.
    3. The effect of the introduction of the residence-based system will be that companies incorporated here will be subject to South African tax on their world-wide income. This legislation will accordingly act as a deterrent to global entrepreneurs and companies who wish to use South Africa as a holding company and headquarter location.
    4. The combination of the proposed legislation and exchange controls also puts South Africa at a disadvantage to its neighbours and other emerging markets in its ability to attract capital and expertise to its shores. This potential aspect is of concern since the country needs to import capital and expertise. The new provisions could well have the opposite effect. In fact, following the Budget proposals (23 February 2000), we are aware that investors are no longer considering SA for these purposes, and there are examples of disinvestment.
    5. We recall that in 1997 the Katz Commission specifically recommended that we should encourage the establishment here of international headquarter companies. Having so far accepted many of the Commission’s other recommendations, we would do well to revisit this particular one.
    6. Therefore, the legislation introducing the residence based system of taxation should include incentives and exemptions to encourage the establishment in South Africa of foreign international headquarter companies, for example, an exemption from capital gains tax, exemption from tax on foreign income and any other measures encouraging holding companies.
    7. Consideration may be given to a "participation regime" that provides an exemption to the designated company from, inter alia, the provisions of the CFE (Controlled Foreign Entity) legislation (section 9D as amended), tax on capital gains in respect of shares in its subsidiaries (when capital gains tax is introduced) and tax on its foreign dividend income. This type of exemption is widely used throughout Europe.

  1. Potential obstacles for local exports, outward investment and African Renaissance

  1. Allow for the relaxation of the 183-day requirement for SA nationals working abroad
    1. The current 183-day requirement in s10(1)(o) will impede the competitiveness of many foreign contract tenders of SA multinationals. Because it is often possible to obtain tax-exempt status for non-resident employees, the reduced cost of employment allows the SA investor to compete on equal terms for large projects in foreign countries.
    2. Since the SA employees involved on these contracts sometimes operate outside SA for less than 183 days, the current draft of s10(1)(o) means that SA investors could be at a competitive disadvantage, which could, in turn lead to the loss of foreign projects.
    3. It is therefore submitted that the 183-day requirement should be reduced to a continuous absence of at least 90 days (subject to incidental visits to South Africa – representation 5 below refers). Alternatively, if this exemption is considered to be too wide, it is submitted that a provision should be included to allow for the 183-day requirement to be relaxed to 90 days in certain cases - perhaps on a ruling basis - to take into account commercial considerations. For example, the ruling process could be restricted to foreign contracts, which are subject to a bid or tender process. The disadvantage of this more controlled approach, is the level of administration it will entail over a simple 90 day exemption.

  2. Incidental visit of SA nationals working abroad
    1. Sub-paragraph (ii) of s10(1)(o) does not take into account the fact that most South Africans working abroad often make non-employment related visits to South Africa during the course of their overseas employment. This might include holidays or attending to personal matters. There may also be an incidental business need, such as attendance at a management meeting, and so forth.
    2. It is therefore submitted that the continuous 183-day requirement should be softened from the current position so as to allow exceptions in the case of incidental visits back home.

  3. CFE exemptions should be broader and should not exclude intra-group transactions
    1. It is suggested that the current draft of s9D is overly complex and will in many respects hinder the international competitiveness of bona fide active trading activities thereby constraining the growth in SA exports. This translates into restricting the growth in the SA tax base, rather than limiting its possible erosion.
    2. The length and complexity of the provisions creates much uncertainty.

    1. You will appreciate that this results in some very convoluted legislation, ending with a wide discretion on the part of the Minister or Commissioner, the wording of which is equally vague. In summary, the legislation lacks certainty and is commercially burdensome.
    2. It would appear that the purpose of the exclusions from s9D is to tax income of an active business on a remittance basis, namely when the income is repatriated to a South African resident by way of dividend, rather than in the hands of a South African resident participator in the year in which the income accrues to the CFE.
    3. At the same time, it is appreciated and accepted that there is the need to target ‘tax diversionary’ transactions and structures. However, it is submitted that this is adequately addressed by the transfer pricing provisions of s31, which appear to be referred to in many provisions of the current draft, including the proviso (b) to the proviso (ii) to s9D(9)(b). In this proviso, the Commissioner may waive the application of proviso (ii) where "the form of the transaction will in all probability not involve pricing that will be inconsistent with the provisions of section 31". It is submitted that the same result will be obtained through the combination of a simplified s9D and s31.
    4. The application of the two sections, triggered by a non-arm’s length (i.e. diversionary) transaction, will result in the doubling up of the taxation, interest and penalties attached to the diversionary transaction(s). This is because the s31 adjustments will almost inevitably be replicated as s9D adjustments for the same transaction(s). This is a particularly harsh result and should be addressed.
    5. It is suggested that this double taxation could be avoided by ensuring that any s31 adjustment is taken into account in reducing the net income determined in accordance with the provisions of s9D.
    6. As indicated above, it is also submitted that the current proviso’s (a) and (b) to s9D(9)(b)(ii) are unclear and create uncertainty. There is simply too much left open to Minister/Commissioner discretion. As a result, the fundamental principle of certainty in taxation is lost. It would appear the purpose is not to invoke the provisions of s9D, where arm’s length prices have been used (proviso (b) refers). If the suggested simplification above is followed, these proviso’s would no longer be required.
    7. The suggested amendment follows the existing draft provision of s9D(9)(b)(i), which excludes transactions from the provisions of s9D, where they relate to a business establishment and the taxpayer has complied with the provisions of s31. The whole exclusion from the business establishment exclusion of s9D found in s9D(9)(b) could therefore be greatly reduced and could read as follows:
    8. "Provided that the provisions of this paragraph shall not apply to any receipts or accruals derived from any international agreement as defined in section 31, between connected persons, unless the consideration in respect of such agreement reflects an arm’s length price that is consistent with the provisions of section 31;"

    9. It is suggested that this more simplified provision would provide an ample deterrent for any one considering a diversionary transaction.

    1. Further certainty could be added by the introduction, in the future, of a formal APA (advance pricing agreement) procedure. This intention could be noted in the Explanatory Memorandum. It is also recommended that until then, the existing practice of taxpayers being able to assume that observance of the guidelines set out in the 1999 transfer pricing practice note, should provide a ‘safe harbour’ in respect of s31, should continue. This too could be explained in the Explanatory Memorandum.

  1. Withdraw 1-year lease period requirement for business establishments
    1. Paragraph (a) of the "business establishment" definition in s9D(1) requires, inter alia, that leased property should be held for at least one year.
    2. However, although it is appreciated that this requirement is seeking an element of permanence, it is submitted that it is restrictive. It does not take into account the fact that initial set-up operations might not satisfy this requirement. For example, the current draft could be construed as disqualifying establishments occupying temporary accommodation in anticipation of a permanent premises. It also prejudices taxpayers who are forced by commercial considerations to (say) sign a short initial lease. There may be several other examples of where the taxpayer intends to, and does indeed, establish a fixed presence exceeding a year but where the actual premises leased do not meet the 1-year test.
    3. It is therefore submitted that the 1-year requirement should be withdrawn or, at the very least, that the definition allow for more than one lease which perhaps "in aggregate" exceeds one year.

  2. Foreign intermediate holding companies for SA multinationals
    1. It is suggested that specific exemptions should be introduced to exclude foreign holding subsidiaries from the CFE rules to the extent that the operation and income of such holding companies actually derive from otherwise exempt bona fide active operating subsidiaries.
    2. The purpose of this exemption is to place South African multinationals in the same tax position as many of their foreign competitors based in developing countries, that recognise the necessary role of an intermediate holding company and so do not penalise such commercial structures.
    3. The draft bill does go some way to recognise this commercial structure in that the exclusions found in S9D(9) (f) and (fA) from the CFE legislation provide for the flow of interest and dividends to such foreign intermediate holding companies. It is submitted that this exclusion should be extended to include royalties as well as other administrative functions of the intermediate holding company, performed for its foreign trading subsidiaries.

  3. Tax sparing
    1. As previously advised, the absence of tax sparing will deter SA multinationals from African Renaissance projects within our less developed neighbours. As noted, unilateral tax sparing is to be preferred – both by SA taxpayers and the SARS – to bilateral agreements.
    2. The introduction of taxation of foreign dividends and now a residence-based tax system without tax sparing provisions, when viewed against the incentives usually offered in developing countries, will have a negative impact on the hoped-for "African renaissance". It will have the effect of moving South African investment funds from developing countries into developed countries, while not increasing the tax collected by the South African fiscus.
    3. Incentives offered by developing countries (usually in respect of infrastructure) effectively compensate the investor for a higher level of risk associated with investment into that territory and/or project. Therefore in terms of the current South African tax system, a South African resident investor is compensated for the additional risk involved in investing into a developing country if it has a tax incentive system, by the lower effective tax rates suffered on income derived therefrom.
    4. If the residence-based system of taxation is introduced without tax sparing provisions, SA investors will be taxed in SA on income derived from foreign investments, with a credit for foreign taxes paid, and will in most cases therefore be taxed at the same rate whether the investment is in a developing country with tax incentives or whether it is into a developed country with rates similar to those of South Africa. There will therefore be no incentive to invest in developing countries. The absence of tax sparing or similar provisions will simply neutralise the incentives offered so that such projects may be abandoned.
    5. Where the absence of tax sparing neutralises this extra return or ‘compensation’ the expected investment funds and expertise would have to be directed to lower-risk, lower-return (because of foreign taxes) projects in more developed countries. Since these incentives are typically for local infrastructure projects, the absence of tax sparing relief in the investor country (i.e. South Africa) means that the infrastructure and the development of our poorer neighbours, and therefore of the region, will be set back even further.
    6. The absence of tax sparing in the introduction of the taxation of foreign dividends and the proposed legislation does not necessarily mean additional direct tax collection for South Africa; rather the tax that would have been saved in the incentive regime will now simply be paid to the authorities in the developed countries.
    7. The introduction of tax sparing provisions would retain the incentive to invest in the developing countries, thus ensuring the continued development of the infrastructure required in these countries, with no effective loss to the SA fiscus.
    8. In addition, tax sparing will usually result in the savings effected by the incentives offered by developing countries being repatriated to South Africa by the investors. In other words, additional foreign currency (i.e. the tax savings) will be repatriated to South Africa, so the tax sparing provisions will result in an increase in the repatriation of foreign currency, which would otherwise have been paid to the tax authorities of developed nations.
    9. It is therefore submitted that South Africa should introduce a tax-sparing regime, either as unilateral provisions as part of our domestic tax law, or as part of existing and new double tax treaties. It is suggested that structuring the provisions as unilateral relief within our law offers the advantage of being able to control the relief on a project-by-project basis rather than on the broader, less ‘police-able’, territory-by-territory basis. From a geographic perspective, such rules could be Africa-focused.
    10. We therefore agree that a possible approach is for the Finance Ministry to meet with other SADC countries and understand that this may have already commenced. We also accept that the required amendments to introduce tax sparing relief do not necessarily need to be included in our domestic legislation. Rather, as SARS has proposed, the matter can be dealt with in the relevant double tax agreements.
    11. However, it is reiterated that unilateral tax sparing (i.e. including the tax sparing provisions in our domestic tax law) would be preferable, because the process, amendments to the terms and conditions of the relief, remain fully under the control of the South African fiscus. Unilateral tax sparing therefore offers more control and efficiency than those of bilateral agreements. It is faster and is less prone to manipulation.
    12. Where a tax sparing provision is abused and results in potential loss to the South African fiscus, the unilateral approach allows the Commissioner to withdraw the relief with immediate effect. Where a treaty approach is adopted, the other contracting state also has to agree to the amendment and this, in practice, may not occur timeously.
    13. As a related matter, we are aware of an infrastructure project that was committed to prior to 23 February 2000 (the date of implementation of the taxation of foreign dividends legislation) that is now no longer commercially viable. We would expect that this is not an isolated problem.
    14. It is submitted that tax sparing relief could be afforded to these taxpayers by providing for such a concession within the discretion of the Commissioner. The basis for the exercise of such discretion could include factors like:

    1. This approach provides the SARS with substantial control over which projects obtain the relief, so that this incentive measure is applied in a controlled manner that meets the objective of developing infrastructure within Africa for the benefit of investment and trade within the region.

  1. Potentially harsher provisions (than the UK for example)

  1. Allow relief for foreign assessed losses
    1. The prohibition in the new para (b) of the proviso to s20(1), against the use of foreign losses against domestic profits is harsh in relation to jurisdictions like the US and the UK. It is also contrary to the fundamental basis of the world-wide system of taxation.
    2. It is understood that there might be tax-avoidance concerns around the possibility that foreign profits in later periods could be excluded after the claiming of foreign losses in the initial period. However, this could be addressed through appropriately drafted recoupment provisions to recoup previous tax losses in circumstances where a trade carried on through a foreign branch is transferred to a foreign subsidiary.
    3. From our initial research, we are not aware of any other jurisdictions that ring-fence foreign branch losses in this manner.

  1. Anomalies / issues requiring further clarification

  1. Meaning of ‘ordinarily resident’
  2. The term "ordinarily resident" remains vague and without objective guidelines, although it is understood to imply a greater level of permanence than simply ‘resident’. It is requested that SARS issue guidelines on this important concept.

  3. Meaning of ‘days’ in definition of ‘resident’
  4. Clarification is required in respect of whether days of arrival and days of departure will be counted as days of physical presence in SA.

  5. Residence in DTAs
  6. Many of SA’s double tax agreements refer to the term ‘ordinarily resident, so that it is not clear whether the individuals who become resident under the terms of the time-based tests in para (a)(ii) of the new ‘resident’ definition will qualify for treaty relief. This will need to be clarified.

  7. Time limits for claims and elections
    1. It is suggested that provisions such as s6quat should allow for the re-opening of assessments to address subsequent changes in circumstances. For example, a s6quat rebate claim might be based on a foreign tax assessment which (at the time of the s6quat claim) might be subject to a dispute with the foreign tax authorities – which dispute is subsequently resolved resulting in a higher or lower foreign tax liability than was originally claimed under s6quat.
    2. Furthermore, the new subsection (4) of s6quat only allows the conversion of foreign taxes at the time of payment (if actually paid in the claim-year) or at the year-end rate (if still unpaid). But in the latter case, it does not take into account subsequent fluctuations when the tax is actually paid.

  8. Currency conversion methods for CFE income – election should be fixed
  9. S9D(6) allows the taxpayer the option of electing either the year-end exchange rate or average rate in converting the apportioned amount. However, this should not be an annual election since different methods in different years will lead to inconsistencies. Therefore, the legislation should either prescribe only a single method, or specify that any election made is a permanent one.

  10. Clarify ‘statutory’ rate in the case of sliding scale rates and incentive regimes
    1. The reference in s9E and the new S9F needs to address specifically the fact that many jurisdictions apply a sliding scale of taxation to individuals, in similar ways to the SA system. In most cases, the lowest rate is below 27% and the maximum rate exceeds 27%, depending on the level of income. In some countries, there is also a graduated rate for certain companies, e.g. the UK’s ‘small companies’ rate.
    2. From our discussions with the SARS, we understand that it is the intention that the highest statutory rate should be used. Perhaps the word "maximum" could then be inserted before the word "statutory" in the actual legislation. Alternatively, the interpretation could simply be noted in the Explanatory Memorandum.
    3. A further question that needs to be clarified – either in the actual wording of the legislation, or in the Explanatory Memorandum – is the issue of incentive rates. Many tax regimes seek to incentivise investment into specific sectors, e.g. manufacturing, by applying a more attractive tax rate that the ‘normal’ statutory rate in that country. We are therefore often faced with a situation where a SA-owned foreign company is in principle, as a tax-paying entity, subject to the normal statutory rate in that country, but having some of its income subject to a lower rate because of the special ‘encouraged’ status of that class of income.
    4. Clarification will be required on how such entities should be treated.

  11. STC exemption for branches and agencies of foreign companies
    1. In terms of the existing section 64B(5)(h) of the Income Tax Act, a company which has its place of effective management outside South Africa and which carries on trade in the country through a branch or agency, will not be liable for STC in respect of dividends declared out of profits derived from the branch or agency. This exemption is to be deleted and the revised section 64B(2) is to provide that STC will only be payable by resident companies. As a resident company is a South African incorporated, established or formed company or a company which has its place of effective management in South Africa, it would appear that the exemption will continue to apply to branches and agencies of non-South African companies.
    2. The definition of "company" in section 1 of the Act however includes companies which are deemed to be incorporated in South Africa in terms of any law in force in South Africa. It may therefore be argued that branches of foreign companies, which are deemed to be incorporated in South Africa in terms of section 322 of the Companies Act, are companies incorporated, established or formed in South Africa and that they are consequently subject to STC on distributions.
    3. In order to clarify the matter, we therefore suggest that consideration be given to retaining the existing exemption in section 64B(5)(h) of the Act. Alternatively, this matter can, to a large degree, be dealt with through refining the definition of residence so that the domestic provisions will follow the tax treaty provisions, where applicable (see next representation).

  12. Dual residence companies
    1. It is possible for a company to be resident of South Africa in terms of the definition of resident and also be a resident of another country in terms of that country's legislation. For example a company may be incorporated in South Africa, but have its place of effective management in the United Kingdom resulting in dual residence status. In terms of the "tie breaker" clause in the South Africa - UK tax treaty, such company's residence will be deemed to be where its place of effective management is situated.
    2. We suggest that a similar "tie breaker" clause be introduced into the Act to clarify how the residence for South African income tax purposes of dual residence companies is to be determined. A similar clause should also be introduced in respect of dual resident individuals.
    3. In both instances the tie breaker clauses may be modeled on the OECD Model Tax Convention or the legislation may simply refer to the Model Tax Convention as has been done with the definition of permanent establishment in the revised section 31.
    4. This refinement to the definition of residence ensures that our domestic law remains consistent with our tax treaties and should avoid tax avoidance that uses South African domestic law residence to obtain reliefs without South African taxation, because of a treaty over-ride. It is noted that the draft bill already includes this type of provision in specific charging sections, e.g. the proposed amendment to 31(1) introducing (d) (ii) to the definition of "international agreement", as follows:

  1. (i)….

  1. any other person who is a resident.

where either of such persons is as a result of the application of the provisions of any agreement entered into by the Republic for the prevention of double taxation, not subject to tax in the Republic;";

    1. It is submitted that the refinement of the residence definition, like the definition of international agreement, ensures that the scenario of dual residence is dealt with in a manner consistent with South Africa’s tax treaties.
    2. For your information, the UK similarly considered this issue and dealt with this anomaly through the introduction of the following provision into its domestic law:

249(1) A company which-

  1. would (apart from this section) be regarded as resident in the United Kingdom for the purposes of the Taxes Acts, and
  2. is regarded for the purposes of any double taxation relief arrangements as resident in a territory outside the United Kingdom and not resident in the United Kingdom


shall be treated for the purposes of the Taxes Acts as resident outside the United Kingdom and not resident in the United Kingdom.

17. Taxation of South African source foreign dividends

17.1 The proposed amendments to s9E, will result in the taxation of foreign dividends which are sourced out of South Africa, e.g. where a South African resident receives dividends from a foreign company which has interests in (and therefore dividend income from) South African companies.

17.2 Under the current s9E such foreign dividends are excluded from the definition of 'foreign dividends' on the basis that the profits (out of which the dividend is declared) are from a SA source. There is no such exclusion in the currently proposed amendments, which simply classify the dividends as 'foreign dividends' when the top declaring company is not SA-resident. It is also not excluded under any of the exemptions. Furthermore, s6quat does not offer relief for underlying SA tax, i.e. it only recognises taxes paid to the governments of countries other than SA.

    1. The easiest solution, it is submitted, would be to include South Africa on the list of designated countries, so that the dividends would be exempt under s9E(7)(d). Alternatively, the definition of "foreign dividends" should be extended to exclude SA-source dividends. Perhaps wording like : " ... excluding dividends to the extent that such dividends are declared out of profits derived from a source which have been subject to tax in the Republic". Another alternative would be to provide a separate exemption under s9E(7) and perhaps the least preferred (from both SARS' and taxpayers’ perspective) would be to allow credit under s6quat.

Further submissions on 6 October 2000

to the Portfolio Committee on Finance

in respect of the 3rd draft Bill (dated 22 September 2000)

introducing the world-wide basis of taxation

By : David Lermer

for and on behalf of


  1. Redefine the ‘official rate’ of interest for foreign loans
    1. The SA ‘official rate’ of interest (Sch 7, para 1) is considered to be quite high and in fact significantly beyond market related rates in other countries.
    2. The official interest rate is particularly relevant in determining whether a loan from an employer to an employee constitutes a fringe benefit (Sch 7, para 2(f) and others), and also when an inter-group loan between companies constitutes a deemed dividend for the purposes of the STC rules (secondary tax on companies) (S64C(3)(d)).
    3. Whilst it is acknowledged that the official rate is usually representative of the market rates in SA, this is not necessarily so. It follows that an intra-group Rand denominated loan, that charges interest at the official rate, can fall foul of a foreign country’s transfer pricing legislation. Yet, if a market related rate is used, that is less than the official rate, to comply with foreign tax legislation, this results in a STC liability on the whole amount of the loan. It is therefore suggested that S64C(3)(d)(i) make reference not only to the official rate but, as it does for foreign currency loans in S64C(3)(d)(ii), to a market related rate.
    4. As regards the fringe benefit provisions, the current position creates an anomaly. This is because the rate used, whilst being a market related rate, so that the employee obtains no fringe benefit, is less than the South African "official rate" of interest. It follows that technically a fringe benefit currently arises on the granting of the foreign currency loan. It is therefore suggested that, like the current wording of S64C(3)(d)(ii), reference should be made in Sch7, para2(f) also to foreign currency loans granted at less than a market related rate of interest. In this way, a foreign currency loan that is granted to an employee on arm’s length (market rate) terms, will not create a fringe benefit.

  2. S10(1)(o) 183-day rule should be based on any 12-month period, not tax year
    1. The current draft of s10(1)(o) exempts foreign income earned during a continuous absence of at least 183 days during the "relevant year of assessment".
    2. It is submitted that this does not take into account positions where the absences for the country spans tax years so that even though the total of 183 days is exceeded in a 12 month period, it still amounts to less than 183 days in each of the tax years. It does not, therefore, provide for the commercial reality that the commencement and termination of offshore contracts are not negotiated and undertaken with reference to the SA tax year.
    3. It is therefore submitted that s10(1)(o) should be amended to refer to a continuous absence during "any 12-month period commencing or ending in the relevant year of assessment", rather than to "the relevant year of assessment".
    4. This refinement follows the treaty provision used in the OECD Model Convention and many of South Africa’s tax treaties in determining the taxing rights of the respective treaty countries in respect of employment income earned from an foreign employer by a non-resident for work performed in the source country.