24 March 2004

To Mr N Nene; MP

Chairperson: Portfolio Committee on Finance

Cape Town

Discussion Paper on Tax Avoidance

Honorable Chairperson

Please find our presentation on the abovementioned subject.



We welcome the opportunity afforded to us by the Parliament of the Republic of South Africa, National Treasury and the South African Revenue Services, to comment on this important aspect of the taxation laws and taxation policy at large.

Indeed the issue of tax evasion versus permissible tax mitigation is as old as the taxation system itself and has occupied the resources of tax enforcement agencies throughout the world. Unfortunately clarity on what is avoidance (permissible) and evasion (not permissible) has eluded both the taxpayer and the tax collector. It is against this background that any efforts to clarify this ambiguity in our tax law are particularly appropriate. We commend the approach taken by the South African Revenue Services ("SARSí), that is, to issue a discussion paper and invite public input and we hope that this trend will continue.

General Problem: Evasion, Avoidance and Planning

The background theory in the discussion paper presented by SARS re-iterates the difficulty experienced by tax authorities worldwide in differentiating between legitimate tax mitigation and impermissible tax evasion. While this difficulty is acknowledged, the temptation to imply or read the intention of lawmakers (i.e. Parliament) when interpreting any law should be left to the lawmaker. Any loopholes identified in the tax system should rather be referred back to Parliament for amendments rather than be left to law enforcement agencies to ratify as this may lead into an uneven application of the law.

The General Anti-Avoidance Rule (GAAR) should, as a rule, not be applied as a general panacea for all "loopholes" in the tax laws.

More importantly, and as acknowledged in the discussion paper, the GAAR is not a charging section and should not be applied as a "when all else has failed" legislation that may penalize legitimate business transactions.


Scope of the Problem

The discussion paper illustrates the extent of the tax evasion problem by citing various international examples. We commend SARS on the ample research provided and the quality of the research. At the same time it is regrettable that all examples cited are exclusively based on experiences of developed economies (USA, UK, Canada, Australia, New Zealand). In using these countries as examples, SARS might have looked at countries that have a similar tax system as that of South Africa, this criteria should however be balanced by also comparing South Africa with countries that have a similar economy and challenges as South Africa, particularly, other developing countries.

Our dual economy challenge should be taken into account when drafting legislation, particularly legislation that has far-reaching economic consequences.

The following challenges particularly affect the taxation policy of developing countries:





Administrative capacity

Radian, A. 1980. makes the following observations, "Two of the most difficult problems are to collect and process the information efficiently and to audit, assess, and enforce the taxes in such a way that compliance is increased and evasion reduced". These problems can be greatly reduced by attracting high-quality staff in sufficient numbers, with the right facilities (essentially a good institutional memory) and well-defined tax law with the means of enforcement"


South Africa has identified skills shortage as one of its challenges; the SARS is also not immune to this challenge. It is therefore important to ensure that sufficient skills are attracted by SARS to avoid an incorrect enforcement of tax laws.

Secondly, the audit process and assessment of taxpayers should be fine-tuned such that on overzealous attempt to penalize legitimate economic transaction is avoided.

Thirdly, the enforcement process should be clear enough and fair for the taxpayer as to anticipate the tax implications of transactions prior to entering into those transactions. We therefore welcome SARSís intention to modify and improve the Advanced Tax Ruling System.








Availability of information regarding the impact of tax avoidance in South Africa

While the discussion paper quotes numerous examples regarding the impact of tax evasion worldwide, it appears as if the extent of the problem has not particularly been quantified in South Africa. In reference the Margo Commission of Enquiry (1986) and the First Interim Report of the Commission of Inquiry into certain aspects of the Tax Structure of South Africa (1984), the tax gap (evasion) in South Africa is estimated to be between 10% and 33%. It should be noted that these estimates are over 10 years old and were conducted against the background of different political dispensation in South Africa.

The trend in South Africa since 1994 has been an increase in revenues collected by the fiscus and an increased level of compliance by taxpayers. Taken alone, these factors could indicate a decline in the level of tax evasion; however, these factors alone are not conclusive. It is therefore recommend that the extent of the evasion be properly researched and quantified in South Africa and be not mainly motivated by the difficulty to apply the current anti- avoidance provision.


Effective tax system

Among other criteria, the GAAR seeks to penalize transactions that have a divergent economic substance to their form, (" commonly referred to as substance over form"). In a similar fashion, it is important to examine the effective tax system rather than the nominal or legally defined system (Newbury et al). The revenue authorities must quantify the structure of the tax system, how much revenue is collected per commodity or income, whether evasion is systematically related to income, location, line of business, etc. This will better equip SARS with appropriate, direct legislative measures to minimize abuse of tax laws instead of general, non- specific measures.

Causes of tax evasion

Some of the factors identified in the discussion paper as forces that increase the incidence of tax evasion are:

While these forces do contribute or make it easier to enter into transactions that have a tax evasion motive, it should also be clearly stated that the intention of GAAR is not to punish the above economic transactions.

It is also acknowledged that in the modern economy, tax competitiveness of countries plays an important role in attracting foreign capital. Therefore, GAAR should not be applied in contravention of other measures that are implemented by the government that are aimed at attracting foreign capital. A transaction should thus not fall foul of GAAR simply because it leads to a lesser tax liability than any other transaction, that is implemented in a manner that results in higher taxes.


The role of tax advisors

A connection is somehow made in the discussion document that the increase in the amount of resources and talent being invested by professional firms in the area of tax laws could directly be linked to the increase in the level of tax evasion.

While some professionals do engage in impermissible tax avoidance activities, the majority of tax practitioners, it is submitted are much more conservative and would rather advise their client to be more tax compliant than tax aggressive. The application of GAAR should therefore not punish legitimate unintentional mistakes by tax advisors.


The impact of tax legislation

The more complex the tax laws, the more there is a likelihood that such laws could be misunderstood or be prone to abuse by taxpayers, mainly due to lack of clarity. In the past five years, several complex and large pieces of legislation have been enacted in South Africa, e.g. the residence basis of taxation, capital gains tax, taxation of controlled foreign entities, company restructuring relief measures, etc, All these provisions are relatively new and their impact or consequences have not been clearly demonstrated as yet. A wholesale change to the GAAR could have an unintended outcome on the application of these new tax laws.

Johnson, Kaufmann, and Zoido-Lobaton (1998b), as quoted by Chamsa in a study of Public Policy and The Shadow Economy (G. Djolov) conclude that it is not higher tax rates per se that increase the size of the shadow economy, but the ineffective and discretionary application of the tax system and regulations by governments. In their study they find a positive correlation between the size of the shadow economy and the corporate tax burden.

SARS should thus consider a phased-in and piecemeal amendment to the GAAR rather a direct importation of measures implemented in other countries.


Administration costs

Reported cases, both in South Africa and other developed countries indicate that in almost all instances where the GAAR has been invoked by the tax authorities, the matter is more likely to end up in a court of law than not. The likelihood of the matter ending in court is increased where an unfair burden of proof, that a scheme is abnormal or solely entered into to avoid taxes, is placed on the taxpayer. Therefore, the successful use of GAAR by the tax authorities is likely to be costly, both in terms of resources dedicated and opportunity costs of resources that could have been applied elsewhere (court time, economic cost to the taxpayer, acrimonious relations between taxpayer and enforcer). Therefore, a cost/benefit analysis should be considered prior to subjecting a tax matter to the GAAR.

Proposed changes

  1. Applicability of GAAR
  2. The proposed changes to section 103 of the Income Tax Act, (the Act) would make the GAAR applicable whenever an arrangement has created a "tax benefit" and the taxpayer has been unable to rebut a presumption of "abnormality" in the arrangement or transaction.

    While factors that could indicate abnormality are listed in the discussion paper, sadly, there are no factors listed that could be objectively applied in determining whether the arrangement or transactionís sole or dominant purpose was the avoidance of tax.

  3. Indicia of abnormality
  4. The proposed section (4)(b) of the GAAR is very stringent. As the discussion paper correctly asserts, the factors listed as indicia of abnormality are not exhaustive and it is submitted, the mere presence of one of the factors in an arrangement cannot provide conclusive evidence that the arrangement is abnormal.

    It is also not clear as to why only factors listed in section 2 (d) to (k) create a presumption, until the contrary is proved, that the arrangement or any step therein or part thereof was entered into or carried out by means or in a manner which would not normally be employed for bona fide business purposes (other than obtaining a tax benefit)

    The proposed section 4(b) could inadvertently lead to a formulaic approach to the implementation of the GAAR in circumstances or arrangements that materially differ from each other.

    We welcome the intention by SARS to expand on these indicia of abnormality in the Explanatory Memorandum.




  5. Application to steps in a larger scheme
  6. It should be emphasized that GAAR is to be applied to transactions or arrangements that have the sole or main purpose of tax avoidance. To select parts of an arrangement without regard to the dominant purpose of the arrangement could lead to absurd results. Therefore, further guidance is required from SARS as to when and how parts of the arrangement could be analysed in isolation.

    The discussion paper creates the impression that this amendment is considered only to bring South Africa in line with Australia and New Zealand.

    There is evidence that some Black Economic Empowerments transactions for instance, have not necessarily been carried out at armsí length or at market values. This has been done intentionally to accommodate the BEE partners (e.g., due to lack of capital), in the process an alternative that results in a lesser tax liability or that postpones the tax liability may be selected. To then subject the whole transaction to the GAAR without considering the dominant purpose of the transaction could be misleading.

  7. Application in the alternative

Where the Commissioner has applied section 103 in addition or as an alternative to any other section of the Act, the Commissioner, it is submitted, should then prove that invocation of section 103 was indeed justified, this should be the case even when the Commissioner has successfully applied the other anti-avoidance sections of the Act. The above is necessary in order to avoid painting every transaction of the taxpayer with a tax evasion brush.


The South African revenue authorities have to take the relevant steps to defend the tax base of the country. Tax evasion schemes and the promoters of such schemes have to be heavily penalized, however, a careful balance between the rights of the taxpayer and those of the tax enforcer should be maintained.