Abstract:
The general anti-avoidance rule (GAAR) has been adopted by South Africa as one of its
methods to combat tax avoidance schemes into which taxpayers enter. Since 1941 when
the South African GAAR was first introduced into the tax legislation, it has been amended
various times as a result of the weaknesses that were highlighted by its failures to stand up
to the rigours of the courts. However, since the most recent amendment to the South African
GAAR in 2006, its efficacy remains unknown due to the fact that it has not been tested by
the courts in its entirety. This study aims to address this concern by determining the
effectiveness of the South African GAAR when compared to its New Zealand counterpart.
This study employed a ‘structured pre-emptive analysis’ research methodology, which is a
combination of doctrinal and reform-oriented approaches. The doctrinal approach was used
in Phase 1 of the research whereby a doctrinal analysis of the South African and New
Zealand GAARs were performed. This approach allowed an understanding of the
interpretation and application of the two GAARs to be obtained, as well as to allow for the
identification of weaknesses in the South African GAAR, while simultaneously making
suggestions for improvement. The reform-oriented approach was used in Phase 2 of the
study in which the South African GAAR was applied to the facts of a case from New Zealand.
Phase 3 of the study contained the triangulation of the findings from both Phases 1 and 2 of
the study, thereby validating the findings of the study.
The findings from Phases 1 and 2 highlighted various weaknesses that exist in the South
African GAAR which indicate that additional guidance should be provided to address the
existing uncertainties currently contained within the interpretation and application, in order
to prevent inconsistencies that may limit its efficacy. The findings of this study indicate that
for a taxpayer to be considered party to an arrangement, they do not need to be aware of
the entire arrangement nor all of its details. Furthermore, it was noted that the sole or main
purpose requirement should be amended to rather require that obtaining the tax benefit was
one of the purposes, provided it is not merely incidental, as opposed to requiring the tax
benefit to be the sole or main purpose of the arrangement. In addition, it is suggested that
the sole or main purpose test be amended to being a purely objective test and not
considering subjective intent of the taxpayer.
