Upload
others
View
9
Download
0
Embed Size (px)
Citation preview
AUGUST 2012 – ISSUE 155
CONTENTS
CAPITAL GAINS TAX
2092. Redemption of redeemable shares
INTERNATIONAL TAX
2096. Breaking ordinary residence
COMPANIES
2093. The venture capital tax regime
VALUE-ADDED TAX
2097. Prepayments retained – time of supply
DEDUCTIONS
2094. Research and development
expenditure
SARS NEWS
2098. Interpretation notes, media releases and
other documents
DIVIDENDS TAX
2095. Dividends in specie to non-residents
CAPITAL GAINS TAX
2092. Redemption of redeemable shares
(Published August 2012)
The decision of the Johannesburg Tax Court in A (Pty) Ltd v Commissioner for the South African
Revenue Service, handed down on 13 February 2012 (Case No. 12644; not yet reported),
addresses for the first time by a South African court hitherto unexplored aspects of key concepts
in the capital gains tax regime laid down in the Eighth Schedule to the Income Tax Act No. 58 of
2
1962 (the Act), in particular, the fundamental nature of a disposal and the essential nature of a
redemption of redeemable shares.
Was the capital loss in question a clogged loss?
In this case, SARS had disallowed a capital loss claimed by the taxpayer company which had
been incurred as a result of the redemption of redeemable preference shares held by it in a
second company in the same group on the basis that the loss was a “clogged loss”, as envisaged
in paragraph 39(1) of the Eighth Schedule to the Act.
The first issue was the proper interpretation of paragraph 39(1) and, in particular, whether that
provision applied to the redemption of redeemable preference shares, thereby rendering the
taxpayer’s capital loss a “clogged loss” because of the connection between the taxpayer and the
company which had issued those shares, for it was common cause that they were part of the same
group of companies and were under common control.
A clogged loss is a loss which, in terms of the Eighth Schedule, must be disregarded in the
determination of the disposer’s aggregate capital gain or aggregate capital loss; such a loss is
ring-fenced and is deductible only from a capital gain arising from the disposal of assets to that
same “connected person”.
Does the redemption of redeemable shares give rise to a recovery of their acquisition cost?
The second issue before the court related to the quantum of the appellant’s capital loss. This
turned on the meaning of the word recovery in paragraph 20(3) of the Eighth Schedule; the
pivotal question was whether the taxpayer had recovered part of the expenditure incurred in
purchasing the preference shares when those shares were redeemed by the issuing company.
Was the redemption of the shares a “disposal to any person”?
If paragraph 39(1) of the Eighth Schedule were applicable, it was clear that the taxpayer’s
capital loss would be a clogged loss, since it was common cause that the taxpayer and the issuing
company were “connected persons”.
The question was whether this paragraph was indeed applicable. In this regard, the language in
which this provision of the Eighth Schedule is expressed was of critical importance. The
spotlight fell on that part of the paragraph which states that a person must disregard –
3
any capital loss ... in respect of the disposal of an asset to any person
(a) who was a connected person in relation to that person immediately before that disposal;
or
(b) which is immediately after the disposal
(i) a member of the same group of companies as that person..
The term disposal is defined in paragraph 11(1) of the Eighth Schedule, and expressly includes a
redemption.
SARS argued that paragraph 39(1) of the Eighth Schedule was applicable and that the
redemption of the preference shares constituted a disposal to the company. The taxpayer
company argued that this paragraph was not applicable because the redemption was not a
disposal “to” any other person.
The nature of a disposal to another person as distinct from a mere extinction of rights
The taxpayer company argued that the kinds of disposal envisaged in paragraph 39(1) are those
in which an asset, or rights in an asset, are transferred from the disposing party to another person;
furthermore, that even though paragraph 11(1) explicitly defines disposal as including
redemption, the language of this provision - and in particular the word “to” in the phrase
“disposal of an asset to any person” had the effect of confining the application of
paragraph 39(1) to those akin to the ones mentioned in paragraph 11(1)(a), such as sales, where
there was a transfer of the asset itself, or of rights in the asset, to another person.
The taxpayer company argued that where shares are redeemed, there is no transfer of the shares
themselves, or of any of the rights represented by the shares, from the shareholder to the
redeeming company, and that the shares or the rights are simply extinguished, and cease to exist.
SARS’s counter-argument was that the redemption of shares is, in essence, a kind of “buy-back”
of the shares and consequently constitutes a disposal to the redeeming company.
The court applied the principles of statutory interpretation
Faced with these arguments and counter arguments as to the proper interpretation of the word
disposal in the context of paragraph 39(1), read with paragraph 11(1), the court looked to the
legal principles regarding the interpretation of fiscal legislation with the spotlight falling on the
4
significance of the word “to” in the phrase “the disposal of an asset to any person” and the
question whether that word could simply be ignored.
In this regard the court (at para [10]) cited the decision in Commissioner for Taxes v Ferera
[1976] 38 SATC 66 for the proposition that –
in interpreting anti-avoidance provisions, such as paragraph 39(1), a wider interpretation is
required so as to suppress the mischief at which the provision is aimed and to advance the
remedy.
The court said (at para [17]) that
The mischief at which paragraph 39(1) is aimed is clearly to prevent the taxpayer from avoiding
or reducing its tax liability by creating a capital loss through the disposal of an asset to a person
(including a company) that it is connected to. To allow such losses ...would provide fertile
ground for the creation of fictitious losses. Tax liability would be reduced while the asset, or the
benefit thereof, would still be retained by the disposer, albeit through the connected person.
The court said (at para [18]) that the wording of paragraph 39(1) clearly covers transactions such
as sales or the transfer of assets (including shares) from the disposer to a connected person or
company. The difficulty, said the court, arises where there is no transfer of the asset or rights in
the asset from the disposer to the connected person.
The court said that, according to established canons of construction, the preposition “to” in the
phrase “the disposal of an asset to any person” could not be ignored unless its inclusion would
result in an absurdity so glaring that it could never have been contemplated by the legislature.
The inclusion of this preposition, said the court-
implies a disposal of a kind in which the asset (or the rights represented therein, in the case of
shares) must be transferred to the connected person.
As to the nature of the redemption of shares, the court held (at para [26]) that
The redemption of shares results in the extinction and not in a transfer of the rights embodied in
the shares to the company redeeming them, or to any other person.
5
The court consequently held that paragraph 39(1) of the Eighth Schedule did not apply to the
redemption of shares in the present case and that the loss incurred by the taxpayer was therefore
not a clogged loss as envisaged in that paragraph.
The quantum of the taxpayer’s loss
The court began its analysis of the quantum of the taxpayer’s loss on the redemption of the
shares in question with the fundamental proposition that a capital gain is the excess of the
proceeds of an asset on its disposal (or, in the case of a deemed disposal, its market value) over
its base cost.
In the present case, the cardinal dispute between the taxpayer company and SARS was whether,
in determining the base cost of the shares, the original purchase price of the shares in question
fell to be reduced by the aggregate preferential dividend and the redemption premium that was
payable in terms of the articles of association of the company which had issued and was now
redeeming the shares.
SARS argued (see para [28] of the judgment) that the dividend and the redemption premium
constituted a recovery as envisaged in paragraph 20(3) of the Eighth Schedule. It is noteworthy
that this argument is in direct contradiction to the guidance given by SARS in its Comprehensive
Guide to Capital Gains Tax (Issue 4) at 8.18, which states that a post-acquisition dividend is not
a recovery in the context of paragraph 20(3). It is indeed surprising that SARS should contradict
itself in contesting an appeal and begs the question whether it considers itself bound by its
published guidance. It is submitted that SARS should adopt a more principled approach in these
situations and concede its publicly disseminated interpretations where taxpayers fall within the
scope of such guidance.
The taxpayer company (see para [30]) argued the contrary, namely that, in the context of the
present case, a recovery as envisaged in paragraph 20, would occur only where it had -got back
into its possession the expenditure (or part of the expenditure) incurred in respect of the
acquisition of the preference shares.
The taxpayer company argued that to treat a dividend or redemption premium as a recovery of
the purchase price of the shares would lead to an absurd result and would be tantamount to
treating rental received by a property owner as a recovery of the purchase price of the property.
6
The company argued further that it had never got back any of the purchase price which it had
paid to the bank from which it had purchased the shares. The redemption premium was received
on account of the redemption of the shares and was not a repayment of part of the purchase price.
It was further argued that paragraph 20(3) referred to an amount that had been recovered and not
to the wider concept of any benefit linked to the acquisition of the asset.
The court held (at para [38]) that
There is merit in the contention that the fruit derived from an asset will, generally, not constitute
a recovery envisaged in paragraph 20(3). On a proper construction of that paragraph, in order
for the amount to be a recovery, the taxpayer must have got back the cost (or part) expended in
acquiring the asset. The fruits of the asset, such as rent, in the case of the assets being a rental
property, or dividends earned in respect of the shares, are, generally, not amounts that have
been recovered as contemplated in paragraph 20, but constitute income earned from the
particular asset.
The court held (at para [43]) that, when calculating the taxpayer company’s capital loss, SARS
had erred in treating the dividend portion and the redemption premium portion of the redemption
payments as recoveries of the cost of acquiring the preference shares.
The decision
Paragraph 39(1) of the Eighth Schedule, said the court, did not apply to the redemption of the
preference shares and the taxpayer company’s loss was not a clogged loss that could be deducted
only in the manner envisaged in that paragraph. As to the quantum of the capital loss, the
dividends and redemption premium paid to the taxpayer company was not treated as a recovery
envisaged in paragraph 20(3).
pwc
IT Act: Eighth schedule paragraph 11(1), 20(3) and 39(1)
Comprehensive Guide to Capital Gains Tax (Issue 4)
COMPANIES
7
2093. The venture capital tax regime
(Published August 2012)
The venture capital company (VCC) tax regime was first created in 2008. Its purpose was to put
in place a structure whereby investors could pool their funds in an intermediate entity, namely
the VCC, which would then channel those funds into investments in small businesses and junior
mining ventures.
The VCC would provide equity capital and supportive management services and, in return,
would acquire a major stake in those ventures until they reached a sustainable level of maturity
(envisaged as being between five and ten years), at which juncture the VCC would sell its stake
at a profit and distribute any profit to its shareholders.
It was, from the outset, realised by SARS and Treasury that a VCC’s investments in this regard
would be high-risk, in which a few large winners would, hopefully, compensate for losses in
other ventures. Investment in the shares of a VCC would be concomitantly high-risk.
As a fiscal inducement to off-set the commercial risk, taxpayers who invested in a VCC by
acquiring its shares would derive an upfront deduction for expenditure incurred for this purpose -
a generous concession, given that investment in equities is normally non-deductible, being
expenditure of a capital nature. On the eventual sale of the equity stake by the VCC, the proceeds
so derived would be a taxable recoupment of the initial deduction.
The VCC fiscal benefits proved insufficient to attract investors
It is now accepted that the VCC tax regime did not succeed in its objective of attracting investor
interest. There were few applications to SARS for approval of venture capital companies, and by
the end of 2011, not a single VCC had been initiated. In hindsight, it seems that the fiscal
benefits were too small, and that the statutory criteria for qualifying small business and mining
ventures were excessively restrictive and complex.
The 2011 tax amendments
To address these concerns, and try to breathe new life into the VCC initiative, the amendments
made by the Taxation Laws Amendment Act No. 24 of 2011 provide, by way of amendments to
section 12J, for a general relaxation of the statutory requirements, balanced by anti-avoidance
requirements which are intended to ensure that the VCC fiscal regime cannot be exploited to
create tax-driven deductions of little or no value to the ostensible beneficiaries.
8
The amendments, summarized below, take effect from years of assessment commencing on or
after 1 January 2012.
Investor criteria
The prior statutory ceiling amounts for deductible expenditure on the acquisition of shares in
VCC companies and the restrictions on qualifying taxpayers have been completely removed. In
addition, the restriction which permitted only natural persons to qualify for the deduction have
been removed and any taxpayer, including legal entities, can now qualify for the deduction of
expenditure incurred in acquiring shares in an approved venture capital company. The previous
ceiling of R750 000 on investments has been removed.
However, three new anti-avoidance provisions have been introduced.
The first will ensure that the deduction is not available to investors who become “connected
persons” as a result of, or on completion of the investment. This is intended to ensure that a
deduction cannot be gained from cycling funds among closely connected parties rather than from
obtaining new independent investments.
The second anti-avoidance provision allows deductions only if the investments in the VCC are
pure equity investment with no debt-like features.
Thirdly, the investor must be genuinely “at risk”, in other words, invested funds which are
derived from a loan or credit facility must be genuinely subject to the economic risks of the
project. An investor is clearly at risk where the funds come from his own resources or from a
loan or credit facility on full-recourse terms, in which the loan must be repaid even if the VCC
does not achieve its objectives and the invested funds are not recouped by a return on the
investment.
The amendments to section 12J provide that the “at risk” requirement will not be satisfied if the
loan or credit facility is directly or indirectly provided by the VCC itself, nor if the period within
which the loan or credit is to be repaid exceeds five years. This is intended to disqualify schemes
where repayment is so long delayed that it becomes meaningless after inflation is taken into
account.
Venture capital company criteria
9
The criteria for qualification as a venture capital company have been significantly relaxed in
order to simplify the system and eliminate unnecessarily burdensome requirements.
Firstly, VCC’s will not be disqualified merely on the ground that they are listed on the JSE: it
has been decided that there were no sound reasons to support the prior restriction in this regard.
Secondly, the VCC is now permitted to be part of a group, regardless of whether it is a
controlling group company or a controlled group company. Some ownership limitations have
been retained in respect of qualifying investments in respect of small business or junior mining
ventures.
Thirdly, the prohibition on having more than 20% of passive income in a single year has been
dropped; henceforth, a temporary cash accumulation will not be a disqualifying factor. However,
the VCC is still required to outlay at least 80% of its expenditure on qualifying small business
and junior mining companies as from the date that the VCC is approved by SARS.
Fourthly, the minimum investment requirements have been dropped, and the VCC is no longer
required to invest a minimum of R30 million in order to acquire a qualifying company that
constitutes a small business, or a minimum of R150 million to acquire a junior mining company.
Fifthly, the diversification requirements have been dropped which previously mandated that a
VCC could not invest more than 15% of its total expenditure on any one qualifying company
engaged in small business or that was a junior mining company; the requisite percentage has
been increased to 20%. In effect, therefore, a VCC can satisfy the criteria by investing in at least
five qualifying companies.
Qualifying investee companies
The 2011 amendments have relaxed the requirements relating to qualifying investee companies.
As was noted above, a VCC is required to invest at least 80% of its expenditure in qualifying
investee companies. Previously, qualifying investee companies engaged in small business could
not have a book value exceeding R10 million and qualifying junior mining companies could not
have a book value exceeding R100 million. These thresholds have been revised to R20 million
and R300 million respectively.
The prior restrictions as to qualifying company ownership have also been relaxed. Previously,
qualifying investee companies could not be more than 50%-owned by the VCC. This
10
requirement was regarded as inappropriate because private equity funds (on which the VCC
regime is modelled) often maintain temporary control of qualifying companies for an incubation
period in order to retain managerial oversight. This threshold has been relaxed so that the
previous prohibition on the VCC can own up to 70%, and the 30% margin can then attract
independent participants.
Finally, the previous prohibition on franchisees has been dropped, and VCCs will be able to
invest in qualifying companies that are small businesses that operate as franchisees.
Overview
Investment in the shares of approved VCC companies offers the prospect of a guaranteed
deduction to taxpayers who find themselves with an embarrassment of otherwise taxable income
as the end of the tax year approaches. Assuming that the shares in the VCC Company more or
less hold their value, the taxpayer will be able to recoup the deduction, piecemeal if so desired,
timed in the most tax-effective manner. However, a taxpayer who is thinking along these lines
would probably only consider shares in a VCC that is listed on the JSE, for only then can there
be an assurance of being able to find a buyer, without delay, when a decision is made to dispose
of the shares.
pwc
IT Act: Section 12J
Taxation Laws Amendment Act No. 24 of 2011
DEDUCTIONS
2094. Research and development expenditure
(Published August 2012)
Treasury and SARS are acutely aware of the need for South Africa to become more actively
engaged in scientific and technological innovation (to save the country having to pay exorbitant
royalties to foreigners) and have perhaps cast envious eyes over the ocean to our antipodean
cousins who have so successfully established themselves as world leaders in various specialist
areas of technological innovation.
11
Some research and development (R & D) expenditure probably qualifies, in any event, for
deduction under the general deduction provisions of section 11(a), but some such expenditure
may well be of a capital nature and therefore not deductible under that provision; the dividing
line in this regard may be difficult to draw.
It was no doubt because of the uncertain ambit of section 11(a) in regard to R & D expenditure
as well as to provide a strong fiscal and financial incentive that section 11D of the Income Tax
was formulated to provide a specific and generous tax allowance for qualifying expenditure in
this regard.
The 150% tax deduction accorded by section 11D to qualifying expenditure on research and
development was first introduced as from 2 November 2006, and has undergone various
amendments since then.
Section 11D has been redrafted in its entirety
The whole of section 11D is now to be replaced for expenditure incurred in respect of “research
and development” (as now defined) incurred on or after 1 April 2012 (or such later date as is
determined by the Minister of Finance) but before 1 April 2022.
Editorial comment: In terms of the Draft Taxation Laws Amendment Bill, 2012, these provisions
are proposed to apply to “research and development” expenditure incurred on or after 1
October 2012.
Implicitly, therefore, SARS is thus promising that this generous deduction will not be withdrawn
before 1 April 2022. Without some such assurance, the deduction would probably have been a
dead letter, as no enterprise is likely to set up expensive infrastructure and personnel for research
and development if the deduction rug could be pulled out from under its feet at any time.
However, section 11D(12)(b)(iv)(bb) explicitly contemplates that this section may undergo an
“amendment” or “adjustment”. It is regrettable that the latter provision did not add “but not
retrospectively and not in such a manner as to deny the deduction of on-going expenditure to
which the taxpayer had committed itself prior to such amendment or adjustment”.
The Taxation Laws Amendment Act No. 24 of 2011 amends section 11D in regard to the
substantive aspects contained in section 11D(1) - (10) while the Taxation Laws Second
12
Amendment Act 25 of 2011 amends the administrative provisions contained in section 11D(11) -
(18).
Both the substantive and the administrative provisions are lengthy and complex and any taxpayer
tempted by the fiscal carrot of a 150% tax deduction should not make any financial commitment
until professional advisers have scrutinised the proposed project and measured it against the
criteria and administrative aspects of section 11D.
The scope of qualifying R & D projects
The clear intent of the new section 11D, as with its soon-to-be-repealed predecessor, is to
provide for a 150% deduction in respect of scientific and technological research and
development.
Clearly, expenditure incurred to try to develop an AIDS vaccine will qualify for the fiscal
incentive. But what about the development of a new bird seed to improve the stamina of racing
pigeons? And how are the tax rules to differentiate between expenditure on research aimed at
developing an improved product and the routine quality control processes that virtually every
manufacturer engages in?
It is extraordinarily difficult to draw these and other lines and to express qualifying criteria for
R & D expenditure clearly and unambiguously. And of course, SARS faces great problems of
monitoring and enforcement to ensure that a taxpayer engaged in genuinely qualifying scientific
research does not include the salaries of his ordinary administrative staff as R & D expenses.
What R & D projects will qualify for the 150% deduction?
As regards the over-arching question as to what kind of research is worthy of the fiscal incentive,
section 11D, in both its old and its new format, adopts the only workable criterion: expenditure
incurred in any research and development for the purpose of discovering scientific or
technological knowledge or for creating any invention that is patentable under the Patents Act
No. 57 of 1978, any design that is registrable in terms of the Designs Act No.195 of 1993, any
computer program as defined in the Copyright Act No. 98 of 1978, or any knowledge essential to
such invention, design or computer program can potentially qualify for the 150% deduction.
Thus, it may seem that expenditure incurred in developing new birdseed will indeed qualify for
the fiscal incentive, provided that it will be patentable.
13
Section 11D in its new format goes further than its previous incarnation and now usefully
provides for the deductibility of expenditure incurred, not merely in creating a new, but in
improving any such invention, design or computer program if the expenditure relates to a new or
improved function, improved performance, improved reliability or improvement of quality in
that invention, design, computer program or knowledge.
In order to be deductible, the expenditure in question must be –
actually incurred;
by the taxpayer;
directly and solely in respect of research and development undertaken in the Republic.
The words “directly and solely” are, it seems, intended to exclude administrative expenses such
as the salary of staff who are not themselves engaged in research and development.
In the production of income and in the carrying on of any trade
In addition to satisfying the aforegoing requirements, qualifying expenditure must be incurred –
in the production of income; and
in the carrying on of any trade.
These are of course also key requirements for the deductibility of expenditure under the general
deduction provisions of section 11(a), but it is noteworthy that, unlike section 11(a), expenditure
is not disqualified from deduction under section 11D merely because it was “of a capital nature”.
Thus, capital expenditure that satisfies the criteria for deductibility in section 11D will also be
deductible.
Conversely, non-capital research and development expenditure that fails to satisfy the criteria in
section 11D may be deductible under section 11(a) - but of course, the incentive for a taxpayer to
bring his expenditure under section 11D is the prospect of a 150% deduction.
The establishment and functions of the statutory committee
Apart from the statutory criteria, outlined above, that must be satisfied for the expenditure on
research and development to qualify for the 150% deduction, there is another significant hurdle
to be cleared by the taxpayer.
14
The amended section 11D provides for the establishment of a statutory committee which will be
required to “evaluate” taxpayers’ applications.
Now, it could indeed make sense to have a technical committee to determine whether the
taxpayer’s application for approval of research that may qualify for the 150% deduction satisfies
the criteria set out in section 11D(1)(a) and (b), namely that it is research directed to discovering
new scientific or technological knowledge or creating a patentable invention, a registrable design
or a computer program.
But the difficulty is that section 11D(9) says that –
“The Minister of Science and Technology …’must approve any research and development being
carried on or funded for the purposes of [the additional 50% deduction provided for in s 11D(3)]
having regard to—
(a) the innovative nature of the research and development;
(b) the extent to which carrying on that research and development requires specialised skills;
and
(c) such other criteria as the Minister of Science and Technology in consultation with the
Minister of Finance may prescribe by regulation.
These are additional criteria, distinct from and now - at the tail-end of section 11D -
superimposed upon those laid down at the outset in section 11D(1)(a) and (b).
It is presumably at this stage that the taxpayer who proposes to engage in R & D to develop new
and improved birdseed will find that his application for approval is turned down. But why, he
will no doubt protest - a new, improved bird seed will be patentable and will therefore satisfy the
criteria in section 11D(1)(a) and (b)?
To which the Committee’s answer, via the Minister (and as foreshadowed in section 11D(9),
quoted above) will no doubt be, ‘yes, but it’s not innovative enough’.
In short, the criteria set out in section 11D(9) are not objective criteria, but involve subjective
views of committee members as to the degree of innovation involved in the proposed research
and the level of “specialised skills” that it involves.
15
Consequently, the opening words of section 11D(9) which say that the Minister “must approve”
– thereby connoting an obligation and apparently heralding criteria which, if fulfilled, will give
the applicant an absolute right to approval of the application - are wholly misleading. The criteria
in this sub-section are so rubbery and subjective that an adverse decision will be very difficult to
challenge by way of judicial review.
Section 11D(9) thus seems to contemplate that the Minister can decline to approve an R & D
application - even though it fulfils the earlier criteria laid down in section 11D(1)(a) and (b),
namely that it was aimed at creating a patentable invention, registrable design or a computer
program.
It is, however, significant that section 11D(16) explicitly provides that the Minister must provide
written reasons for any decision to grant or deny any application for approval of an R & D
project. This clearly foreshadows that an adverse decision can be taken on judicial review by an
unsuccessful applicant.
In conclusion
Expenditure on research and development is, inherently, an expensive gamble, with no guarantee
of worthwhile results. Without a strong fiscal incentive, the business sector may be disinclined to
divert resources to research and development, or may choose to outsource it to countries where it
can be done more cheaply.
It remains to be seen how strong an incentive the revised section 11D turns out to be in this
regard.
It also remains to be seen whether judicial review of a Ministerial decision to decline an
application for research that will qualify for the 150% tax deduction turns out to be an effective
remedy for aggrieved applicants.
pwc
IT Act: Section 11(a), 11D
Taxation Laws Amendment Act No. 24 of 2011
Taxation Laws Second Amendment Act No. 25 of 2011
16
DIVIDENDS TAX
2095. Dividends in specie to non-residents
(Published August 2012)
Mr A is tax resident in the United Kingdom (UK). He owns some shares in X Pty Ltd, a private
company incorporated and tax resident in South Africa (SA). On 15 April 2012, X Pty Ltd
declares and pays a dividend in specie to its shareholders, including Mr A.
What are the dividends tax implications?
In terms of section 64EA of the Income Tax Act No. 58 of 1962 (the Act) the person liable for
the dividends tax in respect of an "ordinary" dividend (that is, a dividend that does not constitute
a distribution of an asset in specie) is the beneficial owner of the relevant share (that is, the
person entitled to the benefit of the dividend attaching to a share, usually the registered
shareholder). However, in terms of that provision, the person liable for the dividends tax in
respect of a dividend which consists of a distribution of an asset in specie is the company that
declares and pays the dividend.
In the Explanatory Memorandum on the Taxation Laws Amendment Bill, 2011 – the Bill that
introduced many changes to the dividends tax regime – the National Treasury says (at page 38)
that "[t] reaty relief is also available for in specie dividends…"
That statement is perhaps a bit of an over-simplification.
As noted above, the dividends tax in the case of a dividend in specie is a tax on the company, and
not a tax on the beneficial owner. So, as a non-resident shareholder is not liable for the dividends
tax in that case, no double tax can arise and one is not able to apply a double taxation treaty. In
this respect, the dividends tax in relation to dividends in specie is akin to secondary tax on
companies, which is also a tax on the company and for which no treaty relief is available.
Section 64FA(2) of the Act says that a company that declares and pays a dividend that consists
of a distribution of an asset in specie is liable for the dividends tax at a reduced rate if the person
to whom the payment is made submits a declaration in the prescribed form to the company "that
17
the portion of the dividend that constitutes a distribution of an asset in specie would, if that
portion had not constituted a distribution of an asset in specie, have been subject to that reduced
rate as a result of the application of an agreement for the avoidance of double taxation…"
Effectively, section 64FA(2) says that, for purposes of dividends tax, a dividend in specie
declared and paid to a shareholder who is not tax resident in South Africa must be treated as if
it were an ordinary dividend which could be reduced by an appropriate treaty.
To return to the example above: X (Pty) Ltd is liable for the dividends tax in respect of the
dividend in specie paid to its shareholder. As Mr. A is not liable for tax, he is not able to simply
by virtue of the UK- SA double taxation treaty require a reduction of the rate of dividends tax.
However, Mr A will be able to fill in the prescribed declaration and require X (Pty) Ltd to reduce
the rate to 10% by virtue of section 64FA(2) read with Article 10 of the SA-UK double taxation
treaty.
To sum up, in the case of a dividend in specie paid to a non-resident shareholder, the shareholder
may be able to reduce the rate of dividends tax not simply by virtue of a double taxation
agreement, but rather by virtue of South African domestic law read with a relevant double
taxation agreement. A subtle, but perhaps important distinction.
Cliffe Dekker Hofmeyr
IT Act: Section 64EA, 64FA(2)
SA-UK DTA – Article 10
INTERNATIONAL TAX
2095. Breaking ordinary residence
(Published August 2012)
We previously commented on the importance of an individual who ceases to be ordinarily
resident in South Africa being able to substantiate factually a clear intention to leave South
Africa permanently and thereby to “break” ordinary residency in South Africa. This article
stirred up much debate and discussion. In particular it is important to debate and discuss the
18
complexities of tax law since it is only through rigorous debate that one crystallises the relevant
issues and either finds common ground or gains an understanding of another party’s position.
This issue is of great importance. We therefore, in this follow-up article, provide further detail on
the “ordinarily resident” test and the “physical presence” test.
The income tax system in South Africa changed from a source-based system of taxation to a
residence basis of taxation with effect from years of assessment commencing on or after
1 January 2001. A resident, for South African tax purposes, is defined in section 1 of the Income
Tax Act No. 58 of 1962 (the Act), as follows:
“resident” means any-
(a) natural person who is-
(i) ordinarily resident in the Republic; or
(ii) not at any time during the relevant year of assessment ordinarily resident in the Republic,
if that person was physically present in the Republic-
(aa) for a period or periods exceeding 91 days in aggregate during the relevant year of
assessment, as well as for a period or periods exceeding 91 days in aggregate during each of the
five years of assessment preceding such year of assessment; and
(bb) for a period or periods exceeding 915 days in aggregate during those five preceding years
of assessment:
in which case that person will be a resident with effect from the first day of that relevant year of
assessment: Provided that–
(A) a day shall include a part of a day, but shall not include any day that a person is in transit
through the Republic between two places outside the Republic and that person does not formally
enter the Republic through a ‘port of entry’ as contemplated in section 9(1) of the Immigration
Act, 2002 (Act No. 13 of 2002), or at any other place as may be permitted by the Director
General of the Department of Home Affairs or the Minister of Home Affairs in terms of that Act;
and
19
(B) where a person who is a resident in terms of this subparagraph is physically outside the
Republic for a continuous period of at least 330 full days immediately after the day on which
such person ceases to be physically present in the Republic, such person shall be deemed not to
have been a resident from the day on which such person so ceased to be physically present in the
Republic; or
(b) person (other than a natural person) which is incorporated, established or formed in the
Republic or which has its place of effective management in the Republic,
but does not include any person who is deemed to be exclusively a resident of another country
for purposes of the application of any agreement entered into between the governments of the
Republic and that other country for the avoidance of double taxation;”
There are a few general observations that may be made in respect of the application of this test to
individuals. Firstly there are two separate tests for residence, namely, the test of “ordinarily
resident” and the “physical presence” test. In other words, two tests are applicable in order to
determine whether or not a person is a resident of South Africa and if either one applies, the
individual will be regarded as being “resident” in South Africa for tax purposes.
This is best illustrated by a simple practical example. Assume an individual passes the physical
presence test, i.e., is not physically present in the Republic for the period stipulated in (aa) and
(bb) or (B) of the definition of “resident”. However, assume the individual is, during that period,
“ordinarily resident” in the Republic, the individual will then qualify as a “resident” in terms of
section 1 of the Act.
A second general observation is that any person who is deemed to be exclusively a resident of
another country for purposes of the application of any agreement entered into between the
governments of South Africa and that other country for the avoidance of double taxation (double
tax agreement) is specifically excluded from the definition of a “resident” as defined in section 1
of the Act. This represents an acceptance of the ‘tie-breaker’ rule commonly found in double tax
agreements to resolve dual residency problems. It is important to note that this effective
exclusion applies only if the person is deemed to be exclusively a resident of the other country
that is a party to the double tax agreement.
20
Again we set out a practical example. If, in terms of the tie-breaker test in Article 4(2) of the
double tax agreement between South Africa and Luxembourg, an individual qualifies as a
resident of Luxembourg, then such person will not qualify as a “resident” of South Africa in
terms of the definition thereof in section 1 of the Act.
The ordinarily resident test
The Act does not define “ordinarily resident” and the question of whether a person is “ordinarily
resident” is one of fact based on the interpretation given by the courts. In Cohen v CIR [1946] 13
SATC 362, it was recognised that it has long been established that a person may have more than
one residence for the purpose of income tax, but that whether a person can be “ordinarily
resident” in more than one country was unclear. The court suggested that a person’s ordinary
residence would be the country to which he would naturally and as a matter of course return
from his wanderings. It would be natural to interpret “ordinarily” by reference to the country of
his most fixed or settled residence and therefore a person can only be “ordinarily resident” in one
country. In CIR v Kuttel [1992] 54 SATC 298, the court noted that a person may have more than
one residence at a time.
In the Cohen case the court held that the question whether an individual was in any one year of
assessment ordinarily resident in South Africa, or elsewhere, was not to be determined solely by
his actions during that particular year of assessment. His conditions of ordinary residence during
that particular year could be determined by evidence as to his mode of life outside the year of
assessment under consideration.
Whilst it does not have any legal effect, it is interesting to note that Interpretation Note No. 3,
2002, issued by the South African Revenue Service (“SARS”) states that a physical presence at
all times is not a requisite to be ordinarily resident in South Africa and that two requirements
need to be present, namely, an intention to become ordinarily resident in a country, and secondly,
steps indicative of this intention having been or being carried out.
The Interpretation Note states further that a natural person may be resident in South Africa even
if that person was not physically present in South Africa during the relevant year of assessment
and that the purpose, nature and intention of the taxpayer’s absence must be established to
determine whether the taxpayer is still ordinarily resident. The Interpretation Note sets out a list
of factors which SARS will take into account in determining whether a person is ordinarily
resident in South Africa as follows:
Most fixed and settled place of residence
21
Habitual abode, i.e. present habits and mode of life
Place of business and personal interest
Status of individual in country i.e. immigrant, work permit periods and conditions, etc.
Location of personal belongings
Nationality
Family and social relations (schools, church, etc.)
Political, cultural or other activities
Application for permanent residence
Period abroad; purpose and nature of visits
Frequency of and reasons of visits.
The above list is not intended to be exhaustive or specific, it is merely a guideline.
The Interpretation Note also states that the circumstances of the individual must be examined as
a whole and the personal acts of the individual must receive special attention. As stated in ITC
1170 [1971] 34 SATC 76 one is entitled to look at the taxpayer’s life beyond the particular
period under consideration.
The physical presence test
As stated above, in terms of the definition of a “resident” in section 1 of the Act, a resident
includes any natural person who is not at any time during the relevant year of assessment
ordinarily resident in South Africa, if that person was physically present in South Africa:
for a period or periods exceeding 91 days in aggregate during the relevant year of
assessment, as well as
for a period or periods exceeding 91 days in aggregate during each of the five years of
assessment preceding such year of assessment, and
for a period or periods exceeding 915 days in aggregate during those five preceding years
of assessment.
It is important to note that in the year of assessment that a taxpayer either ceases to be ordinarily
resident in South Africa or commences being ordinarily resident in South Africa, the physical
presence test cannot be applied. This is because the physical presence test explicitly only applies
if a person was not ordinarily resident in South Africa at any time during the relevant year of
assessment.
22
The effect of the above definition is that a person who is not ordinarily resident in South Africa
is, in terms of the physical presence test, a resident after physical presence in South Africa
exceeding the above limits over a period of six consecutive years of assessment.
A person who becomes tax resident by virtue of the physical presence test will become a resident
from the first day of the year of assessment during which all the requirements of the test are met.
Accordingly, an individual would become tax resident in terms of this test from the beginning of
the sixth consecutive year of assessment in which he has had a physical presence in South Africa
exceeding the above limits.
For the purposes of determining the number of days during which a person is physically present
in South Africa, a part of a day is included as a day. A day spent in transit through South Africa
is not included as a day, provided that the person does not formally enter South Africa through a
“port of entry” as contemplated in section 9(1) of the Immigration Act, 2002.
Where a person who is a resident in terms of the physical presence test is physically outside
South Africa for a continuous period of at least 330 full days immediately after the day on which
such person ceases to be physically present in South Africa, such person is deemed not to have
been a resident from the day on which he ceased to be physically present in South Africa.
Practical example for individuals wishing to lose their “resident” status
Whilst it is clear that there are two separate tests which both need to be separately passed in
order for an individual not to qualify as a “resident”, the practical application of these tests can
lead to confusion. We therefore set out a practical example to illustrate the relevant principles.
In particular the example deals with the significance of the physical presence test for individuals
leaving South Africa in the year following the year in which they lose their “ordinarily resident”
status.
Assume that an individual ceased to be ordinarily resident during the 2010 year of assessment
and that individual then returned to the Republic during the 2011 year of assessment for more
than 91 days in aggregate. This occurs not infrequently, that is, individuals emigrate from South
Africa, but are then required to return to South Africa in the following year of assessment in
order to carry on their business activities, visit remaining family etc.
23
In these circumstances, in respect of the 2010 year of assessment the individual would not be
“ordinarily resident” in South Africa from the date during that year that the individual left South
Africa in the circumstances described in, inter alia, Cohen v CIR (supra). This is itself a topic for
discussion, that is, when precisely an individual loses his ordinarily resident status.
In terms of the physical presence test, such test only applies to an individual who is “not at any
time during the relevant year of assessment ordinarily resident in the Republic”. Therefore this
test does not apply during the 2010 year of assessment and therefore from the time that the
individual loses his ordinary resident status during that year of assessment, he arguably ceases to
qualify as a “resident” as set out in section 1 of the Act.
In respect of the 2011 year of assessment that person, not being ordinarily resident in the 2011
tax year should then test the requirements of the physical presence test. Should that individual
have spent more than 91 days in South Africa during each of the preceding five tax years (i.e. in
2006, 2007, 2008, 2009 and 2010 years of assessment) and more than 915 days in total during
that five-year period (which would in all likelihood be the case if they had been ordinarily
resident in South Africa during that period) then that individual would arguably again qualify as
a “resident” of South Africa by virtue of physical presence with effect from the first day of the
2011 year of assessment. However, where a person is regarded as exclusively resident of
another country in terms of a double tax agreement, they cannot be regarded as a resident for tax
purposes in South Africa.
In terms of this analysis the term “relevant year of assessment” in paragraph (a)(ii) of the
definition of “resident” would then refer only to the 2011 year of assessment and not all previous
years of assessment. In this regard the “relevant year of assessment” is to be contrasted with the
“preceding years of assessment” which therefore supports the interpretation set out above.
Conclusion
The definition of “resident” as it pertains to individuals contains two principal tests. These are
the tests of “ordinarily resident” and the “physical presence” test.
An individual must ensure that neither of these tests applies in order not to fall within the
definition of “resident” in section 1 of the Act.
An individual will not fall within the definition of “resident” if he is deemed to be exclusively a
resident of another country for purposes of a double tax agreement.
24
It is in applying the specific provisions of the “ordinarily resident” test and the physical presence
test to specific factual scenarios that individuals must be careful in planning their affairs.
Edward Nathan Sonnenbergs
IT Act: Section 1
SARS Interpretation Note 3
SA-Luxembourg DTA – Article 4(2)
VALUE-ADDED TAX
2097. Prepayment retained - time of supply
(Published August 2012)
Her Majesty Revenue & Customs (HMRC) recently updated their guidance and practice in the
United Kingdom regarding the VAT status of forfeited deposits, following a judgment of the
European Court of Justice (ECJ) (ECJ Case C-277/05).
In that case the ECJ held that a deposit paid by a client to an hotelier for the reservation of a
room is not subject to VAT when the client does not arrive and forfeits the deposit. The ECJ
viewed the forfeited deposits to be fixed compensation for damages suffered by the hotelier
because the contract was not fulfilled as agreed, and that the hotelier did not make any
identifiable supply of any service to the client.
On 1 September 2011 the Full Federal Court of Australia came to a similar conclusion as that of
the ECJ in Case C-277/05 in Qantas Airways Ltd v Commissioner of Taxation [2011] FCAFC
113.
In the Qantas case the Full Federal Court was required to consider whether Qantas received
consideration for a taxable supply where a person booked and paid for a domestic air ticket but
either cancelled the booking or did not turn up for the flight, and did not collect a refund.
The Commissioner for Taxation argued that the passenger received a reservation of an air ticket
when he made the booking, and that the reservation was the taxable supply for which Qantas
25
received the payment as consideration. However, the Full Federal Court overturned the decision
of the AAT and rejected the arguments of the Commissioner. It found that the only relevant
supply was the supply of air travel, nothing more or less, and that the supply of the air travel was
both the substance and the reality of the transaction, which cannot be split into other supplies.
The Full Federal Court therefore found that since the air travel was not supplied, the amount
retained by Qantas where the person cancelled the booking or did not turn up for the flight, was
not consideration for any taxable supply as no supply was made.
These judgments bring into question the VAT status of all prepayments received by a supplier
and which are retained where the goods or services are never supplied. One would need to
identify the relevant supply for which the payment was received, and if the supply is never made
but the payment is retained, the recipient may not be liable for VAT. Where the recipient
accounted for VAT upon receipt of the payment, he may be entitled to a deduction if the supply
is never made.
The answer as to whether such retained payments are subject to VAT will normally be in the
underlying agreement between the parties and the specific facts of each case must also be
carefully considered.
Edward Nathan Sonnenbergs
SARS AND NEWS
2098. Interpretation notes, media releases and other documents
Readers are reminded that the latest developments at SARS can be accessed on their website
http://www.sars.gov.za
Editor: Mr P Nel
Editorial Panel: Mr KG Karro (Chairman), Dr BJ Croome, Mr MA Khan, Prof KI
Mitchell, Prof L Olivier, Prof JJ Roeleveld, Prof PG Surtees.
26
The Integritax Newsletter is published as a service to members and associates of The South
African Institute of Chartered Accountants (SAICA) and includes items selected from the
newsletters of firms in public practice and commerce and industry, as well as other contributors.
The information contained herein is for general guidance only and should not be used as a basis
for action without further research or specialist advice. The views of the authors are not
necessarily the views of SAICA.
All rights reserved. No part of this Newsletter covered by copyright may be reproduced or copied
in any form or by any means (including graphic, electronic or mechanical, photocopying,
recording, recorded, taping or retrieval information systems) without written permission of the
copyright holders.