Date: 20081212
Dockets: A-297-07
A-298-07
A-299-07
A-300-07
Citation: 2008 FCA 398
CORAM: NOËL J.A.
NADON J.A.
TRUDEL J.A.
BETWEEN:
A-297-07
2529-1915 QUÉBEC
INC.
Appellant
and
HER MAJESTY THE
QUEEN
Respondent
- - - - - -
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - -
A-298-07
2530-1284 QUÉBEC
INC.
Appellant
and
HER MAJESTY THE
QUEEN
Respondent
- - - - - -
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - -
A-299-07
ROBERT LANGLOIS
Appellant
and
HER MAJESTY THE
QUEEN
Respondent
- - - - - -
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - - - - - -
A-300-07
RALPH E. FARAGGI
Appellant
and
HER MAJESTY THE
QUEEN
Respondent
REASONS FOR
JUDGMENT
NOËL J.A.
[1]
These
are four appeals from decisions of Rip A.C.J.T.C., as he then was (“TCC judge”),
confirming, further to a common hearing and based on a single set of reasons,
the assessments made by the Minister of National Revenue (“Minister”) under the
Income Tax Act, R.S.C. 1985, c. 1, (5th
Supp.)
(“Act”), with penalties against each of the appellants.
[2]
The
issue arises from a very elaborate tax plan devised and implemented in 1987 by
the appellants Langlois (tax specialist) and Faraggi (corporate law specialist)
while they were members of the firm Stikeman Elliott in Montréal (“Langlois and
Faraggi” or “the devisers”). The appellants 2529-1915 Québec Inc. and 2530-1284
Québec Inc. are companies that were incorporated by the devisers and played a
key role in the implementation of their plan (“1915 Inc.” and “1284 Inc.” or
“the corporate appellants”).
[3]
The
four appeals before this Court, as those before the TCC, were heard together.
Langlois and Faraggi raised the same arguments in support of their respective
appeals, and the corporate appellants did likewise. These reasons deal with and
dispose of the four appeals and will be filed as reasons for judgment in each
of the Court files.
THE FACTS
[4]
The
objective of Langlois and Faraggi’s plan was to produce capital gains in the
hands of a series of corporations in order to create capital dividend accounts
(“CDA” or “CDAs”) and to transfer, subject to the payment of a monetary
consideration, the tax benefit inherent in these CDAs to unrelated companies in
a position to use them (“third-party corporations”). At the end of the exercise,
Langlois and Faraggi together had pocketed the sum of $12,345,255 (detailed
calculations of this amount are set out in Appendix 3 of the judgment under
appeal).
[5]
The
plan was implemented in two phases involving approximately thirty companies
incorporated for that purpose. Langlois and Faraggi were the directors and had
control of these companies. The first series of transactions began on August
13, 1987, and the second, carried out in two phases, began on September 9, 1987.
The transactions are complex. They are described in detail in the Statement of
Facts filed by the appellants in the Tax Court of Canada (Appeal Book (A‑300‑07),
Volume II, pages 200 to 336), and summarized by the TCC judge at
paragraphs 11 to 17 (first series of transactions), and 19 and 20 (second
series of transactions) of his reasons.
[6]
Without
departing from the TCC judge’s summary save to make two clarifications in the
course of the analysis, I propose to focus only on the aspects of the plan that
are essential to the disposition of the appeals.
[7]
The
first series of transactions was funded by a loan akin to a one-day advance
(described in the evidence as a “daylight overdraft”) of $10,000,100 by the
Royal Bank for a period of time during the day on August 13, 1987. The loan was
made to 2258‑5644 Québec Inc., a company incorporated by Langlois and
Faraggi (“2258 Inc.” or “the first subsidiary”). The second series was funded
by Langlois and Faraggi using the money they had made during the first series
of transactions. In total, the first series of transactions yielded gains of
$109,998,900, thus constituting a cumulative CDA of $54,999,450. The second
series generated gains of $123,688,763 in two phases ($58,899,411 in the first
phase and $64,789,352 in the second), resulting in a cumulative CDA of
$61,844,381. These CDAs, up to $82,712,000, were eventually transferred to
third-party corporations.
[8]
Before
going further, a few words on the operation of CDAs are in order. A CDA is
composed of the non-taxable portion of a capital gain generated by a
corporation (i.e., half of the gain realized during the period at issue). The
legislative objective is to ensure that a capital gain is taxed in the same
manner, whether it is earned directly by an individual or indirectly through a
corporation. To this end, the system provides that, subject to the prescribed
elections being made, the non-taxable portion of a capital gain, once included
in a CDA by the corporation that realized it, remains tax-exempt when
transferred from one corporation to another by way of dividend, until it is
ultimately distributed to an individual in the form of a non-taxable dividend,
also called a “capital dividend”.
[9]
Since
not everyone has access to the benefit provided by a CDA (for example, non-resident
shareholders could not receive tax-free dividends), this system, which dates
back to the 1970s, gradually gave rise to corporations that were rich in CDAs that,
however, could not be used by the corporations’ shareholders. As a result,
planning techniques allowing these CDAs to be transferred to corporations owned
by shareholders able to make use of them were developed. These are the
techniques that inspired the plan conceived by the devisers (Robert Langlois’s
testimony, Appeal Book (A‑300‑07), Volume VIII, pages 61 to 63
(1518 to 1520)).
[10]
Contrary
to what was being done at the time, the arrangement put in place by the
devisers was not based on an existing source of CDAs. It included an additional
phase allowing for the generation of capital gains and the creation of CDAs.
The method used in the first series of transactions shows how the devisers of
the plan went about generating these gains and transferring the resulting CDAs
to third-party corporations. The TCC judge based his analysis on this first
series, and I propose to do the same since the appellants did not take issue
with the approach used by the TCC judge in this regard.
[11]
A
series of 13 corporations (“A Inc.” to “M Inc.” in the TCC judge’s summary) was
first created, each holding the totality of the common shares of the
corporation immediately following it in a vertical line. The loan made by the
Royal Bank to the first subsidiary was deposited in its account (opened at the
same bank) and used to issue a certified cheque in the amount of $10,000,000 to
pay for a block of 10,000 preferred shares issued by the second subsidiary; the
second subsidiary then used the $10,000,000 thus received to issue a certified
cheque in this same amount and acquire an identical block of shares issued by
the third subsidiary, which then acquired the preferred shares issued by the
fourth, and so on, until the twelfth subsidiary had acquired the preferred
shares issued by the thirteenth.
[12]
The
second subsidiary after having issued the preferred shares and after these were
subscribed to by the first subsidiary, declared a stock dividend of 10,000
shares. The shares thus issued had a very high redemption value (i.e.,
$10,000,000) and a very low paid-up capital (i.e., $0.01 per share or $100 in
total). The third subsidiary declared an identical stock dividend in favour of the
second subsidiary as the holder of its preferred shares, and so on, until the
twelfth subsidiary, so that at the end of this part of the transactions, twelve
of the thirteen subsidiaries in the chain were holders of special class shares
with a very low tax cost (i.e., adjusted cost base (“ACB”)) and a very high
redemption value (“gain-making
shares”).
The gain-making shares are identified as the Class “K” shares in the TCC judge’s
summary.
[13]
The
second subsidiary subsequently sold the gain-making shares to the first
subsidiary for a value equal to their redemption value: $10,000,000. This debt
was evidenced by a promissory note payable on demand. Given that the
gain-making shares had a tax cost of $100 ($0.01 x 10,000), the
subsidiary thus realized a capital gain of $9,999,900. Each of the other
subsidiaries in the chain (except for the thirteenth) sold the gain-making
shares to the first subsidiary in the same manner, thereby generating on paper cumulative
gains of $109,998,900.
[14]
To
enable the first subsidiary to reimburse the $10,000,100 bank loan, the
thirteenth subsidiary declared a cash dividend of a corresponding amount in its
favour, paid by certified cheque at the end of the exercise on August 13. The
first subsidiary used this amount to reimburse the bank.
[15]
Further
to the sale of the gain-making shares, eleven of the subsidiaries (the second
to the twelfth inclusively) had a purported CDA equal to half of the gain thus
realized, totalling $54,999,450 (Reasons, paragraph 16 (f)). At this point,
these eleven subsidiaries amended the par value of the gain‑making
shares, all held at that time by the first subsidiary, so as to increase the
par value from $0.01 to $500. In doing so, they triggered deemed
dividends in the amount of $54,998,900 in the hands of the first subsidiary, pursuant
to subsection 84(1) of the Act. This provision provides that when a corporation
increases the paid-up capital of its shares, a dividend equal to the increase
shall be deemed to have been received by the shareholder, that is, in this
case, $500 x 10,000 shares = $55,000,000 less the ACB
of the gain‑making shares, i.e., $1,100 (Reasons, paragraph 16 (h)).
[16]
In
respect of this deemed dividend, each of the subsidiaries, from the second to
the twelfth, made the election provided for in subsection 83(2) of the Act,
according to which the dividend
|
(a) … shall be deemed to be a
capital dividend to the extent of the corporation’s capital dividend account
immediately before the particular time; and
(b) no part of the dividend
shall be included in computing the income of any shareholder of the
corporation.
|
a) […] est
réputé être un dividende en capital jusqu’à concurrence du montant du compte
de dividendes en capital de la corporation immédiatement avant le moment
donné; et
b)
aucune partie du dividende ne doit être incluse dans le calcul du revenu de
tout actionnaire de la corporation.
|
[17]
In
accordance with the requirements of subparagraph 89(1)(b)(ii) of
the Act, the first subsidiary included the deemed dividend in its CDA. Under
this provision, when computing its CDA, a private corporation must include
amounts received
|
(ii) … in respect of a
dividend … on a share of the capital stock of another corporation in the
period, which amount was, by virtue of subsection 83(2), not included in
computing the income of the corporation,
|
(ii) […],à titre de dividende versé sur une
action du capital-actions d’une autre corporation, somme qui, en vertu du
paragraphe 83(2), n’a pas été incluse dans le calcul du revenu de la
corporation.
|
[18]
Thus,
as a result of the elections made by the eleven subsidiaries, the cumulative CDAs
of $54,998,900 were transferred to the first subsidiary (Reasons, paragraph 16
(i)). On or about August 21, 1987, the appellant 1915 Inc. subscribed to a block
of preferred shares issued by the first subsidiary for $55,023,216 (Reasons,
paragraph 16 (k)). Immediately afterwards, the first subsidiary declared a
dividend of $49,566,000 to be paid to 1915 Inc., as the owner of the preferred
shares (Reasons, paragraph 16 (l)), and made the subsection 83(2) election
to indicate that the dividend was paid out of its CDA. At this point, the bulk of
the CDAs generated by the subsidiaries was in the hands of 1915 Inc., ready to be
transferred to the third-party corporations.
[19]
Later,
on September 14, 1987, the devisers arranged for the subsidiaries that had
generated gains to sustain losses equal to the gains (i.e., $110,000,000);
they did so by selling the preferred shares acquired at $10,000,000 to Faraggi
for $100.
[20]
At
various times between September 2 and 22, 1987, 1915 Inc. transferred the CDAs
to third-party corporations. These third-party corporations were owned by
shareholders wishing to receive, tax-free, accumulated surpluses which
otherwise could only be distributed on a taxable basis. They had been
approached by the devisers over the preceding months.
[21]
The
transfer of 1915 Inc.’s CDA to the third-party corporations took place in two
stages. First, each of the third-party corporations subscribed to different
classes of preferred shares of 1915 Inc. These shares all had a minimal par
value – for example, $0.01 – and a high redemption value – for example, $1,000.
After the subscription, 1915 Inc. paid a cash dividend of $999.99, thus
reducing the redemption value to $0.01, while at the same time making the
election set out in subsection 83(2), according to which the dividend was
deemed to have come from its CDA. (At the hearing, counsel for the corporate
appellants explained that, contrary to what the TCC judge said in his reasons
(Reasons, paragraph 16 (n)(i) and paragraph 3), the dividend in question
was not a deemed dividend, but rather a cash dividend. Nothing flows from this error).
The third‑party corporations thereby saw their own CDAs increase accordingly
by virtue of subparagraph 89(1)(b)(ii) of the Act (see paragraph 17
above). In the last stage of the plan, 1915 Inc. would redeem the preferred
shares at their reduced redemption value of $0.01.
[22]
Importantly,
the plan contemplated that the third-party corporations would pay a premium
when subscribing to the preferred shares of 1915 Inc. For example, if the
redemption value was $1,000, the third-party corporation was to pay $1,210 per
share. It is this premium of $210 per share that enabled the two corporate
appellants to accumulate surpluses ($8,105,344 in the case of 1915 Inc. and
$4,677,717 in the case of 1284 Inc.) after having paid the cash dividends and
redeemed the shares (see Appendices 1 and 2 to the judgment under
appeal, setting out the computation of the surpluses generated by the corporate
appellants).
[23]
By
proceeding in the same manner as the third-party corporations (save for the
payment of the premium on the preferred shares), the devisers became shareholders
and were paid dividends by 1915 Inc. (and three other corporations that they
had incorporated). The dividends so paid amounted to $6,085,762 in the case of
Langlois and $6,025,624 in the case of Faraggi. In December of the following
year, they were paid another dividend in the amount of $116,934 ($233,868 in
total). The corporations that paid these dividends made the subsection 83(2) election
so that the dividends received by the devisers were also deemed to be capital
dividends.
[24]
It
is important to note that, at the implementation stage of the plan, the
devisers obtained an opinion (in short and long form) confirming the legal
effects of their plan for the benefit of their clients (the opinions are
reproduced as Appendices 5 and 6 to the judgment under appeal). These opinions
are signed by Mr. Maurice Régnier, a noted
practitioner who was at the time a senior partner in the Stikeman Elliott firm.
[25]
Mr.
Régnier testified before the TCC judge and was questioned regarding the scope
of this opinion. The minutes of a meeting held on May 12, 1989, between Mr.
Régnier and Revenue officials show that he did not concern himself with the
source of the CDAs (Reasons, paragraph 51):
The CDA source was never questioned by Mr. Régnier as to
its source; it was presumed to arise from real and legitimate transactions
giving rise to increases to the CDA accounts; to the extent of any inaccuracies
as to the quantum, the Income Tax Act provided for a Part III tax to the
transferor but a full increase in the CDA account of the transferee. Mr.
Régnier never thought of, to say the least, the possibility of a fabrication.
Mr. Régnier did not question the significant quantum of CDA
referred to in the September 2/87 long opinion given to Mr. Langlois -
approximately $49.6 million. In his mind, he believed that Mr. Langlois had
“found” a source similar in scope to SNC, and also remembered Mr. Langlois's
reference to banking arrangements and an acquaintance at Dominion Securities.
[Citation omitted.]
Mr. Régnier would be on a trip for two weeks during
September /87; during his absence Mr. Langlois would 'close' his CDA deals.
Mr. Régnier never participated at any closings.
The TCC judge accepted Mr. Régnier’s testimony that
he had opined on the provisions relied upon by the devisers to implement their
plan, without expressing any view on the source of the CDAs, since this was not
part of his mandate (Reasons, paragraph 52). The TCC judge’s assessment of Mr. Régnier’s
evidence is not questioned in these appeals.
[26]
Relying
on this opinion, Langlois and Faraggi treated the dividends that they received
as having been paid on a tax-free basis. Their 1987 and 1988 income tax returns
do not reflect any inclusion on that account.
[27]
Similarly,
in their 1987 tax return, the two corporate appellants did not account for the
surpluses that they generated. According to the corporate appellants, the fact
that these surpluses arose from a premium that was part of the price paid by
the third-party corporations to subscribe to shares meant that they were
capital contributions, and therefore tax-free.
The assessments
[28]
By
assessments issued August 16, 1995, the Minister added the surpluses generated during
the corporate appellants’ 1987 taxation year in the computation of their income
for that year (i.e., $8,105,344 in the case of 1915 Inc. and $4,677,717 in the
case of 1284 Inc.), with penalties. According to the Minister, these surpluses
constituted business income in the hands of the corporate appellants which they
had knowingly, or under circumstances amounting to gross negligence, failed to
include in computing their income for that year.
[29]
The
Minister also added to Langlois’s and Faraggi’s income the dividends received during their 1987 and
1988 taxation years, grossed up by one third (the total dividend so assessed
was $8,114,350 in 1987 and $155,912 in 1988 in Langlois’s case, and $8,034,166
in 1987 and $155,912 in 1988 in Faraggi’s case). Penalties were also assessed.
[30]
According
to the Minister, the sale by the subsidiaries of the gain-making shares,
despite all appearances, did not give rise to capital gains. In this respect,
the transactions were shams. It follows that the subsidiaries had no CDAs and
could not pay capital dividends to the corporate appellants. The corporate
appellants could not in turn pay dividends out of their CDAs to the devisers,
since they had no CDAs.
[31]
The
appellants objected to the assessments, which were eventually confirmed by the
TCC judge’s decision dismissing the four appeals.
TCC DECISION
[32]
The
TCC judge began his analysis by quoting the expanded definition of the word
“business” in subsection 248(1) of the Act. He then stated the following (para.
79):
It is not any kind of activity or undertaking that may be
considered a business; there must be some commercial quality to the activity or
undertaking for it to qualify as a business. The “blueprint” for the eventual
“transfer” of CDA to third parties and for the payment of dividends to the
individual appellants, the solicitation, directly or indirectly, of persons who
could benefit from the “transfer” of CDA and other actions by the appellants
were all in the nature of a commercial enterprise, no different from the
development of a product, the manufacture of that product and the sale of that
product by a person carrying on a business. . . .
[33]
He
continued as follows (idem):
. . . That the intent of the parties was cloaked
in purported agreements, in the issuance of shares, in the declaration of
dividends and deemed dividends, in capital gains and in subsection 83(2) elections
does not change what the appellants intended and what they actually did. The
corporate appellants carried on a business of creating dividends, which they
advertised as dividends from their capital dividend accounts, and these
accounts, if nothing else, they transferred in reality to third parties for a
profit. The profits, or parts thereof, were then distributed to the individual
appellants as dividends that are to be included in the incomes of the
individual appellants as assessed. These dividends were not paid out of any
corporation’s CDA. There was no amount that was eligible for election under
subsection 83(2) of the Act.
[34]
The
TCC judge then cited the definition of the term “sham” as set out by
Lord Diplock in Snook v. London & West Riding Investments, Ltd.,
[1967] 1 All E.R. 518 at page 528, and certain other Supreme Court
decisions that have given effect to the concept of “sham” in Canadian law. [The
Minister in his assessments and the TCC judge in his decision used the
expression “trompe‑l’œil” in French to express the concept of “sham”. The
words “artifice”, “faux‑semblant”, “simulacre” and “frime” are
also used in Canadian case law. In my opinion, the word “frime” is most
appropriate, considering its primary meaning, which denotes a conduct that is “volontairement
trompeur . . . en apparence seulement (cf. Pour la galerie)”
[“intentionally deceitful . . . in appearance only (see for
appearances’ sake)”], Le Nouveau Petit Robert, new edition, at page 974.
Moreover, this word was used by the Supreme Court in its first decision dealing
with the concept of “sham”, as set out in Snook, supra (Minister of
National Revenue v. Cameron, [1974] S.C.R. 1062). The French word “frime”
is henceforth used to express the notion of “sham”.] After completing this
review, the TCC judge concluded as follows (paragraph 86):
For a sham to exist, the taxpayers must have acted in such
a way as to deceive the tax authority as to their real legal relationships. The
taxpayer creates an appearance that does not conform to the reality of the
situation.
[35]
The
TCC judge then referred to the decision of the Supreme Court in Canada v.
Antosko, [1994] 2 S.C.R. 312, in particular to Iacobucci J.’s
statement that (page 328)
.
. . In the absence of evidence that the transaction was a sham or an abuse
of the provisions of the Act, it is not the role of the court to
determine whether the transaction in question is one which renders the taxpayer
deserving of a deduction . . . .
[My emphasis.]
According to the TCC judge, this passage would
indicate that “sham” and “abuse of the provisions of the Act” are similar
concepts (Reasons, paragraph 87, note 34).
[36]
The
TCC judge then made the following finding (paragraph 87):
In the appeals at bar, the essential components of the
transactions entered into by the appellants possess the basic elements of sham.
There was an abuse of the provisions of the Act. The initial purported
loan of $10,000,100, the declaration of stock and ordinary dividends, the
corresponding exchange of promissory notes and the capital gains and losses
cloaked an exercise undertaken by the appellants in concert to gain income from
a series of paper transactions. [Citation omitted.]
[37]
In
addition to this finding, which seems to be directed against the plan as a
whole, the TCC judge concluded that the loan granted by the Royal Bank was a
sham (Reasons, paragraph 90). According to the TCC judge, the loan in question
was a loan for consumption and, at the time, the Civil Code of Lower Canada
provided that the borrower became owner of the thing lent (Reasons, paragraphs
88 and 89). According to his analysis, the borrowers did not have “the absolute
enjoyment” of the loaned funds (Reasons, paragraph 90). As well, no interest
was paid (idem).
[38]
The
TCC judge also concluded that the promissory notes, issued by the first
subsidiary in the course of the first series of transactions to acquire the
gain-making shares from the eleven other subsidiaries, were shams. According to
him, those promissory notes did not create any real obligations because they
were never meant to be honoured. Only the nominal sum of $100 was available to
the first subsidiary to fund the payment. The TCC judge concluded from this that
the eleven subsidiaries had no intention of collecting on the notes (Reasons,
paragraphs 91 and 92).
[39]
With
respect to the penalties, the TCC judge found the following (Reasons, paragraph
96):
. . . The appellants knowingly carried out and promoted a
series of transactions they knew were artificial and were an abuse of
provisions of the Act. Yet they prepared and filed the tax returns in issue, or
caused these tax returns to be prepared, knowing full well that the information
contained in the returns for 1987 for all the appellants and the returns for
1988 for the individual appellants contained false statements or omissions.
The corporate appellants knew that they were carrying on a business and their
profits were camouflaged as dividends out of CDAs. And the individual
appellants knew the dividends they received from the corporate appellants were
taxable dividends. . . .
ALLEGED ERRORS IN THE DECISION UNDER APPEAL
[40]
The
four appellants allege that the TCC judge relied on non-existent rules to
support his finding as to the existence of a sham. They submit that, at the
time when the transactions at issue took place, neither the General
Anti-Avoidance Rule (GAAR) nor the rule now provided for in subsection 83(2.1)
of the Act was in force (this last provision, which the appellants say was
enacted in reaction to their plan, provides that as of September 25, 1987, a
dividend subject to the election provided for in subsection 83(2) is not a
capital dividend if its payment is part of a series of transactions having the
purpose of receiving a capital dividend).
[41]
In
their respective memoranda, the appellants all refer to the following excerpts from
the decision under appeal to show that the TCC judge was guided by the GAAR and
the concept of “abuse”:
[80] .
. . The manner in which they wished to achieve
their goal was not consistent with the object, spirit or purpose of, for
example, section 89, subsections 52(3), 83(2) and 84(1) of the Act.
. . .
[87] In the appeals at bar, the essential
components of the transactions entered into by the appellants possess the basic
elements of sham. There was an abuse of the provisions of the Act.
The initial purported loan of $10,000,100, the declaration of stock and
ordinary dividends, the corresponding exchange of promissory notes and the
capital gains and losses cloaked an exercise undertaken by the appellants in
concert to gain income from a series of paper transactions.
. . .
[93] . . . Finally, it is clear that specific
provisions of the Act were abused contrary to their object and spirit.
[Emphasis by the appellants.]
[42]
The
appellants contend that the evidence does not allow for the conclusion that
there was a sham based on the recognized application of this doctrine. They
point out that the legal documentation reflects the parties’ true intentions
and the true effects of the transactions carried out between them. The share
subscriptions took place, actual dividends were paid out, elections were made,
the amounts that had to be paid were paid and the documentation reveals exactly
what occurred. The parties to the transactions had no intention whatsoever of
giving the appearance of a “sham transaction” to cover up a “real transaction”
(Snook, supra, page 545).
[43]
According
to the appellants, the TCC judge’s analysis as a whole was flawed by his
GAAR-based approach and the application of obsolete rules such as “economic reality”,
thus extending the concept of “sham” beyond that which was applicable in 1987,
when the transactions took place. In this regard, they cite several articles
published in various tax journals that sharply criticize the TCC judge’s decision
and in particular his application of the sham doctrine (Timothé Huot, “Sham –
As Bad As It Gets,” Tax Topics, Toronto: CCH, Number
1857, October 11, 2007; Thomas E. McDonnel, “CDA Tax Scheme Voided – Sham
Doctrine Expanded?” Tax for the Owner-Manager, Canadian Tax Foundation,
Volume 7, Number 4, October 2007; Laura Stoddard, “Stretching the Sham
Doctrine?” Canadian Tax Journal (2007), Volume 55, Number 4, pages
850-853).
[44]
As
for the specific finding that the daylight loan was a sham, the appellants contend
that the cashed certified cheques undoubtedly amounted to a valid payment.
According to the appellants, the TCC judge erred in finding that the bank
overdraft granted by the Royal Bank was a sham.
[45]
The
appellants add that, contrary to the TCC judge’s finding, none of the
promissory notes were cancelled without consideration. The promissory notes
were cancelled either following payment by cheque or, wherever two parties held
debts of equivalent amounts against one another, by way of compensation. They
add that, contrary to what the TCC judge appears to understand, the promissory
notes were not intended to serve as a means of payment.
[46]
The
appellants raise the alternative argument that even if there was a sham so that
no capital gain was realized, the assessments are nonetheless invalid. In this
respect, the corporate appellants and the devisers advance different arguments.
[47]
The
appellant corporations submit for their part that, capital gain or not, the
Minister could not treat the premium paid by the third-party corporations when
subscribing to the shares as business income. According to the appellant
corporations, it is well established that share subscriptions are not
commercial transactions, no matter the goal sought by the corporation in
issuing the shares. (The decisions of the Supreme Court in Irrigation
Industries Ltd. v. The Minister of National Revenue, [1962] S.C.R. 346
(paragraph 21); of the TCC in Molstad Development Co. Ltd. v. Her Majesty
the Queen, 97 D.T.C. 913; and of this Court in Queenswood Land Associated Ltd.
v. Her Majesty the Queen, 2000 D.T.C. 6065 (paragraph 30) are cited in support of this proposition.)
The appellant corporations add that within [translation] “well-accepted business principles” and [translation] “generally accepted accounting principles”
there is no rule that suggests or allows for the proposition that a corporation
can make a [translation] “profit” from a sum
received as a result of a share subscription.
[48]
On
the other hand, Langlois and Faraggi emphasize the fact that Parliament has,
under Part III of the Act, specifically provided for the case of a corporation
declaring a dividend when the amount in its CDA is not sufficient to cover the
amount of the dividend. In those circumstances, subsection 184(2) provides that
the corporation shall pay a special tax equal to three quarters of the
shortfall, in which case the excessive dividend remains tax-free in the hands
of the shareholders. According to Langlois and Faraggi, the facts at issue in
this case dictate that Part III should have been applied.
[49]
Langlois
and Faraggi also argue that subsection 83(2) is a mandatory provision according
to which a dividend is “deemed to be a capital dividend”. According to the
devisers, this provision has the effect of treating the deemed dividend
received by the first subsidiary as a capital dividend, regardless of whether
the subsidiaries that paid this dividend generated capital gains.
[50]
In
the same vein, subparagraph 89(1)(b)(ii) provides that a corporation
that receives a dividend further to an election made under subsection 83(2)
“shall” include the amount thus received in its CDA. It follows that even if
the subsidiaries had no CDAs to begin with, the first subsidiary had to include
the dividends received in its CDA. The same reasoning applies to the subsequent
dividends until the ultimate distribution.
[51]
Lastly,
Langlois and Faraggi insist on the fact that the subsidiaries, which generated
capital gains giving rise to CDAs, were never reassessed. In the absence of
assessments nullifying the gains duly declared by the subsidiaries, the
Minister was required to respect the tax implications resulting from the elections
made by the subsidiaries.
Position of the respondent
[52]
Counsel
for the respondent asks that the four appeals be dismissed. He relies entirely
on the reasons given by the TCC judge.
ANALYSIS AND DECISION
[53]
In
their respective memoranda, the parties did not address the standard of review.
Counsel acknowledged during the hearing that questions of law are subject to
the standard of correctness, while questions of fact and of mixed fact and law may
not be reviewed in the absence of palpable error (Housen v. Nikolaisen,
[2002] 2 S.R.C. 235).
The concept of “sham” in
Canadian law
[54]
The
main attack directed against the decision of the TCC judge is that he unduly
stretched the concept of “sham”. In that respect, the appellants are correct in
saying that the state of the law in 1987 did not allow the TCC judge to ignore the
transactions or disregard their effects on the sole ground that they give rise to
an abuse. Only the advent of the GAAR and its invocation in a particular case
allow the Minister to repudiate a transaction on the sole ground that it gives
rise to an abuse of the Act or some of its provisions.
[55]
The
concepts of “sham” and “abuse” are not the same. I do not believe that the few
words of Iacobucci J. in Antosko, supra, cited by the TCC judge
(Reasons, para. 87, note 34), were intended to alter this view. Nowhere in the extensive
case law dealing with the concept of “sham” is it suggested that “sham” and “abuse”
are analogous. Iacobucci J.’s brief comment, which was part of a discussion on
the principles of statutory interpretation, cannot be read as bringing about such
a radical change.
[56]
Subject
to the invocation of the GAAR in a particular case, taxpayers are entitled to
arrange their affairs in such a way as to minimize their tax burden, even if in
doing so, they resort to elaborate plans that give rise to results which
Parliament did not anticipate. This is, I think, what the Supreme Court pointed
out in Shell Canada Ltd. v. Canada, [1999] 3 S.C.R. 622, when it said the
following (paragraph 45):
However, this
Court has made it clear in more recent decisions that, absent a specific
provision to the contrary, it is not the courts' role to prevent taxpayers from
relying on the sophisticated structure of their transactions, arranged in such
a way that the particular provisions of the Act are met, on the basis that it
would be inequitable to those taxpayers who have not chosen to structure their
transactions that way. This issue was specifically addressed by this Court in Duha
Printers (Western) Ltd. v. Canada, [1998] 1 S.C.R. 795,
at para. 88, per Iacobucci J. See also Neuman v. M.N.R., 1998] 1 S.C.R.
770, at para. 63, per Iacobucci J. The courts' role is to interpret and
apply the Act as it was adopted by Parliament. Obiter statements in
earlier cases that might be said to support a broader and less certain
interpretive principle have therefore been overtaken by our developing tax
jurisprudence. Unless the Act provides otherwise, a taxpayer is entitled to be
taxed based on what it actually did, not based on what it could have done, and
certainly not based on what a less sophisticated taxpayer might have done.
[57]
However,
courts have always felt authorized to intervene when confronted with what can
properly be labelled as a sham. The classic
definition of “sham” is that formulated by Lord Diplock in Snook, supra,
and reiterated by the Supreme Court on a number of occasions since. In Stubart
Investments Ltd. v. The Queen, [1984] 1 S.C.R. 536, Estey J. said the following (page
545):
. . . This expression
comes to us from decisions in the United Kingdom, and it has been generally
taken to mean (but not without ambiguity) a transaction conducted with an
element of deceit so as to create an illusion calculated to lead the tax
collector away from the taxpayer or the true nature of the transaction; or,
simple deception whereby the taxpayer creates a facade of reality quite
different from the disguised reality.
This passage
is also quoted with approval in Continental Bank Leasing Corp. v. Canada, [1998] 2 S.C.R.
298, at paragraph 20.
[58]
In Cameron, supra, the Supreme
Court adopted the following passage from Snook, supra, to define “sham”
in Canadian law (page 1068):
. . . [I]t means acts
done or documents executed by the parties to the "sham" which are
intended by them to give to third parties or to the court the appearance of
creating between the parties legal rights and obligations different from the
actual legal rights and obligations (if any) which the parties intend to
create.
The same
excerpt was quoted by Estey J. in Stubart, supra, at page 572.
[59]
It follows from the above definitions that the
existence of a sham under Canadian law requires an element of deceit which generally
manifests itself by a misrepresentation by the parties of the actual
transaction taking place between them. When confronted with this situation,
courts will consider the real transaction and disregard the one that was
represented as being the real one.
Corporate appellants
[60]
The
question in issue, insofar as the corporate appellants are concerned, is
limited to whether the premiums which they extracted from the third-party
corporations as part of the subscription share price can properly be included
in their income. I do not believe that it was necessary to resort to the
concept of “sham” to dispose of this issue.
[61]
The
gist of the TCC judge’s reasoning in confirming the assessments issued against
the corporate appellants can be found at paragraph 69 of his reasons:
. . . What the corporate appellants did was
to enter into a business, at minimum a venture in the nature of trade, and the
difference between what the arm's length third party corporations paid for
preferred shares and the amounts for which the shares were redeemed was
business income to the corporate appellants. . . .
[62]
The
existence of a business under the Act is not subject to any formal requirements
(reference is made to the extensive definition of the word “business” in
subsection 248(1) of the Act). The sole prerequisite is the pursuit of
profit (see Vern Krishna, The
Fundamentals of Canadian Income Tax, 8th ed., page 273). In
this case, under the devisers’ plan, the third-party corporations were required
to pay the corporate appellants a surcharge for the shares which they
subscribed to in order to have access to the CDAs. The third-party corporations
were required to pay $1,210 per share even though it was understood that the
shares would be redeemed for $1,000 (Reasons, paragraph 3). The difference
between the share redemption value and the price paid by the third-party
corporations is described as a [translation] “premium” in the
legal opinion obtained by Langlois and Faraggi when they were implementing
their plan. The word “premium” is here used in its
commercial sense rather than its corporate sense to refer to the additional
negotiated amount over the subscription price.
[63]
The
premiums paid by third-party corporations allowed the corporate appellants to
accumulate surpluses of $8,105,344 and $4,677,717 respectively. In my view, the
fact that these premiums were added to the share subscription price instead of
being charged separately has no effect on the TCC judge’s finding of fact that
these premiums were amounts paid by the third-party corporations to the
corporate appellants for access to their CDAs.The TCC judge’s finding on this
point is strictly based on the application of the Act to the facts found in the
course of his analysis.
[64]
In
this respect, the evidence shows unequivocally that it was agreed at the time
of subscription that the shares had a redemption value of $1,000 and that they would
be redeemed at the price of $0.01, following the payment of a cash dividend of
$999.99. It follows that according to the plan implemented by the devisers,
$1,000 was the maximum value the preferred shares could have had at any time in
the hands of the third-party corporations. The premium was therefore paid for
something other than the shares. This was obviously access to the CDAs and in
particular the corporate appellants’ undertaking to make the required elections
pursuant to subsection 83(2) in order to give the third-party corporations
access to their CDAs (in this regard, see paragraph 3 of the legal opinion
obtained by the devisers (short form)). This is not seriously being called into
question, as even the documents used by the devisers to market their plan
describe the premium as “the price of the CDA” (Appeal Book (A-300-07), Volume
V, page 1020).
[65]
The
alternative argument that a share subscription must always be treated on
account of capital regardless of the circumstances is without merit. The case
law cited by the corporate appellants in that regard (paragraph 47, supra)
does not have the effect attributed to it. Under the Act, no transaction or
operation is systematically excluded from the concept of business. The question
whether a given operation amounts to a business must be determined in
accordance with the particular facts of each case.
[66]
In
the present case, as additional amounts collected by the corporate appellants
were generated in the course of successive operations the purpose of which was
to produce surpluses, all the factors underlying the existence of a business
are present. In my opinion, the TCC judge correctly concluded that the
corporate appellants were engaged in a business and that these additional
amounts constituted business income.
Langlois and Faraggi
[67]
The bulk of the dividends received by the
devisers was paid during the implementation stage of the plan, as and when
dividends were paid to the third-party corporations. It follows, according to
Langlois and Faraggi, that like the dividends paid to the third-party
corporations those that they received were paid out of a CDA and were, to that
extent, capital dividends. However, the TCC judge concluded that the operations
leading to the payment of the dividends were a sham. He noted that, contrary to
what was represented, no capital gains were generated and no CDAs were created,
with the result that the dividends received by Langlois and Faraggi were
ordinary dividends and therefore taxable.
[68]
The parties agree that only the existence of a
sham can justify the conclusion reached by the TTC judge. The dispute centres
on the scope of this concept and its application to the facts at hand.
[69]
As indicated earlier, the mere fact that the
transactions that led to the payment of these dividends may have resulted in an
abuse of the Act is not, in itself, enough to conclude that they are shams. Moreover,
if this concept is applied in conformity with the case law which, as we have
seen, requires an element of deceit, it is difficult to see how the TCC judge
could find that the daylight overdraft granted by the Royal Bank for a short
time on August 13, 1987, was a sham (Reasons, paragraph 90). It is true
that, as the TCC judge pointed out, the funds could not have been used for any
purpose other than for the particular purpose for which they were advanced, and
that, in that sense, the borrowers did not have “absolute enjoyment” of the
funds. But the fact that a loan is extended for a particular purpose is not
exceptional. In that regard, the evidence reveals that the funds were used for
the purpose for which the advance was authorized and that each of the
$10,000,000 cheques certified by the Royal Bank guaranteed the payment of the stated
value (Testimony of Alain Lapointe, Appeal Book (A-300-07), pages 95 and 96
(1916 and 1917)).
[70]
Moreover, the fact that the cost of the loan
was paid by way of a service charge rather than interest is also no reason to
conclude that the loan was bogus. Either way, the cost of the loan was the same
($10,000 was charged as a service fee, while the interest payable on a daylight overdraft
would have been $10,000, that is, a tenth of one per cent (idem, pages
98 et 99 (1937 and 1938)). With respect, it
is not possible to conclude that the Royal Bank, the devisers and the first
subsidiary misrepresented their relationship of lender/borrower. The loan
cannot be held to be a sham.
[71]
Similarly, the promissory notes issued by the
first subsidiary ($110,000,000 in total) cannot be qualified as shams (Reasons,
paragraphs 91 and 92). The TCC judge made the point that first subsidiary did
not have the funds to honour the notes. However, according to the evidence, it
was not contemplated that the promissory notes would be used as a means of
payment. Their only function was to attest to the existence of the debt
incurred by the first subsidiary following the purchase of the gain-making shares
(Appeal Book (A-300-07), Volume II, “Statement of Facts”, paragraphs 125 to
130, 140 to 145, 155 to 160, 170 to 175, 185 to 190, 200 to 205, 215 to 220,
230 to 235, and 245 to 250). The fact that the devisers had planned for the
debts to be discharged by another mode of payment does not make the promissory
notes shams. In fact, the record reveals that all the debts recognized by the
promissory notes were honoured either by payment or by compensation, a
legitimate payment method in Quebec civil law (Appeal Book (A-300-07), Volume V, “Book of
Account”, pages 908 to 923).
[72]
However, I am of the opinion that the
transactions that generated the alleged capital gains amounting to a total of
some $234 million were misrepresented and that the resulting elections
made by the subsidiaries and the corporate appellants were shams in the accepted
meaning of the word. A sale of shares does not result in a capital gain simply
because one decides to call it that. It is the nature of the property in the
hands of the person who disposes of it that determines the tax treatment of the
transaction. As a general rule, property that is capital in nature is property
owned for the long term with a view to its appreciation in value or in order to
earn income from it, such as rent, interest or, in the case of shares,
dividends. At the other extreme, property acquired for resale is held on account
of revenue. Although the characterization of property for tax purposes is
difficult when one moves away from these extremes, this becomes obvious when
the situation falls at one of these extremes.
[73]
In the case at hand, the gain-making shares
were acquired by the subsidiaries for the purpose of their immediate sale. The
terms of the plan allow for no other scenario. This excludes the possibility
that the shares could have been capital property in the hands of the
subsidiaries. As counsel for the respondent pointed out during the proceeding,
the fact that the subsidiaries acquired the gain-making shares through stock dividends
is of no assistance since each of the steps, including the mode of acquisition
of these shares, was decided in advance by the devisers of the plan.
[74]
This is the missing link in the devisers’
strategy. It is not surprising that they did not see fit to ask Mr. Régnier to opine on the manner in which the CDAs were
created. This aspect of the plan was specifically excluded from Mr. Régnier’s
mandate (Testimony of Mr. Régnier, Appeal Book
(A-300-07), Volume X, page 128 (1949), lines 5 and 6). The simple fact that the
acquisition of the gain-making shares by the subsidiaries and their sale had
been pre-ordained eliminates the possibility that a favourable opinion could have
been obtained. When confronted with this during the hearing, counsel for the
appellants was unable to explain how the gain-making shares could be viewed as
capital property in the hands of the subsidiaries.
[75]
The evidence reveals that this issue was raised
during the marketing phase of the plan by Mr. Tom Gillespie, an experienced tax practitioner with the law
firm Ogilvie Renault.
Mr. Gillespie insisted on knowing the source of the capital gains
before issuing a favourable opinion for the benefit of clients interested in
the scheme. No answer was given. Mr. Gillespie was informed that he did not need to know the source
of the CDAs, as this had no bearing on the opinion he was to provide. The
clients in question who were considering a major investment ($50,000,000)
decided not to participate (Testimony of Ralph E. Faraggi, Appeal Book (A‑300‑07),
Volume VII, pages 54 to 57, and 62 (1322 to 1324, and 1329)). The limited scope
of the mandate given to Mr. Régnier and the position taken in response to Mr.
Gillespie’s request tend to suggest that the devisers were aware of the flaw in
their plan.
[76]
Despite the fact that no capital gains were generated,
each of the second to twelfth subsidiaries took the position that it had
generated such gains. The prescribed form and schedule filed with the Minister following
the payment of the deemed dividend to the first subsidiary (i.e., the
elections) set out the computation of the CDA and its source. According to the
prescribed information submitted by the subsidiaries, the disposition of the
gain-making shares had generated capital gains, and these gains were the source
of their CDAs (see, for example, form T-2054 and the schedules filed by
2529-0099 Québec Inc. (the second subsidiary), Appeal Book (A-300-07), Volume
V, pages 841 to 844).
[77]
In making these elections, the subsidiaries misrepresented
to the Minister and to all those affected by these elections that the disposition
of the gain-making shares had resulted in capital gains, half of which formed
part of their CDAs. They further misrepresented that the deemed dividends paid
to the first subsidiary came from these CDAs when they knew that no CDA had
been created in the first place. The first subsidiary, in turn, misrepresented
the facts when it indicated in its election that the dividend declared in
favour of 1915 Inc. came from its CDA, even though it knew that no CDA had been
transferred to it by the subsidiaries. Finally, 1915 Inc. misrepresented the
facts when it indicated in its election that the cash dividends paid to the
devisers came from its CDA, when it knew that no CDA had been transferred to it
by the first subsidiary.
[78]
Despite the impression given, no capital gains
were made and no CDAs were created. In this regard, I agree with the TCC
judge’s statement at paragraph 94 of his reasons:
… To have made an election pursuant to
subsection 83(2) in these circumstances is, to put it mildly, dishonest …
[79]
The initial elections were shams since they misrepresented
the transactions that had taken place between the subsidiaries as being capital
transactions. The subsequent elections were also shams since they represented
that dividends were paid out of a CDA when none had been created and none could
thereafter be transferred. In my opinion, the TCC judge correctly relied on the
real transactions, which produced no capital gains, to conclude that the
dividends ultimately received by the devisers were not capital dividends and were
therefore taxable.
[80]
The devisers’ alternative argument that the
dividends remained capital dividends even though the subsidiaries generated no
capital gains must also fail. It is true that subsection 83(2) provides that
a dividend that is made the subject of an election is deemed to come from the
CDA of the payer corporation. But, according to its wording, this presumption
is effective only “to the extent of” the payer corporation’s CDA. When the payer
corporation does not have a CDA, as was the case of the eleven subsidiaries
that made the first election, the presumption has no effect.
[81]
The devisers’ additional argument that
subparagraph 89(1)(b)(ii) has the effect of creating valid CDAs in
the hands of transferees even where their initial creation was a sham must also
be rejected. The devisers as the authors of the sham cannot rely on subsequent
events to validate the CDAs from which they received the dividends.
[82]
I also do not believe that the Minister was bound
to proceed pursuant to Part III on the facts of this case. The provisions of
Part III deal with excessive elections. They allow corporations that pay
dividends exceeding the value of their CDAs to correct their mistake and avoid
the special tax provided for under Part III. Like the TCC judge, I do not
believe that Parliament intended these provisions to apply when, as here, the
initial CDA was sham, and those claiming the benefit of Part III are the
authors of the sham.
[83]
Lastly, the fact that the Minister has not to
date assessed the subsidiaries to offset the capital gains they claimed to have
generated does not preclude the assessments issued against the appellants.
PENALTIES
[84]
Counsel for the appellants did not dwell on the
TCC judge’s decision regarding the penalties. He explained during the
proceeding that his clients’ fate turns on their liability for the assessed tax
(and interest) and that the offsetting of the penalties, with nothing more,
would be of little use.
[85]
The TCC judge found that the four appellants were
grossly negligent in failing to report their respective incomes/dividends. It
has not been established that he erred in coming to this conclusion.
[86]
I would therefore dismiss the four appeals with
costs.
“Marc Noël”
“I agree.
M. Nadon
J.A.”
“I agree.
Johanne
Trudel J.A.”
Certified
true translation
Susan
Deichert, Reviser
Email this Content