ROGERS ENTERPRISES (2015) INC. (SUCCESSOR BY AMALGAMATION TO CGESR LIMITED), Appellant, v. HER MAJESTY THE QUEEN,, 2020 TCC 92

Docket: 2017-1140(IT)G BETWEEN: ROGERS ENTERPRISES

(2015) INC. (SUCCESSOR BY AMALGAMATION TO CGESR LIMITED), Appellant, and HER MAJESTY THE QUEEN, Respondent , Docket: 2017-3617(IT)G BETWEEN: ROGERS ENTERPRISES

(2015) INC. (SUCCESSOR BY AMALGAMATION TO ESRIL

(1998) LIMITED), Appellant, and HER MAJESTY THE QUEEN, Respondent. Appeals heard on February 25 and 26, 2019, at Toronto, Ontario By: The Honourable Justice Don R. Sommerfeldt Appearances : Counsel for the Appellant: Pooja Mihailovich, Hemant Tilak Counsel for the Respondent: Justine Malone, Marie-Eve Aubry JUDGMENT The Appeals are allowed, with costs, and the Determinations that are the subject of the Notices of Determination dated August 21, 2015 are vacated. The Appellants are entitled to costs.

The Parties shall have 30 days from the date of this Judgment to reach an agreement on costs and to so advise the Court, failing which the Appellants shall have a further 30 days to file written submissions on costs, and the Respondent shall have yet a further 30 days to file a written response. Any such submissions shall be limited to five pages in length. If, within the applicable time limits, the Parties do not advise the Court that they have reached an agreement and no submissions are received from the Parties, costs shall be awarded to the Appellants in accordance with the Tariff.

Signed at Vancouver, British Columbia, this 27th day of August 2020. “Don R. Sommerfeldt” Sommerfeldt J. Citation: 2020 TCC 92 Date:2020 09 17 Docket: 2017-1140(IT)G BETWEEN:

ROGERS ENTERPRISES

(2015) INC. (SUCCESSOR BY AMALGAMATION TO CGESR LIMITED), Appellant, and HER MAJESTY THE QUEEN, Respondent, Docket: 2017-3617(IT)G BETWEEN: ROGERS ENTERPRISES

(2015) INC. (SUCCESSOR BY AMALGAMATION TO ESRIL

(1998) LIMITED), Appellant, and HER MAJESTY THE QUEEN, Respondent. AMENDED REASONS FOR JUDGMENT Sommerfeldt J. I. INTRODUCTION [ 1 ] These Reasons pertain to the Appeals brought by Rogers Enterprises

(2015) Inc. ( " “RE 2015” " ), as successor by amalgamation to CGESR Limited ( " “CGESR” " ) and as successor by amalgamation to ESRIL

(1998) Limited ( " “ESRIL 98” " ), in respect of determinations (the " “Determinations” " ) made pursuant to subsection 152(1.11) of the Income Tax Act (the " “ ITA ” " ), [1] as set out in Notices of Determination dated August 21, 2015 and issued by the Canada Revenue Agency (the " “CRA” " ), on behalf of the Minister of National Revenue (the " “Minister” " ), for the taxation year of CGESR ended September 30, 2010 and the taxation year of ESRIL 98 ended December 31, 2009. II. FACTS [ 2 ] The Parties submitted a Statement of Agreed Facts (the " “SAF” " ), [2] a copy of which is attached as Appendix A. For the convenience of the reader, a concise

summary of the material facts is set out in the following paragraphs. [ 3 ] Edward Samuel (Ted) Rogers was the President and the Chief Executive Officer of Rogers Communications Inc. ( " “RCI” " ), a Canadian public corporation. Several corporations and trusts (collectively, the " “RPC Group” " ) [3] held the interests of the Rogers family in RCI. Members of the RPC Group that are relevant for the purpose of these Appeals were ESR Limited ( " “ESRL” " ), ESRIL 98, CGESR, CGESR

(2009) Limited ( " “CGESR 2009” " ) and Rogers Telecommunications Limited ( " “RTL” " ). ESRIL 98 and CGESR were taxable Canadian corporations, private corporations and Canadian-controlled private corporations ( " “CCPCs” " ). According to paragraph 8 of each Notice of Appeal, ESRL, CGESR 2009 and RTL were also CCPCs. [ 4 ] The September 5, 1984 Rogers Ownership Trust (the " “1984 Trust” " ) and the August 24, 1995 Rogers Ownership Trust (the " “1995 Trust” " ) were family trusts established to hold property for the benefit of Loretta Rogers and the descendants of Mr. Rogers.

In addition, the 1984 Trust held property for the benefit of Mr. Rogers. [ 5 ] From 1981 to 1991, twelve life insurance policies (the " “Policies” " ) insuring the life of Mr. Rogers were issued to various private corporations in the RPC Group or to the 1984 Trust. At the time that each Policy was issued, the policyholder (or owner) of each Policy was also the beneficiary of the Policy.

It is my understanding that the policyholder of each Policy was required to pay the premiums in respect of that Policy. [ 6 ] At a time which is not specified in the SAF but which was before February 25, 2005, the holders of the various Policies changed the beneficiaries of the Policies, in each case designating RTL as the beneficiary. [ 7 ] In 2005, in conjunction with a significant corporate reorganization undertaken by the RPC Group, ESRL became the holder of ten of

the Policies. The other two Policies continued to be held by the 1984 Trust. [ 8 ] Sometime in 2005, the beneficiary of the twelve Policies was changed again, this time to be CGESR.

Notwithstanding the successive changes of beneficiary, ESRL and the 1984 Trust, as the holders of ten and two of the Policies respectively, continued to pay the premiums in respect of their Policies. [ 9 ] In the course of a subsequent reorganization in 2006, CGESR became the direct shareholder of RTL. [ 10 ] By 2009, after CGESR 2009 had been incorporated, ESRL owned Class A and Class D shares in the capital of ESRIL 98, which owned Class A – Series 5 preferred shares and Class A – Series 6 preferred shares in the capital of CGESR, with the 1,000 common shares in the capital of CGESR being owned by CGESR 2009 and the other issued shares in the capital of CGESR being owned by other members of the RPC Group.

CGESR owned shares in the capital of RTL and also owned shares in the capital of RCI. The 1984 Trust and other shareholders which are not identified in the SAF held shares in the capital of ESRL. [ 11 ] Mr. Rogers died on December 2, 2008. On December 17, 2008, the proceeds of the Policies were paid to CGESR, as the designated beneficiary of the Policies. The total amount of the life insurance proceeds was $102,309,794, [4] of which $4,555,420 derived from the Policies held by the 1984 Trust and $97,754,374 derived from the Policies owned by ESRL.

As there was no adjusted cost basis ( " “ACB” " ) of the Policies to CGESR, the entire amount of $102,309,794 was added by CGESR to its capital dividend account ( " “CDA” " ). [ 12 ] Immediately before the death of Mr. Rogers, the total ACB to the 1984 Trust of the two Policies held by it was $1,755,055, and the total ACB to ESRL of the ten Policies held by it was $42,239,105. [5] [ 13 ] On July 15, 2009, CGESR paid dividends in the amount of $702,742 to ESRIL 98 and dividends in the amount of $10,000,000 to CGESR 2009.

CGESR elected under subsection 83(2) of the ITA in respect of the full amounts of the dividends that it paid on July 15, 2009. [ 14 ] On July 15, 2009, CGESR 2009 paid dividends in the amount of $9,999,950.50 to the 1995 Trust. CGESR 2009 elected under subsection 83(2) of the ITA in respect of the full amount of the dividends that it paid on July 15, 2009. [ 15 ] On October 27, 2009, CGESR redeemed some of the class A preferred shares in its capital that were owned by ESRIL 98, as a result of which ESRIL 98 was deemed to have received dividends in the amount of $91,745,839.

CGESR elected under subsection 83(2) of the ITA in respect of the full amount of the deemed dividends of $91,745,839. [6] [ 16 ] On the understanding that the dividends received, or deemed to have been received, by ESRIL 98 from CGESR were capital dividends, the full amount of those dividends was added by ESRIL 98 to its CDA. [ 17 ] According to the records of ESRIL 98, on October 27, 2009, after the above transactions, its CDA balance was $92,448,860. On October 27, 2009, ESRIL 98 paid capital dividends in the amount of $49,998,834 to ESRL.

After the payment of those dividends, the balance remaining in ESRIL 98’s CDA was, according to its records, $42,450,026. I was advised by counsel for RE 2015 that, although it is not so stated in the SAF, the balance of $42,450,026 still remains in ESRIL 98’s CDA.

In other words, there have been no capital dividends paid by ESRIL 98 after October 27, 2009. [7] [ 18 ] Sometime after October 27, 2009, by reason of one or more amalgamations that are not described in detail in the SAF, RE 2015 became the successor by amalgamation to both ESRIL 98 and CGESR. [ 19 ] On August 21, 2015, the Minister issued Notices of Determination to CGESR for its taxation year ended September 30, 2010 and to ESRIL 98 for its taxation year ended December 31, 2009. The Minister considered that, for the purposes of the general anti-avoidance rule (the " “GAAR” " ) in

section 245 of the ITA , the following transactions constituted a series of transactions (the " “Series” " ):

a) the acquisition of the Policies;

b) the designations by ESRL, as the policyholder of the ten Policies owned by it, to make first RTL and subsequently CGESR the beneficiary of those Policies;

c) the receipt by CGESR of the proceeds of the Policies and the interest paid thereon;

d) the payment by CGESR of dividends (actual or deemed) to ESRIL 98; and

e) the elections by CGESR under subsection 83(2) of the ITA , such that those dividends were deemed by paragraph 83(2)(

a) of the ITA to be capital dividends. [ 20 ] The Minister concluded that the Series included one or more avoidance transactions (the " “Subject Transactions” " ), as follows:

a) the designations by ESRL, as the policyholder of the Policies owned by it, to make CGESR the beneficiary of those Policies;

b) the receipt by CGESR of the proceeds of the Policies;

c) the payment by CGESR of dividends (actual or deemed) to ESRIL 98; and

d) the elections by CGESR under subsection 83(2) of the ITA in respect of the dividends (actual or deemed) paid to ESRIL 98. [ 21 ] Relying on subsection 245(2) of the ITA , the Minister determined what she considered to be the reasonable tax consequences to CGESR and ESRIL 98 in order to deny the tax benefits that had been identified by the Minister. In particular:

a) in respect of CGESR, the Minister determined that: i . to the extent of $42,239,100, the dividends paid by CGESR to ESRIL 98 were taxable dividends, rather than capital dividends; ii . CGESR’s CDA as at October 28, 2009 was nil; and

b) as a consequence, in respect of ESRIL 98, the Minister determined that: i . the dividends received by ESRIL 98 fromCGESR, to the extent of $42,239,100, were, pursuant to subsection 112(1) of the ITA , deductible in computing the taxable income of ESRIL 98; and ii. ESRIL 98’s CDA balance as at December 31, 2009 was reduced by $42,239,100. III. ISSUES [ 22 ] The issues in respect of these Appeals are:

a) Was there a tax benefit that, but for

section 245 of the ITA , resulted, directly or indirectly, from the Series?

b) If there was a tax benefit, may it reasonably be considered that any of the transactions in the Series, if the ITA were read without reference to

section 245, resulted directly or indirectly in a misuse of the provisions of the ITA , or resulted directly or indirectly in an abuse having regard to the provisions of the ITA , other than

section 245, read as a whole? IV. ADMISSIONS [ 23 ] For the purposes of these Appeals, RE 2015 admits that:

a) the transactions defined in paragraph 19 as the " “Series” " were a " “series of transactions” " for the purposes of

section 245 of the ITA ; and

b) each of the Subject Transactions was part of the Series. [8] [ 24 ] While RE 2015 takes the position that there was no tax benefit resulting from any of the Subject Transactions, if the Court finds that there was a tax benefit, RE 2015 admits that:

a) but for

section 245 of the ITA , one or more of the Subject Transactions resulted, directly or indirectly, in the tax benefit; and

b) each such Subject Transaction was an avoidance transaction within the meaning of subsection 245(3) of the ITA . [9] V. ANALYSIS A. General Principles [ 25 ] To analyze the Minister’s application of the GAAR, three steps are required:

a) The Court must determine whether there is a tax benefit arising from a transaction, as contemplated by subsections 245(1) and (2) of the ITA .

b) The Court must determine whether the transaction is an avoidance transaction, within the meaning of subsection 245(3) of the ITA .

c) The Court must determine whether the avoidance transaction would result directly or indirectly in a misuse of certain enumerated statutory or treaty provisions or would result directly or indirectly in an abuse, as more fully particularized in subsection 245(4) of the ITA . [10] B. Tax Benefit [ 26 ] Subsection 245(1) of the ITA defines the term " “tax benefit” " as follows: “tax benefit” means a reduction, avoidance or deferral of tax or other amount payable under this Act or an increase in a refund of tax or other amount under this Act …. [ 27 ] According to the Crown, the following were the tax benefits resulting from the Series:

a) the increase in the amounts of the respective CDAs of CGESR and ESRIL 98;

b) the reduction in the computation of income pursuant to subsection 83(2) of the ITA , by an amount which would otherwise have been included as taxable dividends for ESRIL 98 pursuant to sections 3 and 9, subsection 82(1) and paragraph 12(1)(

j) of the ITA ;

c) the avoidance of tax under

Part III of the ITA by ESRIL 98 and the reduction of tax under

Part I of the ITA by the ultimate shareholders of ESRIL 98; and

d) the avoidance of tax under

Part III of the ITA by CGESR. [11]

(1) CDA Increase [ 28 ] With respect to the first of the above alleged tax benefits, the Crown submits that the increase in the respective CDAs of CGESR and ESRIL 98 fell within the phrase " “an increase in … [an] other amount under this Act ,” " as contained in the statutory definition of " “tax benefit.” " The actual more complete phrase in the definition of " “tax benefit” " in subsection 245(1) is " “an increase in a refund of tax or other amount under this Act .” " There are two ways of interpreting the statutory provision. One

interpretation ( " “Interpretation A” " ) is: “tax benefit” means … an increase in[:] • a refund of tax[,] or • other amount under this Act …. The other

interpretation ( " “Interpretation B” " ) is: “tax benefit” means … an increase in a refund of[:] • tax[,] or • other amount under this Act …. [ 29 ] The French version of the ITA does not assist in clarifying the above ambiguity. In 2009, the French definition of " “tax benefit” " (i.e., avantage fiscal) read as follows: «avantage fiscal» Réduction, évitement ou report d’impôt ou d’un autre montant exigible en application de la présente loi ou augmentation d’un remboursement d’impôt ou d’un autre montant visé par la présente loi. As in English, the concluding portion of the French definition may be read in one of two ways. The first (i.e., the French equivalent of

Interpretation

A) is: «avantage fiscale» … augmentation[:] • d’un remboursement d’impôt[,] ou • d’un autre montant visé par la présente loi. The other way to interpret the French provision (i.e., the French equivalent of

Interpretation

B) is : «avantage fiscale» … augmentation d’un remboursement[:] • d’impôt[,] ou • d’un autre montant visé par la présente loi. [ 30 ] The Crown prefers

Interpretation A. The Crown submits that a tax benefit includes " “an increase in … [an] other amount under this Act .” " For the reasons set out below, I prefer

Interpretation B, i.e., the particular portion of the definition of " “tax benefit” " refers to an increase in a refund of tax or an increase in a refund of another amount under the ITA . [ 31 ] My first reason for preferring

Interpretation B is based on the wording of the relevant portion of the definition of " “tax benefit” " as shown in paragraph 28 above. If the Crown’s

interpretation were correct, one would expect that the word " “an” " would have appeared before the phrase " “other amount,” " or that the word " “another” " would have been used rather than the word " “other.” " [ 32 ] The second reason is based on the Explanatory Notes released by the Department of Finance in June 1988, when the GAAR was introduced. Those notes stated the following about the definition of " “tax benefit” " : Generally, for the purposes of

section 245, a transaction, to be an avoidance transaction, must result in a “tax benefit”. This expression is defined as a reduction, avoidance or deferral of tax or other amount payable under the Act or an increase in a refund of tax or other amount under the Act .

The references in this definition to “other amount payable under this Act ” and “other amount under this Act ” are intended to cover interest, penalties, the remittance of source deductions, and other amounts that do not constitute tax. [12] Like the French and English statutory provisions themselves, the Explanatory Notes do not clarify whether the term " “tax benefit” " is to be read as including an increase in a refund of tax or an increase in another amount under the ITA , or as including an increase in a refund of tax or an increase in a refund of another amount under the ITA .

However, if the ejusdem generis rule can be applied to the last sentence of the above quotation from the Explanatory Notes , [13] it would seem that the phrase " “other amount under this Act ” " is intended to refer to amounts that are payable or refundable under the ITA , such as interest, penalties and source deductions. [ 33 ] The third reason derives from Canada Trustco , in which the Supreme Court of Canada stated the following in respect of the meaning of " “tax benefit” " :

“Tax benefit” is defined in s. 245(1) as “a reduction, avoidance or deferral of tax” or “an increase in a refund of tax or other amount” paid under the Act . [14] [ Emphasis added .] The above statement is similar to the following statement made in 1988 by Brian Arnold and James Wilson: A “tax benefit” is defined very broadly in subsection 245(1) to be a reduction, avoidance, or deferral of tax or other amount payable (such as interest and estimated tax instalments) or an increase in a refund of tax or other amount payable under the Act . [15] [ Emphasis added .] The phrase " “paid under the Act ,” " as used by the Supreme Court, and the phrase “payable under the Act ,” as used by Arnold and Wilson, after the phrase " “an increase in a refund of tax or other amount,” " suggest that the words " “other amount” " refer to something that a particular taxpayer has paid, or is required to pay, under the ITA . [ 34 ] The fourth reason for preferring

Interpretation B, rather than

Interpretation A, is based on an analysis of the ITA . The ITA refers to many amounts, some of which are typically more advantageous to a taxpayer as the amounts increase, and others of which are typically more advantageous to a taxpayer as the amounts decrease, at least for the purposes of the ITA (but not necessarily for economic purposes).

For instance, an increase in the amount of a deduction, an increase in the amount of a loss or an increase in the amount of a tax attribute (such as the paid-up capital in respect of shares or the adjusted cost base to a taxpayer of a capital property) may well be viewed as a tax benefit. On the other hand, it would seem peculiar to consider that an increase in the amount of income or an increase in the amount of profit would constitute a tax benefit. Thus, it does not seem reasonable that increases in all amounts under the ITA are to be viewed as tax benefits. [16] [ 35 ] A fifth reason is based on an

article written by Robert Couzin, which focused on subsection 245(3) of the ITA (which defines " “avoidance transaction” " ). Couzin made the following comment about the definition of " “tax benefit” " in subsection 245(1) of the ITA : One reason to define “tax benefit” is to extend the scope of the concept beyond what might otherwise be comprehended. A tax benefit, undefined, would include a reduction in or avoidance of tax, but not necessarily the mere deferral of tax.

The extended definition is certainly needed to reach a reduction in interest or penalty, since these are not “tax.” While an increase in a refund may, in some circumstances, be the result of a reduction in tax, that is not always the case (for example, the increase in a refundable credit).

Thus, the definition is required to catch such additional refundable amounts . [17] [ Emphasis added .] The above statement seems to suggest that the phrase " “other amount under this Act ” " refers to an amount that could be refundable under the ITA . [ 36 ] On the other hand, one could argue that, because Parliament included the word " “payable” " in the first portion of the definition of " “tax benefit” " (when referring to " “a reduction, avoidance or deferral of tax or other amount payable under this Act ” " ), the failure to include the word " “refundable” " in the concluding portion of the definition must have some significance.

In other words, perhaps it is significant that the concluding portion does not read " “an increase in a refund of tax or other amount refundable under this Act .” " The counter argument, which I prefer, is that, because " “tax” " and " “other amount” " are both objects of the preposition " “of” " , which follows the word " “refund,” " it is not necessary to use both " “refund” " and " “refundable” " in the same phrase, i.e., " “refundable” " would be redundant. [ 37 ] For the reasons set out above, [18] the increases in the amounts of the respective CDAs of CGESR and ESRIL 98 were not tax benefits.

(2) Reduction in Computation of Income [ 38 ] The Crown submits that the reduction in the computation of income pursuant to subsection 83(2) of the ITA (i.e., the treatment of the actual or deemed dividends as capital dividends, rather than as taxable dividends) was a tax benefit.

In making this argument, the Crown is not suggesting that there was a reduction of tax, but merely a reduction in the computation of income. [ 39 ] In a GAAR context, the computation of income is relevant for the purposes of the definition of " “tax consequences,” " defined in subsection 245(1) of the ITA as follows: “tax consequences” to a person means the amount of income, taxable income, or taxable income earned in Canada of, tax or other amount payable by or refundable to the person under this Act , or any other amount that is relevant for the purposes of computing that amount…. [19] However, the concepts of income, taxable income and taxable income earned in Canada do not appear in the definition of " “tax benefit.” " To constitute a tax benefit, it is insufficient that there be a reduction in the computation of income.

Rather, in this context, a reduction of tax is required. Admittedly, in many situations a reduction in the computation of income will generally result in a reduction of tax.

However, in this particular situation, where capital dividends (actual or deemed) aggregating $92,448,581 were paid by CGESR to ESRIL 98 and subsequently a capital dividend aggregating $49,998,834 was paid by ESRIL 98 to ESRL, although those capital dividends were not included in computing the recipient corporation’s income, there was no less tax paid than if those dividends had been taxable dividends, which would have been included (pursuant to subsection 82(1) of the ITA ) in computing the income of the recipient corporation, but which would have been deductible (pursuant to subsection 112(1) of the ITA ) in computing the taxable income of the corporation.

In other words, whether the dividends in question were capital dividends or taxable dividends, there would have been no tax payable. Thus, while there was a reduction in income, there was no reduction in tax. Therefore, the reduction in the computation of income did not constitute a tax benefit.

(3) Avoidance of

Part III Tax by ESRIL 98 and Reduction of

Part I Tax by Its Ultimate Shareholders

(

a) Avoidance of

Part III Tax by ESRIL 98 [ 40 ] To perform the first of the three steps in a GAAR analysis, some form of measurement or comparison is typically required in order to ascertain whether there is a tax benefit. As Justice Bonner explained in McNichol : There is nothing mysterious about the subsection 245(1) concept of tax benefit. Clearly a reduction or avoidance of tax does require the identification in any given set of circumstances of a norm or standard against which reduction is to be measured.

Difficulties may exist in other cases in identifying the standard but in this case there is no such difficulty…. [20] [ 41 ] After discussing the " “tax benefit” " analysis undertaken by Justice Bonner in McNichol and by Justice Bowman (as he then was) in Equilease , [21] in the above-referenced

article Couzin observed: None of this eliminates the problem of finding a comparable transaction (or series of transactions) in order to judge whether the transaction (or series) actually undertaken resulted in a tax benefit. Judge Bonner admitted as much, as both he and Judge Bowman did in fact identify the comparable: a taxable dividend distribution. What is perhaps less clear is how one might formulate a general rule for finding such a comparable. In these particular cases, the evidence before the court, and the long history of surplus stripping in Canada, may have pointed the way.

But in other cases that path is likely to be less evident. In a sophisticated tax plan, based on arcane tax rules, judges may not find the normative transaction quite so obvious. [22] As will be noted below [23] , these Appeals by RE 2015 are among those envisioned by Couzin, i.e., a comparable normative transaction is not readily apparent. [ 42 ] In Canada Trustco , the Supreme Court of Canada stated the following in respect of the comparison of alternative arrangements: If a deduction against taxable income is claimed, the existence of a tax benefit is clear, since a deduction results in a reduction of tax.

In some other instances, it may be that the existence of a tax benefit can only be established by comparison with an alternative arrangement…. In such cases, the existence of a tax benefit might only be established upon a comparison between alternative arrangements.

In all cases, it must be determined whether the taxpayer reduced, avoided or deferred tax payable under the Act . [24] [ 43 ] In Copthorne , the Supreme Court of Canada reiterated and expanded upon the above point made in Canada Trustco , as follows: As found in Trustco , the existence of a tax benefit can be established by comparison of the taxpayer’s situation with an alternative arrangement…. If a comparison approach is used, the alternative arrangement must be one that “might reasonably have been carried out but for the existence of the tax benefit” (D. G. Duff, et al.

Canadian Income Tax Law (3 rd ed. 2009), at p. 187). By considering what a corporation would have done if it did not stand to gain from the tax benefit, this test attempts to isolate the effect of the tax benefit from the non-tax purpose of the taxpayer. [25] [ 44 ] As indicated above, on October 27, 2009, ESRIL 98 redeemed certain shares, made an election under subsection 83(2) of the ITA , and was deemed to have paid capital dividends in the amount of $49,998,834 to ESRL. At the time, ESRIL 98 understood that its CDA balance was $92,448,860.

According to the CRA, this balance was inflated to the extent of $42,239,105 (being the ACB to ESRIL of the ten Policies held by it). [26] If the CRA’s position is correct, the proper CDA balance would have been $50,209,755 (i.e., $92,448,860 − $42,239,105). Thus, even if the CRA’s position were correct, the amount of the capital dividend paid by ESRIL 98 to ESRL was less than the amount of the available CDA balance. Accordingly, it is difficult to see how ESRIL 98 avoided the payment of

Part III tax when it redeemed the shares, made the subsection 83(2) election, and was deemed to have paid the capital dividend on October 27, 2009. [27] [ 45 ] ESRIL 98 has not paid any capital dividends after October 27, 2009. Therefore, there could not have been an avoidance of

Part III tax by ESRIL 98 subsequent to October 27, 2009, nor was there a tax benefit, based on this argument. (

b) Reduction of

Part I Tax by Ultimate Shareholders [ 46 ] The CRA also takes the position that the ultimate shareholders of ESRIL 98 have reduced the amount of

Part I tax that would be payable by them if they were to receive taxable dividends, rather than capital dividends, derived from ESRIL 98. [28] However, there is no evidence to indicate that any capital dividends in excess of $49,998,834 have been paid by ESRIL 98 to ESRL or that any capital dividends have been paid by ESRL to its shareholders. [ 47 ] In Wild , the Federal Court of Appeal, in conducting a GAAR analysis, distinguished between the creation of the potential for a tax- free distribution of a corporation’s retained earnings and the realization of that potential. [29] The Court indicated that, similar to the pre- packaging of tax losses in OSFC , [30] the transactions in Wild that resulted in the increased paid-up capital in respect of certain shares did not result in a tax benefit. [31] [ 48 ] In commenting on the Wild decision, Brian Arnold stated: Apart from the taxpayer’s concession [that he had obtained a tax benefit from the transactions in question], it seems clear to me that the increase in paid-up capital was not a tax benefit.

A deduction, allowance, credit, exemption or exclusion is clearly a tax benefit because they all result in a reduction of tax payable. However, an increase in a tax attribute does not result in any reduction, avoidance or deferral of tax; it may be a benefit in a general sense because it allows an amount to be distributed tax-free in the future. [32] After making the above comment, Arnold went on to discuss the significance of the phrase " “directly or indirectly” " in subsections 245(2), (3) and (4) of the ITA . He stated:

Although it is not completely clear, it would seem that the phrase “directly or indirectly” is used out of excess caution to ensure that the application of the GAAR cannot be avoided by technical arguments about whether a transaction or a series of transactions has resulted in a tax benefit. For example, where a taxpayer engages in a transaction that results in an increase in the cost of depreciable property, there is no tax benefit until the taxpayer claims capital cost allowance.

Once the taxpayer claims capital cost allowance, there is a tax benefit; however, does the tax benefit result directly from the transaction that increased the capital cost of the property or from the taxpayer’s claim for capital cost allowance? This issue is resolved by the use of the term “indirectly” because the tax benefit results indirectly from the transaction that increased the cost of the property. However, in most cases, the concept of a series of transactions will capture tax benefits that are achieved circuitously.

Therefore, it is unclear what, if anything, the inclusion of the phrase “directly or indirectly” in the definition of an avoidance transaction adds to the GAAR. Nevertheless, it is clear that the phrase does not extend the GAAR to tax benefits that have not yet resulted in any tax saving but may do so in the future. [33] [ Emphasis added .] Arnold then said the following in respect of the applicability of subsection 152(1.11) of the ITA : Subsection 152(1.11) is also relevant to this discussion.

The purpose of subsection 152(1.11) would appear to be to allow the Minister to make a determination of an amount, such as the paid-capital of shares or the adjusted cost base of property, rather than an assessment of tax, since tax may not be payable until a subsequent year. As Jim Wilson and I pointed out back in 1988 ( Canadian Tax Journal, vol. 36, no. 5 at 1176), the wording of the GAAR and subsection 152(1.11) is inconsistent with this purpose.

Subsection 152(1.11) does not explicitly refer to a tax benefit and subsection 245(2) does not apply until there is an avoidance transaction and a tax benefit; as a result, subsection 152(1.11) applies only where the Minister ascertains the tax consequences under subsection 245(2). Therefore, the earliest time that subsection 152(1.11) would allow the Minister to make a determination (for example, a determination of the amount of paid-capital of shares or the adjusted cost base of capital property) is the first taxation year in which the taxpayer claims a tax benefit.

This result is problematic because it forces the CRA to maintain a watching brief on the taxpayer and the taxpayer’s successors in order to apply the GAAR when the increased paid-capital is used to shelter an otherwise taxable distribution. Take the simple example where the capital cost of depreciable property is increased through an abusive avoidance transaction, but the taxpayer does not claim any capital cost allowance in the year in which the transaction occurs. The CRA must wait until the first year in which the taxpayer claims capital cost allowance before it can apply the GAAR.

Once the GAAR applies, however, the Minister can make a determination under subsection 152(1.11) to reduce the capital cost of the properties; the Minster does not have to wait and apply the GAAR every time the taxpayer claims capital cost allowance. [34] [ 49 ] Accordingly, the future reduction of tax under

Part I of the ITA by the ultimate shareholders at ESRIL 98, as suggested by the Crown, is not a tax benefit at this time.

(4) Avoidance of

Part III Tax by CGESR [ 50 ] In Canadian Pacific , Justice Bonner revisited the question of identifying a standard against which to measure an alleged reduction in tax, as follows: The definition of tax benefit in s. 245(1), by referring to “a reduction, avoidance or deferral of tax…”, assumes the existence of a standard amount of tax against which reduction may be measured….

The standard against which reduction is to be measured is not a transaction which is theoretically possible but, practically speaking, unlikely in the circumstances. [35] [ Footnote omitted. ] [ 51 ] After CGESR received life insurance proceeds on December 17, 2008 in respect of the twelve Policies in the total amount of $102,309,794 and interest in the amount of $84,090 (which CGESR mistakenly treated as life insurance proceeds), CGESR understood that the amount of its CDA was increased by $102,393,885 (i.e., $102,309,794 + $84,090, with rounding), given that the ACB of the Policies to CGESR was nil. [36] As the balance of CGESR’s CDA had been $54,701 on December 2, 2008, CGESR understood that its CDA balance on December 17, 2008 was $102,448,586 (i.e., $54,701 + 102,393,885).

On July 15, 2009, CGESR paid capital dividends in the total amount of $10,702,742. On October 27, 2009, CGESR redeemed some of the class A preferred shares in its capital that were owned by ESRIL 98, resulting in a deemed dividend in the amount of $91,745,839, in respect of which CGESR made a capital dividend election under subsection 83(2) of the ITA . Therefore, during the four-month period from July 2009 to October 2009, CGESR paid capital dividends (actual or deemed) in the aggregate amount of $102,448,581 (i.e., $10,702,742 + $91,745,839), which was $5 less than the understood balance of its CDA.

However, as indicated in footnote 6 above, this balance was excessive to the extent of $84,090, representing interest paid on the insurance proceeds by the insurer to CGESR and inadvertently included in computing the balance of CGESR’s CDA. CGESR was assessed, and paid,

Part III tax in respect of the $84,090. [ 52 ] According to the Crown’s theory, if the ACB to ESRL of the ten Policies held by it (i.e., $42,239,105) were subtracted from the amount of the actual life insurance proceeds (not including interest) added to CGESR’s CDA, the CDA balance would have been only $60,070,689 (i.e., $102,309,794 − $42,239,105), and the aggregate capital dividends (actual or deemed but excluding the capital dividend in respect of the interest of $84,090) paid by CGESR in 2009 would have exceeded the CDA to the extent of $42,293,807 (i.e., $102,364,496 [37] − $60,070,689). It is the position of the Crown that CGESR avoided tax under

Part III in respect of that excess. [38] [ 53 ] The Crown’s position assumes that CGESR would have paid capital dividends (actual or deemed) in the total amount of $102,364,496, even if its CDA balance had been only $60,070,689. This is the alternative transaction which the Crown submits should be used in determining whether there was an avoidance of tax for the purposes of the definition of " “tax benefit.” " However, in my view, this is not a reasonable alternative transaction, as it is unlikely that CGESR would have paid capital dividends (actual or deemed) in the aggregate amount of $102,364,496 if its CDA balance had been only $60,070,689.

[ 54 ] The above view is consistent with the mechanism set out in subsections 184(3) and (4) (which are in

Part III) of the ITA , whereby, if a corporation makes a capital dividend election under subsection 83(2) of the ITA and the amount of the dividend exceeds the corporation’s capital dividend account, the corporation may, with the concurrence of all its shareholders, elect under subsection 184(3), whereupon the excess portion of the dividend is deemed to be a separate taxable dividend. Therefore, while it was unlikely that CGESR would have knowingly made an election under subsection 83(2) in respect of a dividend that exceeded the balance of its capital dividend account by $42,293,807, if such had occurred, by making an election under subsection 184(3), no

Part III tax would have been payable in any event. [ 55 ] The Notice of Determination issued by the CRA on behalf of the Minister for the taxation year of CGESR ended September 30, 2010 did not actually impose any

Part III tax (or tax under any other part of the ITA ). Rather, the Minister determined that, to the extent of $42,239,100, the dividends (actual or deemed) paid by CGESR to ESRIL 98 were taxable dividends, rather than capital dividends, and that those taxable dividends were deductible under paragraph 112(1)(

a) of the ITA in computing the taxable income of ESRIL 98. This is consistent with the result that would have occurred if the CRA had assessed

Part III tax under subsection 184(2) of the ITA and if CGESR had made an election under subsection 184(3) of the ITA (as explained in the preceding paragraph). [ 56 ] As noted above, [39] in Copthorne, the Supreme Court indicated that, if a taxpayer’s situation is compared with an alternative arrangement, “the alternative arrangement must be one that ‘might reasonably have been carried out but for the existence of the tax benefit’.…” [40] I am of the view that, if the balance of CGESR’s CDA had been only $60,070,689 on October 27, 2009, it would have been unreasonable for CGESR to have made capital dividend elections under subsection 83(2) of the ITA in the aggregate amount of $102,364,491. [ 57 ] Applying the principle enunciated by Justice Bonner in Canadian Pacific , [41] while it is theoretically possible that CGESR may have paid capital dividends in the aggregate amount of $102,364,491 [42] if the balance of its CDA was only $60,070,689, practically speaking, it is unlikely that CGESR would have done so.

Rather, it is my view that, if the balance of CGESR’s CDA had been only $60,070,689, CGESR would have paid a capital dividend equal to or less than that amount, and any additional dividends would have been structured as taxable dividends. In fact, as noted above, that is the way that the CRA processed the Determinations. Accordingly, I am of the view that the Crown’s view as to the comparative alternative arrangement was only a theoretical possibility that was, practically speaking, unlikely in the circumstances. Accordingly, there was no tax benefit in this regard.

(5) Summary [ 58 ] It is my view that the Subject Transactions did not result in a tax benefit; however, in case that view is mistaken, I will proceed to the other two steps of a GAAR analysis. C. Avoidance Transaction [ 59 ] Subsection 245(3) of the ITA defines an " “avoidance transaction” " as follows: An avoidance transaction means any transaction (

a) that, but for this section, would result, directly or indirectly, in a tax benefit, unless the transaction may reasonably be considered to have been undertaken or arranged primarily for bona fide purposes other than to obtain the tax benefit; or (

b) that is part of a series of transactions, which series, but for this section, would result, directly or indirectly, in a tax benefit, unless the transaction may reasonably be considered to have been undertaken or arranged primarily for bona fide purposes other than to obtain the tax benefit. [ 60 ] While I have found that the Subject Transactions did not result in a tax benefit, in the event that my finding is incorrect (i.e., if there was a tax benefit), RE 2015 has admitted that:

a) but for

section 245 of the ITA , one or more of the Subject Transactions resulted, directly or indirectly, in the tax benefit; and

b) each such Subject Transaction was an avoidance transaction within the meaning of subsection 245(3) of the ITA . [43] Therefore, nothing more needs to be said about this step of the GAAR analysis. D. Abusive Tax Avoidance

(1) Applicable Principles [ 61 ] Subsection 245(2) of the ITA provides that, where a transaction is an avoidance transaction, the tax consequences to a person shall be determined in order to deny a tax benefit that would otherwise result from that transaction. By reason of subsection 245(4) of the ITA , subsection 245(2) applies only if there is a misuse of a provision of (in the context of these Appeals) the ITA or an abuse of the provisions of the ITA (other than

section 245) read as a whole. In Canada Trustco , the Supreme Court of Canada stated: 44. The heart of the analysis under s. 245(4) lies in a contextual and purposive

interpretation of the provisions of the Act that are relied on by the taxpayer, and the application of the properly interpreted provisions to the facts of a given case. The first task is to interpret the provisions giving rise to the tax benefit to determine their object, spirit and purpose. The next task is to determine whether the transaction falls within or frustrates that purpose…. 45. This analysis will lead to a finding of abusive tax avoidance when a taxpayer relies on specific provisions of the Income

Tax Act in order to achieve an outcome that those provisions seek to prevent. As well, abusive tax avoidance will occur when a transaction defeats the underlying rationale of the provisions that are relied upon. An abuse may also result from an arrangement that circumvents the application of certain provisions, such as specific anti-avoidance rules, in a manner that frustrates or defeats the object, spirit or purpose of those provisions.

By contrast, abuse is not established where it is reasonable to conclude that an avoidance transaction under s. 245(3) was within the object, spirit or purpose of the provisions that confer the tax benefit. [44] [ 62 ] In Copthorne , the Supreme Court of Canada provided additional guidance concerning the use of the textual, contextual and purposive analysis to determine the object, spirit or purpose of a statutory provision, as follows: 70. The object, spirit or purpose can be identified by applying the same interpretative approach employed by this Court in all questions of statutory

interpretation — a “unified textual, contextual and purposive approach”…. While the approach is the same as in all statutory

interpretation, the analysis seeks to determine a different aspect of the statute than in other cases. In a traditional statutory

interpretation approach the court applies the textual, contextual and purposive analysis to determine what the words of the statute mean. In a GAAR analysis the textual, contextual and purposive analysis is employed to determine the object, spirit or purpose of a provision. Here the meaning of the words of the statute may be clear enough. The search is for the rationale that underlies the words that may not be captured by the bare meaning of the words themselves. However, determining the rationale of the relevant provisions of the Act should not be conflated with a value judgment of what is right or wrong nor with theories about what tax law ought to be or ought to do. [45]

(2) Object, Spirit and Purpose [46] [ 63 ] Central to the analysis required to resolve these Appeals is an

interpretation of the object, spirit and purpose of the provision (which I will call the " “Reduction Provision” " ) in the definition of " “capital dividend account” " that reduces the amount of the proceeds of a life insurance policy received by a corporation that may be added to the corporation’s capital dividend account. As will be shown below, the wording of the Reduction Provision has changed over the years. It is my view that the policy underlying the Reduction Provision has also changed from time to time. Before embarking on an

interpretation of the Reduction Provision, I will review the legislative history of the Reduction Provision and other relevant provisions in the ITA , with a focus on several significant amendments. (

a) Legislative History [ 64 ] Counsel for the Crown took me through the legislative history of the definition of " “capital dividend account” " and a couple of the legislative provisions pertaining to life insurance policies. [47] (i) 1969 [ 65 ] The taxation of income earned in respect of a life insurance policy was inserted in the former Income Tax Act (the " “ Former Act ” " ) [48] in 1969, when the following provision was added as subsection 79D(1): 79D.

(1) There shall be included in computing the income for a taxation year of a policyholder, (

a) in respect of any life insurance policy other than an annuity contract, the amount, if any, by which the proceeds of the disposition of an interest in the policy that he became entitled to receive in the year exceeds the adjusted cost basis of the policy to the policyholder as of the time of the disposition…. [49] [ Emphasis added. ] The above provision indicates that Parliament was aware of the need to determine the adjusted cost basis of a policy by reference to a particular person, in this case the policyholder. [ 66 ] At the same time, the following

definitions of " “adjusted cost basis” " and " “cash surrender value” " were included in subsection 79D(10) of the Former Act :

(10) In this section, " (a) “adjusted cost basis” to a policyholder as of a particular time of an interest in a life insurance policy means the amount, if any, by which " " (

i) the aggregate of the cost to him of acquiring all his interests in the policy and all amounts paid by him or on his behalf before that time as or on account of premiums under the policy, " " exceeds " " (ii) all proceeds of disposition of his interests in the policy that he became entitled to receive before that time, except to the extent that such proceeds were required to be included in computing his income for a taxation year by virtue of paragraph (

a) of subsection (1); " " (b) “cash surrender value” at a particular time of a life insurance policy means its cash surrender value at that time computed without regard to any policy dividends payable thereunder or any interest payable on such dividends; …. " [50] " " It is noteworthy that " “cash surrender value” " was defined without any reference to the holder of the particular life insurance policy, whereas " “adjusted cost basis” " was defined by reference to the particular policyholder.

In other words, as the adjusted cost basis of an interest in a life insurance policy could be determined only by reference to a particular person (in this case the policyholder), the policyholder was part of the context to be considered in applying the definition of " “adjusted cost basis.” " It is also significant that, insofar as premiums were concerned, the definition of " “adjusted cost basis” " recognized the premiums paid by the policyholder or on

behalf of the policyholder. (ii) 1972 [ 67 ] With the advent of tax reform, effective as of January 1, 1972, paragraph 79D(1)(

a) of the Former Act became paragraph 148(1)(

a) of the ITA (without any changes in the wording of that provision). Similarly, the definition of the term " “adjusted cost basis” " in paragraph 79D(10)(

a) of the Former Act , in essence, became paragraph 148(9)(

a) of the ITA . [51] By reason of capital gains becoming taxable in 1972, the definition of " “capital dividend account” " was added as paragraph 89(1)(

b) of the ITA . For the purposes of these Appeals, the following portions of that definition are relevant: (b) “ capital dividend account ” of a corporation at any particular time means the amount, if any, by which the aggregate of … (iv) the amount, if any, by which (

A) the proceeds of any life insurance policy received by the corporation in the period and after 1971 in consequence of the death of any person whose life was insured under the policy, exceeds (

B) all amounts paid as or on account of premiums paid under the policy ,…. [52] [ Emphasis added. ] Thus, in the initial version of the definition of " “capital dividend account,” " the Reduction Provision consisted of the word " “exceeds” " and clause 89(1)(b)(iv)(B). Clause 89(1)(b)(iv)(

B) referred to amounts paid as or on account of premiums, without any reference to the person by whom or on whose behalf those premiums may have been paid. In other words, insofar as the payor of the premiums was concerned, clause 89(1)(b)(iv)(

B) was broader than subparagraph (

i) in the definition of " “adjusted cost basis” " in paragraph 148(9)(

a) of the ITA , as it read at that time. (iii) 1977 [ 68 ] In 1977, Parliament amended clause 89(1)(b)(iv)(

B) of the ITA , so as to include in the Reduction Provision a reference to the adjusted cost basis of a life insurance policy to the corporate beneficiary (rather than referring only to the premiums paid under the policy). The amended definition of " “capital dividend account” " read as follows: (b) “ capital dividend account ” of a corporation at any particular time means the amount, if any, by which the aggregate of … (iv) the amount, if any, by which (

A) the proceeds of any life insurance policy received by the corporation in the period and after 1971 in consequence of the death of any person whose life was insured under the policy, exceeds (

B) the adjusted cost basis of the policy (within the meaning of paragraph 148(9)(

a) to the corporation immediately before that person’s death,…. [53] [ Emphasis added. ] [ 69 ] In 1977, Parliament also amended paragraph 148(9)(

a) of the ITA so as to expand the definition of " “adjusted cost basis” " to a policyholder of an interest in a life insurance policy. The expanded definition read as follows: (a) “adjusted cost basis” to a policyholder as at a particular time of his interest in a life insurance policy means the amount, if any, by which the aggregate of (

i) the cost of each interest in the policy acquired by him before that time but not including an amount referred to in subparagraph (ii) or (iv), (ii) all amounts each of which is an amount paid before that time, by him or on his behalf, in respect of a premium under the policy, (iii) all amounts, each of which is an amount in respect of the disposition of an interest in the policy before that time that was required by paragraph (1)(

a) as it read for the 1977 taxation year, subsection (1),

section 16 or paragraph 56(1)(

d) to be included in computing his income for a taxation year, (iv) all amounts, each of which is an amount in respect of the repayment before that time and after March 31, 1978 of a policy loan in respect of the policy but not including any payment of interest thereon or any repayment of a loan that was deductible pursuant to paragraph 20(1)(hh), and (

v) the amount, if any, by which the cash surrender value of the policy as at its first anniversary date after March 31, 1977, exceeds the adjusted cost basis (determined under this Act as it would have read on that date if subsection (8) had not been applicable) of his interest in the policy on that date exceeds the aggregate of (vi) the aggregate of amounts of which is proceeds of the disposition of his interest in the policy that he became entitled to

receive before that time, (vii) the amount payable on March 31, 1978 in respect of a policy loan in respect of the policy; and (viii) the aggregate of all amounts each of which is an amount received before that time in respect of the policy that he was entitled to deduct under paragraph 60(

a) in computing his income for a taxation year;…. [54] The expanded definition of " “adjusted cost basis” " contained a number of additional factors, as well as the three original factors, i.e., the cost of acquiring an interest in a policy, the premiums paid under the policy and the proceeds of disposition of an interest in the policy. [ 70 ] The 1977 amendment also added the definition of " “proceeds of the disposition” " to the ITA , as paragraph 148(9)(e.2). [55] [ 71 ] Also in 1977, Parliament amended subsection 148(1) of the ITA . Until that time, paragraph 148(1)(

a) had remained unchanged since its original enactment as paragraph 79D(1)(

a) of the Former Act .

The relevant portions of the subsection, as amended in 1977, read as follows: 148(1) There shall be included in computing the income for a taxation year of a policyholder in respect of the disposition of an interest in a life insurance policy other than, (a) [n/a], or (b) [n/a] the amount, if any, by which the proceeds of the disposition of that interest in the policy that the policyholder, beneficiary or assignee, as the case may be, became entitled to receive in that year exceeds the adjusted cost basis to the policyholder of that interest immediately before the disposition. [56] [ Emphasis added. ] It is noteworthy that in 1977 Parliament continued, in subsection 148(1), to use the same concept that had been in place since 1969, by using the phrase " “the adjusted cost basis to the policyholder of that interest,” " notwithstanding that, in the 1977 definition of " “capital dividend account” " in clause 89(1)(b)(iv)(B), as amended by the same statute, Parliament had adopted the phrase " “the adjusted cost basis of the policy … to the corporation….” " To me, this suggests that Parliament had deliberately chosen to use two different entities as the complementary reference points (i.e., the objects of the preposition “to”) for the term “adjusted cost basis,” i.e. the corporation in the context of clause 89(1)(b)(iv)(

B) and the policyholder in the context of subsection 148(1) and paragraph 148(9)(a). [ 72 ] In 1977, as noted above, paragraph 148(9)(

a) of the ITA continued to define the " “adjusted cost basis” " to a policyholder of his interest in a life insurance policy by reference to (among other things) the amounts paid by or on behalf of the policyholder in respect of a premium under the policy, while paragraph 89(1)(

b) defined " “capital dividend account” " by reference to the adjusted cost basis of a policy to the corporation (notwithstanding that subsection 148(1) and paragraph 148(9)(

a) continued to refer to the adjusted cost basis to the policyholder of the particular interest in the policy). By reading paragraphs 89(1)(

b) and 148(9)(

a) and subsection 148(1) together, there was an implication that, for the purpose of determining the amount to be added to a corporation’s capital dividend account, one was to consider the premiums paid by the corporation, but not by the policyholder (assuming that the policyholder was not the beneficiary). [57] This was a change from the legislative situation that had existed before April 1, 1977, and indicated that in 1977 there was a change in legislative policy. [ 73 ] To summarize, from 1972 to March 31, 1977, clause 89(1)(b)(iv)(

B) of the ITA had referred to " “all amounts paid as or on account of premiums paid under the policy,” " regardless of the identity of the payor of those premiums. With the 1977 amendment, upon a corporation receiving the proceeds of a life insurance policy, the amount to be added to the corporation’s capital dividend account was now calculated by reference to the adjusted cost basis of the policy to the corporation.

Thus, after March 31, 1977, only premiums paid by or on behalf of the corporation, as well as the cost of acquiring each interest in the policy, any proceeds of the disposition of an interest in the policy and several other factors, were to be considered in calculating the amount to be added to the corporation’s capital dividend account. This clearly signalled a change in the factors to be considered in determining the amount to be added to a corporation’s capital dividend account upon the corporation receiving proceeds of a life insurance policy.

As noted above, this was indicative of a change in legislative policy in 1977. (iv) 1982-83 [ 74 ] In 1983, when Parliament introduced the concept of a life insurance capital dividend account effective as of June 28, 1982, it also amended the definition of " “capital dividend account” " to read as follows: (b) “capital dividend account” of a corporation at any particular time means the amount, if any, by which the aggregate of … (iv) the amount, if any, by which (

A) the aggregate of all amounts each of which is the proceeds of a life insurance policy of which the corporation was a beneficiary on or before June 28, 1982 received by the corporation in the period and after 1971 in consequence of the death of any person whose life was insured under the policy, exceeds (

B) the aggregate of all amounts each of which is the adjusted cost basis of the policy (within the meaning of paragraph 148(9)(a)) to the corporation immediately before that person’s death,… [58] [ Emphasis added; see footnote below. ] [ 75 ] The following definition of " “life insurance capital dividend account” " was enacted by Parliament at the same time:

(b.2) “life insurance capital dividend account” of a corporation at any particular time means the amount, if any, by which the aggregate of (

i) the amount, if any, by which (

A) the aggregate of all amounts each of which is the proceeds of a life insurance policy of which the corporation was not a beneficiary on or before June 28, 1982 received as a consequence of the death of a person whose life was insured under the policy by the corporation in the period (

I) commencing on the first day of the first taxation year commencing after the time the corporation last became a private corporation and ending after 1971, and (II) ending immediately before the particular time exceeds (

B) the aggregate of all amounts each of which is the adjusted cost basis (within the meaning assigned by paragraph 148(9) (a)) to the corporation immediately before that person’s death of a policy referred to in clause (A), and …. [59] [ Emphasis added. ] There was a similarity between the manner of calculating a corporation’s life insurance capital dividend account and the manner in which a corporation’s capital dividend account had previously been calculated. In particular, there was symmetry between clauses 89(1)(b)(iv) (

B) and 89(1)(b.2)(i)(B). If Parliament had intended that premiums paid by the policyholder would be taken into consideration in calculating the amount of insurance proceeds to be added to the life insurance capital dividend account, it could easily have so provided, rather than continuing to use the phrase " “adjusted cost basis … to the corporation.” " [ 76 ] In 1983, Parliament also amended the definition of " “adjusted cost basis” " in subsection 148(9), so as to exclude from that term the net cost of pure insurance in respect of a life insurance policy acquired by a policyholder after December 1, 1982.

The relevant portion of the amended definition was the following: (a) " “adjusted cost basis” " to a policyholder as at a particular time of his interest in a life insurance policy means the amount, if any, by which the aggregate of (

i) the cost of each interest in the policy acquired by him before that time but not including an amount referred to in subparagraph (ii) or (iv), (ii) all amounts each of which is an amount paid before that time, by him or on his behalf, in respect of a premium under the policy,…. exceeds the aggregate of (vi) the aggregate of amounts each of which is proceeds of the disposition of his interest in the policy that he became entitled to receive before that time,…. (ix) in the case of an interest in a life insurance policy (other than an annuity contract) that was last acquired after December 1, 1982 by the policyholder, the aggregate of all amounts each of which is the net cost of pure insurance in respect of the interest, as determined in prescribed manner, immediately before the end of the calendar year ending in a taxation year commencing after May 31, 1985 and before that time,…. [60] [ 77 ] Turning to subsection 148(1) of the ITA, although amendments were made in 1983 to the opening and middle portions of that subsection, the concluding portion contained only a minor amendment (the phrase " “in that year” " was replaced with the phrase " “in the year” " ).

In particular, the phrase " “exceeds the adjusted cost basis to the policyholder of that interest” " remained unchanged. [61] (v) 1986 [ 78 ] In 1986, in conjunction with the repeal of the short-lived rules relating to the life insurance capital dividend account, the definition of " “capital dividend account” " in paragraph 89(1)(

b) of the ITA was amended to read as follows: (b) “ capital dividend account ” of a corporation at any particular time means the amount, if any, by which the aggregate of … (iv) the amount, if any, by which the aggregate of (

A) all amounts each of which is the proceeds of a life insurance policy of which the corporation was a beneficiary on or before June 28, 1982 received by the corporation in the period and after 1971 in consequence of the death of any person, and (

B) all amounts each of which is the proceeds of a life insurance policy of which the corporation was not a beneficiary on or before June 28, 1982 received by the corporation in the period and after May 23, 1985 in consequence of the death of any person exceeds the aggregate of all amounts each of which is the adjusted cost basis (within the meaning assigned by paragraph 148(9)(a)) of a policy referred to in clause (

A) or (

B) to the corporation immediately before that person’s death,… [62] [ Emphasis added. ]

The 1986 amendment did not change the general import of the Reduction Provision that had been in place since 1977, i.e., the amount to be added to a corporation’s capital dividend account, upon receiving proceeds of a life insurance policy, was to be reduced by the adjusted cost basis of the policy to the corporation. (vi) 1994 [ 79 ] The ITA was brought within the Revised Statutes of Canada 1985 (“RSC 1985”) on March 1, 1994, as

chapter 1 of the Fifth Supplement, with the result that defined terms were no longer identified as particular paragraphs of a

section or subsection, but rather were listed alphabetically (in both official languages) in the particular

section or subsection. This led to some internal renumbering of paragraphs, subparagraphs and the like in many of the defined terms. Consequently, the relevant portion of the definition of " “capital dividend account” " was reworded as follows: “ capital dividend account ” of a corporation at any particular time means the amount, if any, by which the total of … (

d) the amount, if any, by which the total of (

i) all amounts each of which is the proceeds of a life insurance policy of which the corporation was a beneficiary on or before June 28, 1982 received by the corporation in the period and after 1971 in consequence of the death of any person, and (ii) all amounts each of which is the proceeds of a life insurance policy of which the corporation was not a beneficiary on or before June 28, 1982 received by the corporation in the period and after May 23, 1985 in consequence of the death of any person exceeds the total of all amounts each of which is the adjusted cost basis (within the meaning assigned by subsection 148(9)) of a policy referred to in subparagraph (

i) or (ii) to the corporation immediately before that person’s death,…. [63] [ Emphasis added. ] Thus, notwithstanding the conversion to the formatting of the " RSC 1985 " , the substance of the life insurance portion of the definition of " “capital dividend account” " remained essentially the same as previously. [64] [ 80 ] More significant changes in form were visible in the RSC 1985 version of the definition of " “adjusted cost basis,” " now listed alphabetically in subsection 148(9) of the ITA . Notably, a formulaic approach to the definition was used.

In addition, shortly after the RSC 1985 version took effect and before the commercial publishers could release a new edition of the ITA , Parliament made a few amendments. For the purposes of these Appeals, the relevant portions of the definition of " “adjusted cost basis,” " in RSC 1985 format and with an amendment made in 1994, read as follows: “ adjusted cost basis ” to a policyholder as at a particular time of the policyholder’s interest in a life insurance policy means the amount determined by the formula (A + B …) − (H + …

L) where A is the total of all amounts each of which is the cost of an interest in the policy acquired by the policyholder before that time but not including an amount referred to in the description of B or E, B is the total of all amounts each of which is an amount paid before that time by or on behalf of the policyholder in respect of a premium under the policy, other than amounts referred to in clause (2)(a)(ii)(B), in subparagraph (iii) of the description of C in paragraph (

a) of the definition “proceeds of the disposition” or in subparagraph (b)(

i) of that definition, … H is the total of all amounts each of which is the proceeds of the disposition of the policyholder’s interest in the policy that the policyholder became entitled to receive before that time,… L is (

a) in the case of an interest in a life insurance policy (other than an annuity contract) that was last acquired after December 1, 1982 by the policyholder, the total of all amounts each of which is the net cost of pure insurance, as defined by regulation and determined by the issuer of the policy in accordance with the regulations, in respect of the interest immediately before the end of the calendar year ending in a taxation year commencing after May 31, 1985 and before that time,… [65] While there were formatting and numbering changes and a few other minor changes in the above portions of the definition of " “adjusted cost basis,” " the overall substance remained essentially the same. (vii) 2013 [ 81 ] In 2013, Parliament enacted amendments designed to address concerns relating to leveraged insurance annuity ( " “LIA” " ) policies and 10/8 arrangements.

Accordingly, paragraph (

d) of the definition of " “capital dividend account” " was amended to exclude, from the computation of a corporation’s capital dividend account, the proceeds of an LIA policy, and to provide that, if an insurance policy is a 10/8 policy, the outstanding amount of a described borrowing is not to be included in computing the capital dividend account. The amended definition read as follows: “capital dividend account” of a corporation at any particular time means the amount, if any, by which the total of …

(

d) the amount, if any, by which the total of (

i) all amounts each of which is the proceeds of a life insurance policy of which the corporation was a beneficiary on or before June 28, 1982 received by the corporation in the period and after 1971 in consequence of the death of any person, and (ii) all amounts each of which is the proceeds of a life insurance policy (other than an LIA policy) of which the corporation was not a beneficiary on or before June 28, 1982 received by the corporation in the period and after May 23, 1985 in consequence of the death of any person exceeds the total of all amounts each of which is: (iii) the adjusted cost basis (within the meaning assigned by subsection 148(9)) of a policy referred to in subparagraph (

i) or (ii) to the corporation immediately before the death, or (iv) if the policy is a 10/8 policy immediately before the death and the death occurs after 2013, the amount outstanding, immediately before the death, of the borrowing that is described in subparagraph (a)(

i) of the definition “10/8 policy” in subsection 248(1) in respect of the policy,…. [66] [ Emphasis added. ] The 2013 amendments did not make any significant changes to the relevant provisions of subsection 148(1) and the definition of " “adjusted cost basis.” " (viii) 2016 [ 82 ] In 2016, Parliament amended subparagraph (d)(iii) of the definition of " “capital dividend account” " in subsection 89(1) of the ITA , so as to preclude the type of planning that was done by the RPC Group in 2005.

After the amendment, which took effect as of December 15, 2016, subparagraphs (d)(i)–(iii) of that definition read as follows: “ capital dividend account ” of a corporation at any particular time means the amount, if any, by which the total of … (

d) the amount, if any, by which the total of (

i) all amounts each of which is the proceeds of a life insurance policy of which the corporation was a beneficiary on or before June 28, 1982, received by the corporation in the period and after 1971 in consequence of the death of any person, and (ii) all amounts each of which is the proceeds of a life insurance policy (other than an LIA policy) of which the corporation was not a beneficiary on or before June 28, 1982 received by the corporation in the period and after May 23, 1985 in consequence of the death of any person exceeds the total of all amounts each of which is (iii) the “ adjusted cost basis ” (in this paragraph as defined in subsection 148(9)), immediately before the death, of (

A) if the death occurs before March 22, 2016, a policy referred to in subparagraph (

i) or (ii) to the corporation , and (

B) if the death occurs after March 21, 2016, a policyholder’s interest in a policy referred to in subparagraph (

i) or (ii),… [67] [ Emphasis added. ] [ 83 ] The amendment referred to in the preceding paragraph was announced on March 22, 2016, in conjunction with the Budget Speech delivered by the Minister of Finance on that day. The proposed legislation set out in the budget documents released on March 22, 2016 was slightly different from the legislation that was actually enacted (as set out above).

Specifically, in the proposed legislation released on March 22, 2016, the parenthetical phrase, after the words " “adjusted cost basis” " in the initial portion of subparagraph (d)(iii), read simply " “(as defined in subsection 148(9)).” " The words " “in this paragraph” " were added to that parenthetical phrase in a Notice of Ways and Means Motion tabled in the House of Commons in October 2016. [ 84 ] The Notice of Ways and Means Motion tabled in October 2016 made further amendments to paragraph (

d) of the definition of " “capital dividend account,” " which resulted in the term " “adjusted cost basis” " being used three additional times in the amended provision. [68] Curiously, in the four situations in paragraph (

d) of the amended definition of " “capital dividend account” " where the term " “adjusted cost basis” " is used, three different wordings are used, as follows: (

a) In clause (d)(iii)(B), the wording is: … adjusted cost basis … of … a policyholder’s interest in a policy…. (

b) In subclause (d)(v)(A)(I), the wording is: … the adjusted cost basis to the policyholder of the interest…. (

c) In subparagraph (d)(vi), in defining variable A (as used in the formula in that subparagraph), the wording is: … the adjusted cost basis to the policyholder of the interest…. (

d) In subparagraph (d)(vi), in defining variable B (as used in the formula in that subparagraph), the wording is: … the adjusted cost basis … of the interest in the policy….

It is difficult to determine whether the three variations in the language of the provisions set out above were intended by Parliament to have any significance in interpreting those provisions or in ascertaining the legislative policy underlying those provisions. [ 85 ] The 2016 version of paragraph (

d) of the definition of " “capital dividend account” " was the first time that the term " “adjusted cost basis” " was enclosed within quotation marks in that definition " . " In all previous iterations of that definition, from 1977 to 2015, the term " “adjusted cost basis” " was not enclosed within quotation marks. Perhaps the reason for this difference was that, in previous iterations, the parenthetical phrase in the definition was " “(within the meaning assigned by subsection 148(9)) ,” " whereas in the 2016 iteration the parenthetical phrase was " “(in this paragraph as defined in subsection 148(9)) .” " [ 86 ] The wording in clause (d)(iii)(

B) of the definition of " “capital dividend account,” " as enacted by the 2016 budget legislation, is peculiar in that it deviates from the general practice in the ITA of describing an adjusted cost basis as being the adjusted cost basis to a particular person. [69] For instance, the definition of " “adjusted cost basis” " in subsection 148(9) of the ITA actually defines the " “‘adjusted cost basis’… to a policyholder of the policyholder’s interest in a life insurance policy…” " [ emphasis added ].

Similarly, both before and after the 2016 budget amendments, subsection 148(1) of the ITA described, and continues to describe, the income inclusion contemplated by that subsection (in the context of a disposition of a life insurance policy) as being determined by reference to " “the amount … by which the proceeds of the disposition of the policyholder’s interest in the policy … exceeds the adjusted cost basis to the policyholder of that interest…”[ emphasis added ]. " Even in two of the other three situations in paragraph (

d) of the definition of " “capital dividend account” " where the term " “adjusted cost basis” " was inserted by the 2016 amendments, the traditional wording (i.e., " “adjusted cost basis to the policyholder ” " ) [ emphasis added ] is used. [70] Only in the definition of variable B, in subparagraph (d)(vi), is the wording (i.e., " “the adjusted cost basis … of the interest in the policy” " ) somewhat consistent with (but not identical to) the wording in clause (d)(iii)(B). (

b) Effect of a Statutory Amendment (

i) Applicable Principles [ 87 ] Subsections 45(2) and (3) of the

Interpretation Act state:

(2) The amendment of an enactment shall not be deemed to be or to involve a declaration that the law under that enactment was or was considered by Parliament or other body or person by whom the enactment was enacted to have been different from the law as it is under the enactment as amended.

(3) The repeal or amendment of an enactment in whole or in part shall not be deemed to be or to involve any declaration as to the previous state of the law. [ 88 ] In commenting on the above provisions, the Federal Court of Appeal has stated: 46. … I note for now that subsequent amendments cannot be assumed to alter or confirm the prior state of the law (see subsections 45(2) and (3) of the

Interpretation Act …). The recent decision of this Court in Univar Holdco Canada ULC … illustrates the point that in a GAAR context, the provisions used to obtain the tax benefit must first be construed on their own. Only then can one say whether a subsequent amendment that touches upon the same subject matter confirms or alters the prior state of the law…. 86. Whether an amendment clarifies the prior law or alters it turns on the construction of the prior law and the amendment itself. As explained, the

Interpretation Act prevents any conclusion from being drawn as to the legal effect of a new enactment on the prior law on the sole basis that Parliament adopted it.

Keeping this limitation in mind, the only way to assess the impact of a subsequent amendment on the prior law is to first determine the legal effect of the law as it stood beforehand and then determine whether the subsequent amendment alters it or clarifies it. [71] [ 89 ] The Federal Court of Appeal went on to caution that, in a GAAR context, when determining the effect of an amendment, it is not sufficient simply to compare the wording of the former provision to the wording of the amended provision. The Court stated: 88.

The difficulty with [the lower Court’s] reasoning is that it is based on the wording of the former provisions rather than on their object, spirit and purpose. As was stated in Copthorne , the GAAR contemplates that the meaning and legal effect of the provisions of the Act can vary depending on whether they are construed according to a traditional, textual, contextual and purposive construction focused on the meaning of the words of the Act , or on the basis of an analysis focused on discerning their underlying rationale or reason for being ( Copthorne at para. 70).

While one cannot rule out the possibility that the underlying rationale for a provision will be fully captured by the words, this must still be demonstrated by inquiring into the provision’s reason for being ( Copthorne at paras. 110-111). [72] (ii) Pre-2016 Amendments [ 90 ] When the definition of " “adjusted cost basis” " was added to the ITA in 1969, there were only three factors to be considered in computing the adjusted cost basis to a policyholder of an interest in a life insurance policy, namely, the cost of acquiring each of the policyholder’s interests, the premiums paid under the policy by or on behalf of the policyholder, and the proceeds of disposition (if any) of interests in the policy.

An almost identical definition of " “adjusted cost basis” " was included in the 1972 version of the ITA . [ 91 ] When the definition of " “capital dividend account” " was added to the ITA in 1972, as part of tax reform, the Reduction Provision in the definition provided that, in determining the portion of life insurance proceeds received by a corporation to be added to the corporation’s capital dividend account, all amounts paid as or on account of premiums under the policy were to be subtracted.

This situation continued until the definition was amended in 1977, when the Reduction Provision was changed, such that it no longer referred to total premiums paid, and instead r

[…]