[1997] 3 F.C. 441
A-491-95
ACTRA Fraternal Benefit Society (Appellant)
v.
Her Majesty the Queen (Respondent)
Indexed as: ACTRA Fraternal Benefit Society v. Canada (C.A.)
Court of Appeal, Strayer, Robertson and McDonald JJ.A.—Toronto, March 18; Ottawa, April 30, 1997.
Income tax — Exemptions — Fraternal benefit society life insurance fund — Where assets in life insurance fund exceed amount necessary for purposes of life insurance business, only investment income earned on necessary amounts should be treated as taxable income under Income Tax Act, s. 149 — Canadian and British Insurance Companies Act, s. 81(1) examined — Decision not to withdraw surplus monies out of fund not determinative of appeal — Inference all assets in fund necessary for life insurance operation rebuttable — Objective standard applicable.
Insurance — Fraternal benefit society life insurance fund — Income tax payable on investment income — Where assets in life insurance fund exceed amount necessary for purposes of life insurance business, tax payable on investment income earned on necessary amounts only — Canadian and British Insurance Companies Act, s. 81(1) examined — Decision not to withdraw surplus monies out of fund not determinative of appeal — Inference all assets in fund necessary for life insurance operation rebuttable.
The appellant taxpayer was a fraternal benefit society incorporated under the Canadian and British Insurance Companies Act (Insurance Act) which carried on several activities on behalf of its members. One of these was the provision of life insurance benefits through a life fund. The taxpayer maintained two other funds for such things as accident and sickness insurance benefits, for alcohol and drug rehabilitation services and for scholarships for members and their families. Fraternal benefit societies were subject to federal income tax only on taxable income attributable to their life insurance business.
In 1987, the taxpayer decided to adopt a new method for calculating investment income, the investment base method. This new method had the effect of increasing the taxpayer’s tax liability not only with respect to the 1988 and subsequent taxation years but also for the preceding years. It was determined, by an actuary, that the life fund had a surplus of approximately $2.6 million (that amount was considered unnecessary to the operation of the life insurance business) and, with the approval of the Superintendent of Insurance, it was decided to transfer that amount out of the life fund. The taxpayer therefore contacted Revenue Canada, proposing to pay more taxes, including taxes for previous years, but to exclude the investment income earned from the surplus assets in the past. However, the Minister reassessed the taxpayer for the taxation years 1984 to 1987 on the basis that all of the investment income earned from the assets held in the life fund in each of those years was to be included in computing the taxpayer’s income from its life insurance business.
On appeal from that assessment, the Tax Court Judge found in favor of the Minister. He concluded that for the purposes of the Income Tax Act, the assets necessary for the taxpayer to carry on its life insurance business were those which the taxpayer determined should be in the fund at any one time. Also, that as long as assets remained in the life fund they were affixed with a “statutory condition or charge” pursuant to subsection 81(1) of the Insurance Act. This was an appeal from that decision.
Held (McDonald J.A. dissenting), the appeal should be allowed.
Per Robertson J.A.: The principal issue raised herein was whether all of the investment income derived from assets contained within the life insurance fund was to be included when computing taxable income from the fraternal society’s life insurance business. This depended on whether all of the assets in the life fund were “necessary” to the taxpayer’s life insurance business.
Contrary to what the Tax Court found, subsection 81(1) of the Insurance Act could not be deemed to impose a charge on assets contained within the life fund. That provision did not have the legal effect of “locking in” all of the assets held in that fund. There was no restriction on the transfer of funds—no approval of the Superintendent of Insurance was needed. It was a matter of judgment within the jurisdiction of the fraternal society.
The argument that the decision not to withdraw surplus monies from the life fund was based on concern over the possibility of AIDS-related claims was not persuasive. The evidence revealed that the taxpayer’s concern over the effect on its insurance business of the spread of AIDS did not surface until 1988 and not in the earlier taxation years. And even if that conclusion were wrong, the decision of the taxpayer not to withdraw monies from the life fund was not deteminative of the issue as to whether all of the assets were necessary for purposes of carrying on the taxpayer’s life insurance business.
The argument that all of the assets retained within the life fund were committed to and employed in carrying on the taxpayer’s life insurance business and, therefore, were “risked” in that business could not stand. While the inference of “necessity” could be drawn from the taxpayer’s leaving the monies in the fund and from the financial statements submitted to the Superintendent of Insurance, that inference could be, and was, displaced by the taxpayer adducing persuasive evidence to the contrary (the actuary’s determination that $2.6 million was unnecessary to the operation of the life insurance business).
The comments of Dickson C.J. in Bronfman Trust v. The Queen, [1987] 1 S.C.R. 32 that “the courts must deal with what the taxpayer actually did, and not what he might have done” have no application herein. The question was not whether the taxpayer could have withdrawn the surplus assets from the fund, but whether all of the assets therein were necessary for the life insurance component of the taxpayer’s business.
McCutcheon Farms Ltd. v. M.N.R., [1991] 1 C.T.C. 50 (F.C.T.D.) and Lutheran Life Insurance Society of Canada v. Canada, [1991] 2 C.T.C. 284 (F.C.T.D.) were authority for the proposition that the conduct of, or business judgment exercised by, a taxpayer is not determinative of what is necessary. This would be a subjective standard. McCutcheon reaffirmed the general understanding that the onus is on the taxpayer to establish, on an objective (financial, actuarial) basis, what is or is not necessary for purposes of carrying on a business (or, in this case, what proportion of the investment income related to the life insurance business).
The proper remedy herein was to remit the matter to the Minister on the understanding that the original assessments were valid.
Per McDonald J.A. (dissenting): The appeal should be dismissed and the Minister’s assessment should stand.
The decision to keep additional assets in the life account was a business decision by the taxpayer which should not be questioned by this Court. Once the decision had been made, the taxpayer benefitted from its decreased risk position. The taxpayer could not now attempt to revisit the past and reassess its tax burden.
Even accepting that the existence of excess assets in the life fund raised a rebuttable presumption that the assets were risked or employed in the course of the business, the presumption has not been rebutted in this case. While actuarial evidence may be helpful, the taxpayer decided to maintain excess reserves “in the event of an unforeseen economic catastrophe”. It cannot later attempt to remove these reserves from tax liability simply because the unforeseen catastrophe did not materialize.
The case of Ensite Ltd. v. R., [1986] 2 S.C.R. 509 was distinguishable. But even if the test, applied therein, of whether assets were “risked or employed” in a business were to be applied herein, that test must necessarily be more flexible when that business is a life insurance fund. The risk structure in the insurance business is necessarily different than most businesses. While actuaries are able to prepare risk estimates, unforeseen events can occur which may result in claims exceeding even generous reserves.
The Court must deal with what the taxpayer actually did, and not what it might have done (Bronfman Trust). Where the taxpayer has made an educated business decision to retain assets in a taxable fund, the Court should not look back with the benefit of hindsight to effectively remove assets from that fund with retroactive effect in order to put the taxpayer in a more advantageous tax position.
STATUTES AND REGULATIONS JUDICIALLY CONSIDERED
Canadian and British Insurance Companies Act, R.S.C. 1970, c. I-15, ss. 81(1),(7), 95(2), 97.
Income Tax Act, S.C. 1970-71-72, c. 63, ss. 129(1), (4), 149(1)(k),(3).
CASES JUDICIALLY CONSIDERED
APPLIED:
Lutheran Life Insurance Society of Canada v. Canada, [1991] 2 C.T.C. 284; (1991), 91 DTC 5553; 47 F.T.R. 25 (F.C.T.D.); McCutcheon Farms Ltd. v. M.N.R., [1991] 1 C.T.C. 50; (1991), 91 DTC 5047; 40 F.T.R. 180 (F.C.T.D.).
DISTINGUISHED:
Bronfman Trust v. The Queen, [1987] 1 S.C.R. 32; (1987), 36 D.L.R. (4th) 197; [1987] 1 C.T.C. 117; 87 DTC 5059; 25 E.T.R. 13; 71 N.R. 134.
REFERRED TO:
Ensite Ltd. v. R., [1986] 2 S.C.R. 509; (1986), 33 D.L.R. (4th) 491; [1986] 2 C.T.C. 459; 86 DTC 6521; 70 N.R. 189; R. v. Marsh & McLennan, Limited, [1984] 1 F.C. 609 [1983] CTC 231; (1983), 83 DTC 5180; 48 N.R. 103 (C.A.); Matheson, J A v The Queen, [1974] CTC 186; (1974), 74 DTC 6176 (F.C.T.D).
APPEAL from a Tax Court of Canada decision ([1995] 2 C.T.C. 2671; (1995), 96 DTC 1722 (T.C.C.)) confirming the Minister’s assessment including all of the investment income derived from assets contained within a “life insurance fund” when computing taxable income from a fraternal society’s life insurance business. Appeal allowed.
COUNSEL:
Joseph Groia and Mark I. Jadd for appellant.
Luther P. Chambers, Q.C. for respondent.
SOLICITORS:
Heenan Blaikie, Toronto, for appellant.
Deputy Attorney General of Canada for respondent.
The following are the reasons for judgment rendered in English by
Robertson J.A.: The appellant taxpayer is a fraternal benefit society incorporated under the provisions of the Canadian and British Insurance Companies Act, R.S.C. 1970, c. I-15 as amended (hereinafter the Insurance Act). Pursuant to paragraph 149(1)(k) and subsection 149(3) of the Income Tax Act [S.C. 1970-71-72, c. 63] such entities are subject only to federal income tax on taxable income attributable to their life insurance business. Income derived from other sources is exempt. The principal issue raised in this appeal is whether all of the investment income derived from assets contained within a “life insurance fund” (hereinafter the life fund) is to be included when computing taxable income from a fraternal society’s life insurance business. Before the Minister of National Revenue and the Tax Court, the taxpayer argued that in cases where the assets in the life fund exceed what is necessary for purposes of the life insurance business, only the investment income earned on the necessary amounts should be treated as taxable income. Both the Minister and the Tax Court Judge rejected this argument. In my respectful opinion, the appeal from the decision of the Tax Court must be allowed. As that decision is now reported it is unnecessary to offer an exhaustive analysis of the facts as did the learned Tax Court Judge: see [1995] 2 C.T.C. 2671.
A fraternal benefit society offers its members a variety of benefits. In the present case the taxpayer maintains three funds from which it is able to cover the costs of same for members of ACTRA, the Alliance of Canadian Cinema Television and Radio Artists. The above-noted life fund is directed at providing life insurance benefits to members. A second fund provides for accident and sickness insurance benefits. The third fund is labelled the fraternal fund. It provides for a multitude of benefits ranging from alcohol and drug rehabilitation services to scholarships for members and their families.
As a result of various collective agreements, the taxpayer receives a percentage of income earned by its members from their employers or, as they are referred to in the entertainment business, “engagers”. These engagers are also required to remit additional amounts on account of non-members. The taxpayer negotiated for such additional payments in order to ensure that it would not be less expensive for engagers to hire non-members. Such payments are partially refundable to non-members with the taxpayer retaining the remainder. Non-members in fact receive very few benefits.
Over time, some of the monies received on account of non-members found its way into the life fund. In addition to the amounts transferred to the life fund on account of non-member contributions, this so-called surplus generated significant investment income resulting in a “surplus” of approximately $3.6 million in the life fund at the end of 1987.
For its 1980 to 1987 taxation years inclusive, the taxpayer computed its investment income from the life insurance business on the basis of a formula devised by the taxpayer’s then auditor, Roger Gauvin. This formula was set out in a letter written by Mr. Gauvin and sent to officials with the Department of National Revenue in 1981. At no time did the Department take objection to the Gauvin formula even though it resulted in a nominal tax liability on the part of the taxpayer. At the same time as the taxpayer was submitting tax returns and calculating investment income with respect to its life fund on the basis of the Gauvin formula, the taxpayer was submitting financial statements to the Office of the Superintendent of Financial Institutions (hereinafter the Superintendent of Insurance). The latter financial statements reflected all of the investment income earned by the assets held in the life fund. For example, in 1985 the taxpayer reported investment income of $335,000 to the Superintendent of Insurance. That amount reflected, in part, investment income earned from surplus amounts attributable to contributions from non-members deposited in the life fund. By comparison, in filing its 1985 tax return the taxpayer reported investment income of $8,500 pursuant to the Gauvin formula. The Minister now seeks to add the difference between the two amounts of investment income to the taxpayer’s liability for each of the taxation years in question.
In 1987 the taxpayer retained Coopers & Lybrand to act as the former’s auditor. Acting on the auditor’s behalf, A. E. John Thompson undertook a review of the taxpayer’s accounting methods. Mr. Thompson concluded that the Gauvin formula did not provide the truest picture of the appellant’s life insurance business in that it underreported investment income. He also concluded that the investment income being reported to the Superintendent of Insurance was based on assets far in excess of what the life insurance operation required. Mr. Thompson proposed that the appellant adopt a new method for calculating investment income, the investment base method. Under this method the taxpayer’s investment income, for example, for the 1985 taxation year increased from $8,500 to $74,000. The adoption of the Thompson formula had the immediate effect of increasing the taxpayer’s tax liability not only with respect to the 1988 and subsequent taxation years but also for the preceding years.
In undertaking the review, Mr. Thompson retained the services of an actuary to determine the amount that was actually necessary to support the life insurance business and to recommend to the appellant the amount which could be prudently transferred out of the life insurance fund. Of the approximately $3.6 million held in the life fund as of 1987, it was determined that approximately $2.6 million was unnecessary to the operation of the life insurance business. Of the $1 million remaining in the life fund, approximately $350,000 represented a special allocation (the minimum surplus amount) to protect against the unknown risk of AIDS-related deaths. Before transferring the $2.6 million out of the life fund, the taxpayer consulted with the Superintendent of Insurance who, on the basis of the actuarial evidence presented, voiced no objection to the proposed transfer.
The taxpayer accepted Mr. Thompson’s proposed method of calculating investment income and on its own initiative raised the matter with officials at the Department of National Revenue in 1988. In acknowledging that it had underreported investment income for the 1985, 1986 and 1987 taxation years (and inferentially the 1984 taxation year), the taxpayer agreed that it would have to pay additional taxes in the vicinity of $80,000 based on the Thompson formula. A field auditor with the Department recommended accepting both the taxpayer’s proposal to transfer the funds out of the life fund and the Thompson formula for calculating investment income. Initially, therefore, the Minister agreed that the assets found within the life fund during the period 1985 to 1987 did not accurately or reasonably represent those related to the life insurance business. However, the matter was referred to the Tax Avoidance and Audit Applications Division of the Department. Ultimately, it was decided that the taxpayer’s proposal should be rejected. Accordingly, the Minister reassessed the taxpayer for the 1984, 1985, 1986 and 1987 taxation years on the basis that all of the investment income earned from the assets held in the life fund in each of those taxation years was to be included in computing the taxpayer’s income from its life insurance business. In support of its position the Minister relied on a recent decision of the Federal Court Trial Division, a case which the field auditor felt could be distinguished easily: Lutheran Life Insurance Society of Canada v. Canada, [1991] 2 C.T.C. 284.
The Tax Court Judge made two critical findings. First, he concluded that for purposes of the Income Tax Act the assets necessary for the taxpayer to carry on its life insurance business are those which the taxpayer determines should be in the fund at any one time. In other words, it is simply a business decision to be made by the taxpayer subject to any direction that might be made by the Superintendent of Insurance. Second, as long as assets remain in the life fund they are affixed with a “statutory condition or charge” pursuant to subsection 81(1) of the Insurance Act. In the opinion of the Tax Court Judge that subsection makes all of the assets in the life fund a constituent part of the taxpayer’s life insurance business. Alternatively expressed, subsection 81(1) was held to have the effect of “lock[ing] the investment income producing assets in a fund like the life fund into a fundamental purpose of a life insurance business, the payment of claims” (reasons, at page 2686). Subsection 81(1) reads as follows:
81. (1) Subject to subsection (2), where any company, in the exercise of its powers, combines such business with other classes of insurance business, it shall maintain separate and distinct accounts, funds and securities in respect of its life insurance business, and such funds and securities shall be available only for the protection of the holders of its policies of life insurance, and shall not be liable for the payment of claims arising from the other class or classes of business that the company transacts.
Having decided that all of the investment income derived from the assets of the life fund were to be treated as taxable income for the taxation years in question, the Tax Court Judge went on to conclude that it was unnecessary to deal with the jurisprudence outlining the tests to be applied when determining whether certain property or assets are used or held by a corporation in the course of carrying on a business. Specifically the taxpayer relied on the decision of the Supreme Court of Canada in Ensite Ltd. v. R., [1986] 2 S.C.R. 509 and the judgment of this Court in R. v. Marsh & McLennan, Limited, [1984] 1 F.C. 609(C.A.).
I pause here to note that the test outlined in the two cases noted above is expressed in terms of whether property is “employed and risked in the business”. In turn, for a property to be so employed and risked it is generally accepted that it must be integral to the continued operation of the business in question. For purposes of deciding this appeal, and for reasons which will be made evident, I shall continue to pursue the principal issue as stated at the outset in terms of whether all of the assets in the life fund were “necessary” to the taxpayer’s life insurance business.
In my respectful opinion, the learned Tax Court Judge erred in his findings. My analysis begins with a consideration of the effect of subsection 81(1) of the Insurance Act.
At the outset of the oral hearing, this Court expressed the view to counsel for the parties that regardless of what may be the true legal effect of subsection 81(1) of the Insurance Act it could not by any stretch of the imagination be deemed to impose a charge on assets contained within the life fund. Not even the most liberal of readings of that subsection could support such a position and I do not think that the Tax Court Judge intended to use the word “charge” as a term of art. In short, this is not a case where it is appropriate to introduce principles of real or personal property security law in order to resolve a tax issue. It is sufficient to speak of subsection 81(1) in terms of imposing a statutory condition or directive. The proper true question is whether that provision has the legal effect of “locking in” all of the assets held in the life fund as determined by the Tax Court Judge. The Minister recasts the question in terms of the allegation that a fraternal benefit society can only withdraw funds from a life fund with the approval of the Superintendent of Insurance and, therefore, until such time as permission is granted all of the assets must be deemed to be necessary to the life insurance business. In my opinion, subsection 81(1) cannot be construed in the manner advocated.
First, where a restriction on the transfer of funds is intended the Insurance Act clearly sets out the restriction. One need go no further than subsection 81(7) which restricts the transfer of segregated funds (e.g. RRSP monies) without the consent of the Superintendent. No similar provision exists for the transfer of unsegregated amounts in the life fund to other funds. Second, with respect to fraternal benefit societies the Superintendent merely requires a declaration from the society’s actuary that the assets of each fund are sufficient to provide for the payment of all of the obligations of the fund without deductions or abatement. This is provided for in subsection 95(2). Third, and furthermore, section 97 outlines the manner in which the Superintendent is to respond to a society which has unwisely transferred monies out of any one fund. Where it appears to the Superintendent that the assets in a fund are insufficient a special report is to be forwarded to the Minister who is entitled to request the society to make good the deficiency within a period not exceeding four years. A failure to comply results in forfeiture of status as a fraternal benefit society. This is a convenient place to reproduce subsections 81(7), 95(2) and section 97 of the Insurance Act:
81. …
(7) Where a separate and distinct fund with separate assets is maintained pursuant to subsection (6), the assets of the fund so maintained shall be available only to meet the liabilities arising under policies in respect of which the fund is maintained, except that
(a) amounts transferred to the separate and distinct fund from another fund of the company may, subject to the approval of the Superintendent, be withdrawn from the separate and distinct fund and transferred to such other fund as the directors may determine; and
(b) any assets remaining in the separate and distinct fund after the discharge of all of the company’s liabilities in respect of the policies for which the fund is maintained, may be transferred to such other fund as the directors may determine.
[A subsequent amendment to this subsection appears to eliminate the requirement that amounts transferred be approved by the Superintendent.]
…
95. …
(2) Such report shall include a declaration by the actuary that in his opinion the assets of the society applicable to each fund, taken at the value accepted by the Superintendent, together with the premiums, dues and other contributions to be thereafter received from the members according to the scale in force at the date of the valuation, are sufficient to provide for the payment at maturity of all the obligations of the fund without deductions or abatement.
…
97. (1) Where it appears to the Superintendent, from the annual statement filed with him or from any examination or valuation made in pursuance of this Act, that the assets of any fraternal benefit society registered under this Act, or of any benefit fund thereof, are insufficient to provide for the maturity of its policies without deduction or abatement or without increase of premiums or additional premiums, he shall make a special report to the Minister on the condition of the society and shall in such report state the amount of the deficiency in the society’s assets, shown by the annual statement or by any examination or valuation made as aforesaid; but, before so reporting to the Minister, the Superintendent may make a special valuation of the liabilities of the society under the said policies.
(2) Where the Minister after consideration of the special report concurs in the opinion of the Superintendent, he shall request the society within such time, not exceeding four years, as he may prescribe, to make good the deficiency.
(3) Where the society does not within the time so prescribed comply with the request of the Minister, the certificate of registry of the society may be withdrawn.
Reading the provisions of the Insurance Act collectively it is apparent that a decision to transfer assets out of a fund at any one time is a matter of judgment within the discretion of the fraternal society. No prior approval by the Superintendent is required as submitted by the Minister. This conclusion is reinforced by the reasons of MacKay J. in Lutheran Life, supra, which outline the true purpose of subsection 81(1) (at page 296):
Subsection 81(1) directs that a society combining operations of a life insurance business and another insurance business shall maintain separate accounts for its life insurance business, and funds and securities held in those accounts are to be available only for the protection of life insurance policy holders and are not liable for claims arising from other insurance operations carried on by the Society.
In concluding that subsection 81(1) does not have the legal effect imputed by the Tax Court Judge and the Minister, it remains to be decided whether the fact that the taxpayer elected not to withdraw surplus monies out of the life fund until 1988 is determinative of the outcome of this appeal. In my respectful opinion it is not.
As I understand it, the Minister’s alternative submission before both the Tax Court and this Court is that all of the assets retained within the life fund were committed to and employed in carrying on the taxpayer’s life insurance business and, therefore, were “risked” in that business. In the opinion of the Minister, this is so because the taxpayer permitted the assets to remain in the fund as a matter of business judgment. Specifically it is argued that the decision not to withdraw surplus monies from the life fund was based on concern over the possibility of AIDS-related claims. I shall turn to the AIDS issue first.
The taxpayer counters the above argument by maintaining that it did not turn its mind to the growing surplus in the life fund until 1988 when concerns were first expressed about the possibility of increased life insurance claims arising from the spread of AIDS. The taxpayer acknowledges that an additional $350,000 was set aside in 1988 to cover this extraordinary contingency. With respect to the Minister’s allegation that the AIDS concern was a factor with respect to all of the taxation years in question and not just in 1988, the taxpayer points out that in 1985 the threat of AIDS had not even been recognized or fully appreciated.
In my opinion, the evidence clearly supports the position of the taxpayer. Both the documentary evidence and transcript reveal that the taxpayer’s concern over the effect on its insurance business of the spread of AIDS did not surface until 1988 and not in the earlier taxation years: see Appeal Book, Vol. III, at page 390 and Appeal Book, Vol. II of Appendix, at page 294. The sole passage referred to us by counsel for the Minister cannot be construed as a contrary admission: see Appeal Book, Vol. II, at pages 300-301.
Even if I were to come to a different conclusion on the AIDS issue, the question would remain whether the decision of the taxpayer not to withdraw monies from the life fund is determinative of the issue as to whether all of the assets were necessary for purposes of carrying on the taxpayer’s life insurance business. In my opinion, the answer to that question must be framed in the negative.
I am prepared to accept that it would have been open to the Tax Court Judge to draw an inference from the fact that the taxpayer was content to leave all of the monies in the life fund until 1988. The inference would be that monies or assets left in the life fund are necessary for purposes of the taxpayer’s life insurance business. I would also be prepared to accept that a similar inference could be drawn from the fact that the financial statements submitted to the Superintendent of Insurance declare all of the investment income as related to the life fund. However, this is not the stance adopted by the Minister who argues simpliciter that the business decision not to remove assets from the life fund is determinative of the principal issue raised on the appeal.
In effect, the Minister’s argument hinges on the understanding that the sole basis for determining whether monies are necessary, that is to say risked and employed in the life insurance business, is to look at what the taxpayer did and not what it might have done. Both the Tax Court Judge and the Minister rely on a passage from Bronfman Trust v. The Queen, [1987] 1 S.C.R. 32, at page 55, in which Dickson C.J. states that “the courts must deal with what the taxpayer actually did, and not what he might have done”.
In my respectful opinion, the comments of Dickson C.J. in Bronfman are inapplicable to the present case for the reason that the former Chief Justice’s remarks are cited out of context. In Bronfman the taxpayer was seeking to deduct interest payments on a loan, the proceeds of which were used to make a capital allocation to a beneficiary under a trust. The trustees argued that had they sold off some of the trust’s assets to make the allocation and then turned around and repurchased the assets with a bank loan, the trust would have been entitled to deduct the interest payments on the loan. Therefore, the trustees maintained that the trust should be entitled to deduct the interest payments even if the trustees did not go so far as to structure the transaction in a manner which would allow for the deduction of interest. It is to this context that Dickson C.J. directed his remarks.
In my opinion, the comments of the former Chief Justice have no application to cases such as the one under consideration. The question before this Court is not whether the taxpayer could have withdrawn the surplus assets so as to avoid investment income being attributed to the life insurance business. Clearly, the taxpayer could have done so. Rather the question is whether all of the assets in the life fund were necessary to the life insurance component of the taxpayer’s business. The second question stands alone from the first.
I need go no further than two cases to find authority or support for the proposition that the conduct of, or business judgment exercised by, a taxpayer is not determinative of the issue of whether certain assets were necessary for purposes of carrying on a business. The first is a decision of Strayer J. (as he then was): McCutcheon Farms Ltd. v. M.N.R., [1991] 1 C.T.C. 50 (F.C.T.D.). The second is Lutheran Life referred to earlier in these reasons.
In McCutcheon the corporate taxpayer had received interest income with respect to substantial term deposits. The question to be decided was whether the interest income could be characterized as income from an active business. A positive response would have permitted the taxpayer to claim the “small business deduction” in respect thereof. Strayer J. concluded that since the taxpayer had never needed to redeem any of the term deposits in carrying on its business and had not provided sufficient evidence that even in times of emergency it would be required to do so the interest income could not be characterized as income from an active business.
In the present case, if I were to accept the Minister’s argument then I would also have to conclude that McCutcheon was wrongly decided on the basis that the corporate taxpayer’s decision to set aside certain funds in case of emergencies should have been determinative of the taxation issue. That is to say, one need go no further than the business decision taken by the taxpayer. In the end McCutcheon reaffirms the general understanding that the onus is on the taxpayer to establish what is or is not necessary for purposes of carrying on a business. On reflection it should be apparent that in effect the Minister is arguing for a subjective standard of “necessity” and not an objective one as established in McCutcheon. The objective standard was also applied by MacKay J. in Lutheran Life.
In Lutheran Life the taxpayer was a fraternal benefit society which extended life insurance, sickness and accident insurance, and certain fraternal benefits to its members. The society, however, only maintained one fund, a life fund. It argued that a portion of the assets that were credited to the life fund could be traced to contributions related to the fraternal operations of the society. Accordingly, the taxpayer argued that it was entitled to deduct from the investment income attributed to its life insurance business a portion of the investment income alleged to be attributable to the society’s fraternal assets. In its financial statements filed with the Superintendent of Insurance the taxpayer did not distinguish between investment income attributable to fraternal assets and investment income from the life insurance business. The distinction, however, was made in the income returns submitted to the Minister.
MacKay J. concluded that the Insurance Act did not impose an obligation on the taxpayer to establish a separate fund for fraternal activities. The failure to do so had the effect of placing the onus on the taxpayer to establish that the deduction from investment income of monies earned from assets for fraternal purposes other than insurance was not income from its life insurance business as reported. On the facts, MacKay J. concluded “that investment income attributable to fraternal activities other than insurance has not been proven for the purpose of exempting it from life insurance income subject to taxation” (at page 297).
In summary, neither subsection 81(1) of the Insurance Act nor the decision of the taxpayer not to withdraw assets out of the life fund is determinative of what assets are necessary for purposes of carrying on a life insurance business. The onus, however, rests on the taxpayer to establish what proportion of the investment income relates to the life insurance business. The fact that the taxpayer submitted financial documents to the Superintendent of Insurance allocating all of the investment income to the life fund and that the taxpayer declined to remove surplus assets from that fund are at best factors which support the inference that all of the assets held in the life fund were necessary for the life insurance operation. Such an inference, however, can be displaced by the taxpayer adducing persuasive evidence to the contrary.
At the hearing of the appeal a question was raised as to the appropriate remedy, as the Tax Court did not decide whether the taxpayer had met the onus of establishing that not all of the investment income was attributable to the life insurance business. Specifically, the Tax Court Judge made no findings as to the validity of the Thompson formula and the allocations made by the taxpayer as outlined at pages 2680-2682 of the Tax Court Judge’s decision. In my opinion, there is simply no need to remit the matter to the Tax Court Judge for further consideration. Let me explain.
From the outset and in the pleadings the Minister has taken the position that all of the investment income is taxable. Paragraph 14 of the Minister’s reply to the taxpayer’s notice of appeal makes this clear (Appeal Book, Vol. I, at page 14):
B. THE ISSUES TO BE DECIDED
14. The Respondent submits that the only issue to be decided is whether all of the income of the Appellant from the fund and securities which formed part of the Appellant’s life insurance fund in the 1985, 1986 and 1987 taxation years was income of the Appellant from carrying on its life insurance business, within the meaning of subsections 138 (1) and (2) of the Income Tax Act.
In support of his position the Minister has invoked only two legal arguments. Had this Court accepted either one then the decision of the Tax Court Judge would have to stand. But nowhere can I find any allegation by the Minister that the Thompson formula was flawed or, more importantly, that the taxpayer failed to meet its onus of establishing what portion of the investment income was attributable to the taxpayer’s insurance business. I agree with counsel for the taxpayer that the proper remedy is to remit the matter to the Minister on the understanding that the original assessments are valid. If the taxpayer remains intent on paying more tax for the 1985 to 1987 taxation years it may do so even though no legal obligation exists.
I would allow the appeal with costs here and below, set aside the judgment of the Tax Court of Canada dated June 21, 1995, allow the appeal from the reassessments and remit the matter to the Minister for reconsideration in a manner consistent with these reasons.
Strayer J.A.: I agree.
* * *
The following are the reasons for judgment rendered in English by
McDonald J.A. (dissenting): I have had the advantage of reading my brother Robertson J.A.’s reasons for judgment, and with respect, I must disagree. I am of the view that the Minister’s assessment for the taxation years 1984 through 1987 should stand.
I should note at the outset that I am in agreement with Robertson J.A.’s conclusion that subsection 81(1) of the Canadian and British Insurance Companies Act (hereinafter, the Insurance Act) does not affix a statutory charge on the assets held in the life fund. I also share my brother’s view that this finding is not determinative of the issue before this Court.
The appellant is a fraternal benefit society which carries on several activities on behalf of its members. One of these activities is the provision of life insurance through a life fund. Under the Income Tax Act (the Act), the income of a fraternal benefit society is generally exempt from Part I taxation. One exception to this general exemption is found in subsection 149(3) of the Act which states that the exemption from Part I taxation does not apply to taxable income which is acquired in carrying on a life insurance business.
The facts of this case disclose that the appellant retained more assets in its life insurance fund than may have actually been needed. The issue in this case is whether income from the “surplus” assets which the appellant identified with its life fund ought to have been included in computing the appellant’s income from its life insurance business for the taxation years 1985, 1986, and 1987.
The evidence in this case discloses that the taxpayer carried on three types of activities: the provision of life insurance, the provision of sickness and accident insurance, and the provision of fraternal benefits. Under subsection 81(1) of the Insurance Act, the taxpayer was required to maintain separate and distinct accounts in respect of its life insurance business. This taxpayer chose to establish three funds: the life fund, the accident fund, and the fraternal fund. The taxpayer’s board of directors decided in which funds assets were to be placed, and reported assets as belonging to a specific fund when the taxpayer filed its annual reports with the Superintendent of Insurance.
The board of directors of the taxpayer decided that its objective was “to provide a strong financial base thus ensuring in the event of unforeseen catastrophic economic event(s) occurring within the performing arts industry, the Society [would] be able to withstand such economic shocks and be in a position to continue to provide for [its] members’ needs” (letter dated October 5, 1988 from representative of the taxpayer to the Office of Superintendent of Financial Institutions, Appeal Book, at page 393). While this is a laudable objective, it is one which carries consequences: the taxpayer wilfully and knowingly maintained a surplus in the life fund in excess of what may have actually been required. In my view, the decision to keep these additional assets in the life account was a business decision by the taxpayer which should not be questioned by this Court. Once the decision had been made, the taxpayer benefitted from its decreased risk position. The taxpayer cannot now attempt to revisit the past and reassess its tax burden.
My brother Robertson J.A. seems to suggest that the existence of excess assets in the life fund raises a rebuttable presumption that the assets are risked or employed in the course of the business. Accepting this proposition for the moment, I have not been persuaded that the taxpayer has rebutted that presumption in this case. The insurance business is fraught with risks and uncertainties. While actuarial evidence may be helpful, the taxpayer in this case decided to maintain excess reserves “in the event of an unforeseen economic catastrophe.” I cannot see how the taxpayer can later attempt to remove these reserves from tax liability simply because the unforeseen catastrophe did not materialize.
The evidence in this case discloses that the taxpayer’s board of directors made conscious decisions about the allocation of assets into the life fund. There has been no suggestion that had these assets been needed in the event of some “unforeseen catastrophe” they would not have been used. How, then, can these assets not be subject to taxation? The taxpayer attempted to rely on the Federal Court Trial Division case of Lutheran Life Insurance Society of Canada v. Canada, [1991] 2 C.T.C. 284 (hereinafter Lutheran Life) and the Supreme Court of Canada case of Ensite Ltd. v. R., [1986] 2 S.C.R. 509 (hereinafter Ensite). I have considered the application of these two cases to the case at bar, and must, with respect, come to a different conclusion than that of Robertson J.A.
The case of Lutheran Life is, at first blush, quite similar to the case at bar. However, I am of the view that it is easily distinguishable on its facts. In Lutheran Life, the taxpayer had only one fund from which it carried out all of its business including its life insurance business. As a result, it was necessary to look into the purposes to which the assets of the fund were put, as the assets could have been used for any one of a number of purposes, of which the life insurance business was but one. As noted by the Trial Judge in that case, there was no requirement for the society to establish a separate fund for activities other than insurance, so failure to establish a separate fund could not be used as a consideration against the society.
In the case at bar, however, the taxpayer had established three separate funds: the life fund, the accident and disability fund, and the fraternal fund. Assets were deposited in each of the funds according to the decision of the taxpayer. There is no need to look beyond these allocation decisions into the individual funds as was required in Lutheran Life, as the taxpayer in this case had the opportunity to allocate its assets at the outset, or even to move assets from one fund to another at a later time. This division of assets by the taxpayer was a luxury not afforded to the Court in Lutheran Life, and in order to avoid unfair taxation of the taxpayer, the Trial Judge had to look within the fund to attempt to allocate assets after the fact. Here, there is no need for such speculation, as the assets were divided by the taxpayer according to its best judgment and after consultation with professionals in the accounting, actuarial, and legal fields.
It was further submitted by counsel for the taxpayer in this case that the Supreme Court of Canada decision in Ensite was determinative of the issue in this case. In Ensite, the Supreme Court of Canada considered whether interest earned by the taxpayer qualified as foreign investment income within the meaning of what was then subsection 129(4) of the Act for the purpose of the dividend refund under (then) subsection 129(1) of the Act. Wilson J. speaking for the Court, held that only assets which could be said to be “risked or employed” in the course of the taxpayer’s business could properly qualify for the dividend refund. Further, the “risk” contemplated had to be more than a remote risk. In the words of Wilson J. [at page 520]:
The threshold of the test is met when the withdrawal of the property would “have a decidedly destabilizing effect on the corporate operations themselves” …. The test is not whether the taxpayer was forced to use a particular property to do business; the test is whether the property was used to fulfil a requirement which had to be met in order to do business.
It should be noted that in reaching this conclusion, Wilson J. undertook an analysis of the underlying purpose of the legislative scheme and concluded that its purpose was to draw a distinction between active business income and other sources of income. Certainly, it is not contentious that in order to properly be considered active business income, interest must come from an asset that is risked or employed in the course of the business.
In this case, however, the issue is not whether income in the fund is properly considered “active business income” for the purposes of the Act. Instead, this Court is being asked to consider whether the assets in the life fund and the income derived from them are properly taxable as being a part of the life fund. The Act provides that the income from a fraternal benefit society is not taxable except where the income is from carrying on a life insurance business. Interest earned from assets held in the life fund, especially where the taxpayer has made a conscious decision to maintain those assets in that fund, is prima facie income from a life insurance business. In my view, this is not the same situation addressed by the Court in Ensite. The proper question is not whether the assets were risked or employed in the course of the business of the life fund, but instead whether the assets are properly considered part of the life fund.
The taxpayer in this case argued that the case is applicable because assets can only be considered “part of the life fund” when they are integrally risked in the life insurance business. Thus, the fact that assets are in the life fund and the taxpayer chose to put them in the life fund would not mean that the assets were actually part of the life fund. In effect, this proposed application of the Ensite test would allow the taxpayer to derive the benefit of extra security from having “surplus” assets in a taxable fund, and later escape taxation on those assets as long as actuarial evidence could establish that those assets were not integrally risked or employed. I do not accept this line of argument. I do not think it was the intention of the Court in Ensite to exempt portions of a life fund from taxation even where deliberate business decisions were made to keep specific assets in specific funds.
Even if I am in error and the reasoning in Ensite is applicable to this case, I am of the view that the test of whether assets are “risked or employed” in a business must necessarily be more flexible when that business is a life insurance fund. The insurance business necessarily has a different risk structure than most businesses. While actuaries may give risk estimates, unforeseen events may occur which may exceed even generous reserves. As I read the facts of this case, the taxpayer decided to maintain excess reserves in order to cover off some unforeseen contingency or contingencies. This was a reasonable position: as noted by the Tax Court Judge, at page 2684, the testimony in this case demonstrated that “there is nothing approaching mathematical certainty about what should be in a life fund”.
By choosing to maintain higher reserves in the life fund, the taxpayer fulfilled its stated goal of minimizing its risk position. The taxpayer expected to benefit and indeed did benefit from this decreased risk position. Its decreased risk position was beneficial not only for its own piece of mind, but that of the Superintendent of Insurance as well. In the words of one witness, as reported by the Tax Court Judge, at page 2684, “the greater the excess, the better people in the Superintendent’s office would sleep”.
Any number of events may have occurred which would have drawn upon the “excess” reserves. I do not believe it is proper to now look back in time to find that because such reserves were in excess of what was actually needed or what the actuarial evidence suggests may have been needed, they are not properly considered part of the life fund. Estimates of adequate reserves were available to the taxpayer at the time the taxpayer decided how best to allocate assets among the funds, and the taxpayer made a conscious decision to keep additional assets in the taxable life fund instead of moving them or allocating them to another fund.
In addition, I do not find it persuasive that the taxpayer is now able to say that the “surplus” funds could have been removed earlier than the taxpayer actually chose to remove them. The fact that the taxpayer could have moved the assets out of the life fund in the past does not lead inescapably to the conclusion that these assets were not ever part of the life fund. If anything, it supports the position that leaving the assets in the life fund was a business decision from which the taxpayer expected to derive some benefit. On this point, the words of Dickson C.J. in Bronfman Trust v. The Queen, [1987] 1 S.C.R. 32, at page 55 and cited by this Court on numerous occasions, are instructive:
… the courts must deal with what the taxpayer actually did, and not what he might have done: Matheson v. The Queen, 74 D.T.C. 6176 (F.C.T.D.), per Mahoney J.
While Robertson J.A. is correct when he points out that the facts of this case are different than those considered in Bronfman or indeed in Matheson [Matheson, J A v The Queen, [1974] CTC 186 (F.C.T.D.)], the general principle has been subsequently applied by this Court numerous times in many different fact situations. I believe that principle is applicable to this case; where the taxpayer has made an educated business decision to retain assets in a taxable fund, the Court should not look back with the benefit of hindsight to effectively remove assets from that fund with retroactive effect in order to put the taxpayer in a more advantageous tax position.
While I am sympathetic to the fact that the taxpayer may have shouldered a lesser tax burden had it not approached Revenue Canada, my sympathies do not change the facts of the case: this taxpayer chose to leave assets in the only fund administered by it which was subject to taxation. That is a decision which I cannot now question with the benefit of hindsight. The taxpayer cannot benefit from its decreased risk position at a time in the past and then seek to benefit again in the present by attempting to reduce its tax burden on the basis that the assets were not integrally “risked or employed” in the course of its life insurance business.
I am of the view that the decision of the Tax Court should be upheld.