COURT OF APPEAL FOR ONTARIO

CITATION: Jeffery v. London Life Insurance Company, 2014 ONCA 87

DATE: 20140203

DOCKET: C56665

Cronk, Blair and Strathy JJ.A.

BETWEEN

James Jeffery and D’Alton S. Rudd

Plaintiffs (Respondents)

and

London Life Insurance Company and The Great-West Life Assurance Company

Defendants (Appellants)

Proceeding Under the Class Proceedings Act, 1992, S.O. 1992, c.6

BETWEEN

John Douglas McKittrick

Plaintiff (Respondent)

and

The Great-West Life Assurance Company and Great-West Lifeco Inc.

Defendants (Appellant)

Proceeding Under the Class Proceedings Act, 1992, S.O. 1992, c.6

Sheila R. Block and Crawford G. Smith, for the Appellants

Paul J. Bates, Earl A. Cherniak, Q.C., David B. Williams, Jonathan J. Foreman and Robert L. Gain, for the Respondents

Heard: September 4, 2013

On appeal from the order of Justice Johanne N. Morissette of the Superior Court of Justice dated January 24, 2013, with reasons reported at 2013 ONSC 347.

The Court:

Background

[1] The genesis of these class proceedings lies in the 1997 acquisition of London Life Insurance Company by The Great-West Life Assurance Company. The members of the two plaintiff classes are holders of participating life insurance policies in the two companies. Their complaints center on an aspect of the acquisition known as the Participating Account Transactions (“PATs”).

[2] Participating insurance policies entitle the holders to share in the profits of life insurance companies. The companies are required to maintain accounts in respect of such policies separately from those maintained in respect of the companies’ other businesses. These participating insurance policy accounts are known as “PAR accounts”.

[3] In the case of the London Life acquisition it was anticipated that the merger would generate substantial savings resulting from the synergies of the transaction and the expected elimination of overlapping expenses. The PAR accounts would, as a result of the participating insurance policies’ profit-sharing qualities, benefit from those savings along with the regular shareholder accounts. Therefore, the PATs were developed as a mechanism to ensure that the PAR accounts also shared in the cost of obtaining those savings benefits. The PATs called for the payment of $220 million from the PAR accounts to help finance a portion of the acquisition in exchange for what were called pre-paid expense assets (“PPEAs”) in the same amount. The $220 million represented the present value of the anticipated expense savings to be attributed to the PAR accounts over a 25-year period (the PPEAs), using a discount rate of 6.91 per cent per annum. In other words, the PAR accounts were to receive a 6.91 per cent rate of return per annum on those savings as they flowed in. The $220 million purchase price – $180 million from the London Life Par accounts and $40 million from the Great-West Life Par accounts – was paid into the shareholder accounts (the “Share accounts”) and a deferred revenue liability was established in the Share accounts in the same amount to make up any difference if the merger expense savings turned out to be less than anticipated.

[4] The underlying purpose of the PATs and their elements are set out in the reasons of the trial judge following the first trial (the “First Trial Decision”) and in this Court’s reasons on appeal from that decision (“The First Appeal Decision”). They are reported, respectively, at Jeffery v. London Life Insurance Co., 2010 ONSC 4938, 74 B.L.R. (4th) 83, and Jeffery v. London Life Insurance Company, 2011 ONCA 683, 90 B.L.R. (4th) 1. For present purposes, it is only necessary to note the following.

[5] After a lengthy trial, Morissette J. ruled in favour of the class plaintiffs. She accepted their submissions that the PATs violated the Insurance Companies Act, S.C. 1991, c. 47 (the Act) in several ways. The nature of those breaches is not particularly important for present purposes. With one exception, this Court upheld her findings of breach[1]. It is one aspect of the remedy she crafted to rectify the breaches, and the rejection of that remedy by this Court, that led to the second hearing before the trial judge and, ultimately, to this appeal.

[6] Where there has been a breach of the Act, s. 1031, empowers a court to order that an insurance company comply with the Act. The trial judge interpreted this provision as enabling her not only to require that the $220 million plus interest (approximately $390 million) be returned to the PAR accounts, but also to create “litigation trusts” for the purposes of distributing those amounts directly to the participating policyholders – in effect, empowering her to order a monetary award of general damages payable to the individual class plaintiffs.

[7] This Court rejected that approach on the basis that it interpreted the power to order “compliance” too broadly. In place of the remedy chosen by the trial judge, the Court substituted an order unwinding the PATs as of an Effective Date to be determined (but approximating “the present” as opposed to the date of the acquisition transaction in 1997) and developed a formula for calculating the amount to be returned to the PAR Accounts. In doing so, this Court rejected the argument of the class plaintiffs that the PATs should be unwound as of 1997. On the basis of the “no contribution/no benefit” principle – accepted on appeal by the appellants – it also rejected the notion that the PAR accounts would have been entitled to the benefits of the merger even without a contribution to the cost of the merger. The Court then remitted the matter to the trial judge to determine the amounts according to the formula and to fix the Effective Date for the unwinding of the PATs.

[8] The trial judge did so, and released her second decision on January 24, 2013 (“the Second Trial Decision”). The appellants contend that she erred in interpreting the formula in arriving at that decision.

Analysis

[9] At the heart of this second appeal is the meaning to be attributed to paragraph 200 of the First Appeal Decision. Paragraph 200 provided as follows:

The remedy we impose calls for the PAR accounts to receive:

a) Their original contributions of $220 million;

Plus:

b) Forgone investment income to the date of trial in the amount of $172.7 million as calculated by the trial judge:

Plus:

c) A further amount of foregone investment income to the present, calculated on the same basis;

Less:

d) An amount representing the merger expense savings received by the PAR accounts to date (including, in the case of the London Life PAR account, the additional expense savings to date flowing from the ERA report, but not including the $27.1 million associated with the 2008 review[2]);

Plus:

e) An amount that represents a 6.91 percent return in relation to the merger expense savings received to date.

[10] The parties agreed on the amounts to be returned pursuant to paragraphs (a), (b) and (c) above. They were not able to agree on the remaining calculations, however, or on the appropriate Effective Date. These matters were dealt with by the trial judge. The argument centered on three questions:

(i) Were amortization charges in the amount of $139,437,000 to form part of the paragraph 200(d) calculation with the result that the amounts to be returned to the PAR accounts would be increased by that amount?

(ii) Should the amount set out in paragraph 200(e) be calculated on a before tax or after tax basis?

(iii) Does the paragraph 200 formula call for the amount in paragraph 200(e) to be added to the total formula amount rather than be added to the paragraph 200(d) amount (i.e., to express the formula in mathematical terms instead of in words, is the formula A + B + C – D + E rather than A + B + C - [D +E])?

(iv) What should the Effective Date be?

[11] The trial judge answered questions (i), (ii) and (iii) in the affirmative. She concluded that the appropriate Effective Date was December 31, 2011, but concluded, alternatively, that if she were mistaken about the deduction of the amortization expenses, the most appropriate Effective Date would be December 31, 2010. Respectfully, in our view, she misinterpreted the First Appeal Decision and the provisions of paragraph 200 in arriving at those conclusions.

The Purpose of the Decision to Unwind the PATs

[12] The purpose of this Court’s decision to unwind the PATs was to put the PAR accounts in the same position they would have been in had the PATs not occurred. In other words, the PAR accounts were to receive back their original investment of $220 million in full, together with the additional 6.91 per cent annual return on that investment as of the Effective Date; but they were required, at the same time, to account for any benefits they had received from the Par Account transactions since their inception in 1997 through the Effective Date.

[13] Paragraph 200 of the First Appeal Decision was designed to achieve that overall result. As the Court explained:

196. The PAR accounts have received a portion of the merger synergy expense savings over the years since the implementation of the PATs. Each year their share of the annual value of those savings has been credited to them and as a result the expenses in the PAR accounts have been reduced.

197. What this means is that if the PATs are to be unwound as of now, the monies to be returned to the PAR accounts by the appellants must be adjusted to account for the merger expense savings received by the accounts. Put another way, the PAR accounts are not entitled [to] get back all of their $220 million plus interest, but rather a discounted version of that amount to reflect the “purchase price” for the benefits already received prior to the date of unwinding. These benefits will include the additional expense savings identified by the ERA report that have flowed to the PAR accounts to date.

198. Based on the foregoing formula[3], the amount returned to the PAR accounts is to be reduced by the total merger expense savings received in the PAR accounts and to which the PAR accounts would be called upon to contribute, in order to give effect to the “no contribution/no benefit” principle and to ensure that the PATs are effectively unwound. [Emphasis in original.]

[14] The trial judge’s calculations do not effect this result, however.

The Amortization Issue

[15] Para. 200(d) of the First Appeal Decision directed that the following be deducted from the monies to be returned to the PAR accounts:

d) An amount representing the merger expense savings received by the PAR accounts to date (including, in the case of the London Life PAR account, the additional expense savings to date flowing from the ERA report, but not including the $27.1 million associated with the 2008 review)

[16] The purpose of this provision was to unwind the PATs as of the Effective Date, by putting the PAR accounts and the shareholder accounts in the same position in which they would have been in had the PATs not been implemented in 1997. The PAR accounts were to recover the full purchase price that they had paid for their assets ($220 million for the PPEAs), together with a rate of return to the Effective Date. But to unwind the PATs fully and put the PAR accounts in the position they would have been in had the PATs never been implemented, it was necessary for the PAR accounts to give credit for the actual amount of the merger expense savings they had received to that date. The parties agree that those merger expense savings amounted to $250,062,000.

[17] Accepting the submissions of the respondent class plaintiffs, however, the trial judge interpreted paragraph 200(d) as permitting her to apply an amortization factor to the “amount representing the merger expense savings received.” The amortization factor – a sum of $139,437,000 – represented the amount by which the asset purchased by the PAR accounts (the PPEAs) had been amortized over the years since the PATs were implemented. As a result, instead of ordering the PAR accounts to recognize the roughly $250 million in merger expense savings received under paragraph 200(d), the trial judge deducted the amortization amount from that sum and fixed the sum to be credited under paragraph 200(d) of the formula at $110,625,000. The effect of this approach was to increase the amount to be repaid to the PAR accounts by $139,437,000.

[18] The respondents submit that the trial judge was correct in taking the approach she did. They argued the following. The PATs were structured in a way that was designed to shift the benefit of the merger expense savings to the shareholders’ accounts for the first 25 years of the transaction, and this was done by offsetting the merger expense savings allocated to the PAR accounts by the annual amortization amounts expensed against the PPEAs. The amortization charges were “real, actual expenses of the PAR accounts each year”, and had the effect of neutralizing the impact of the merger expense savings allocated to the PAR accounts. Therefore, the argument goes, the PAR accounts had “contributed” the accumulated amortization charges of $139,437,000 to the merger expense savings.

[19] We recognize that, in the context of how the PATs were to operate, the PATs were structured to have the effect described above. There is really no dispute about that. As an expense item on the PAR accounts’ financial statements, the amortization charges served to reduce the surplus in those accounts, and they did so in a fashion designed to offset the effect of the revenue item stemming from the allocation of merger expense savings.[4] At the same time, there was a corresponding reduction in the deferred revenue liability established in the Share accounts, to the immediate advantage of the Share accounts.

[20] There was ample evidence to support the trial judge’s finding that the purpose of the PATs was to transfer to the Share accounts the benefits of the projected merger synergies for the first 25 years of the London Life acquisition transaction. Respectfully, however, the trial judge erred in moving from that finding to the conclusion – in the context of how the PATs were to be unwound – that she should deduct the amortization charges from the merger expense savings allocated to the PAR accounts on the basis that “PAR has contributed $139 million in amortization toward the $250 million savings to December 31^st, 2011”.

[21] We say this for several reasons.

[22] First, the trial judge incorrectly relied upon the evidence of Ms. O’Malley – one of the appellants’ experts at trial – who acknowledged that it was necessary for the amortization charges to be reversed, if the PATs were set aside, “in order to make PAR whole.” However, Ms. O’Malley’s evidence to that effect was premised on the PATs being unwound as of 1997 and not “as of the present,” which is what this Court ordered.

[23] Secondly, the trial judge misinterpreted this Court’s meaning in the passage from the First Appeal Decision (paras. 196-97) cited above, and in particular the statement:

Put another way, the PAR accounts are not entitled [to] get back all of their $220 million plus interest, but rather a discounted version of that amount to reflect the “purchase price” for the benefits already received prior to the date of unwinding.

[24] The trial judge interpreted this statement to mean that the PAR accounts were only required to account for a discounted portion of the merger expense savings received – hence the reduction for amortization. The foregoing statement must be read in the context of the entirety of paragraph 197 of the First Appeal Decision, however. What this Court was saying was that the PAR accounts were not entitled to receive the full amount of their initial $220 million payment plus interest because actual merger expense savings had been realized by them since the inception of the PATs. These merger expense savings included a return of part of the initial investment and a rate of return on that investment. As the Court noted in the First Appeal Decision (at paragraph 15):

In brief, the PATs involved a contribution by the PAR accounts of Great-West Life and London Life to the financing of the acquisition in exchange for PPEAs in the same amounts as the contributions plus a return on investment of 6.91 percent per annum. [Emphasis added.]

[25] To unwind the PATs effectively “as of the present” the PAR accounts had to repay all the benefits they had received. These included the portion of the merger expense savings attributable to the rate of return on the initial investment. It was in this sense that the PAR accounts were only entitled to receive a “discounted version” of the $220 million purchase price plus interest. They were not entitled to be credited with a “discounted version” of the savings received.

[26] Thirdly, while the trial judge properly recognized that the amortization charges were “real actual expenses of the PAR accounts each year”, she misunderstood the nature of those expenses in the context of the paragraph 200 formula, which is concerned with cash amounts. Amortization charges are “real actual expenses” for accounting purposes. But, they are a “non-cash” expense, not a “cash” expense. As the respondents’ expert, Mr. Thornton, testified, “there was no movement of cash [from the PAR accounts to the Share accounts], “just a movement of the numbers”. In short, the PAR accounts did not pay the Share accounts the amortized amounts. Amortization is simply an accountant’s way of recognizing the receipt of the merger expense savings and writing down the value of the PPEAs accordingly over time. The PAR accounts nonetheless received the cash benefit of the $250 million merger expense savings allocated to them.

[27] The appellants provide a simple example. Suppose a person subscribes to a magazine and pays the annual subscription price of $120 in advance for 12 monthly issues. The subscriber has a pre-paid expense asset (a magazine subscription), but no magazine at the outset. The publishing company has a matching deferred revenue liability, but cannot yet recognize the revenue. As the magazines are delivered each month, the publishing company’s deferred revenue liability is reduced by $10 as it recognizes the revenue received and the subscriber’s pre-paid expense asset is reduced by $10 to account for the fact that the magazine has been received. This does not mean that the subscriber has not received the magazine.

[28] The same may be said here. As the merger expense savings were received by the PAR accounts, the PPEAs were written down to reflect the receipt of the savings, and the deferred liability in the Share accounts was also reduced to reflect the decrease in that liability as a result of the payment of the merger expense savings. It does not mean that the PAR accounts did not receive those savings.

[29] Finally, the trial judge’s decision to apply an amortization factor to the paragraph 200(d) calculation would only make sense if the PATs had been ordered unwound as of 1997 or if the unwinding mechanism “as of the present” had only called upon the appellants to repay to the PAR accounts the reduced amortized value of the PPEAs as of the Effective Date. Neither was the case, however. To repeat, the PAR accounts are to receive the return of their full initial $220 million payment plus the 6.91 per cent rate of return to the Effective Date. The effect of that provision is to “fold” the amortized amount back into that recovery.

[30] The purpose of the paragraph 200 formula was to make things as if the PATs had never happened. Accordingly, while it is true that, for purposes of the way in which the PATs operated as structured, the annual amortization charges offset the annual merger expense savings allocated to the PAR accounts, it does not follow that the former should be deducted from the latter for purposes of unwinding the PATs. The PAR accounts are already to be reimbursed for the reduced amortization value of the PPEAs through the mechanism of requiring the appellants to repay their full initial payment (representing the value of the estimated expense savings discounted at 6.91 per cent), plus a similar rate of return on that purchase price in the intervening years through the Effective Date. To credit the PAR accounts with the amortization charges would result in the PAR accounts receiving a $139 million windfall and would be contrary to the “no contribution/no benefit” principle on which the transaction was premised and on which its unwinding is to proceed.

The Tax Issue

[31] The trial judge calculated the amount to be applied pursuant to paragraph 200(e) of the formula using a before-tax rate of 6.91 per cent. The appellants submit that the amount should be determined on the basis of the after-tax equivalent of a 6.91 per cent return (on their calculation, an after-tax effective rate of 4.35 per cent).

[32] We agree with the appellants on this point, for the reasons they advance. First, as the trial judge acknowledged, the paragraph 200(e) amount should be calculated in a manner that is consistent with the Mercer Report and the ERA studies[5]. It seems apparent from these documents that the PATs were structured overall on an after-tax basis. Secondly, as counsel for the respondents acknowledged during oral argument, the paragraph 200(a), (b) and (c) amounts – each agreed to by the parties – and the paragraph 200(d) amount, are all calculated on an after-tax basis. For purposes of consistency, the paragraph 200(e) amount should be determined on the same basis, in our view.

[33] We would allow this ground of appeal as well.

The Plus/Minus Issue

[34] For the sake of convenience, we reiterate the paragraph 200 formula:

The remedy we impose calls for the PAR accounts to receive:

a) Their original contributions of $220 million;

Plus:

b) Forgone investment income to the date of trial in the amount of $172.7 million as calculated by the trial judge:

Plus:

c) A further amount of foregone investment income to the present, calculated on the same basis;

Less:

d) An amount representing the merger expense savings received by the PAR accounts to date (including, in the case of the London Life PAR account, the additional expense savings to date flowing from the ERA report, but not including the $27.1 million associated with the 2008 review);

Plus:

e) An amount that represents a 6.91 percent return in relation to the merger expense savings received to date.

[35] Working on the premise that “plus” means “plus” and “minus” means “minus”, the trial judge concluded that the paragraph 200(e) amount should be added to the monies to be returned to the PAR accounts rather than subtracted from those amounts. Interpreting the paragraph 200 formula, and paragraph (e) in particular, in a way she described as “literally and purposively”, she determined that the formula should be applied, in mathematical terms, as A + B +C – D + E.

[36] Read “literally”, and without context, we agree that the formula could be read as the trial judge did. It could also be read literally as if the word “plus” before (e) were the word “and”. Respectfully, we do not agree that the formula can be read “literally and purposively” in the fashion adopted by the trial judge when paragraph 200(e) is taken in context. It is not a question of whether “plus” means “plus” and “minus” means “minus”. Undoubtedly, they do. Rather, the question concerns how the paragraph 200 formula is to be given effect in a way that conforms to the purpose and nature of the remedy crafted by this Court in the First Appeal Decision. Looked at from the latter perspective, the literal and purposive reading of the formula – expressed in mathematical terms – is not A + B + C – D + E; it is A + B+ C – [D + E].

[37] The interpretation given by the trial judge to the application of paragraph 200(e) misconceives what was to be balanced on each side of the “unwinding ledger”, and how that balancing was to be accomplished. It failed to give effect to the purpose, nature and intent of the remedy provided for in the First Appeal decision, as already discussed in these reasons.

The Nature of the PAT Investment in this Context

[38] In simple terms, the PAR accounts invested $220 million in 1997 to fund part of the Great West Life acquisition of London Life. What that investment “bought” them was the right to receive a stream of anticipated merger expense savings – the equivalent of a revenue stream – over the next 25 years. The estimated value of that stream of savings was expected to equate to the return of the original $220 million investment plus a 6.91 per cent annual rate of return on that investment.

[39] The parties agree that, over the years between the inception of the PATs and the Effective Date of the unwinding, merger expense savings in the amount of $250,062,000 have in fact been allocated to the PAR accounts. Those savings, as they came in, constituted a benefit to the PAR accounts that represented both reimbursement for a portion of the $220 million investment to date and the 6.91 per cent annual rate of return on that portion of the investment to date[6]. To repeat the observation in paragraph 15 of the First Appeal Decision, the PAR accounts’ investment was made “in exchange for PPEAs in the same amounts as the contributions plus a return on investment of 6.91 percent per annum.”

[40] How, then, to unwind the PATs as of the Effective Date?

The Remedy Created

[41] As stated earlier in these reasons, the purpose and intent of the remedy created by this Court was to ensure that the PAR accounts and the Share accounts were made whole – effective the date of the unwinding – as if the PATs had never been implemented. Neither was to be placed in a better position.

[42] For the unwinding to be accomplished successfully, the PAR accounts were to recover their entire initial investment of $220 million, plus a 6.91 per cent annual rate of return on that full investment to the Effective Date (one side of the “unwinding ledger”), and they were to credit back to the appellants all the benefits they had received from the transaction in the interim, again assuming a 6.91 per cent annual rate of return (the other side of the “unwinding ledger”).

Application of Paragraph 200(e)

[43] The PAR accounts are entitled to the return of their initial $220 million investment in full (paragraph 200(a)), “bumped up” to its value as of the Effective Date by applying a 6.91 per cent annual rate of return on that investment (paragraphs 200(b) + (c)). That is the “account to PAR” side of the unwinding ledger. But the PAR accounts are required to credit the appellants with the total of the merger expense savings received by them to the Effective Date. That is the “account by PAR” side of the ledger.

[44] Those benefits constituted a portion of the stream of savings the PAR accounts had purchased in 1997 for $220 million. As noted above, they incorporated both a partial recovery of the original $220 million “purchase price” and a partial payment of the 6.91 per cent annual rate of return that had been built into the PATs. This understanding is important to the interpretation exercise.

[45] Furthermore, in addition to having received a rate of return component as part of the merger expense savings allocated to them, the Par accounts have also enjoyed the benefit of those savings, as they came in, over the period during which the PATs have been operative. It follows that they must account for the rate of return they would have earned on the merger expense savings during the period from when each “piece” of the savings flowed in until the Effective Date in order to make the “account by PAR” side of the unwinding ledger complete. Paragraph 200(e) effectuates this obligation by “bumping up” the merger expense savings received to their value as of the Effective Date, much as paragraphs 200(b) and 200(c) do with respect to the amount to be recovered by the PAR accounts pursuant to paragraph 200(a).

[46] To put it another way, paragraphs 200(d) and (e), when combined, represent the Effective Date valuation of the merger expense savings allocated to the PAR accounts to that date. Therefore, their combined amount is to be deducted from the amounts to be repaid by the appellants to the PAR accounts, not added to them.

[47] This conclusion accords with the wording of paragraph 1(d) of this Court’s order of November 3, 2011, which requires the appellants to pay into the PAR accounts “the sum of $220 million, plus foregone investment income to the effective date (paragraphs 200 (a) + (b) + (c)) less an amount … in accordance with the formula set out in para. 200 of this Court’s reasons for decision” (i.e., less paragraphs (d) and (e), the only remaining portions of the paragraph 200 formula) (emphasis in original).

[48] The respondents submit that the 6.91 per cent annual rate of return went to the Share accounts instead of to the PAR accounts as a result of the way in which the PATs were structured. They therefore contend, and the trial judge agreed, that the paragraph 200(e) amount should be deducted from the amount of the actual merger expense savings (i.e., added to the amounts to be returned to the PAR accounts). For the reasons set out in paragraphs 26-28 above, rejecting the amortization argument, this argument must fail as well. The structure of the PATs did not mean that the PAR accounts did not receive the full $250 million in merger expense savings, including the rate of return.

[49] Were the respondents’ submission accepted, the PAR accounts would be recouping their 6.91 per cent rate of return threefold: first, as a component of the merger expense savings allocated; secondly, in the form of a rate of return on those merger expense savings to the Effective Date; and thirdly, as part of the return of their entire $220 million investment plus an additional rate of return on that investment to the Effective Date.

[50] The respondents’ interpretation might make sense if this Court had simply ordered the return of the original $220 million investment alone, with no interest to date. In that case, the merger expense savings received would have to be discounted to their 1997 value – i.e., paragraph 200(e) subtracted from paragraph 200(d) – in recognition of the fact that the PAR accounts had received no rate of return on their investment, which was a part of what they had “purchased” in 1997. That is not what the Court ordered, however. The original investment of $220 million is to be returned to the PAR accounts at its Effective Date value. So, too, must the merger expense savings received (including, as they do, a rate of return component) be “bumped up” to their Effective Date value.

[51] In paragraph 201 of the First Appeal Decision this Court observed that “[t]he 6.91 percent factor referred to in [paragraph 200(e)] is reflective of the fact that the PAR accounts were not required to pay 100 cents on the dollar for the benefits received.” The trial judge and the respondents point to this observation in support of the view that the PAR accounts are entitled to retain the benefit of the paragraph 200(e) rate of return rather than the reverse. We do not think the comment supports this contention. It merely recognizes that the PAR accounts paid $220 million in 1997 for a stream of expense savings worth more than that (enough more to generate a 6.91 per cent annual rate of return). In that sense, the PAR accounts were not required to pay 100 cents on the dollar for the benefits received. In the paragraph 200(e) context, the comment simply recognizes that some recovery of the rate of return has already been incorporated into the merger expense savings realized – the PAR accounts did not pay 100 cents on the dollar for those savings – and therefore must be accounted for in the application of the paragraph 200(e) rate of return.

[52] In a further comment in paragraph 202 of the First Appeal Decision, this Court stated that “[t]he amount returned to the PAR accounts would still be reduced by the total expense savings (less the 6.91 per cent factor referred to above.” This comment was made in a somewhat different context, but to the extent the comment suggests something contrary to the foregoing, this Court simply misspoke. Given the reasoning regarding the remedy as a whole, this Court’s meaning might have been more accurately expressed by using the word “and” rather than the word “less” in the portion of paragraph 202 cited above.

[53] For the reasons outlined above, we conclude that the interpretation accepted by the trial judge regarding the manner in which paragraph 200(e) is to be integrated into the operation of the paragraph 200 formula is not in accord with the purpose and intent of the remedy developed by this Court and does not give effect to the wording of paragraph 1(d) of this Court’s order of November 3, 2011, properly construed. Interpreted correctly, paragraph 200(e) requires the amount calculated under it to be subtracted from the amount to be transferred to the PAR accounts.

The Effective Date

[54] The decision as to what date should be chosen as the Effective Date of the unwinding was referred to the trial judge by the First Appeal Decision. She chose December 31, 2011, as opposed to December 31, 2010, on the basis that the former was more convenient for the paragraph 200 calculations and closer in time to this Court’s decision. However, her opinion was that, in the event that she was mistaken about the deduction of amortization expenses in determining the paragraph 200(d) amount, the appropriate Effective Date should be December 31, 2010.

[55] We conclude that the appropriate Effective Date is December 31, 2011, for the reasons expressed by the trial judge, regardless of the amortization issue. As discussed above, the notion of amortization is not relevant to the paragraph 200(d) calculation; it therefore should not affect the choice of an appropriate date for the unwinding.

The Stare Decisis Argument

[56] The respondents argued forcefully that this Court is bound by the First Appeal Decision on a stare decisis basis and therefore has no ability to interfere with the decisions made by the trial judge based on the evidence before her.

[57] We would not give effect to this submission. The issue is not whether the First Appeal Decision is binding on us. Clearly it is. The issue is the correct interpretation of that binding decision. For the reasons we have articulated, the trial judge’s interpretation was flawed because it failed to give effect to the language of paragraph 1(d) of this Court’s order of November 3, 2011.

Disposition

[58] We would accordingly allow the appeal and order as follows:

(i) Paragraphs 2 and 3 of the trial judge’s order dated January 24, 2013 are set aside;

(ii) The amount to be repaid by the appellants to the PAR accounts resulting from the formula set out in paragraph 200 of the First Appeal Decision is $51.6 million, calculated on the following basis:

(a)+(b)+(c)

(a) Original contribution

$220 million

(b) Forgone investment income to date of trial (as determined by the trial judge)

$109.7 million

(c) Further amount of forgone interest to the Effective Date, calculated on the same basis

$42.1 million

(a) + (b) + (c) total

$371.8 Million

LESS (d) + (e)

(d) An amount representing the merger expense savings received by the PAR accounts to December 31, 2011

$250.1 million

(e) An amount representing a 6.91 per cent rate of return in relation to the merger expense savings received to date to December 31, 2011

$70.1 million

(d) + (e) total

$320.2 million

Paragraph 200 Amount to be Repaid to the PAR Accounts as of December 31, 2011: $371.8 - $320.2

$51.6 million

(iii) The amounts referred to in paragraph (ii) above shall be updated from December 31, 2011 to March 31, 2014 (the yearly quarter immediately following the release of this decision), unless the parties agree or this Court orders otherwise.

[59] Counsel advised that in all likelihood they will be able to agree upon costs once this decision has been released, whatever the outcome. If that is not the case, they may contact the Registrar of this Court to schedule the filing of brief written submissions as to costs for the panel’s consideration.

“E.A. Cronk J.A.”

“R.A. Blair J.A.”

“G.R. Strathy J.A.”

Released: February 03, 2014

[1] The one exception related to her finding that the directors of the companies were individually liable under s. 166(2) of the Act. This Court set aside that finding.

[2] This clarification referred to an aspect of the transaction specific to London Life and to a report detailing the results of an Expense Rating Adjustment (“ERA”) done earlier that revealed that the merger expense savings were greater than had been anticipated at the time of the merger. The ERA report is not an issue on this appeal.

[3] That is, the paragraph 200 formula.

[4] The fact that the actual amortization amount does not equate to the actual expense savings allocated to the PAR accounts is a function of timing. The merger expense savings were realized earlier, and in greater amounts, than initially anticipated.

[5] The Mercer Report was an independent actuarial opinion on the fairness of certain aspects of the financing of the London Life acquisition, including the PATs, that was procured by Great-West Life at the request of the Office of the Superintendent of Financial Institutions (“OSFI”), and which OSFI considered as part of its review of the PATs. As noted above, the ERA report discussed unanticipated merger expense savings from the London Life acquisition.

[6] The reason these benefits do not constitute the full stream of savings attributable to the initial purchase amount or the full amount of the savings attributable to the rate of return is that, as the PATs were structured, these savings were structured over a 25-year period. Because the Effective Date of the unwinding is December 31, 2011, only a portion of the benefits were received by the PAR accounts and must be returned in order to complete the unwinding of the PATs.