Moore v. Sweet

Ontario Reports

Court of Appeal for Ontario

Strathy C.J.O., Blair and Lauwers JJ.A.

March 2, 2017

134 O.R. (3d) 721 | 2017 ONCA 182

Case Summary

Equity — Unjust enrichment — Deceased and applicant entering into post-separation oral agreement that deceased would maintain applicant as beneficiary of his life insurance policy and that applicant would continue to pay premiums — Deceased and respondent forming relationship and living together for 13 years until deceased's death — Deceased revoking designation of applicant as beneficiary and designating respondent as irrevocable beneficiary under policy — Applicant not advised of change and continuing to pay premiums — Application judge finding that applicant was entitled to proceeds of policy as oral agreement amounted to equitable assignment by deceased to applicant of his equitable interest in proceeds — Application judge erring in relying on doctrine of equitable assignment as that doctrine was not pleaded or argued — Respondent not unjustly enriched as irrevocable beneficiary designation provisions of Insurance Act provided valid juristic reason for her receipt of insurance proceeds — Circumstances of this case not providing basis for "good conscience" constructive trust in favour of applicant — Insurance Act, R.S.O. 1990, c. I.8.

The applicant was the named beneficiary of the deceased's life insurance policy during their lengthy marriage. After they separated, the applicant and the deceased entered into an oral agreement that the deceased would maintain the applicant as his beneficiary and that the applicant would continue to pay the premiums. The deceased formed a relationship with the respondent after separating from the applicant, and they lived together in the respondent's apartment until the deceased's death 13 years later. Both had disabilities, and they were supportive of each other. The deceased executed a change of beneficiary form and made the respondent the irrevocable beneficiary under the policy. He did not advise the applicant of the change, and she continued to pay the premiums. After the deceased's death, the applicant claimed entitlement to the insurance proceeds. She asserted that the respondent would be unjustly enriched if she received the proceeds, asked the court to impose a constructive trust on the proceeds in her favour and invoked the doctrine of unjust enrichment. The application judge ruled in her favour. He found that the oral agreement between the deceased and the applicant took the form of an equitable assignment to the applicant of the deceased's equitable interest in the proceeds, or alternatively, of his entire interest in the policy, in return for her agreement to pay the premiums in the future. The respondent appealed.

Held, the appeal should be allowed.

Per Blair J.A. (Strathy J.A. concurring): The application judge erred in relying on the doctrine of equitable assignment, as that doctrine was neither placed in issue nor argued before him. Because the record was not created on the basis that the parties had joined issue on the ground of equitable assignment, the application judge's findings with respect to it were unreliable and had to be set aside. Absent equitable assignment or another exceptional circumstance to a similar effect, the irrevocable beneficiary designation provisions of the Insurance Act operated to provide a valid juristic reason for the respondent's receipt of the insurance proceeds, making a finding of unjust enrichment unavailable. There was nothing in the circumstances of this case that would provide the basis for a "good conscience" constructive trust when the facts did not support such a trust based on unjust enrichment or wrongful act, and it was therefore unnecessary to determine whether such a third category of remedial constructive trust continues to be available in Canada.

Per Lauwers J.A. (dissenting): The Supreme Court of Canada in Soulos v. Korkontzilas did not limit the remedy of a constructive trust to the two defined situations of unjust enrichment and wrongful acts and did not otherwise abolish the doctrine of good conscience constructive trusts as a general source of equitable jurisdiction. The application judge did not err in finding that the applicant had made out her claim in unjust enrichment.

Overview

[1] The issue on this appeal is whether the appellant, Risa Lorraine Sweet, or the respondent, Michelle Constance Moore, is entitled to the proceeds of a policy of insurance on the life of Lawrence Anthony Moore. The proceeds of the policy have been paid into court pending resolution of these competing claims.

[2] The Moores were married for more than 20 years and Ms. Moore was the named beneficiary of the policy during their marriage. She was not designated as an irrevocable beneficiary, however. Following the separation, Mr. Moore established a relationship with Ms. Sweet. They lived together in Ms. Sweet's apartment until his death 13 years later. Shortly after Mr. Moore and Ms. Sweet began living together -- and contrary to an oral agreement he had with Ms. Moore, who continued to pay the premiums -- Mr. Moore revoked the designation of Ms. Moore as beneficiary and executed a change in beneficiary form designating Ms. Sweet as the irrevocable beneficiary under the policy. The change of beneficiary was properly recorded with the insurer.

[3] The application judge ruled in favour of Ms. Moore. He held that the policy proceeds of $250,000 plus interest were held in trust and that Ms. Moore was entitled to recover them on the basis of unjust enrichment. Ms. Sweet appeals.

[4] For the reasons that follow, I would allow the appeal.

Factual Background

[5] Ms. Moore and Mr. Moore were married in 1979. They have three children together, born in 1983, 1984 and 1988. Although they were minors at the time the oral agreement was made, all were adults by the time of Mr. Moore's death in 2013.

[6] In October 1985, Mr. Moore applied for and obtained a policy of insurance on his life in the principal amount of $250,000 from a predecessor of RBC Life Insurance Company. He was named as the owner of the policy. Ms. Moore was named as a beneficiary, but not as an irrevocable beneficiary. Annual premiums of $507.50 were paid from the Moores' joint account prior to their separation. Thereafter, the premiums were paid by Ms. Moore alone. The policy remained in effect until Mr. Moore's death on June 20, 2013. His estate has no significant assets.

[7] According to Ms. Moore, "[t]he Policy was entered into due to the Deceased's financial irresponsibility and the understanding the Deceased and [she] shared, which was that the Policy was required in order to support [their] Children in the event of the Deceased's passing".

[8] The Moores separated in December 1999. Ms. Moore says that shortly thereafter the deceased told her and the children "that he would ensure we were taken care of". Without stating whether the agreement occurred at the time the policy was taken out, or at the time of separation, or on some other occasion, she deposed that

It was expressly agreed between the Deceased and I that in paying for the Policy premiums, I would be entitled to receive the Policy benefits upon his death.

[9] On the basis of this evidence, the application judge found that there was an oral agreement between the Moores "made in 2000 to the effect that [Ms. Moore] would pay the premiums and be entitled to the proceeds of the Policy on [Mr. Moore's] death". This finding is not contested on appeal.

[10] Mr. Moore and Ms. Sweet met when they were both attending the Donwood Institute, an addiction treatment centre. He was being treated at the time for a history of drug, alcohol and substance abuse problems and mental health issues. She also had a history of medical issues including substance abuse. They became supportive of each other and soon developed a close relationship.

[11] The application judge found that Ms. Sweet and Mr. Moore began living together in the summer of 2000. However, according to Ms. Sweet and her sister, Ann McAdam, it was sometime in 1998 after they met at the Donwood Institute. In any event, they lived together in Ms. Sweet's apartment until his death.

[12] Contrary to his agreement with Ms. Moore, Mr. Moore changed the beneficiary under the policy in September 2000 from Ms. Moore to Ms. Sweet by executing a change of beneficiary form that designated Ms. Sweet as an irrevocable beneficiary under the policy pursuant to the Insurance Act, R.S.O. 1990, c. I.8. He filed the form with the insurer, which recorded the change of beneficiary a few days later. He did not advise Ms. Moore of this change and she continued to pay the policy premiums.

[13] That said, the change in beneficiary was not carried out surreptitiously without consultation. It was effected through the offices of, and after discussions with, Ms. Moore's brother-in-law, a life insurance broker who is married to Ms. Moore's sister. The sister also works for a life insurance company.

[14] According to Ms. Sweet, who was present when the decision was made at the broker's office, the rationale for the change in beneficiary was that

Mr. Moore did not want me to worry about how I was going to pay the rent or buy my medications in the unlikely event he passed away before me. He wanted to make sure I was able to live my remaining years in the building where I have resided for the past 40 years. And he wanted me to live worry and debt free.

[15] In May 2002 -- some time after the oral agreement and the subsequent change in beneficiary status -- the Moores entered into a separation agreement. Mr. Moore agreed to convey his one-half interest in the matrimonial home to Ms. Moore. Each waived and released the other from all equalization of net family property claims (with a minor exception regarding household contents). Mr. Moore agreed to pay child support in accordance with the Federal Guidelines. Ms. Moore waived and released outstanding child support arrears.

[16] The separation agreement was silent about the policy or anything related to it.

[17] The Moores were divorced in October 2003.

[18] It is evident from the materials that the marriage breakdown between the Moores was related to Mr. Moore's struggles with chronic pain and his alcohol and substance abuse issues, which no doubt contributed to what Ms. Moore refers to as his financial irresponsibility, and build up of debts burdening them both. They clearly went through a very difficult time. In fact, the financial difficulties -- including a very substantial debt of Mr. Moore to Revenue Canada of over $70,000 -- led to both declaring bankruptcy in early 2000.

[19] Between 1999 and 2000, Mr. Moore lost his driver's licence and his job, for medical reasons, and for the rest of his life his source of income was a long-term disability pension from his former employer in the gross amount of $5,000 per month (with some indexing). We do not know much about Ms. Moore's financial circumstances post-separation, except that it appears she continues to reside in the former matrimonial home in Mississauga. We also know that there were disputes post-separation about the payment by Mr. Moore of child support, that arrears accumulated and that beginning in 2005 the Family Responsibility Office ("FRO") was garnishing his disability payment. After an additional deduction for income tax, Mr. Moore was left with a net disability amount of $2,125 per month, with the latest information available.

[20] Ms. Moore says that Mr. Moore made an offer to settle the arrears of child support in January 2005, but that arrears in the amount of over $10,000 remained at the time of his death. There is some dispute in the materials, however, over whether this amount has been paid.

[21] Ms. Sweet and Mr. Moore cared for each other during their 13 years of cohabitation. She is disabled as well. She suffers from chronic pain, is on numerous medications, requires the use of a walker and cannot travel by public transit. Her sister, Ms. McAdam, has been given power of attorney over her personal property. According to Ms. Sweet, Mr. Moore "helped out" with the rent and assisted with chores and running errands for her; when Mr. Moore's disability income was garnished by FRO, it caused them financial hardship and led to them borrowing from friends and relatives to make ends meet.

The Issues

[22] As noted, Ms. Sweet does not contest the application judge's finding that there was an oral agreement between Ms. Moore, the respondent and her former husband, the deceased Larry Moore, to the effect that Ms. Moore would continue to pay the premiums on the policy and would be entitled to the proceeds on his death. Nor is it disputed that she paid the premiums and that he did not tell her of the change in beneficiary.

[23] The two central issues raised on the appeal relate to equitable assignment and unjust enrichment, although Ms. Moore's end position is that a constructive trust in her favour is justified on the basis of "good conscience".

[24] In the result, there are four questions for consideration:

(i) Did the application judge err in relying on the doctrine of equitable assignment, which was not pleaded or argued before him?

(ii) If he were entitled to rely on the doctrine, did the application judge err in finding that the contract here constituted an equitable assignment of the deceased's interest in the policy proceeds?

(iii) Did the application judge err in failing to hold that there was a valid juristic reason for Ms. Sweet's receipt of the policy proceeds and therefore in holding that the proceeds were impressed with a trust in Ms. Moore's favour based on unjust enrichment?

(iv) Is there some other "good conscience" basis for the creation of a remedial constructive trust in Ms. Moore's favour?

[25] For the reasons that follow, my answer to questions (i), (ii) and (iii) is yes, and to (iv) is no.

[26] In summary, I am satisfied that

(i) because it was neither placed in issue nor argued before him, the application judge erred in relying on the doctrine of equitable assignment and applying it on his own initiative;

(ii) because the record was not created on the basis that the parties had joined issue on the ground of equitable assignment, the application judge's findings with respect to it are unreliable and must therefore be set aside;

(iii) absent equitable assignment or another exceptional circumstance to a similar effect, the provisions of the Insurance Act, pursuant to which Ms. Sweet was designated an irrevocable beneficiary, operate to provide a valid juristic reason for her receipt of the insurance proceeds, making a finding of unjust enrichment unavailable; and

(iv) there is nothing in the circumstances of this case that would provide the basis for a "good conscience" constructive trust when the facts do not support such a trust based on unjust enrichment or wrongful act, and it is therefore unnecessary to determine whether such a third category of remedial constructive trust continues to be available in view of the Supreme Court of Canada's decision in Soulos v. Korkontzilas.

Discussion and Analysis

Equitable Assignment

[27] The application judge held that the oral contract described took the form of an equitable assignment to Ms. Moore of Mr. Moore's equitable interest in the proceeds of the policy, or alternatively, of his entire interest in the policy, in return for her agreement to pay the premiums in the future.

[28] Respectfully, he erred in doing so. I say this for two reasons.

[29] First, it was procedurally unfair to found his decision on a principle that was neither put in play nor relied on by Ms. Moore on the application, and on which the parties had not joined issue.

[30] Second, in addition to the fairness concerns arising from the foregoing, the introduction of a new theory of liability in the reasons for decision raises concerns about the inherent unreliability of any finding based on that untested theory: see Rodaro v. Royal Bank of Canada, at paras. 60-63; Labatt Brewing Co. v. NHL Enterprises Canada, at paras. 5-7. In respect of this analysis, it is not a matter of determining whether the application judge made a palpable and overriding error in his findings of fact or of mixed fact and law regarding the assignment. By introducing the equitable assignment issue himself in the course of arriving at his decision, he erred in law by depriving the parties -- and Ms. Sweet in particular -- of the ability to make a proper record relating to it, thereby rendering the result unreliable and justifying appellate intervention.

Procedural Fairness

[31] Equitable assignment was not before the court on the application. Ms. Sweet had no notice that it was in play. It was not asserted in the notice of application. It was not "pleaded" in any other way or raised as such in the evidence presented or other materials filed by Ms. Moore. It was not addressed or in any way argued by counsel on her behalf. Not surprisingly, it was not met or addressed by Ms. Sweet either.

[32] Ms. Moore's position was essentially that (i) there was a contract and (ii) she was entitled to the funds based on the equitable principle of unjust enrichment. The concept of equitable assignment was introduced by the application judge in his reasons for decision.

[33] None of this is contested by Ms. Moore on the appeal. She argues, however, that there was no procedural unfairness because (i) the facts "are capable of" supporting an assignment and (ii) the agreement itself entailed an equitable assignment of the policy.

[34] I disagree.

[35] I shall return to these submissions below when dealing with the substantive issues relating to equitable assignment. Suffice it to say here that neither addresses the procedural unfairness created by resting the decision, at least in significant part, on that doctrine. Indeed, as I shall explain, the concerns are reinforced by the factors raised by those submissions.

[36] Equitable assignment is not a straightforward concept to understand or apply, procedurally or substantively, even when a party is represented by counsel. It raises many questions. Here are some, for example. Does the transaction in question constitute an equitable assignment of a "legal" right of action? Or is it an equitable assignment of an "equitable" right of action? Are there procedural implications arising out of the difference? Would Ms. Moore, as the equitable assignee, be required to add Mr. Moore's estate as a co-applicant (something she did not do)? What evidence is required to establish the assignor's intention to assign the property interest or right of action in question? What exactly is it that was intended to be assigned -- here, the policy in its entirety or the right to the proceeds? Was the assignment to be absolute or something else? Was there consideration? What form of consideration is necessary?

[37] None of this was touched upon in Ms. Moore's application materials or in argument.

[38] Ms. Sweet was unrepresented at the hearing. She is disabled. Although the Public Guardian and Trustee was satisfied that she was capable of representing herself at the hearing, she requires her sister, Ms. McAdam, to act as her power of attorney in respect of her personal property. Where is the appearance of fairness when a party in such circumstances learns for the first time, on receipt of the decision, that she has lost what she thought was her entitlement to $250,000 needed to sustain her for the rest of her life, on a basis that she had no way of knowing would affect the outcome or that she had to meet? To pose the question, in my opinion, is to answer it.

[39] This court has made it clear on a number of occasions that lawsuits are to be decided within the boundaries of the pleadings (i.e., the documents framing the issues), and based on findings and conclusions that are "anchored in the pleadings, evidence, positions or submissions of any of the parties". Otherwise, they are "inherently unreliable" and "procedurally unfair, or contrary to natural justice". When a judge steps outside of the case as it was "developed by the parties" to decide a given issue, the parties are deprived of the opportunity to make submissions and to "address that issue in the evidence". See Rodaro, at paras. 60-63; Labatt Brewing, at paras. 5-7; 460635 Ontario Ltd. v. 1002953 Ontario Inc., at para. 9; A-C-H International Inc. v. Royal Bank of Canada.

[40] The circumstances of this case affirm those very risks and lead to the second concern, the inherent unreliability of the findings.

The Inherently Unreliable Findings

[41] With respect to the oral agreement and equitable assignment, the application judge held that

(i) each of Mr. and Ms. Moore had an equitable interest in the proceeds of the policy from the time it was taken out in 1985;

(ii) the oral agreement in 2000 took the form of an equitable assignment by Mr. Moore to Ms. Moore of his equitable interest in the proceeds in return for her agreement to pay the ongoing premiums;

(iii) if he were wrong in finding that Ms. Moore had an equitable interest in the proceeds of the policy from 1985, with the result that Mr. Moore had the entire equitable interest in the proceeds until 2000, the effect of the 2000 oral agreement was to assign his entire interest in the policy to her at that time, in exchange for her commitment to pay the ongoing premiums.

[42] The bases for these conclusions are set out in the application judge's reasons, at paras. 18-23. In summary, he found that

(i) the Moores "jointly" took out the policy in 1985 to provide for Ms. Moore and the children in the event of his death and, because the premiums were paid out of their joint account, they each had an equitable interest in the proceeds which "[a]s a practical matter . . . took the form of a right to determine the beneficiary of the Policy" (para. 18);

(ii) the agreement had a significant, continuing purpose when it was made, namely, allowing Mr. Moore to fulfill his obligations to his children and provide for Ms. Moore (para. 20);

(iii) Ms. Moore paid the premiums until Mr. Moore's death and would not have done so if she had not obtained an equitable assignment of the insurance proceeds (para. 21);

(iv) Mr. Moore had no assets in 2000 and relied on disability benefits to cover his living expenses. He was aware the policy was kept in force by virtue of Ms. Moore paying the premiums and "could not reasonably have expected [Ms. Moore] to continue to pay the premiums on the Policy unless it was for the benefit of [her] and their children" (para. 22); and

(v) the 2002 separation agreement was silent with respect to the policy, even though both parties were aware of its existence. "The most probable explanation for the silence is that the parties had already reached, and implemented, the agreement respecting the Policy" (para. 23).

[43] In Fraser v. Imperial Bank of Canada, Davies J. (concurring) stated that

No form of words is necessary to create [an equitable] assignment. It is always a question of fact and of the intention of the parties to be gathered from what they said and did and from all the surrounding circumstances.

[44] It may be that, on a record properly developed on the basis that the parties had joined issue on the question of equitable assignment, the foregoing findings would have been open to the application judge. But the parties did not join issue on this question and, with respect to the application judge, many questions needed to be addressed and remained open and unanswered on the record. As this court stated in Rodaro, at paras. 62-63:

In addition to fairness concerns which standing alone would warrant appellate intervention, the introduction of a new theory of liability in the reasons for judgment also raises concerns about the reliability of that theory. We rely on the adversarial process to get at the truth. That process assumes that the truth best emerges after a full and vigorous competition amongst the various opposing parties. A theory of liability that emerges for the first time in the reasons for judgment is never tested in the crucible of the adversarial process. We simply do not know how [the trial judge's theory] would have held up had it been subject to the rigours of the adversarial process[.]

[The trial judge] erred in finding liability on a theory never pleaded and with respect to which battle was never joined at trial. This error alone requires reversal.

[45] Consider the following areas of potential controversy, for example.

[46] First, what exactly were the intentions of Mr. Moore and Ms. Moore at the time the oral agreement was entered into, or at the time the policy was taken out or at any time between? The only evidence presented on the application was contained in three affidavits and attached exhibits. There was no cross-examination on the affidavits. From these materials, there is nothing to indicate that Mr. Moore and Ms. Moore put their minds to the concept of an assignment of the policy or of its proceeds. There is no evidence of what words were actually spoken between the Moores to express Mr. Moore's intention to assign or, if there were, to express whether the purported assignment was to be absolute or conditional (something that could affect the priorities in a subsequent dispute over the proceeds between Ms. Moore, as assignee, and any subsequent assignee or beneficiary without notice of the earlier assignment). Unlike unjust enrichment, a finding of an assignment requires evidence of intention. See, generally, Fraser v. Imperial Bank of Canada; Bank of Nova Scotia v. Newfoundland Rebar Co., at para. 12; Chudy v. Merchant Law Group, at paras. 41 and 47; Nadeau v. Caparelli, at para. 19.

[47] Counsel for Ms. Moore did not argue that she had an equitable interest in the proceeds of the policy from its inception. Nor did he submit that Mr. Moore had assigned his equitable interest in the policy entirely, or at all, pursuant to the 2000 oral agreement, or that Ms. Moore's equitable interest took the form of Ms. Moore's right to determine the beneficiary of the policy.

[48] Ms. Moore did not state that the policy had been taken out "jointly", as the application judge finds it was. It was not her evidence that the policy was taken out in 1985 to provide for the children and for her. Her evidence was that the policy was taken out to provide for the children:

The Policy was entered into due to the Deceased's financial irresponsibility and the understanding the Deceased and I shared, which was the Policy was required in order to support our Children in the event of the Deceased's passing.

(Emphasis added)

[49] Ms. Moore did not require an equitable ownership interest in the policy to achieve that purpose. She is not named as the applicant on the policy application. Or the owner.

[50] There was no evidence that, from the outset, she would have "a right to determine the beneficiary of the Policy", as the application judge observed. What would the insurer have done had she purported to file a designation of someone as a beneficiary? Could she herself, for example, have designated someone from her family as a revocable, or even an irrevocable, beneficiary under the policy in her stead? At any time? With or without Mr. Moore's permission? We do not know the answers to such questions because no evidence was led directed to this issue.

[51] An assignment involves the transfer of a proprietary right, including a right of action. As noted above, although there is no particular form required for an equitable assignment, it is essential that there be sufficiently clear evidence to establish the intention of the assignor to make such an assignment, that is, to establish that the assignor clearly intended the contractual right to become the property right of the assignee: see Nadeau, at para. 19. The application judge did not direct his mind to the question of Mr. Moore's intention, however. He seems to have leaped directly from the fact there was an oral agreement that if Ms. Moore continued the payments she would continue as a beneficiary, to the conclusion that Mr. Moore intended to make an absolute assignment of the property in the policy or its proceeds. There is little, if any, other evidence to support such an intention.

[52] A further example may assist in highlighting the frailty in making the direct analytical jump in terms of whether Mr. Moore had the necessary intention to make an assignment. Suppose that, in breach of the oral agreement, Ms. Moore paid no premium, or stopped after paying one or two; and assume the children still needed and were entitled to support at that point. Mr. Moore would have been faced with two options in such an event, assuming he assigned his rights as the application judge found.

[53] First, he could continue to pay premiums in order to keep the policy alive, but if he died Ms. Moore would receive the proceeds even though she was in breach of the agreement. Would Mr. Moore have said "but of course I accept that would be the case" if he had been asked that question at the time the agreement was made?

[54] Second, he could let the policy lapse, in which case the children would be deprived of the protection they would have in the event of his death. Would he have accepted such a turn of events if asked during the discussion leading up to the oral agreement? We do not know the answers to these questions because nobody considered the need to examine them, or to consider other potentially similar questions, at the time of the application, and one might wonder if Mr. Moore's answers would have been positive if faced with those choices.

[55] I make two final observations before leaving the subject of equitable assignment.

[56] There is some authority to suggest that payment of premiums by an assignee may not, of itself, constitute valuable consideration for the assignment of a policy and that such cases may be treated as assignments by way of a gift, in which case it must be shown that the assignor has done everything possible to make the transfer of the rights under the policy irrevocable: see Craig Brown and Thomas Donnelly, Insurance Law in Canada, loose-leaf (Toronto: Carswell, 2016), at para. 15.2(b). Without determining whether the law is truly to that effect -- because it is not necessary to do so -- I observe only that this potential difficulty was not considered by the application judge and, of course, was not addressed by any of the parties.

[57] Finally, there are provisions in the Insurance Act that may bear on the assignment issue as well, but were not addressed.

[58] It is not clear, even if the oral agreement did constitute an equitable assignment, that Ms. Moore would have been protected as an assignee, in a contest with Ms. Sweet, the designated irrevocable beneficiary. Had the issue of equitable assignment been in play, and had it been brought to the application judge's attention, s. 200(1)(b) of the Insurance Act may well have affected his determination on this point. Section 200(1)(b) states:

200(1) Where an assignee of a contract gives notice in writing of the assignment to the insurer at its head or principal office in Canada, the assignee has priority of interest as against,

(b) a beneficiary other than one designated irrevocably as provided in section 191 prior to the time the assignee gave notice to the insurer of the assignment in the manner prescribed in this subsection.

[59] This provision would appear to indicate that, as an assignee of the policy or of its proceeds, Ms. Moore would have had to give the insurer notice in writing of her assigned interest in order to obtain a priority of interest against another named beneficiary, much less as against an irrevocable beneficiary. She did not give any such notice. While there may be an argument that s. 200(1)(b) relates to the effectiveness of an assignment as against the insurer, and not as between competing beneficiaries, the Insurance Act does not say this is the case, and there was no argument on this point either, since equitable assignment was not made an issue on the application.

[60] For all the foregoing reasons, the application judge's finding that the oral agreement constituted an equitable assignment cannot stand, in my opinion. Absent equitable assignment or some other basis upon which it could be said that Mr. Moore had placed the policy or its proceeds out of his control and into the hands of Ms. Moore, Ms. Moore must look to a remedial constructive trust for relief. On the application, she focused on unjust enrichment in this regard.

[61] I turn to the trust issues raised on the appeal now.

Constructive Trust

[62] I begin this portion of the analysis with the observation that this is not one of those cases -- in spite of what it may seem at first impression -- where the "equities" are heavily weighted in favour of one party or the other.

[63] It is the case that Ms. Moore had an oral agreement with Mr. Moore that if she paid the premiums she would receive the proceeds of the policy. It is the case that she paid the premiums. And it is the case that Mr. Moore breached the agreement by designating Ms. Sweet as the irrevocable beneficiary under the policy.

[64] On the other hand, Mr. Moore was a man of limited means, living in the post-separation period on a disability pension, and suffering from the disabilities associated with his physical, mental and substance abuse issues. Ms. Sweet -- who is herself disabled -- took care of Mr. Moore and, for practical purposes, provided him with a home, a place to live and a supportive family during the 13 years of their relationship.

[65] There is little, if any, evidence on the record as to Ms. Moore's present financial needs. She continues to live in the former matrimonial home after Mr. Moore's transfer of his one-half interest at the time of separation. Ms. Sweet would appear from the record to be in financial need. Indeed, her evidence is that she was made a beneficiary of the policy because Mr. Moore wanted to ensure that she would be able to remain in the apartment home that she had occupied for 40 years by the time of his death.

[66] On these facts, it cannot be said that Ms. Sweet is no more than a volunteer who gave nothing in exchange for being named irrevocable beneficiary, or that she is simply the recipient of a windfall. She was a 13-year spouse with heavier than normal caregiving duties (both she and Mr. Moore were disabled in varying degrees) and was the person primarily responsible for the home that they lived in.

[67] On the application, Ms. Moore's position was that she was entitled to the policy proceeds on the basis of unjust enrichment. On appeal, her approach was more nuanced. She continued to rely on unjust enrichment but embraced the application judge's finding of equitable assignment in support of the claim, as well. She submitted that the application judge was correct in holding that the irrevocable designation of beneficiary provisions in the Insurance Act did not provide a juristic reason for Ms. Sweet's receipt of the proceeds. In the end, she fell back upon the more expansive view that a constructive trust may be imposed "where good conscience requires it".

[68] I have already concluded that, on the way the case was framed and argued before the application judge and on the record as it currently exists, it is not open for the court to determine whether the oral agreement constituted an equitable assignment. I turn, then, to a consideration of whether the claim for unjust enrichment can otherwise stand or, if not, whether a remedial constructive trust should be found on some other "good conscience" basis in these circumstances.

[69] In my view, Ms. Moore's claim cannot succeed on either basis.

Unjust Enrichment

[70] The elements of a claim for unjust enrichment are well known and were addressed by the application judge. The claim requires that there be (a) an enrichment; (b) a corresponding deprivation; and (c) the absence of any juristic reason for the enrichment: Pettkus v. Becker, at p. 848 S.C.R.; Kerr v. Baranow, at para. 32.

[71] Here, the application judge found that there was an enrichment (Ms. Sweet's receipt of the policy proceeds), a corresponding deprivation (Ms. Moore would not receive them and, yet, there was a causal link between her payment of the premiums and the availability of the proceeds), and that the provisions of the Insurance Act did not provide a juristic reason for Ms. Sweet's retention of the policy proceeds. Ms. Sweet argues that he applied both the corresponding deprivation and the juristic reason parts of the test incorrectly.

Corresponding Deprivation

[72] Ms. Sweet is right when she says that the application judge's finding of a corresponding deprivation to the extent of the full value of the policy proceeds -- as opposed to simply the extent of the premiums she paid -- is rooted in his finding that the oral agreement constituted an equitable assignment. Since I have concluded that the equitable assignment finding must be set aside, the finding of a corresponding deprivation based on it must fail as well.

[73] However, there may be other grounds on which it could be said that Ms. Moore suffered a corresponding deprivation to Ms. Sweet's enrichment. Ms. Moore did continue to pay the premiums following the separation, unaware that she had been replaced as beneficiary by Ms. Sweet, to a total amount of approximately $7,000. Each payment kept the policy alive. It might be said that since she paid the final premium before death, Ms. Moore's deprivation corresponded to Ms. Sweet's enrichment to the extent of the full policy proceeds.

[74] Ms. Sweet argues, on the other hand, that the "deprivation" elements of an unjust enrichment claim are measured on the basis of a "straightforward economic approach", and that "other considerations, such as moral and policy questions, are appropriately dealt with at the juristic reason stage of the analysis": Kerr, at para. 37; Garland v. Consumers' Gas Co., at para. 31; Peel (Regional Municipality) v. Canada, at p. 803 S.C.R. She contends that any corresponding deprivation suffered by Ms. Moore is limited to the loss of the moneys she paid ($7,000); that Ms. Moore is not entitled to count her "expectation losses" (a breach of contract concept) as deprivation, for purposes of unjust enrichment (see Pacific National Investments Ltd. v. Victoria (City)); and that the amount of the premiums paid ($7,000) is sufficiently insignificant, compared to the amount of the insurance proceeds ($250,000) to warrant the imposition of a constructive trust: Richardson Estate v. Mew, at para. 40.

[75] In the end, I do not find it necessary to resolve this issue. In my view, on the record here, Ms. Moore's unjust enrichment claim must fail because the irrevocable beneficiary designation provisions of the Insurance Act provide a juristic reason justifying the receipt by Ms. Sweet of the insurance proceeds.

A Valid Juristic Reason

[76] Respectfully, the application judge made two errors in his assessment of the juristic reason element of unjust enrichment. First, he failed to recognize the significance of Ms. Sweet's designation as an "irrevocable" beneficiary. Second, he failed to apply the two-step analysis to the juristic reason assessment that has been mandated by the Supreme Court of Canada in Garland, at paras. 44-45; and in Kerr, at para. 43-45.

The Irrevocable Beneficiary Designation

[77] Mr. Moore designated his second partner, Ms. Sweet, as the irrevocable beneficiary under the policy. Ms. Moore's designation as beneficiary, on the other hand, was revocable. The change of beneficiary form was properly executed and filed with the insurer pursuant to s. 191(1) of the Insurance Act.

[78] There is a significant difference between the status of a revocable and an irrevocable beneficiary, in my view. The former may be altered or revoked by the insured at any time; the latter may not, except with the consent of the named irrevocable beneficiary. Part V of the Insurance Act has within it a specific regime governing the manner in which a person may be designated as an "irrevocable" beneficiary -- a statutory framework for the benefit of the insured, the insurer and beneficiaries. The designation must be done either in the life insurance policy itself or by way of a declaration. It must be filed with the insurer. And the effect of attempting to designate someone as an irrevocable beneficiary without complying with the provisions of the Insurance Act is that the designation is revocable.

[79] This is clear from the relevant provisions of the Insurance Act. Section 190 stipulates that:

190(1) Subject to subsection (4), an insured may in a contract or by a declaration designate . . . a beneficiary as one to whom or for whose benefit insurance money is to be payable;

(2) Subject to section 191, the insured may from time to time alter or revoke the designation by a declaration.

[80] Section 191 provides that:

191(1) An insured may in a contract, or by a declaration . . . filed with the insurer at its head or principal office in Canada during the lifetime of the person whose life is insured, designate a beneficiary irrevocably, and in that event the insured, while the beneficiary is living, may not alter or revoke the designation without the consent of the beneficiary and the insurance money is not subject to the control of the insured, is not subject to the claims of the insured's creditor and does not form part of the insured's estate.

(2) Where the insured purports to designate a beneficiary irrevocably . . . in a declaration that is not filed as provided in subsection (1), the designation has the same effect as if the insured had not purported to make it irrevocable.

[81] As this court has observed, a beneficiary designation in a life insurance policy is "normally unassailable" absent exceptional circumstances and may provide a juristic reason for the named beneficiary's enrichment: Richardson Estate, at paras. 28 and 61. This is even more forcefully the case, in my opinion, where the designation is irrevocable (which was not the case in Richardson Estate).

[82] Once an irrevocable designation is made in compliance with the Insurance Act -- as it was here, in Ms. Sweet's favour -- the legislation imposes a regime over the policy and its proceeds that gives rights and protection to both the insurer and the named irrevocable beneficiary: the insurer is protected in making payment to the beneficiary, the irrevocable beneficiary by the statutory right to remain as the named beneficiary entitled to receive the insurance moneys unless he or she consents to being removed. The regime is designed to provide an element of certainty and predictability for those statutory rights and protections, which are reinforced by the legislature's decision, reflected in s. 191(1) that, once an irrevocable designation is properly effected, the insured loses control over the insurance moneys and the moneys are immune from attack by the insured's creditors and do not form part of the insured's estate, and in s. 191(2) that a purported attempt to designate an irrevocable beneficiary without complying with s. 191(1) renders the designation revocable.

[83] Simply put, I read this legislative regime as leaning heavily in favour of payment of the proceeds of life insurance policies to those named as irrevocable beneficiaries, whereas it continues to recognize the right of an insured, at any time prior to such a designation, to alter or revoke a beneficiary who does not fall into that category. In short, on the record here, it provides a juristic reason for Ms. Sweet's receipt of the policy proceeds. Disposition of law is well established as a category of juristic reason: Garland, at para. 49.

[84] Authorities holding that the Insurance Act regime may be superseded by provisions in a separation agreement or in some other contract requiring an insured to name the other party as a designated beneficiary must be read with this in mind. As must authorities pre-dating the Supreme Court of Canada's decisions in Soulos and Garland.

[85] Although it was not put this way in argument, a number of these authorities might be explained on the basis of a theory somewhat akin to the equitable assignment argument -- namely that there was a contractual arrangement which was tantamount to an irrevocable beneficiary designation, thereby effectively depriving the insured of any further ability to deal with the policy.

[86] Given the foregoing discussion, I believe that proposition to be questionable at best. However, authorities such as Shannon v. Shannon may be open to that interpretation.

[87] Shannon is frequently referred to for the proposition that courts are willing to override a beneficiary designation on the basis of a prior contract. There, the deceased specifically agreed in the prior separation agreement that he would continue to name his former wife as beneficiary and agreed not to revoke that beneficiary designation at any time in the future. He subsequently removed her as beneficiary, however, and named his nephew and niece as revocable beneficiaries, without their having provided any consideration for his act. McKinlay J. (as she then was) held that the wife was entitled to the proceeds. In an oft-cited passage, she said, at p. 461 O.R.:

The Insurance Act does not specifically preclude the existence of rights outside its provisions. It would go against all conscience to hold that because of the provisions cited, the plaintiff's clear rights under an agreement with her husband for good consideration can be taken away in favour of a niece and nephew who have given no consideration for those rights.

The plaintiff is entitled to a declaration that the proceeds of the policy are impressed with a trust in her favour.

[88] McKinlay J. did not say whether the trust she found was express or constructive. Either might be inferred from her analysis. An important factor in the case, however, was the fact that the separation agreement provided that the wife's designation would not be revoked (or, to put it positively, that it was to be irrevocable). Here, we have no evidence suggesting that the oral agreement contained such a stipulation.

[89] Bielny v. Dzwiekowski; and Fraser v. Fraser, at para. 18, could also be explained on the basis that there was an agreement to name the plaintiff as an irrevocable beneficiary. That said, Shannon and Fraser v. Fraser pre-date the decision of the Supreme Court of Canada in Soulos -- a seminal case dealing with remedial constructive trusts that will be discussed below -- and Bielny did not refer to Soulos.

[90] Moreover, in none of the authorities to which we have been referred, or which my research has unearthed, has the unsuccessful competing beneficiary been designated as the irrevocable beneficiary of the policy proceeds. To the extent that Shannon and other similar authorities may be read as suggesting that a simple contractual promise to make someone a beneficiary under a life insurance policy, albeit a promise for valuable consideration, is sufficient to circumvent the clear statutory requisites for an irrevocable beneficiary designation, I respectfully disagree.

[91] I need not go so far as to say that the designation of a beneficiary as an irrevocable beneficiary under the Insurance Act invariably trumps a prior claimant. There may be cases where the insured, by agreement, has placed the policy or its proceeds beyond his or her ability to deal with them, and therefore beyond his or her ability to make the purported irrevocable designation. This, however, is not one of those cases.

The Two-Step Analysis

[92] Further, the application judge failed to apply the required two-step analysis to the juristic reason assessment. There is no dispute that "[a]t the heart of the doctrine of unjust enrichment lies the notion of restoring a benefit which justice does not permit one to retain": Kerr, at para. 31, citing Peel, at p. 788 S.C.R. Courts, however, have been careful to guard against the remedial constructive trust turning into a completely unprincipled, open-ended remedy, the application of which depends on a particular judge's subjective view of what is unjust or goes against good conscience -- or, to put it in its classic formulation, that is dependent "upon the length of the Chancellor's foot". In Canada, the Supreme Court has developed the "absence of juristic reason" as a mechanism for instilling a principled approach to the remedy.

[93] Iacobucci J. addressed this point in Garland, at para. 40:

It is not of great use to speculate on why Dickson J. in Rathwell expressed the third condition as absence of juristic reason but I believe that he may have wanted to ensure that the test for unjust enrichment was not purely subjective in order to be responsive to Martland J.'s criticism in his reasons that application of the doctrine of unjust enrichment contemplated by Dickson J. would require "immeasurable judicial discretion" (p. 473). The importance of avoiding a purely subjective standard was also stressed by McLachlin J. in her reasons in Peel, at p. 802, in which she wrote that the application of the test for unjust enrichment should not be "case by case palm tree justice".

[94] In continuing with his careful analysis of the juristic reason element of unjust enrichment in Garland, Iacobucci J. acknowledged the need for courts to be able to expand the categories of juristic reason in order that the concept not become too inflexible. At para. 43, he wrote:

As McLachlin J. wrote in Peel, at p. 788, the Court's approach to unjust enrichment, while informed by traditional categories of recovery, "is capable, however, of going beyond them, allowing the law to develop in a flexible way as required to meet changing perceptions of justice". But at the same time there must also be guidelines that offer trial judges and others some indication of what the boundaries of the cause of action are. The goal is to avoid guidelines that are so general and subjective that uniformity becomes unattainable.

[95] In the following paragraphs, Iacobucci J. then articulated the two-step analysis that Canadian courts are required to engage in when considering the juristic reason element. At paras. 44-46, he said:

The parties and commentators have pointed out that there is no specific authority that settles this question. But recalling that this is an equitable remedy that will necessarily involve discretion and questions of fairness, I believe that some redefinition and reformulation is required. Consequently, in my view, the proper approach to the juristic reason analysis is in two parts. First, the plaintiff must show that no juristic reason from an established category exists to deny recovery. By closing the list of categories that the plaintiff must canvass in order to show an absence of juristic reason, Smith's objection to the Canadian formulation of the test that it required proof of a negative is answered. The established categories that can constitute juristic reasons include a contract (Pettkus, supra), a disposition of law (Pettkus, supra), a donative intent (Peter, supra), and other valid common law, equitable or statutory obligations (Peter, supra). If there is no juristic reason from an established category, then the plaintiff has made out a prima facie case under the juristic reason component of the analysis.

The prima facie case is rebuttable, however, where the defendant can show that there is another reason to deny recovery. As a result, there is a de facto burden of proof placed on the defendant to show the reason why the enrichment should be retained. This stage of the analysis thus provides for a category of residual defence in which courts can look to all of the circumstances of the transaction in order to determine whether there is another reason to deny recovery.

As part of the defendant's attempt to rebut, courts should have regard to two factors: the reasonable expectations of the parties, and public policy considerations.

(Emphasis added)

[96] This approach was re-affirmed in Kerr, at paras. 43-44; and in Pacific National, at paras. 22-26.

[97] Here, however, the application judge did not conduct that requisite two-step analysis. A review of his reasons demonstrates that his analysis was focused on the more sweeping principles underlying equitable trusts without first considering whether there was an existing category of juristic reason -- a disposition of law or statutory obligation -- that would justify Ms. Sweet's receipt of the policy proceeds. He appears to have overlooked that the juristic reason was developed to inject a measure of discipline into a court's analysis and to avoid the dangers of "immeasurable judicial discretion" and the determination of an unjust enrichment analysis in a way that involves "case by case palm tree justice".

[98] Instead, the application judge began by referring to Cromwell J.'s comment in Kerr, at para. 50, that "the constructive trust is a broad and flexible equitable tool used to determine beneficial entitlement to property", without acknowledging that Cromwell J. had already reaffirmed Garland's two-step approach, at paras. 43-44. Later in his reasons, after distinguishing Richardson Estate and other authorities, he noted that "[o]f greater relevance [than those authorities] is the more general principle that the test for a juristic reason requires a flexible contextual approach in the particular circumstances of each case", and -- again citing Kerr, at paras. 44 and 41 -- that "[o]verall, the test for juristic reason is flexible, and the relevant factors to consider will depend on the situation before the court", and that consideration may also be given to "the legitimate expectation of the parties, the right of parties to order their affairs by contract", and that, "the test for a juristic reason is flexible and involves discretion and questions of fairness".

[99] All of these principles are correct, in their proper context. However, in relying only on them the application judge lost sight of the need to consider first whether there is an existing category of juristic reason that precludes recovery on the basis of unjust enrichment. As I have indicated above, absent exceptional circumstances, the existence of the statutory regime relating to revocable and irrevocable beneficiaries outlined above falls into an existing recognized category of juristic reason: it constitutes both a disposition of law and a statutory obligation, in my view.

"Good Conscience" Trusts

[100] There has been considerable debate in the jurisprudence and in academia about whether resort to the remedial constructive trust in Canada is now limited to two categories since the Supreme Court of Canada's decision in Soulos -- unjust enrichment and wrongful acts -- thereby eliminating resort to a more elastic "good conscience" trust, i.e., one based on no more than a sense of fairness to the effect that it would be "against all good conscience" to deny a plaintiff recovery in the circumstances of a particular case. At the end of the day, Ms. Moore submits that good conscience is satisfied by giving effect to the oral agreement without which the policy would not have continued to exist.

[101] It has long been accepted that equity is quintessential never-say-never terrain, and that concepts respecting its application develop with the times and to meet the needs of particular circumstances. This long-standing principle may work against establishing a completely closed set of categories as the foundation for imposing a remedial constructive trust.

[102] At the same time, McLachlin J. was pretty clear in Soulos that, while a constructive trust "may be imposed where good conscience so requires" (para. 34), "[t]he situations which the judge may consider in deciding whether good conscience requires imposition of a constructive trust may be seen as falling into two general categories" (unjust enrichment and situations where property had been obtained by a wrongful act) (para. 36). It was her view that "[w]ithin these two broad categories, there is room for the law of constructive trust to develop and for greater precision to be attained, as time and experience may dictate" (para. 43) [emphasis added]. Rothstein J. reaffirmed this view in Professional Institute of the Public Service of Canada v. Canada (Attorney General), at paras. 144-45.

[103] I do not think it is necessary to resolve this debate for purposes of this appeal.

[104] Most of the authorities in which courts have been willing to override a beneficiary designation can be explained on the basis of an agreement between one of the claimants and the insured that removed the insured's ability to designate a later beneficiary. As noted earlier, Shannon involved a separation agreement in which the insured undertook to name his first spouse as a beneficiary irrevocably. In Bielny, the separation agreement required the insured to name the children of the first marriage as irrevocable beneficiaries. In Fraser v. Fraser, the trial judge found on the facts that the terms of the separation agreement requiring the insured to maintain the plaintiff as beneficiary were tantamount to an irrevocable designation.

[105] Whether these authorities need to be re-examined in light of Soulos, as suggested in some authorities -- see, for example, Love v. Love -- is not something that need be determined here. As I have concluded above, it was not open to the application judge on this record to hold that the oral agreement between the Moores constituted an equitable assignment, or that it was tantamount to an irrevocable beneficiary designation.

[106] Absent those considerations, I do not see anything in the circumstances of this case that would place it in some other "good conscience" category not caught with the rubric of either wrongful act (not asserted here) or unjust enrichment. For that reason, I do not see the need to resolve the foregoing debate about whether Soulos has restricted the categories for imposing a remedial constructive trust to unjust enrichment or wrongful act or whether there remains some additional "good conscience" basis.

[107] Simply because wrongful act is not asserted, and unjust enrichment is unsuccessful, does not mean that some other "good conscience" basis must exist on the facts. To engage in such an exercise, on this record at least, it seems to me, would undermine the rationale for creation of the juristic reason element in the first place.

Disposition

[108] For the above reasons, I would allow the appeal and set aside the judgment below. With the exception of the disposition in the following paragraph, I would dismiss the application and order that the proceeds of the policy of insurance on the life of Lawrence Anthony Moore, and accrued interest, be paid out of court to the appellant.

[109] The appellant argued that, at most, Ms. Moore was entitled to the return of the $7,000 in premiums that she paid following the separation. I agree that she should receive that amount, with interest, out of the proceeds held in court. I would not extend this to include the premiums paid prior to separation. Those premiums were paid from the joint account she held with Mr. Moore prior to their separation. She and the children received the protection that the payment of those premiums afforded them during that period -- she was the designated beneficiary in the event of his death -- and he was as entitled to the money in the account as she was.

[110] In my view, this is an appropriate case for the parties to bear their own costs. Accordingly, I would make no order as to costs, here or below.

Dissent

[111] LAUWERS J.A. (dissenting): -- I take a different view of the law regarding unjust enrichment and constructive trusts than my colleague, and therefore dissent.

[112] As my colleague notes, this case is about who should receive the $250,000 proceeds of a life insurance policy on the life of Lawrence Anthony Moore who died in 2013. Should the proceeds be paid to the appellant, Risa Lorraine Sweet, with whom the deceased lived from the summer of 2000 until his death? Or should they be paid to his former spouse, the respondent Michelle Constance Moore?

[113] The application judge decided that the respondent should receive the life insurance proceeds and ordered that the life insurance proceeds be impressed with a trust in favour of the respondent. I would dismiss the appeal for the following reasons.

A. The Factual and Legal Context

[114] The life insurance policy was issued in 1985 by a predecessor of RBC Life Insurance Company. From its inception until the breakup of their marriage in 2000, the deceased and the respondent paid the insurance premiums on the policy from their joint account. Thereafter, the respondent paid the premiums under what the application judge found, at para. 47, to be "an oral agreement with the [respondent] in 2000 to ensure that the premiums payable in respect of the policy would be paid and the proceeds would be available to support his former spouse and children on his death".

[115] On September 21, 2000, despite his oral agreement with the respondent, the deceased signed an irrevocable change of beneficiary form in favour of the appellant and then sent it to the insurer. The appellant signed as a witness. The deceased did not tell the respondent he had changed the beneficiary designation on the policy, and thus she continued to pay the premiums.

[116] The deceased and the respondent entered into a separation agreement dated May 1, 2002 dealing with the distribution of assets and child support. It was silent about the life insurance policy. Except for net equalization property claims, the separation agreement did not contain a general release of any property claims by each party against the other.

[117] The deceased and the respondent divorced on October 3, 2003.

[118] The respondent did not learn the deceased had changed the beneficiary designation on the life insurance policy in favour of the appellant until she received correspondence from the insurance company on July 5, 2013, several weeks after his death.

[119] This case is the regrettable outcome of the fact, noted by the application judge, that "[e]ach of the parties trusted the Deceased, who betrayed that trust by acting duplicitously." Neither is at fault. My colleague points out that "Ms. Sweet -- who is herself disabled -- took care of Mr. Moore and, for practical purposes, provided him with a home, a place to live, and a supportive family during the 13 years of their relationship." True enough, but these benefits to the deceased would not be as significant if he had died the day after signing the irrevocable designation in the appellant's favour. In my view, they are not equities that are relevant to the outcome.

[120] The effect of an irrevocable beneficiary designation is set out in s. 191(1) of the Insurance Act, R.S.O. 1990, c. I.8. The appellant submits the application judge should have given effect to the irrevocable designation signed in her favour by the deceased, despite the oral agreement the deceased made with the respondent by which she paid the premiums in the expectation she would receive the proceeds on his death.

B. The Decision Under Appeal

(1) The Proceeding Below

[121] The respondent stated in her affidavit: "It was expressly agreed between the Deceased and I that in paying for the Policy premiums, I would be entitled to receive the Policy benefits upon his death." She added:

The Policy was entered into due to the Deceased's financial irresponsibility and the understanding the Deceased and I shared, which was that the Policy was required in order to support our Children in the event of the Deceased's passing.

[122] According to the respondent, the financial irresponsibility of the deceased, including borrowing against the respondent's credit, related to his "history of medical issues including substance abuse", which led to the respondent's bankruptcy in March 2000. She was not cross-examined on this evidence.

[123] The issue before the application judge expressed in the notice of application was whether "[t]he Deceased and the Applicant entered into an express, valid and binding contract where for consideration of the Applicant paying all Premiums for the Policy, the Applicant would receive the Benefits". However, nothing in the notice of application addressed the remedial route to recovery, via constructive trust, express trust, equitable assignment, perfectionary trust or some other equitable mechanism.

[124] In the court below, the respondent simply asserted the appellant "would be unjustly enriched if she receives the deceased's life insurance policy". The respondent asked the court to impose a constructive trust on the proceeds of the insurance policy in her favour, and invoked the doctrine of unjust enrichment developed in a line of cases including Soulos v. Korkontzilas.

(2) The Application Judge's Findings

[125] As requested by the respondent, the application judge imposed a constructive trust on the proceeds of the insurance policy in her favour.

[126] He found there was a valid and binding oral agreement between the deceased and the respondent under which the respondent would pay the premiums and would be entitled to the proceeds of the insurance policy on the deceased's death. The application judge gave four reasons for this finding. First, he found there was "a significant continuing purpose for the agreement", which was "to provide for the children of the Applicant and the Deceased as well as the Applicant", since the deceased continued to have obligations to his children at the time of the agreement.

[127] Second, the application judge noted the respondent started paying the premiums herself and continued to do so until the date of the deceased's death. He found she "would not have paid the premiums out of her own account if she had not obtained an equitable assignment of the proceeds of the policy in her favour from the deceased".

[128] Third, the evidence was that the deceased had no assets in 2000 and was dependent upon disability payments for his own living expenses. He was not paying the insurance premiums and he knew full well the respondent was. The application judge found:

The Deceased was aware that the [policy] was being kept in force at the time that he signed the designation in favour of the Respondent. Otherwise, his execution of the change in designation of beneficiary form would have been meaningless.

[129] Fourth, the application judge found the separation agreement's silence about the life insurance policy, even though both parties were aware of its existence, "further confirms the existence of an agreement respecting the Policy that was implemented by the Applicant and the Deceased in 2000".

[130] This lawsuit arose "because the deceased made two incompatible promises before his death", as the application judge noted, adding that "[e]ach of the parties trusted the Deceased, who betrayed that trust by acting duplicitously."

(3) The Application Judge's Legal Determinations

[131] The application judge pointed out that there would have been no insurance proceeds "but for the Applicant's performance of her obligations under the agreement". He held:

Equity should enforce the agreement that the Deceased made assigning his interest in the Policy in the absence of any other agreement between the Respondent and the Deceased and in the absence of any evidence that the Respondent gave any consideration for the Deceased's change of designation in her favour.

[132] The application judge held: "the Deceased agreed to an equitable assignment of his interest in the proceeds of the Policy to the Applicant in 2000 in exchange for her commitment to pay the premiums of the Policy". He repeated this holding and found the oral agreement between the deceased and the respondent functioned as an equitable assignment that thwarted the deceased's irrevocable beneficiary designation under the Insurance Act, because the designation was made in breach of the agreement.

[133] The application judge imposed a constructive trust on the life insurance proceeds in favour of the respondent. I note that he did not mention Soulos in his decision even though it was referred to in the applicant's factum before him. However, he did refer to Kerr v. Baranow, which was built on the Soulos foundation: see Kerr, at para. 71.

C. The Positions of the Parties

[134] The parties take radically different positions on the interpretation and application of the Supreme Court's decision in Soulos.

[135] The appellant argues Soulos limits the remedy of constructive trusts to unjust enrichment and wrongful acts. Therefore, the case abolished the doctrine of good conscience as a general source of equitable jurisdiction. In her view, since the facts do not substantiate either unjust enrichment, or wrongful acts as the basis for a constructive trust, one should not have been imposed in this case.

[136] Further, the appellant takes the position that, as a matter of law, the respondent is not entitled to a proprietary interest in the insurance proceeds, but has only an action in damages against the estate based on the deceased's breach of his oral agreement with the respondent. (This would be an empty remedy since the estate has no assets.)

[137] Accordingly, the appellant asserts the application judge erred in finding that the doctrine of unjust enrichment permitted him to craft an equitable remedy to defeat the appellant's entitlement to the life insurance proceeds as the named irrevocable beneficiary under s. 191(1) of the Insurance Act. She argues doing so was not consistent with this court's decision in Richardson Estate, which was binding on the application judge.

[138] The respondent argues, to the contrary, that in view of the plainly duplicitous and inequitable conduct of the deceased on the facts as found, the court has equitable jurisdiction, recognized in Soulos, to impose a constructive trust over the insurance proceeds in the respondent's favour, despite s. 191 of the Insurance Act. The respondent also submits Richardson Estate is distinguishable. Further, the respondent relies on a group of similar cases in which the court imposed a remedial constructive trust that I will describe as the "disappointed beneficiary cases". These cases were wrongly decided, the appellant argues, since Soulos "abolished the doctrine of good conscience constructive trusts".

[139] As framed by the parties, this appeal raises squarely, for the first time in this court, whether the Supreme Court intended in Soulos to confine the availability of remedial constructive trusts to instances of unjust enrichment and wrongful gains only. Although the facts are simple, the law is not.

D. The Issues

[140] The appeal raises the following issues:

(1) What is the correct interpretation and application of the Supreme Court's decision in Soulos?

(2) Did the application judge err in finding the respondent had made out her claim in unjust enrichment?

(3) Did the application judge err in finding that the deceased had made an equitable assignment of the insurance proceeds to the respondent?

[141] I address the third issue briefly. I agree with my colleague, for the reasons he gives, that the application judge erred in finding that the deceased made an equitable assignment of the policy and its proceeds to the respondent. But, in my view, the error does not affect the outcome. As I will explain, once the application judge found unjust enrichment, it was open to him to immediately impose a remedial constructive trust on the life insurance proceeds. He did not need to find some other equitable mechanism like equitable assignment before doing so.

E. Analysis

[142] The application judge's critical findings are very clear and well supported by the evidence: the deceased and the respondent entered into an oral agreement by which she would pay the premiums on the life insurance policy and receive the proceeds upon his death; and the deceased behaved duplicitously in irrevocably designating the appellant as the beneficiary of the policy.

[143] I begin with some observations on the underlying policy tension in this area of the law, which in my view suffuses the analysis of the substantive issues. I then consider the Soulos decision and the issues in sequence.

The Tension Between Law and Equity

[144] In order to contextualize the ruling in Soulos, it is necessary to acknowledge a perennial tension in the common law tradition between, on the one hand, the formal demands of the law as dispensed by the old common law courts, and, on the other hand, the mercies of equity dispensed by the old chancery courts. (Indeed, this tension helps explain some of the differences between my colleague and me.)

[145] The way equity operated is well known. When the common law courts reached a particularly harsh result flowing from the demands of formality and rigorous logic, chancery courts could intervene and mitigate the result in certain circumstances by exercising authority over the defendant's conscience and compelling the defendant not to act on his or her full legal rights.

[146] Law claims the virtues of certainty and predictability, while equity claims the virtues of doing justice and upholding fairness in particular cases. As my colleague observes, the ancient criticism of equity is that its mercies varied arbitrarily with the length of the Chancellor's foot. The fundamental tension lives on, even though law and equity have been fused in Ontario since the late 19th century.

[147] The constructive trust cases show this tension. Judges imposing constructive trusts sometimes cite the magnanimous words of Lord Denning in Hussey v. Palmer, who spoke of a constructive trust, at p. 747 All E.R., as

a trust imposed by law whenever justice and good conscience require it. It is a liberal process, founded on large principles of equity, to be applied in cases where the defendant cannot conscientiously keep the property for himself alone, but ought to allow another to have the property or a share in it.

[148] The tension between the legal and equitable impulses is evident in Pettkus v. Becker. Dickson J., at pp. 847-48 S.C.R., speaking for the majority, described the attributes of a remedial constructive trust:

The principle of unjust enrichment lies at the heart of the constructive trust. "Unjust enrichment" has played a role in Anglo-American legal writing for centuries. Lord Mansfield, in the case of Moses v. Macferlan put the matter in these words: "the gist of this kind of action is, that the defendant, upon the circumstances of the case, is obliged by the ties of natural justice and equity to refund the money". It would be undesirable, and indeed impossible, to attempt to define all the circumstances in which an unjust enrichment might arise. . . . The great advantage of ancient principles of equity is their flexibility: the judiciary is thus able to shape these malleable principles so as to accommodate the changing needs and mores of society, in order to achieve justice. The constructive trust has proven to be a useful tool in the judicial armoury.

(Citations omitted and emphasis added)

[149] However, Martland J., in dissent, saw the majority's extension of the constructive trust remedy in Pettkus as undesirable, because he found, at p. 859 S.C.R., that "[i]t would clothe judges with a very wide power to apply what has been described as palm tree justice without the benefit of any guidelines." He then asked: "By what test is a judge to determine what constitutes unjust enrichment?" He answered: "The only test would be his individual perception of what he considered to be unjust." This was what he resisted.

[150] Justice McLachlin (as she then was) said that the tension between the legal and equitable impulses was "jurisprudential in nature", and described it in Peel (Regional Municipality) v. Canada, at para. 34 (p. 785 S.C.R.), as follows:

I refer to the tension between the need for certainty in the law and the need to do justice in the individual case; the tension between the need for predictable rules upon which people can predicate their conduct and the desire to allow recovery where, and only where, retention of the benefit would in all the circumstances be unjust. This jurisprudential tension corresponds to the doctrinal tension discussed above. An approach based on traditional categories has the advantage of being predictable. Its supporters conjure up "frightful images . . . of judges roaming willy-nilly over the restitutionary landscape with only their inner voices to guide them." (McInnes, "Incontrovertible Benefits and the Canadian Law of Restitution" (1990) 12 Advocates' Q. 323, at p. 352). Those advocating the approach of general principle, on the other hand, are more ready to concede that in some cases, the court may have to make decisions based on the equities of the particular case before them.

[151] In Garland v. Consumers Gas Co., at para. 40, Iacobucci J. noted this concern as he provided an explanation for the way Dickson J. articulated the "juristic reason" element of the test for unjust enrichment in Rathwell v. Rathwell:

. . . I believe that he may have wanted to ensure that the test for unjust enrichment was not purely subjective in order to be responsive to Martland J.'s criticism in his reasons that application of the doctrine of unjust enrichment contemplated by Dickson J. would require "immeasurable judicial discretion" (p. 473). The importance of avoiding a purely subjective standard was also stressed by McLachlin J. in her reasons in Peel, at p. 802, in which she wrote that the application of the test for unjust enrichment should not be "case by case palm tree justice".

[152] Other Supreme Court cases raising this tension include Pacific National Investments Ltd. v. Victoria (City), at para. 13; and Kerr, at para. 43. As will be seen, the tension between the legal and equitable impulses burbles up occasionally in the analysis of the substantive issues, like an underground stream.

F. Issue One: What is the Correct Interpretation and Application of the Supreme Court's Decision in Soulos?

[153] Both parties proceeded on the basis that Soulos was the governing precedent and that unjust enrichment was the basis of any possible constructive trust claim in this case. But their positions on the proper interpretation and application of Soulos, described above, could not be further apart.

[154] For the reasons that follow, I would reject the appellant's argument that Soulos limits the remedy to the two defined situations of unjust enrichment and wrongful acts, and otherwise abolished the doctrine of good conscience constructive trusts as a general source of equitable jurisdiction.

(1) The Fact Situation in Soulos

[155] In Soulos, a real estate broker entered into negotiations to purchase a commercial building on behalf of his client, but then arranged for his wife to purchase the building instead. The client brought an action to compel the broker to convey the property to him on the basis that the broker breached his fiduciary duty and that breach gave rise to a constructive trust. However, in the meantime, the property's value had declined so that the client had suffered no financial loss from the broker's perfidy. Nonetheless, the Supreme Court affirmed this court's decision requiring the broker to convey the property to the client.

(2) The Issue Determined in Soulos

[156] The specific legal issue determined by the Supreme Court in Soulos was "whether a constructive trust over property may be imposed in the absence of enrichment of the defendant and corresponding deprivation of the plaintiff", as McLachlin J. stated, writing for the majority, at para. 1. She delineated, at para. 45, the circumstances in which a court could impose a constructive trust based specifically on wrongful acts including fraud and breach of loyalty.

(3) The Appellant's View of Soulos

[157] The appellant argues that Soulos limits the remedy to the two defined situations of unjust enrichment and wrongful acts, and otherwise abolished the doctrine of good conscience constructive trusts as a general source of equitable jurisdiction. She draws support for her argument from a line of cases starting with Soulos and continuing with Gorecki v. Canada (Attorney General); Professional Institute of the Public Service of Canada v. Canada (Attorney General) ("PIPSC"); and Sun Indalex Finance, LLC v. United Steelworkers ("Sun Indalex"). The appellant also relies on several disappointed beneficiary cases I will consider later.

[158] I note by way of preface that all three cases involve a party pursuing a remedial constructive trust, based on unjust enrichment or wrongful acts. None of the cases raised squarely or addressed the issue of whether remedial constructive trusts are confined to these two bases. However, the appellant quotes snippets of each in support of her argument to that effect.

[159] First, the appellant cites Gorecki. This was a proposed class action in which the class members sought interest payments from the federal government on retroactive lump sum disability pension payments under the Canada Pension Plan. This court agreed with the motion judge's conclusions that the government did not owe the plaintiff a fiduciary duty and that there was no basis for an unjust enrichment claim, leaving only the claim for a constructive trust based on a wrongful act. However, this court disagreed with the motion judge's conclusion that, since the constructive trust is an evolving remedy, it would be wrong to dismiss the wrongful act claim on a pleading motion. Sharpe J.A., writing for the court, dismissed the wrongful act claim, explaining, at para. 11:

As I have already explained, the Crown is under no equitable obligation in relation to CPP benefits. Nor, for reasons already explained, does the Crown owe the appellant any equitable obligation or duty of loyalty. As the claim for interest relates to a claim for a statutory benefit, it is difficult to see any basis for the assertion of a proprietary remedy. Finally, it would be unjust to impose a constructive trust in the face of a comprehensive statutory scheme that precludes the award of interest.

[160] The appellant relies on a single sentence of the decision, where, after citing para. 43 of Soulos, Sharpe J.A. stated, at para. 9:

Accordingly, there are two possible ways of acquiring a constructive trust: first, by establishing unjust enrichment; and, second, by demonstrating the type of wrongful conduct by the defendant that is capable of giving rise to a constructive trust.

[161] However, there is no suggestion in Gorecki that any other basis for a constructive trust was advanced, so I do not see this language as determinative.

[162] Second, the appellant cites PIPSC. Certain unions and employee associations were beneficiaries of defined benefit pension plans administered by the federal government as the employer. They sued to recover $28 billion in pension surplus the government had debited from the actuarial surplus. Justice Rothstein, writing for the court, noted, at para. 144:

Since this Court's decision in Soulos . . . there have been two grounds on which a court can impose a constructive trust: (1) breach of an equitable obligation, and (2) unjust enrichment. The appellants have argued both in this appeal.

[163] Candidly, I see this statement not so much as prescriptive, but more as descriptive of the basis of the appellant's case in PIPSC.

[164] The court in PIPSC rejected a wrongful act as the basis for a constructive trust on the basis that the government did not owe its employees a fiduciary duty (para. 147). The court rejected unjust enrichment because there was no property and no transfer of wealth to the government, and therefore "no link between the alleged deprivation and the property right they seek" (paras. 157-58). The reason was that "the Superannuation Accounts are mere accounting records, and do not contain assets in which the appellants have an interest, [therefore] no enrichment and corresponding deprivation can be found", as Rothstein J. noted, at para. 154. As in Gorecki, there was no issue raised about imposing a constructive trust outside of unjust enrichment and wrongful act.

[165] Third, the appellant cites Sun Indalex. It was an infernally complicated case about the respective priorities of creditors and pension beneficiaries when an enterprise is no longer operating in the ordinary course of business, but is under the Companies' Creditors Arrangement Act, R.S.C. 1985, c. C-36 ("CCAA") and heading for bankruptcy.

[166] It is important to note that the claim in Sun Indalex proceeded under the wrongful acts branch of Soulos, particularly the criteria laid out in para. 45. These criteria have no application to this appeal. I note that the prospect of another basis for imposing a constructive trust was not in issue.

[167] Indalex sought protection from its creditors under the CCAA. Indalex was both the corporate employer and the administrator of two employee pension plans. Such a dual role is permissible under s. 8(1)(a) of the Pension Benefits Act, R.S.O. 1990, c. P.8, which also provides, in s. 22(4), that a plan administrator must not permit its interests to conflict with its duties in respect of the pension fund.

[168] Indalex's decision to seek protection under the CCAA gave rise to the prospect that its financial interests could conflict with the interests of the pension beneficiaries, thus jeopardizing its ability to discharge its fiduciary duties properly as the plan administrator.

[169] All three Sun Indalex opinions in the Supreme Court acknowledged that the permissible dual role could potentially morph into a conflict of interest: Deschamps J., at para. 64; Cromwell J., at paras. 196-98; and LeBel J., at para. 272. The issue that divided the Supreme Court was the point at which Indalex was required to address the prospect of this conflict between its corporate duties and its duties as pension administrator, and what it ought to have done to address the conflict.

[170] The pension beneficiaries argued that Indalex had breached its fiduciary duties, and claimed a constructive trust over the proceeds of the company's sale in priority to other creditors. This court agreed with the pension beneficiaries, but the majority of the Supreme Court reversed.

[171] Cromwell J. observed, at para. 228 of Sun Indalex: "As the Court of Appeal recognized, the governing authority concerning the remedial constructive trust outside the domain of unjust enrichment is Soulos." The claim to a remedial constructive trust foundered on the lack of a causal connection [at para. 236]: "There is no evidence to support the contention that Indalex's breach of its fiduciary duty as pension administrator resulted in the assets retained in the reserve fund."

[172] I do not read Gorecki, PIPSC or Sun Indalex as ruling definitively that Soulos abolished the doctrine of good conscience constructive trusts beyond the two defined situations of unjust enrichment and wrongful acts. The issue was not squarely before the courts and its determination was not necessary to the decisions. Further, such a reading would be inconsistent with the words of McLachlin J. in Soulos.

(4) The Reach of Soulos

[173] In Soulos, McLachlin J. worked through the history of constructive trusts in order to discern whether and how to recognize a remedial constructive trust for wrongful acts, which was the particular problem in the case before her. In my view, she did not purport to restate and reframe the law of constructive trusts for all purposes, and she said nothing to close the categories of constructive trusts. Had she intended to abolish good conscience constructive trusts beyond the categories of unjust enrichment and wrongful acts, then one would have expected clear and definitive language to that effect, but there is none. Instead, McLachlin J.'s choice of language justifies the conclusion that the court expected constructive trust law to continue to develop beyond the categories of unjust enrichment and wrongful act.

[174] Justice McLachlin disagreed with the position that a constructive trust cannot be imposed where there has been no unjust enrichment, at para. 16, which was the real issue in the case. She stated, at para. 17:

The history of the law of constructive trust does not support this view. Rather, it suggests that the constructive trust is an ancient and eclectic institution imposed by law not only to remedy unjust enrichment, but to hold persons in different situations to high standards of trust and probity and prevent them from retaining property which in "good conscience" they should not be permitted to retain.

[175] Justice McLachlin made several other pertinent observations of more general application, at paras. 20-22:

Canadian courts have never abandoned the principles of constructive trust developed in England. They have, however, modified them. . . .

This Court's assertion that a remedial constructive trust lies to prevent unjust enrichment in cases such as Pettkus v. Becker should not be taken as expunging from Canadian law the constructive trust in other circumstances where its availability has long been recognized. The language used makes no such claim. A. J. McClean, . . . describes the ratio of Pettkus v Becker as a "a modest enough proposition". He goes on: " It would be wrong . . . to read it as one would read the language of a statute and limit further development of the law ".

Other scholars agree that the constructive trust as a remedy for unjust enrichment does not negate a finding of a constructive trust in other situations.

(Emphasis added)

[176] Justice McLachlin also noted, at para. 17, that the law of England and Canada could well be diverging:

In England, the trust thus created was thought of as a real or "institutional" trust. In the United States and recently in Canada, jurisprudence speaks of the availability of the constructive trust as a remedy; hence the remedial constructive trust.

She embarked on a review of English law, noting, at para. 24: "In sum, the old English law remains part of contemporary Canadian law and guides its development."

[177] Justice McLachlin was well aware that English law was evolving. She described the opposition in England to Lord Denning's expansive view of constructive trust, at paras. 30-31. She added, at para. 37, that "[i]n England the law has yet to formally recognize the remedial constructive trust for unjust enrichment, although many of Lord Denning's pronouncements pointed in this direction." She also surveyed the law in New Zealand and the United States.

[178] Justice McLachlin expressed a critical conclusion, at para. 25:

I conclude that the law of constructive trust in the common law provinces of Canada embraces the situations in which English courts of equity traditionally found a constructive trust as well as the situations of unjust enrichment recognized in recent Canadian jurisprudence.

[179] This led McLachlin J. to the heart of the matter, at paras. 34-35:

It thus emerges that a constructive trust may be imposed where good conscience so requires. The inquiry into good conscience is informed by the situations where constructive trusts have been recognized in the past. It is also informed by the dual reasons for which constructive trusts have traditionally been imposed: to do justice between the parties and to maintain the integrity of institutions dependent on trust-like relationships. Finally, it is informed by the absence of an indication that a constructive trust would have an unfair or unjust effect on the defendant or third parties, matters which equity has always taken into account. Equitable remedies are flexible; their award is based on what is just in all the circumstances of the case.

Good conscience as a common concept unifying the various instances in which a constructive trust may be found has the disadvantage of being very general. But any concept capable of embracing the diverse circumstances in which a constructive trust may be imposed must, of necessity, be general. Particularity is found in the situations in which judges in the past have found constructive trusts. A judge faced with a claim for a constructive trust will have regard not merely to what might seem "fair" in a general sense, but to other situations where courts have found a constructive trust. The goal is but a reasoned, incremental development of the law on a case-by-case basis.

(Emphasis added)

[180] These statements, in my view, decisively refute the appellant's argument that a court can impose a constructive trust now in only two categories of cases: to remedy an unjust enrichment or a wrongful act.

[181] I acknowledge that McLachlin J. then turns, at para. 36, to address specifically the two dominant categories of unjust enrichment and wrongful acts as bases for imposing a constructive trust. She innovated somewhat by more firmly establishing remedial constructive trusts as a remedy for wrongful acts, at para. 45, which was the basis on which she decided Soulos itself.

[182] The Supreme Court plainly knew Soulos was carving out new ground. This was acknowledged by Sopinka J., in dissent, who criticized the majority's decision, at para. 72, because it "fails to refer to a single Canadian case where a constructive trust was ordered despite the absence of unjust enrichment".

[183] But I note that nowhere did McLachlin J. abandon as spent the continuing development of the law of constructive trusts: see paras. 20-25 and 34-35 of Soulos, quoted above.

(5) Concluding Observations on Soulos

[184] In my view, the Supreme Court left open four routes by which a court could impose the "ancient and eclectic institution" of a constructive trust: (1) unjust enrichment; (2) wrongful acts or wrongful gain; (3) circumstances where its "availability has long been recognized" (para. 21), such as "situations where a constructive trusts have been recognized in the past" (para. 34) or "other situations where courts have found a constructive trust" (para 35); and (4) otherwise, where good conscience requires it. In relation to this last point, the Soulos court anticipated that the law of remedial constructive trusts would continue to develop, consistent with the words of Dickson J. in Pettkus, at p. 847-48 S.C.R.: "the judiciary is thus able to shape these malleable principles so as to accommodate the changing needs and mores of society, in order to achieve justice".

[185] I see no other way to give meaning to the words of McLachlin J. in Soulos, at para. 21: "This Court's assertion that a remedial constructive trust lies to prevent unjust enrichment in cases such as Pettkus v. Becker should not be taken as expunging from Canadian law the constructive trust in other circumstances where its availability has long been recognized." See, also, BNSF Railway Co. v. Teck Metals Ltd., at paras. 45-55.

[186] I observe that McLachlin J. expected the tension between the legal impulse and the equitable impulse to be reconciled and disciplined in specific instances by considering and analogizing to past situations in which the courts have imposed constructive trusts.

[187] As a roadmap to what follows, I begin with the first route to a constructive trust: to remedy unjust enrichment. The case was argued on this basis. (The second route, wrongful act or wrongful gain, is not engaged.) But I am alive to the ambiguous way the concept of "unjust enrichment" is sometimes used in the authorities to describe, on the one hand, restitution for unjust enrichment, that is, the idea of "giving back" what was taken, and, on the other hand, disgorgement, that is, the idea of "giving up" a gain that properly belongs to another. I would enlist both the third route to a constructive trust -- circumstances where the availability of a trust has previously been recognized -- and the fourth route -- where good conscience otherwise demands it -- to resolve the ambiguity.

[188] I now turn to the specific issues in this appeal.

G. Issue Two: Did the Application Judge Err in Finding the Respondent had Made Out Her Claim in Unjust Enrichment?

[189] I begin by reviewing the elements of an unjust enrichment claim, and then turn to the appellant's arguments that the application judge erred in finding the respondent made out her claim in unjust enrichment. Finally, I consider the jurisprudence in the disappointed beneficiary cases, which are similar.

(1) The Elements of Unjust Enrichment and the Application Judge's Findings

[190] For the court to find unjust enrichment, three elements must be present, as the Supreme Court held in Pettkus: first, the defendant's enrichment; second, the plaintiff's corresponding deprivation; and third, the absence of any juristic reason for the defendant's enrichment. To succeed on the first element, the plaintiff must demonstrate that the defendant was enriched. For the second element, the plaintiff must show that he or she sustained a corresponding deprivation. Finally, the plaintiff must show that the defendant's enrichment occurred without a juristic reason. If there are no existing categories of juristic reason that can justify the defendant's enrichment, then the court "may take into account the legitimate expectations of the parties . . . and moral and policy-based arguments about whether particular enrichments are unjust": Kerr, at para. 44.

[191] The application judge had no difficulty in finding the respondent met the first two elements of the test. He noted:

The Respondent is enriched to the extent of the proceeds of the Policy. . . . The Applicant has suffered a corresponding deprivation not only to the extent she paid the premiums but also, more fundamentally, to the extent the proceeds of the Policy were paid to the Respondent notwithstanding the prior equitable assignment of such proceeds to her.

[192] As for the third element, the absence of a juristic reason for the enrichment, the application judge found:

[W]hile the designation of a beneficiary [under the Insurance Act] may be a juristic reason in some cases, the issue for the Court is whether the designation of the Respondent is a juristic reason in the particular circumstances of this case.

On the basis of the foregoing, I conclude that a court is not precluded from imposing a constructive trust on proceeds of an insurance policy paid to a named beneficiary based on the principle of unjust enrichment merely because the deceased policyholder executed a designation in favour of the named beneficiary.

[193] In reaching this conclusion, the application judge distinguished this court's decision in Richardson Estate. He found it was not a "useful authority" and was not dispositive.

(2) The Appellant's Arguments

[194] The appellant does not contest the first element of unjust enrichment: her enrichment. She does contest the second element, the respondent's deprivation, and the third, the absence of a juristic reason for her enrichment.

(a) The Deprivation Element

[195] The appellant disputes the deprivation finding on three grounds, which I address in the sequence in which she makes her arguments.

[196] First, the appellant submits the deprivation must be measured using a "straightforward economic approach", drawing on the Supreme Court's language in Pacific National Investments, at para. 20. The appellant submits that the respondent's deprivation, in restitutionary terms, should be limited to the premiums she paid from 2000 until the deceased's death, which is about $7,000. She asserts this is too small an amount to justify the remedy of a constructive trust, drawing on this court's observation in Richardson Estate, where the amount of premiums paid was $1,161.

[197] I would reject this submission. I observe that the respondent's contribution to maintaining the policy since separation was more than $7,000. From its inception in 1985, the respondent's contribution was considerably greater, about $30,000 according to the evidence. It is not reasonable to discount her contribution on the basis that she had the benefit of the policy during her marriage. On either amount, this would not be a minimal deprivation unworthy of the court's attention.

[198] Second, the appellant argues that the remedy for the deprivation is not to give the insurance proceeds to the respondent, since that would wrongly reward her "expectation interest", not her "restitutionary interest", which is to be measured by the direct expenditures made. Third, the appellant makes the related argument that the respondent is not entitled to a proprietary remedy for the deceased's breach of the oral agreement.

[199] I would reject both arguments. This is not, contrary to the appellant's argument, a simple case of breach of contract.

[200] I pause here to address a lurking concern that is found in some of the disappointed beneficiary cases, which I analyze more intensely below, and in my colleague's reasons. This is the fear described by Professor McInnes of "judges roaming willy-nilly . . . with only their inner voices to guide them", quoted by McLachlin J. in Peel. It is also reflected in the statement, with which I agree, that Soulos counselled courts to be "careful to avoid inviting judges to impose such trusts whenever and wherever they believe justice requires" in Love v. Love, para. 47. There is a fear, which I do not share, that granting a constructive trust remedy to the former spouse whose claim was rooted in breach of contract only would make it difficult to resist the migration of the concept to the general law of contract: see Ladner v. Wolfson, at para. 52. These expressions simply reflect the tension I referred to earlier between law and equity.

[201] To be clear, I do not take the position that any breach of contract would or should lead to the imposition of a constructive trust. In my view, the disappointed beneficiary cases must be treated differently than ordinary breach of contract cases. I discuss this in detail below, but a useful starting point is to recognize that the court must take into account the family context within which the dispute arose, as Cromwell J. noted in Kerr, at para. 34. Although the principles behind imposing a constructive trust are meant to apply to all kinds of cases, with no special modifications for family cases, he noted in Kerr that the family context affects the application of the principles.

[202] The oral agreement in this case came about in the context of a marriage unravelling in dire economic circumstances, in which the deceased's addiction-fuelled spending habits had forced the respondent into bankruptcy. The respondent and the deceased were salvaging the best they could for their family from a desperate situation. As the application judge observed, the purpose of their agreement was to create an asset that would eventually assist the respondent and the children, and would be free of attachment by the deceased's creditors. It is clear that the deceased had no money to pay the premiums; had the respondent not done so, there would have been no policy.

The Domestic Cases Provide a Useful Analogy

[203] This case can be analogized somewhat to the domestic cases such as Kerr. These cases engage the dichotomy referred to earlier between a restitutionary "giving back", and "giving up" a wrongful gain, although the cases are understandably not framed quite that way. In that case, Cromwell J. began his discussion of the jurisprudence, at para. 46 of Kerr, by observing that "[r]emedies for unjust enrichment are restitutionary in nature; that is, the object of the remedy is to require the defendant to repay or reverse the unjustified enrichment." However, he noted, at para. 50, that where property is involved, it is appropriate for the court to consider a constructive trust remedy where there is a link between the acquisition of the asset in the defendant's hands and the plaintiff's contribution. In such circumstances, giving the plaintiff a share of the property proportionate to her contribution could be appropriate.

[204] The general rule in the domestic cases is that a plaintiff must establish that "a monetary award would be insufficient in the circumstances": Kerr, at para. 52; and Peter v. Beblow, at para. 31 (p. 999 S.C.R.). In Peter, McLachlin J. noted, at para. 31, that, "in looking at whether a monetary award is insufficient the court may take into account the probability of the award's being paid as well as the special interest in the property acquired by the contributions". In deciding whether to grant a proprietary remedy instead of a monetary award, "the court may take into account the probability of recovery, as well as whether there is a reason to grant the plaintiff the additional rights that flow from recognition of property rights": Kerr, at para. 52. There is no doubt that, if the respondent is not given the proceeds of the insurance policy in this case, she will get nothing, since the estate has no assets to pay a damages award for breach of contract.

[205] In my view, consideration of these factors supports the imposition of a remedial constructive trust in this case. Contrary to the appellant's argument, the respondent's financial contribution is not the correct measure of her deprivation. The appellant's entitlement under the designation crystallized at the moment of the deceased's death under the beneficiary designation. At that moment, both had a claim to the same asset. In my view, the deprivation the respondent would suffer if the appellant were successful is the full payout of the life insurance proceeds, not the amount of the premiums she paid. That is the asset -- the chose in action -- for which she paid, of which she stands deprived and which she seeks to have restored to her. The appellant contributed nothing to the acquisition of the asset in dispute.

[206] In my view, the deprivation element of the unjust enrichment test has been met to the full extent of the insurance payout in the respondent's favour. In the context, the relevant enrichment is to be measured by the possible benefit to the appellant, and the deprivation to the respondent must correspondingly be measured in the same terms: Kerr, at para. 39. That is the full value of the life insurance proceeds. To use that measure would be to take the relevant "straightforward economic approach" required by the Supreme Court in Pacific National Investments, as adapted by the court in Kerr.

(b) The Juristic Reason Element

[207] According to the appellant, the application judge's most serious error is his finding there was no juristic reason for the appellant's enrichment. The appellant argues this case turns on the application of one of the accepted categories of juristic reasons that justify the respondent's enrichment, being a "disposition of law", referred to in Rathwell, at p. 455 S.C.R.; and Garland, at para. 44. The irrevocable designation in her favour under s. 191 of the Insurance Act was just such a juristic reason, the appellant asserts, based on this court's decision in Richardson Estate, which is binding and is sufficient to dispose of the appeal.

[208] I would reject this submission, for three reasons: first, a purposive approach to the interpretation and application of s. 191 of the Insurance Act does not lead to the conclusion that it was intended to operate as a juristic reason for the appellant's enrichment on the facts of this case. Second, the application of the principled approach to unjust enrichment leads to the contrary conclusion. Third, Richardson Estate does not drive the outcome of this case and should be distinguished. I explain each in turn, after a short description of the juristic reason analysis.

(i) The Juristic Reason Element of the Unjust Enrichment Test

[209] The "juristic reason" element has evolved since it was first expressed by Dickson J. in Rathwell, and repeated by him in Pettkus. In Garland, Iacobucci J., at para. 46, acknowledged that "this area is an evolving one" and predicted that new cases "will add additional refinements and developments": see, also, Pacific National Investments, per Binnie J., at paras. 23-24. To succeed, it must be shown that "there is no reason in law or justice for the defendant's retention of the benefit conferred by the plaintiff, making its retention unjust in the circumstances of the case", as Cromwell J. noted in Kerr, at para. 40.

[210] In Garland, the court established a two-step analysis for the absence of juristic reason, taking its lead from the decision of McLachlin J. in Peel, at para. 32, p. 784 S.C.R., where she contrasted the old "categorical approach" to claims for unjust enrichment with the new "principled" approach the court took in Pettkus. At the first step, the court considers the established categories of juristic reasons. If none applies, then at the second step the court assesses "the reasonable expectations of the parties and public policy considerations to assess whether recovery should be denied": Kerr, at para. 43.

[211] Justice Cromwell listed the categories of juristic reasons in Kerr, at para. 41, as the existence of a contract or a donative intent, and a disposition of law such as common law, equitable or statutory obligations. The appellant invokes disposition of law as the relevant juristic reason.

[212] Justice Cromwell elaborated on the second step of the juristic reason analysis, at para. 44 of Kerr: "[I]f the case falls outside the existing categories, the court may take into account the legitimate expectations of the parties . . . and moral and policy-based arguments about whether particular enrichments are unjust". He added that "the test for juristic reason is flexible, and the relevant factors to consider will depend on the situation before the court". Finally, he noted, at para. 45: "questions regarding how (and when) factors relating to the manner in which the parties organized their relationship should be taken into account".

(ii) The Disposition of Law Category of Juristic Reason Does Not Apply to This Case

[213] In Kerr, Cromwell J. included common law, equitable and statutory obligations in the category of "disposition of law". This category operates, he noted, where "the enrichment of the defendant at the plaintiff's expense is required by law, such as where a valid statute denies recovery" (para. 41). The appellant argues this is precisely how s. 191 of the Insurance Act was intended to operate in the circumstances presented by this case. I do not agree.

The First Step of the Juristic Reason Analysis is Purposive

[214] It is necessary to carefully consider the relevant purpose of the legislation that is asserted to constitute a juristic reason for refusing equitable relief, in this case ss. 190 and 191 of the Insurance Act, which I reproduce for convenience.

[215] Section 190 provides:

90(1) [A]n insured may in a contract or by a declaration designate . . . a beneficiary as one to whom or for whose benefit insurance money is to be payable. (2) Subject to section 191, the insured may from time to time alter or revoke the designation by a declaration.

Section 191 provides:

191(1) An insured may in a contract, or by a declaration . . . filed with the insurer at its head or principal office in Canada during the lifetime of the person whose life is insured, designate a beneficiary irrevocably, and in that event the insured, while the beneficiary is living, may not alter or revoke the designation without the consent of the beneficiary and the insurance money is not subject to the control of the insured, is not subject to the claims of the insured's creditor and does not form part of the insured's estate.

[216] The task at the first step of considering whether a juristic reason exists to deny the plaintiff's recovery is not a mere exercise in mechanical pigeonholing. For example, in Peter, it was argued that the legislative decision to exclude unmarried couples from property division legislation was the juristic reason that prevented the court from using the equitable doctrine of unjust enrichment to address their property dispute. The Supreme Court rejected the argument in Peter, where McLachlin J., said, at p. 994 S.C.R.: "It is precisely where an injustice arises without a legal remedy that equity finds a role." As Cromwell J. explained in Kerr, at para. 45, this argument "misapprehended the role of equity".

[217] The purpose of what are now ss. 190 and 191 of the Insurance Act was discussed in Shannon v. Shannon, Shannon v. Shannon Estate. Justice McKinlay (as she then was) imposed a trust in circumstances somewhat similar to those in this case. The deceased was obliged under a separation agreement "to continue to name his wife a beneficiary" and "not to revoke such beneficiary designation at anytime in the future". A year later, the deceased filed a "new revocable beneficiary designation form" under the Insurance Act naming the defendants. He died shortly thereafter. The assets in his estate would not have satisfied a judgment in favour of his former spouse for breach of contract.

[218] The plaintiff took the position that by virtue of the provision in the separation agreement (para. 5):

[T]he insurance policy and the proceeds became impressed with a trust, and that the designation by the insured of [the defendants] as beneficiaries constituted a disposition of trust property to volunteers who could not take the trust property in the face of a prior trust for value in favour of the wife of the deceased.

[219] Justice McKinlay found, at para. 10, that language in a separation agreement could function as a revocable designation under what is now s. 190 of the Insurance Act. She then addressed the defendant's argument, described at para. 13 (p. 461 O.R.), "that any finding of the court of a trust in favour of the plaintiff would have the effect of the court's attempting to overrule a clear statutory provision", being the insured's right to change the designation under the predecessor to s. 190 of the Insurance Act.

[220] Justice McKinlay rejected this argument. She accepted, at para. 14 (p. 461 O.R.), that "the Insurance Act provides a statutory framework for the protection of the insured, the insurer and beneficiaries". She explained, at para. 15 (p. 461 O.R.): "There is no doubt that had the insurer paid the proceeds to the [defendants] before receiving notice of the claim of the plaintiff, the insurer would have been protected by the provisions of the Act."

[221] But McKinlay J. acknowledged that this was not the issue before her. Instead, the contest was between the defendants, who were designated as revocable beneficiaries under the predecessor section to s. 190 of the Insurance Act, and the plaintiff, whose claim to the insurance proceeds was by way of a separation agreement entered into before the deceased made the beneficiary designations.

[222] Justice McKinlay found the Insurance Act had nothing to say about that contest, especially since the Insurance Act did not specifically preclude the type of claim contemplated by the plaintiff. Consequently, she noted, at para. 16: "Any rights of the niece and nephew as volunteers are in equity subject to the prior rights of the plaintiff to the proceeds." She explained the relationship between the Insurance Act and equity, at para. 14 (p. 461 O.R.):

[E]quity imposes duties of conscience on parties based on their relationship and dealings one with another outside the purview of the statute. When he concluded the separation agreement with his wife, the deceased bound himself to maintain the policy in good standing, which he did; he also bound himself to maintain it for the benefit of his wife, which he did not.

[223] In declaring that the proceeds of the policy were impressed with a trust in the plaintiff's favour, McKinlay J. did precisely the task that equity set before her. She explained the outcome, at para. 17 (p. 461 O.R.):

The Insurance Act does not specifically preclude the existence of rights outside its provisions. It would go against all conscience to hold that because of the provisions cited, the plaintiff's clear rights under an agreement with her husband for good consideration can be taken away in favour of a niece and nephew who have given no consideration for those rights.

[224] This court cited Shannon with approval in Stewart v. Ryerson Polytechnic University and dismissed the appeal in Bielny v. Dzwiekowski, without reasons, affirming the decision below, which relied on Shannon: Bielny v. Dzwiekowski.

[225] As I see it, the purposes of an irrevocable designation under s. 191 of the Insurance Act are to creditor-proof the insurance proceeds, to permit them to pass outside the deceased's estate, to prevent the deceased from controlling the proceeds of the insurance policy and to prevent the deceased from making another designation.

[226] The critical point to observe is that the Act assumes the deceased had the right to irrevocably designate in the first place. By entering into a binding agreement with the appellant, the deceased relinquished his ability to designate another party as either a revocable or irrevocable beneficiary under his insurance policy, since a person cannot give what he does not have.

[227] I conclude that a purposive approach to the interpretation and application of s. 191 of the Insurance Act does not lead to the conclusion that the Act provides a juristic reason for the appellant to retain the insurance proceeds, contrary to the view of my colleague. As in Shannon, the respondent's right to the insurance proceeds was created by an agreement that preceded the designation and existed outside the provisions of the Insurance Act. The result is that, on the facts, the respondent's entitlement to the insurance proceeds as against the appellant is neither precluded nor affected by the operation of the Insurance Act. The case therefore falls outside the category of disposition of law as a juristic reason to permit the appellant to retain the life insurance proceeds.

The Second Step of the Juristic Reason Analysis

[228] In view of that conclusion, I must move on to consider the second step of the Garland analysis, in which, as Cromwell J. explained in Kerr, at paras. 43-44, the court assesses (1) "the reasonable expectations of the parties and public policy considerations to assess whether recovery should be denied"; (2) the "moral and policy-based arguments about whether particular enrichments are unjust"; and (3) [at para. 45] "questions regarding how (and when) factors relating to the manner in which the parties organized their relationship should be taken into account".

[229] I need not repeat the facts that led the application judge to find the oral agreement between the deceased and the respondent, which the appellant seeks to obviate. Neither the reasonable expectations of the parties nor public policy reasons support the appellant's success. Since both the appellant and respondent have a claim to the proceeds, this court must determine the outcome on equitable principles. In equity, the appellant, having provided no consideration for her designation as beneficiary, is merely a volunteer. Indeed, as McKinlay J. observed in Shannon, equity will not prioritize a gratuitously benefited volunteer like the appellant over a party with a prior right to an asset like the respondent. I would conclude that these factors favour the respondent, not the appellant.

This Court's Decision in Richardson Estate Does Not Determine the Outcome of This Appeal

[230] The appellant relies on this court's decision in Richardson Estate. There, the deceased had designated his former spouse as the beneficiary of a life insurance policy. Their separation agreement obliged the deceased to maintain her as the beneficiary under the policy until February 28, 1995. He told his second spouse that he would re-designate her as the beneficiary at a later date but he never did. The second spouse continued to pay the insurance premiums from the joint bank account she held with the deceased, wrongly believing she was the beneficiary.

[231] When the former spouse in Richardson Estate refused to release her claim to the death benefit, the second spouse brought a motion claiming a constructive trust over the death benefit. The motion judge found there was no unjust enrichment because the first spouse's designation as a beneficiary under the policy was a juristic reason for the second spouse's enrichment: Richardson Estate v. Mew. He recognized he had the power to rectify the policy, but he was not satisfied the evidence established the deceased's failure to change the beneficiary designation was the result of a mistake.

[232] This court Richardson Estate v. Mew found the separation agreement was not dispositive, since it only obliged the deceased to maintain the policy until the end of February 1995. The agreement contemplated that the amount of life insurance the deceased was to hold in favour of his first spouse in the future was to be determined in a later variation proceeding, but that never happened. Based on the case law, Gillese J.A., writing for this court, found, at para. 55:

A former spouse is entitled to proceeds of a life insurance policy if his or her designation as beneficiary has not changed. This result follows even where there is a separation agreement in which the parties exchange mutual releases and renounce all rights and claims in the other's estate. General expressions of the sort contained in releases do not deprive a beneficiary of rights under an insurance policy because loss of status as a beneficiary is accomplished only by compliance with the legislation. The general language used in waivers and releases does not amount to a declaration within the meaning of the Insurance Act.

[233] This court agreed with the motion judge that it would not be unjust for the first spouse to take the death benefit, noting, at para. 61: "[T]here is a juristic reason for Ms. Mew's enrichment -- she was the named beneficiary of the Policy."

[234] I would distinguish this case from Richardson Estate on several grounds, despite some factual similarities. First, and most importantly, there was no contractual obligation in Richardson Estate to his second spouse that he would change the beneficiary designation on the life insurance policy. Indeed, the motion judge in Richardson Estate stated specifically, at para. 61:

. . . I am unable to say, with anything approaching the level of confidence necessary to grant rectification, that the evidence establishes that the continuation of Stephenie as the beneficiary of the policy was a mistake and that Michael's true intent was to name Anne as the beneficiary.

The motion judge added, at para. 64, that "the evidence does not satisfy me that the designation of Stephenie was a result of a mistake or that Michael had a contrary intent". This is a long way from the application judge's conclusion in this case that there was an oral contract.

[235] Second, there was no impediment to a re-designation of the beneficiary in Richardson Estate, such as the oral agreement in this case.

[236] Third, the respondent made a much more robust contribution to the life insurance policy by way of premiums than her counterpart in Richardson Estate.

[237] Accordingly, I would not consider the outcome of Richardson Estate as dispositive of the matter in the appellant's favour.

[238] In my view, s. 191 of the Insurance Act does not provide a juristic reason to oust the availability of a constructive trust over the life insurance proceeds based on the Supreme Court's decision in Soulos, despite the Richardson Estate decision.

The Disappointed Beneficiary Cases

[239] In this section of the reasons, I consider the disappointed beneficiary cases. They are not on all fours with this case, but resemble it in material details. Many of them have taken the unjust enrichment route to a constructive trust, so it is not surprising that this case proceeded as it did.

[240] I begin with two pertinent observations. First, I note that the disappointed beneficiary cases fit rather awkwardly into the structure of the unjust enrichment analysis. The typical pattern in familial unjust enrichment cases involves a dispute between spouses in which the plaintiff seeks to acquire a property interest in something owned by the defendant. This is a two-party structure that makes the plaintiff's restitutionary claim especially compelling, since the plaintiff seeks a remedy directly from the defendant who has been enriched at the plaintiff's expense. These cases include Pettkus, Sorochan v. Sorochan, Rathwell, Peter and Kerr.

[241] However, this case falls into a set of fairly common disappointed beneficiary cases concerning the right to life insurance proceeds. Typically, the plaintiff pursues the life insurance proceeds on the death of the deceased former spouse, in a situation where the deceased had named a new beneficiary, often a new spouse, who is the defendant. The disappointed beneficiary cases are different in that they take effectively a three-party structure: the plaintiff's target is the defendant beneficiary and not the deceased. The claim against the defendant, who is often innocent of any involvement in the deceased's actions, seems less compelling than it does in the normal two-party dispute between spouses.

[242] I observe that in none of the disappointed beneficiary cases, except for Love, at para. 42, did the court advert to the fact that what was at issue was not the typical two-party relationship in an unjust enrichment case between the plaintiff and the defendant, but a three-party relationship. The cases utilize the rubric of unjust enrichment, but the structural difference is glossed over.

[243] Second, the disappointed beneficiary cases constitute a type of case that, to my mind, fits comfortably with those in the third and fourth categories I mentioned earlier, described in Soulos, as "situations where constructive trusts have been recognized in the past", or "other situations where courts have found a constructive trust" or "circumstances where its availability has long been recognized": paras. 21-22 and paras. 34-35. I return to this discussion below after discussing the cases in order to better set the context.

The Categories of Disputed Beneficiary Cases

[244] In general terms, the disappointed beneficiary cases fall into three categories. The first are those in which the plaintiff succeeds in securing a constructive trust over the insurance proceeds. These include Steeves v. Steeves; Roberts v. Martindale; Bielny; Mitchell v. Clarica Life Insurance Co.; Schorlemer Estate v. Schorlemer; and Holowa Estate v. Stell-Holowa. The list is not exhaustive.

[245] The second category are those dismissing the plaintiff's claim to a constructive trust, taking Soulos into account. This category includes Ladner v. Wolfson and Milne Estate v. Milne, which followed Ladner.

[246] The third category of disappointed beneficiary cases are those in which the plaintiff was unsuccessful in securing a constructive trust, largely on the basis that the deceased's intention to benefit the plaintiff was not sufficiently clear. These include Kang v. Kang; Richardson Estate; Chanowski v. Bauer; and Love. This category of cases is not relevant.

[247] The appellant argues that the cases in the first category have been overtaken by Soulos and are not good law. I will attend to this argument after describing cases from each category.

Cases Granting a Remedial Constructive Trust

[248] Shannon is a pre-Soulos decision. The details were discussed above. Despite what she implicitly found was a revocable insurance designation, McKinlay J. imposed a trust on the life insurance proceeds, relying on a provision in the separation agreement. I see no meaningful difference between the effects of the words used in the separation agreement in Shannon and the oral agreement found by the application judge in this case.

[249] In Steeves, the husband was ordered by the court to maintain a life insurance policy in favour of his former wife. Before his death, he designated his second wife as the beneficiary of the policy. Since the policy was provided as an employment benefit to the deceased, the plaintiff had not contributed to it. Steeves is also pre-Soulos, but the trial judge invoked the Supreme Court jurisprudence that led to Soulos including Rathwell, Pettkus and Peter in reaching the conclusion that a constructive trust was warranted.

[250] Creaghan J. noted, at para. 31, that the deceased's failure to comply with the court's order was wrongful. The estate lacked assets, pointing to "the absence of an adequate remedy to the plaintiff at law". He also noted that the defendant "is not a bona fide purchaser for value, but rather, is in the nature of what has been referred to at law as an innocent volunteer". He concluded, at para. 35, that the three-part test for unjust enrichment from Pettkus was met. Finally, at para. 36, he described the constructive trust impressed on the policy proceeds:

The constructive trust that arises in this instance is that Roger Steeves held the inchoate proceeds of the insurance policy in the event of his death in trust for the Plaintiff and that upon his death, where the proceeds could be traced and had not come into the hands of a bona fide purchaser for value, an equitable lien attached to them in the hands of the Defendant Dorilla Harper Steeves.

[251] The facts giving rise to a constructive trust in Roberts differ. The husband renounced his rights in his former wife's property. He remained the policy beneficiary when his former wife died because she did not take steps to change the designation to her sister and caregiver, Ms. Roberts, whom she wished to be the beneficiary. The British Court of Appeal did not permit the insurance designation to stand in the way of imposing a constructive trust. Speaking for the court, Southin J.A. stated, at paras 24-27:

Constructive trusts have been imposed in many cases where the defendant has done nothing which could properly be characterized as "fraudulent". There was no fraud, for instance, in Pettkus v. Becker, or in Hussey v. Palmer, in which Lord Denning spoke of imposing a trust "whenever justice and good conscience require it".

There is in this proposition the very great danger of judges invoking their personal predilections. "Good conscience" is an infinitely variable concept upon which reasonable men and women, and therefore reasonable judges, may have widely divergent opinions.

But I am comfortable in this case in saying that it would be against good conscience for the appellants to keep this money because Mr. Martindale had, by the separation agreement, surrendered any right he might have had to the property of the deceased. A policy of life insurance is a species of property of the insured, albeit the amount payable under the contract of insurance does not fall into possession until the insured's death, and, by law, cannot be taken by the insured's creditors.

For the appellant, Mr. Martindale, to claim from the insurer the proceeds was a breach of the separation agreement and such a breach is sufficient, in my opinion, to call in aid the doctrine of the remedial constructive trust. To put it another way, it is not the mistaken belief of Mrs. Martindale which gives rise to a remedy; it is the bargain which Mr. Martindale made.

[252] The court did not refer to Soulos.

[253] The Alberta Court of Queen's Bench reached a similar result in Holowa Estate. Ms. Hanrahan was the estranged spouse of the deceased. She was defined legislatively as his "pension partner" under the applicable pension plan and was entitled to the proceeds on his death. The deceased had failed to get a matrimonial property order that would have changed Ms. Hanrahan's status. The application judge found, at para. 50, that

[Ms. Hanrahan] could hardly have more explicitly waived her right to the [Local Authorities Pension Plan], she did it both in the context of the 2001 agreement during her marriage, and again in the 2003 agreement on separation, where she expressly released any rights arising on death, or out of any Act of the Province of Alberta.

[254] There was also evidence that the deceased attempted to designate his two children as his beneficiaries for the pension, who were also his estate beneficiaries.

[255] The application judge undertook a lengthy exposition of the law of constructive trust, including Soulos. He imposed a constructive trust on the pension benefit in favour of the estate on the basis of unjust enrichment. He then went on to consider the alternative claim advanced, as he said, at para. 62, "under constructive trust principles, independent of unjust enrichment". He considered para. 43 of Soulos, where he noted: "Expansive language is used in relation to the ability of this equitable remedy to develop." He concluded, at para. 65:

Surely it would be against good conscience for Hanrahan to retain this benefit, when there is no evidence to show any equitable basis for her to receive the benefit, or be entitled to it in contravention of the two matrimonial contracts between the parties. She specifically waived her right to the LAPP, it was a specific contractual agreement, during the marriage, affirmed again on separation.

[256] There are several Ontario Superior Court cases that follow Shannon and do not mention Soulos. These are Bielny (separation agreement prevails over designation), Mitchell (original common intention respecting beneficiaries prevails) and Schorlemer Estate (separation agreement prevails). Remedial constructive trusts were imposed in these cases.

Cases Refusing a Remedial Constructive Trust

[257] Ladner is unusual, in that it was a lawyer's negligence case. The wife sued her lawyer and the law firm handling her case for failing to pursue a trust claim for a remedial constructive trust against her former husband's estate at the same time as a damages claim. The trust claim was merged in the damages claim, but by that time the husband's estate was insolvent. The defendants asserted that a trust remedy would not have been available in the action against her husband's estate, so the lawyer had not been negligent. The trial judge found that a constructive trust claim would have succeeded, but the court of appeal reversed the decision.

[258] In the underlying fact situation, the husband failed to keep in good standing the existing life insurance policies designating his former spouse, as he was obliged to do under a separation agreement. Instead, he purchased other life insurance policies and designated his estate as the beneficiary. The Court of Appeal considered the former spouse's claim to fall under the "wrongful act" aspect of Soulos. The trial judge had found there was no fiduciary relationship between the spouses, and the Court of Appeal accepted that finding. The result, Garson J.A. noted, at para. 47, was that "the first of the Soulos conditions for the imposition of a good conscience trust [resulting from a wrongful act] cannot be fulfilled".

[259] Justice Garson went on to find, at para. 53, there was "an insufficient proprietary connection between the wrongful act and the existing life insurance policies". The life insurance proceeds at issue were not proven to be generated from the policies referred to and required under the separation agreement. As a result, there was an insufficient proprietary connection between the wrongful act and the existing life insurance policies. Garson J.A. considered a proprietary connection to be essential, since to grant the remedy of a constructive trust without such a connection "would risk extending equitable remedies to almost any common law breach of contract": Ladner, at para. 52.

[260] The trial judge, said Garson J.A. at para. 57, had wrongly relied on Roberts as having "extended the law as enunciated in Soulos so as to permit the imposition of a constructive trust" in Ms. Ladner's favour. This was in error, since Soulos was not cited in Roberts, and for this reason he cast doubt on its correctness: Ladner, at paras. 58-59. She then distinguished Roberts on the basis that, having surrendered any claim to the property of his former wife, the husband in Roberts "was therefore not entitled to receive the property over which the remedy of constructive trust was sought".

[261] In Love, the contest was between the former wife, who was designated under his insurance policy, and the deceased's son, whom the deceased had wanted to designate. The Saskatchewan Court of Appeal found sufficient evidence to support rectification. In obiter, Richards J.A. reflected on the appellant's unjust enrichment argument that was based on the British Columbia Court of Appeal's decision in Roberts. He observed, at para. 45, that Roberts had not been followed by any other appellate-level court, including this court in Richardson Estate. He noted, at paras. 46-47, that his difficulty with Roberts is that it made no reference to Soulos, which had not invited judges to impose constructive trusts "whenever and wherever they believe justice requires".

[262] In Milne Estate, the deceased husband disobeyed a court order requiring him to maintain a life insurance policy with the wife as beneficiary. Instead, he designated his new partner. The court followed Ladner and found no basis for a remedial constructive trust based on breach of fiduciary duty, as there was no such relationship between the spouses. Instead, he granted judgment against the estate for the full value of the policy.

Analysis

[263] The disappointed beneficiary cases utilize the rubric of unjust enrichment but gloss over the structural difference. This approach exemplifies the common law's inclination to use old tools for new tasks, even if they do not quite fit. We will always be feeling our way. As McLachlin J. noted in Soulos, at para. 35: "The goal is but a reasoned, incremental development of the law on a case-by-case basis." That method is part of the genius of the common law.

[264] In managing the deployment of remedial constructive trusts in aid of good conscience, the discipline of particularity is important, as McLachlin J. noted. To repeat her words, in para. 35 of Soulos: "Particularity is found in the situations in which judges in the past have found constructive trusts." She noted: "A judge faced with a claim for a constructive trust will have regard not merely to what might seem fair in a general sense, but to other situations where courts have found a constructive trust." She saw this careful approach as essential to "a reasoned, incremental development of the law on a case-by-case basis".

[265] I recapitulate the findings that must be made for a court to impose a constructive trust on life insurance proceeds, which have emerged so far in the cases involving disappointed beneficiaries. These serve as limits to discipline judicial discretion. First, the defendant has been enriched and the plaintiff deprived in a family context, not in the market world. Second, the deceased's ruling intent, before resiling, was to benefit the plaintiff. That intent can be found in an oral agreement, a separation agreement or in a court order, but it must comprise an obligation. Third, there is a proprietary link between the plaintiff and the life insurance proceeds. It is this life insurance policy that is in issue, not some other. Finally, providing the plaintiff with the remedy of a constructive trust does not breach any law. Experience with constructive trusts in the disappointed beneficiary context would undoubtedly add other refinements.

[266] The disappointed beneficiary cases represent a distinct type of case in which the constructive trust remedy is disciplined by the common structure and elements of the dispute, which ought to serve to assuage the concern that equity is off on a frolic of its own, paying no attention to the law. Equity follows the law; the imposition of a constructive trust does not block the law's operation, which in this case is the operation of the Insurance Act; it imposes an obligation in conscience on the appellant the moment her entitlement to the proceeds attaches, one that requires her to hold the proceeds in trust for the respondent.

[267] To my mind, the disappointed beneficiary cases constitute a genus in which a constructive trust can be imposed on life insurance proceeds consistently with the reasoning of McLachlin J. in Soulos. They are situations in which courts have found a constructive trust.

[268] I do not agree with the appellant's argument that those disappointed beneficiary cases in which the court granted a remedial constructive trust have been overtaken by Soulos and are not good law, for several reasons.

[269] First, I am not persuaded by the obiter in Ladner and Love that the British Columbia Court of Appeal's reasoning in Roberts is to be doubted in light of Soulos. Neither court explained the basis for the doubt apart from the lack of a mention of Soulos in the decisions under review.

[270] Second, I would distinguish Ladner itself on two bases. The case proceeded under the rubric of wrongful act, which has a detailed set of elements according to Soulos that do not apply to the unjust enrichment rubric. Further, the proprietary connection between the new and old insurance policies that was missing in Ladner was very much present in this case between the policy and the payment of the premiums by the respondent.

[271] Finally, in light of the limits I discussed earlier, I do not share the Ladner court's underlying concern that constructive trusts would become unmanageable.

[272] Equity asks a pertinent question in the difficult dilemma posed in the disappointed beneficiary cases: Which of the two claimants has the superior claim to the life insurance proceeds? Equity's answer, all things being equal, is to assist the one with the superior right in equity, as McKinlay J. pointed out in Shannon. In this case, that is the respondent, as the application judge found.

Conclusion on Unjust Enrichment

[273] In my view, the application judge did not err in finding that the respondent's deprivation consisted of the life insurance proceeds. He also did not err in finding that the deceased's irrevocable designation in the appellant's favour under s. 191 of the Insurance Act, given his prior oral agreement with the respondent, did not provide a juristic reason to oust the availability of a constructive trust over the life insurance proceeds, despite the Richardson Estate decision.

[274] But I would go further and add that, to the extent that they fit awkwardly under the rubric of unjust enrichment, the disappointed beneficiary cases are perhaps better understood as a genus of cases in which a constructive trust can be imposed via the third route in Soulos -- circumstances where the availability of a trust has previously been recognized -- and the fourth route -- where good conscience otherwise demands it, quite independent of unjust enrichment.

Disposition

[277] I would dismiss the appeal with no order as to costs.

Appeal allowed.