Court File and Parties

COURT FILE NO.: CV-22-88458 DATE: 2024/10/22

ONTARIO SUPERIOR COURT OF JUSTICE

BETWEEN:

Greg Burwell, The Burwell Family Trust and 8442568 Canada Inc. Applicants – and – Marta Wozniak Respondent

Counsel: Anne Tardif, James Plotkin, and Miriam Vale Peters for the Applicants Geoffrey Cullwick and Ronald Prehogan for the Respondent

HEARD: April 19, 2024

REASONS FOR JUDGMENT

Rees J.

I. Overview

[1] The applicant Greg Burwell and the respondent, Marta Wozniak, were in a volatile relationship. At one point the couple separated. Mr. Burwell wanted to reconcile. In an email, he promised Ms. Wozniak shares in his company, which would be given effect through a family trust agreement.

[2] The couple later separated permanently. Before the sale of Mr. Burwell’s company, Ms. Wozniak claimed an interest in the shares. Her claim was based in trust and, in the alternative, proprietary estoppel. The parties agreed to arbitrate Ms. Wozniak’s claim after the sale.

[3] The arbitrator found in favour of Ms. Wozniak. The arbitrator held that Mr. Burwell’s email constituted an express trust that preceded the creation of the Burwell Family Trust. This court set aside the trust award in an earlier appeal and remitted Ms. Wozniak’s proprietary estoppel claim to the arbitrator.

[4] The applicants now appeal the arbitrator’s proprietary estoppel award. The arbitrator held that Ms. Wozniak was entitled to 50% of all amounts paid for corporate shares held personally by Mr. Burwell and by the Burwell Family Trust based on proprietary estoppel, which the arbitrator found arose from Mr. Burwell’s email.

[5] The applicants ask that I set aside the arbitrator’s proprietary estoppel award. I agree that the award must be set aside. The award impermissibly departs from Ontario law by holding that proprietary estoppel can apply beyond interests in land. That is not presently the law in Ontario.

[6] Even if proprietary estoppel were to apply to other forms of property, such as shares, the arbitral decision contains two errors. The email from Mr. Burwell on which Ms. Wozniak rests her claim of proprietary estoppel is internally inconsistent. The nature of the promise was unclear and ambiguous. The promise was also conditional on the execution of a family trust agreement. This family trust agreement modifies, and ultimately overtakes, the promises in the email. As a result, the email cannot properly ground a claim for proprietary estoppel.

[7] The appeal is allowed and Ms. Wozniak’s claim is dismissed.

II. Background facts

[8] Mr. Burwell and Ms. Wozniak were in a relationship from September 2010 until January 1, 2014. In June 2011, Mr. Burwell and his business partner launched Fusebill Inc., which developed and marketed subscriber-based billing management software.

[9] In September 2011, Mr. Burwell hired a lawyer to establish a family trust. About a month later, Mr. Burwell and Ms. Wozniak separated. Wishing to reconcile, Mr. Burwell sent Ms. Wozniak an email on October 18, 2011, which is now central to their dispute. In it, Mr. Burwell wrote:

If you can ever find it in your beautiful soul to forgive me and move forward, I vow to you the following… this is my list of actions:

(this one is in progress) I am appointing you the sole beneficiary of the family trust which will hold all Fusebill shares, with you being the sole beneficiary you are entitled to 50% of the holdings if I am alive, if not you are entitled to 100%, once we sign this document 50% of the 4.5 million shares of Fusebill are yours regardless of what happens. You are right, you have been by my side for this and until today I didn’t realize how much of a rock you were for me, so I consider you to be cofounder. So you have earned it.

I will support all of this with a will and the legal trust documents.

[10] On December 16, 2011, the couple signed a trust agreement, creating the Burwell Family Trust. It appointed Mr. Burwell and Ms. Wozniak as co-trustees. Mr. Burwell’s business partner was the settlor of the family trust, which he settled by transferring $20 to the co-trustees. The family trust’s subject matter consisted of this $20, plus whatever property might later be assigned to the trustees. The family trust’s beneficiaries were Mr. Burwell, Ms. Wozniak, any of their future children, and several family members.

[11] The beneficiaries’ shares were not specified and were at the trustees’ discretion. The trust agreement provides that the trustees powers and discretion “are absolute and not to be controlled or reviewed”.

[12] The family trust subscribed for 4.5 million Fusebill Class ‘A’ Common Shares. Mr. Burwell continued to hold 4.5 million Class ‘A’ Common Shares of his own, and he subscribed for 231,525 more in the spring of 2012. He never transferred his personally held shares to the family trust.

[13] In March 2013, Fusebill reorganized. Mr. Burwell’s 4,731,525 shares were converted into 1,737,203 shares. The 4.5 million held by the family trust shares were converted into 450,000 shares. This reorganization is not disputed by the parties.

[14] Mr. Burwell later transferred the shares he held personally to his holding company, the applicant 8442568 Canada Inc.

[15] Mr. Burwell and Ms. Wozniak separated on January 1, 2014.

[16] Just over seven years later, with a sale of Fusebill pending, Ms. Wozniak asserted a 50% interest in Mr. Burwell’s personal shares and the shares held by the family trust. Mr. Burwell disputes Ms. Wozniak’s claim.

[17] To allow the Fusebill sale to proceed without prejudice to Ms. Wozniak’s claim, she and Mr. Burwell entered into an interim agreement concerning her claim. The interim agreement provided for mediation, and if unsuccessful, arbitration.

[18] The mediation was unsuccessful and the parties proceeded to arbitration. Ms. Wozniak’s claim was based in trust and, in the alternative, proprietary estoppel.

The trust award

[19] On January 1, 2022, the arbitrator delivered his first award. He held that the October 18 email was a valid declaration of trust. He found that the email evidenced Mr. Burwell’s intention to create a family trust and that the Family Trust Agreement, with some modifications, formalized Mr. Burwell’s promises and intentions. Accordingly, the arbitrator found that Ms. Wozniak was entitled to 50% of all amounts paid for the Fusebill shares held personally by Mr. Burwell and by the family trust.

The Superior Court set aside the trust award

[20] Mr. Burwell appealed the trust award to this court. On March 14, 2023, the court set aside the trust award: Burwell v. Wozniak, 2023 ONSC 1685. Justice Jensen held that the arbitrator erred in law by permitting the October 18 email to overwhelm the terms of the Burwell Family Trust. Justice Jensen held that the arbitrator essentially created a new trust based on the email, which conflicted with the Burwell Family Trust agreement. She held that there was only one trust: the Burwell Family Trust. Justice Jensen concluded that the email did not create a valid trust. As a beneficiary of the family trust, Ms. Wozniak has a beneficial interest in the trust shares.

[21] Justice Jensen remitted the proprietary estoppel issue to the arbitrator, directing that he make the appropriate final disposition, “taking into account my decision with respect to the trust issue”.

The proprietary estoppel award

[22] The arbitrator released his proprietary estoppel award on August 30, 2023. He held that the doctrine of proprietary estoppel was not limited to interests in land and could apply to Ms. Wozniak’s claim. In doing so, he considered the leading case on proprietary estoppel in Canada, Cowper‑Smith v. Morgan, 2017 SCC 61, [2017] 2 S.C.R. 754. The arbitrator held that although the Supreme Court of Canada suggested that proprietary estoppel is commonly understood to be concerned with interests in land, it refrained from determining the issue in Cowper-Smith. He explained that he did not read the case or any other jurisprudence as limiting proprietary estoppel to an interest in land or restricting the definition of property to real property.

[23] The arbitrator determined that Ms. Wozniak had established a proprietary estoppel based on the October 18 email. Although the arbitrator acknowledged that the email “may have ‘internal inconsistencies’ and ‘contradictions’”, he also found that the email “is unequivocal as the underlying intent and promise are clear”. The arbitrator reasoned that, despite an absence of specificity about what constituted 50% ownership of the Fusebill shares that Mr. Burwell owned at the time or would own in the future, Mr. Burwell intended Ms. Wozniak to have 50% of them.

[24] The arbitrator awarded Ms. Wozniak 50% of all amounts paid for corporate shares held personally by Mr. Burwell and by the Burwell Family Trust based on proprietary estoppel. Indeed, he concluded that “any remedy pursuant to the doctrine of proprietary estoppel applies in favour of Marta and her claim should be allowed as ordered in [the trust award dated] January 2022”.

III. Issues

[25] The appeal raises the following issues:

a. Should leave to appeal be granted?

b. What is the applicable standard of review?

c. Did the arbitrator err in law or mixed fact and law in concluding that Ms. Wozniak was entitled to 50% of all amounts paid for the Fusebill shares held personally by Mr. Burwell and by the family trust?

IV. Analysis

Leave is granted

[26] This is an appeal under s. 45 of the Arbitration Act, 1991, S.O. 1991, c. 17. Under the arbitration agreement, the parties agreed that they could appeal a question of law with leave, and may appeal a question of fact or a question of mixed fact and law as of right.

[27] There is therefore no question that the applicants can appeal the application of the principles of proprietary estoppel to the facts of the case. But because the applicants dispute that proprietary estoppel can even apply, as a matter of law, they must seek leave on this issue. This is a purely legal question.

[28] Under s. 45(1) of the Arbitration Act, 1991, S.O. 1991, c. 17 the court shall grant leave only if it is satisfied that (a) the importance to the parties of the matters at stake in the arbitration justifies an appeal; and (b) determination of the question of law at issue will significantly affect the parties’ rights. The first criterion is met where the award has a “significant impact” on a party’s financial position: McAsphalt Marine Transport Ltd. v. Liberty International Canada (2005), 22 C.C.L.I. (4th) 129, at para. 34; and Ledore Investments Ltd. v. Ellis-Don Construction Ltd., 2015 ONSC 6536, at para. 14. The second criterion is met where correcting the identified errors would decisively alter the outcome: McAsphalt, at para. 35; Ledore, at para. 15.

[29] Ms. Wozniak does not dispute that the applicants have met the test for leave. I agree. The arbitrator’s award of roughly $1.5 million significantly impacts the applicants’ financial position. Correcting the errors alleged would decisively alter the outcome: if the applicants are successful, it would entirely defeat Ms. Wozniak’s claim.

Appellate standards of review apply

[30] The applicants argue that appellate standards of review apply rather than reasonableness review. The respondent disagrees.

[31] In Sattva Capital Corp. v. Creston Moly Corp., 2014 SCC 53, [2014] 2 S.C.R. 633, at para. 106, the Supreme Court held that the standard of review on appeals from private arbitrations restricted to questions of law will be reasonableness unless the question is one that would attract the correctness standard, such as constitutional questions or questions of law of central importance to the legal system as a whole and outside the arbitral tribunal’s expertise. The Court confirmed this in Teal Cedar Products Ltd. v. British Columbia, 2017 SCC 32, [2017] 1 S.C.R. 688.

[32] Then came the Supreme Court’s reset of its standard of review jurisprudence in administrative law in Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65, [2019] 4 S.C.R. 653. In Vavilov, the Supreme Court held that the appellate standards of review apply to statutory appeals of administrative decisions, at para. 17.

[33] Vavilov concerned judicial review, not arbitration. So far, a majority of the Supreme Court has not clarified whether courts should continue applying Sattva reasonableness review to arbitral awards or should follow Vavilov and apply the appellate standards of review to arbitral appeals: Wastech Services Ltd. v. Greater Vancouver Sewerage and Drainage District, 2021 SCC 7, [2021] 1 S.C.R. 32. While noting, at para. 45, that “Vavilov does not advert either to Teal Cedar or Sattva, decisions which emphasize that deference serves the particular objectives of commercial arbitration”, the majority in Wastech left the issue for consideration for a future appeal. In doing so, the majority cautioned that it should not be understood as agreeing with the minority’s view on the standard of review. I will return to the minority’s view below.

[34] There is no clear consensus in the Court of Appeal for Ontario on whether Sattva remains good law after Vavilov.

[35] One panel of the Court of Appeal held that correctness applied to a question of law decided by an arbitrator: Travelers Insurance Company of Canada v. CAA Insurance Company, 2020 ONCA 382, 151 O.R. (3d) 488, at para. 14. Travelers Insurance considered an arbitrator’s decision regarding statutory accident benefits under s. 268 of the Insurance Act, R.S.O. 1990, c. I.8. Although these arbitrations are statutorily mandated under the Insurance Act, they are conducted under the framework of the Arbitration Act, 1991, S.O. 1991, c. 17: Insurance Act, s. 267.8(11). In holding that the correctness standard applied, the court relied on Vavilov. The court also held that, in any event, the errors were constitutional, jurisdictional, and exceptional — in respect of which deference is not due. Further, it held that the standard of correctness has always applied to questions of law. The court did not discuss whether Sattva and Teal Cedar remained good law, however.

[36] A later panel of the Court of Appeal declined to answer whether Sattva continues to apply to arbitral awards or whether appellate standards of review should apply after Vavilov: Ontario First Nations (2008) Limited Partnership v. Ontario Lottery and Gaming Corporation, 2021 ONCA 592, at para. 37. Since then, another panel of the Court of Appeal did not expressly consider the standard of review but applied the reasonableness standard from Sattva to the arbitrator’s interpretation of a contract: Tall Ships Development Inc. v. Brockville (City), 2022 ONCA 861, 476 D.L.R. (4th) 500, at para. 40. Neither of the Court of Appeal’s later decisions considered Travelers Insurance.

[37] Elsewhere, courts of appeal in Alberta and the Northwest Territories have held that appellate standards of review apply after Vavilov: Northland Utilities (NWT) Limited v. Hay River (Town of), 2021 NWTCA 1, 456 D.L.R. (4th) 278, at para. 44; TransAlta Corporation v. Alberta (Utilities Commission), 2021 ABCA 232, at para. 27. But there is no appellate consensus yet: lululemon athletica canada inc. v. Industrial Color Productions Inc., 2021 BCCA 108, at para. 33 (characterizing the question as “unsettled at the appellate level”).

[38] This court has applied reasonableness after Vavilov in a case where the parties agreed that the standard of review was reasonableness: Taseko Mines Limited v. Franco-Nevada Corporation, 2023 ONSC 2055, 166 O.R. (3d) 600, at paras. 46-52.

[39] With respect to those of a different view, I will apply the appellate standards of review.

[40] There is no clear consensus in the Court of Appeal. Travelers Insurance gives some direction, however. Although the appeal did not arise from a private arbitration, the arbitration was governed by the Arbitration Act, 1991, S.O. 1991, c. 17. From a statutory interpretation perspective, this distinction should not lead to a different result.

[41] In my view, Sattva and Teal Cedar are no longer binding since there has been a significant development in the law since they were released: Canada (Attorney General) v. Bedford, 2013 SCC 72, [2013] 3 S.C.R 1101, at para. 42; Carter v. Canada (Attorney General), 2015 SCC 5, [2015] 1 S.C.R. 331, at para. 44. Specifically, the standard of review framework on which Sattva was built has changed. Sattva expressly borrowed from the Supreme Court’s judicial review framework developed in Dunsmuir v. New Brunswick, 2008 SCC 9, [2008] 1 S.C.R. 190: Sattva, at paras. 104-106. But Vavilov reconsidered the court’s Dunsmuir framework. Vavilov sought to simplify and bring coherence to the standard of review jurisprudence. It also put judicial review on a firmer theoretical footing by respecting legislative intention.

[42] The revised standard of review analysis begins with a presumption that reasonableness is the applicable standard in all cases. This presumption is rebutted where the legislature explicitly prescribes the applicable standard of review, or where—as here—the legislature has provided a statutory appeal mechanism. In providing for a statutory appeal, the legislature “has subjected the administrative regime to appellate oversight and indicated that it expects the court to scrutinize such administrative decisions on an appellate basis”: Vavilov, at para. 36. The appropriate standard of review is then determined on the appellate standard.

[43] The arbitral context does not change the need to give effect to the legislature’s intention in creating a statutory appeal mechanism. I agree with the reasons of Brown and Rowe JJ. (Côté J. concurring) in Wastech Services, at paras. 119-120. The differences between arbitration and administrative decision-making “do not… affect the standard of review where the legislature has provided for a statutory right of appeal. Appellate standards of review apply as a matter of statutory interpretation.” As they explain:

Factors that justify deference to the arbitrator, notably respect for the parties’ decision in favour of alternative dispute resolution and selection of an appropriate decision‑maker, are not relevant to this interpretive exercise. What matters are the words chosen by the legislature, and giving effect to the intention incorporated within those words.

[44] Applying the appellate standards of review gives effect to the legislature’s intention under the Arbitration Act, 1991, S.O. 1991, c. 17. The appeal provision in s. 45 expressly uses the word “appeal”. As the majority observed in Vavilov, at para. 44, “there is no convincing reason to presume that legislatures mean something entirely different when they use the word “appeal” in an administrative law statute than they do in, for example, a criminal or commercial law context”. Then why assume the legislature means something different when it uses the word “appeal” in the arbitral context?

[45] The strongest counterarguments against applying appellate standards of review are policy arguments. These policy arguments suggest that applying appellate standards of review would undermine the central aims of arbitration—efficiency and finality—and would not respect the relative expertise of arbitrators, chosen by the parties, over courts: Teal Cedar, at para. 1; and Sattva, at para. 105.

[46] But these policy arguments are rebutted, where—as here—the parties have agreed to an appeal to the superior court in their arbitration agreement. Applying appellate standards of review respects the party autonomy principle inherent in s. 3 of the Arbitration Act, 1991, S.O. 1991, c. 17. Section 3 allows parties to contract out of appeals altogether, except in family arbitrations. This leaves it up to the parties to decide whether the arbitral tribunal will have the last word. Here, the parties agreed that they could appeal a question of law with leave, and could appeal a question of fact or a question of mixed fact and law as of right. This agreement should be given effect by applying the appellate standards of review: see Northland Utilities, at para. 43.

[47] Even if I were wrong, it would not alter the outcome. The award cannot withstand review on a reasonableness standard either.

The parties’ positions on the merits

[48] The applicants advance three grounds of appeal. First, the applicants argue that the arbitrator erred in law in extending proprietary estoppel beyond interests in land. Second, the applicants contend that the arbitrator erred in finding that an inconsistent and contradictory email constituted an unequivocal promise, and further erred by failing to recognize that the promise was conditional on the execution of the family trust agreement. Third, the applicants argue that even assuming proprietary estoppel is available and made out on the facts, the arbitrator failed to apply the required test in selecting a remedy. The applicants contend that the arbitrator impermissibly constrained the discretion of the Burwell Family Trust trustees without notice to the other discretionary beneficiaries, over whom he lacked jurisdiction.

[49] In response, Ms. Wozniak contends that the Supreme Court opened the door in Cowper-Smith to apply proprietary estoppel beyond interests in land. She also argues that the arbitrator correctly applied the principles of proprietary estoppel to the facts. Finally, she argues that the arbitrator reasonably exercised his discretion in crafting an appropriate remedy.

The arbitrator erred in law in extending proprietary estoppel beyond interests in land

[50] The question of whether the law of Ontario applies proprietary estoppel to interests other than in land is a pure question of law. The appellate standard of review is therefore correctness.

[51] In my view, the arbitrator erred in law in extending proprietary estoppel to the shares.

[52] As discussed, the Supreme Court considered proprietary estoppel in Cowper-Smith. In that case, it set out the elements required to establish a claim of proprietary estoppel. It also held that in Canadian law, proprietary estoppel is generally understood to apply to interests in land. The Court noted academic criticism that this limitation to land may be arbitrary and that English courts have gone “much further”, allowing proprietary estoppel claims against other forms of property. Ultimately, the court concluded that it did not need to decide whether proprietary estoppel may attach to an interest in property other than land: see paras. 21-22. The court neither opened nor shut the door. It simply left the resolution of the question for another day.

[53] The Court of Appeal for Ontario has consistently stated the test for proprietary estoppel as requiring an interest in land: Oakville (Town) v. Sullivan, 2021 ONCA 1, 10 M.P.L.R. (6th) 163, at para. 32; Clarke v. Johnson, 2014 ONCA 237, 318 O.A.C. 186, at para. 52; Schwark v. Cutting, 2010 ONCA 61, 261 O.A.C. 262, at para. 34; and Eberts v. Carleton Condominium Corporation No 396 (2000), 136 O.A.C. 317, at para. 23. So too have appellate courts in other provinces: see e.g., Sabey v. Rommel, 2014 BCCA 360, 64 B.C.L.R. (5th) 221, at para. 32; Burgsteden v. Long, 2014 SKCA 115, 446 Sask. R. 207, at para. 25; Bellton Farms Ltd. v. Campbell, 2016 NSCA 1, 369 N.S.R. (2d) 302, at para. 46.

[54] True, in none of the Ontario cases did the Court of Appeal expressly hold that proprietary estoppel could only apply to an interest in land. Still, at some point, the absence of appellate authority applying proprietary estoppel beyond interests in land signals the Court of Appeal’s understanding that it only applies to land. In my view, the Court of Appeal’s view of proprietary estoppel is binding.

[55] I now turn to three cases which Ms. Wozniak contends support the extension of proprietary estoppel to other types of property.

[56] The arbitrator relied on Aiello et al. v. Bleta et al., 2022 ONSC 2798, 79 E.T.R. (4th) 62. In Aiello, the court applied proprietary estoppel to a transfer of corporate shares. This was overturned by the Court of Appeal on the basis that the claim was statute-barred: Aiello v. Bleta, 2023 ONCA 525, 88 E.T.R. (4th) 198, at para. 38. As a result, Aiello cannot stand as authority for proprietary estoppel extending beyond land. Although released before the arbitrator’s award, the Court of Appeal’s decision does not appear to have been brought to his attention.

[57] In this court, Ms. Wozniak relies on Hepburn v. Jannock Ltd. (2008), 40 B.L.R. (4th) 165, at para. 122, as an instance where the court applied proprietary estoppel to an interest in funds owned by a trust. But the court’s holding in Hepburn was based on promissory estoppel, not proprietary estoppel, and was upheld by the Court of Appeal on that basis: Hepburn v. Jannock Ltd., 2008 ONCA 847, 305 D.L.R. (4th) 571, at para. 18. The superior court’s discussion of proprietary estoppel was obiter.

[58] Ms. Wozniak also relies on Sweetnam v. Lesage, 2016 ONSC 4058, which considered a claim of proprietary estoppel against shares in a corporation. But Sweetnam does not assist Ms. Wozniak. The court concluded that two elements of proprietary estoppel had not been established and dismissed this claim. The Court did so without any discussion about whether proprietary estoppel could in principle apply to shares in a corporation. Thus, it does not stand as authority for proprietary estoppel extending beyond land.

[59] Consistent with Cowper-Smith, my review of the case law leads me to conclude that Ontario courts have not extended proprietary estoppel beyond interests in land. This is not to say that in an appropriate future case, a court could not recognize that proprietary estoppel applies to other forms of property. But that is not the law as it now stands.

[60] It was not open to the arbitrator to extend proprietary estoppel beyond interests in land. I come to this conclusion for two reasons.

[61] First, s. 31 of the Arbitration Act, 1991, S.O. 1991, c. 17 requires arbitral tribunals to decide a dispute in accordance with the law. In reading s. 31, consider how it both empowers and constrains: “[a]n arbitral tribunal shall decide a dispute in accordance with law, including equity, and may order specific performance, injunctions and other equitable remedies”. It empowers arbitral tribunals to apply statute law, common law and equity to disputes; it grants them remedial authority over only the parties to order specific performance, injunctions and other equitable remedies; but it also constrains them to decide a dispute in accordance with law, including equity.

[62] In considering s. 31, this court held that:

“[A]rbitration represents a process to address a dispute; it does not confer jurisdiction to ignore or rewrite the law and established legal principles. Put differently, the arbitration provision does not confer on the arbitrators the ability to do what they please unencumbered by applicable legal principles”: Omers Realty Corp. v. Sears Canada Inc. (2005), 74 O.R. (3d) 423 (S.C.), at para. 22, aff’d , 80 O.R. (3d) 561 (C.A.); see also Taseko Mines, at paras. 80-81.

This means that the arbitrator was required to apply the law as it stood at the time of the arbitration. He could not make new law.

[63] Second, where—as here—the parties have designated the substantive law of the dispute, the arbitrator must apply it. This is spelled out in s. 32(1) of the Arbitration Act, 1991, S.O. 1991, c. 17: “[i]n deciding a dispute, an arbitral tribunal shall apply the rules of law designated by the parties or, if none are designated, the rules of law it considers appropriate in the circumstances”.

[64] Here, the arbitration agreement provided that the “arbitration will be conducted in accordance with the law of Ontario and the law of Canada as it applies to Ontario”. This is critical given the contractual nature of private arbitration. An arbitration clause “derives its legitimacy from the consent of the parties to subject their dispute to an arbitrator”: Frambordeaux Developments Inc. v. Romandale Farms Ltd., 2007 CarswellOnt 8102 (S.C.), at para. 30. Departing from the law of Ontario, as the arbitrator did here, undermines the agreement between the parties.

[65] I recognize that arbitrators need not be lawyers (though the arbitrator here was) and that arbitral tribunals need not reason like courts nor exercise their discretion the same way a court would. But they must still identify and apply the correct legal principles: National Ballet of Canada v. Glasco (2000), 49 O.R. (3d) 230 (S.C.), at para. 63.

[66] I am not imposing a straitjacket on arbitrators. They are free to reason from precedent. Nor am I requiring some rigid adherence to analogous cases. Rather, there is a qualitative difference between the incremental application of a precedent to the facts of a dispute and extending a principle of law or equity in an entirely new direction. The former is permitted to arbitrators; the latter is not.

[67] Here, the arbitrator untethered proprietary estoppel from interests in land and held that it could apply to shares. Extending proprietary estoppel beyond interests in land is a major expansion of proprietary estoppel and a significant departure from precedent in Ontario. Whether proprietary estoppel ought to remain confined to interests in land is a question that the Supreme Court expressly chose not to decide in Cowper-Smith. It is the kind of change to the law that our courts—particularly our appellate courts—only make after careful deliberation and for compelling reasons. It was not open to the arbitrator to rewrite the law and established legal principles.

The arbitrator erred in finding a clear and unequivocal promise

[68] On their second ground of appeal, the applicants argue the arbitrator correctly identified the legal test but failed to apply it. They say that this amounts to an error of law: Canada (Director of Investigation and Research) v. Southam Inc., [1997] 1 S.C.R. 748, at para. 39. I agree.

[69] Proprietary estoppel has three conjunctive requirements:

a. a representation or assurance is made to the claimant, on the basis of which the claimant expects that they will enjoy some right or benefit over property;

b. the claimant relies on that expectation by doing or refraining from doing something, and their reliance is reasonable in all the circumstances; and

c. the claimant suffers a detriment as a result of their reasonable reliance, such that it would be unfair or unjust for the party responsible for the representation or assurance to go back on their word: Cowper-Smith, at para. 15.

[70] For the claimant’s reliance to be reasonable, the promise must be clear and unambiguous: Cowper-Smith, at para. 26. The arbitrator failed to apply this element of the test. This was an error of law.

[71] The email cannot satisfy the requirement of an unequivocal promise for two reasons. First, as the arbitrator recognized, the email contained “internal inconsistencies” and “contradictions”. These are the same inconsistencies and contradictions that Jensen J. found the arbitrator ignored in his trust award.

[72] Justice Jensen found the following internal inconsistencies in the email (at para. 103):

For example, in the October 18 email, Greg states that the family trust will hold “all Fusebill shares”. Yet, in the same sentence he states: “once, we sign this document 50% of the 4.5 million shares of Fusebill are yours”. It is unclear from this sentence whether Greg meant that the subject matter of the Burwell Family Trust would be the 4.5 million Personal Shares he was granted upon the incorporation of Fusebill, or the 4.5 million Trust Shares to which the Burwell Family Trust subscribed. In addition, Greg would seem to have contradicted himself in the October 18 email when he stated that Marta was entitled to 50% of the 4.5 million shares “no matter what happens” and then in the same sentence he said she would be entitled to 100% of the shares if he dies.

[73] In his proprietary estoppel award, the arbitrator did not address the inconsistencies that Jensen J. identified. He simply found that the underlying intent and promise were clear, and concluded that—despite lacking specificity regarding what constitutes his 50% ownership of Fusebill shares—Mr. Burwell intended Ms. Wozniak to have 50% of the Fusebill shares that he owned, regardless of what happens.

[74] But Jensen J. concluded that the internal inconsistencies in the October 18 email “effectively undermine the certainty of the subject matter of the trust created by the October 18 email”: Burwell, at para. 103. If the email lacked the certainty of the subject matter required to create a valid trust, it follows that the email cannot constitute a sufficiently clear and unambiguous promise about the same subject matter to ground a claim to in proprietary estoppel.

[75] The October 18 email does not answer the following questions: Which shares were Ms. Wozniak promised? Mr. Burwell’s personal shares or the shares to which the Burwell Family Trust subscribed, or all of them? The promises in the email are neither clear nor unambiguous, and therefore cannot ground a successful claim for proprietary estoppel.

[76] The arbitrator committed the same legal error that Jensen J. identified in his earlier trust award—he simply substituted proprietary estoppel for an express trust.

[77] Second, the arbitrator overlooked the fact that the promises in the October 18 email were contingent on the execution of the family trust agreement. Ms. Wozniak acknowledged before the arbitrator that the promises in the email were contingent on the execution of a family trust agreement. The family trust agreement modifies, and ultimately overtakes, the promises in the email. It is the family trust agreement that must be given effect, not the email.

[78] Failing to consider the contingent nature of the October 18 email is a legal error. This is particularly so given that Jensen J. held that the arbitrator “could not ignore the inconsistencies between the October 18 email and the Burwell Family Trust document. They are part of the same factual matrix and must be interpreted accordingly”: Burwell, at para. 99.

[79] Given my conclusions on the applicants’ first two grounds of appeal, I need not consider their third ground of appeal.

Disposition

[80] I grant leave to appeal and set aside the decision of the arbitrator.

[81] It would serve no purpose to remit the matter to the arbitrator. Ms. Wozniak’s claim is therefore dismissed.

[82] Mr. Burwell is presumptively entitled to costs. If the parties cannot agree on costs, they can each make written submissions to me of no more than 1,000 words within two weeks of the release of these reasons for judgment. These are to be sent to scj.assistants@ontario.ca to my attention.

Justice Owen Rees Released: October 22, 2024