COURT FILE NO.: CV-22-00680613-00CL DATE: 20230417

SUPERIOR COURT OF JUSTICE – ONTARIO COMMERCIAL LIST

IN THE MATTER OF AN ARBITRATION UNDER THE ARBITRATION ACT, 1991, S.O. 1991, c. 17

RE: TASEKO MINES LIMITED, Claimant (Respondent)

AND:

FRANCO-NEVADA CORPORATION, Respondent (Appellant)

BEFORE: Penny J.

COUNSEL: Andrew Gray and Jonathan Silver, counsel for the Appellant Junior Sirivar and Connor Bildfell, counsel for the Respondent

HEARD: December 12, 2022

REASONS ON APPEAL

Overview and Issues

[1] Franco-Nevada appeals from a 2022 commercial arbitration award which found that an agreement between Franco-Nevada and Taseko, concerning the development of a gold mine on what is called the “Prosperity Property” in British Columbia, had been frustrated.

[2] On May 12, 2010, Franco-Nevada entered into a metal purchase and sale agreement (also referred to as a streaming agreement) with Taseko. In exchange for Franco-Nevada’s commitment to provide a cash payment deposit of US$350 million, Taseko agreed to sell Franco-Nevada a portion of the gold extracted from the proposed Prosperity gold mine at a pre-determined price for a period of 40 years. Although structured as a purchase and sale of gold, it is common ground that the agreement was also a method of project financing for the development of the Prosperity mine.

[3] Federal environmental approval for the proposed mining development was denied in the fall of 2010. A revised proposal was again denied by the federal government in early 2014. The Prosperity Property (which contains the gold reserves) is the subject of unresolved claims by the Tsilhqot’in First Nation (TFN). It is common ground that a significant factor in the federal government’s decision to refuse approval was TFN’s opposition to any development of the Prosperity Property. Taseko unsuccessfully sought judicial review of the government’s 2014 refusal. The Supreme Court of Canada denied leave in May 2020. No further approvals have been sought since that time. There has been no development of the proposed Prosperity mine. No funds have been advanced and, for obvious reasons, no gold has been delivered and no streaming payments for refined gold have been made. Discussions between Taseko, relevant government authorities and the TFN are ongoing but, as yet, there have been no concrete developments and the TFN position remains that it is opposed to any development of the Prosperity Property. Taseko has considered selling its rights. However, one of the obstacles to any sale is that the agreement purports to bind subsequent purchasers of the Prosperity project.

[4] There are four issues raised on the appeal. The first issue is the standard of review. It is common ground that the standard of review is reasonableness.

[5] The second and most important issue is whether the arbitrator’s finding that the agreement has been frustrated was unreasonable.

[6] The third issue is whether the arbitrator’s Award of substantial indemnity costs against Franco-Nevada was unreasonable.

[7] The final issue is whether portions of the Appeal Record should be sealed under the test set out in Sherman Estate v. Donovan, 2021 SCC 25.

[8] As I will explain in the reasons which follow, the arbitrator’s declaration that the agreement had been frustrated was unreasonable; Franco-Nevada’s appeal of the arbitrator’s decision is granted. Taseko’s motion for a sealing order in relation to specified portions of the Appeal Record (which was not opposed by Franco-Nevada) is also granted.

Background

[9] Taseko is a mining company that holds the rights to a mineral deposit located in the interior of British Columbia. Franco-Nevada is a prominent royalty and streaming company in the mining industry. In the mid-1990s, Taseko initiated the permitting process for the proposed Prosperity mining project. The Prosperity project required both provincial and federal environmental assessments and approvals. Taseko’s impact statements and applications were submitted in 2009.

[10] Later in 2009, Taseko and Franco-Nevada entered into discussions about possible stream financing for the Prosperity project. A streaming agreement is a form of metal purchase agreement where, in exchange for an upfront deposit payment, a streaming company has the right to purchase all or a portion of one or more metals subsequently produced from a mine at a preset price.

[11] Before signing the agreement, Franco-Nevada conducted due diligence on Taseko and its mine plans.

[12] As noted in the overview, the required federal environmental approvals were denied in 2010 and, following a resubmission in 2011, again in 2014. Taseko has since been negotiating with the TFN and the relevant government authorities but there has been no change to the status quo. Taseko has also been seeking to sell the Prosperity Property.

[13] In January 2021, Taseko took the position that the federal government’s refusal to issue the required environmental approvals had frustrated the agreement. An arbitration notice was issued in June 2021.

[14] An evidentiary hearing took place before Gerald W. Ghikas, Q.C. on January 31 and February 1, 2022. Written submissions were delivered on February 18, 2022 and oral submissions took place on March 7, 2022. The Award was issued on April 5, 2022.

[15] By virtue of Schedule L to the agreement, either party is entitled to appeal the decision of the arbitrator “on a question of law or a mixed question of fact and law.”

The Award

[16] The central legal issue in dispute before the arbitrator was whether the test for frustration required showing that the “supervening event” was not foreseen by the parties.

[17] The arbitrator applied the test for frustration set out in the decision of the Supreme Court of Canada in Naylor Group Inc. v. Ellis-Don Construction Ltd., 2001 SCC 58, [2001] 2 S.C.R. 943, at para. 53. Frustration occurs when a situation has arisen for which the parties made no provision in the contract and performance of the contract becomes “a thing radically different from that which was undertaken by the contract”. Here, the Supreme Court was citing from Peter Kiewit Sons’ Co. v. Eakins Construction Ltd., [1960] S.C.R. 361, per Judson J., at p. 368, who was, in turn, quoting from Davis Contractors Ltd. v. Fareham Urban District Council, [1956] A.C. 696 (H.L.), at p. 729.

[18] The Supreme Court in Naylor went on to observe that earlier cases of frustration proceeded on an “implied term” theory. Under that approach, the court was to ask itself a hypothetical question: if the contracting parties, as reasonable people, had contemplated the supervening event at the time of contracting, would they have agreed that it would put the contract to an end? The implied term theory, the Court found, is now largely rejected because of its reliance on fiction and imputation. More recent case law, including Peter Kiewit, has adopted a more candid approach. The court is asked to intervene, not to enforce some fictional intention imputed to the parties, but to relieve the parties of their bargain because a supervening event has occurred without the fault of either party.

[19] The arbitrator went on to delve more deeply into the earlier decisions in Peter Kiewit and Davis Contractors. He noted that, in another passage from the reasons of the House of Lords in Davis Contractors (which was also adopted by the Supreme Court in Peter Kiewit), Lord Radcliffe said there was “no uncertainty as to the materials upon which the Court must proceed. The data for decision are, on the one hand, the terms and construction of the contract, read in the light of the then existing circumstances, and on the other hand the events which have occurred”. Lord Reid, in his concurring judgment, added that “on this view there is no need to consider what the parties thought or how they or reasonable men in their shoes would have dealt with the new situation if they had foreseen it. The question is whether the contract which they did make is, on its true construction, wide enough to apply to the new situation: if it is not then it is at an end”.

[20] The arbitrator considered a decision of the Saskatchewan Court of Queen’s Bench in Industrial Overload Ltd. v. McWatters (1972), 24 D.L.R. (3d) 231, which adopted the approach of the House of Lords. In that case, Sirois J. said, at para. 5, that: “The fact that the parties, at the time of contracting, actually foresaw the possibility of the event or new circumstances in question does not necessarily prevent the doctrine of frustration from applying when that event takes place.” The arbitrator noted that the same view was expressed by the British Columbia Court of Appeal in Wightman Estate v. 2774046 Canada Inc., 2006 BCCA 424, 276 D.R.L. (4th) 492, where it said, at para. 26, that: “since the implied-term theory has been rejected, it is no longer necessary to speculate as to what the parties, as reasonable persons, would have agreed if they had thought about the supervening event at the time they made their contract. Thus, whether the parties foresaw the event is irrelevant.”

[21] Franco-Nevada relied on a passage from the Supreme Court’s decision in Naylor which expressly referred to an event that was foreseen as not qualifying as a “supervening event”. Binnie J., writing for a unanimous Court, wrote (at para. 56):

While the second approach (“a radical change in the obligation”) is to be preferred and is now the established test, the appellant’s argument would fail under either view. There has been no “supervening event” in the sense required by either approach to the doctrine of frustration and in fact the OLRB ruling against the appellant was a foreseeable outcome.

The arbitrator found, however, that the Supreme Court’s decision in Naylor did not turn on a finding that the outcome of the OLRB decision was foreseeable. It turned on findings that there had been no supervening event and that the parties had, in any event, provided in their agreement for the contractual consequences of the alleged “supervening event.”

[22] Franco-Nevada also relied on three decisions of the Court of Appeal for Ontario as support for its contention that a reasonably foreseeable risk could not qualify as a “supervening event” sufficient to make performance something radically different from that which was undertaken in the contract. In Bang v. Sebastian, 2018 ONSC 6226, aff’d 2019 ONCA 50 [1], the risk of the purchaser being unable to obtain financing was “contemplated by the parties” and the purchaser “deliberately chose” not to provide for this risk in the agreement. As such, the alleged supervening event was not “unforeseen” and did not constitute a frustrating event.

[23] Perkins v. Sheikhtavi, 2019 ONCA 925 was factually similar to Bang. A legislative change suppressed property values in the purchaser’s area such that she no longer wished to conclude the purchase. The Court of Appeal for Ontario held that the supervening event did not constitute frustration of the agreement because it was “contemplated by the parties at the time of contracting and was provided for or deliberately chosen not to be provided for in the contract”: at para. 16.

[24] Fram Elgin Mills 90 Inc. v. Romandale Farms Limited, 2021 ONCA 201 is another case concerned with the purchase of land where government planning changes occurred after the agreement was entered into. The Court of Appeal held, at para. 239, that “the possibility of planning changes was within the parties’ contemplation when they entered into the 2005 August Agreement. Despite that, they made no provision for such a possibility – as, for example, through the insertion of a “drop-dead” provision. Therefore, even if the planning changes were a supervening event, the 2005 August Agreement is not frustrated”.

[25] The arbitrator listed several reasons for rejecting the approach of the Court of Appeal for Ontario in these three cases. These reasons were:

• this approach adds, without explanation and without regard to the conceptual underpinnings of the modern doctrine, to the criteria stated by the Supreme Court of Canada in Peter Kiewit and Naylor
• the proposition that “deliberately choosing” not to address a foreseeable risk is a basis for denying frustration was said to originate with the “seminal” decision of the Court of Appeal in Capital Quality Homes Ltd. v. Colwyn Construction Ltd. (1975), 9 O.R. (2d) 617, at p. 626. However, this principle does not appear in Capital Quality. To the contrary, a requirement that frustration should be denied if the invoking party foresaw the risk of, but failed to provide in the contract for, the supervening event, is the antithesis of the approach endorsed by the Ontario Court of Appeal in Capital Quality
• this approach ignores the practical, commercial reality that what is included and what is not included in a commercial agreement depends on the agreement of both parties and is not a unilateral decision. It introduces a notion of “fault” into the analysis
• this approach, where relief is denied simply because the risk of the supervening event was foreseen or reasonably foreseeable and not provided for, is simply a re-cycling of the original harsh and outmoded position under common law which the modern test for frustration was intended to moderate, and
• in none of Bang, Perkins or Fram was the finding concerning foreseeability or deliberate choice essential to the disposition of the case.

[26] After reviewing the arguments of both parties, the arbitrator concluded that the following legal principles apply to the doctrine of frustration:

a. the party seeking relief by virtue of frustration has the burden to prove that frustration has occurred;

b. frustration of an otherwise binding contract occurs when:

i. a supervening event has occurred; and

ii. the parties made no provision in the contract for the supervening event; and

iii. as a result of the supervening event, performance of the contract has become a thing radically different from that which was undertaken by the contract;

c. a “supervening event” is a post-contractual event that was not the fault of either party, and,

i. while there is authority to the contrary, the better view is that to constitute a supervening event, it is not necessary that the possibility of the event be unforeseen by both parties or by the party invoking the doctrine; and

ii. while there is authority to the contrary, the better view is that to constitute a supervening event, it is not necessary that the possibility of the event had been unforeseeable by a reasonable person in the position of the parties; and

d. to determine whether the contract provides for the supervening event, the contract is to be interpreted according to the ordinary rules for construction of contracts whereby the language of the agreement as a whole is to be taken into account and the words used are to be given their ordinary and grammatical meaning, consistent with the surrounding circumstances known to the parties at the time of formation of the contract;

e. in determining whether the performance of the contract has become a thing radically different from that which was undertaken by the contract:

i. the performance undertaken by the contract is, again, to be determined by applying the ordinary rules for construction of contracts, including recourse to evidence of surrounding circumstances;

ii. when assessing whether the supervening event has caused performance to be a thing radically different from that which was undertaken by the contract:

1. it is not necessary to show that performance has become impossible;
2. hardship or inconvenience or material loss alone is not sufficient to make performance radically different;
3. it is relevant to consider whether the alleged supervening event is merely the continuation of normal cyclical events (such as market fluctuations) or is transient, and not permanent;
4. it is relevant to consider whether the event destroyed the very root or foundation of the agreement;

f. While there is authority to the contrary, the better view is that, if the contract does not provide for the supervening event:

i. it is not relevant that the party invoking frustration “deliberately chose” not to provide for the supervening event in the contract by actually considering, but not insisting on, such provision; and

ii. it is not relevant that the parties could have provided for the supervening event in the contract.

[27] The arbitrator dealt first with the question of whether the refusal of the federal government to grant environmental approval was a “supervening event”. He found that it was. It was not caused by the fault of either party and took place after the contract was concluded. Despite the arbitrator’s finding that the doctrine of frustration is not dependent on proof that the supervening event was not foreseen, he went on to review the evidence and make findings of fact on this issue. The evidence clearly showed, he went on to say, that there was a risk or possibility that the federal government might refuse to grant approval. Despite this risk, he found as a fact that both parties expected, with different levels of confidence, that federal approval would be granted; that is, that it was more likely than not that the approval would be granted. A reasonable person, he also found, would have believed as Taseko and Franco-Nevada did.

[28] The arbitrator then dealt with the question of whether the agreement “provided for” the event of a federal government refusal to grant environmental approval. He found that it did not. The arbitrator concluded that no single provision in the agreement explicitly stipulated what was to happen if the federal government refused to grant environmental approval. He went on, however, to review those provisions of the agreement that were tied to the environmental approvals and to the contemplated timeline for the Prosperity project.

[29] First and foremost, section 3.2(h) of the agreement states that before Franco-Nevada is obliged to provide the cash payment deposit, Taseko must have received all relevant permits, including the provincial and federal environmental approvals. Section 5.1(c) provides that Franco-Nevada can terminate the agreement if the conditions precedent for advances of the cash payment deposit (which include the environmental approvals) have not been met within two years after the date the agreement was entered into (i.e., by May 2012). Thus, since May, 2012, Franco-Nevada has had the right, but not the obligation, to terminate the agreement. Section 3.6 grants Franco-Nevada the right to waive any condition precedent “in whole or in part” in its sole discretion. To date, Franco-Nevada has not exercised its right to terminate.

[30] Further, the consideration required from Franco-Nevada, once the conditions precedent were met, also included the delivery of 2 million Franco-Nevada warrants. Under the terms of the agreement, however, those warrants expired and became valueless on June 16, 2017. Thus, by June 16, 2017, a component of the consideration to be paid by Franco-Nevada was no longer payable.

[31] Under s. 7.1(a), Taseko may “amend the mine plan for the Prosperity Project at any time.” Section 7.9 also contemplates Taseko making “material amendments” to the project’s operating plan, development plan, or schedule. Importantly, the agreement does not require Taseko “to construct, operate, or continue operating the Prosperity project, or to explore or develop the Prosperity Project” at all (s. 7.1(b)).

[32] Reference is made in the agreement and its schedules to a 2009 Technical Report in the context of the Prosperity Processing Plant, which is defined as a plant for processing ore “substantially in the form contemplated” in the 2009 Technical Report. The report contemplated, among other things, a 344 million tonne reserve and a capacity to mill 70,000 tonnes per day of ore capable of producing copper-gold concentrate.

[33] According to Schedule D to the agreement, as of May 12, 2010 the projected date for receipt of federal approvals was September 2010. Part of Schedule B to the agreement is a schedule, as of April 16, 2010, which projects that site works would begin on October 4, 2010.

[34] Although Taseko did not need Franco-Nevada’s consent to make changes to the development plans, it was required to give notice to Franco-Nevada of all material changes. Materiality was defined to mean, regarding project costs, a change of 5%, regarding operating volumes, a change of 2.5%, and regarding timing delays, a delay of 1 month or more.

[35] The arbitrator also reviewed evidence of surrounding circumstances, which included:

a. when negotiations began, Taseko had already completed detailed cost and production models, development plans, environmental studies, technical reports, and other materials for the Prosperity project outlined in the 2009 Technical Report, all of which had been the subject of extensive due diligence by Franco-Nevada;

b. Franco-Nevada and Taseko both based their financial modeling on the costs and timing associated with that planned project;

c. the federal environmental approval of the project had been sought and the federal government’s decision was expected before the end of 2010; and

d. Taseko and Franco-Nevada both expected (in varying degrees) that the federal environmental approval would be obtained for the project.

[36] Based on all this, the arbitrator concluded that the agreement made no provision for what was to happen if, as had clearly happened by 2014, the particular Prosperity project that is the subject matter of the agreement was not approved within the particular timeline contemplated by the agreement.

[37] The arbitrator then turned to the final question of whether the 2014 federal refusal of environmental approval and its sequelae “radically altered” performance obligations under the agreement.

[38] The arbitrator concluded that the primary reason why federal environmental approval was refused, and the reason why the Prosperity project contemplated by the agreement had not progressed, was the long-standing and continuing opposition of the TSN to any mining project that had negative impacts on Fish Lake and its environs which had sacred, cultural significance. Any mine development on the Prosperity Property requires federal and provincial approvals. The federal and provincial governments have a legal duty to consult with and accommodate Aboriginal Peoples who may be affected by governmental decisions such as whether to approve a proposed mining project. The enactment of federal and provincial legislation implementing the United Nations Declaration on the Rights of Indigenous People serves to emphasize the need for and importance of this duty.

[39] The arbitrator observed that unlike many cases, whether contractual performance is impossible or radically different can in this case be decided with the benefit of evidence of what has actually occurred over 8 years since the supervening event occurred. The evidence, he said, proved conclusively that the 2014 federal refusal of environmental approval was not and will never be overturned by the courts. Taseko took significant steps to attempt to mitigate or avoid the consequences of the 2014 federal refusal, both through court proceedings and through discussions with the TSN and the province. There was no suggestion Taseko’s efforts were insufficient.

[40] Despite these efforts, the Prosperity project has not progressed at all. The provincial environmental permit that was granted expired. No material progress has been made to overcome the TSN opposition. There is no pending application for either provincial or federal environmental approval. The payment deposit warrants have expired, without Taseko ever having the opportunity to exercise them. It is certain that the Prosperity project contemplated by the agreement will not be achieved within the particular time frame upon which the agreement was based.

[41] The arbitrator rejected Franco-Nevada’s argument based on what he called inferences drawn from possible future events. At its highest, the arbitrator held, the evidence of what may lie ahead shows no more than that there is a possibility that at some future time a mining project different from that contemplated by the agreement might receive provincial and federal approval. In order to be able to call upon Franco-Nevada for an advance of funds, Taseko would have to design and obtain all necessary approvals for a revised mine design and a new mine development plan substantially different from that envisaged by the agreement. As a practical matter this would almost certainly require reaching a compromise with the TSN. The revised development plan necessarily would result in revised construction and permitting schedules, and revised estimates of capital and operating costs, different from those contemplated by the agreement. And, all of this activity would have to be undertaken by Taseko knowing that Franco-Nevada could terminate the agreement at any time.

[42] Taseko’s only other option, if the agreement were to continue in its present form, would be to cease to pursue any mine development for the remainder of the 40 year term. Although Taseko is permitted under the agreement to do this, simply abandoning the Prosperity project altogether would not result in either party receiving the fundamental benefit for which it bargained. In other words, it was not the purpose of the agreement for Franco-Nevada to agree to provide financing to facilitate Taseko’s intended mine development and then for all mine development to cease. Ceasing any effort toward mine development, because performance has become impossible or radically different from what was intended, frustrates the commercial purpose of the agreement. The commercial objective that formed the foundation of the agreement would not be achieved.

[43] On the basis of this reasoning, the arbitrator concluded that performance of the bargain as intended was no longer possible and that any attempt now to perform the agreement would necessarily require performance radically different from what the parties intended when the agreement was made.

[44] Thus, on the three key questions, the arbitrator found:

(1) the 2014 federal refusal of environmental approval was a supervening event;

(2) the agreement makes no provision for what is to occur if the particular Prosperity project that is the subject matter of the agreement is not approved within the particular timeline contemplated by the agreement; and,

(3) the evidence establishes that performance of the bargain as intended is no longer possible and any attempt now to perform the agreement would necessarily require performance radically different from what the parties intended when they made the agreement.

[45] Franco-Nevada sought substantial indemnity costs of $576,133.55. Taseko sought full indemnity costs of $492,591.41. Both parties agree that an award of costs by an arbitral tribunal is not bound by Rule 57.01 of the Ontario Rules of Civil Procedure. Taseko was the successful party. The arbitrator found that counsel for Taseko managed the case efficiently and that the amounts charged to Taseko for legal fees were reasonable. On this basis he awarded Taseko’s full indemnity costs.

Analysis

Standard of Review

[46] The parties agree that the standard of review on this appeal is reasonableness: Creston Moly Corp. v. Sattva Capital Corp., 2014 SCC 53, [2014] 2 S.C.R. 633, at paras. 105–106. Reasonableness is a highly deferential standard “concerned mostly with the existence of justification, transparency and intelligibility within the decision-making process [and] also with whether the decision falls within a range of possible, acceptable outcomes which are defensible in respect of the facts and law”: Sattva, at para. 119; Canada (Minister of Citizenship and Immigration) v. Vavilov, 2019 SCC 65, [2019] 4 S.C.R. 653 at para. 86.

[47] Courts apply this highly deferential standard to commercial arbitration in recognition of the “important link between the fact of a private consensual arbitration and the need for judicial deference to the result of the arbitration”: Ledore Investments Limited (Ross Steel Fabricators & Contractors) v. Ellis-Don Construction Ltd., 2017 ONCA 518 at para. 17. “The parties’ selection of their forum implies both a preference for the outcome arrived at in that forum and a limited role for judicial oversight of the award made in the arbitral forum. The application judge’s decision to not set aside the award is consistent with the well-established preference in favour of maintaining arbitral awards rendered in consensual private arbitrations”: Popack v. Lipszyc, 2016 ONCA 135, at para. 26. Also, as Gascon J. explained in Teal Cedar Products Ltd. v. British Columbia, 2017 SCC 32, [2017] 1 S.C.R. 688, “[t]he preference for a reasonableness standard dovetails with the key policy objectives of commercial arbitration, namely efficiency and finality” (para. 74).

[48] Under this highly deferential approach, the issue is not whether a rigorous reading of the decision will reveal some sort of flaw or misstep but whether the shortcomings in the decision are sufficiently serious so that the decision no longer exhibits the requisite degree of justification, intelligibility and transparency: Vavilov, paras. 100 and 102. “An error in the arbitrator’s analysis would not necessarily lead to an unreasonable decision”. Reasonableness review “is not an examination of the weakest link of a chain of analysis in isolation from the reasons as a whole”. Nor is it a “line-by-line treasure hunt for error”. Indeed, “a global analysis of the arbitrator’s decision can lead to the conclusion that the result falls within acceptable, defensible outcomes, and that the reasons—despite their flaws—fulfill the criteria for justification, transparency and intelligibility”: Ontario (Finance) v. Elite Insurance Company, 2018 ONCA 809 at para. 40.

[49] Additionally, as expertise is a factor in judicial review, it is a factor in commercial arbitrations: where parties choose their own decision-maker, it may be presumed that such decision-makers are chosen either based on their expertise in the area which is the subject of dispute or are otherwise qualified in a manner that is acceptable to the parties: Sattva, at para. 105.

[50] The role of courts in these circumstances is to refrain from deciding the issue themselves. Accordingly, a court applying the reasonableness standard does not ask what decision it would have made in place of that of the decision maker of first instance, attempt to ascertain the “range” of possible conclusions that would have been open to the decision maker, conduct a de novo analysis or seek to determine the “correct” solution to the problem. The Federal Court of Appeal noted in Delios v. Canada (Attorney General), 2015 FCA 117, 472 N.R. 171 that, “as reviewing judges, we do not make our own yardstick and then use that yardstick to measure what the administrator did”: at para. 28. Instead, the reviewing court must consider only whether the decision made by the original decision maker—including both the rationale for the decision and the outcome to which it led—was unreasonable: Vavilov, at para. 83.

[51] This approach applies to a finding of frustration. Determining whether a particular contract has been frustrated by a particular event involves a “highly fact-driven exercise” rooted in “findings concerning the foundation or fundamental purpose of the contract”. This highly fact-driven exercise must be “focused on the contract that is alleged to have been frustrated”: Interfor Corporation v. Mackenzie Sawmill Ltd., 2022 BCCA 228 at paras. 76-79. The deference owed to the decision maker at first instance is enhanced where the decision maker heard live testimony and made critical determinations about the surrounding circumstances and the fundamental purpose, root or foundation of the agreement on the strength of that evidence.

[52] However, as I will discuss below, one of the issues on this appeal is whether the arbitrator failed to follow binding precedent. The applicable law under the agreement is the law of Ontario. Franco-Nevada maintains that a consideration of both the foreseeability and the permanency of the supervening event is a requirement under the Ontario law of frustration, and that the arbitrator refused to follow that law by refusing to consider the foreseeability and impermanence of the federal permit refusals. The majority of the Supreme Court of Canada in Vavilov made clear that assessing whether a decision maker has appropriately followed binding precedent is part of the reasonableness analysis: Vavilov at paras 111–112. Failure to follow binding precedent, or failure to reasonably explain departure from it, could render a decision unreasonable: Vavilov at para 112. Whether an administrative decision that departs from binding precedent on the applicable legal test is considered “incorrect” or “unreasonable” matters less than the fact that “the decision cannot stand”. Vavilov leaves no doubt that the analytical approach is one of reasonableness, with binding precedent acting as a “legal constraint” on the decision maker that may render the decision unreasonable: Browne v. Canada (Citizenship and Immigration), 2022 FC 514. See also Vavilov at paras 96 to105, 111–112.

Was the Arbitrator’s Conclusion That the Contract Had Been Frustrated Unreasonable?

[53] Franco-Nevada’s first argument is that the arbitrator misapplied the test for frustration; he concluded, contrary to binding Ontario authority, that foreseeability plays no role in the frustration doctrine. Had the arbitrator applied the proper principles to his factual findings, Franco-Nevada says, he would have dismissed Taseko’s claim. By omitting any consideration of foreseeability from the test for frustration, the arbitrator declined to follow recent binding precedent from the Ontario Court of Appeal. This was unreasonable. The parties agreed that Ontario law governed the agreement and the arbitration. The arbitrator was bound by that choice and was required to follow Ontario’s highest court. Further, while it may have been transparent, the arbitrator’s decision was neither justified nor intelligible. The arbitrator not only applied the wrong legal test for frustration; his reasoning was inconsistent with and confused a fundamental principle of contract interpretation (the content and role of evidence of factual matrix) and was internally inconsistent, both with respect to his findings of fact and his application of the law.

[54] Second, Franco-Nevada says the arbitrator concluded, contrary to Ontario law, that Taseko did not need to prove that the contractual impact of the federal rejections was “permanent”. Ontario authority and other leading sources make permanency part of the test for frustration. Courts must distinguish between “a disruption that is permanent, vis-à-vis the contract (albeit that it may not be permanent in other respects) and one that is temporary or transient”: Cowie v. Great Blue Heron Charity Casino, [2011] O.J. No. 5573 at paras. 19-23 (Div. Ct.), citing G.H.L. Fridman, The Law of Contract in Canada, 4th ed. (Scarborough: Carswell, 1999) at 679-680. Frustration applies in the former situation, not the latter. According to Franco-Nevada, Taseko had to prove that the federal permitting rejections permanently prevented Taseko from performing under the contract. By excluding permanency from the test, the arbitrator unreasonably relieved Taseko of its evidentiary burden.

[55] Third, Franco-Nevada argues that the arbitrator interpreted the 40-plus-year agreement as being timeline- and project-specific to a particular mine plan, a plan that Taseko never submitted for approval. Franco-Nevada says the arbitrator’s conclusion that the agreement was project and timeline-specific was entirely unsupported by the evidence and contrary to the plain words of the contract and was therefore unreasonable.

[56] Taseko argues that the arbitrator’s reasons for rejecting Franco-Nevada’s first two arguments about the role of “foreseeability” and “permanency” demonstrate justification, transparency, and intelligibility. There was an arguable basis in law upon which he reached his conclusions. In the alternative, it says, even if the arbitrator’s decision to reject these proposed additional requirements was unreasonable, that decision did not affect the outcome because he went on to find as a fact that: the supervening event was neither foreseen nor reasonably foreseeable; was permanent; and, the supervening event prevented the parties’ bargain from being achieved as intended. Taseko also argues that the arbitrator’s conclusion, that the agreement was project and timeline specific, was rooted in, and reasonably available on, the evidence.

The Test for Frustration Under Ontario Law

[57] Para. 13.4 of the agreement provides that the agreement “shall be governed by and construed under the laws of the Province of Ontario and the federal laws of Canada applicable therein”.

[58] The arbitrator relied heavily on the decisions of the Supreme Court of Canada in Peter Kiewit and the Court of Appeal for Ontario in Capital Quality Homes Ltd. v. Colwyn Construction Ltd. (1975), 9 O.R. (2d) 617 for his conclusion that consideration of whether a supervening event was “foreseen” is not a requirement or necessary factor. However, in Peter Kiewit, Judson J., for the majority, rejected the plaintiff’s theory that the contract had been frustrated by the defendant’s insistence on additional specifications for the work the plaintiff was to perform. Judson J. adopted, as a correct statement of the law, the now famous quote from the House of Lords decision in Davis Contractors Ltd v. Fareham Urban District Council, [1956] A.C. 696 at p. 729:

Frustration occurs whenever the law recognizes that without default of either party a contractual obligation has become incapable of being performed because the circumstances in which performance is called for would render it a thing radically different from that which was undertaken by the contract.

[59] But then, Judson J. went on to conclude that on any view of the facts, frustration could not be made out. That was because the kind of “extra work” said to have been performed was “a contingency covered by the express contract and does not afford a ground for its dissolution”. Even if extra work had to be performed (as the plaintiff maintained), the character and extent of the obligation to pay for such work were “fully covered in the contract”. Thus, even on the plaintiff’s own view of the case, its performance was not “radically different from that called for by the contract”.

[60] This passage emphasizes that, in order to determine whether something is “radically different from that which was undertaken by the contract”, it is necessary to determine what was undertaken in the first place. In other words, whether a supervening event causes performance to become something radically different must be measured against the performance that was contemplated by the parties at the time of their agreement.

[61] This is precisely the point made by the Court of Appeal for Ontario in Capital Quality, described by the arbitrator as the “seminal” case on frustration from Ontario’s highest court. In Capital Quality, the purchaser agreed to purchase from the defendant, 26 building lots within a registered plan of subdivision. Both parties were aware that the purchaser was buying building lots for the purpose of erecting a home on each lot with the intention of selling the homes by way of separate conveyances. Before closing, the Legislature enacted what was then s. 29 of the Planning Act, which restricted an owner’s right to convey adjacent lots without obtaining a consent from the Committee of Adjustment. After reviewing the development of the law, including Davis, Evans J.A., writing for the Court stated:

There can be no frustration if the supervening event results from the voluntary act of one of the parties or if the possibility of such event arising during the term of the agreement was contemplated by the parties and provided for in the agreement [emphasis added]. In the instant case the planning legislation which supervened was not contemplated by the parties, not provided for in the agreement and not brought about through a voluntary act of either party.

The reason for the proviso that there can be no frustration if the supervening event is contemplated by the parties and provided for in the agreement is obvious — if the event was contemplated and provided for, performance post-supervening event cannot logically be something radically different.

[62] While the word “foreseen” was not used in either case, I cannot discern any rational or meaningful difference between an event that was “contemplated” and one that was “foreseen”.

[63] This point is made by the Supreme Court of Canada in Naylor. In Naylor, the issue was whether Ellis-Don could subcontract electrical work with a contractor, Naylor, whose workers were not affiliated with the IBEW union. Ellis-Don had been in a relationship with the IBEW for over 30 years. Knowing there was a pending IBEW grievance against Ellis-Don asserting the existence of a collective agreement, and knowing Naylor had a collective agreement with a different union, Ellis-Don nevertheless contracted with Naylor. The OLRB subsequently held that Ellis-Don was bound to a valid and subsisting collective agreement with the IBEW. Ellis-Don sought to escape liability to Naylor on the basis that the finding of the OLRB had frustrated its contract with Naylor. The Supreme Court rejected that argument.

[64] Binnie J., writing for the Court, found that Ellis-Don knew the IBEW had taken the position since 1962 that its workers had been promised the electrical work on Ellis-Don’s projects. The OLRB found that Ellis-Don had largely observed this collective bargaining obligation over the intervening 30 years. Thus, Justice Binnie found, the OLRB decision merely recognized and affirmed Ellis-Don’s obligation to the IBEW; it did not create it. When Ellis-Don took on Naylor as the electrical subcontractor, therefore, it was promising work that had already been bargained away to the IBEW. Binnie J., after reviewing both the older, implied contract theory and the modern formulation for frustration in Peter Kiewet and Davis, concluded (at para. 56) that:

There has been no “supervening event” in the sense required by either approach to the doctrine of frustration and in fact the OLRB ruling against the appellant was a foreseeable outcome.

[65] The arbitrator, however, concluded that whether the supervening event was “foreseen” was not part of the test articulated by the Court in Naylor and that Justice Binnie’s decision “did not turn” on that determination, implying that the reference to “foreseen” was mere surplusage or obiter dicta (at para. 104). Later in his decision (at para. 135), the arbitrator referred to Justice Binnie’s conclusion, quoted above, as containing a “passing reference” to the OLRB outcome having been foreseen.

[66] With the greatest of respect to the arbitrator, both these observations are wrong in law. The very reason given by Justice Binnie for why the OLRB decision did not constitute a supervening event is that the result — Ellis Don had to use IBEW subcontractors — was foreseen. And this is precisely why the “event” of the OLRB’s decision could not make enforcement of Ellis Don’s subcontract with Naylor something radically different from what was originally contemplated.

[67] It is true that Justice Binnie set out a second reason why frustration could not apply. But the second reason involves a completely separate and distinct finding. This is made clear by the introductory words to para. 59: “There is another reason why the doctrine of frustration is inapplicable” (emphasis added). The existence of a second reason in no way undermines the authority of the first.

[68] This brings me to the three controversial decisions from the Court of Appeal for Ontario that were rejected by the arbitrator.

[69] In Bang, a purchaser failed to close a residential real estate agreement. She sought to avoid liability on the basis that the contract had been frustrated by: a) her inability to obtain mortgage financing; and b) a decline in the market price of her existing property, which she had to sell in order to enable her purchase under the agreement. In his decision on a motion for summary judgment, the motion judge held that failure to obtain financing was a known risk the purchaser accepted when she decided not to include a financing condition in her final, accepted offer. The motion judge also held that the purchaser “knew that property values could go up, and from this she knew that property values could go down”. Although the alleged property value decline was beyond her control and not what she “expected would occur”, this did not make it “unforeseen”. The motion judge concluded that frustration had not been established because the alleged decline in property value “was not an unforeseen supervening event”. An appeal to the Court of Appeal for Ontario was dismissed. However, as no appeal was taken from the conclusion that the agreement had not been frustrated, that issue was not addressed in the Court of Appeal’s reasons.

[70] In Perkins, however, frustration was squarely addressed on appeal. Like Bang, Perkins involved a purchaser who failed to close a residential property deal. She pleaded that a change in government policy had caused real estate prices to decline and that her subsequent inability to sell her own home at an acceptable price, and to obtain financing, had frustrated the agreement.

[71] Upholding the decision of the motion judge, who rejected the frustration argument, the Court of Appeal held that “a contract is not frustrated if the supervening event was contemplated by the parties … and was provided for or deliberately chosen not to be provided for in the contract”: at para. 16. The purchaser chose not to include a condition in her offer that she had to have sold her home and obtained mortgage financing. The purchaser “would reasonably have known there was a risk her home would not sell at the price she sought but made an unconditional offer” because she wanted her offer to be accepted: Perkins, at para. 19. Accordingly, the test for frustration was not met. Although the change in government policy was supervening event, it did not force the purchaser “to do something radically different from what the parties agreed”: at para. 21.

[72] Fram is a complicated case involving many issues. In simple terms, Romandale agreed in 2005 to sell two blocks of real estate development lands to Kerbel. In 2009, changes in government land development policy significantly changed the timelines and development prospects for the lands, deferring potential development, in the form of secondary planning approval, for a decade and more. Among other things, Romandale sought to escape its obligations under the agreement on the basis that the change in policy (together with the consequent changes in development potential) frustrated the agreement. The trial judge held the agreement had been frustrated. The Court of Appeal reversed.

[73] The Court of Appeal held that “a contract is not frustrated if the supervening event results from a voluntary act of one of the parties or if the parties contemplated the supervening event at the time of contracting and provided for, or deliberately chose not to provide for, the event in the contract”: at para. 230.

[74] Applying this principle to the circumstances, the Court of Appeal found that neither the change to the timing of the development of the Lands nor the fact that the development paths of the two blocks now diverged “render Romandale’s obligations under the 2005 August Agreement radically different from that to which it agreed”: at para. 233. Although most of the parties’ obligations were to be performed in short order, there were conditional provisions, one of which required secondary planning approval (referred to as “SPA” in the decision) to have been granted in respect of the Lands. The Court of Appeal held at para. 235:

While the planning changes altered the timing horizon for the development of the Lands and the development paths of the Snider and McGrisken Farms, those changes did not radically alter what the parties had agreed to under the 2005 August Agreement. In fact, the planning changes did not alter the parties’ obligations under the Conditional Provision in any way. What changed were the parties’ expectations about when SPA would be obtained for the Lands. Romandale remained obliged to sell its Remaining Interest to Kerbel, either by obtaining Fram’s consent to the transaction or by using the buy-sell provisions in the COAs, once SPA for the Lands was achieved. And Kerbel remained obliged to pay Romandale $160,000 per acre for the Remaining Interest. The fact that the expected timing for SPA changed did not alter those obligations – and nothing in the 2005 August Agreement suggests otherwise. For example, there is no “drop-dead date” provision in the agreement.

[75] Thus, the Court of Appeal concluded, because the parties’ obligations under the conditions in the agreement were not altered by the planning changes, it could not be said that compelling performance of the agreement would be to require Romandale to do something “radically different” from that to which it agreed: at para. 236.

The Arbitrator Failed to Apply Binding Precedent

[76] The first four of the five reasons (listed above in para. 25) given by the arbitrator for rejecting the approach taken to the role of foreseeability in the applying the doctrine of frustration in these three Ontario decisions are all different ways of saying that these decisions were wrongly decided.

[77] The arbitrator was bound to apply the law of Ontario and applicable federal law. The law of frustration in Ontario is as articulated by this province’s highest court, read together with any binding authority on the matter issued by the Supreme Court of Canada. For reasons I have set out in paras. 57-67 above, there is no conflict between what the Supreme Court said in Peter Kiewit and Naylor, or what the Court of Appeal for Ontario said in Capital Quality, and a consideration of whether, in determining if a supervening event makes performance of a contract radically different from what was contemplated, the supervening event was foreseen.

[78] The role of foreseeability is, in any event, not just an Ontario-specific point of view. The same approach has been taken in Alberta and in British Columbia as well. The Alberta Court of Queen’s Bench described foreseeability as “an essential element” in cases of frustration: Troika Land Development Corp. v. West Jasper Properties Inc., 2009 ABQB 590 at para. 41. The British Columbia Court of Appeal found that neither Peter Kiewit nor Davis Contractors “obviate the need to show that the supervening event was not foreseeable”: Ballenas Project Management Ltd. v. P.S.D. Enterprises Ltd., 2007 BCCA 166 at paras. 27 to 29.

[79] Ontario and other Canadian courts have carefully defined the test for frustration. This is both proper and necessary because contracts are presumptively enforced. Frustration is a limited exception to this presumption and applies only when “the occurrence of the unexpected event is so much outside the range of risks that the agreement allocates…that the values favouring enforcement are outweighed”: Stephen Waddams, The Law of Contracts, 7th ed. (Toronto: Thomson Reuters Canada Limited, 2017), p. 249.

[80] On the basis of the forgoing analysis, I am driven to the conclusion that the arbitrator’s decision is unreasonable because it failed to undertake the proper legal analysis as defined under Ontario law. This rendered the arbitrator’s decision incompatible with binding precedent: Echelon General Insurance Company v. Ontario (Minister of Finance), 2018 ONSC 5029 at paras. 32-33; Intact Insurance Company v. Allstate Insurance Company of Canada, 2016 ONCA 609, 131 O.R. (3d) 625, at para. 65.

[81] Arbitrators must decide disputes “in accordance with law”. The parties agreed that Ontario law governs their agreement and the arbitration. The arbitrator was bound by that choice and was required to follow Ontario’s highest court. As the Court of Appeal has confirmed, “arbitration represents a process to address a dispute; it does not confer jurisdiction to ignore or rewrite the law and established legal principles”: Omers Realty Corp. v. Sears Canada Inc. (2005), 74 O.R. (3d) 423 (Sup. Ct.), para. 22, aff’d , 80 O.R. (3d) 561 (C.A.).

[82] The Supreme Court of Canada in Naylor, and the Court of Appeal for Ontario in Perkins and Fram, have said unequivocally that foreseeability is a key element in the analysis of frustration. This is because, whether an event was reasonably foreseen is important, if not necessary, in determining whether performance has become something radically different from what was agreed to in the contract.

[83] The arbitrator spent a great deal of time and analysis on the question of whether a failure to provide for a foreseen risk in a contract constituted grounds for denying frustration of that contract.

[84] The arbitrator grounded his finding that the agreement made no provision for the denial of federal environmental approval on the basis that there is no provision which expressly says “if federal approval is denied the contract will/will not remain valid and enforceable” or words to that effect. While this is strictly true, using this approach as the test establishes an impossible standard for whether provision has been made in the agreement for supervening events. No contract can make express provision for every possible eventuality. If that were required to preclude a finding that an agreement had been frustrated, frustration would become an everyday occurrence. A simple example demonstrates this point. In simple terms, the agreement:

• sets the amount of the payment deposit of $350 million;
• sets the amount of gold being purchased at 22% of production; and,
• sets the price to be paid for the gold as the lower of: a) the fixed price of $400 per ounce; and, b) the market price.

Thus, fluctuations in the market price of gold have a material impact on the value of the agreement to the parties. If gold is above $400, it benefits Franco-Nevada and disbenefits Taseko. Yet, there is no provision in the agreement which says “this agreement remains in full force and effect regardless of the market price of gold”. This does not mean that a fluctuation in the price of gold could become a frustrating event merely because there is no provision for price fluctuations stipulated in the agreement. It would be contrary to all commercial sense to require such a stipulation in circumstances like these, where ongoing fluctuations in the price of gold were obviously known to, contemplated, and reasonably foreseen by these two sophisticated contracting parties.

[85] It was in any event unreasonable, in my view, for the arbitrator to conclude that the agreement made no provision for the risk that federal environmental permits might be refused. To the contrary, it is incontrovertible that the agreement did make provision for the parties’ rights and obligations in the event of final a failure to obtain federal approval. The grant or denial of federal approval expressly conditions Franco-Nevada’s obligation to advance funds (section 3.2(h)) and its right (but not obligation) to terminate the agreement (section 5.1(c)). The agreement simply cannot be read in any other way. As in Fram, the agreement provides no “drop dead” date upon which both parties’ obligations under the contract are at an end; it provides only a “drop dead” date upon which Franco-Nevada may elect to terminate the agreement.

[86] While it is true that the Court of Appeal in Quality Capital did not determine that a deliberate choice not to provide for the supervening event in the contract precluded resort to the doctrine of frustration (because on the facts, that issue did not arise), the Court in both Perkins and Fram did make such a determination. The test as articulated by the Court of Appeal in Fram and Perkins (and consistent with the Supreme Court of Canada in Naylor) is the law of Ontario and was binding on the arbitrator. It was unreasonable for the arbitrator to have rejected Ontario law.

[87] Taseko’s real objection is that the manner in which the parties provided for the risk of a denial of the federal permits in the agreement relieved Franco-Nevada of its obligation to advance funds and, after two years, gave Franco-Nevada the right (but not the obligation) to terminate, but conferred no similar rights on Taseko. Whether that was because Taseko could not unilaterally do so, or chose not to, is of no importance. Sophisticated parties negotiating bespoke agreements with the benefit of legal and other professional assistance must be taken to have consciously and carefully evaluated the risks and benefits as allocated in their agreement. Taseko, of course, has the unilateral right not to proceed with the mining development at all. But, if Taseko did not want Franco-Nevada to have the only “drop dead” date by which to terminate, Taseko’s option, in the absence of being unable to negotiate such a provision for its own benefit, was to walk away from negotiations and look for another source of financing for the Prosperity project. There is nothing “harsh or antiquated” about this. It is the everyday reality of sophisticated commercial negotiations.

[88] It is not the function of the court (or arbitrator) to rewrite a contract for the parties nor is it the court’s (or the arbitrator’s) role to relieve one of the parties against the consequences of an improvident contract: Pacific National Investments Ltd. v. Victoria (City of), 2004 SCC 75, [2004] 3 S.C.R. 575, at para. 31. However, by finding the agreement had been frustrated and relieving Taseko of the burden of the agreement which it entered into with Franco-Nevada, this is exactly what the arbitrator has done.

Did the Agreement Involve a Specific Project No Longer Capable of Being Achieved?

[89] The arbitrator also interpreted the agreement as being specific to one project, a 33-year mine project described in Taseko’s 2009 Technical Report, and one timeline, which assumed federal approval and commencement of site preparation in late 2010.

[90] There are two significant problems with the arbitrator’s analysis of this issue which render his conclusion unreasonable.

[91] First, there were three versions of the Prosperity project discussed at various points and in various documents:

(a) Taseko’s initial 20-year mine plan, which was submitted for federal approval in 2009 and rejected in 2010;

(b) Taseko’s 33-year mine plan discussed in its 2009 Technical Report, which Taseko never submitted for provincial or federal approval; and

(c) Taseko’s revised 20-year mine plan, known as New Prosperity, which was submitted for federal approval in 2011 and rejected in 2014.

[92] The 33-year plan which the arbitrator held was the “purpose” of the agreement, could not have been approved in 2010 or in 2014 because it was never submitted to either government authority. It was the 20-year mine plan, not the 33-year plan, that was submitted for review.

[93] This confusion between the different mine plans makes sections of the arbitrator’s Award illogical and commercially unreasonable. For example, the arbitrator focused on Schedule D of the agreement, which he said targeted September 2010 as the date for receipt of federal permitting. But Schedule D is based on the 20-year mine plan. Thus, Schedule D could have no bearing on the timeline for approval of the 33-year plan in the 2009 Technical Report.

[94] The arbitrator was aware of this anomaly but failed to explain how it could be reconciled with his overall conclusion about the agreement’s purpose. If the parties’ obligations were tied to the 20-year project, how could the agreement’s purpose be exclusively to facilitate the 33-year project on the 2009 Technical Report?

[95] The logical consequence of inconsistencies such as this is that, on the arbitrator’s theory of the agreement and its purpose, even if Taseko’s New Prosperity project proposal submitted in 2011 had been approved by the federal government in 2014, the agreement would still have been frustrated because: a) it would have been an approval for a 20 year mine, not the 33 year mine proposal that the arbitrator found was the whole basis of the agreement; and, b) the 2011 proposal submitted for federal approval was already after the 2010 permitting window that the arbitrator also identified as being the basis of the agreement. The arbitrator variously considered that the agreement had been frustrated: when federal approval was refused in 2010; when federal approval was refused in 2014; and, when the expiry of the warrants occurred in 2017 without federal approval having been obtained. These cannot all be correct, yet there is no coherent, rational explanation for how, or when, the alleged frustration actually occurred.

[96] These inconsistencies in the arbitrator’s reasoning also illustrate how the arbitrator’s approach is divorced from the agreement itself. The metal purchase agreement was tied to neither a 20-year or a 33-year mine, nor to forecast approvals in September 2010, or forecast commencement of site preparation in October 2010. Rather, the agreement is tied to the gold taken from the Prosperity Property over a 40 plus year term. Three provisions in the Agreement make this clear:

(1) Section 7.1(a) gives Taseko unilateral discretion to “amend the mine plan for the Prosperity Project at any time” and to make all decisions about the timing of the project. Section 7.1(b) stipulates that Taseko is under no obligation to proceed with the development of the mine at all. Point-in-time estimates of when certain milestones might occur must be read in the context of the controlling contractual term—that Taseko had the unilateral right to amend the project plan and the timelines or not to proceed at all. If the parties intended to limit the agreement to one mine proposal on one timeline, section 7.1(a) and (b) make no sense;

(2) Section 5.1(a) defines the agreement’s term as 40 years, plus extensions, and only Franco-Nevada has a right to early termination. Thus, Taseko agreed to be bound for at least 40 years, not just for the few years of its permitting attempts; and,

(3) Section 2.1(a) requires Taseko to sell Franco-Nevada 22% of all “Produced Gold”. Section 1.1 defines “Produced Gold” as all gold that is “extracted or otherwise recovered from the Prosperity Property…”. This definition is not tied to any specific project plan, and certainly not the 2009 Technical Report which governs the construction of the ore processing plant. Rather, it requires Taseko to sell 22% of all gold extracted from the Prosperity Property to Franco-Nevada until at least 2050, irrespective of which mine plan might have been approved, or when.

[97] The arbitrator was obliged to consider all material terms in the agreement and avoid an interpretation that would render a provision meaningless or redundant. His interpretation of the agreement rendered meaningless Taseko’s absolute discretion to change the mine plan. And his Award simply does not address how his timeline-specific interpretation tied to federal approval by 2010 (or even 2014) can be reconciled with the lengthy term of the agreement and the express rights and obligations of the parties contained therein.

[98] The issue that most appears to have concerned the arbitrator and Taseko is an alleged anomaly resulting from the denial of the federal approvals in light of the specific rights and obligations of the parties contained in the agreement. Taseko has no obligation to proceed with the development of the mine, but if it does within the 40-year term, it would have to honour its commitment to sell 22% of its gold production to Franco-Nevada at the agreed price. Franco-Nevada, since two years elapsed without required approval in 2012 under section 5.1(c), effectively has an option for the duration of the contract term. If the prevailing price of gold when a future approval to develop the mine might be obtained is favourable, Franco-Nevada can insist on its bargain. If the prevailing price of gold is unfavourable, Franco-Nevada can exercise its latent right to terminate under section 5.1(c). This result, however, is not radically different performance from that set out in the agreement. Both parties have respective rights which could involve the prevailing market price for gold. If Taseko decides the price is too low to warrant proceeding with the Prosperity project, it does not need to proceed. This is exactly what was contemplated by the parties by the express terms and conditions they agreed to.

[99] The federal government’s denial of the necessary permits was a risk known to both parties at the time of contracting. The agreement contains provisions allocating the burden of that risk. The federal government’s refusal did not change the parties’ obligations from those contemplated by the agreement in any way. Thus, Taseko’s obligations under the agreement cannot be said to have become, as a result of the denial of the federal permits in 2010, or in 2014, something radically different from what was agreed to. The arbitrator’s Award to the contrary was, for all these reasons, unreasonable.

Did the Arbitrator Nevertheless Deal with the Frustration Scenario in the Alternative?

[100] Taseko makes the alternative argument that the arbitrator’s decision about foreseeability did not affect the outcome because he went on to find, as a fact, that the supervening event was, in any event, neither foreseen nor reasonably foreseeable.

[101] The arbitrator acknowledged during his initial review of contract interpretation principles that evidence of the factual matrix in which the agreement was made is admissible. Factual matrix evidence, however, he also acknowledged, is evidence of surrounding circumstances reasonably known to both parties at the time the contract was made; subjective evidence of intention is not admissible to assist with the interpretation of a contract.

[102] The arbitrator found on the evidence that both Taseko and Franco-Nevada were aware, when the agreement was made, that there was a risk that the federal government might refuse to grant approval. In addition to the fact that this was expressly contemplated in sections 3.2(h) and 5.1(c) of the agreement, he also noted that Taseko’s witnesses candidly acknowledged that they knew that there was always a “permitting risk” — a risk that a required permit would be refused or delayed. The same risk was identified by Franco-Nevada in the course of its due diligence. In para. 154 of the Award, the arbitrator expressly found that “the alleged supervening event was not unforeseen—both parties knew there was some possibility of its occurrence.”

[103] Despite this finding, the arbitrator went on to consider the subjective assessments of the parties as to the likelihood of this risk being realized. Taseko’s subjective assessment, the arbitrator found, was that the federal approvals were almost certain to be granted. Franco-Nevada’s subjective assessment was “50/50” or just “more likely than not”. The arbitrator noted that there was no evidence of what a “reasonable person” would have thought was the likelihood of a denial of the federal permits. Nevertheless, he concluded that a “reasonable person” would have believed as the parties did. He then engaged in some consideration of what “reasonably foreseeable” means in this context. He found that a “reasonable person” would have known there was some possibility of the federal permits being denied. If that were the test, the denial of the federal permits was reasonably foreseeable. However, he also found that these parties thought it more likely than not that the federal permits would be approved, such that if a balance of probability was the test, a refusal of the federal permits was not reasonably foreseeable (para. 156).

[104] The arbitrator did not, however, actually resolve this issue and make a finding that the federal permit denial was not reasonably foreseeable. His purpose in engaging in this analysis seems to have been to cast further doubt on the introduction of the concept of foreseeability into the analysis of frustration at all. He refers, in para. 157, to the concept of foreseeability creating a “jarring disconnect” between “normal principles of objective contractual interpretation” and an interpretation which turns on the “parties’ subjective expectations”. He went on to find that the apparent purpose of an analysis of reasonable foreseeability is “to attribute a form of fault to a party who otherwise would be entitled to frustration relief, by finding, in effect, that the party either knowingly (but not expressly) accepted the risk or could have and should have provided for it.”

[105] There are two fundamental problems with Taseko’s alternative argument. First, the substance of Taseko’s submission is that whether the arbitrator was right about foreseeability playing a role in the analysis of frustration does not matter to the reasonableness of the outcome because, even accepting reasonable foreseeability has a role, the arbitrator went on to find that the denial of the federal permits was not reasonably foreseeable. This is not, however, what the arbitrator did. Having posed two alternative outcomes based on two alternative understandings of what reasonable foreseeability means, he did not resolve which interpretation of reasonable foreseeability was applicable, and made no finding on the point.

[106] The second, and more fundamental, problem arises from the nature of the arbitrator’s analysis, viewed in light of the reasonableness test set out by the Supreme Court in Vavilov. At para. 96 of Vavilov, the Supreme Court held that where the reasons given by a decision maker contain a fundamental gap or reveal that the decision is based on an unreasonable chain of analysis, it is not ordinarily appropriate for the reviewing court to fashion its own reasons in order to buttress the decision even if the outcome of the decision could be reasonable under different circumstances. To be reasonable, a decision must be based on reasoning that is both rational and logical. It follows that a failure in this respect may lead a reviewing court to conclude that a decision must be set aside. While reasonableness review is not a “line-by-line treasure hunt for error”, the reviewing court must be able to trace the decision maker’s reasoning without encountering any fatal flaws in its overarching logic: Vavilov, at para. 102. A decision will be unreasonable if the reasons for it fail to reveal a rational chain of analysis or if they reveal that the decision was based on an irrational chain of analysis: Vavilov, at para. 103.

[107] In essence, having recognized that evidence of subjective intention has no role in contract interpretation, it was illogical and irrational for the arbitrator to then delve into the parties’ subjective intentions or expectations, or to ground any kind of analysis of the role of reasonable foreseeability on such subjective expectations.

[108] The answer is simple, based on well-settled, first principles. Whether an event was, as the Court of Appeal said in Quality Capital, “within the contemplation of the parties”, or, as the Supreme Court and the Court of Appeal said more recently, “foreseeable”, is determined from an interpretation of the agreement, reading the language of the agreement: (i) in accordance with its ordinary and grammatical sense; (ii) in the context of the agreement as a whole; (iii) in conjunction with the factual matrix reasonably known to both parties at the time, without allowing factual matrix to overwhelm ordinary and grammatical meaning; (iv) excluding subjective evidence of intention; and, (v) in a manner that achieves commercial efficacy: Sattva.

[109] The arbitrator found, on the evidence, that both Taseko and Franco-Nevada were aware, when the agreement was made, a) the federal government had not yet granted permit approval and, b) there was a risk that the federal government might refuse to do so. The arbitrator also found that this risk was reflected in several important provisions of the agreement, in particular as a precondition to Franco-Nevada advancing the $350 million payment deposit (section 3.2(h)), and as a basis, if the approvals were not obtained within two years, for Franco-Nevada to terminate the agreement unilaterally (section 5.1(c)). The agreement provides for what happens if the approval is not granted: Franco-Nevada would be under no obligation to advance the payment deposit; and, after two years, Franco-Nevada would have the right, but not the obligation, to terminate the agreement. If Taseko had obtained the federal permits within two years, and satisfied the other conditions precedent, Franco-Nevada would have been obliged to make the payment deposit and would have had no right to terminate. But, because Taseko failed to obtain federal approval within two years, Franco-Nevada has not been required to advance the payment deposit and can, in the event of a future approval, either waive its conditions and pay the deposit or terminate the agreement.

[110] In concluding that the agreement does not contemplate risk of failure to obtain federal permits, the arbitrator failed to give effect to his own interpretation of sections 3.2(h) and s. 5.1(c) and to the evidence which he found was reasonably known to both parties at the time of contracting. As a result, his decision is unreasonable—the arbitrator reached a conclusion that “cannot follow from the analysis undertaken.” Further, Taseko knew that Franco-Nevada had the federal permit approval precondition to advancing the payment deposit and the “drop dead” right to terminate after two years, yet was unable, or chose not, to incorporate concomitant protections into the agreement for itself. Thus, in keeping with well settled principles of contract interpretation, it was reasonably foreseeable to both parties, without regard to any evidence of subjective intentions or expectations, that the federal government might not grant the permit approvals. The parties allocated this foreseeable risk and made provision for that event (or, in Taseko’s case, failed to make provision) in the agreement. This case is, in these respects, indistinguishable from the analysis in Naylor, Perkins and Fram.

[111] Franco-Nevada’s appeal is, for these reasons, allowed.

Permanency

[112] In light of my conclusion that the arbitrator’s refusal to follow Ontario law regarding the role of foreseeability, which was binding on him, rendered the Award unreasonable, it is not necessary to deal with the question of permanency.

Was the Arbitrator’s Cost Award Unreasonable?

[113] In light of my conclusion on the first issue, it is not necessary to determine whether the cost award of the arbitrator was unreasonable.

[114] Had it been necessary to decide the matter, I would have dismissed Franco-Nevada’s appeal as to costs.

[115] The agreement did not prescribe any scale of costs for the arbitration. Instead, it left costs to the arbitrator’s discretion (Schedule L, section 4(c)). The agreement included a clause related to appeal costs (section 4(d)) which specifically says costs shall be awarded on a substantial indemnity basis. But that clause relates, by definition, only to court proceedings, not to the arbitration itself. The rules of court do not apply to the discretion of a private arbitrator to award costs.

[116] The ordinary rule in arbitration proceedings is that the successful party should recover reasonable legal fees without reference to any court scale, absent the agreement of the parties or specific legislation. Had I not granted Franco-Nevada’s appeal on the first issue, I would have found that the arbitrator’s decision to award Taseko “its reasonable legal fees and expenses” was reasonable.

Should Specific Portions of the Appeal Record be Sealed?

[117] Taseko brought a motion for a limited confidentiality order over confidential and commercially sensitive information filed in this appeal. Franco-Nevada took no position on Taseko’s motion.

[118] Taseko seeks an order to seal or redact three discrete categories of confidential and commercially sensitive information from the public record: (1) non-public information related to ongoing facilitated discussions between Taseko, the TFN, and the Province of British Columbia; (2) non-public information related to Taseko’s prior discussions with third parties about a potential sale of Taseko’s Prosperity mineral rights; and (3) non-public technical, scientific, or proprietary information related to the Prosperity project.

[119] Subsection 137(2) of the Courts of Justice Act provides that this Court may order that any document filed in a civil proceeding be treated as confidential, sealed, and not form part of the public record. The Superior Court of Justice also has inherent jurisdiction to grant sealing orders and other forms of confidentiality orders.

[120] However, it has been well established that the “open court” principle is protected by the constitutional guarantee of freedom of expression and is essential to the proper functioning of our democracy. In order to grant an order of the nature sought, the applicant must show that the information sought to be excluded from the public record meets the test established by the Supreme Court of Canada, first in Sierra Club of Canada v. Canada (Minister of Finance), 2002 SCC 41, [2002] 2 S.C.R. 522, and later refined in Sherman Estate v. Donovan, 2021 SCC 25. A party seeking a confidentiality order must establish three elements: (1) public disclosure would pose a serious risk to an important public interest; (2) no reasonably alternative measures would prevent this risk; and (3) the benefits of the order outweigh any negative effects.

Serious Risk to an Important Public Interest

The Information is Confidential

[121] The ongoing discussions between Taseko, the TFN, and British Columbia are protected by contractual confidentiality requirements. In December 2019, Taseko and the TFN entered into a Standstill and Preservation of Rights Agreement containing a confidentiality clause providing that “the terms and existence of this Agreement are and shall remain strictly confidential”, subject to limited exceptions, none of which apply. Later, Taseko, the TFN, and British Columbia signed a Communications Protocol and Confidentiality Agreement containing a confidentiality clause providing that “the Parties expressly acknowledge and agree that (a) the terms of the Standstill Agreement and this Agreement; and (b) the substance of any negotiations between them with respect to exploring a potential Resolution, are and shall remain strictly confidential”, subject to limited exceptions, none of which apply. The parties have extended their agreement multiple times, and it remains in effect today.

[122] Taseko’s prior discussions with third parties about a potential sale of Taseko’s Prosperity mineral rights were made with a reasonable expectation of confidentiality and are subject to contractual confidentiality requirements. In 2018-20, Taseko entered into three confidentiality agreements with third parties to facilitate discussions about a potential purchase of Taseko’s Prosperity mineral rights. Each of these agreements contained a standard non-disclosure clause prohibiting the parties from disclosing “the existence or the terms and conditions of this Agreement or of the proposed Transaction that is the subject of this Agreement”, subject to limited exceptions. Although these agreements have since terminated, the parties’ prior discussions were nonetheless treated as confidential at all relevant times. Moreover, the Agreement’s confidentiality provisions protect Taseko’s prior discussions with third parties about a potential sale of Taseko’s Prosperity mineral rights. These confidentiality provisions were designed precisely to cover this kind of commercially sensitive information.

[123] The non-public technical, scientific, or proprietary information related to the Prosperity project is confidential. For example, Franco-Nevada’s appeal record includes Taseko’s Resource Block Model, which shows the “blocks” comprising the mineralized ore body for the Prosperity deposit. This information is confidential and proprietary, belonging to Taseko alone, and is the product of years of investment and study by Taseko. This is the type of non-public technical, scientific, or proprietary information over which courts have routinely granted confidentiality orders. By contrast, any public technical, scientific, or proprietary information related to the Prosperity project, such as permitting documents that have already been posted on the internet by Taseko or a regulatory authority, would remain in the public record.

Disclosure Would Harm Taseko

[124] All of the information sought to be protected is commercially sensitive, in the sense that disclosure of this information would expose Taseko to harm.

[125] Public disclosure of non-public information related to the ongoing facilitated discussions between Taseko, the TFN, and British Columbia, would pose a serious risk of compromising those ongoing discussions, to the detriment of all three parties concerned. Further, disclosure would pose a serious risk of damaging Taseko’s goodwill and relationships with the TFN and British Columbia, which would undermine a key premise underpinning these discussions in the first place. This damage could have long-lasting impacts on Taseko and its relationships with First Nations more broadly.

[126] Similarly, public disclosure of non-public information related to Taseko’s prior discussions with third parties about a potential sale of Taseko’s Prosperity mineral rights would pose a serious risk of compromising Taseko’s ability to maintain and maximize the market value of its assets. Specifically, it would pose a serious risk of giving any future potential purchasers insight into how many other potential purchasers have considered purchasing Taseko’s mineral rights and the nature of their discussions with Taseko. These discussions are analogous to the sales process situation in receiverships and insolvencies, where the public interest of maximizing value is routinely recognized and such information protected.

[127] And, also in a similar vein, disclosure of non-public technical, scientific, or proprietary information related to the Prosperity project would also pose a serious risk of harming Taseko’s legitimate commercial interests, including by giving competitors, potential future business partners, and others information that could be used to Taseko’s detriment.

Taseko and Others Had a Reasonable Expectation of Confidentiality

[128] The information was accumulated with a reasonable expectation that it would be kept confidential. The first two categories of information were accumulated under contractual confidentiality agreements prohibiting disclosure. These confidentiality agreements reflected the parties’ shared expectation that the information would be kept confidential. This expectation was reasonable given the sensitivity of the information, the nature of the parties’ relationships, and the principle that “a court will generally enforce the terms of a written contract”.

[129] The third category of information comprises non-public technical, scientific, or proprietary information related to the Prosperity project, all of which Taseko reasonably expected would not be made public. This category of information excludes any documents that have already been made public, for which Taseko would have no reasonable expectation of confidentiality.

No Reasonably Alternative Measures

[130] The second criterion “asks whether there are reasonably alternative measures; it does not require the adoption of the absolutely least restrictive option”. Depending on the circumstances, potential alternative measures may include sealing orders, publication bans, orders excluding the public from a hearing, and redaction orders.

[131] Here, Taseko proposes the least restrictive measure possible: a limited confidentiality order over only the specifically identified information. The order will authorize redaction of only that specifically identified information, thereby preserving public access to the vast majority of the record. This order sought is, therefore, tailored to protect the public interest in open courts as much as reasonably possible.

[132] No lesser measures would prevent the serious risk to the important public interest in preserving confidential information. For example, a publication ban would not prevent members of the public from accessing confidential information about the ongoing discussions between Taseko, the TFN, and British Columbia; prevent any future potential third-party purchasers from accessing confidential information about Taseko’s prior discussions with other potential third-party purchasers; or prevent Taseko’s competitors or potential future business partners from accessing and using Taseko’s non-public technical, scientific, or proprietary information to Taseko’s detriment. Likewise, an order excluding the public from the hearing—on its own—would leave the information in the record completely unprotected. Therefore, a confidentiality order over the information, which is achieved by a redaction of the information from the documentary record filed, is necessary.

The Limited Order Is Proportionate in that its Benefits Outweigh Any Negative Effects

[133] The confidentiality order being sought has been carefully tailored to protect the public interest in open courts as much as possible while still protecting the public interest in preserving confidential information. This limited order will authorize redaction of only the specifically identified information, thereby preserving public access to the vast majority of the record.

[134] Further, the subject matter of the appeal involves a dispute over whether a private commercial agreement for the purchase and sale of minerals has been frustrated, and whether the arbitrator’s decision was reasonable. This dispute does not engage the public interest, as that term is usually employed. The impact of the order on the public interest is minimal. By contrast, the public interest in preserving confidentiality of the specifically identified information in this case weighs heavily in the analysis. As outlined above, disclosure of the information poses a serious risk to another important public interest.

[135] Finally, the confidentiality order being sought will not prejudice Franco-Nevada in any way. Franco-Nevada already has full access to the information in question and has been permitted to rely on it (to the limited extent necessary) as part of its appeal. Likewise, the Court has also had full access to that information in a complete, unredacted version of the record. Therefore, the order does not interfere with the proper adjudication of this dispute in any way.

[136] For these reasons, the motion is granted. The order providing for redaction of the specifically identified information shall issue. The parties shall file the full redacted Appeal Record with the Court and post it on CaseLines. An unredacted hard copy of the full Record shall be placed in a sealed envelope and filed with the Court. A copy of this endorsement and the related order, and any subsequent court order which may affect the materials, shall be attached to the envelope.

Conclusion

[137] Franco-Nevada’s appeal from the arbitrator’s Award declaring the agreement has been frustrated, is granted.

[138] Taseko’s motion to seal specified portions of the appeal record is granted.

Costs

[139] The parties reached an agreement on appeal costs such that: if Taseko is successful and the appeal is dismissed, it will receive substantial indemnity costs on the appeal of $105,000 under s. 4(d) of Schedule L to the agreement; if Franco-Nevada succeeds, it will receive partial indemnity costs on the appeal of $95,000 under the Rules of Civil Procedure. Partial indemnity costs are therefore awarded to Franco-Nevada in the amount of $95,000 (all inclusive).

Penny J.

Date: April 17, 2023

[1] In Bang, the motion judge rejected the defence of frustration because the supervening event was within the contemplation of the parties and was foreseen. There was an appeal but the appeal did not raise, and the Court of Appeal did not address, the foreseeability issue.