COURT OF APPEAL FOR ONTARIO

CITATION: SS&C Technologies Canada Corp. v. The Bank of New York Mellon Corporation, 2024 ONCA 675

DATE: 20240912

DOCKET: C69439

Tulloch C.J.O., Hourigan and Miller JJ.A.

BETWEEN

SS&C Technologies Canada Corp.

Applicant

(Respondent/Appellant by Cross-Appeal)

and

The Bank of New York Mellon Corporation* and CIBC Mellon Global Securities Services Company

Respondents

(Appellant*/Respondent by Cross-Appeal*)

J. Thomas Curry, Eli Mogil, Brandon Kain, Brian Kolenda, Erin Chesney, Christopher Yung, and Andrew Locatelli, for the appellant/respondent by way of cross-appeal

Chris G. Paliare, Ren Bucholz, Glynnis Hawe, Catherine Fan, and Janet-Lee Song, for the respondent/appellant by way of cross-appeal

Heard: April 24 and 25, 2024

On appeal from the judgment of Justice Markus Koehnen of the Superior Court of Justice, dated July 11, 2023, with reasons reported at 2021 ONSC 2657, 18 B.L.R. (6th) 113; 2022 ONSC 1652, 31 B.L.R. (6th) 112; and 2023 ONSC 4083.

Tulloch C.J.O.:

PART A: LIABILITY

I. INTRODUCTION

[1] Every day, sophisticated parties enter into commercial contracts to achieve their business objectives and provide certainty and stability for ongoing business relationships. Sometimes the parties’ subjective understandings of the deal they made diverge. In those circumstances, the courts are asked to intervene, and to hold parties to the bargain they struck by objectively determining the meaning of the words the parties chose within the context of the contract’s surrounding circumstances and commercial purpose. If applying those tools leaves any doubt, then courts can also look to the parties’ course of conduct in performing the contract to determine what they meant.

[2] This case involves a contract dispute between the Canadian affiliate of the world’s largest hedge fund and private equity administrator, SS&C Technologies Canada Corp. (“SS&C”), and the world’s largest custodial bank, Bank of New York Mellon Corporation (“BNY”), and requires us to determine and enforce the bargain they struck.

[3] The parties agreed in 1999 that SS&C would provide market pricing data to a singular client, Mellon Trust, but, after more than one and a half decades of doing business together, they now cannot agree on who that client is. SS&C argues that “Mellon Trust” is a single legal entity, namely BNY, and that BNY breached the contract by redistributing the data SS&C provided to CIBC Mellon Global Securities Company (“CIBC Mellon”) and other BNY affiliates. BNY, in turn, claims that the singular word “Client” actually means many different entities, namely both BNY and its numerous affiliates.

[4] The trial judge was partially correct in his liability ruling. He correctly ruled that the “Client” did not include CIBC Mellon or other affiliates that BNY and its predecessor acquired after entering into the relevant contract. But he incorrectly ruled that the “Client” included the entities BNY’s predecessor owned when it entered the contract.

[5] After carefully considering the contract’s text, the commercial purpose, and the course of performance, the trial judge ruled that the term “Client” meant a single legal entity and, thus, held that BNY breached the contract by redistributing data to CIBC Mellon and other affiliates it acquired after contracting. He was right to find that BNY’s claim that the contract entitled an ever-shifting group of dozens of unspecified affiliates to access data was inconsistent with the text, the commercial objectives it reflected, and both parties’ pre-dispute actions. I would accordingly dismiss BNY’s appeal on liability.

[6] But the trial judge failed to give full effect to his thorough findings concerning the parties’ bargain. His conclusion in his liability ruling that the contract authorized BNY to share data with its original affiliates contradicted his findings that the text referred to a single legal entity, that SS&C neither knew nor should have known that Mellon Trust referred to a group of entities, and that both parties’ actions confirmed that BNY alone was SS&C’s counterparty. His subsequent justification for this conclusion, that restricting the client to a single legal entity would be unfair to BNY, improperly used a one-sided approach to commercial reasonableness to rewrite the contract by sweeping away the constraints the text and surrounding circumstances imposed, and is based on an obvious factual error that BNY and its predecessor did not need any data themselves. He also overlooked that, even if there was any doubt about what the text meant, both parties’ pre-dispute actions resolved it by unequivocally confirming that BNY alone was SS&C’s counterparty. I would thus allow SS&C’s cross-appeal on liability and rule that the contract authorized only BNY to access data.

[7] Finally, this decision reaffirms this court’s longstanding precedent that trial courts can change their orders before entering them in the court record to account for outcome-determinative matters that they overlooked. This flexible test, which this court has never overturned, furthers the justice system’s efficient operation and access to justice by enabling trial courts to promptly correct mistakes that would otherwise go unremedied or lead to needless and costly appeals. It also respects litigants’ interest in certainty, finality, and meaningful appellate review by preventing the re-argument of matters courts considered and rejected, barring reconsideration of non-outcome-determinative points, and requiring trial judges to give clear reasons for making changes to their orders. In this case, the trial judge reasonably changed his unentered judgment because he initially overlooked and did not address SS&C’s claim that BNY breached the contract by redistributing data to CIBC Mellon.

[8] My colleague, Hourigan J.A., writes separately to address the issue of damages. I concur entirely with his reasons and disposition.

II. BACKGROUND

[9] BNY is a custodial bank that, together with its many subsidiary and affiliate custodial entities, holds funds in safekeeping for its customers. A key component of its operations is the utilization of market pricing data to calculate the value of its customers’ holdings. SS&C, among other things, sells that data to other custodial banks, similar to BNY.

[10] SS&C’s predecessor, the Securities Valuation Company Inc. (“Securities Valuation”), agreed in 1999 to sell data services to two clients. First, it made the Mellon Trust Agreement with “Mellon Trust.” Mellon Trust is not a legal entity. Instead, it was the brand name under which BNY’s predecessor, Mellon Financial Corporation (“Mellon Financial”), and its custodial entities, carried on business. Second, Securities Valuation made the CIBC Mellon Agreement with CIBC Mellon, a joint venture of Mellon Financial and the Canadian Imperial Bank of Commerce.

[11] The Mellon Trust Agreement and the CIBC Mellon Agreement, which I refer to as the “Agreements,” are very similar. Each contract is between Securities Valuation and a singular “Client,” Mellon Trust and CIBC Mellon, respectively. Each provides that “either party” can terminate the contract if the “other party” becomes insolvent. Each also requires the “Client” to exclusively use the data that Securities Valuation provides and prohibits it from redistributing data to any entities other than its customers. Finally, each contract stipulates that each party’s “successors and permitted assigns” will be bound by and benefit from the contract and prohibits non-consensual assignment.

[12] From 1999 onwards, Mellon Financial redistributed the data SS&C provided under the Mellon Trust Agreement to CIBC Mellon and its other custodial entities. Mellon Financial had 14 such entities at the time of contracting and acquired more between 1999 and 2006.

[13] In 2003, Securities Valuation and CIBC Mellon amended the CIBC Mellon Trust Agreement’s definition of “Client” to include a CIBC Mellon affiliate. The amendment authorized this CIBC Mellon affiliate to access specified amounts of SS&C’s data. The parties never expanded the Mellon Trust Agreement’s definition of “Client.”

[14] In 2005, SS&C acquired Securities Valuation and assumed its rights and duties under the Agreements.

[15] In 2007, Mellon Financial merged with Bank of New York to form BNY. Before the merger closed, Mellon Financial’s senior executive responsible for market pricing data sent SS&C’s chief representative a signed letter affirming that the parties to the Mellon Trust Agreement were Mellon Financial and SS&C. She told him that BNY would succeed to Mellon Financial’s rights and duties under that contract. She did not mention that either Mellon Financial’s custodial entities or the additional Bank of New York custodial entities that the merged entity would own were or would become parties to the Mellon Trust Agreement. SS&C’s chief representative confirmed his agreement by countersigning the letter.

[16] In the same letter, Mellon Financial inaccurately told SS&C that the merger would not expand BNY’s need for SS&C’s data and that, if BNY needed more data, it would inform SS&C. In fact, BNY expanded its data use by redistributing the Mellon Trust Agreement data to numerous newly acquired former Bank of New York custodial entities, all without informing SS&C.

[17] Next, in 2011, CIBC Mellon stopped paying SS&C for its data and began surreptitiously obtaining it from BNY. It terminated the CIBC Mellon Agreement and, inaccurately, told SS&C that it no longer needed its services. But behind SS&C’s back, BNY redistributed the data SS&C shared with it under the Mellon Trust Agreement to CIBC Mellon.

[18] SS&C discovered this scheme in Fall 2016 when a Canadian Imperial Bank of Commerce representative forwarded CIBC Mellon’s complaints about SS&C’s data delivery to SS&C. In response, SS&C asked BNY to disclose whether it was redistributing SS&C’s data to other custodial entities. When BNY refused, SS&C terminated the Mellon Trust Agreement and sued BNY for breaching it.

III. THE TRIAL JUDGE’S RULINGS

[19] The central liability question was simple: Who was the “Client” under the Mellon Trust Agreement? SS&C argued that the answer was only BNY, as, in SS&C’s view, the contract prohibited BNY from redistributing data to CIBC Mellon and its other custodial entities. BNY claimed that all its custodial entities, including CIBC Mellon, were also the “Client” and entitled to receive data, even entities BNY acquired after the contract was signed.

[20] The trial judge held that the contract only authorized Mellon Financial’s non-CIBC Mellon custodial entities at the time of contracting and BNY itself to receive data: 2021 ONSC 2657, 18 B.L.R. (6th) 113 (“Liability Decision”). He found that the contract’s text, the commercial purpose, the surrounding circumstances, and the course of performance showed that it prohibited data sharing beyond the named “Client,” Mellon Trust, a term which he found the parties used to refer to a single legal entity. After rejecting BNY’s limitations and equitable defences, he ruled that BNY breached the Mellon Trust Agreement by redistributing data to other custodial entities. He also ruled that CIBC Mellon breached the CIBC Mellon Agreement by obtaining SS&C’s data from BNY after 2011.

[21] The trial judge’s liability ruling overlooked one of SS&C’s claims and did not explain a critical conclusion. First, instead of addressing SS&C’s claim that BNY breached the Mellon Trust Agreement by redistributing data to CIBC Mellon, the trial judge ruled that CIBC Mellon was liable for breaching the CIBC Mellon Agreement. SS&C never advanced this claim. Second, the trial judge did not explain his conclusion that the Mellon Trust Agreement authorized BNY to share data with Mellon Financial’s custodial entities at the time of contracting. SS&C asked the trial judge to correct these points before beginning the damages trial and entering his judgment.

[22] The trial judge changed his unentered judgment to correct the first point and explained his conclusion concerning the second point: 2022 ONSC 1652, 31 B.L.R. (6th) 112. He removed his mistaken conclusion that CIBC Mellon had breached the CIBC Mellon Agreement and replaced it with a finding that BNY breached the Mellon Trust Agreement by redistributing data to CIBC Mellon. Next, he cited commercial reasonableness to justify his determination that BNY could share data with Mellon Financial’s custodial entities as of the time of contracting. He found that BNY’s predecessor, Mellon Financial was a holding company and thus, in his view, had no use for the data itself. He accordingly concluded that it would undermine the contract’s commercial purpose to limit the definition of “Client” to BNY and prevent it from redistributing data to those entities.

IV. ISSUES

[23] BNY argues in its liability appeal that the trial judge should have interpreted the Mellon Trust Agreement to authorize CIBC Mellon and all other BNY custodial entities to access data. It also submits that SS&C’s claims are limitations-barred, and the trial judge lacked jurisdiction to change his unentered judgment to hold BNY liable for redistributing data to CIBC Mellon. SS&C, in turn, cross-appeals on liability. It maintains that the trial judge should have ruled that the Mellon Trust Agreement only authorized BNY, and not also Mellon Financial’s custodial entities at the time of contracting, to access data.

(a) The Mellon Trust Agreement Only Authorized BNY To Obtain Data

[24] I begin with the crux of the parties’ dispute: Who is the “Client” under the Mellon Trust Agreement? Is it only BNY as SS&C contends, BNY along with Mellon Financial’s affiliates at the time of contracting as the trial judge ruled, or BNY, CIBC Mellon, and all other BNY or Mellon Financial affiliates as BNY submits? I would dismiss BNY’s appeal and allow SS&C’s cross-appeal because, in my view, SS&C has the better argument.

(i) Governing Interpretive Principles and Standard of Review

[25] The trial judge correctly instructed himself to ground his interpretation in the contract’s text, consider the contract’s commercial purpose, and examine the parties’ course of performance to resolve any ambiguity. Parties are presumed to mean what they say in the text, which is the focal point of contract interpretation: Earthco Soil Mixtures Inc. v. Pine Valley Enterprises Inc., 2024 SCC 20, 492 D.L.R. (4th) 389, at para. 63; Heritage Capital Corp. v. Equitable Trust Co., 2016 SCC 19, [2016] 1 S.C.R. 306, at para. 47. Courts also favour interpretations that advance the contract’s commercial purpose and are commercially reasonable: 2484234 Ontario Inc. v. Hanley Park Developments Inc., 2020 ONCA 273, 150 O.R. (3d) 481, at para. 65. Finally, the parties’ course of performance can help determine what the parties intended if the text, considered together with the surrounding circumstances, is ambiguous: Prism Resources Inc. v. Detour Gold Corporation, 2022 ONCA 326, 162 O.R. (3d) 200, at para. 18.

[26] Because the Mellon Trust Agreement is a negotiated agreement between two sophisticated parties, the standard of review is highly deferential. As a general rule, this court can only interfere if the trial judge committed a palpable and overriding error. However, extricable errors of law are reviewed for correctness: Ledcor Construction Ltd. v. Northbridge Indemnity Insurance Co., 2016 SCC 37, [2016] 2 S.C.R. 23, at para. 21.

(ii) The Contract Barred Redistributing Data to CIBC Mellon

[27] BNY first argues that the Mellon Trust Agreement authorized CIBC Mellon to access data that SS&C provided under that contract. The crux of BNY’s submission is that the trial judge’s rejection of this argument is internally inconsistent. BNY notes that CIBC Mellon was a Mellon Financial custodial entity in 1999. It argues that the trial judge’s ruling that CIBC Mellon was not the “Client” conflicts with his conclusion that Mellon Financial’s custodial entities at the time of contracting were the “Client.”

[28] I reject this argument because, as I explain later in these reasons, Mellon Financial’s custodial entities at the time of contracting were not the “Client.” This conclusion eliminates any potential contradiction between CIBC Mellon’s contractual rights and the rights of other Mellon Financial custodial entities.

[29] This argument also fails because the trial judge reasonably distinguished CIBC Mellon from Mellon Financial’s other custodial entities at the time of contracting. Under the related contracts principle, when parties make multiple contracts as part of an overall transaction, they must be read together and harmonized to give effect to the parties’ intentions, prevent clashes, and avoid rendering any of them meaningless: Ottawa (City) v. ClubLink Corporation ULC, 2021 ONCA 847, 159 O.R. (3d) 255, at para. 54, leave to appeal refused, [2022] S.C.C.A. No. 17; Samson Cree Nation v. O'Reilly & Associés, 2014 ABCA 268, 375 D.L.R. (4th) 663, at para. 82. Consistent with this principle, the trial judge looked to the CIBC Mellon Agreement to interpret the Mellon Trust Agreement because Mellon Financial and its affiliate, CIBC Mellon, made these Agreements with SS&C in the same year to meet their data needs. He also found that treating CIBC Mellon as the “Client” under the Mellon Trust Agreement would render the CIBC Mellon Agreement redundant. This finding justified treating CIBC Mellon differently from Mellon Financial’s other custodial entities, which did not have their own data purchase contracts with SS&C.

[30] I reject BNY’s challenge to this finding. BNY submits that the CIBC Mellon Agreement was not redundant because it provided different data for different purposes. But the trial judge considered this evidence and found that it supported his conclusion that the parties intended for CIBC Mellon to be covered by its own contract that provided distinct services and data from the Mellon Trust Agreement. BNY’s argument that the trial judge should have drawn a different conclusion from this evidence does not establish a palpable and overriding error: 3091‑5177 Québec inc. (Éconolodge Aéroport) v. Lombard General Insurance Co. of Canada, 2018 SCC 43, [2018] 3 S.C.R. 8, at para. 35.

[31] BNY next argues that the trial judge overlooked CIBC Mellon’s need to access the data when interpreting the Mellon Trust Agreement provision authorizing the “Client” to share data with its “customers.” This argument fails because the trial judge rejected its premise that CIBC Mellon required a separate contract to access the data it needed. He did not need to rehash his finding on that point when interpreting the customer data sharing provision. Nor does BNY challenge his conclusion that the term “customer” in that provision refers to arm’s length end users rather than other legal entities within the Mellon Trust brand.

(iii) The Contract Barred Redistributing Data to BNY Affiliates

[32] BNY next submits that the contract authorized affiliates that Mellon Financial did not own when the contract was entered to access data. I disagree and would reject BNY’s largely fact-based challenge to the trial judge’s finding that those affiliates were not entitled to access the relevant data.

[33] First, BNY argues that the trial judge failed to consider whether SS&C should have known that Mellon Trust was a brand that included numerous BNY subsidiaries. I disagree. BNY is correct that the surrounding circumstances include background facts known to one party that its counterparty should have but did not know: Sattva Capital Corp. v. Creston Moly Corp., 2014 SCC 53, [2014] 2 S.C.R. 633, at para. 58. But the trial judge considered whether SS&C should have known this fact and found that it would be “asking too much” of SS&C’s sales personnel to know it. He did not err by also considering SS&C’s lack of subjective knowledge because the surrounding circumstances also include background facts that both parties subjectively knew: Sattva, at para. 58.

[34] Second, BNY argues that the trial judge’s conclusion was commercially unreasonable because the affiliates of Mellon Trust also needed data to service customers. This submission fails because it takes a one-sided approach to commercial reasonableness, which is an objective interpretative tool that requires considering both parties’ perspectives: Harvey Kalles Realty Inc v. BSAR (Eglinton) LP, 2021 ONCA 426, at para. 6, leave to appeal refused, [2021] S.C.C.A. No. 314; Atos IT Solutions v. Sapient Canada Inc., 2018 ONCA 374, 140 O.R. (3d) 321, at para. 60, leave to appeal refused, [2018] S.C.C.A. No. 218. As the trial judge found, while paying the same Mellon Trust Agreement rates would have been convenient for BNY, this interpretation was unreasonable from SS&C’s perspective: Kentucky Fried Chicken Canada v. Scott’s Food Services Inc. (1998), 1998 CanLII 4427 (ON CA), 114 O.A.C. 357 (C.A.), at para. 27. It would have permitted BNY to “cannibalize” SS&C’s business by selling data to an unknown and ever-shifting group of after-acquired entities.

[35] Third, BNY turns to the Mellon Trust Agreement’s successorship clause, which provides that the contract binds and grants rights to each party’s “successors”. BNY argues that this clause demonstrates that the number of custodial entities that the term “Client” included was intended to change over time because mergers could add additional subsidiaries to the successor entity.

[36] I disagree because the premise of this argument, that subsidiaries automatically succeed to rights that their parent company succeeds to, is wrong. The term “successors” in the clause presumptively carries its legal meaning and thus refers to corporate successors, which means corporations that assume the rights and liabilities of other corporations through merger, amalgamation, or other means: Heritage Capital, at para. 47; Montreal Trust Co. of Canada v. Birmingham Lodge Ltd. (1995), 1995 CanLII 438 (ON CA), 24 O.R. (3d) 97 (C.A.), at p. 104, citing National Trust Co. v. Mead, 1990 CanLII 73 (SCC), [1990] 2 S.C.R. 410, at p. 423.[^1] Because companies have separate legal personalities, a parent company does not become a successor simply because its subsidiary succeeds another company. Doing so would require piercing the corporate veil, which BNY has not alleged: Malik (18A application of Talisman Energy Inc.), 2007 BCSC 739, at paras. 25-29. Likewise, a merger between two parent companies, without more, does not cause one parent company’s subsidiaries to succeed the other parent company or its subsidiaries.

[37] These principles are fatal to BNY’s argument. Successorship is a factual question that BNY, as the party relying on the successorship clause, bore the burden to prove: Genstar Development Partnership v. The Roman Catholic Episcopal Corporation of the Diocese of Hamilton in Ontario, 2019 ONCA 506, 3 R.P.R. (6th) 1, at paras. 17-18, 21. BNY has not met that burden. While BNY succeeded Mellon Financial when the 2007 merger closed, this did not make the former Bank of New York affiliates that BNY acquired successors to Mellon Financial and its rights under the Mellon Trust Agreement. And even if the custodial entities that Mellon Financial owned in 1999 had rights under the Mellon Trust Agreement, BNY has not shown that those former Bank of New York affiliates succeeded them either. Having chosen to obtain the benefits of conducting its business through numerous subsidiaries, BNY must also bear the corresponding burdens of their separate legal personality: Brunette v. Legault Joly Thiffault, s.e.n.c.r.l., 2018 SCC 55, [2018] 3 S.C.R. 481, at para. 39.

(iv) The Contract Barred Mellon Financial Affiliates from Accessing Data

[38] This brings me to SS&C’s argument that the Mellon Trust Agreement only authorizes BNY itself, not also the custodial entities that Mellon Financial owned at the time it entered into the contract, to access the data. I agree.

[39] As I mentioned, the trial judge correctly instructed himself to interpret the contract’s text in light of its commercial purpose and examine the parties’ course of performance to resolve any ambiguity. If these three factors all favour a given interpretation, then the court should adopt it unless other countervailing principles justify a different result: Prism, at paras. 27-39; Hanley Park, at paras. 61-65.

[40] The trial judge’s findings concerning these factors show that the Mellon Trust Agreement only authorized a single legal entity, BNY, to access data. Beginning with the text, he found that the term “Client,” which the contract defined as “Mellon Trust,” referred to a single legal entity, not a brand or corporate group containing multiple legal entities. He next considered the commercial purpose and found that, because data sellers try to control buyers’ distribution of data to protect their business, it was commercially reasonable to interpret the contract as barring the named data buyer from sharing the data with other entities in its corporate group. He found that this interpretation was consistent with the contract’s text, which provided that the “Client” could not share the data with any parties other than its customers. Finally, he found that the parties’ course of performance showed that only BNY and its predecessor Mellon Financial, not a brand containing multiple legal entities, was the contracting party.

[41] Because these findings all pointed in the same direction, applying the law of contract interpretation to them should have led the trial judge to conclude that the contract only authorized BNY to receive data. But instead of drawing this conclusion, he determined that the contract authorized not only BNY but also Mellon Financial’s custodial entities at the time of contracting to access data. The trial judge candidly acknowledged that his liability ruling did not explain how his findings justified that determination. The trial judge tried to supply that missing justification in his amending ruling by reasoning that it would be commercially unreasonable to restrict the definition of “Client” to BNY. But this justification cannot stand because it used commercial reasonableness to erase the restrictions that his findings and the legal principles concerning the text, surrounding circumstances, and course of performance imposed, and is based on an obvious factual error that Mellon Financial did not need any data itself. These extricable legal errors and palpable and overriding factual error require our intervention: Fuller v. Aphria Inc., 2020 ONCA 403, 4 B.L.R. (6th) 161, at paras. 49-50.

[42] First, the trial judge used BNY’s individualized version of the background facts to interpret the contract. Sattva held that courts can only consider background facts within the parties’ common knowledge, meaning those facts that both parties knew or should have known at the time of entering into the contract, as surrounding circumstances: at paras. 58-60; see also Earthco, at para. 65. This rule prohibits considering a party’s subjective intention (NexJ Systems Inc. (Re), 2023 ONCA 451, 43 B.L.R. (6th) 224, at para. 10), as well as other facts that one party knew but that its counterparty neither knew nor should have known: Taggart v. McLay, 1998 CanLII 5541 (B.C.C.A.), at para. 7; Geoff R. Hall, Canadian Contractual Interpretation Law, 4th ed. (Markham: LexisNexis, 2020), at p. 34. The trial judge violated this rule. He found that Securities Valuation neither knew nor should have known what Mellon Financial knew, namely that Mellon Trust was a brand made up of Mellon Financial and its many subsidiaries. But he then relied on those same individualized background facts to find that it would be commercially unreasonable to prohibit Mellon Financial and its successor, BNY, from redistributing data to those subsidiaries.

[43] The trial judge’s conclusion that courts can consider individualized background facts to establish commercial purpose is incorrect. Rather, Sattva’s rule against individualized background facts applies globally, including when parties seek to use them to advance commercial reasonableness arguments: Sattva, at para. 58; Geophysical Service Incorporated v. Plains Midstream Canada ULC, 2023 ABCA 277, at para. 10. This makes sense because, like background facts, commercial reasonableness must be assessed objectively and from both parties’ perspective, not from one party’s subjective perspective: Harvey Kalles, at para. 6; Atos, at para. 60.

[44] The trial judge was also incorrect to suggest that not considering these individualized background facts is unfair. Rather, Sattva’s rule against individualized background facts is fair to both parties because it prevents either party from relying on undisclosed background facts that the other party had no reason to know. This ensures that courts determine the meaning of contracts based on both parties’ shared, objective understanding of the background facts. In this case, for instance, BNY did not tell SS&C that it sought to buy data access for a brand of many subsidiaries. This uncommunicated material information could have led SS&C to demand higher fees because, as the trial judge found, data sellers seek to protect their business by restricting their clients’ redistribution of data.

[45] Second, even if I considered these subjective background facts, I would still disagree with the trial judge that treating Mellon Financial as the “Client” is clearly commercially unreasonable. BNY has not met this high standard. See Simex Inc. v. Imax Corporation (2005), 2005 CanLII 46629 (ON CA), 206 O.A.C. 3 (C.A.), at paras. 23, 52-54.

[46] To begin with, I disagree with the trial judge that Mellon Financial itself did not need data. The trial judge found that Mellon Financial was a holding company, meaning that it was formed to control its subsidiaries in the Mellon Trust brand through stock ownership and supervising their management. See Manitoba Netset Ltd. v. High Speed Mole Inc., 2013 MBQB 205, 296 Man. R. (2d) 162, at para. 20, aff’d, 2015 MBQB 60. But the trial judge’s inference, that because Mellon Financial was a holding company, its subsidiaries exclusively serviced customers and it did not need data itself, simply does not follow. As BNY concedes, holding companies are not necessarily shell companies. Rather, Mellon Financial had its own market data employees, including BNY’s two lead witnesses in this case. One of those employees, the same senior executive who signed the May 11, 2007 letter, deposed that Mellon Financial itself, not merely its affiliates, provided services under the Mellon Trust brand name. And even the witness whose evidence BNY relies on to defend the trial judge’s finding deposed that BNY provides some services directly to external customers. This evidence, which the trial judge did not refer to, establishes that he made an obvious, outcome-determinative error that undermines the factual basis of his commercial reasonableness analysis.

[47] Further, this is not a case where, under SS&C’s interpretation, Mellon Financial and BNY would be required to pay it for nothing or next to nothing. Commercial reasonableness considerations would carry more weight if, for instance, SS&C was not required to provide any service in exchange for payment, or, alternatively, if it was providing a service known not to work or that could not work because it was incomplete. See Anderson v. Bell Mobility Inc., 2015 NWTCA 3, 379 D.L.R. (4th) 682, at paras. 29, 64-65, leave to appeal refused, [2015] S.C.C.A. No. 92. But no one disputes that the data SS&C delivers is a real, valuable service that SS&C expended time, money, and labour to provide. Rather, the issue here is merely which entities can access those services. As I have explained, answering that question requires considering both parties’ perspectives, not merely Mellon Financial’s preference to obtain data for its entire corporate group.

[48] Third, the trial judge overlooked the focal point of contract interpretation, the text, and the constraints it imposes on commercial reasonableness. Courts can legitimately use commercial reasonableness as a tool to interpret the text because parties are unlikely to have intended to strike bargains that make no business sense: Earthco, at paras. 64-65, 72; Bhasin v. Hrynew, 2014 SCC 71, [2014] 3 S.C.R. 494, at para. 45. But the text, in turn, helps courts determine which interpretations are commercially reasonable and prevents using this factor to rewrite contracts that, in hindsight, appear improvident or inconvenient: Atos, at paras. 60-61; Thunder Bay (City) v. Canadian National Railway Company, 2018 ONCA 517, 424 D.L.R. (4th) 588, at para. 52, leave to appeal refused, [2018] S.C.C.A. No. 358; Jardine v. General Hydrogen Corp., 2007 BCSC 119, 68 B.C.L.R. (4th) 162, at para. 37. Thus, commercial reasonableness is not always determinative (Canadian Premier Holdings Ltd. v. Winterthur Canada Financial Corp. (2000), 2000 CanLII 5724 (ON CA), 132 O.A.C. 172 (C.A.), at para. 13), nor is it a licence to depart from the text to engineer whatever result seems “fair” to the court: Kilitzoglou v. Curé, 2018 ONCA 891, 143 O.R. (3d) 385, at paras. 57-59. Further, because parties are presumed to intend the legal consequences of the words they chose, it is not commercially unreasonable to hold them to those words’ meaning: John E. Dodge Holdings Ltd. v. 805062 Ontario Ltd. (2003), 2003 CanLII 52131 (ON CA), 63 O.R. (3d) 304 (C.A.), at paras. 14-16, 35-36, leave to appeal refused, [2003] S.C.C.A. No. 145; Heritage Capital, at para. 47.

[49] Because the trial judge did not consider these textual constraints, he misinterpreted the contract. He overlooked his findings that the text showed that “Client” referred to a single legal entity that, unlike an amorphous brand comprising numerous entities, had a specific address and could become insolvent and assign its assets. The text’s reference to “Client,” “either party,” and “the other party” in the singular instead of the plural confirmed this conclusion: 202135 Ontario Inc. v. Northbridge General Insurance Corporation, 2022 ONCA 304, 29 B.L.R. (6th) 183, at paras. 26-31. By overlooking these textual restrictions, he overstepped the legitimate use of commercial reasonableness to inform the interpretation of the text and improperly used it to rewrite the text.

[50] Fourth, the trial judge failed to square his conclusion with his course of performance findings. While the law approaches this type of evidence cautiously, it can attract significant weight if it is unequivocal, consistent, mutual, close to the time of contracting, is based on the actions of the contracting companies’ chief representatives for a given matter, and is not self-serving: Shewchuk v. Blackmont Capital Inc., 2016 ONCA 912, 404 D.L.R. (4th) 512, at paras. 45, 52-56; Prism, at paras. 37-39; Riddell Kurczaba Architecture Engineering Interior Design Ltd v. University of Calgary, 2019 ABCA 195, 89 B.L.R. (5th) 175, at para. 11; British Columbia Hydro & Power Authority v. Cominco Ltd., 1985 CarswellBC 2783 (S.C.), at para. 65, aff’d, (1989), 1989 CanLII 2667 (BC CA), 34 B.C.L.R. (2d) 60 (C.A.). In these circumstances, course of performance evidence can be the next-best-evidence of the parties’ intent at the time of contracting after the text, just as parties’ subsequent conduct is sometimes the best evidence of whether they objectively intended to form a contract: Geoff R. Hall, Canadian Contractual Interpretation Law, 1st ed. (Markham: LexisNexis, 2007), at p. 70; Angus v. CDRW Holdings Ltd., 2023 BCCA 330, 53 R.P.R. (6th) 173, at para. 51.

[51] The trial judge failed to consider these principles because he misapprehended his course of performance findings. He concluded that they only showed that the contract did not authorize data sharing with entities that BNY acquired after 1999. But he overlooked his finding that the parties’ chief representatives agreed in a 2007 co-signed letter that Mellon Financial and, subsequently, its successor BNY itself, rather than a brand containing many entities, was SS&C’s contracting party. As the trial judge found, this conduct was consistent with the parties’ 2003 decision to expand the definition of “Client” in the CIBC Mellon Agreement to add an affiliated entity, which showed that amending this definition was necessary to entitle affiliates to access data.

[52] This overlooked finding concerning the 2007 letter resolves any ambiguity and confirms that the Mellon Trust Agreement only authorized BNY to access data. That letter is persuasive evidence of the parties’ intent because both companies’ chief representatives signed it reasonably close to the time of contracting, it unequivocally communicated that BNY and SS&C were the sole contracting parties, it was not self-serving, and it was consistent with the amendment to the CIBC Mellon Agreement.

(b) BNY’s Other Arguments Fail

[53] BNY also argues that, even if it breached the contract, SS&C’s claims should be dismissed as limitations-barred and because the trial judge lacked authority to change his unentered judgment. I disagree. BNY fails to identify any legal errors and has not met the high bar to interfere with the trial judge’s factual findings and discretionary decision to change his judgment.

(i) BNY Failed to Prove Its Limitations Defence

[54] Section 4 of the Limitations Act, 2002, S.O. 2002, c. 24, Sch. B (the “Act”) establishes a default limitation period that bars plaintiffs from commencing claims more than two years after they discover them. As subsections 5(1)(a)-(iv) of the Act provide, discovery requires three elements. They are as follows: (1) knowing that damage has occurred; (2) knowing that the defendant’s act or omission caused or contributed to it; and 3) knowing that a legal proceeding would be an appropriate means to remedy it.

[55] Subsections 5(1)(a) and (b) of the Act establish that the plaintiff discovers a claim, and, thus, section 4’s two-year clock begins to run, when either the plaintiff or a reasonable person with its abilities and in its circumstances becomes aware of the above three elements. Determining the plaintiff’s knowledge is a subjective test, while determining the reasonable person’s knowledge is a modified objective test: Apotex Inc. v. Nordion (Canada) Inc., 2019 ONCA 23, 431 D.L.R. (4th) 262, at paras. 75-81. The trial court’s determination concerning whether a limitation period has expired and the factual findings on which it is based are both reviewed for palpable and overriding error, absent an extricable error of law or principle: AssessNet Inc. v. Taylor Leibow Inc., 2023 ONCA 577, 168 O.R. (3d) 276, at para. 30.

[56] The expiry of a limitations period is an affirmative defence that the defendant must prove. Subsection 5(2) of the Act establishes a limited exception to this rule by creating a presumption that the plaintiff discovered the claim on the date the act or omission on which the claim is based took place: AssessNet, at paras. 34-35. But the threshold to rebut this presumption is low: Presley v. Van Dusen, 2019 ONCA 66, 144 O.R. (3d) 305, at para. 24. The plaintiff need only show that it discovered the claim on a different date: Apotex, at para. 83. If the court so finds, then the burden shifts back to the defendant to prove that the claim is limitations-barred: AssessNet, at para. 35.

[57] BNY argues that the trial judge erred in law by (1) failing to apply subsection 5(2)’s presumption of knowledge, (2) failing to consider when a reasonable person in SS&C’s position would have discovered the claim, (3) applying the wrong test to impute the knowledge of SS&C’s employees to SS&C, and (4) failing to give effect to his alleged finding that SS&C learned of BNY’s breach in 2009. BNY also submits that the trial judge made factual errors in finding that SS&C did not discover BNY’s breaches until Fall 2016. I disagree. In my respectful view, the trial judge reasonably concluded that BNY failed to prove its limitations defence.

[58] First, the trial judge’s findings displace subsection 5(2)’s presumption. SS&C’s claims are based on BNY’s breaches of the Mellon Trust Agreement by redistributing data to CIBC Mellon and other unauthorized entities. Because BNY began committing these breaches in 1999, subsection 5(2) triggered a presumption that SS&C discovered its claims then. But the trial judge found that SS&C only discovered the breaches in Fall 2016. This finding displaced the presumption by establishing that SS&C discovered the claims on a different date from when the breaches began and shifted the burden of proof back to BNY: Apotex, at para. 83; AssessNet, at para. 35. It also makes clear that the trial judge concluded that SS&C had rebutted the presumption, even though he did not expressly reference it: Kumarasamy v. Western Life Assurance Company, 2021 ONCA 849, 464 D.L.R. (4th) 700, at para. 38, leave to appeal refused, [2022] S.C.C.A. No. 12.

[59] I disagree with BNY’s counterarguments. BNY first argues that the trial judge made the same error as in Clarke v. Sun Life Assurance of Canada, 2020 ONCA 11, 149 O.R. (3d) 433, by failing to make findings concerning the presumption. But Clarke is distinguishable. That case required those findings concerning the presumption because, unlike here, the motion judge did not find that the plaintiff discovered the claim on a date other than when the breach began: at paras. 23-25. Because the trial judge made that finding in this case, the presumption did not apply, and the trial judge did not need to make any further findings concerning it. BNY next submits that the trial judge did not hold SS&C to its burden to rebut the presumption because he found that it discovered the claim in 2016 when considering BNY’s equitable defences, which BNY bore the burden to prove. I disagree because he first made this finding when considering SS&C’s breach of contract claim, which SS&C bore the burden to prove: Porter v. Metropolitan Life Insurance Co. (1985), 1985 CanLII 5744 (NS CA), 70 N.S.R. (2d) 248 (C.A.), at para. 11.

[60] Second, I decline to consider BNY’s submission that SS&C should have discovered its claims sooner. This is a new issue because, as BNY concedes, it only raised SS&C’s subjective knowledge under subsection 5(1)(a) at trial, not what SS&C should have known under subsection 5(1)(b)’s modified objective test. Appellate courts generally decline to consider new issues if they require making findings of fact or mixed fact and law, especially if the appellant has not explained its failure to raise those issues at trial: Kaiman v. Graham, 2009 ONCA 77, 245 O.A.C. 130, at paras. 18, 21-22. That is the case here. Because knowledge under this modified objective test is a question of mixed fact and law (Espartel Investments Limited v. Metropolitan Toronto Condominium Corporation No. 993, 2024 ONCA 18, 492 D.L.R. (4th) 659, at para. 16 & n. 5), resolving this issue would require evidence and findings concerning what steps SS&C should have taken to detect BNY’s breaches: Service Mold + Aerospace Inc. v. Khalaf, 2019 ONCA 369, 146 O.R. (3d) 135, at para. 31. SS&C lacked the opportunity to call that evidence and the trial judge did not make those findings because BNY, for unexplained reasons, did not raise this issue at trial. It would not be fair to SS&C to consider this argument on appeal, just as it is generally unfair to allow parties to raise unpled limitations defences on appeal: Streisfield v. Goodman (2004), 8 E.T.R. (3d) 130 (Ont. C.A.), at para. 12, leave to appeal refused, [2004] S.C.C.A. No. 356.[^2]

[61] BNY submits that, because it raised SS&C’s subjective knowledge at trial, the trial judge was also required to address what SS&C should have known as part of a single integrated assessment. I disagree. BNY was responsible for determining whether to rely on its limitations defence (Abrahamovitz v. Berens, 2018 ONCA 252, 140 O.R. (3d) 161, at para. 33), and which of its elements to put in play. By making those choices, BNY defined the issues for the trial judge to resolve. The trial judge was entitled to decide the case on that basis and was not obligated to consider elements of the limitations defence that BNY did not put in issue, just as he was not required to raise that defence on BNY’s behalf: McDiarmid Lumber Ltd v. Letandre, 2003 MBQB 99, 174 Man. R. (2d) 104, at paras. 13-16.

[62] Third, the trial judge applied the correct legal test to determine whether to impute the knowledge of SS&C’s employees to SS&C. Rocco v. Northwestern National Insurance Co. (1929), 1929 CanLII 407 (ON CA), 64 O.L.R. 559 (C.A.), is the governing case on this question: Bank Leu AG v. Gaming Lottery Corp. (2003), 2003 CanLII 28360 (ON CA), 231 D.L.R. (4th) 251 (Ont. C.A.), at paras. 48-49. The trial judge stated, citing Rocco, that information known to non-management employees is not automatically imputed to the company. This statement accurately summarized Rocco’s holding that those employees’ knowledge is only imputed to the company if they “would be likely to communicate [it] to those in charge of the affairs of the company”: at p. 562; see also Cataraqui Cemetery Company v. Cyr, 2017 ONSC 5819, at para. 219.

[63] I do not agree with BNY’s argument that the trial judge erred by stating that a corporation’s “directing mind” must know the material facts in question. These comments merely tried to paraphrase Rocco’s test that non-management employees must be likely to communicate information to corporate directors or managers. The trial judge did not mean that those employees’ knowledge could never be imputed to the corporation because they are not directors or managers.

[64] Fourth, I reject BNY’s argument that the trial judge failed to give effect to his alleged finding that SS&C learned of the breaches in 2009. The premise of this argument, that the trial judge made this finding, is wrong. At trial, BNY argued that a 2009 email discussion between a SS&C client support representative and his superior, whom BNY claims was in a senior management role, showed that SS&C knew that BNY was sharing data with CIBC Mellon. SS&C argued that the discussion did not show this and concerned a different topic. The trial judge found that he could not determine which interpretation was correct. This finding shows that BNY failed to meet its burden to prove its limitations defence.

[65] Because this finding is outcome-determinative, I need not consider the trial judge’s alternative conclusion that, assuming without deciding that BNY’s interpretation of the email was correct, that merely showed that SS&C was granting BNY a “limited indulgence” concerning some information it shared with CIBC Mellon. While the trial judge repeated these remarks in his damages ruling, this does not show that, contrary to his liability ruling, he found that SS&C knew that BNY was redistributing data to CIBC Mellon. To the contrary, he made clear that this was an alternative conclusion rather than a finding, and reiterated his primary conclusion that BNY had failed to prove “that SS & C knew that [BNY] was sharing data with CIBC Mellon.”

[66] Because BNY has failed to establish any legal error, I would defer to the trial judge’s finding that SS&C did not discover BNY’s breaches until Fall 2016 and, thus, that its claims are not limitations-barred. BNY’s challenges to this finding of mixed fact and law fail to establish any palpable and overriding error.

[67] To begin with, BNY argues that SS&C knew it was sharing data because, in August 2012, BNY told SS&C to send Mellon Trust Agreement invoices to CIBC Mellon. But the trial judge reasonably declined to draw this inference because BNY never told SS&C that it made the billing change because CIBC Mellon was using the data. Next, BNY submits that it forwarded CIBC Mellon’s complaints about the data BNY shared with it to SS&C technical support staff between 2012 and 2015. But the trial judge, who applied the correct legal test from Rocco, reasonably found that those employees’ knowledge could not be imputed to SS&C because they were not responsible for SS&C’s contracting and, thus, were unlikely to communicate this information to the SS&C managers responsible for that matter. His finding on this factual question is entitled to deference: Concentra Financial Services Assn. v. Rawling, 2017 ONCA 348, at para. 10. BNY finally points to SS&C’s 2002 agreement to add Scotia McLeod pricing data, which Mellon Financial wanted to serve CIBC Mellon’s needs, to the Mellon Trust Agreement. This argument fails because the trial judge found that neither Mellon Financial nor BNY told SS&C that they intended to redistribute data to CIBC Mellon.

(ii) The Trial Judge Properly Changed His Unentered Judgment

[68] BNY next argues that the trial judge lacked authority to change his judgment to find that it breached the Mellon Trust Agreement by sharing data with CIBC Mellon. I disagree. The trial judge had broad discretion to reconsider and change his judgment before its entry. He properly exercised that discretion to correct his initial conclusion that CIBC Mellon breached its separate contract with SS&C, which overlooked SS&C’s actual breach of contract claim against BNY and was an outcome-determinative error.

[69] Trial courts sometimes make mistakes in their orders.[^3] While parties can appeal these mistakes and appellate courts can correct them, trial courts also have a time-limited opportunity, between making an order and entering it, to change it to rectify those mistakes. Montague v. Bank of Nova Scotia, following longstanding precedent, affirmed that courts have “very broad discretion” to make such changes to “better serve the ends of justice,” either on their own initiative or on a party’s request: (2004), 2004 CanLII 27211 (ON CA), 69 O.R. (3d) 87 (C.A.), at paras. 34, 36, leave to appeal refused, [2004] S.C.C.A. No. 79, citing Holmes Foundry Ltd. v. Village of Point Edward, 1963 CanLII 197 (ON CA), [1963] 2 O.R. 404 (C.A.), at p. 407. This discretion remains broad until the order is entered. Then, trial courts can only amend it if authorized by statute, to correct a drafting slip-up or error in expressing their manifest intention, or if the matter has not been heard on its merits: Canadian Broadcasting Corp. v. Manitoba, 2021 SCC 33, [2021] 2 S.C.R. 785, at para. 33; Chandler v. Alberta Association of Architects, 1989 CanLII 41 (SCC), [1989] 2 S.C.R. 848, at p. 860.

[70] Montague permits trial judges to change their orders if they overlook outcome-determinative matters: at paras. 21, 38. By focusing on whether trial judges have overlooked a point of law, argument, or fact, this test discourages litigants from rearguing points the trial judge considered and rejected or raising new arguments or facts that they could have raised earlier: Meridian Credit Union Ltd. v. Baig, 2016 ONCA 942, 6 C.P.C. (8th) 33, at para. 8, leave to appeal refused, [2016] S.C.C.A. No. 173; Mujagic v. Kamps, 2015 ONCA 360, 125 O.R. (3d) 715, at para. 13, leave to appeal refused, [2015] S.C.C.A. No. 330. Likewise, because the overlooked point must be outcome-determinative, this test deters litigants from focusing on lesser errors that would not change the result: First Elgin Mills Developments Inc. v. Romandale Farms Ltd., 2015 ONCA 54, 381 D.L.R. (4th) 114, at para. 9, leave to appeal refused, [2014] S.C.C.A. No. 442.

[71] Montague also requires trial courts to clearly explain changes to orders, especially important changes: at para. 40; see also 1711811 Ontario Ltd. (AdLine) v. Buckley Insurance Brokers Ltd., 2014 ONCA 125, 371 D.L.R. (4th) 643, at para. 72. If trial courts do so, then their discretionary decision to change their orders is owed deference absent a legal error, a palpable and overriding factual error, or an unreasonable exercise of discretion: Montague, at para. 39; Canada (Transportation Safety Board) v. Carroll-Byrne, 2022 SCC 48, 475 D.L.R. (4th) 274, at para. 41.

[72] Montague has stood the test of time. It accounts for the reality that busy trial judges sometimes make mistakes and overlook outcome-determinative points by giving them an opportunity to correct those mistakes before their orders are entered. I agree with the foundational English precedent Montague followed, Harrison v. Harrison, that denying judges this opportunity is “unworkable”: [1955] 1 Ch. 260 (C.A.), at p. 276; see Montague, at para. 34. Doing so would undermine access to justice and the justice system’s efficient operation by requiring needless appeals and consequent delays and legal costs to remedy overlooked points that trial judges could promptly correct: Gore Mutual Insurance Co. v. 1443249 Ontario Ltd. (2004), 2004 CanLII 27736 (ON SC), 70 O.R. (3d) 404 (S.C.), at para. 8, per Karakatsanis J. (as she then was), aff’d, (2004), 2004 CanLII 43772 (ON CA), 75 O.R. (3d) 477 (C.A.). At the same time, focusing the inquiry on whether trial judges have overlooked an outcome-determinative matter appropriately accounts for litigants’ interest in certainty and finality: ATU, Local 615 v. Saskatoon (City), 2021 SKCA 93, 86 C.L.R.B.R. (3d) 1, at para. 83. Finally, Montague’s clear reasons requirement permits meaningful appellate review and helps negate any suspicions that judges are making changes for improper reasons: Montague, at para. 40; Buckley, at para. 73. This respects the principle that discretionary decisions require adequate reasons: Penate v. Martoglio, 2024 ONCA 166, at para. 21.

[73] BNY argues that a stricter and narrower standard applies. It submits that courts can only change unentered orders to fix technical errors or avoid a miscarriage of justice. In its view, technical errors are restricted to errors that are either mathematical or obvious and inadvertent, and miscarriages of justices require threats to the integrity of the litigation process such as failing to review a party’s materials. It contends that outcome-determinative legal or factual errors do not meet this high threshold, which it maintains this court adopted in FS Partnership/UPI Energy FS v. Mr. Refuel Inc., 2022 ONCA 612.

[74] I disagree because BNY’s argument clashes with precedent and unnecessarily seeks to replace a test that has long worked well with an unworkable stricter standard. BNY attempts, in effect, to transplant the strict restrictions on changing entered orders into the test for changing unentered ones. This contravenes the Supreme Court of Canada’s direction that those strict restrictions only apply after orders are entered: Canadian Broadcasting Corp., at para. 35; Chandler, at p. 860.[^4] It contradicts Montague’s holding that courts have wide discretion to amend their orders before entry to further justice: at paras. 34, 36. It is also inconsistent with Montague’s affirmance, without finding a threat to the integrity of the litigation process, of the trial judge’s decision in that case to change her judgment to account for outcome-determinative legal and factual points that she overlooked: at paras. 21, 34, 38. Moreover, the obviousness test it introduces is unworkable and inconsistent with the rule of law, which disfavours requiring parties to accept erroneous decisions simply because the error is not obvious enough: Dunsmuir v. New Brunswick, 2008 SCC 9, [2008] 1 S.C.R. 190, at paras. 40-42. Finally, this stricter standard is unnecessary because Ontario trial courts can be trusted to distinguish meritorious reconsideration motions from meritless ones, which adverse cost consequences will help deter.

[75] BNY’s argument that FS Partnership established a stricter test is wrong. Rather, consistent with Montague’s deferential standard of review, FS Partnership simply deferred to the motion judge’s decision not to change an order: at paras. 26-27. FS Partnership, following the Divisional Court’s Brown decision, also referenced a “miscarriage of justice” test: FS Partnership, at para. 26, citing Brown, at para. 20. But because Brown recognized Montague’s binding force and followed its clear reasons rule (at paras. 20, 24), I read this reference to a “miscarriage of justice” test as paraphrasing Montague’s flexible “better serve the ends of justice” test rather than adopting a stricter standard like the undermine the integrity of the litigation process or obviousness tests that BNY proposes: Montague, at para. 36. Thus, the trial court decisions adopting those stricter tests that BNY cites are no longer good law on this point: see, e.g., Dnaagdawenmag Binnoojiiyag Child and Family Services v. V.S. et al, 2021 ONSC 5562, at paras. 26-28, 36, and Scott, Pichelli & Easter Ltd. v. Dupont Developments Ltd. et al., 2019 ONSC 6789, 59 C.P.C. (8th) 187, at para. 10-13, 20, rev’d, 2021 ONSC 6579, 157 O.R. (3d) 772 (Div. Ct.), aff’d, 2022 ONCA 757, 475 D.L.R. (4th) 364.

[76] FS Partnership’s comment that courts’ authority to change an order is “narrow” also does not change Montague’s test. This merely references Brown’s statement that re-argument is inappropriate, subject to “narrow” exceptions: FS Partnership, at para. 27; Brown, at para. 19. This is consistent with Montague, which also does not favour re-argument but permits reconsideration.

[77] Applying Montague’s test, I would defer to the trial judge’s well-reasoned discretionary decision to change his judgment before its entry. As he explained, he overlooked SS&C’s actual claim that BNY breached the Mellon Trust Agreement by sharing data with CIBC Mellon and instead adjudicated an unpled, never-advanced claim that CIBC Mellon breached the CIBC Mellon Agreement. This mistake was outcome-determinative because it impacted whether BNY or CIBC Mellon bears liability and that liability’s contractual basis. The trial judge reasonably determined that, to correct his mistake, he had to address SS&C’s actual claim that BNY breached the Mellon Trust Agreement. His decision to do so furthered justice as Montague requires by ensuring that the parties could know which entity was liable under which contract before the damages trial.

[78] I reject BNY’s arguments that the trial judge lacked discretion to change his judgment. BNY’s submission that the change did not remedy a technical error or a threat to the integrity of the litigation process fails because Montague requires neither. SS&C also properly sought reconsideration to address its overlooked claim rather than seeking re-argument as BNY wrongly contends.

V. DISPOSITION

[79] I would dismiss BNY’s appeal on liability and allow SS&C’s corresponding cross-appeal. I would accordingly vary paragraph 1 of the judgment by adding a declaration that no entities other than Mellon Financial and BNY were entitled to access the data SS&C provided under the Mellon Trust Agreement.

“M. Tulloch C.J.O.”

Hourigan J.A.:

PART B: DAMAGES

I. introduction

[80] I concur entirely with my colleague Tulloch C.J.O.’s reasons and disposition on the liability portion of the appeal and cross-appeal. As Tulloch C.J.O. indicated, however, the appeal and cross-appeal also require addressing the trial judge’s damages award: 2023 ONSC 4083. These reasons address that award.

[81] When an appellate court reviews a damages award it does not require perfection. This is for the practical reason that damage awards often involve the exercise of discretion and do not always admit to a single correct figure; they are more often art than science. Even in the context of commercial disputes, where an award should presumably be subject to more mathematical rigor, trial judges are sometimes placed in a position where it is difficult to quantify a damages award with precision because the evidentiary record contains gaps. In such circumstances, what is required of a trial judge is to make a damages award that is rooted in the evidentiary record and free of legal error and palpable and overriding factual error or error of mixed fact and law. A trial judge who produces such an award will not be second guessed by an appellate court on the basis that it would calculate damages differently.

[82] In the case at bar, the litigants are highly sophisticated corporate entities who have engaged in hard fought, high-stakes litigation for over six years. On the issue of damages, it is difficult to envision two more diametrically opposed positions. SS&C submits that it has suffered approximately $890 million USD in damages, while BNY takes the position that SS&C is entitled to no damages. Given the history of this litigation, this divergence is not surprising. What is surprising is that, despite the length of this litigation and the sophistication of the parties, there are significant evidentiary gaps regarding damages. The existence of these gaps was part of the reason why the trial judge rejected the damages theories advanced by the parties and devised his own theory of damages.

[83] On appeal, both sides remain rigid in their positions. According to them, this case is worth either over a billion Canadian dollars or nothing. Both sides point to each other and complain that required disclosure was not made. Both sides argue that the trial judge’s award is fundamentally flawed. These reasons explain why I would not interfere with the damages award, except as it applies to damages for sharing data with CIBC Mellon.

[84] In short, the trial judge’s approach is to be commended. He identified the evidentiary gaps in the record and the flaws in the parties’ damages theories, and correctly concluded that neither theory should prevail. The trial judge then applied common sense, guided by commercial reasonableness, in fashioning a fair and reasonable damages award that is generally rooted in the evidence. With the exception of the damages for sharing data with CIBC Mellon, it is free from error, and accordingly, immune from appellate interference. I have no doubt that the award is imperfect, but a standard of perfection is not required. This is especially so in a situation such as the case at bar where the parties clung to damages positions that were legally untenable and unsupported by the record.

II. Procedural Background

[85] The trial judge recognized the disparity in the parties’ positions on damages and was frank about the difficulties he was encountering in trying to fix an appropriate damages award based on a paper record. In the Liability Decision, he concluded:

The damages issue has caused me quite some difficulty. This matter proceeded as an application, not as a trial. The issues both parties raised about damages are complex, factually intense and are questions with respect to which the parties have diametrically opposed views … I have struggled long and hard to determine whether I could come to a responsible view about damages on a paper record. I have reluctantly come to the conclusion that I cannot.

[86] The trial judge ordered a trial on damages. This was a sensible approach on the record before him. The stated purpose of the damages trial was to better understand and absorb the conflicting evidence. The trial judge ordered that it proceed on the current record unless both sides agreed to otherwise supplement the record. This ruling, which effectively froze the evidentiary record, would become a bone of contention for BNY. However, as I will discuss, BNY’s complaints in this regard are of no moment as the exclusion of evidence complained of had no impact on the trial judge’s decision.

III. Decision below

(a) SS&C Damages Claim

[87] As noted, SS&C seeks damages of approximately $890 million USD. Its claim is based on the theory that unauthorized users of the data would have entered into agreements with SS&C on substantially the same terms as the Mellon Trust Agreement. SS&C took the actual monthly amounts paid by BNY and multiplied that sum by the number of entities that were wrongly accessing the data (which they say is 65) by the number of months those entities were in existence. This formula generated a “theoretical maximum model” of lost revenue of $150 million USD. SS&C also calculated damages based on 65 entities paying the minimum monthly fees under the Mellon Trust Agreement, which generated a figure for the “theoretical minimum model” of approximately $35 million USD.

[88] In addition to these sums, SS&C claimed for late payment fees. It was submitted that because the Mellon Trust Agreement permitted it to claim a penalty of 1.5% compounded monthly on any outstanding amounts for an invoice not paid within 60 days of receipt, it was entitled to claim late fees on its lost revenue. For the “theoretical maximum model” the late fees totaled approximately $740 million USD and for the “theoretical minimum model” the late fees totaled approximately $218 million USD. Therefore, based on 65 entities wrongfully accessing SS&C’s data, the damages inclusive of lost revenue and late fees ranged between approximately $253 million and $890 million USD.

[89] Given the trial judge’s finding in the Liability Decision that the sharing of data was permissible for entities with custodial businesses existing as of the time of the execution of the Mellon Trust Agreement, SS&C also calculated the damages figures using the same methods cited above on the basis of 44 entities. The result was an adjusted claim for damages inclusive of lost revenue and late fees that ranged between approximately $114 million to $482 million USD. This submission was in the alternative, as SS&C maintained throughout the proceedings below and on appeal that it is entitled to $890 million USD.

[90] The trial judge concluded that the report tendered by SS&C’s damages expert was based on an instruction from counsel that he should assume that the 65 entities entered into contracts with substantially the same terms and rates as the Mellon Trust Agreement. According to the trial judge, all the expert did was calculate damages based on that instruction. He did not accept that evidence, concluding that the expert “is not an expert assisting the court in devising the appropriate way to assess damages, he is simply carrying out mathematical calculations as instructed by counsel.” Therefore, the expert evidence was rejected on the basis that it was not “a true opinion on damages arrived at by independently applying his expertise to the facts.”

[91] The trial judge went on to reject the theory propounded by SS&C on the basis that it was inconsistent with the parties’ conduct. He noted that when the Mellon Trust Agreement and CIBC Mellon Agreement were amended to add new entities or groups of securities, the pricing for the supplemental information was lower than the pricing under the initial agreement. Regarding the SS&C theory, the trial judge observed that:

While it is conceivable that such an approach is appropriate, there is no evidence before me about the basis on which data service providers like SS&C charge for their data. Like most businesses, there is probably some relationship between volume and price. It may also be that the fee to share data with additional entities within a corporate group would be lower than the fee payable by the initial corporate signatory. There is simply no factual basis on which I can conclude that each of the 65 (or 44 or some lower number of) entities would be paying the same monthly fee as BNY did other than the fact that applicant’s counsel instructed Mr. McGavock to so assume.

[92] In addition, the trial judge also considered the damages theory in the context of the commercial relationship between the parties. He noted that the total fees paid under the Mellon Trust Agreement over 16 years was $4,586,273 USD and the total fees paid under the CIBC Mellon Agreement was $3,102,162 CAD. These relatively modest figures led to his observation that “it is difficult to accept that damages for sharing that data within the BNY family should come to $889 million or anything approaching that without a specific evidentiary basis for such a conclusion.” Further, he found that the notion of billing services on an entity-by-entity basis was contrary to the evidence at trial that the practice in the asset servicing industry was to centralize pricing data in order to take advantage of economies of scale.

(b) BNY’s Defence

[93] BNY’s position on damages centered on its contention that CIBC Mellon was the user of more than 95% of the data. According to BNY, the other users made only de minimis use of the data. On this basis, it submitted that SS&C suffered no damages.

[94] BNY also asserted that the cost of the data had fallen since the execution of the Mellon Trust Agreement. Specifically, BNY advised the court that when its contractual relationship with SS&C was terminated, it entered into a new agreement with a different supplier at a rate that was 25% less than it was previously paying SS&C. In addition, BNY argued that if it were required to enter into 65 contracts at the same rate as the Mellon Trust Agreement, it would have ended its business relationship with SS&C.

[95] The trial judge rejected the central contention in BNY’s defence that CIBC Mellon was the overwhelming user of data. He found that the evidence BNY tendered in support of this submission was unreliable and that its investigation regarding which entities were using the data was “not an investigation into the facts, it is simply a bald assertion masquerading as an inquiry.” The trial judge concluded that BNY could not account for over half of the data that SS&C delivered. He also rejected the assertion that the predominance of quotes for Canadian securities lead to the conclusion that the majority of quotes must have been for CIBC Mellon.

[96] Other arguments advanced by BNY were also rejected on the basis that BNY had not produced a sufficient evidentiary basis to support its assertions. For example, the trial judge declined to accept the expert evidence adduced by BNY that generally the price of data had declined and that service providers are reluctant to raise contract amendment issues because they often lead to a lower price. He also dismissed the submission that after the parties’ business relationship ended, BNY was able to negotiate a new contract at a cost that was 25% less than what SS&C was charging. In that regard, he noted that BNY had not produced the new contract or any of the bids it received for the contract.

[97] BNY’s failure to produce relevant documents was a significant concern for the trial judge. Shortly after the unauthorized data sharing was first discovered, SS&C wrote to BNY asking it to preserve records that would enable the parties to better understand which entities were using the data. BNY sent the request to its legal department. However, it declined to preserve and produce those records. SS&C took the position that BNY had committed the tort of spoliation with respect to this information. The trial judge was not prepared to find that the tort was committed, but he did draw the following adverse inference:

The adverse inference I draw is twofold: First, I infer that the unaccounted for data was used by unauthorized entities within the BNY group other than CIBC Mellon. Second, I decline to conclude that non-CIBC Mellon users made only de minimis use of the data as the respondent asks me to.

[98] The trial judge also dismissed other arguments raised by BNY. He concluded that he could not reduce the damages to a nominal amount – as requested by BNY – on the basis that BNY would have ended the business relationship immediately had it been forced to execute 65 contracts. According to the trial judge, had BNY been forthright regarding how it was using the data, it would have been in a position to terminate the business relationship. It was not in that position because of its lack of transparency, and it should bear the consequences of its misconduct.

[99] The trial judge refused to reduce the damages award on the basis that SS&C would have had to pay vendors for the data and would have paid sales commissions. On the former point, the trial judge relied on 1854329 Ontario Inc. v Cairo, 2022 ONCA 744, for the proposition that where a claimant may have an obligation to pass on part of a damages award to a third party, a court should not reduce a damages award for the contingent obligation. Regarding the commissions point, the trial judge found that BNY should not be entitled to a windfall in the form of a reduction in the amount that it would have otherwise been obligated to pay by deducting the notional commissions.

(c) Trial Judge’s Assessment of Damages

(i) Rationale

[100] The trial judge articulated the evidentiary shortcomings in the record and how the lack of evidence impacted his ability to determine the appropriate damages award:

The calculation of damages in this case is difficult because neither side has given the court the sort of record that would be helpful to assess damages. On the one hand, the applicant has not provided detailed information about the basis on which fees are calculated and has not provided examples of situations in which a multi-enterprise entity like BNY is charged a separate fee for each entity equal to the fee that the primary entity was paying. On the other hand, the respondent has failed to provide complete information about the use of the data within BNY and cannot account for 55.4% of the data it used. Similarly, although the defendant asserts that data prices have fallen since the contracts were entered into, it has provided nothing but bald assertions to that effect. It has not provided actual agreements, nor has it explained the parameters on which those sorts of contracts are based today.

[101] Given these concerns, the trial judge considered whether the evidentiary record afforded a reasonable and reliable alternative basis to quantify damages. He determined that it did and proceeded to calculate damages in two categories: damages for sharing data with CIBC Mellon and damages for sharing data with other BNY entities.

(ii) CIBC Mellon

[102] At the start of his analysis, the trial judge rejected the submission by BNY that it entered into a separate contract with CIBC Mellon by mistake. He noted that, among other things, BNY was not candid or transparent when the CIBC Mellon Agreement was cancelled because it did not tell SS&C that it intended to share data under the Mellon Trust Agreement.

[103] The trial judge found that BNY was liable for sharing data with CIBC Mellon in an amount equal to the revenue SS&C would have earned from CIBC Mellon between April 1, 2011 (the termination date of the CIBC Mellon Agreement) to February 28, 2017, when the business relationship between BNY and SS&C ended. He directed that the amount should be calculated by taking the average revenue that CIBC Mellon paid to SS&C for the three months preceding April 1, 2011, and subjecting it to an annual increase equal to the percentage by which payments under the Mellon Trust Agreement increased annually. Those monthly fees should then be notionally charged throughout the period and a final figure arrived at for lost revenue. This formula generated a damages figure of $922,887 with respect to the sharing of data with CIBC Mellon.

(iii) Rateable Approach

[104] Regarding the damages flowing from the unauthorized use of data by the other BNY entities, the trial judge determined that the best approach was to hold BNY liable for the data it could not account for by calculating its value having reference to the price BNY paid for data for which it could account. He labelled this damages theory the “rateable approach.”

[105] The total fees paid under the Mellon Trust Agreement amounted to $4,586,273 USD. BNY was able to account for 44.6% of the data that SS&C delivered, which was used by CIBC Mellon, and the trial judge inferred that BNY used the remaining 55.4% by sharing it with entities with whom it was not authorized to share data. Therefore, BNY effectively paid $4,586,273 for 44.6% of the data it received, which meant that 100% of the data would have cost $10,283,123 USD. Under the rateable approach, the delta between the fees for 100% of the data and the fees that were actually paid was $5,696,850 USD. The trial judge found BNY liable for that amount as damages for sharing data with other unauthorized entities.

[106] The trial judge recognized that both parties objected to the rateable approach, but he justified his decision this way:

To my mind, however, it strikes a balance between what I view as the commercially unrealistic approach of the applicant of assuming that each of 65 entities would pay the price paid each month under the Mellon Trust agreement and the somewhat too self-interested approach of the respondent which is to base damages on the minimum payment it can find without ever providing evidence of its actual usage. As noted earlier, I draw an adverse inference from the respondent’s failure to produce usage records. Part of that adverse inference applies here to support the rateable approach.

The rateable approach simply assumes that a third group of entities would have been prepared to pay a similar price for directionally similar data in directionally similar volumes. That strikes me as more commercially realistic than saying damages are 65 times what was paid under the Mellon Trust agreement or that damages are the absolute minimum that a single additional entity paid for a restricted amount of data. The rateable approach is also more consistent with the evidence about centralized pricing being a directional norm. The rateable approach applies centralized pricing to the unaccounted for data at the same rate as BNY was prepared to pay for the data for which it could account.

(iv) Late Penalties and Prejudgment Interest

[107] The final issue was whether SS&C was entitled to interest or damages in substitution for the prejudgment interest provided for in the Courts of Justice Act, R.S.O. 1990, c. C.43. SS&C claimed for damages based on the late fees in the Mellon Trust Agreement. Alternatively, it claimed for its weighted average cost of capital (i.e., the average rate of all debt and equity financing that SS&C paid to finance its assets) or a prejudgment interest rate of 5%.

[108] As noted above, the late charges amounted to 1.5% of the outstanding balance compounded monthly, commencing 60 days after the receipt of the unpaid invoice. Given the lengthy period of unauthorized sharing of data, the late payments amounted to significantly more than what was alleged to be owing for lost revenue. The trial rejected this claim on the grounds that there was no late payment because the respondent paid all invoices when rendered in the amount billed. Further, s. 128(4)(b) of the Courts of Justice Act precludes compound interest and generally there is no equitable jurisdiction to award compound interest in the absence of a wrongful retention of funds.

[109] Regarding the claim for the weighted average cost of capital, the trial judge found that it was inappropriate because the calculation of the amount claimed assumed that the entire revenue stream received would have been reinvested in the business when only net profit can be reinvested. In addition, there was no evidence regarding the weighted average cost of capital for the business unit in question. Finally, the trial judge rejected the claim for 5% prejudgment interest because it was based on s. 3 of the Interest Act, R.S.C. 1985, c. I-15, which provides for a 5% rate where no rate is fixed by law. However, the prejudgment rate is fixed by the Courts of Justice Act.

IV. ISSUES

[110] Regarding damages, BNY’s appeal raises the following issues:

• Whether the trial judge erred in awarding more than nominal damages after SS&C waived negotiating damages at trial;

• Whether the minimum performance principle was ignored;

• Whether SS&C was entitled to an award of damages to compensate it for lost profits; and

• Whether the damages awarded regarding sharing of data with CIBC Mellon was free of any error in principle.

[111] On its cross-appeal, SS&C raises the following grounds of appeal regarding damages:

• The trial judge erred in not awarding damages based on the Mellon Trust Agreement, which required an award of damages tied to the number of unlicensed entities;

• The trial judge compounded that error by not finding, as part of drawing an adverse inference, that the unlicensed entities made extensive use of the data;

• The trial judge erred in finding that there was insufficient evidence regarding how a multi-enterprise entity like BNY is charged for data; and

• The trial judge erred in failing to award damages for the late fees.

[112] Both parties submit that this court should vary the damages award and not send the issue of damages back to the Superior Court. As I have stated at the outset of my reasons and will explain in more detail below, I would not vary the damages award, except the damages for sharing data with CIBC Mellon. The parties’ submissions are otherwise unpersuasive and not in accord with the law or the evidence adduced at trial. Before considering the grounds of appeal, I will first analyze the applicable standard of review.

(a) Standard of Review

[113] Appellate review of an award of damages is subject to the principles articulated in Housen v. Nikolaisen, 2002 SCC 33, [2002] 2 S.C.R. 235: see Southwind v. Canada, 2021 SCC 28, [2021] 2 S.C.R. 450, at para. 154. This means that errors of law are reviewed on a standard of correctness, while errors of fact and errors of mixed fact and law are reviewed on a standard of palpable and overriding error.

[114] Despite this principle, appellate courts have expressed a general reluctance to interfere with damages awards. Appellate interference with a damages award is said to be “justified only where the trial judge made an error in principle, misapprehended the evidence, failed to consider relevant factors, considered irrelevant factors, made an award without any evidentiary foundation, or otherwise made a wholly erroneous assessment of damages”: TMS Lighting Ltd. v. KJS Transport Inc., 2014 ONCA 1, 314 O.A.C. 133, at para. 60. See also: Whitefish Lake Band of Indians v. Canada (Attorney General), 2007 ONCA 744, 87 O.R. (3d) 321, at para. 28; Extreme Ventures Partners Fund I LP v. Varma, 2021 ONCA 853, 24 B.L.R. (6th) 38, at para. 53, leave to appeal refused, [2022] S.C.C.A. No. 61.

[115] The reluctance to interfere with a trial judge’s damages assessment is perhaps best evidenced in the statement made by Wilson J. in Guerin v. The Queen, 1984 CanLII 25 (SCC), [1984] 2 S.C.R. 335, at p. 363: “…I do not think it is the function of this Court to interfere with the quantum of damages awarded by the trial judge if no error in principle in determining the measure of damages has been demonstrated ... The trial judge’s task was not an easy one but I think he “did the best he could”.”

[116] This approach to damages stems from two principal sources. First, damages quantification engages the exercise of discretion by a trial judge. Indeed, it often involves a series of discretionary decisions. Take for example a trial judge fixing damages in a personal injury case. They might have to determine damages for pain and suffering, attendant care, future care, and lost earnings based on their understanding of the facts of the case and the relevant caselaw. There is no single correct answer because the analysis is dependent on how a particular trial judge evaluates the impact of the injury on various categories of damages.

[117] Second, even in commercial disputes, where a trial judge may have fewer discretionary decisions to make, a judge may encounter difficulty in fixing a damages award because the evidentiary record does not provide an adequate basis to answer all the issues raised in the calculation of damages. However, while damages quantification may be inherently difficult, this does not excuse the claimant from leading evidence that is supportive of its claim. Finlayson J.A. described that obligation this way in Martin v. Goldfarb (1998), 1998 CanLII 4150 (ON CA), 41 O.R. (3d) 161 (C.A.), at p. 187, leave to appeal refused, [1998] S.C.C.A. No. 516:

I have concluded that it is a well-established principle that where damages in a particular case are by their inherent nature difficult to assess, the court must do the best it can in the circumstances. That is not to say, however, that a litigant is relieved of his or her duty to prove the facts upon which the damages are estimated. The distinction drawn in the various authorities, as I see it, is that where the assessment is difficult because of the nature of the damage proved, the difficulty of assessment is no ground for refusing substantial damages even to the point of resorting to guess work. However, where the absence of evidence makes it impossible to assess damages, the litigant is entitled to nominal damages at best.

[118] Applying these principles, we must review the trial judge’s damages award with considerable deference. The single overarching question that must be considered is: Did the trial judge do his best to assess damages on the available evidence or did he impermissibly award damages when none had been proven? In analyzing that question, we must be alert to errors in principle. I turn now to a consideration of the specific grounds of appeal.

(b) BNY Grounds of Appeal

(i) Negotiating Damages

[119] The leading case on negotiating damages or negotiated damages is from the Supreme Court of the United Kingdom in Morris-Garner and another v. One Step (Support) Ltd., [2018] UKSC 20, [2018] 3 All E.R. 659. There, the court recognized that in limited circumstances, a court may award damages based on what the parties might have negotiated as an acceptable price for an asset that has been wrongfully acquired. In Atlantic Lottery Corp. Inc. v. Babstock, 2020 SCC 19, [2020] 2 S.C.R. 420, the Supreme Court of Canada recognized the availability of such damages in Canada.

[120] It is beyond dispute that SS&C was initially seeking damages on this basis, but it later advised the court unequivocally that it was not claiming negotiating damages. In a lengthy ruling on objections, the trial judge found that SS&C was not seeking damages based on this theory and, therefore, prohibited BNY from adducing evidence in this regard: 2022 ONSC 5834. BNY submits that the trial judge erred because, having rejected the applicability of negotiated damages, he proceeded to award damages on that basis. It argues that the trial judge calculated damages based on what the parties would have negotiated had BNY sought SS&C’s authorization to share the data.

[121] I do not find this argument persuasive. I note first that negotiating damages were unavailable in the circumstances of this case. At para. 92 of Morris-Garner, the court described the circumstances where such damages could be awarded:

As the foregoing discussion has demonstrated, such circumstances can exist in cases where the breach of contract results in the loss of a valuable asset created or protected by the right which was infringed, as for example in cases concerned with the breach of a restrictive covenant over land, an intellectual property agreement or a confidentiality agreement. Such cases share an important characteristic with the cases in which Lord Shaw’s “second principle” and Nicholls LJ’s “user principle” were applied. The claimant has in substance been deprived of a valuable asset, and his loss can therefore be measured by determining the economic value of the asset in question. The defendant has taken something for nothing, for which the claimant was entitled to require payment. [Emphasis in original.]

[122] Thus, this type of damage award may be applicable in cases where a contract creates or protects an asset, and its breach deprives the wronged party of the use of the asset. In this case, the contract was for the supply of a service. Even if it could be argued that the data is property that is protected by the contract, there is no suggestion that SS&C was deprived of that asset.

[123] Further, and in any event, I am also not persuaded that the trial judge awarded damages based on the negotiating damages theory. BNY contends that what the trial judge did “was to base his award on the different terms which he found SS&C could have hypothetically negotiated with BNY Mellon under the Mellon Trust Agreement or with CIBC Mellon or the Other Affiliates under other contracts.” In fact, the trial judge explicitly found that there was insufficient evidence to understand what the parties might have negotiated in circumstances where a multi-entity enterprise entered into an agreement for the provision of data.

[124] The fact that the trial judge rejected the damages theory propounded by SS&C does not render the service it provided valueless. What the trial judge was obliged to do – and what he did do – was to determine whether there was an evidentiary basis to quantify the loss. If there was no evidentiary basis to quantify damages, then SS&C would have failed to discharge its onus and it would be entitled to only nominal damages. The trial judge explicitly undertook this analysis: “Having rejected both sides’ theories of damages, I ask myself, as I am obliged to, whether the evidentiary record affords a reasonable and reliable alternative basis for the quantification of these damages. I conclude that it does.”

[125] That was the correct analytical approach. The trial judge considered SS&C’s damages theory for expectation damages and rejected it. He then examined the record to determine if there was an alternative basis to quantify damages, found that there was, and awarded damages on that basis.

[126] It was reasonable for SS&C to expect to be paid for the service it provided. That is its business. The business only works if the service has value. That is why, as the late Professor Stephen Waddams and Patrick Healy wrote, courts have long calculated damages by reference to the value of the services the defendant wrongfully acquired from its contract counterparty: The Law of Damages (Toronto: Thomson Reuters, 1991) (loose-leaf 2023-Rel.1), at § 9.5. Those authors cited Alkok v. Grymek, 1968 CanLII 10 (SCC), [1968] S.C.R. 452, in which the Supreme Court required an owner who wrongfully acquired building services from a contractor to pay him “the actual value of the work performed”: at p. 458 (quotation omitted). As in Alkok, the trial judge was attempting to quantify the value of the service wrongfully acquired. That valuation was part of the quantification of damages for the use of data by unauthorized entities. In determining the appropriate damages, the trial judge did not seek to impose hypothetical contractual terms. Quite the opposite. He calculated the value of the service of providing the data based on what the parties had historically paid and extrapolated that figure to cover all of the services provided.

[127] BNY also argued that the trial judge erred in ordering that the evidentiary record for the damages trial could only be expanded on the consent of the parties. I see no error in the trial judge’s approach.

[128] As the trial judge noted, the application was meant to address both liability and damages, and the trial was intended to clarify the conflicting evidence on damages. The trial judge correctly observed that the parties had several years to adduce the necessary evidence on liability and damages. I also agree with his statement that parties are entitled to assume that when a case comes on for hearing, the record is closed. The trial judge was under no obligation to permit the parties to expand the evidentiary record.

[129] Further, and in any event, the evidence that BNY attempted to tender was primarily directed at undermining SS&C’s damages theory. That theory was rejected by the trial judge. Therefore, the failure to allow this evidence was of no moment in the determination of damages.

[130] For the foregoing reasons, I would reject this ground of appeal.

(ii) Minimum Performance

[131] The doctrine of minimum performance applies where there has been a repudiation of a contract by a defendant and there are several ways the defendant could have performed the contract. In such circumstances, the plaintiff is only entitled to damages based on the minimum performance provided for under the agreement.

[132] The doctrine has long been part of English law and is well established in the American jurisprudence regarding damages for breach of contract. It has also been accepted in Canada. It was analyzed by the Supreme Court in the seminal case Hamilton v. Open Window Bakery Ltd., 2004 SCC 9, [2004] 1 S.C.R. 303. There, at para. 8, the court cited Cockburn v. Alexander (1848), 6 C.B. 791, 136 E.R. 1459 (Ct. Com. Pl.), for the following description of the doctrine: “Generally speaking, where there are several ways in which the contract might be performed, that mode is adopted which is the least profitable to the plaintiff, and the least burthensome to the defendant.”

[133] The court at para. 13 of Open Window also cited Withers v. General Theatre Corp., [1933] 2 K.B. 536, [1933] All E.R. Rep. 385 (C.A.), for the following illustration of minimum performance:

Now where a defendant has alternative ways of performing a contract at his option, there is a well settled rule as to how the damages for breach of such a contract are to be assessed .… A very common instance explaining how that works is this: A. undertakes to sell to B. 800 to 1200 tons of a certain commodity; he does not supply B. with any commodity. On what basis are the damages to be fixed? They are fixed in this way. A. would perform his contract if he supplied 800 tons, and the damages must therefore be assessed on the basis that he has not supplied 800 tons, and not on the basis that he has not supplied 1200 tons, not on the basis that he has not supplied the average, 1000 tons, and not on the basis that he might reasonably be expected, whatever the contract was, to supply more than 800 tons. The damages are assessed … on the basis that the defendant will perform the contract in the way most beneficial to himself and not in the way that is most beneficial to the plaintiff.

[134] Underlying the doctrine is the enforcement by the court of the bargain struck by the parties. When a plaintiff agrees to accept a minimum level of performance, the court will hold it to its bargain. The plaintiff is not entitled to more than the minimum performance it agreed to in the contract. This is a commercially sensible approach and one that provides certainty to contracting parties. It avoids a result where the plaintiff obtains more than what it bargained for in the agreement.

[135] BNY submits that the doctrine applies in this case and the trial judge erred in failing to rely on it in his analysis. Specifically, BNY argues that had it known that SS&C would insist on separate contracts for each entity using the data, it would have exercised its rights under the Mellon Trust Agreement to end the contract on 90 days’ notice. Indeed, it notes that when SS&C discovered the unauthorized use of the data, this is precisely what happened. SS&C gave notice under the Mellon Trust Agreement and the business relationship ended. Therefore, it submits that SS&C should only be entitled to nominal damages.

[136] This submission is meritless. As I will explain, this case does not trigger the doctrine because BNY never wrongfully repudiated the contract, and the expansion of the doctrine to fit the facts of this case would lead to absurd and perverse results that are contrary to the policy underlying the doctrine. In order to analyze this submission, it is helpful to consider the facts in Open Window.

[137] In Open Window, the appellant, Jane Hamilton, had entered into a contract with the respondent, Open Window Bakery Limited (“OWB”), for a term of 36 months as an exclusive agent for the marketing and sale of OWB’s baked goods in Japan. The contract provided that OWB could terminate the agreement without notice for cause and without cause after the commencement of the 19th month of the term on three months’ notice. Approximately 16 months after the contract was executed, OWB repudiated the agreement alleging cause. The trial judge held that OWB wrongfully repudiated the contract and awarded damages reflecting the payments that would have been made under the full 36-month term of the contract, less an allowance of 25% to reflect the possibility that OWB might, at some later point, have validly exercised its right to terminate the agreement with notice.

[138] On appeal, a majority of this court held that the early termination clause with three months’ notice constituted the minimum guaranteed benefits under the contract. Therefore, it also constituted OWB’s maximum exposure for damages. The Supreme Court dismissed the appeal of the damages award based on the doctrine of minimum performance, upholding this court’s decision.

[139] The situation in the case at bar is distinguishable from the facts of Open Window. Here, BNY has had the benefit of the contract for 17 years and did nothing in that period to repudiate it. This is not a situation where the breaching party wrongfully terminated a contract without notice and there was an option of lawfully terminating the contract with reasonable notice. Thus, the minimum performance doctrine is not triggered because BNY never wrongfully repudiated the contract: Hamilton, at para. 11; Radikov v. Premier Project Consultants Ltd., 2017 ONSC 7192 (Div. Ct.), at paras. 64, 72-73. Nor is it a situation where there was a minimum level of contractual performance. BNY was obliged not to share data with unauthorized entities; there was no alternative permissible performance that permitted it to do so.

[140] It is not open to BNY as the breaching party to say that had it been caught sooner it would have mitigated damages. The trial judge’s decision is based on what happened in this case and not in an alternative reality where BNY’s wrongful conduct was discovered sooner or BNY had acted honourably and disclosed its misconduct.

[141] Part of the search for a commercially reasonable damages award is the avoidance of absurd results. Expanding the minimum performance doctrine to cover these circumstances leads to an absurd result because it would permit a party who has breached a contract for many years by wrongfully obtaining the benefits of the contract to escape liability on the basis that it would have terminated the agreement had it been caught earlier.

[142] The argument advanced by BNY would be an absurd result. It would leave the aggrieved party without a remedy and would encourage businesspeople to engage in wrongful conduct because there are no economic consequences if their misconduct is discovered. This is also inconsistent with the policy underlying the doctrine of minimum performance, which was designed to prevent a claimant from obtaining a windfall by holding them to their bargain. The interpretation of the doctrine urged upon us by BNY would sway the pendulum to the other extreme by permitting defendants to wrongfully receive benefits under a contract at little or no cost and force a result on a defendant that it never bargained for.

[143] Nowhere in the Mellon Trust Agreement did SS&C agree to a result where BNY could wrongfully use its data for years and pay SS&C nothing for the wrongly shared data. BNY’s argument would turn the doctrine of minimum performance on its head by granting a breaching party contractual rights that it never bargained for in its agreement. The trial judge correctly rejected this argument.

(iii) Gross Revenue

[144] BNY submits that the trial judge erred in his damages award by measuring SS&C’s damages based on revenues rather than net profits. It argues that SS&C would have had to pay fees to the vendors of data it aggregated and commissions to its salespeople. I am not persuaded that the trial judge erred in this regard.

[145] Where an aggrieved party to a contract claims for damages for a contract that was not performed, they are generally entitled to only the net profits for that contract: Jaremko v. A.E. LePage Real Estate Services Ltd. (1989), 1989 CanLII 4163 (ON CA), 69 O.R. (2d) 323 (C.A.), at p. 326. For example, if a customer orders 1,000 units of a product that costs $10 per unit to manufacture and is valued under the contact at $50 per unit and then cancels the contract before the products are manufactured, the aggrieved party may only claim for its net profits and not its gross revenue (i.e., 1,000 x $40 = $40,000). This is because it should not receive a windfall given that it did not have to expend $10 per unit to manufacture the product. Instead, it should be put in the position it would have been had the contract been performed.

[146] The facts of the present case are different. Here, SS&C performed the contract. In order to obtain the data, it had to pay its suppliers and it did so. Thus, the court does not have to determine what its notional costs are because the costs were actually incurred. Therefore, there is no need to make a deduction to put SS&C in the position it would have been had the contract been performed because it performed the contract and incurred the cost of obtaining data from its suppliers.

[147] As I understand it, BNY’s argument is based on the notion that SS&C did not pay its vendors for the data and, therefore, full payment would result in a windfall to SS&C. To the extent that data was obtained without paying its suppliers – which is not evident from the record – I see no error in the trial judge’s reliance on Cairo for the proposition that there should be no deduction to reflect the potential future obligation to suppliers. To hold otherwise runs the risk that SS&C would not be compensated to cover a claim that may be asserted by its suppliers.

[148] I am also not prepared to find that the trial judge erred by failing to deduct commissions. Even if it is assumed that no commissions were paid for all or part of the data, the trial judge was not provided with sufficient evidence to quantify the deduction. If BNY wanted to challenge the quantum of the claim made by SS&C on this basis, it should have obtained the necessary evidence from SS&C.

(iv) CIBC Mellon

[149] As noted, the trial judge found that 44.6% of the data was used by CIBC Mellon. He further found that BNY paid $4,586,273 USD for the data used by CIBC Mellon. However, despite these findings, the trial judge also held that BNY should be subject to a separate damages award for sharing data with CIBC Mellon, fixed in the amount of $922,887. This is the amount that CIBC Mellon would have paid under the CIBC Mellon Agreement had it not been terminated.

[150] BNY did not focus its submissions on damages from sharing data with CIBC Mellon. Instead, it submitted that SS&C was only entitled to nominal damages. However, the trial judge’s award of damages for sharing data with CIBC Mellon amounts to an error in principle because the damages award in addition to the contractual payments constitutes double compensation. SS&C was not entitled to be paid under the Mellon Trust Agreement for the usage of this data and at the same time also have the judge require CIBC Mellon to continue to make payments in accordance with its previous contractual arrangements. Therefore, the CIBC Mellon damages award of $922,887 must be struck.

(c) SS&C Grounds of Appeal

(i) Damages Based on the Mellon Trust Agreement

[151] SS&C submits that because the Mellon Trust Agreement is a licensing agreement, each unauthorized entity that received data should have paid fees to SS&C equal to the fees required under the Mellon Trust Agreement. It argues that this was a real-world contract that generated the theoretical maximum and theoretical minimum models, which the trial judge should have followed in assessing damages.

[152] In support of this argument, SS&C relies in this appeal – as it did at trial – on two licensing contract decisions: Geophysical Service Incorporated v. Total SA, 2020 ABQB 730, and XY, Inc. v. International Newtech Development Incorporated, 2012 BCSC 319, 2 B.L.R. (5th) 19, aff’d, 2013 BCCA 352, 366 D.L.R. (4th) 443, leave to appeal refused, [2013] S.C.C.A. Nos. 376, 378 & 380. I see no error in the trial judge’s determination that these decisions were not apt because in neither case was the court being asked to assume that multiple, multi-year contracts would be executed. Instead, in those cases, the relevant contracts quantified the obligation that the defendants breached. This is distinguishable from the case at bar because the Mellon Trust Agreement did not set a price for wrongful data sharing with affiliates.

[153] The difficulty the trial judge faced was that there was no quantification of damages within the Mellon Trust Agreement for the impermissible use of data. These were sophisticated parties and there is no reason why such a term could not have been included in the Mellon Trust Agreement. Further, SS&C failed to produce compelling evidence regarding how multi-entity enterprises are charged for data. Consequently, the trial judge was forced to develop his own theory of damages.

(ii) Spoliation and Adverse Inference

[154] SS&C asserted a claim of spoliation regarding the missing usage data. Spoliation is a common law doctrine, which finds its origins in the legal system of ancient Rome. It was first developed in Canada in 1896 by the Supreme Court in St. Louis v. The Queen, 1896 CanLII 65 (SCC), [1896] 25 S.C.R. 649, and has not changed much in the past 128 years: Gideon Christian, “A ‘Century’ Overdue: Revisiting the Doctrine of Spoliation in the Age of Electronic Documents” (2022) 59:4 Alta. L. Rev. 901. It is an evidentiary rule that allows the court to remedy abuses of its process, although whether it is also a free-standing tort claim remains unresolved: Trillium Power Wind Corporation v. Ontario, 2023 ONCA 412, 167 O.R. (3d) 321, at paras. 21-22, 24, leave to appeal refused, [2023] S.C.C.A. No. 363.

[155] In St. Louis, the court found that the doctrine creates a rebuttable presumption that evidence destroyed would have been unfavourable to the party who destroyed it: at pp. 652-665. More recently, the constituent elements of spoliation have been described as follows: “to prove spoliation, a party must prove: (i) that relevant evidence was destroyed; (ii) that legal proceedings existed or were pending; and (iii) that the destruction was an intentional act indicative of fraud or intent to suppress the truth”: Stamatopoulos v. The Regional Municipality of Durham, 2019 ONSC 603, 85 M.P.L.R. (5th) 31, at para. 606, aff’d, 2022 ONCA 179, 26 M.P.L.R. (6th) 1, leave to appeal refused, [2022] S.C.C.A. No. 12. Thus, the unintentional destruction of documents is not spoliation, although it may still attract sanctions or remedies: Christian, at p. 912; McDougall v. Black & Decker Canada Inc., 2008 ABCA 353, 302 D.L.R. (4th) 661, at paras. 24-25.

[156] Intent has two elements. It is not enough that the destruction of the document be proven to be intentional. In addition, the claimant must also prove “a mala fides desire to prevent the use of the document in litigation, to suppress the truth, and hence impact the outcome of the litigation”: Christian, at pp. 911-12.

[157] An important factor relevant to the issues of intent and whether litigation is contemplated, is whether a party is served with a preservation notice. Such a notice demands that the served party preserve documentation relevant to active litigation or anticipated litigation. In the present case, counsel for SS&C, Chris Paliare, wrote to senior executives of BNY asserting his client’s position regarding the anticipated litigation. In that letter, Mr. Paliare stated “We trust that you will ensure preservation of all communications, documents, and files related in any way to BNY’s relationship with SS&C and BNY’s provision or sharing of Data to any third-parties.” This was a sensible approach because it removed any doubt that BNY understood that it had an obligation to preserve relevant documents.

[158] BNY’s then counsel, a lawyer then at McCarthy Tétrault (who did not appear on this appeal), responded to Mr. Paliare’s letter. He stated that “SS&C’s management has also known that CIBC Mellon has been by far the largest user of the Data. In fact, the usage by the other Mellon Trust business lines over the years was such that, but for CIBC Mellon’s need for the Data, the Agreement would never have been signed…In sum, BNY Mellon categorically denies that a breach of the Agreement has occurred. Accordingly, BNY Mellon declines to accede to the demands set forth in your letter.”

[159] As we know, the trial judge rejected BNY’s position that the use of the data by entities other than CIBC Mellon was de minimis. Therefore, the basis for the rejection of the preservation and production of the data was invalid. What is more concerning is that the obligation to preserve relevant documents was eschewed on the basis of BNY’s views about SS&C’s claim. It is not open to lawyers or parties to ignore their obligations under the Rules of Civil Procedure, R.R.O. 1990, Reg. 194, and at common law based on their opinion of the merits of a potential claim. As found by the trial judge, BNY knew or ought to have known that the data was required for the litigation, yet it was never produced, and no explanation was ever proffered by BNY or its counsel regarding why it was not preserved.

[160] SS&C complains that the trial judge drew an adverse inference against BNY but made no factual findings about the extent of the unauthorized use of data. Instead, all he found was: “that the unaccounted for data was used by unauthorized entities within the BNY group other than CIBC Mellon”, and that he could not find that “non-CIBC Mellon users made only de minimis use of the data.”

[161] The trial judge rejected SS&C’s spoliation claim on the grounds that “spoliation as a standalone tort is one for which the wrongdoer can be liable for damages. The applicant seeks no such remedy here. The applicant seeks an adverse inference.” He also stated that “adverse inferences of that sort have nothing to do with spoliation but have to do with party’s failure to produce evidence to support a proposition that it is advancing.” The trial judge proceeded to draw the adverse inference noted above.

[162] If the trial judge meant that the remedy for spoliation was limited to damages or that an adverse inference is not an available remedy where spoliation is established, he erred in law. Regardless of whether spoliation is only an evidentiary rule or is also a standalone tort claim, an adverse inference is an available remedy where spoliation has been established: Doust v. Schatz, 2002 SKCA 129, 227 Sask. R. 1, at para. 29; Trillium Power Wind, at para. 24. Indeed, in St. Louis, the Supreme Court held that where spoliation has been established, a rebuttable presumption is created that the evidence destroyed would have been unfavourable to the party who destroyed it. As the trial judge found, BNY has not explained why the data was not preserved. Therefore, it has not rebutted that presumption. On this appeal, SS&C asserts that the trial judge erred in not drawing a more comprehensive adverse inference. As I will explain, I am not persuaded that there is any basis for this court to interfere with the adverse inference found by the trial judge. Moreover, in my view, SS&C’s adverse inference ground of appeal amounts to nothing more than an alternative attack on the trial judge’s damages theory.

[163] As for the merits of SS&C’s spoliation argument, although the trial judge did not offer a view regarding whether SS&C had established spoliation in this case, I have no doubt that it did. Having been warned to preserve data and with full knowledge that litigation was going to be commenced, BNY refused to do so on the basis that it rejected the allegations made against it. The reasonable inference is that it did so to suppress the truth in the litigation. In short, it failed to preserve important data that was highly relevant to the issue of data sharing, and it chose to ensure that the data would not be available in any legal proceeding. BNY has offered no compelling evidence to rebut this inference.

[164] It is evident that the trial judge was troubled by BNY’s conduct in the litigation. His concerns were well founded. The failure to preserve and produce relevant documents is conduct worthy of censure and the drawing of an adverse inference was appropriate in the circumstances of this case. Our civil justice system cannot function when parties do not comply with their disclosure obligations. That said: “whether to draw an adverse inference is a highly discretionary fact-based assessment which must be accorded deference”: The Cambie Malone’s Corporation v. British Columbia (Liquor Control and Licensing Branch), 2016 BCCA 165, 87 B.C.L.R. (5th) 219, at para. 40. See also: Parris v. Laidley, 2012 ONCA 755, at para. 2.

[165] A different judge may have drawn another adverse inference or ordered a different remedy. For example, where spoliation has been established, it is open to a judge to strike a pleading, including a statement of defence. Thus, the tactical decision to ignore production obligations is fraught with danger. Moreover, I am troubled by the position taken by BNY in this litigation. It smacks of contempt for the justice system. I hasten to add that my criticism is not directed at BNY’s former counsel. He is an experienced litigator, and I must assume he advised his client that its position was unsustainable.

[166] SS&C submits that the finding that the non-CIBC Mellon entities usage was not de minimis is not a positive finding about who used the data, how much, or over what period. Further, it argues that the findings leave open an array of possibilities when there is actually only one reasonable conclusion. Once the trial judge rejected BNY’s theory that the other entities only made minimal use of the data, he had to conclude that they all made widespread use of the data. It says that the trial judge should have presumed that the destroyed and non-produced records would reveal widespread usage by the unauthorized entities and applied the negotiated fee schedule under the Mellon Trust Agreement. Accordingly, he is alleged to have erred by stopping short of deciding the ultimate factual issue before him.

[167] I reject this argument. In the circumstances of this case, BNY’s conduct and the trial judge’s errors regarding the law of spoliation make no difference to the ultimate result. Instead, SS&C’s argument regarding the scope of the adverse inference is a red herring. The trial judge’s calculation of damages was based on data use and his conclusion that whatever data that was unaccounted for should be deemed to have been wrongfully shared. Therefore, the complaint on appeal that the trial judge should have gone farther and provided a detailed calculation of the extent of the sharing is unavailing. The real gravamen of SS&C’s argument is not that a different data calculation should have been made, but that the trial judge failed to impose a series of agreements with substantially the same terms as the Mellon Trust Agreement. As discussed above, the trial judge correctly rejected this argument on the basis that there was insufficient evidence supporting the notion that a multi-enterprise entity would enter into a series of agreements.

(iii) Misapprehension of the Record

[168] On this ground, SS&C’s position is that when the trial judge rejected SS&C’s damages model, he erred by faulting SS&C for not leading evidence that each entity would pay the same as the primary licensed entity, and in suggesting that the agreements before him contradicted that assertion. SS&C submits that those were palpable and overriding errors that conflicted with the record and his own findings. The agreements it references are the CIBC Mellon Agreement and the Mellon Trust Agreement.

[169] According to SS&C, the CIBC Mellon Agreement was a real-world example between these parties. Within short succession, CIBC Mellon and Mellon Trust entered into two separate agreements with SS&C’s predecessor with comparable fee schedules. The two agreements otherwise bore no relation to each other and neither entity was legally entitled to share under the other agreement simply because they were affiliates. According to SS&C, no further evidence was needed to show that two related entities would enter into different agreements paying similar amounts under each agreement.

[170] I am not persuaded by this submission. SS&C’s position that BNY would have paid roughly 65 times what it was paying for the supply of data by entering into multiple contracts strikes me as unrealistic. The onus lay with SS&C to produce evidence of such contractual arrangements with a multi-entity corporation. It failed to do so. Therefore, I see no error in the trial judge’s rejection of this claim.

(iv) Late Fees and Prejudgment Interest

[171] SS&C argues that the trial judge erred by finding there was no late payment, by mischaracterizing the late payment as compound interest, and by failing to engage with the analysis on the enforceability of penalty clauses.

[172] I see no error in the trial judge’s approach. Plainly, the late fee provision applied in circumstances when an invoice is issued, and it was designed to ensure prompt payment of invoices. It was not meant to apply to notional invoices, which provide no notice to the payor of an amount due. SS&C’s reliance on the late payment provision is misplaced.

V DISPOSITION

[173] I would dismiss the parties’ appeal and cross-appeal on damages, except I would delete paragraph 4 of the judgment and revise paragraph 5 of the judgment to fix CIBC Mellon damages at $0.

[174] If the parties cannot agree on the issues of the costs of this appeal or any variation of the costs award below, I would order as follows:

• BNY shall serve and file its costs submissions of no more than 10 pages plus a bill of costs within 10 days of the release of these reasons;

• SS&C shall serve and file its costs submissions of no more than 10 pages plus a bill of costs within 10 days of the receipt of BNY’s costs submissions; and

• BNY shall serve and file any reply costs submissions of no more than 5 pages within 10 days of the receipt of SS&C’s costs submissions.

“C.W. Hourigan J.A.”

Miller J.A.:

[175] I concur entirely with the reasons of Tulloch C.J.O. and Hourigan J.A.

Released: September 12, 2024 “M.T.”

“B.W. Miller J.A.”

[^1]: BNY’s reliance on Brown v. Belleville (City), 2013 ONCA 148, 114 O.R. (3d) 561, is misplaced because in that case, unlike here, the text and surrounding circumstances showed that the parties intended the term “successor” to have a broader meaning: at paras. 80-88; see also Resolute FP Canada Inc. v. Ontario (Attorney General), 2019 SCC 60, [2019] 4 S.C.R. 394, at paras. 158-162, per Côté and Brown JJ. (dissenting, but not on this point).

[^2]: I also decline to consider the other new issue BNY attempted to raise, that the trial judge did not find that SS&C had rebutted the presumption as to its corporate predecessor, Securities Valuation, as section 12 of the Act requires. Like subsection 5(1)(b)’s modified objective knowledge test, this issue raises questions of mixed fact and law that this court cannot resolve due to the lack of factual findings and BNY has not explained why it failed to raise it at trial: Kaiman, at paras. 18, 21.

[^3]: I use the term “orders” to refer to both judgments, meaning decisions that finally dispose of actions or applications on their merits, and other orders. See Rules of Civil Procedure, R.R.O. 1990, Reg. 194, r. 1.03 (definitions of “judgment” and “order”).

[^4]: Courts also take a stricter approach where, unlike here, changing an unentered judgment requires reopening a trial or would effectively extend a limitation period: 671122 Ontario Ltd. v. Sagaz Industries Canada Inc., 2001 SCC 59, [2001] 2 S.C.R. 983, at paras. 59-61; Brown v. The Municipal Property Assessment Corp., 2014 ONSC 7137, 31 M.P.L.R. (5th) 77, at paras. 25-27.