The Supreme Court of Canada has issued its first comprehensive statement on the doctrine of spoliation since 1896, in SS&C v. BNY Mellon, 2026 SCC 29 an appeal arising from a seventeen-year data-licensing dispute between a data provider (SS&C) and a global custodial bank (BNY Mellon). The Court held that the respondent bank committed spoliation by refusing to preserve evidence after receiving a preservation notice, and that the courts below committed reviewable legal and factual errors in their assessment of damages flowing from that spoliation. The takeaway: the company that hides the ball does not get to walk away clean, but it also does not automatically hand the other side a blank cheque either. The Court sided with SS&C on the core wrongdoing but sent the question of damages back to the Superior Court of Justice, requiring the trial judge to draw stronger adverse inferences to quantify damages.
The Players and the Business at Stake
SS&C Technologies Canada Corporation (“SS&C”) is a data provider that gathers and processes specialized pricing information used by financial institutions to value securities, and it is now part of the world's largest hedge fund and private equity administrator, with nearly US$45 trillion in assets under management. On the other side is Bank of New York Mellon Corporation (“BNY”), the world's largest custodial bank, holding more than US$35 trillion in assets under custody.
Their relationship began back in 1999, when SS&C's predecessor company signed two separate data licensing agreements with BNY’s predecessors: the “Mellon Trust" and CIBC Mellon. The Mellon Trust Agreement between SS&C and BNY stayed in force for nearly seventeen years, under which BNY paid SS&C roughly US$4.6 million over its lifetime.
SS&C's licensing model was volume-based, and it charged per security, meaning BNY paid once for the data tied to a specific security, even though that same pricing data could be valuable to any number of BNY's customers. Crucially, and the reason behind this case reaching the Supreme Court of Canada, the agreement barred BNY from sharing, redistributing, or selling that data to other parties, including its own subsidiaries and affiliates.
In 2011, CIBC Mellon terminated its separate agreement with SS&C, claiming it no longer needed the data services. In reality, CIBC Mellon still needed the data, but had simply started receiving it for free through BNY instead, without ever telling SS&C. That workaround let CIBC Mellon dodge SS&C's licensing structure entirely.
SS&C did not discover this arrangement until October 2016. A technical glitch disrupted SS&C's data deliveries, and CIBC Mellon's complaint about the resulting interruption inadvertently disclosed that it had been receiving SS&C's data through BNY. A CIBC Mellon employee even wrote to SS&C admitting, in effect, that they might have been receiving information they never paid for. SS&C subsequently terminated the agreement with BNY and commenced an action for breach of contract in 2017.
As the litigation unfolded, the scale of the problem became staggering. The trial judge ultimately found that BNY may have shared SS&C's proprietary data with as many as sixty-five affiliated entities over the full seventeen-year span of the agreement, and that BNY had even been paid by some of those affiliates for passing the data along.
The Missing Evidence Problem
The spoliation issue that ultimately drove this appeal arose from the parties' conduct after the breach was discovered. Once SS&C learned of the scheme, it formally demanded on December 23, 2016, that BNY preserve every record showing which affiliates had received the data, how much data each one received, and how much money BNY had earned from redistributing it. BNY refused. Its lawyers took the position that because CIBC Mellon was assumed to be the primary user of the data, there had been no breach at all, and so BNY declined to preserve anything. It offered no other justification for ignoring the request.
As defined by the Court, spoliation is “the intentional destruction, alteration, mutilation, or concealment of evidence aimed at undermining the truth-finding process of a lawsuit.”
To prove spoliation, a party must establish four things on a balance of probabilities:
- that the evidence was intentionally destroyed, altered, mutilated, or concealed;
- that litigation was already underway or reasonably expected at the time;
- that the evidence was actually relevant to that litigation; and
- that it is reasonable to infer that evidence was destroyed specifically to affect the case.
If all four elements are met, a presumption kicks in that the missing evidence would have hurt the spoliator's case, and the burden then shifts to the wrongdoer to prove otherwise. If they cannot rebut that presumption, the court is required to draw an adverse inference against them.
The Supreme Court agreed that every element of the spoliation test was satisfied, noting that the data was clearly relevant, was destroyed or never produced after the preservation notice went out, and that BNY had full knowledge litigation was coming when it made that choice. Importantly, BNY's rationale for refusing to preserve the data, essentially deciding for itself that SS&C's claim lacked merit, was flatly rejected. As the Court of Appeal put it, parties and their lawyers are not free to ignore preservation obligations based on their own opinion of how strong the case against them might be.
Two Separate Legal Battles: Liability and Damages
The case generated two distinct rounds of appellate scrutiny, one over whether BNY breached the contract at all, and a second over how much money it owed as a result.
On the liability question, BNY argued that the Mellon Trust Agreement was what the industry calls a "line of business agreement," meaning it authorized data sharing across BNY's entire corporate family, rather than an "enterprise agreement" that would apply only to specific named entities. Both the trial judge and the Court of Appeal rejected this argument. The Court of Appeal found that the contract's defined term "Client" referred to a single entity, that the commercial purpose of the deal was to let SS&C control exactly who accessed its data, and that the conduct of the parties over the years pointed to BNY alone, not an entire corporate family, as the actual contracting party. The trial judge took a factual misstep in finding that a certain Mellon entity, Mellon Financial Corporation, did not need any data under the agreement, when in fact it had its own market data staff and provided real services. Correcting for that error, the Court of Appeal ruled that only BNY and Mellon Financial Corporation, not the roughly twenty other affiliates that existed when the contract was signed, were ever entitled to access SS&C's data.
The damages issue proved considerably more contested, and it is what ultimately brought the case before the Supreme Court. SS&C claimed it had suffered US$889,752,087 in total damages, a figure that included a hefty late payment fee compounding at 1.5 percent monthly under the contract. BNY, unsurprisingly, argued SS&C deserved nothing, or at most a nominal amount, pointing out that data pricing had fallen over time and that SS&C's damages theory was unrealistic. Faced with these wildly divergent positions, the trial judge crafted his own middle-ground solution, which he called a "rateable approach".
The trial judge found that BNY could only account for 44.6 percent of the data it had received from SS&C. He then inferred that the unaccounted-for 55.4 percent had been shared with unauthorized entities and ordered BNY to pay for that unaccounted-for share at what he believed to be the contract's standard rate. That math produced a damages award of US$5,696,850, a figure dramatically smaller than what SS&C sought. The Court of Appeal largely left this approach untouched, though it did trim one component of the award tied to CIBC Mellon's usage after finding it amounted to double-counting money SS&C had already been paid under a separate agreement.
Why the Supreme Court Sent the Case Back – Mandatory Adverse Inference
The court agreed that BNY had committed spoliation but found that both the trial judge and the Court of Appeal got the consequences of that spoliation wrong. Once spoliation is established, the Court explained that there is no room for judicial discretion about whether to draw an adverse inference at all. Judges must draw one, and that inference has to be strong enough to genuinely fill the gap left by the missing evidence. What is discretionary, however, is the specific shape and size of that inference, since trial judges are best positioned to weigh the surrounding facts and craft a response proportionate to the harm.
The problem in this case was that the trial judge's inferences did not result in any material consequences. The trial judge found that the unaccounted-for data had been used by unauthorized BNY entities, and that this use was not minimal, but the Court found these were essentially restatements of what the existing evidence already showed rather than genuine responses to the evidentiary vacuum BNY had created. In the Court's words, the inferences "failed to level the evidentiary playing field and failed to provide a factual basis on which the trial judge could award damages".
Compounding that problem, the rateable approach used to calculate the actual dollar figure for the damages was, in the Court's assessment, fundamentally disconnected from the facts of the case. It ignored how many entities may have accessed the data and how many times each one used it and instead rested on an arbitrary assumption that all sixty-five unauthorized users would have paid BNY's original contract rate, an assumption unsupported by the evidentiary record. The Court noted a particularly damning inconsistency: the trial judge's own findings recognized that up to sixty-five entities had improperly used the data over many years, yet that finding played essentially no role in the final damages number. A damages award, the Court emphasized, does not need to be perfect, since courts often have to do the best they can with imperfect evidence. But it does need to be tethered to the facts actually found, and here it was not.
The Court was also careful to reject SS&C's more aggressive argument, which asked the justices to adopt a rule that whenever spoliation is proven, the wrongdoer should automatically be on the hook for the non-spoliating party's maximum theoretical damages claim. The Court declined to go that far, siding with BNY's position that remedies for spoliation should remain flexible and tailored to the specific facts of each case. The practical effect is that a finding of spoliation compels a genuine evidentiary reckoning with the missing evidence but does not entitle the non-spoliating party to automatic maximum recovery.
What Happens Next?
Despite the Supreme Court's finding that the damages calculation was legally flawed, it did not reassess damages but instead sent it back to the trial judge for reassessment. The trial judge was ordered to draw proper mandatory adverse inferences that actually account for the scope and impact of BNY's spoliation, resulting in concrete factual findings about how many entities accessed the data and how many times each one used it. Those factual findings, along with proper pricing information, must then be used to calculate an actual damages figure.
The Bigger Picture: Why This Case Matters Beyond Banking and Data
This decision confirms that once a party is on notice that litigation is coming, ignoring a preservation request because you personally disagree with the merits of the claim against you is never a valid excuse, no matter how sophisticated or well-resourced. The Court went out of its way to note that this kind of conduct is especially troubling when it occurs between large, well-advised institutions that should know better and have parameters in place to preserve evidence.
At the same time, the ruling should reassure defendants that a spoliation finding is not an automatic path to a windfall judgment for the other side. Courts still must connect any damages award to actual, demonstrable facts, even when those facts have to be reconstructed through adverse inferences rather than direct evidence.
The doctrine of spoliation was also constrained by the Supreme Court’s decision to leave to another day whether spoliation is a tort for which one can be sued in and of itself, absent some other legal wrong.
For businesses that rely on licensing arrangements, particularly involving proprietary data or intellectual property distributed among complex corporate families, the case also serves as a cautionary tale about how ambiguity in contract drafting around terms like "Client" or "affiliate" can snowball into decades of undetected non-compliance. And for litigators, it is now the leading modern authority on exactly how to prove spoliation, what remedy naturally follows, and how a damages methodology must remain grounded in the record even when key evidence has been intentionally destroyed.
amount of damages ultimately owed to SS&C remains undetermined and will depend on the factual findings the Superior Court makes on remand. The governing legal principle, however, is now settled: a party that destroys or withholds evidence in the face of contemplated or ongoing litigation will not be permitted to benefit from the resulting evidentiary gap. An adverse finding against a spoliator is mandatory. However, a finding of spoliation does not relieve the innocent party of the requirement to establish damages that are connected to the evidentiary record as reconstructed through mandatory adverse inferences.
Birpal Benipal is a member of Blaneys’ Commercial Litigation Group and Restructuring & Insolvency subgroup.