Regulatory Searches and Seizures in the Digital Age: Binance Holdings Limited v. Ontario Securities Commission, 2025 ONCA 751

In Canadian law, the protections of the Charter of Rights and Freedoms apply with relatively little force in areas of economic regulation. The following passage from R. v. Wholesale Travel Group Inc., [1991] 3 SCR 154 is representative of the courts’ concern that giving too expansive a scope to the Charter might be socially damaging:

Regulatory legislation is essential to the operation of our complex industrial society; it plays a legitimate and vital role in protecting those who are most vulnerable and least able to protect themselves.  The extent and importance of that role has increased continuously since the onset of the Industrial Revolution.  Before effective workplace legislation was enacted, labourers — including children — worked unconscionably long hours in dangerous and unhealthy surroundings that evoke visions of Dante’s Inferno.  It was regulatory legislation with its enforcement provisions which brought to an end the shameful situation that existed in mines, factories and workshops in the nineteenth century.  The differential treatment of regulatory offences is justified by their common goal of protecting the vulnerable.

Successful challenges to regulatory schemes or enforcement action that impact Charter rights are few and far between (see, e.g., here and here for some unsuccessful Supreme Court challenges). But a recent challenge was successful and may be the harbinger of significant developments in this area of the law. In Binance Holdings Limited v. Ontario Securities Commission, 2025 ONCA 751, the Ontario Court of Appeal held that a summons issued by the Commission violated s. 8 of the Charter, because it was overbroad and amounted to “a fishing expedition of the entire business in a speculative search for documents where there is no reasoned basis for believing that they may be relevant to the inquiry that is being undertaken” (at para. 98).

I will begin with some background on s. 8 before introducing the facts of the case. I will then discuss Paciocco JA’s s. 8 analysis before concluding with some thoughts on the procedural posture of this case (including the relevance of the Doré requirement to balance statutory objectives against Charter rights).

Section 8 and Regulatory Searches and Seizures

Section 8 of the Charter provides protection against unreasonable searches and seizures. Indeed, it guarantees “a broad and general right to be secure from unreasonable search and seizure” (Hunter et al. v. Southam Inc., [1984] 2 SCR 145, at p. 158). Hunter is the high watermark of judicial enforcement of s. 8 in a regulatory context. For it contained within it the roadmap for a retreat from that high watermark. Dickson J noted that reasonableness requires “an assessment as to whether in a particular situation the public’s interest in being left alone by government must give way to the government’s interest in intruding on the individual’s privacy in order to advance its goals, notably those of law enforcement”.

In Thomson Newspapers Ltd. v. Canada (Director of Investigation and Research, Restrictive Trade Practices Commission), [1990] 1 SCR 425, La Forest J leveraged context-sensitivity to fundamentally reorient the principles of s. 8 as they apply in a regulatory context:

But the degree of privacy the citizen can reasonably expect may vary significantly depending upon the activity that brings him or her into contact with the state.  In a modern industrial society, it is generally accepted that many activities in which individuals can engage must nevertheless to a greater or lesser extent be regulated by the state to ensure that the individual’s pursuit of his or her self‑interest is compatible with the community’s interest in the realization of collective goals and aspirations.  In many cases, this regulation must necessarily involve the inspection of private premises or documents by agents of the state.  The restaurateur’s compliance with public health regulations, the employer’s compliance with employment standards and safety legislation, and the developer’s or homeowner’s compliance with building codes or zoning regulations, can only be tested by inspection, and perhaps unannounced inspection, of their premises.  Similarly, compliance with minimum wage, employment equity and human rights legislation can often only be assessed by inspection of the employer’s files and records. It follows that there can only be a relatively low expectation of privacy in respect of premises or documents that are used or produced in the course of activities which, though lawful, are subject to state regulation as a matter of course. 

As Wilson J put it in R. v. McKinlay Transport Ltd., [1990] 1 SCR 627, at p. 645: “Since individuals have different expectations of privacy in different contexts and with regard to different kinds of information and documents, it follows that the standard of review of what is “reasonable” in a given context must be flexible if it is to be realistic and meaningful” (see also British Columbia Securities Commission v. Branch, [1995] 2 SCR 3, at paras. 55-63).

Regulatory statutes typically now provide simply that an officer may inspect a premises (which includes examining equipment, books and records, and taking samples) where the officer has reasonable grounds to suspect regulatory non-compliance. Only where a private dwelling is targeted, access to a premises is refused or a premises can only be accessed by use of force is it necessary to apply for a warrant (see e.g. Canadian Environmental Protection Act, 1999, SC 1999, c 33, s. 218).

It is theoretically possible that a regulatory search or seizure might violate s. 8 but in practice a violation will be hard to establish. To begin with, the test for a reasonable expectation of privacy will be difficult to meet. There are four components: “(1) an examination of the subject matter of the alleged search; (2) a determination as to whether the claimant had a direct interest in the subject matter; (3) an inquiry into whether the claimant had a subjective expectation of privacy in the subject matter; and (4) an assessment as to whether this subjective expectation of privacy was objectively reasonable, having regard to the totality of the circumstances” (R. v. Cole, 2012 SCC 53, [2012] 3 SCR 34, at para. 40). Evidently, as long as the officer is searching or seizing things related to regulated activity, criteria (1), (3) and (4) will favour a conclusion that the expectation of privacy is minimal at best. This is certainly borne out by the case law on regulatory searches and seizures.

Next, a search “will be reasonable if it is authorized by law, if the law itself is reasonable and if the manner in which the search was carried out is reasonable” (R. v. Collins, [1987] 1 SCR 265, at para. 23). Given that regulatory statutes authorize searches and that such statutes will generally be found to be reasonable in s. 8 terms (see e.g. Sazant v. College of Physicians and Surgeons of Ontario, 2012 ONCA 727; Law Society of Alberta v Sidhu, 2017 ABCA 22; A Lawyer v. The Law Society of British Columbia, 2021 BCCA 437), in most cases it is only where a regulatory search was carried out in an unreasonable manner that a constitutional violation might occur. This, the Court of Appeal concluded, was what happened in Binance.

The Binance Case

Binance is a crypto asset trading platform. It claims to be the largest such platform in the world. The Commission made a general announcement in 2021 that it considered that it had jurisdiction over crypto asset trading platforms and required any platforms trading in Ontario to enter into compliance discussions. The Commission contacted Binance, who responded that they intended to shut down operations within Ontario by the end of 2021. This did not quite happen and some negotiations between the Commission and Binance followed.

In 2023, the Commission became aware of enforcement action being taken against Binance by American regulators on the basis that the company was circumventing compliance controls. Commission staff feared that Binance was doing something similar in Ontario. The Commission commenced an inquiry under s. 11 of the Securities Act, RSO 1990, c S.5. Invoking the compulsion power in s. 13 of the Securities Act,the Commission made four demands:

• Confirm the total fees and other revenue earned on all Ontario accounts since the inception of Binance.com;
• Confirm the number of Ontario accounts remaining open on the Binance Trading Platform and the aggregate value holdings in these accounts on various dates in 2021 and 2022
• Provide a description of the methodology used to answer the first two demands; and
• For the period of January 1, 2021 to present, provide all communications regarding Ontario (or Canada generally) among directors, officers, employees, contractors, agents and consultants of Binance Holdings Limited and related entities [in relation to compliance with the Securities Act and its Binance’s undertakings].

As Paciocco JA explained:

This fourth demand went on to provide a wide non-exclusive description of the form of “communications” that could include “e-mails, letters, [and] chats/texts on messaging platforms (e.g. Signal, WhatsApp, Telegram, Slack, etc.)”. The Summons directed that the requested communications “shall include”, but again, not be “limited to”, “those among directors, officers, employees, contractors, agents and consultants of Binance regarding” a list of events and activities. The non-exclusive list of events and activities included account opening and trading procedures, staff training, hiring and the preparation of quarterly reports pursuant to the Undertaking (at para. 14).

It was the breadth of this last demand that Paciocco JA found to be unreasonable:

the Summons demanded production, without limitation, of all communications between virtually anyone that may have managed, been employed by, or done work for either Binance or its related entities over a two-and-a-half-year period relating not only to Ontario but to all of Canada, regardless of the subject matter of those communications. This demand, enforceable by the threat of contempt proceedings, is staggering in its breadth and in my view was made without apparent concern about the relevance of what was being demanded, beyond mere speculation that there could be something relevant that would otherwise be missed (at para. 81).

He summarized his legal analysis as follows:

I am persuaded that the power of an appointed investigator to issue a production order under s. 13 is limited to documents that may be relevant to an inquiry that is properly undertaken pursuant to the Securities Act. A seizure that purports to compel the production of documents in the absence of a reasonable foundation to believe they may be relevant is therefore overbroad and unreasonable, contrary to s. 8 of the Charter (at para. 80).

The difficulty here is that the second sentence does not follow from the first and, indeed, is incompatible with it (see also at para. 96 for a similar formulation). The first sentence is correct in my view: a production order that is within the scope of a valid regulatory inquiry can properly require production of documents that may be relevant to that inquiry. But the second sentence requires “a reasonable foundation to believe” that documents may be relevant.

There are principled and practical reasons to doubt the correctness of the proposition that a regulator needs a “reasonable foundation to believe” that documents may be relevant to an inquiry.

From a principled perspective, the powers of a regulatory body like the Commission are set out in statute. Before availing itself of its power to compel the production of documents, the Commission must first establish a basis for an inquiry. As Paciocco JA recognized, the Commission had done exactly this: “the inquiry that was being undertaken was described by the Commission in commendable detail in the Investigation Order, which set out both the events of interest and the Commission’s concerns about Binance’s regulatory non-compliance” (at para. 81). An inquiry is a condition precedent to a production order: the scope of the inquiry sets the boundaries of what is to be produced, namely anything that is relevant to the inquiry.

The test, however, is relevance, not “a reasonable foundation to believe”. Consider this passage from Paciocco JA’s reasons: “Even in the regulatory context, ‘The material sought must be relevant to the inquiry in progress.’ See Thomson, at p. 530. It will not be if the demands made in the subpoena are broader than needed for the purposes of that inquiry” (at para. 91). But this is a gloss on what was said in Thomson, not an interpretation of it: Thomson states that relevance is the test but Paciocco JA adds that the demands can be no “broader than needed” for the purposes of the inquiry — this introduces an element of necessity and, elsewhere in his reasons, Paciocco JA imposes a requirement for a “reasonable foundation to believe” that documents may be relevant. (We can, I appreciate, argue about whether necessity arises from Thomson, where there was no clear majority on the point, but in all events, the “reasonable foundation to believe” concept is a novelty in the regulatory context.)

In short, the statute sets the bounds of what the Commission may inquire into, the exercise of the inquiry power sets the bounds of any summons that the Commission may issue, and the summons will be reasonable if it seeks documents relevant to the inquiry.

Here, the Commission considered that it could inquire into whether Binance’s platform was being operated in accordance with the Securities Act; it exercised its inquiry power by setting out a detailed factual basis for its concerns; and its summons sought documents relevant to the question of whether Binance was complying with applicable statutory and regulatory requirements. The summons was undoubtedly broad, but how could it be otherwise? The gravamen of the Commission’s case was that regulated entity was not complying with its obligations, which is precisely why the Commission has expansive powers of compulsion, and the means of carrying out its inquiry had to be the production of records relevant to compliance. In a situation of alleged non-compliance, the scope of what is relevant will be broad. Its (inevitable) breadth cannot make it unconstitutionally overbroad, as the inquiry is precisely what is mandated by the Securities Act, and there is no doubt that the Securities Act itself complies with s. 8 of the Charter. There is nothing “unreasonable” about the “manner” in which the search was conducted (Collins, at para. 23), as it flowed from the statutory scheme. A search might become unreasonable if the Commission executed the search abusively, negligently, or in a way that would impinge upon protected interests (see, e.g. for a recent example, albeit in a criminal context, R. v. Fox, 2026 SCC 4) but a broad search is not per se unreasonable in the regulatory context.

Practically speaking, as the Commission rightly argued, it will often be impossible to know whether a document is relevant to a regulatory inquiry without having actually seen the document. Paciocco JA recounted and rejected this argument in the following passage:

the Commission argued that it cannot know if a document is relevant without seeing it, and it relies on Wilson J.’s statement in McKinlay Transport Inc., at p. 650, that, “The Minister has no way of knowing whether certain records are relevant until he has had an opportunity to examine them.” I accept that it will be impossible for the Commission to identify whether a particular document is relevant before examination, and that a strict relevance standard therefore cannot be imposed on regulatory production orders. But there is no reason why regulators cannot target their inquiries to categories of documents that are reasonably likely to contain relevant documents or otherwise describe the scope of their inquiry in a way that would enable the regulated party and the courts to identify relevant documentation. Wilson J.’s comment cannot, in the context of her decision, be read as supporting the Commission’s position that there is no relevance requirement in regulatory cases where production orders are used. Taking that view would sanction “fishing expeditions”, which the Thomson court made clear could not be countenanced. Moreover, as I have explained, Wilson J. stressed in McKinlay that s. 231(3) of the Income Tax Act had been interpreted narrowly to authorize only limited seizures of documents that “may be relevant”. It is evident that when read in context, Wilson J. offered this comment as an explanation for why a “may be relevant standard” is appropriate in the regulatory context, not as a rejection of any relevance limitation.

It is not clear to me that the Commission’s position was “that there is no relevance requirement”, so much as that the scope of relevance was determined by the lawful exercise of the Commission’s inquiry powers and that a broad array of documents were relevant to alleged non-compliance. True, many of (say) the messages sent by Binance officers, employees, etc on social media platforms might turn out not to be relevant (in which case, for the search to be performed reasonably, the Commission would have to discard the messages). But that hardly means that the messages could not be relevant. And, in any event, the summons did seem (based on the summaries available in the reported decisions) to specify what exactly the Commission was interested in, certainly enough to permit Binance “to identify relevant documentation”.

For these principled and practical reasons, then, I think the imposition of a “reasonable foundation to believe” standard that is not provided for in a constitutionally valid regulatory statute is problematic.

Now, all of that said, it is worth considering whether Paciocco JA’s test is normatively appropriate. Although Paciocco JA does not approach the issue in these terms, it is appropriate to engage in debate and discussion about how apt the 1980s and 1990s jurisprudence on regulatory searches is in the contemporary context. At the time of the foundational decisions on s. 8, regulatory searches were mostly confined to searches of physical spaces and seizures of physical documents: there were clear boundaries between workplaces and private homes. Regulatory searches in that era would rarely run up against privacy concerns. Today, however, regulatory searches are no longer physical but electronic. Recent decades have also seen the collapse of clear boundaries between home and office, between personal and work communication devices and between personal time and work time — employees often work remotely, use devices for professional and personal purposes, and are reachable at all times electronically regardless of their location. As such, a regulatory search in 2026, seeking the same information about compliance as a regulatory search in 1986, sweeps much more broadly. And, in other contexts, the Supreme Court of Canada’s jurisprudence on digital privacy has limited the scope of investigative powers. Even, therefore, if Paciocco JA’s analysis is novel in this particular context, it might be a creature of its time.

Interestingly, to the extent Paciocco JA engages such general considerations about social and legal evolution, he comes down in favour of the status quo. Notably, he rejected the proposition that Binance had a relatively higher expectation of privacy because of the enmeshment of the personal and professional:

I reject Binance’s supplementary claim that its reasonable expectation of privacy is heightened in this case because the demand in the Summons includes chats and texts on social media platforms which may include personal messages sent or received using these platforms. It is true that there is a higher reasonable expectation of privacy in personal documents; however, Binance has done nothing to show that any such personal messages were sent or received. Its claim that they were is a bald assertion. Moreover, I agree with the Commission that a regulated party cannot be permitted to increase the intensity of its expectation of privacy in its business platforms by allowing them to be used for personal purposes (at para. 39).

Be that as it may, Paciocco JA was certainly concerned about overbreadth. As I will discuss in the next section, the procedural posture of this litigation was suboptimal, featuring arguably less oversight than desirable, which makes Paciocco JA’s underlying concerns understandable.

Procedural Problems

In a perfect world, the Commission’s exercise of its authority would be subject to review or appeal at various stages, in particular when launching an inquiry and when issuing a summons. Now, the form of review or appeal has to be adapted to the context — one would not want several years of tribunal or court wrangling over preliminary matters. But there should be some form of review or appeal about whether the Commission’s powers are properly exercised.

In this case, remarkably enough, Paciocco JA was the first judge to consider Binance’s s. 8 arguments, despite almost half a dozen separate proceedings before the Capital Markets Tribunal (which concluded it did not have jurisdiction) and the Divisional Court. It is entirely understandable that Paciocco JA would consider it appropriate to make a final determination on the s. 8 question, Binance having waded gamely and at considerable expense through a procedural quagmire. Equally, though, Paciocco JA was in a difficult position not having had the benefit of full argument and analysis before lower courts. Indeed, he held that the Commission properly found that it did not have the authority to vary or revoke a summons issued under s. 13 (at paras. 44-53). I do not have a strong view on whether this interpretation is correct, but even if it is the resulting situation is quite unfortunate, as it means that a court considering an alleged s. 8 breach by the Commission will come to the matter cold, without the benefit of an adjudicative-type decision from the Commission or the Tribunal. A “reasonable foundation to believe” standard might be easier to swallow if there were an efficacious means of ventilating the relevant facts and argument about the application of the standard. This is certainly a matter for potential legislative reform (at para. 52).

This unfortunate procedural posture created a further complication. Binance argued that the Doré framework applied: before issuing a summons, the Commission should have balanced the values underlying s. 8 of the Charter with its statutory objectives. Any failure to do so would, on Binance’s argument, mean that the summons was invalid. Paciocco JA rejected this argument:

In my view, the issues before us are not governed by the Doré test but by the legal standards established under s. 8 of the Charter, which provide a dedicated methodology for assessing the constitutional validity of searches and seizures. This s. 8 jurisprudence already balances competing interests and addresses the proportionality issue through its assessment of the intensity of the reasonable expectation of privacy and its relevance inquiry. I do not understand Doré to have overtaken this body of law by substituting a looser deferential examination of the proportionality of the balance that has been achieved between the Charter implications of the discretionary decision to exercise the power of search and seizure and the competing administrative interests. In my view, any state agent who authorizes or exercises a search or seizure is bound to comply correctly with the requirements of s. 8 of the Charter, even when exercising a statutory discretion. It would not be fitting to hold that even though the s. 8 requirements have not been met, it was reasonable for the party authorizing the seizure to strike the balance they did. The Doré standards add nothing to the instant case and can only confuse the law if applied in this context (at para. 116).

I have argued (see here) that Doré has a procedural and substantive aspect. The substantive aspect is that a decision-maker should comply with Charter rights. Paciocco JA correctly observes that any summons issued by the Commission has to comply with s. 8. But there is an additional, procedural aspect to Doré, namely that the decision-maker should give due consideration to Charter values in the exercise of a statutory power. Here, that would mean that the Commission would have to demonstrably turn its mind to the values underpinning s. 8, such as the protection of private communications that might be swept up in a regulatory search, and consider whether these were outweighed by the statutory objectives. If this was Binance’s argument, then it was correctly advanced in my view. It is also the best place, I would suggest, to consider whether a regulatory search sweeps too broadly in view of its implications for the personal privacy interests of Binance employees whose communications might have to be disclosed. Again, however, the procedural posture obscured the issue: in the absence of any detailed analysis by the Commission itself or by the Tribunal, it is understandable that Paciocco JA would look askance at the Doré argument.

Conclusion

The reasonableness of administrative searches and seizures rarely comes before the courts. Paciocco JA’s reasons make a strong case for a more robust judicial approach in the s. 8 context. To return to a theme I touched on above, the dawn of the digital era (and the consequent evolution of s. 8 in the criminal context) poses difficult questions for debate and discussion. (1) Should regulated entities’ reasonable expectations of privacy be readjusted upwards given how employees’ professional and personal lives have become enmeshed? (2) Should the reasonableness test for the validity of a statutory inquiry power be more demanding, in terms of statutory limitations and oversight powers, given evolution in the workplace and in the law of s. 8? (3) Should each and every regulatory inquiry that rubs up against private interests receive close judicial scrutiny? (4) Should each inquiry require a balancing-type analysis before any summons is exercised?

Paciocco JA said “No” to (1) and (4), but “Yes” to (3). My preference is to answer “Yes” to (4). Perhaps the most interesting question, though, is (2), which did not arise on the facts of this case. Maybe our regulatory statutes need to be updated for the digital era. And maybe the analysis in this case will provide an impetus to do so.

Further reading

For more on this topic, I commend to you Professor François Tanguay-Renaud’s paper on the Supreme Court decision in Bykovets.

This content has been updated on February 22, 2026 at 20:58.