You Don't Have the Power: Securities Investigations in Québec

In the context of an ongoing investigation of the embattled engineering firm, SNC-Lavalin, Québec's securities regulator compelled an executive to produce certain documents. In the same letter, however, the regulator purported to prevent the executive from telling anyone else about the documents (apart from the company's lawyers). Revealing the existence of an ongoing investigation was permitted, but not any details of the requested documents.

After some back and forth, the prohibition on disclosure was modified somewhat, so as to allow the company to comply with the request. However, the company was forbidden from communicating the details of the requested documents to its auditors. The auditors cried foul, arguing that they could not sign off on the company's books without knowing the details of the investigation, which they could piece together once they had identified the requested documents. On (internal) appeal to the Bureau de décision et révision, the company won, and the auditors had the necessary information (almost) at their fingertips, at which point the regulator too cried foul and appealed to the ordinary courts.

The saga came to an end yesterday (barring the granting of leave to appeal to the Supreme Court of Canada), with defeat for the regulator in Autorité des marchés financiers c. Groupe SNC-Lavalin Inc., 2013 QCCA 204.

Interestingly, the decision did not turn on the legality of the prohibition on disclosure as such. Section 245 of the Québec Securities Act gives it the power to impose confidentiality obligations in the context of an investigation. Rather, the substantive issue was whether the Bureau had the authority to modify the prohibition. If not, the executive should have sought judicial review in the ordinary courts (which would, at the very least, have applied a standard of review much more congenial to the regulator).

Here, the key provision was s. 322 of the Act, which gives the Bureau the jurisdiction to review any "decision rendered by the [regulator]". One might think, as the regulator argued, that "decision" means a final decision reached after investigation, consideration of the evidence, representations of the parties and adjudication. The imposition of a prohibition in the context of an ongoing investigation, though clearly the sort of coercive action that is subject to judicial review, is not of the same nature. But the Bureau thought differently.

The regulator argued that this was a jurisdictional determination by the Bureau, which should be subject to review on a standard of correctness. However, noting the narrow scope given by the Supreme Court of Canada to jurisdictional questions, Dalphond J.A. held that the Court of Appeal had to defer to the Bureau's interpretation of the term "decision":

[55]        En d'autres mots, la rédaction de l'art. 322 L.V.M. et une analyse de l'ensemble de la L.V.M. peuvent raisonnablement donner lieu à l'interprétation retenue par le Bureau que l'émission d'une ordonnance de confidentialité constitue une décision au sens de la L.V.M. et que le refus de la modifier et de fournir des renseignements additionnels aux vérificateurs constitue une décision de l'Autorité au sens de la L.V.M.

The next issue was the reasonableness of the Bureau's decision to modify the prohibition on disclosure (thereby allowing the auditors to know the details of the requested documents). Before the Bureau, the regulator had confined its submissions to questions of competence and had not explained the likely effects on its investigation of any modification to the prohibition on disclosure. But this made it difficult to subsequently attack the reasonableness of the modification before the ordinary courts:

[68]        Je souligne que la position de l'Autorité devant le Bureau était axée sur son absence de compétence et qu'elle n'a pas voulu indiquer quelles pourraient être les conséquences sur son enquête de la communication en toute confidentialité de la correspondance entre elle et l'intimée aux seuls vérificateurs et aux membres du comité de vérification.

[69]        Il ressort par contre de la preuve de l'intimée devant le Bureau que, conformément aux pratiques suivies depuis des années entre la société, représentée par son comité de vérification, et ses vérificateurs et aux lettres d'engagement de ces derniers, les projets de réponses doivent être communiqués aux vérificateurs pour revue. En l'absence de toute preuve relative à des risques de collusion ou d'entrave, le Bureau a permis, dans ce contexte, la continuation des pratiques établies et le respect des termes de la lettre d'engagement.

[70]        Dans ces circonstances, rien ne permet de dire que les modifications apportées à l'ordonnance de confidentialité ne sont pas raisonnables.

One curious omission from Dalphond J.A.'s judgment is the standard of review that the Bureau applied to the prohibition on disclosure. One would think that the Bureau should not intervene too lightly in the decisions of the investigators as to when disclosure of information should be prohibited. Some internal deference is surely warranted. I suppose that the regulator's position before the Bureau might have militated against the Court of Appeal making a finding on this point, but it remains an important issue.

The case was heard on an accelerated timetable because the auditors have to sign off on the company's accounts early this year. That may make Supreme Court review unlikely, but given the importance of the issues -- jurisdictional questions, agencies' ability to impose constraints on disclosure, and standard of review within administrative structures -- the regulator may wish to push for leave to appeal.