|
Competition - Discovery. Google Canada Corporation v. Canada (Commissioner of Competition)
In Google Canada Corporation v. Canada (Commissioner of Competition) (Fed CA, 2026) the Federal Court of Appeal dismissed an appeal, this brought against a discovery refusal ruling within "proceedings in the Competition Tribunal concerning some of the appellants’ online advertising services".
Here Stratas JA waxes large on the purpose and practice of 'discovery' (particularly proper questions in light of their "relevance, materiality and proportionality"), here in a federal tribunal context:[1] The Commissioner of Competition brought proceedings in the Competition Tribunal concerning some of the appellants’ online advertising services. The Commissioner says they are anti-competitive, contrary to section 79 of the Competition Act, R.S.C. 1985, c. C-34.
[2] Discoveries have taken place. During the discoveries, certain questions were refused on the basis of relevance and litigation privilege, among other things. The Tribunal (per Andrew D. Little, Chairperson) ruled on the refusals: 2026 CanLII 29605 (CT), 2026 Comp. Trib. 17. The appellants now appeal from that ruling.
[3] Discoveries in competition proceedings under the Act are governed by the same principles that apply to civil proceedings in the Federal Court: see the Tribunal’s decision at para. 6; see also Competition Tribunal Rules, S.O.R./2008-141, Rule 34, which incorporates Rules 3, 240 and 242 of the Federal Courts Rules, S.O.R./98-106. This Court has seldom discussed those principles. For a rare example, see the brief discussion in Canada v. Lehigh Cement Limited, 2011 FCA 120 at paras. 33-37. To resolve the issues in this appeal, a little more needs to be said.
[4] Discoveries are not an end in themselves; they are a means to an end. That end is the fair, efficient and informed determination of a dispute.
[5] When done well, discoveries separate the wheat from the chaff. Evidence supporting allegations with possible merit comes to light; allegations without support are cast aside. Real issues stay on the table; the rest go in the bin. Adjournments, delays and costly unproductive trial days become less likely. Procedural fairness, especially knowledge of the case to meet, is enhanced. Settlement can become more likely because the parties have better information about the dispute. And if the trial runs its full course, the judge’s fact-finding is more likely to be accurate.
[6] But here, too much of a good thing can be bad. If relevance—whether the evidence sought has any tendency to make an allegation more or less probable—alone governed every question and production request during discoveries, parties would have to rummage through files scattered across many offices. They would have to scrounge for anything logically related to an issue, no matter how tangential or minor, and produce it. Re-examinations, further searches and still more productions would follow. In many cases, the insights gained would be too slight to justify the time and expense.
[7] Thus, relevance alone cannot govern whether a discovery question or a production request is proper. Materiality and proportionality must also be brought to bear.
[8] Materiality asks whether the evidence sought is aimed at something the law cares about in the case. Given the pleadings, the elements of the claim or defence, and the governing substantive law, does anyone really need to prove or disprove a particular fact?
[9] Proportionality asks whether the evidence sought is worth the cost and burden it imposes. Information may be logically relevant and material to a pleaded claim or defence and still be off-limits if the expense, delay, or intrusion is out of line with what the case actually needs.
[10] Discovery questions that go too far often fall into three categories:The question is only logically relevant to a fact that is not material, a fact of no real consequence under the pleadings.
The fact is material, but the particular question is so remote that it will not significantly help determine the issue, or it is not proportional to the needs of the case.
The cumulative burden of the whole line of inquiry—searching for, collecting, reviewing, and producing megabytes or even terabytes of electronically stored information, along with other relevant considerations such as privacy—outweighs its likely benefit, even if individual questions, taken alone, could be defended. No doubt there are other categories too.
[11] Sometimes questions seeking a party’s legal positions are proper. They can further the purposes of discovery, many of which are set out in paragraphs 4 and 5 above.
[12] Though an overbroad discovery has its pitfalls, so does an overly restrictive approach. Undue restrictions on the scope of discovery can lead to serious problems, delays or even injustice at trial. See Canada v. CHR Investment Corporation, 2021 FCA 68 at para. 20, citing Montana Band v. Canada, 1999 CanLII 9366 (FCTTD), [2000] 1 F.C. 267 at para. 5. Thus, in cases of real doubt, it is often best for a regulator or court to allow a discovery question to be asked.
....
[14] Reading the Tribunal’s reasons properly (see Mahjoub v. Canada (Citizenship and Immigration), 2017 FCA 157, [2018] 2 F.C.R. 344 at paras. 69 and 153, Canada v. South Yukon Forest Corporation, 2012 FCA 165 at paras. 49-51, and Halton (Regional Municipality) v. Canada (Transportation Agency), 2024 FCA 122 at paras. 23-25), we see that, in careful, complete and logical reasons concerning the refusals, the Tribunal was alive to accepted principles of discovery. Throughout its reasons, sometimes explicitly, sometimes implicitly, the Tribunal considered relevance, materiality and proportionality—principles baked into Rules 3, 240 and 242 of the Federal Courts Rules. It applied these principles without any palpable and overriding error.
|
