Citation: 2019 FC 281
CANADIAN NATIONAL RAILWAY COMPANY
Defendant by Counterclaim
BNSF RAILWAY COMPANY
Plaintiff by Counterclaim
REASONS FOR ORDER
 The plaintiff by counterclaim, BNSF Railway Company (BNSF), moves for a protective order. The defendant by counterclaim, Canadian National Railway Company (CN), consents to the terms of the proposed protective order. Despite the parties’ agreement, I dismissed the motion with reasons to follow (Order dated February 13, 2019). These are the reasons.
 CN commenced a patent infringement action against BNSF in 2017. As part of its defence, BNSF counterclaimed alleging invalidity of the patents in suit. CN has since discontinued its main claim in this action, but BNSF’s counterclaim continues.
 The discovery phase of the present action is about to begin. This phase will involve the exchange of documents and information between the parties. As is common in actions involving patents, some of the documents and information to be exchanged are confidential. In addition, a subset of such confidential material is of such sensitivity that disclosure thereof to the other party could reasonably be expected to cause competitive or commercial harm. This is a particular concern in a case such as this where the parties are direct competitors.
In anticipation of the discovery process, the parties discussed and agreed on terms for a protective order that would define two levels of confidential documents and information,
“Confidential Information” and
“Confidential Information – Counsel’s Eyes Only,” and that would prescribe how a party receiving such documents and information may deal therewith. BNSF then submitted the draft protective order to the Court as part of its motion record. CN subsequently submitted its own motion record providing additional representations in support of BNSF’s motion. In addition, prior to the hearing of the motion, BNSF provided a supplemental motion record with supplementary written representations.
 The hearing of the motion was somewhat unusual because both parties argued the same side of the debate. No one argued against the motion.
II. CONFIDENTIAL NATURE OF THE INFORMATION IN ISSUE
 The facts relevant to this motion are unremarkable. It is not at all surprising that the parties should wish to take measures to protect their confidential information in the context of discovery. I readily accept that confidential information will be exchanged during discovery. I also readily accept that some such information will be so sensitive that it will merit the stricter treatment that is contemplated for information that is designated Counsel’s Eyes Only.
 It is also not unusual for the parties to agree on terms for a protective order and make a request on consent for its issuance by the Court. Motions on consent to issue protective orders have traditionally been granted by this Court, especially in patent actions. Recent jurisprudence, however, has put that practice into question. These recent developments are discussed in the next section.
III. LAW APPLICABLE TO ISSUANCE OF PROTECTIVE ORDERS
 Two recent decisions are of particular note concerning the issuance of protective orders: Live Face on Web, LLC v Soldan Fence and Metals (2009) Ltd, 2017 FC 858 [Live Face on Web], and Seedlings Life Science Ventures LLC v Pfizer Canada Inc, 2018 FC 443 [Seedlings 1]. In both of these decisions, Prothonotary Mireille Tabib provided a history of the use of protective orders, a discussion of the current legal test for issuance of a protective order, and a description of the implied undertaking rule which applies even in the absence of a protective order. The implied undertaking rule is described in the next section. In both Live Face on Web and Seedlings 1, Prothonotary Tabib refused to issue the requested protective order. Importantly, Seedlings 1 was reversed by Justice Shirzad S. Ahmed in Seedlings Life Science Ventures LLC v Pfizer Canada Inc, 2018 FC 956 [Seedlings 2]. Justice Ahmed issued the requested protective order.
 I accept BNSF’s assertion that the legal test for the issuance of a protective order is found in the decision of the Supreme Court of Canada in Sierra Club of Canada v Canada (Minister of Finance), 2002 SCC 41 [Sierra Club]. However, there appears to be some doubt as to what the test is.
 Before discussing the legal test in Sierra Club, it is apt to clarify some nomenclature. As used herein, consistent with modern use, a protective order is an order which prescribes the treatment of confidential information, but does not provide for the filing of confidential information with the Court. A confidentiality order is one that does address the filing of confidential information with the Court. There are also hybrid orders which have provisions that are characteristic of both protective orders and confidentiality orders.
 The order sought in the present motion is referred to herein as a protective order because it concerns mainly the treatment of confidential information not filed with the Court. However, I note that two paragraphs of the proposed protective order do provide for the filing of confidential information with the Court. These two paragraphs are not material to this decision since, even if I had been willing to issue the requested Order in the present motion, I would have excluded these two paragraphs. The question of filing documents under seal may be addressed at the appropriate time. The Court is generally willing to issue an order permitting the filing of documents under seal for the limited purpose of motions to compel answers to questions refused on discovery. This subject can be discussed with the case management judge in due course.
 Returning now to the legal test for the issuance of a protective order, Justice Ahmed in Seedlings 2 rejected a distinction that was drawn by Prothonotary Tabib in Seedlings 1 between the test for issuance of a confidentiality order and the test for a protective order. Prothonotary Tabib was of the view that only confidentiality orders (and not protective orders) were contemplated in Sierra Club. Justice Ahmed found nothing in Sierra Club to support Prothonotary Tabib’s distinction, and concluded that Sierra Club applies to both types of orders. At paragraph 26 of Seedlings 2, he stated:
… The rationale behind the Sierra Club test, whether one is considering a confidentiality order, a protective order, or a hybrid order, is the same: that is, the protection of sensitive information – whether from the general public or other business adversaries – from abuse or use in activities collateral to the litigation. In this sense, the Prothonotary’s attempt to read Sierra Club as exclusive to confidentiality orders is effectively an exercise in splitting hairs.
 This passage gives the indication that Justice Ahmed felt that the same test should apply whether the request is for a confidentiality order, a protective order, or a hybrid order. However, it appears that Justice Ahmed actually applied a different test within Sierra Club for protective orders.
 At paragraph 53 of Sierra Club, the Supreme Court of Canada stated:
… A confidentiality order under Rule 151 [of the Federal Courts Rules, SOR/98-106] should only be granted when:
(a) such an order is necessary in order to prevent a serious risk to an important interest, including a commercial interest, in the context of litigation because reasonably alternative measures will not prevent the risk; and
(b) the salutary effects of the confidentiality order, including the effects on the right of civil litigants to a fair trial, outweigh its deleterious effects, including the effects on the right to free expression, which in this context includes the public interest in open and accessible court proceedings.
The Court added that
“three important elements” are subsumed in the first branch of this test:
(a) The risk in question must be real and substantial, in that the risk is well grounded in the evidence, and poses a serious threat to the commercial interest in question (para 54);
“important commercial interest”in question cannot merely be specific to the party requesting the order; the interest must be one which can be expressed in terms of a public interest in confidentiality (para 55); and
(c) The phrase
“reasonably alternative measures”requires the judge to consider not only whether reasonable alternatives to a confidentiality order (or a protective order) are available, but also to restrict the order as much as is reasonably possible while preserving the commercial interest in question (para 57).
 Justice Ahmed does not seem to have applied this test. Instead, it appears that he adopted an argument by the defendant in that case that the test for protective orders is provided for in paragraph 60 of Sierra Club, reproduced here:
Pelletier J. noted that the order sought in this case was similar in nature to an application for a protective order which arises in the context of patent litigation. Such an order requires the applicant to demonstrate that the information in question has been treated at all relevant times as confidential and that on a balance of probabilities its proprietary, commercial and scientific interests could reasonably be harmed by the disclosure of the information: AB Hassle v. Canada (Minister of National Health and Welfare) (1998), 83 C.P.R. (3d) 428 (F.C.T.D.), at p. 434. To this I would add the requirement proposed by Robertson J.A. that the information in question must be of a “confidential nature” in that it has been “accumulated with a reasonable expectation of it being kept confidential” as opposed to “facts which a litigant would like to keep confidential by having the courtroom doors closed” (para. 14).
 The applicable test, as stated by the defendant in Seedlings 2, and as apparently adopted by Justice Ahmed, was that:
(a) The information has been treated as confidential and, on a balance of probabilities, the disclosing party’s proprietary, commercial and scientific interests could reasonably be harmed by the disclosure of the information; and
(b) The information is of a confidential nature with a reasonable expectation that it be kept confidential. (para 24)
 While this definition of the test considers the importance of the interest at issue, it fails to consider whether the requested order is necessary because reasonably alternative measures will not prevent the risk to that interest. In Seedlings 2, Justice Ahmed did not address this criterion, except, arguably, in obiter dicta, in the last paragraph of his decision. I will discuss that paragraph later.
 In my view, a request for a protective order should be considered using the same criteria as set out in paragraphs 53 and following of Sierra Club for a confidentiality order.
IV. IMPLIED UNDERTAKING RULE
There is no dispute that, in the context of litigation in the Federal Court, there exists a rule whereby documents and information that are exchanged in the course of discovery are received subject to an implied undertaking that such documents and information will not be used or disclosed for any purpose other than the proceeding in which it is disclosed. This is known as the
“implied undertaking rule.” The rationale for this rule is discussed in Juman v Doucette, 2008 SCC 8 at paras 23 and following [Juman].
 There is also no dispute that the implied undertaking is made to the Court (Juman at para 27), and therefore it may be enforced by contempt of court proceedings in the event of a breach (Juman at para 29). The implied undertaking arises even in the absence of any order of the Court.
 As indicated above, I accept that confidential information will be exchanged during discovery in this action. I accept that the criteria as defined by the defendant in Seedlings 2, and as apparently adopted by Justice Ahmed, are satisfied. My principal concern is whether the requested order is necessary because reasonably alternative measures will not prevent the risk to the parties’ interest in that confidential information.
 The reasonable alternative measure at issue here is the implied undertaking that arises automatically, supplemented by the terms agreed to by the parties as set out in the requested protective order. Specifically, I must consider whether, instead of obtaining the requested protective order, the parties’ interests in their confidential information would be adequately protected by an express undertaking to the Court which supplements the implied undertaking. For simplicity, I will refer to such an express undertaking as a protective agreement.
 The parties have raised a number of concerns with reliance on a protective agreement as an alternative to a protective order. It is convenient to address these concerns under the following headings:
- Enforceability of a Protective Agreement;
- Applicability of a Protective Agreement to Third Parties;
- Imprecision and Lack of Certainty in the Scope of the Implied Undertaking Rule;
- Parties’ Discomfort in the Absence of a Protective Order;
- Added Heft of a Court Order; and
- Important Change to Longstanding Practice
A. Enforceability of a Protective Agreement
 Two key concerns expressed by BNSF with the idea of a protective agreement are:
(i) It is not clear that the Federal Court would have jurisdiction to enforce such an agreement (since the Federal Court does not normally have jurisdiction to rule on cases of breach of contract), and
(ii) Even if the Court does have jurisdiction to address a breach of a protective agreement, a separate order of the Court will be required in order to enforce the agreement by contempt proceedings.
 These concerns were addressed by Prothonotary Tabib in Live Face on Web at paragraphs 21-23:
I fail to see why additional undertakings given to but not specifically endorsed by the Court, that aim to achieve the same goals as the implied undertaking rule, should be any less binding on the parties or amenable to enforcement by the Court’s contempt power than the general rule, so long as they are voluntarily given by the parties and their solicitors in the mutual belief that they are lawful and appropriate, in the circumstances, to protect the parties’ legitimate privacy interests during the conduct of the litigation.
The Federal Court’s jurisdiction to enforce compliance with the implied undertaking rule arises from the Court’s inherent power to control its own process. The implied undertaking rule itself was born from the recognition that it would promote compliance with the Court’s discovery process and help prevent abuses or misuses of this process by affording protection for the parties’ privacy interests. The implied undertaking arises automatically and may be enforced as soon as a person receives discovery information, even outside of the presence of the Court and without its knowledge. In that, it is unlike undertakings given by parties in relation to their substantive rights, such as settlement agreements that include an undertaking not to make use of a trademark or other intellectual property. These substantive undertakings cannot be enforced by contempt unless they have been communicated to and acknowledged by the Court in an order (Williams Information Services Corp. v Williams Telecommunications Corp.,  FCJ No 594, 142 FTR 76).
Undertakings that are strictly related to the procedural aspects of litigation and aim to assist in regulating the Court’s process, such as those restricting the number of persons who can have access to designated discovery information or requiring prior notice of an intention to file, are of the same kind as the implied undertaking rule. As such, they should not need to be expressly acknowledged by the Court in order to be amenable to enforcement by the Court as part of its inherent jurisdiction to control its process, including by contempt in appropriate cases.
 I agree entirely with this reasoning. I see no reason for concern that the Federal Court could not enforce an agreement between the parties regarding the control of its own process. Also, I see no reason that a breach of an express undertaking to the Court would be less enforceable by contempt proceedings than a breach of the implied undertaking which it supplements. Moreover, I see no difficulty transforming the terms on which the parties agreed for a protective order into a protective agreement incorporating an express undertaking to the Court.
B. Applicability of a Protective Agreement to Third Parties
 In his obiter dicta at the end of Seedlings 2, Justice Ahmed expressed concern that there is insufficient authority to be confident that the implied undertaking rule applies to third parties. He noted that the single case cited by Prothonotary Tabib in support of this idea, Winkler v Lehndorff Management Ltd (1998), 28 CPC (4th) 323,  OJ No 4462 (QL) (Ont Ct J (Gen Div)) [Winkler], is over 20 years old, from a different court, and unrelated to patents. Though this is true, it should be noted that Winkler was addressing the implied undertaking rule as it exists in common law, and that its reasoning was unimpeachable and equally applicable in patent matters. Winkler cited the Ontario Court of Appeal’s decision in Goodman v Rossi (1995), 24 OR (3d) 359, 125 DLR (4th) 613, which found that the implied undertaking rule extends to third parties, and which adopted the following reasoning:
… The purpose of the undertaking is to protect, so far as is consistent with the proper conduct of the action, the confidentiality of a party’s documents. It is in general wrong that one who is compelled by law to produce documents for the purpose of particular proceedings should be in peril of having those documents used by the other party for some purpose other than the purpose of the particular legal proceedings and, in particular, that they should be made available to third parties who might use them to the detriment of the party who has produced them on discovery.…
 I note also that Winkler was followed on the issue of the implied undertaking rule applying to third parties in St Elizabeth Home Society v Hamilton (City), 130 ACWS (3d) 48,  OJ No 1420 (QL) (Ont Sup Ct) at para 7. Further, a more recent decision of the Ontario Court of Justice, Canadian National Railway v Holmes, 2014 ONSC 593, reaches a similar conclusion, though without reference to Winkler.
 BNSF refers to the decision of the Newfoundland and Labrador Court of Appeal in Power v Parsons, 2018 NLCA 30 [Power], in which the Court found that a third party accountants’ regulatory body that obtained documents concerning the accountant defendant that had been disclosed as part of a discovery process was not restrained from using those documents in a separate investigation of the accountant defendant as part of its mandate.
 White JA, with the concurrence of the other members of the panel in Power on this point, and citing Juman at paragraph 55, stated that the implied undertaking rule does not bind third parties (para 10). It should be noted firstly that the cited passage from Juman was referring to those who are not parties to the implied undertaking, not those who are not parties to the litigation. Therefore, the conclusion in Power would not apply to experts who received discovery materials from counsel after giving an undertaking of the kind contemplated in the draft protective order submitted in the present case.
White JA, still with the concurrence of the other judges, went on to state at paragraph 15 that, even though third parties are not bound by the implied undertaking rule, they are affected by it and must respect it to some degree. In support this statement, White JA cited the decision of the Supreme Court of Canada in MacMillan Bloedel Ltd v Simpson,  2 SCR 1048, 137 DLR (4th) 633 [MacMillan Bloedel], though that case concerned the effect of a court order on third parties rather than the implied undertaking rule. The Supreme Court of Canada found (at para 27) that only parties are bound by an injunction, but
“anyone who disobeys the order or interferes with its purpose may be found to have obstructed the course of justice and hence be found guilty of contempt of court.”
 White JA went on to state at paragraph 17 that an implied undertaking does not make the documents privileged, but exists to minimize the risk of prejudice which may be suffered by a person who is compelled to produce documents on discovery. He concluded as follows:
… The purpose of the undertaking is not therefore, to prevent third parties from using the information (which is the purpose of privilege), but to impede access to information to which they may not otherwise be entitled and to prevent collateral use of the information thereby obstructing justice by creating a disincentive for parties to provide full and frank disclosure.
 On the facts in Power, White JA found that, once the third party regulatory body had received the documents in question without having provided an undertaking as to the use thereof, it was empowered by law to investigate the information, and therefore such a use of the documents was justified and not an obstruction of justice.
 BNSF argues that Power demonstrates that an implied undertaking is not reliable to protect parties’ confidential information from collateral use by third parties. Although, I do not disagree with this argument as far as it goes, I am not convinced that BNSF would be any better protected with a protective order. The reasoning in Power that left the regulatory body free to use the information in question for its investigative purposes was based on MacMillan Bloedel which concerned the effect of a court order on third parties.
 I conclude that BNSF’s concern about the possibility of collateral use of its confidential information if such information were somehow communicated to third parties under a protective agreement would be equally justified if there were a protective order.
 Apart from the result in Power, BNSF notes that the members of the panel issued three separate sets of reasons, thus showing that they were not even able to agree among themselves on the appropriate reasoning. Again, I do not disagree. However, I remain unconvinced that the information in question in Power would have been better protected under a protective order.
 As regards the enforceability of the implied undertaking on third parties (such as experts) who have been retained on behalf of a party and have obtained discovery material for the purposes of the litigation, I agree with Prothonotary Tabib that it is clear that such persons are subject to the implied undertaking rule. I have been shown no authority that suggests otherwise. Accordingly, I conclude that this is not a reason to conclude that a protective agreement would not protect the parties’ confidential information as adequately as a protective order.
C. Imprecision and Lack of Certainty in the Scope of the Implied Undertaking Rule
 BNSF expresses concern that the scope of the implied undertaking rule is uncertain and imprecise. Firstly, there is less merit to that concern today than there once was. The jurisprudence over the last few decades has added much clarity to the implied undertaking rule since the time that its very existence was in doubt.
 More importantly, any perceived gaps in the implied undertaking rule can be filled by the terms of the protective agreement. The parties have come to the Court with terms for a protective order, which they clearly agree are sufficiently clear. I see no loss of precision or clarity if those same terms are put into an agreement instead.
 BNSF also argues that there may be added difficulties if, because of a change in circumstances, it later becomes necessary to amend the terms of their agreement. I am not convinced. If one of the parties forms the view that circumstances have changed and that an amendment must be made, then the other party will either agree or disagree. If the other party agrees, then amendment of the protective agreement could be done quickly between the parties and without involving the Court. On the other hand, if the parties cannot agree on the change of circumstances, then the party who believes that an amendment to the agreement is necessary may make a motion to the Court. I accept that a protective agreement is based on a meeting of the minds of the parties, and therefore the Court cannot impose terms for an amended protective agreement on an unwilling party. However, just as with the question of a protective order in the first place, if the parties are not able to agree, a Court order may indeed be justified, at least to address the terms on which the parties are unable to agree. In the end, I do not accept that amendment of a protective agreement would be more difficult than amendment of a protective order.
D. Parties’ Discomfort in the Absence of a Protective Order
 CN’s submissions on the present motion focus on practicalities in view of the fact that many litigants before the Federal Court, particularly those debating issues of intellectual property (IP), are based in the United States where there is no implied undertaking rule. There, parties to litigation are not restricted in their use of information and documents obtained in discovery unless a protective order is put in place. Accordingly, it is routine in the United States to issue a protective order to protect confidential information exchanged during discovery, especially in litigation involving IP. In addition, the existence of a protective order in United States proceedings has often been cited as a consideration in the issuance of a similar order in parallel proceedings in Canada.
CN devotes considerable energy in its submissions to the concerns of litigants, especially those based in the United States, that their sensitive information will not be adequately protected unless a protective order is put in place. CN notes that much IP litigation is international in nature, and a party in another country making decisions about where to commence proceedings may hesitate to do so in Canada if it is not confident that its confidential information will be adequately protected. It may decide to avoid Canada altogether. CN argues that the Federal Court has become a
“go to” destination for efficient disposition of IP litigation, but that this position would be threatened if the Court were to limit the routine issuance of protective orders.
 In my view, it is not necessary for the Court to address this argument in the way that it is framed. As I see it, parties’ concerns with not having a protective order in place are either justified or they are not. If they are justified, then the reason for these concerns can be considered and assessed by the Court, and it is not necessary to concentrate on parties’ anxiety or discomfort. On the other hand, if parties’ concerns are not justified (i.e. a protective agreement protects confidential information as well as a protective order), then CN’s argument is based solely on the parties’ unfounded anxiety or discomfort. In that case, the solution is not to pander to such erroneous thinking, but rather for competent counsel to explain to their clients that a protective agreement protects their confidential information as well as a protective order, and to cite jurisprudence on the subject.
 To the extent that either party is concerned that its opponent in litigation may not be sufficiently aware of its obligations under a protective agreement, or of the fact that the terms of an express undertaking in such an agreement are just as binding as the terms of a protective order, then I suggest that this concern be addressed in the text of the agreement itself.
 CN also expresses the concern that many parties in IP litigation are direct competitors and may have considerable difficulty reaching agreement on the terms of a protective agreement. While I do not doubt that this may be true in some cases, I am not convinced that the difficulty would be any less if the parties were negotiating the terms of an order rather than an agreement. Either they will be able to agree or they will not. If not, then a protective agreement may not represent a reasonable alternative measure to a protective order, and the Court may be favourably disposed to granting a motion for a protective order to address at least the terms on which the parties are unable to agree.
E. Added Heft of a Court Order
Both CN and BNSF argue that a Court-issued order is viewed more seriously by parties than an agreement between the parties. BNSF refers to the
“added heft” of a Court order. In my view, this argument is reason for serious concern. It suggests that parties to IP litigation in Canada are routinely exchanging discovery material with an erroneous impression of the obligations of the receiving party under the implied undertaking rule. As demonstrated in these reasons, the implied undertaking rule, as supplemented by a protective agreement, provides protection for parties’ confidential information equal to that provided by a protective order. Any misunderstanding on that score by a receiving party should be corrected without delay.
 Moreover, I am concerned that the continued routine issuance of protective orders in the circumstances similar to those in the present case risks perpetuating some parties’ misunderstanding of their obligations in respect of discovery material.
F. Important Change to Longstanding Practice
 CN argues that the Federal Court’s longstanding practice of issuing protective orders on consent should not be changed without a corresponding amendment to the Federal Courts Rules. Justice Ahmed agreed at paragraph 30 of Seedlings 2.
 In my view, this is the strongest argument for granting the present motion. A desire to maintain the status quo has prompted me to challenge my views on all of the arguments put forward by the parties for granting the motion for a protective order.
 However, I am satisfied that the test provided by the Supreme Court of Canada in Sierra Club is applicable and binding on me. Moreover, I have not been convinced that this test is met in the present case. A protective agreement appears to be a reasonable alternative measure which adequately addresses the risk to the parties’ confidential information. I note that the law relating to the implied undertaking rule has evolved over time such that earlier uncertainty which may have led to the longstanding practice is no longer such a concern.
 Another argument put forward by CN is that, without the involvement of the Court in issuing protective orders, there will be a wider variety of terms on which parties reach agreement, and hence a deterioration in the uniformity of such terms. Over time, CN argues, the negotiation of protective agreements will become more difficult as compared to the negotiation of protective orders. Though I am not prepared to predict with certainty that this will not happen, I find this argument speculative. There exist many precedents for protective orders and they are equally practical for use in protective agreements. It is not clear to me that, even if some parties are able to negotiate more inventive terms for their protective agreements than currently exist in protective orders, the process of negotiating protective agreements thereafter will thereby become more time-consuming and less efficient. The precedents accepted by the Court will remain available.
 In the present case, I conclude that the reasonable alternative measure of a protective agreement will protect the parties’ confidential information as well as the requested protective order.
 Because of this conclusion, it is not necessary for me to consider the second part of the Sierra Club test which balances the salutary and deleterious effects of granting the requested order. However, I would expect that, in the context of a protective order (which does not address documents filed with the Court), this part of the test would typically not be a challenging obstacle since a protective order has no deleterious effects on the principle of open and public courts.
“George R. Locke”
March 7, 2019
SOLICITORS OF RECORD
STYLE OF CAUSE:
CANADIAN NATIONAL RAILWAY COMPANY v BNSF RAILWAY COMPANY
PLACE OF HEARING:
DATE OF HEARING:
February 11, 2019
REASONS FOR ORDER:
MARCH 7, 2019
For The Plaintiff/
DEFENDANT BY COUNTERCLAIM
For The Defendant/
PLAINTIFF BY COUNTERCLAIM
SOLICITORS OF RECORD:
Smart & Biggar
Barristers and Solicitors
For The Plaintiff/
DEFENDANT BY COUNTERCLAIM
Baker & McKenzie LLP
Barristers and Solicitors
For The Defendant/
PLAINTIFF BY COUNTERCLAIM