chevron-down Created with Sketch Beta.

Litigation Journal

Spring 2022: Embracing Change

Protecting Confidential Information

Mark A Klapow, Raija Horstman, Laura Schwartz, and Joshua Mark Rychlinski


  • How should we address the conflict between a litigant’s desire for confidentiality and the principles of an open judicial system?
  • Courts are routinely asked to seal business records submitted with filings.
  • Citing the public nature of the judicial system and the public’s right to access, courts can be hostile to sealing requests.
  • Here’s a proposal for a new approach.
Protecting Confidential Information
malerapaso via Getty Images

Jump to:

There must be a better way. We think there is.

In most business litigation, particularly trade secret cases, discovery materials involve confidential or sensitive information. To shield discovery materials from use and disclosure beyond the case, counsel typically negotiate—and courts routinely enter—confidentiality protective orders. Indeed, to streamline that process, many courts now use sample or model orders. Regardless, at least some of that information ultimately comes before the courts. Parties need to cite evidence in briefing, particularly on summary judgment. Hearings may arise along the way. Trials, of course, always involve evidence.

Given the scope of discovery in modern business litigation, it’s unsurprising that courts are routinely asked to seal business records submitted with filings. That’s especially true in cases with trade secrets and competitively sensitive information, often the most valuable resources a business has. As a matter of principle, citing the public nature of the judicial system and the public’s right to access, courts can be hostile to sealing requests. Many judges are wary on pragmatic grounds, too.

How should we address the conflict between a litigant’s desire for confidentiality and the principles of an open judicial system? Is an ad hoc, document-by-document review a worthwhile use of limited judicial resources? Can it ever lead to consistent results?

Years of experience have shown that the uncertainty of the sealing process consumes time and resources, and creates opportunities for gamesmanship. It’s a barrier to justice, consistency, and predictability. Busy judges don’t want to hear discovery disputes, much less sealing requests that are merely tangential to the merits of a case. Enterprising parties sometimes use sealing battles to gain leverage or commercial advantage.

The inability of our judicial system to ensure the confidentiality of business information has a chilling effect. Why sue to protect stolen trade secrets if, in the processing of that lawsuit, those trade secrets must be filed on the public record?

The Competing Interests

Finding a better way starts by looking at the competing interests.

On the one hand, the public has a presumptive right of access to court proceedings. See, e.g., Oregonian Publ’g Co. v. District Court, 920 F.2d 1462, 1467 (9th Cir. 1990); Baxter Int’l, Inc. v. Abbot Labs., 297 F.3d 544 (7th Cir. 2002) (“Information transmitted to the court of appeals is presumptively public. . . .”). Even nonparties, including the media, may intervene on behalf of the public to oppose motions to seal. See, e.g., In re Google Inc. Gmail Litig., No. 13-MD-02430-LHK, 2014 U.S. Dist. LEXIS 136420, at *1 (N.D. Cal. Aug. 6, 2014). From this foundational principle, several courts have stated the presumption in broad terms, including the Fifth Circuit in an important case this year: “The working presumption is that judicial records should not be sealed.” Binh Hoa Le v. Exeter Fin. Corp., 990 F.3d 410, 419 (5th Cir. 2021).

From that premise, an increasing number of courts have created regimes hostile to sealing business records. Full public access is the rule, they say; sealing is the exception, reserved for a compelling showing of good cause—whatever that means. That approach misstates the public interest, undervalues the private interests, invites delay and gamesmanship, and is wildly inefficient in forcing courts and litigants to spend time addressing matters that are not relevant to the merits of the dispute.

The principle of presumptive open access stems from and is always supported by cases of legitimate public interest, such as a case to hold a government official accountable. The public interest is often obvious in criminal matters or matters involving public institutions. In most modern civil cases, however, the argument for the public interest is harder to discern. Moreover, the public interest in openness has always been described as relating to the public’s right to understand the basis for judicial decisions on the merits and the evidence that supports those decisions. The public has the right to know—it is argued—how and why the court decided the case.

Modern civil litigants recognize that most disputes over the confidentiality of business records do not happen in that context. Rather, most are about nothing more than the records themselves. Not every attachment to every filing by the parties forms the basis for a judicial decision on the merits of a case. Courts that refuse to acknowledge that difference misperceive the grounds for the presumption of openness.

The interests of private parties in the confidentiality of their information can be acute. The value of most enterprises is in their ideas and information. Without protections of those ideas and information, incentives to create value diminish. Across industries, there is a real public interest in protecting information on privacy and economic grounds.

In an action, for example, in which a supplier sues to protect its secret formula and pricing strategy from an alleged misappropriator, the supplier will have to disclose that confidential information in its discovery, creating the prospect that, at some point, it will be attached to a court filing, most likely by the alleged misappropriator.

Without assurances that the confidential information will not be disclosed to the public, including its competitors, the supplier has to think twice about filing the lawsuit in the first place. When the supplier does file suit, the alleged misappropriator has tremendous unearned leverage—the capacity to force, at every turn, an unfair fight over the confidentiality of the information at issue. In that way, defendants in private suits can use the openness of the court system to press leverage wholly unrelated to the public interest.

Broad pronouncements about the presumption of openness in the courts and tests to weigh private interests in confidential information run head-on into the reality that most judges have both large dockets and little patience for sealing motions. All of that creates a more acute problem in trade secret cases. Unsealing the information will destroy its value as a trade secret. In trade secret cases, a sealing motion is not tangential; it can be the whole case. That a sealing motion can devolve into the merits of the trade secret claim, without the benefit of witnesses, evidence, or a jury, presents a serious due process problem with which our courts have not really grappled.

How Confidentiality Issues Arise

Finding a better way means understanding how these issues arise, how courts approach them, and the shortcomings of those approaches.

Issues of confidentiality arise in most cases of any complexity. Today, we create and receive huge amounts of information in emails, text messages, and the like, both personally and in our business and professional settings. In discovery, litigation and subpoenaed parties must produce those materials, which may include trade secrets, financial records, pricing, strategies, formulas, processes, and other confidential business information. Shielding that information from competitors in the marketplace and from the public is of vital importance to those who created it. As a result, the threat or risk of disclosing that information is a leverage point that can be weaponized.

The problem merely worsens over time. The legal press is constantly innovating, and the appetite for hot-off-the-press court filings and analysis of them always grows. Here’s just one example: USA Today runs a Twitter profile called @big_cases (with the handle “Big Cases Bot”) that provides “real-time updates on the latest filings in major cases in U.S. District Courts and the U.S. Supreme Court” ( The bot provides real-time updates, directly from PACER, in dozens of cases involving the president, the Department of Justice, the Federal Bureau of Investigation, and others. That bot and many other similar services reflect a diverse audience for information, the expectation that it be immediately available, and the destructive power of a filing being posted on social media. Once millions of Twitter users have seen a filing, that bell cannot be unrung.

Often when a case begins, the parties are aligned in their desire to protect their own sensitive materials. With relative ease, they may agree upon a confidentiality protective order. Traditionally, courts have welcomed such orders. “Protective orders are, obviously, an ever–expanding feature of modern litigation. . . . Protective orders recognize that parties engaged in litigation do not sacrifice all aspects of privacy or their proprietary information simply because of a lawsuit.” In re Mirapex Prods. Liab. Litig., 246 F.R.D. 668, 672–73 (D. Minn. 2007).

Typical protective orders offer a multi-tiered framework based on sensitivity. While the terms of the protective orders are public, nearly all of the materials designated pursuant to them are not. That is how it must be. Judges do not review documents produced in discovery en masse, nor should they. Rather, courts rely on the parties to identify the key documents and testimony and submit them before the court renders decisions.

Once those materials are prepared for submission to the court, however, the parties’ interests in confidentiality may no longer align. Owners of confidential information remain interested in keeping it confidential. The other party usually does not and may even see leverage in pressing to make it public. That is particularly true in trade secret cases, in which the information, once publicly disclosed, is no longer a trade secret.

That misalignment may result in briefing on a sealing motion, which is becoming ever more common. In many courts, even if unopposed, a sealing motion requires the court to weigh the public’s right to access against the parties’ privacy interest in the materials sought to be sealed. As part of that calculus, courts often require at least a showing that the proposed information warrants protection. All that drains a court’s most precious resource—its time. The requirement of sealing motions and the different ways in which courts and individual judges approach them lead to an unpredictable, time-consuming, and expensive process to determine a confidentiality issue that may have no relevance to the underlying dispute. That’s simply a waste.

Due Process Issues

Moreover, there are due process problems. In trade secret cases, a decision denying some or all of a sealing request at any stage of the litigation could end the case just as thoroughly as a jury verdict for the defendant, perhaps more so, as the effect of denying a sealing motion exposes the secret to the public, which a jury verdict may not do. A sealing motion may become a decision on the entire case. Several courts have adopted sealing procedures that are cumbersome. Those courts may well hope that, by that process, the parties will be dissuaded from pursuing judicial sealing or, at least, if the parties do move to seal, that the process will produce more predictable and efficient results. Unfortunately, there is little evidence of either.

The Eastern District of Michigan, for example, recently adopted a multipart process for sealing documents submitted on the public record. A party seeking to seal a document that has been designated confidential must obtain a sealing order. E.D. Mich. L.R. 5.3(b)(1). Even if it were the opposing party who designated it, this requires briefing. Each document must come with a “detailed analysis, with supporting evidence and legal citations, demonstrating that the request to seal satisfies controlling legal authority.” E.D. Mich. L.R. 5.3(b)(3)(a)(iv).

As it may have been the other party that made the designation, the moving party may have no incentive to put forth the best argument. Even then, the court will grant the sealing motion only “upon a finding of a compelling reason why certain documents or portions thereof should be sealed.” E.D. Mich. L.R. 5.3(b)(3)(c)(i). In practice, that can lead to a highly fact-specific inquiry, rather than a simple procedural motion. See, e.g., In re FCA US LLC Monostable Elec. Gearshift Litig., 377 F. Supp. 3d 779 (E.D. Mich. 2019).

The Northern District of California has a similar process requiring the court to first enter an order permitting sealing. N.D. Cal. L.R. 79-5(b). The attendant briefing must establish that “the document, or portions thereof, are privileged, protectable as a trade secret or otherwise entitled to protection under the law.” Id. In addition, the motion must be supported by a “declaration establishing that [each] document sought to be filed under seal, or portions thereof, are sealable. Reference to a stipulation or protective order that allows a party to designate certain documents as confidential is not sufficient to establish that a document, or portions thereof, are sealable.” N.D. Cal. L.R. 79-5(d)(1)(A).

While the Northern District does not explicitly reference Ninth Circuit precedent in its local rules, the declaration requirement likely reflects a standard that the moving party must present “compelling reasons supported by specific factual findings that outweigh the general history of access and the public policies favoring disclosure.” Kamakana v. City & Cnty. of Honolulu, 447 F.3d 1172, 1178–77 (9th Cir. 2006).

The inclusion of more process has simply led to more problems without ensuring predictability. The courts that start with the immutable right of the public to access all judicial records and then focus on process have created more work without a countervailing benefit. The identified standards are too amorphous. As with any function that includes more process, smaller entities are disadvantaged. Innovators, new market entrants, and undercapitalized businesses may face hard choices about protecting information at the expense of other priorities. Those with deeper pockets will not feel those decisions as acutely.

More process also creates opportunities for gamesmanship. Disputes between commercial competitors can be multifaceted, including on the public relations front. A party’s public relations strategy may involve bringing into public view bad facts about its competitor that have emerged through discovery. The unsealing process can represent an avenue to attempt to bring protected information to light by attaching mostly irrelevant material to a pleading. That same party can then move to unseal, stating that because the document is on the court’s docket (by, of course, that same party’s hand), the public has a presumptive right to access it.

Courts sometimes see through such efforts. See, e.g., Accenture Glob. Servs. GMBH v. Guidewire Software Inc., 631 F. Supp. 2d 504, 509–10 (D. Del. 2009) (denying motion to unseal because “[i]t is clear from the record here that maintaining the amended answer and counterclaim under seal will promote the integrity of the judicial process. This litigation, like many, is part of a larger business dispute between competitors. Nevertheless, although related, there should always be a clear distinction made between the parties’ conduct in the business world and the parties’ conduct in the court. . . . [D]efendant’s answer and counterclaims recites hearsay out of context from documents designated as confidential that may not otherwise ever be exposed to public scrutiny at trial. Defendant uses such materials not merely for purposes of its litigation strategy, but (were the pleading to be unsealed) in a larger public relations fight. Defendant seeks to inflame, not inform, and the attempted use of judicial process to accomplish that end is not to be tolerated.”); see also City and Cnty. of Honolulu, 447 F.3d at 1179 (“In general, ‘compelling reasons’ [to seal documents] . . . exist when such ‘court files might have become a vehicle for improper purposes,’ such as the use of records to gratify private spite, promote public scandal, circulate libelous statements, or release trade secrets.”).

Nevertheless, the longer and more cumbersome the sealing process, the greater the opportunity for mischief. There is no evidence that more process leads to more predictability in result or less work for the courts themselves. Lawyers and clients are often left to guess whether their sealing motions will be well received. In the meantime, they “paper the record” with a flurry of sealing motions.

The District of Delaware’s Approach

The District of Delaware takes a different approach. There, once the court issues a protective order, a party has all the authority it needs to file a document under seal. Once the sealed document has been filed, the court requires a second electronic filing of a redacted version within seven days. That approach strikes a balance between the protection of parties’ confidential information and the public’s right to access. Delaware is also the nation’s busiest patent docket. That less stringent policy may reflect the court’s recognition that sensitive confidential information in patent cases is common and of little legitimate public interest or that busy judges should focus on the merits of active disputes, not sealing motions.

In a case reaffirming the view that all files are presumptively public, the Fifth Circuit recently criticized Delaware’s regime and others like it. Binh Hoa Le v. Exeter Finance Corp. reviewed (and affirmed) summary judgment in a matter in which nearly three-quarters of the record—3,202 of 4,391 pages!—was sealed. The Fifth Circuit harshly criticized that fact. “The public deserves better. . . . [W]hat happens in the halls of government happens in public view. Americans cannot keep a watchful eye, either in capitols or in courthouses, if they are wearing blindfolds.” 990 F.3d at 417.

The Fifth Circuit also noted that the case was not unique. Rather, it was “consistent with the growing practice of parties agreeing to private discovery and presuming that whatever satisfies the lenient protective-order standard will necessarily satisfy the stringent sealing-order standard.” While recognizing that some documents or parts of them can, in the abstract, be sealed because of trade secret issues or other privacy reasons, the Fifth Circuit still called for a “case-by-case,” “document-by-document,” “line-by-line” review of requests for sealing.

As in so many appellate decisions on this topic, how best to implement any of those procedures was glossed over, left as an exercise for the district courts in the Fifth Circuit to determine. That approach will result in uncertainty, unnecessary distraction, expense, and games.

Proposal for a New Approach

What’s needed is a new approach. Here’s a proposal:

Simultaneously with the entry of a protective order, the court should decide if the dispute is of a type that has legitimate public interest. In those cases—which should be a small minority of modern business disputes—the presumption of complete openness and document-by-document briefing and review of confidentiality should continue. For the vast majority of civil cases that do not meet that standard, the courts should adopt a streamlined approach in which the bulk of sealing issues would not be addressed during the litigation itself unless the court finds compelling circumstances to do so.

Instead, at the same time the court enters a protective order, it should direct the parties to file information under seal with an attestation that they have a good-faith basis for confidentiality to be tested later. Then, at the end of a case, once a decision on the merits is reached, all sealing requests would be addressed by the parties and a magistrate judge, perhaps in a “Binh Hoa Le” hearing as a tip of the cap to the Fifth Circuit.

Of course, any litigant or third party could challenge the initial sealing of a record for the case, but that party would bear the burden of showing, first, that the sealing issue not only is the basis for a judicial decision but also must be addressed at that time. The challenge process would act as on off-ramp from that approach, bringing only the specific document at issue before the court. Once the need for contemporaneous review has been established, the party seeking to seal the docket would again bear the burden of proof. The court would always have discretion to consider unsealing sua sponte.

Under the first prong of this approach, the parties could simply seal court filings as necessary to protect their confidential information as is the rule in Delaware. While no party would be denied the right to apply for unsealing during the proceeding, the general rule would be to take up those issues in a hearing or briefing once the case ends, absent anyone taking the public interest off-ramp. That would stem what the Fifth Circuit recognized as the rising tide in cases with extensively sealed records.

This proposal has several major advantages. It would dramatically lessen the burden on the court system and align judicial resources with resolving merits disputes rather than issuing sealing orders. Courts could spend less time on sealing issues, except for rare cases in which there is a genuine need or an important public interest at stake. It also would cut down on the opportunities and temptations for malfeasance by attaching documents or testimony of public relations value but borderline relevance to the issue at hand. In most cases, sealing would be assumed. If sealing is not challenged, the court would simply move on. If sealing is challenged, the movant would first be required to establish that the document at issue is the basis for a judicial decision on the merits. Before summary judgment, that should be rare.

Moreover, this new approach would lessen any chilling effect and give litigants confidence that the material they produce will be protected, even when filed with the court, at least until a decision on the merits is made. It also would make for an orderly determination of sealing motions. As many courts have said, the principle behind open access to court records is so the public can understand the basis for judicial decisions, not so the public or specific competitors can root around in a party’s files. Unsealing should be required by public policy considerations only when the document is the basis for a substantive judicial decision.

In most cases, sealing issues come up well before any substantive judicial decision and thus may have to be revisited as more decisions are reached. That makes no sense. Instead, by tying decisions on sealing to the resolution of the case, courts would be armed with the knowledge of what information, if any, is the basis for a substantive judicial decision. Under such an approach, the party whose confidentiality is at issue would have to advocate for sealing, not the party that happens to be filing the brief to which the allegedly confidential information is attached. That should result in more effective advocacy.

As most cases do not go to trial or even reach a merits decision, this new approach would avoid the courts and litigants spending copious time on sealing issues for information that never becomes the basis for a judicial determination. Once settled, the parties may come into agreement on the sealing of documents previously briefed. Considering sealing en masse at the end of a case would curtail the aggregate number of disputes. Indeed, parties considering settlement before trial might also weigh the prospect of lengthy and expensive resolution of unsealing issues if the parties have to argue after trial what information was the basis for the court’s or jury’s decision. If a case is settled before any merits decisions, then there are no records that have formed the basis for a judicial determination.

Under this approach, all the sealing issues in each case would be decided at the same time. That would be fairer and more efficient, and it would give rise to greater consistency and predictability. It’s a change we should embrace.