Produce It Now, Ask for It Back Later


As any company that has been involved in litigation knows, fact discovery is often the most time-intensive and expensive phase of a civil case. Traditionally, at the front end of the fact discovery phase of any civil litigation, the parties negotiate document custodians (i.e., the persons whose files will be searched for relevant and responsive documents) and keyword search terms (i.e., terms applied to collected documents to isolate a subset of the data to be reviewed and analyzed prior to production) for purposes of identifying documents potentially responsive to the other side’s written discovery requests. Keyword searches, even those narrowed through negotiation (e.g., using hit rates and similar narrowing mechanisms), can—and often do—find thousands (if not millions) of documents in the agreed-upon custodians’ files, many of which are completely irrelevant to the case and contain highly sensitive business and personal information.

Given their tendency to return large numbers of irrelevant documents, applying search terms is typically just the first step in a document review process. Many parties elect to manually review all documents that contain the agreed-upon terms for both relevance/responsiveness and privilege before producing all non-privileged, relevant/responsive materials to the opposing party. Others may negotiate the use of technology-assisted review (“TAR”) or some other form of artificial intelligence to narrow the population of documents for production, often allowing them to avoid search terms altogether.

But with the lifting of a temporary stay order pending a petition for writ of certiorari before the US Supreme Court, civil litigants in a large antitrust case are now facing an unprecedented discovery order requiring them to produce any and all documents that hit on a search term prior to conducting any relevance review of those documents and then attempt to claw irrelevant documents back from the opposing party at some later date. This novel process—if allowed to stand and if adopted by other federal courts—will have significant implications, both strategic and monetary, for current and future litigants.

On March 6, 2020, the US Supreme Court vacated a February 28, 2020, order entered by Justice Alito temporarily staying the portion of a federal district court’s case management order (“CMO”) in a sprawling antitrust suit that prohibited the parties from “withhold[ing] prior to production any documents based on relevance of responsiveness.”1

Defendants—dozens of generic pharmaceutical companies—sought a writ of mandamus from the Third Circuit after the entry of the CMO, arguing that it contradicted the Federal Rules of Civil Procedure, which impose a mandatory duty on courts to limit discovery to material that is “relevant to any party’s claim or defense.”2 A divided Third Circuit panel refused the writ, stating that district courts have “wide latitude in controlling discovery” and “broad parameters” to “compel the production of documents.”3 The majority further stated that the clawback and confidentiality provisions included in the CMO would adequately protect the parties, who could “claw back trade secrets, unrelated business information, and unrelated personal or embarrassing information.”4

In a petition for writ of certiorari, the defendants asked the Supreme Court to review the Third Circuit’s decision, which they characterized as “contravene[ing] Rule 26’s text, structure, and history” and “squarely conflict[ing] with [the Supreme Court’s] precedent holding that Rule 26(b)’s relevance requirement is to be ‘firmly applied.’”5 The defendants further argued that the Third Circuit’s requirement of an “unprecedented use of ‘clawbacks’” encouraged the “abuse of discovery by seeking material—including sensitive business information—unrelated to any claim or defense” and using the threat of having to produce that material to coerce settlement.6

With the March 6 vacation of Justice Alito’s February 28 stay, discovery in the lower court resumes in full, subject to all of the restrictions (or lack thereof) in the CMO. Parties must begin the process of producing documents without a relevance/responsiveness review, subject to a vague opportunity to claw irrelevant documents back from opposing parties at some point later in the litigation.

It almost goes without saying that “[t]he implications of the CMO [] extend far beyond this case.”7 If courts allow parties to seek broad discovery of irrelevant materials, the discovery process may shift from being focused on finding relevant and probative evidence, to improperly motivating settlement. Defendants faced with the choice of paying hundreds of thousands of dollars to process and produce highly sensitive, irrelevant documents (which they may or may not be able to clawback after production) may be incentivized to settle even frivolous lawsuits for large amounts of money. The special master who issued the Actavis CMO acknowledged that one reason for allowing such “extensive and broad-ranging discovery” was that it was “essential for any meaningful settlement.”8 However, not all plaintiffs are entitled to settlement; rather, “the purpose of discovery is to help determine whether plaintiffs are entitled to relief.”9 The threat of having to turn over confidential, competitively sensitive or personal information that is irrelevant to any claim or defense may have an effect of placing a thumb on the scales of justice.

A number of large corporations—litigants in numerous cases and jurisdictions—have taken issue with this seismic shift in discovery. On March 16, 2020, a dozen companies jointly filed an amicus brief with the Supreme Court asking the Court to grant Defendants’ Petition for Certiorari.10 In the amicus brief, the companies argued that the CMO may lead to the “unnecessary production” of “irrelevant but valuable trade secrets and confidential information” of “third parties such as amici [who] may never discover that their proprietary and confidential information was disclosed.”11 Because of the likely impact to the privacy interests of companies and individuals unrelated to the litigation, the companies asserted that “[t]he sweep of the [CMO] compromises the ability of our civil litigation system to coexist with the personal privacy regimes of other countries, most notably the European Union.”12

Tips for Future Litigants

While litigants, particularly in the Third Circuit, may not be able to avoid expansive discovery orders, such as the one described above, parties can (and should) take proactive steps to limit their potential harm.

  • Consider Negotiating the Use of Technology-Assisted Review (“TAR”). While search terms are still common in litigation, TAR is quickly becoming a popular e-discovery practice. One form of TAR (typically referred to as “TAR 1.0”) is a process that allows a small team of expert reviewers to classify a subset of documents for relevance. A computer algorithm is then applied to the remaining documents, tagging potentially responsive documents based on keywords and other metadata. While documents identified as potentially relevant by the computer algorithm are usually manually reviewed for relevance by the producing party, TAR 1.0 typically will decrease the number of irrelevant materials being produced even without the secondary manual review.

    One benefit of TAR is its potential to achieve large cost savings due to the significantly smaller volume of documents that require manual review by lawyers. Even if a secondary review of documents that are identified as potentially relevant is conducted, the need to double-check documents identified by the algorithm as non-relevant is greatly diminished. TAR can also decrease the amount of time between when documents are collected/processed and when they are ultimately produced to the opposing party—a benefit that may help sway opposing counsel during negotiations. TAR is not without costs and risks, however. If a manual review will not be conducted prior to production of documents identified as relevant by the algorithm, TAR is likely to increase the total number of documents being produced compared to a traditional review (i.e., applying search terms and reviewing documents for relevance/responsiveness before production).

    If CMOs, like that entered in Actavis, are going to become the new normal (and it appears it may be at least in the Third Circuit), litigants should consider whether TAR is a better option—even if a manual responsiveness review will not be allowed.
  • Consider Completing a Relevance Review Prior to Production. Companies that find themselves subject to expansive discovery orders, like the Actavis CMO, should nonetheless consider conducting a relevance review before handing the documents over to opposing counsel (i.e., within the time afforded by the CMO to make production). Identifying what material should be clawed back prior to production (and having the clawback filing prepared) will limit the amount of time opposing counsel has with the irrelevant documents and therefore the likelihood that competitively sensitive or highly personal material will be seen by multiple people.
  • Confidentially Designations Are Key. Counsel should ensure that any applicable protective order or confidentiality order includes a three-tier confidentiality designation structure, which limits the dissemination of commercially sensitive or personal information to outside counsel only. A structure that distinguishes between “Highly Confidential,” “Confidential” and “Not Confidential” documents will allow the producing party to keep “Highly Confidential” documents outside of the hands of businesspeople at their opponent, who may be a competitor.
  • Know the Rule of Your Jurisdiction. When it comes to producing irrelevant materials, location matters. The Third Circuit’s Actavis decision—which upheld a discovery order that required the production of irrelevant materials, where there was no accusation or finding of misconduct to justify departing from ordinary discovery procedures—contradicts decisions from the Fifth, Eighth, Ninth, Tenth and Eleventh Circuits.13 Companies facing litigation in the future should be aware of these differences and, if the litigation is in a circuit that has previously overturned similarly expansive orders, be ready to petition the circuit court for relief.


1 Actavis Holdco U.S. Inc. et al. v. Connecticut et al., No. 19-1010 (Feb. 28, 2020) (order granting temporary stay). Section 3(b) of the Case Management Order states, “Defendants shall apply the agreed search terms to the agreed custodial files and may review the identified documents for privilege, but may not withhold prior to production any documents based on relevance of responsiveness. ” In Re: Generics Pharmaceuticals Pricing Antitrust Litig. (E.D. Pa. Oct. 24, 2019) (Case Management Order and Discovery Schedule).

2 Fed. R. Civ. P. 26(b).

3 In Re: Actavis Holdco U.S, Inc., et al., No. 19-3549 (3rd Cir. Nov. 22, 2019) (order denying writ of mandamus and motion to stay discovery).

4 Id.

5 Petition for Writ of Certiorari for Defendants at 2, Actavis Holdco U.S. Inc. et al. v. Connecticut et al., (Feb. 28, 2020) (No. 19-1010).

6 Id. at 2-3.

7 Id. at 3.

8 In Re: Generics Pharmaceuticals Pricing Antitrust Litig. (E.D. Pa. Oct. 24, 2019) (Report and Recommended Order).

9 Petition for Writ of Certiorari for Defendants at 35, Actavis Holdco U.S. Inc. et al. v. Connecticut et al., (Feb. 28, 2020) (No. 19-1010).

10 See Motion for Leave to File Brief for Amici Curiae and Brief of Twelve Companies and Robert D. Owen as Amici Curiae in Support of Petitioners, Actavis Holdco U.S. Inc. et al. v. Connecticut et al., (Mar. 16, 2020) (No. 19-1010). Amicus briefs also have been filed by the Chamber of Commerce of the US, the National Association of Manufacturers, the American Tort Reform Association, the Pharmaceutical Research and Manufacturers of America, Lawyers for Civil Justice, and the Voice of the Defense Bar.

11 Motion for Leave to File Brief for Amici Curiae and Brief of Twelve Companies and Robert D. Owen as Amici Curiae in Support of Petitioners at 9, Actavis Holdco U.S. Inc. et al. v. Connecticut et al., (Mar. 16, 2020) (No. 19-1010).

12 Id. at 7.

13 Id. at 23-26.