E-Discovery Update: Revisiting ESI Agreements and Court Orders

Under the Federal Rules of Civil Procedure (and in an increasing number of state courts), litigants must meet early in a dispute - generally within the first 60-90 days of a case—to discuss the scope of discovery and to hopefully reach agreement on how best to proceed with the discovery of potentially relevant electronically stored information (“ESI”). Results of this meet and confer session are formalized in a court order following a Scheduling Conference. What happens, though, when fundamental assumptions used to reach agreement at that early stage in the case turn out to be incorrect?

Recovering from discovery disclosure errors is much trickier than talking through potential issues before they arise. Finding an equitable solution involves, at a minimum, weighing a number of factors. First, of course, who made the error—the requesting party or the producing one? Second, what type of due diligence was performed by the erroneous party prior to making incorrect statements? Third, what is the impact of the incorrect assumptions and the best way to move the litigation forward, given the actual state of discovery?

Both Requesting and Producing Parties Make Mistakes

Most e-discovery jurisprudence discussing erroneous initial discovery assumptions focuses on the actions of the producing party. However, requesting parties also make mistakes. For example, negotiating counsel for the requesting party may not be fully briefed on the technology that his team intends to use, leading to an agreement for ESI to be produced in a format that cannot be read by their tools. Sometimes, too, in an effort to reach agreement, negotiating counsel may defer or waive production of specific ESI that later turns out to have critical importance. And of course, negotiating counsel may make completely miss the target when estimating the cost of achieving certain goals, leading to later unexpected pressure on a client-approved budget.

Producing parties, of course, have many opportunities to make inadvertent misstatements regarding discovery of their ESI. They may significantly over or under-estimate the amount of ESI that must be searched for potential relevance. They may discover that their data harvesting protocol relies on staff and hardware assumptions that turn out to be wrong. They may find that their data repositories are partially or wholly corrupt, preventing the party from retrieving ESI that they believed they had. They may discover that their pre-production document review effort requires much more time than anticipated. The list of potential problems is nearly endless.

Courts have been increasingly stern with both requesting and producing parties that fail to do adequate due diligence regarding ESI before the initial meet and confer and Rule 26(f) conferences and that must subsequently change promises that they made. Long before the revised Rules took effect in 2006, requesting parties have been on notice that they must lie in the discovery bed that they have made (E.g., Bills v. Kennecott Corp., Bills v. Kennecott Corp., 108 F.R.D. 459, 1985 U.S. Dist. LEXIS 15310, 42 Empl. Prac. Dec. (CCH) P36732, 40 Fair Empl. Prac. Cas. (BNA) 1182 (D. Utah Oct. 3, 1985) or, at the very least, pay the additional expense the producing party would incur to comply with their revised request for production of documents (e.g., Anti-Monopoly, Inc. v. Hasbro, Inc., 1995 WL 649934 (S.D. N.Y. 1995)). Producing parties, in turn, have been disciplined—sometimes severely—when it appears that they didn’t look particularly hard before making binding representations to opposing counsel and to the court (e.g., Coleman (Parent) Holdings, Inc. v. Morgan Stanley & Co., 2005 WL 679071 (Fla. Cir. Ct. Mar. 1, 2005), rev'd on other grounds, Morgan Stanley & Co. Inc. v. Coleman (Parent) Holdings Inc., 955 So.2d 1124 (Fla. Dist. Ct. App. 2007)).

But what happens when litigants have negotiated in good faith, with seemingly adequate due diligence - and one or both still end up fundamentally wrong in their assumptions? For example, what if the vintage hardware retained to restore obsolete backup tapes turns out to have failed—and alternative methods of reading this data are exponentially more expensive than what had been expected? Can a producing party now claim that this ESI, explicitly promised to the requesting party, has fallen outside the “reasonably accessible” standard articulated by the Federal Rules of Civil Procedure? What happens when a complaint—or answer—is amended to the point where entirely new data repositories that originally fell outside the scope of the original vision of the case should now be retained and searched—to the extent that they still exist? Or, as was recently discussed in DC Circuit appellate opinion, In re Fannie Mae Securities Litigation, _ F.3d _, 2009 WL 21528 (C.A.D.C., Jan. 6, 2009), what are the consequences of entering into a stipulated order designed to resolve some of the problems encountered over the course of complying with a subpoena issued under Federal Rule of Civil Procedure 45?

Beware the Power of the Stipulation

In re Fannie Mae demonstrates the harsh consequences than can flow from strategic decisions that seemed to make sense at the time. Appellant Office of Federal Housing Enterprise Oversight (OFHEO) was served with several broad subpoenas requesting documents that might be relevant in numerous lawsuits consolidated into multi-district litigation in United States District Court for the District of Columbia. After unsuccessfully moving to quash the subpoenas, OFHEO was able to persuade the requesting litigants to limit the scope of their requests. However, even after reaching some level of agreement as to scope of the document requests, OFHEO was unable to timely meet its promises. Further complicating matters, the Agency also offered seemingly less than accurate explanations for its actions.

In response to OFHEO’s actions, the requesting parties filed a motion to hold the Agency in contempt for its failure to honor its legal obligations. However, part-way through the multi-day contempt hearing, OFHEO entered into a stipulated order that held the contempt proceedings in abeyance. Among other things, OFHEO agreed to search for electronic stored documents using terms supplied by the requesting parties. Significantly, the order stated, “[T] the Individual Defendants will specify the search terms to be used.”

Pursuant to the stipulated order, the requesting parties submitted approximately 400 search terms, which flagged approximately 660,000 documents as potentially responsive. OFHEO, in turn, hired fifty (50) contract attorneys to review these documents for relevance and privilege, ultimately spending over six million dollars ($6,000,000) on this project, or approximately nine percent (9%) of its annual budget. Even with this substantial effort, however, the Agency was unable to meet the deadlines that it had negotiated with the requesting parties, even after several extensions. Upon motion, the district court from which the subpoena was issued found OFHEO in contempt and ordered, as a sanction, that OFHEO produce all documents withheld on the sole basis of the qualified deliberative process privilege that had not been logged by the final, court-sanctioned deadline.

On appeal, the Court of Appeals for the District of Columbia Circuit denied OFHEO’s request for relief and upheld the District Court’s sanctions. Reviewing the sanctions from an abuse of discretion standard, the Court of Appeals found that OFHEO had voluntarily bound itself to the specific terms of the stipulated order, even though those terms were aggressively exploited by the requesting parties. Notwithstanding the fact that OFHEO was not a party to the underlying litigation and notwithstanding the uncontested fact that OFHEO had spent nine percent of its entire annual budget trying to comply with the subpoenas, the Court found that the Agency’s considerable and expensive efforts to meet the very high burden that had been placed on the Agency were nonetheless sufficiently flawed that a District Court could have concluded that sanctions were appropriate.

Assessing the Significance of In re Fannie Mae

At first glance, the In re Fannie Mae decision seems patently unfair. A non-party to a litigation matter searched its active and backup tape ESI repositories using 400 search terms compiled by the requesting parties and spent a significant portion of its annual operating budget collecting, processing, reviewing and otherwise preparing these materials for production—yet was still sanctioned on grounds that it hadn’t done enough to meet its legal obligations. It is not unreasonable to wonder whether this decision will empower requesting parties to raise the stakes in litigation by serving similarly broad discovery requests in hopes of creating additional pressure to settle litigation due to the cost of complying with the demands.

Upon reflection, though, In re Fannie Mae is a reminder that quality of lawyering can make or break a case. It’s unclear from the appellate record exactly what arguments OFHEO’s counsel made in its unsuccessful attempts to quash the initial subpoena, but the Court of Appeals made no mention of a detailed, well-supported explanation of the burden that requesting parties sought to impose on the Agency. It’s somewhat more apparent that OFHEO’s counsel gradually lost credibility with the District Court by repeatedly promising production dates, only to flatly state at each deadline that the work had not been completed. OFHEO and its counsel lost even more credibility when the requesting parties were able to demonstrate that the Agency had not been completely truthful to the District Court regarding its efforts to search its ESI repositories for potentially relevant documents. And finally, one can infer that the stipulated order that laid the foundation for the imposition of sanctions was drafted as a last-ditch attempt by OFHEO to avoid a contempt hearing that the Agency felt it was unlikely to win.

Given this case history, it’s not surprising that the presiding District Court finally reached a breaking point and held OFHEO to the literal terms of a stipulated order that the Agency itself helped draft. Given the Agency’s well-documented history of responding unevenly and inadequately to a subpoena that the Court had previously found valid, even strong factual arguments about the cost of subpoena compliance could not outweigh the fact that none of this expense appears to have produced the results repeatedly promised by the Agency’s legal team. In re Fannie Mae should remind practitioners yet again that unqualified representations and statements in e-discovery practice are a poor strategy—and that it can be difficult if not impossible to recover from such hard and fast positions once they have been asserted.