Editor: Why should the scope of e-discovery be limited? Should it presumptively be limited to information available in the ordinary course of business?
Schulman: E-discovery continues to be one of the most costly stages of any litigation. The very first rule in the Federal Rules – Rule 1 – tells us that our courts are supposed to achieve “the just, speedy, and inexpensive determination of every action and proceeding.” That goal is getting harder. Lately there has been much discussion surrounding the concept of proportionality, which is reflected in Federal Rule 26 (and also, increasingly, in state and local rules). The idea is that discovery should be limited so that the burden or expense of the discovery does not outweigh the likely benefit of that discovery to the resolution of the case. Information that is available in the ordinary course of business is usually the least burdensome to collect and produce, so that’s often the first set of ESI to be collected. I am not sure that a “presumption” is warranted in every case, but in many cases I suggest a staged approach; let’s produce ordinary business records first and see what else is needed for the case in light of Rule 26 and similar concepts of proportionality. I once heard that someone had calculated that only a fraction of one percent of documents produced are actually used at depositions or trial. We are producing a lot of discovery that’s simply not needed, and there are ways to do it smarter and more efficiently while being fair to all parties. These are subjects that can be discussed in advance of collection and production with the adversary at preliminary conferences.
Editor: Why should e-discovery costs be borne by the requesting party?
Schulman: In many other countries, discovery and other litigation costs are borne by the losing party. In the United States, we feel the courthouse door should be open even for plaintiffs who would not be able to pay costs if they were to lose. This is particularly significant in the class action context, where you might have individual plaintiffs. That guiding philosophy existed long before anyone envisioned the dramatic costs of just getting past discovery and to the merits of the case. So there’s now a better argument for some kind of cost-sharing arrangement, though the question of ability remains. Some of the new emerging technologies, like predictive coding or technology-assisted review, might present the requesting party with a choice: agree that the responding party can use one of the new efficient approaches to the document review, or help pay the costs of another approach.
Additionally, there have been some recent court decisions allowing taxation of e-discovery costs by the prevailing party under 28 U.S.C. § 1920, but there is inconsistency across the country on how that provision is interpreted. The Third Circuit has taken a narrow view on how to read the statute and what costs are recoverable. California courts have taken a much broader view. The statute has not been applied in New York in a reported decision that I’m aware of, so there remains a huge opportunity for a prevailing party to seek reimbursement of various types of e-discovery expenses after the conclusion of the case. This could pose a serious challenge for class action plaintiffs and their counsel who are counting on not incurring much in the way of discovery costs in order to bring their case to a verdict.
Editor: Why should the preservation obligation be triggered by the commencement of litigation?
Schulman: Traditionally, the preservation obligation has been triggered by the “reasonable anticipation” of litigation. That doctrine was developed at common law, concerning physical evidence such as wrecked automobiles when the custodian really knows there is likely to be a lawsuit. This standard has become challenging in e-discovery, because any manner of incidents or complaints could be considered a step towards a litigation of some kind, and locking down various data systems in response to those events can be disruptive and expensive.
A colleague at a major corporation once told me that of the dozens of litigation holds they put into place each year, perhaps 10 percent turn into actual litigation. A lot of time and effort is spent to secure ESI that is never examined. And, of course, a large part of the information isn’t all that important to the case if one is commenced. A trigger that turns on the commencement of litigation would obviously lessen the preservation burden but would also increase the chance that relevant ESI is lost. The key is that the anticipation must be reasonable. It could be that in many instances, including certain class actions, there is no “reasonable” anticipation of litigation until a lawsuit is actually commenced because it is so common for a corporation to receive complaints from consumers or investors that never materialize into actual litigation.
The Sedona Conference Commentary on Legal Holds explains that a “credible threat” of litigation is required. I think that’s an important way to think about the trigger. In 2009, the New York City Bar Joint Committee on Electronic Discovery issued a proposal to amend the CPLR to specify the preservation obligation as arising when “future litigation is likely” – a higher threshold than the common law federal standard. Since then, however, the First Department has adopted the federal “reasonable anticipation” standard in its January 2012 VOOM HD Holdings v. Echostar decision. So, we are probably stuck with the existing framework, absent legislative action.
Editor: Why should the preservation obligation be limited primarily to information relevant, material and proportional to the claims and defenses with an emphasis on what is necessary to the conduct of the business?
Schulman: There has been a lot of discussion about amending the Federal Rules to apply some aspect of its existing concepts of proportionality, found in Rule 26, to the preservation stage. The idea is that it is wasteful and disruptive to preserve vast quantities of ESI when only a small portion will be called for in discovery. In some cases, the preservation exercises can be quite expensive and arguably in disproportion to the benefit of preserving all that data. In the Pippins v. KPMG decision out of the Southern District of New York in early 2012, a party was ordered to preserve over 2,500 hard drives of all its nationwide former employees who were potential class members of a putative employment class action regarding overtime pay. The court ordered broad preservation even though the estimated cost was $1.5 million. The outcome might have been different if there were a new rule requiring preservation principles to be considered including the size, scope and value of the prospective litigation. Right now the doctrine seemingly says that every potentially relevant source of ESI must be preserved, although many local rules now contemplate an early discussion with your adversary about the scope of preservation. The Federal Rules Advisory Committee is considering proposed amendments to the rules that might incorporate a proportionality standard for preservation, but such an amendment is years away, even if the Committee achieves consensus on what it should say.
Editor: Why should spoliation sanctions be imposed only where willful conduct was carried out for the purpose of depriving another party of the use of the destroyed evidence and the destruction results in actual prejudice to the other party?
Schulman: The Second Circuit has traditionally viewed spoliation sanctions as a matter of restoring a level playing field. So if one party has destroyed material evidence, a sanction should be imposed that places the “cost” or consequence of that loss on the party who was in a position to prevent it. You wouldn’t want a culpable party to benefit from its own negligence. That’s the Residential Funding case, and it reflects a non-punitive view of sanctions. The problem is that the word “sanction” is pretty harsh, both for parties and their law firms, and when juries are instructed about loss of emails or other electronic files, and given the option of drawing an adverse inference, there is a disproportionate impact on the merits of a case. Or at least, that’s the fear. Email loss may be treated with suspicion, even if the loss was simply the result of an oversight. That is the main reason to implement a requirement of purposeful conduct or bad faith. Indeed, other courts, such as those in the Fifth Circuit, do require a showing of bad faith before severe sanctions are imposed. If the goal is to make the e-discovery process fair for all parties, I think it makes sense to recognize that in a rapidly changing computer environment, ESI is sometimes easily lost through inadvertence. The resulting corrective action in the litigation shouldn’t disproportionately impact the party who was otherwise acting in good faith. The Federal Rules Advisory Committee is looking at amending the rules to provide for a national standard. It is a point of great contention, I think, in part because you could view such sanctions either as corrective or as punitive, and an effective judicial system needs to leave room to impose either kind of consequence under the right circumstances. It doesn’t make much sense to me that this standard would be different depending on the federal circuit your case is in, particularly because the preservation exercise may be required before the lawsuit is even brought. That is perhaps the best reason to try to codify some national standard governing spoliation sanctions.
N.B. The answers reflect my personal views and are not that of the firm or any of its clients, nor should they be construed as legal advice.