True North.

Back to Blog

ESI Gaps?

Pause before assuming what’s missing is adverse – or even evidence.

In the current round of proposed changes to the Federal Rules, the change to Rule 37e related to the standard for an adverse inference instruction has raised some interesting controversy in the legal community.  Some think that a showing of negligence or gross negligence should be the standard. But in this era of voluminous and uncontrolled data, the proposed rules appropriately articulate a standard of bad faith.

An adverse inference instruction influences a fact-finder’s determination of the merits of a case. The instruction should be given by a court only when it is clear that the lost information would have been 1) admissible, material evidence, and 2) adverse to the party that lost it. Absent bad faith, the supposition that those conditions are met and the outcome-skewing impact of the instruction on the justice system are both unwarranted.

Lost Information Should Not be Presumed to be Lost Admissible Evidence. In the age of electronic information, a vast amount of information is created daily. It is copied, shared, and stored in multiple locations, devices, and forms. In the age of social media and “bring your own device to work,” the locations are increasingly disparate. The variety of communication devices individuals use and the sharing opportunities of which they avail themselves at home are migrating to the work place; workers want to use means of communication with which they are comfortable and businesses reach out to consumers where they spend their time.

Further, the systems and cycles and rules for saving or backing up these locations, and for subsequently storing or disposing of them, are varied and often complex. There is so much information that managing information item-by-item based on perusal of content is difficult absent sophisticated search techniques. Often, information is managed based on high-level considerations such as date of creation (e.g. a time-based retention system), or employment status (e.g. handling of data of terminated employees). Therefore, if a source is lost or deleted, absent evidence of a document-by-document determination, there is no inherent reason to presume that the content would have been adverse rather than favorable regardless of the care used in deciding to overwrite or delete it.

Finally, information is commonly created and stored in locations that contain content on myriad topics. As litigators and in-house litigation counsel know, only a small percentage of information for a particular custodian or from a particular source is likely to relate to a claim or defense in a specific dispute.  A far smaller percentage of that discoverable information is actually identified as a trial exhibit, and still less is actually admitted into evidence at trial. And a yet smaller percent is case-determinative.

As a result:

  • Information lost from a particular device, location, or custodian may discoverable by other means. Content that is on point, whether duplicative or not, is likely to exist in multiple locations and forms. Witnesses can testify.
  • There is low likelihood that information from any particular source would have been admitted into evidence at trial.
  • There is an even lower likelihood that the lost information would be case-determinative against the party that lost it.

In these circumstances, without evidence that a loss of information was in bad faith, there is no basis to infer that the lost information would have been admitted into the trial, much less propelled the opponent to a win. In contrast, where information was deleted to prevent its detection, there is a rational basis to conclude that it would have been harmful to the deleting party. A standard based on bad faith properly protects opponents from intentional deletion of content known or believed to assist the opponent.

A Bad Faith Standard Best Serves Our Justice System. In the electronic era, document discovery is reported to constitute the largest single cost in litigation. The cost to preserve electronic information for potential discovery is substantial as well, in both dollars and risk. Given the myriad sources and the relative invisibility of the electronic information (when boxes-worth of data are stored on a single thumb drive and much information is stored offsite or in a third-party’s social media platform), preservation scope and method is often complex to determine and implement. The larger the litigant, the more sources and system setting are likely to be implicated; the smaller the litigant, the less likely it has the expertise to preserve properly its electronic information in the face of system functions that may change it.

Preservation involves judgment: what content, what time periods, what custodians, what sources. For an in-house counsel, those judgments are often made long before there is a stated claim, perhaps as often as they are first made after a claim is brought. While the very fact of pre-litigation issuance of a legal hold evidences good faith, an opponent may ask a court to evaluate the scope of the hold years later, when information outside the scope may be long-gone.  Consider, for example, a claim brought on the eve of expiration of a three- or six-year statute of limitations, when a dispute on preservation occurs once the case is underway.

The standard on which an adverse inference instruction could issue directly impacts in-house preservation decisions. If in the future, with 20-20 hindsight, a judge could determine that a well-intentioned judgment on preservation boundaries were grossly negligent and issue an adverse inference instruction on data counsel never even looked at, counsel could manage the risk only by preserving every imaginable source of information. Yet some risk remains, as the actual complaint or litigation may still take an unanticipated turn. The expansion of resources needed to identify, locate, and preserve all those sources is unwarranted. It is important to note that not every anticipated litigation turns into an actual case; but the resources are expended and the data maintained until the statute of limitations expires nonetheless.

As a result:

  • A lower standard for adverse inference instructions will drive up the cost of discovery and the attendant risks.
  • Higher costs and risks will perpetuate the trend for cases to settle on factors other than the merits.

We serve the justice system poorly to let a court tell a jury that the information that was of unknown, and perhaps irrelevant or immaterial, content, would have hurt the side that lost it. To the contrary, the justice system should place a premium on just and speedy and cost effective disposition on the merits. The rules amendment proposal permits information about a loss of information to go to a jury in appropriate circumstances, for the jury to make its own assessment – based on existing evidence and witness credibility — of the significance of the event. In the absence of bad faith, considering the factors above, there is no basis for adverse inference instruction to turn preservation and discovery mistakes into case-dispositive events.


Thank you for subscribing to the H5 blog, True North.

We strive to provide quality content on a variety of topics related to search, eDiscovery and the legal realm.

Please check your email inbox for your subscription confirmation!