AMENDED RULE 37(e) OF THE FEDERAL RULES OF CIVIL PROCEDURE (“RULE 37(e)”) BECAME EFFECTIVE ON DEC. 1, 2015. It emerged as a pithy and focused restatement of the best thinking on spoliation remedies after an original, flawed attempt to over-think the process in 2013.1 As revised, it reflects the fact that many public comments had urged “an alternative focus on ‘curative measures’ in the absence of bad faith.”2
The chair of the subcommittee that drafted the rule explained that “[w]e felt that this approach would promote reasonable steps to preserve ESI, cure any prejudice, and deter intentional failure to preserve ESI.”3 The primary purpose is to bring “consistency and coherence” to the ways that courts handle claims of failure to preserve ESI4 and to resolve conflicts among the circuits.5 This article assesses the impact of the rule and its relationship to using inherent sanctioning authority in addressing spoliation.
Amended Federal Rule 37(e) provides guidance to courts dealing with the spoliation of electronically stored information (“ESI”). According to the Committee Note, it specifies the findings needed to justify the use of measures it authorizes and forecloses use of inherent authority or state law for that purpose. It provides (new material underlined):
Failure to Preserve Electronically Stored Information. If electronically stored information that should have been preserved in the anticipation or conduct of litigation is lost because a party failed to take reasonable steps to preserve it, and it cannot be restored or replaced through additional discovery, the court: (1) upon finding prejudice to another party from loss of the information, may order measures no greater than necessary to cure the prejudice; or (2) only upon finding that the party acted with the intent to deprive another party of the information’s use in the litigation may: (A) presume that the lost information was unfavorable to the party; (B) instruct the jury that it may or must presume the information was unfavorable to the party; or (C) dismiss the action or enter a default judgment.
The rule has been referred to or applied by four appellate and approximately 100 district court opinions.6 When challenged, most courts have concluded that it is equitable to apply the rule because it incorporates the existing common law duty and, in some respects, is more lenient than the rules that apply to “nonelectronic” spoliation.7 However, over 60 decisions, including those of appellate courts, have ignored the rule — often without explanation.8
A core difference from practice in the nonelectronic context is that a showing of culpability is not required for courts to address any prejudice suffered when ESI is lost which should have been preserved. That requirement, defined so to require proof of an “intent to deprive,” now cabins the use of case-dispositive measures and overrules Residential Funding v. DeGeorge and its progeny.9
As explained by the Committee Note, “the better rule for the negligent or grossly negligent loss of [ESI] is to preserve a broad range of measures to cure prejudice caused by its loss, but to limit the most severe measures to instances of intentional loss or destruction.” However, sanctions are not appropriate based solely on a finding of an intent to deprive. The note was revised prior to the meeting at the request of a member of the Standing Committee to emphasize that culpability alone is not sufficient for harsh measures.10
Spoliation involves the destruction or significant alteration of evidence, including the failure to preserve property for another’s use as evidence in pending or reasonably foreseeable litigation.11 The duty to preserve — owed to courts, not parties — is an offshoot of this doctrine and is essential to assuring the integrity of the trial process.
Rule 37(e) incorporates existing common law principles. It does not alter existing federal law as to whether evidence should have been preserved or when the duty to preserve attaches. Only if relevant ESI is “lost” because of a failure to take “reasonable steps” to preserve and the missing ESI cannot be restored or replaced through additional discovery are measures available to address the failure to preserve.
ESI that “Should Have Been Preserved.” The rule applies to the loss of relevant ESI while in the custody and control of a party prior to or after commencement of litigation if the duty has attached. It embraces ESI in any form so long as it is stored in a medium from which it can be obtained. As is the case with nonelectronic information, relevancy is broadly defined. It includes what a party knows, or reasonably should know, is or may be relevant to potential claims or defenses, or is the subject of a pending discovery request. While the burden of proof to demonstrate discovery relevancy is typically that of the moving party, courts have shifted it upon a showing of egregious conduct.
In GN Netcom v. Plantronics, for example, the nonpreserving party was required to prove that the missing ESI was not relevant after its executive deleted a massive amount of email in bad faith and with an intent to deprive.12 As noted in FTC v. DirectTV Inc., however, it is not enough to argue that “potentially relevant” ESI may have existed.13
“In the Anticipation or Conduct of Litigation.” The onset (“trigger”) of the duty to preserve is determined in the prelitigation context by whether “litigation is reasonably foreseeable.” This involves the extent to which a party is on notice that litigation is likely and that the information would be relevant. In the case of the party that initiates the action, this means that once it has decided to proceed, it has an obligation to preserve.
The issue is intensely fact-specific. The Committee Note observes that “a variety of events may alert a party to the prospect of litigation,” but cautions that they may provide only “limited information.” In the absence of such notice, according to Cache La Poudre Feeds v. Land O’Lakes, the duty must be “predicated on something more than an equivocal statement of discontent.”14
A duty to preserve may also arise from statutory requirements, administrative regulations, or a party’s own information-retention protocols. In CTB v. Hog Slat, a failure to implement a litigation hold required by an internal requirement was sufficient to trigger the duty.15 However, the mere fact of an independent obligation to preserve does not mean that the party also had such a duty with respect to the litigation.16
“Reasonable Steps.” The rule provides that if a party can show that it took reasonable steps to preserve, no measures are available even if some ESI has been lost. Although negligent conduct can negate a conclusion that reasonable steps were taken, the rule is not a “strict liability rule” that applies automatically when ESI is lost. As the court noted in Matthew Enterprise v. Chrysler Group, it provides a “genuine safe harbor” for parties that take reasonable steps to preserve.17
In Zubulake v. UBS Warburg (“Zubulake IV”), the court famously held that once a party anticipates litigation, it “must suspend its routine document retention/destruction policy and put in place a ‘litigation hold’ to ensure the preservation of relevant documents.”18 A failure to do so is a factor to weigh in the analysis and courts are prepared to order discovery about the steps actually undertaken after a party anticipates litigation.19 In O’Berry v. Turner, a “minimal” and ultimately unsuccessful effort to preserve the only paper copy of electronic data did not pass muster.20 In Marten Transport v. Plattform Advertising, however, failing to preserve a computer’s internet history during a routine replacement did not constitute a failure to take reasonable steps. The court cautioned against assessing conduct by a “perfection” standard or with hindsight as to the circumstances at the time the decision was made.21
The effort involved in implementing preservation obligations is also impacted by proportionality concerns. In Rimkus v. Cammarata, the court stated that what is reasonable “depends on whether what was done — or not done — was proportional to that case.”22
“Restore or Replace.” The loss of ESI from one source may be harmless when substitute information can be found elsewhere. Thus, even if a party fails to take reasonable steps, measures are not available if the missing ESI can be restored or replaced through additional discovery. Under those circumstances, ESI is not “lost” within the meaning of the rule, according to G.P.P. v. Guardian Protection Products.23
The rule requires courts to order additional discovery first and to not impose sanctions that would not be required. In ILWU-PMA Welfare Plan v. Connecticut General, the court ordered additional discovery at the cost of the nonmoving party in a case where it had not yet been established if the lost ESI could be “restored or replaced.”24 In Fiteq v. Venture Corporation, the moving party’s failure to take steps to depose a key witness barred application of the rule.25 In Living Color v. New Era Aquaculture, the court declined to order further discovery given “the abundance of preserved information” to meet the needs of the moving party.26
Prejudice. Absent a showing of prejudice, there is no basis for relief under the rule. Subdivision (e)(1) necessarily assumes the existence of prejudice. The finding of “intent to deprive” required by subdivision (e)(2) imparts a rebuttable inference of prejudice for measures listed there.27 In Global Material Tech. v. Dazheng Metal Fibre, for example, the court ordered a default judgment on liability under Rule 37(e) only after presuming prejudice because the nonmoving party had acted with an intent to deprive.28
There must be more than “minimal” prejudice, however. It is not enough to argue that a party must “piece together” information from various sources, as was contended in the case of In re Ethicon decision, 2016 WL 5869448 (S.D. W.Va. Oct. 6, 2016). The court in Simon v. City of New York emphasized that the loss must be of sufficient consequence to have made a difference in pursuit of claims or presentation of defenses.29
Courts have discretion in assigning the burden of proof on the topic. In some cases, it may be fair to place the burden on the nonmoving party. In McQueen v. Aramark Corporation, for example, it was enough for the court that the loss “may very well have an effect” on the ability of the moving party to purse their claim.30
In contrast to the 2006 version of Rule 37(e), which merely “carved out a safe-harbor — a good faith operation of an electronic information system — from a court’s power to impose sanctions [under the Federal Rules],” amended Rule 37(e) affirmatively provides authority to impose measures for failures to preserve covered by the rule.31 It does not, however, definitively catalog the measures. According to the Committee Note, the use of inherent authority to determine when certain measures are available is foreclosed by the rule. (See “Inherent Powers,” infra). The note does not purport to preclude reliance on other provisions of Rule 37 where applicable.32
The rule leaves the selection process to the sound discretion of the court, subject only to the cabining of use of certain harsh measures by a predicate requirement of “intent to deprive.” Proportionality is important. Subdivision (e)(1) instructs courts to select measures that are “no greater than necessary” to cure prejudice and the Committee Note cautions that the “remedies should fit the wrong.” Courts are not to apply the measures authorized by subdivision (e)(2) when “lesser” measures are sufficient to “redress the loss.”
The following measures have been employed in decisions applying the rule. There is no hierarchical ranking involved, but their use satisfies the traditional rationales of the spoliation doctrine.33
Monetary Sanctions. Monetary sanctions (other than reimbursement of attorney’s fees and costs) have been sparingly imposed under the rule. In GN Netcom v. Plantronics, the court levied a “punitive” monetary sanction” of $3 million in response to the bad-faith conduct of a senior executive and an unwillingness of the party to initially acknowledge wrongdoing.34
The court appears to have regarded that measure as authorized under the rule. It may also be indicative of an appropriate exercise of inherent powers where a party has deceived a court or abused the process at a level that is utterly inconsistent with the orderly administration of justice. (See discussion in “Inherent Powers,” infra.)
Attorney’s Fees and Costs. The reasonable costs — including attorney’s fees — associated with bringing a Rule 37(e) motion have been awarded under the rule. In CAT3 v. Black Lineage, such an award was made to address “the burden and expense of ferreting out the malfeasance and seeking relief from the court.”35 In GN Netcom v. Plantronics, the court awarded attorney’s fees and costs at least partially to address prejudice.36 This use of Rule 37(e) — despite its silence on the topic — seems appropriate given that, by definition, a moving party that meets the threshold requirements of the rule has already shown (or is presumed to have suffered) a form of economic prejudice.37
Some courts favor use of the more explicit authority of Rule 37(a)(5)(A).38 The court in Best Payphones v. City of New York, justified such an award under that rule because the spoliation motion resulted in production of additional materials.39 In Wal-Mart Stores v. Cuker Interactive, however, a court rejected use of Rule 37(a) because it pertains only to motions to compel disclosure or discovery.40
Yet other courts rely on their inherent power to justify reimbursement of fees, provided that bad-faith conduct exists.41 Richards v. Healthcare Resources, 2016 WL 7494292 (C.D. Cal. Sept. 29, 2016), is such an example. In Friedman v. Philadelphia Parking Authority, the court asserted its authority to act under its inherent power but ultimately relied on Rule 37(a) because the rule was a more “tailored” remedy.42
Preclusion of Evidence. According to the Committee Note, preclusion of evidence is available to courts without a showing of “intent to deprive” in order to deal with prejudice. In Ericksen v. Kaplan Higher Education, the use of an email at trial was precluded because of the failure to preserve ESI that might have rebutted its authenticity.43 In Cahill v. Dart, a party was prohibited from offering testimony about what had been seen on a missing videotape segment.44
Taking Unproven Facts as Established.
Courts may deem facts as established when evidence of the fact is destroyed as a result of a failure to preserve that is subject to Rule 37(e). In Morrison v. Veale, M.D.,45 a former employee was barred from challenging the accuracy of time cards she had prepared but had not preserved. The court in Security Alarm Financing Enterprises v. Alder Holdings46 entered a partial summary judgment as to liability utilizing evidence deemed to be established by an inference arising from a failure to preserve.
Admitting Spoliation Evidence Instructions. According to the Committee Note, courts may allow parties to introduce evidence about a failure to preserve ESI. They may also allow argument about the inferences the jury may draw in the absence of a finding of “intent to deprive.” This may include giving the jury instructions to assist in its evaluation of such evidence or argument.
This provides an alternative remedy, frequently invoked, when a moving party has been unable to prove an “intent to deprive.” In Nuvasive v. Madsen Medical, the court allowed both parties to present evidence regarding the loss, while instructing the jury it could consider the evidence along with other evidence in the case.47 According to Mali v. Federal Insurance, “[s]uch an instruction is not a punishment. It is simply an explanation to the jury of its fact-finding powers.”48
In BMG Rights Management v. Cox Communications, a court gave an instruction “alerting the jury to the ‘fact’ of spoliation, identified the missing evidence, and permitted them to consider that fact in their deliberations.”49 In Abdul-Hamza Wali Muhammad v. Mathena, the court planned to instruct the jury that a missing video was lost through no fault of the requesting party and that jurors should not assume that the lack of corroborating objective evidence undermined [the moving party’s] version of the events at issue.50
In Shaffer v. Gaither, where missing text messages had been read by witnesses, the court decided to allow testimony and cross-examination of those who had seen them, with the jury free to decide whether to believe the testimony.51
This practice can be a slippery slope.52 It can undermine the limitations on adverse inference instructions based on “intent to deprive.” Once a jury hears evidence of spoliation, it may see the parties in a light that unduly prejudices its ability to fairly resolve the issues on the merits. In Delta/AirTran Baggage Fee Antitrust Litigation, the court barred introduction of such evidence because it would “transform what should be a trial about [an] alleged anti-trust conspiracy into one on discovery practices and abuses.” FRE 403 cautions that courts must not “overemphasize” the importance of the missing evidence.53
Adverse Inferences. Subdivision (e)(2) of Rule 37 requires that a court seeking to instruct a jury that it may or must presume missing ESI to be unfavorable should do so only upon finding that the nonpreserving party acted with “an intent to deprive the other party of the use of the evidence in the litigation.” (See “Intent to Deprive,” infra.) The same constraint applies to inferences drawn by the court itself as to missing evidence during a bench trial or other court proceedings.
In a typical case, Virtual Studios v. Stanton Carpet, the court declined to draw an adverse inference or authorize a jury instruction since, at most, the party was “negligent or careless.”54 Some courts defer a decision on the issue until after presentation of evidence.55 However, not all district courts have “gotten the message.” In Bordegaray v. County of Santa Barbara, for example, a court failed to acknowledge Rule 37(e) and imposed an adverse inference without a showing of the requisite culpability.56
It is not unusual or inappropriate for courts to draw additional guidance from existing circuit precedent provided it is not contrary to any express grant of or limitation on the district court’s power established by Rule 37(e). In Edelson v. Cheung, for example, the decision to choose an adverse inference rather than the requested default judgment was based on circuit precedent that cautioned against the use of harsh measures.57
Dismissals or Defaults. The imposition of a dismissal or default judgment (including summary judgments) under Rule 37(e) also requires a showing of an ‘intent to deprive.” In Global Material Tech. v. Dazheng Metal Fibre, a default judgment was imposed where the nonmoving party had been affirmatively deceitful to the opposing party and the court and an adverse inference would not be sufficient to punish the party for their dishonesty.58 Similarly, in Roadrunnrer Transportation v. Tarwater, the Ninth Circuit affirmed a dismissal by the lower court because, had the rule been applied, the lower court findings justified the conclusion that the requirements of Rule 37(e) were met.59
Circuit precedent is often referenced when assessing requests for dismissal under Rule 37(e). In BMG Rights Management v. Cox Communications, supra, a district court relied on Fourth Circuit caselaw for the conclusion that dismissal is “reserved for only the most egregious circumstances,” given the strong policy preference to allow cases to be resolved by trial on the merits.60 Use of circuit authority that does not decrease the minimum requirements under Rule 37(e) is appropriate.
Striking Pleadings/Barring Claims or Defenses. In Feist v. Paxfire, a court barred a party from asserting evidence on a narrow issue likely to have been the subject of the missing ESI and barred an argument that statutory damages should be awarded.61 In Newman v. Gagan, a district judge barred a defense based on the argument that personal devices that had been wiped did not contain documents at issue in a discharge case. Depending on the degree to which these measures are deemed the functional equivalent of dismissals or defaults, the need for a finding of “intent to deprive” is implicated.62
INTENT TO DEPRIVE
Only upon finding that “the [nonmoving] party acted with the intent to deprive another party of the information’s use in the litigation” may a court invoke the harsh measures listed in subdivision (e)(2). This mandate rejects Residential Funding v. DeGeorge, supra, where the court suggested that a “culpable state of mind” requirement in effect in the Second Circuit could be satisfied by a finding of negligence.63
The showing of an “intent to deprive” under subdivision (e)(2) is satisfied by proof similar to that employed to satisfy the “bad faith” requirement in use in some circuits, but is defined even more precisely. In Bracey v. Grondin, for example, the Seventh Circuit noted that bad faith requires destruction “for the purpose of hiding adverse information.”64
The test is not whether the information was intentionally destroyed, but the reason for the destruction. In CTB v. Hog Slat, the court explained that “willful” conduct did not necessarily indicate that the party acted for the purpose of depriving the adversary of the evidence.65 As explained by the Sixth Circuit in Applebaum v. Target, a showing of negligence or even gross negligence will not do the trick under the rule.66
The test moves away from the negligence standard under which any intentional destruction suffices. Indeed, had the rule been in effect, it could well have barred use of instructions such as that utilized in Zubulake v. UBS Warburg (“Zubulake V”), where an adverse inference was authorized merely because employees acted “wilfully” in destroying ESI.67 In Mazzei v. The Money Store, the Second Circuit affirmed that the rule “superseded” in part existing circuit precedent on adverse inferences.68
Where a corporate entity is involved, courts attribute the intent of the employees and agents to the entity when acting within their scope of employment. In First Financial Security v. Freedom Equity Group, an entity was found responsible for acts of its employees based on the logic of a “shared intent” to deprive the moving party of the use of the deleted text messages.69
Examples. In DVComm v. Hotwire, the court found sufficient circumstantial evidence to conclude that the deletion of crucial information was done with an intent to deprive.70 In Brown Jordan v. Carmicle, the requisite intent was found to exist because an individual party with substantial IT experience deleted substantial amounts of information without credible explanation.71
On the other hand, conformance in good faith to a routine policy weighs against such a finding. In Barnett v. Deere & Co, the court refused to find that the destruction of documents in electronic form occurred in bad faith (or with an intent to deprive under Rule 37(e)) where it occurred pursuant to a routine document retention policy.72
Role of the Jury. The Committee Note implies that juries may “play a role” in determining if an intent to deprive exists. One commentator suggests that this is appropriate only if a court first determines that a reasonable jury could conclude that the nonpreserving party acted in that manner.73 In Cahill v. Dart, the court allowed the jury to decide if the prison officials had intentionally allowed a videotape to be overwritten because it was also an element of a malicious prosecution claim that had to be resolved by the jury.74
Standard of Proof. It is not enough to find “intent to deprive” based on “equivocal evidence” about the state of mind.75 Some argue, therefore, that proof of “intent to deprive” should be proven by “clear and convincing” evidence. In CAT3 v. Black Lineage, supra, the court utilized a clear and convincing evidence standard, a choice that was rejected in Friedman v. Philadelphia Parking Authority, 2016 WL 6247470, at ¶58-59 (E.D. Pa. March 10, 2016).
In Chambers v. NASCO,76 the Supreme Court acknowledged that courts have inherent authority to deal with “the full range of litigation abuses.” While a court must exercise caution and restraint in doing so, this authority exists even if procedural rules sanction the same conduct. While courts should “ordinarily” rely on the rules, they may resort to use of inherent power when statutes or rules are not “up to the task.”77 Subsequently, in Dietz v. Bouldin, the Court clarified that the use of inherent authority is not appropriate when “contrary to any express grant of or limitation on the district court’s power contained in a rule or statute.”78
The Committee Note to Rule 37(e) aims to “foreclose” use of inherent authority to determine when “certain measures” should be used.79 The rule itself is silent, however, on that topic. The extent to which the Note compels that conclusion is debatable since Committee Notes are regarded as persuasive but not authoritative.80
The Southern District of New York extensively explored the interplay of inherent authority with Rule 37(e) in CAT3 v. Black Lineage.81 The court interpreted the Committee Note to bar reliance on inherent authority to “dismiss a case for a sanction for merely negligent destruction of evidence.”82 However, it also endorsed use of inherent authority if there is a particularized showing of bad faith spoliation that threatens the integrity of the judicial proceedings, especially when the conduct is not covered by the rule.83 It posited that dismissal or adverse inference was available “under either Rule 37(e) or the court’s inherent authority,” which some see as a rejection of the note’s attempt to restrict inherent judicial power.84 A similar conclusion was reached in Cohn v. Guaranteed Rate, where the court noted that while some of the remedies are listed in Rule 37(e), a court “also has broad, inherent power to impose sanctions for failure to produce discovery and for destruction of evidence over and above the provisions of the Federal Rules.”85
Subject to further clarification by the Supreme Court in Haeger v. Goodyear Tire & Rubber, it seems clear that Rule 37(e) does not preclude reliance on inherent powers to fill “gaps” in the rule or to supplement it by measures that do not contradict its core requirements.86 Such use will, at a minimum, require a finding that the party acted in bad faith. The note, properly construed, merely reflects that principle.
In Shaffer v. Gaither (“Shaffer II”), a court refused to dismiss a case based on its inherent authority where the alteration of text messages did not “rise to the level” of decisions interpreting Chambers to allow dismissal when a party “deceives a court or abuses the process at a level that is utterly inconsistent with the orderly administration of justice or undermines the integrity of the process.”87
Courts have not, as hoped, drawn on experience under Rule 37(e) to provide guidance in the nonelectronic context. This leads to rulings in cases like Czuchaj v. Conair, where a mere delay in instituting a litigation hold was sufficient, without proof of prejudice or of an intent to deprive, to justify imposition of an adverse inference about the immaterial loss of tangible property.88
There is no principled reason why the assessment of spoliation should turn on the form of the evidence. Earlier concerns about the drafting difficulties are no longer valid. Experience since Dec. 1, 2015, has shown that a loss of tangible property that is “central” to the case only occurs rarely and, when it does, it can be dealt with either through the severe measures available to remediate prejudice or, if the rule is not “up to the task,” by the appropriate use of the court’s residual inherent powers.
The limitation to losses of ESI encourages perverse behavior at the margins. Some courts do not apply the rule when digital information is part of a physical medium, even when they acknowledge its digital aspects.89 Expanding the rule would end the uncertainty and confusion about surveillance videos, cell phones, digital photographs, and the like. It also would deal with cases where courts and parties now are required to apply “two standards for spoliation” — often with different outcomes — when documents and ESI are not preserved in the same case.90 As the then-Chair of the Rules Committee noted at the time of Rule 37(e)’s adoption, “if it works [as to ESI], we can think seriously about extending it to other forms of information.”91 Now is the time to do so.