Here is my third annual list of the top 10 e-discovery developments and trends from the past year.
1. The growth of Bring Your Own Device (BYOD) policies and work-related text messaging is creating litigation hold challenges. A Cisco survey found that 89% of companies are currently enabling employees to use their own electronic devices for work. Gartner predicts that by 2017 a half of all employers will require employees to provide their own devices. The growing prevalence and convenience of personal devices in the workplace is leading more employees to use text messaging for work-related purposes.
With these trends, it is no wonder that there were a number of decisions last year addressing whether an employer must produce ESI (mainly text messages) from its employees’ devices (mainly cell/smart phones). One of the key issues in these cases is whether the employer has “possession, custody, or control” over the devices. To decide this issue, courts have looked at whether the employer provided the devices, whether the employees used the devices for work-related purposes, and whether the employer otherwise had any legal right to obtain ESI from the devices on demand. Other issues that have been raised are the privacy rights of the employees and the employer’s obligations if its employees refuse to turn over their devices during discovery.
In ordering the production of business-related text messages on employees’ cell/smart phones, a court rejected the argument that the failure to preserve text messages should not be sanctioned because they are a “less prominent form of communication” and that “the production of text messages is too burdensome.” According to the court, the defendants “had a duty to ensure that their employees understood that text messages were included in the litigation hold” because “[i]t is certainly common knowledge that texting has become the preferred means of communication.” In Re Pradaxa (Dabigatran Etexilate) Products Liability Litigation, MDL No. 2385 (S.D. Ill. Dec. 9, 2013).
When the defendants raised the reluctance of employees to turn over their personal cell phones for examination as an issue, the court rejected that concern. It said that “[t]he litigation hold and the requirement to produce relevant text messages, without question, applies to that space on employees’ cell phones dedicated to the business which is relevant to this litigation.” The court further said that employees refusing to provide their phones to the defendants would be subject to a show cause order to appear personally before the court to explain why they should not be held in contempt.
Other courts reached different conclusions based on the evidence before them. In Cotton v. Costco Wholesale Corp., Case No. 12-2731 (D. Kan. July 24, 2013), the court held that the defendant-employer did not have “possession, custody, or control” over text messages sent or received by its employees on their personal cell phones. The court found that there was no evidence that the employer issued the cell phones to the employees, the employees used the cell phones for any work-related purpose, or the employer otherwise had any legal right to obtain employee text messages on demand.
Likewise, in Kicapoo Tribe of Indians of the Kickapoo Reservation in Kansas v. Nemaha Brown Watershed Joint District No. 7, No. 06-CV-2248 (D. Kan. Sept. 23, 2013), the court found that the plaintiff failed to meet its burden of showing that the defendant had “possession, custody, or control” of computers and other electronic devices personally owned by the defendant’s former and current members and employees. The court also expressed “significant concerns regarding the intrusiveness of the request and the privacy rights of the individuals to be affected” and found that the plaintiff’s “broad, non-specific request to inspect” the computers did not outweigh the privacy concerns of the individuals affected.
In Ewald v. Royal Norwegian Embassy, No. 11-cv-2116 (D. Minn. Nov. 20, 2013), the court issued a mixed ruling. It held that the plaintiff was entitled to receive responsive text messages and voice messages contained on the mobile phone the defendant-employer had provided to her and another employee. However, it refused to order the production of the work-provided mobile phones of 10 other employees of the defendant because the plaintiff had failed to show that any relevant information was likely to exist on those devices.
Other courts did not directly rule on the discoverability of ESI contained on employees’ personal devices, but they issued decisions that had implications for BYOD situations. In Puerto Rico Tel. Co. v. San Juan Cable LLC, No. 11-2135 (D.P.R. Oct. 7, 2013), the court found that the defendant had a duty to preserve relevant emails that came from the personal email accounts of its former officers because it “presumably knew” that the officers used these accounts to conduct company business.
In PTSI, Inc. v. Haley, No. 684 WDA 2012 (Pa. Super. Ct. May 24, 2013), a Pennsylvania appellate court declined to impose spoliation sanctions against two defendants who deleted text messages from their smartphones. The court found that the defendants routinely deleted text messages from their phones to free up storage space and there was “no showing that the innocent cleanup of personal electronic devices to allow them to function was unusual, unreasonable or improper under the circumstances.”
In Christou v. Beatport, LLC, No. 10-2912 (D. Colo. Jan. 23, 2013), the court found that a defendant had a duty to preserve text messages on his iPhone as potential evidence, but it refused to sanction the defendant with an adverse inference jury instruction. The court reasoned it had no basis to assume that the failure to preserve was anything other than negligence. However, the court did permit the plaintiff to introduce evidence at trial of the defendant’s failure to preserve text messages and argue “whatever inference they hope the jury will draw.”
In Barrette Outdoor Living, Inc. v. Michigan Resin Representatives, No. 11-13335 (E.D. Mich. April 26, 2013), the court sanctioned a former employee of the plaintiff because he disposed of his cell phone after his duty to preserve arose.
These decisions illustrate the challenges faced by employers in implementing a litigation hold in a BYOD environment and the need to determine whether key custodians are using their own cell/smart phones or other electronic devices for work-related purposes. If they are, there is a risk that the employer could be held responsible for the preservation and production of text messages, emails, and other ESI created or stored on these devices, whether or not the employer has officially adopted a BYOD program.
2. Predictive coding has moved from an unknown technology without judicial approval to an accepted method for finding responsive documents. A court awarded more than $2.8 million in predictive coding expenses to the defendants based on its finding that the plaintiffs had pursued objectively baseless patent and trade secret misappropriation claims in bad faith. The court found that this method of document review reduced the overall fees and attorney hours incurred by the defendants and, therefore, found the requested amount of fees to be reasonable. Gabriel Technologies Corp. v. Qualcomm Inc., No. 08 CV 1992 (S.D. Cal. Feb. 1, 2013).
Other courts rejected undue burden objections based in part on the availability of predictive coding technology. See, e.g., Chevron Corp. v. Donziger, No. 11-0691 (S.D.N.Y. Mar. 15, 2013) (noting that predictive coding is “an automated method that credible sources say has been demonstrated to result in more accurate searches at a fraction of the cost of human reviewers” and relying in part on its availability to reject a law firm’s undue burden objection to a subpoena served on it); Harris v. Subcontracting Concepts, LLC, No. 12-mc-82 (N.D.N.Y. Mar. 11, 2013) (rejecting an undue burden argument and stating that predictive coding and other technologies could dramatically reduce the time and cost to produce large reams of documents).
It was reported that the Department of Justice’s Antitrust Division approved the use of predictive coding to respond to the DOJ’s request for information during its investigation of Anheuser-Busch Inbev’s acquisition of Mexican brewer Grupo Modelo. Predictive coding permitted counsel for the parties to review over a million documents within two weeks, and the parties spent 50% less than they would have using more traditional review methods. It also was announced that the Securities and Exchange Commission was deploying predictive coding to its Division of Enforcement staff to search and analyze hundreds of terabytes of ESI.
Courts and litigants also recognized, however, that predictive coding is not suitable for every case. In EORHB, Inc. v. HOA Holdings, LLC, No. 7409-VCL (Del. Ch. May 6, 2013), for example, the Delaware Court of Chancery reconsidered its sua sponte order requiring the use of predictive coding. The parties agreed that based on the low volume of relevant documents expected to be produced in discovery by the plaintiffs, “the cost of using predictive coding assistance would likely be outweighed by any practical benefit of its use.” As a result, the court ruled that the plaintiffs could conduct their document review using “traditional methods.”
As predictive coding becomes more common, courts will provide more guidance to litigants regarding how they can use this technology to meet their discovery obligations. In In Re: Biomet M2A Magnum Hip Implant Products Liability Litigation, MDL No. 2391 (N.D. Ind. Apr. 18, 2013), the court considered whether the defendant met its discovery obligations through its use of predictive coding. The defendant used keyword searching and de-duplication to reduce a data set of 19.5 million documents down to 2.5 million documents before it applied predictive coding. The plaintiffs argued that the defendant’s use of keyword searching tainted the whole process and requested an order requiring the defendant to use predictive coding with the full data set. Relying on principles of proportionality, the court rejected the plaintiffs’ argument and found that the defendant’s procedure complied fully with its discovery obligations under Civil Rules 26(b) and 34(b)(2).
3. A lawyer agreed to a 5-year suspension of his license for advising his client to “clean up” his Facebook photos. Courts have largely moved past deciding the issue of whether information posted on social media websites is discoverable. Courts expect parties to preserve this information if it is relevant to the claims and defenses in a case, just like other forms of ESI. See, e.g., Giacchetto-v-Patchogue-MedfordUnion, No. CV 11-6323 (E.D.N.Y. May 6, 2013) (“The fact that Defendant is seeking social networking information as opposed to traditional discovery materials does not change the Court’s analysis.”). If parties do not preserve such information, courts will sanction them.
A court ordered that a “spoliation instruction” be given to a jury at trial, permitting the jury to draw an adverse inference against a plaintiff who permanently deleted his Facebook account during a case. According to the court, the permanent deletion of the plaintiff’s account prejudiced the defendants “because they have lost access to evidence that is potentially relevant to Plaintiff’s damages and credibility.” Gatto v. United Airlines, Inc., No. 10-cv-1090 (D.N.J. Mar. 25, 2013).
The Supreme Court of Virginia affirmed monetary sanctions of $180,000 against a plaintiff and $542,000 against his attorney for spoliation of evidence on the plaintiff’s Facebook page. Allied Concrete Co. v. Lester, 736 S.E.2d 699 (Va. 2013). Following the death of his wife in a car accident, the plaintiff brought a wrongful death action seeking compensatory damages for economic and non-economic losses, including mental anguish. After learning that the plaintiff’s Facebook page had a picture showing the plaintiff holding a beer can while wearing a T-shirt that said, “I ♥ hot moms” across the front, the plaintiff’s attorney instructed a paralegal to tell the plaintiff to “clean up” his Facebook page to avoid “blow-ups of this stuff at trial.” The plaintiff then deactivated his Facebook account, deleted 16 photographs from his Facebook account after reactivating his account, and then lied about having deactivated his account.
As part of the fallout from the Allied Concrete case, the Virginia State Bar Disciplinary Board suspended the lawyer who represented the plaintiff for 5 years for violating professional rules that govern candor toward the tribunal, fairness to opposing party and counsel, and misconduct. The lawyer agreed to the suspension.
Although these cases involved the intentional destruction of Facebook evidence by individuals, corporate litigants should not discount the possibility of spoliation sanctions. Depending on the circumstances in a case, the duty of preservation could include corporate social media accounts and the personal accounts of key players. Accordingly, parties should include social media in their litigation hold notices as a potential source of discoverable information, and they should ask about employee use of social media during custodian interviews.
4. Courts permit “discovery about discovery.” A number of courts compelled parties to disclose the search methods they used to find discoverable ESI after refusing to do so voluntarily. See, e.g., Apple Inc. v. Samsung Electronics Co. Ltd., No. 12-CV-630 (N.D. Cal. May 9, 2013) (ordering non-party Google to disclose the list of search terms and custodians used to find and produce documents responsive to Apple’s subpoena); Ruiz-Bueno v. Scott, No. 12-cv-0809 (S.D. Ohio Nov. 15, 2013) (ordering defendants to answer an interrogatory asking them to explain what procedures or methods were used to search for responsive ESI); Uelian de Abadia-Peixoto v. U.S. Dept. of Homeland Security, No. 11-4001 (N.D. Cal. Aug. 23, 2013) (ordering defendants to disclose the “search parameters” — i.e., custodians, sources, and search terms — used to respond to plaintiffs’ document requests); American Home Assurance Co. v. Greater Omaha Packing Co. Inc., No. 11-CV-270 (D. Neb. Sept. 11, 2013) (ordering defendant to disclose the ESI sources it searched and search terms it used after producing only 25 emails); Viteri-Butler v. University of California, Hastings College of Law, No. CV 12-02651 (N.D. Cal. Sept. 30, 2013) (ordering defendant to identify the location and types of information systems with potentially discoverable ESI and the search methods used to identify discoverable ESI).
This “trend” of permitting “discovery about discovery” has led to lively discussions about whether the selection of search terms should be protected from disclosure based on the attorney-client privilege or work product doctrine, whether there needs to be some showing that a party failed to produce responsive ESI before a court orders the disclosure of search methods, and whether parties should similarly have to disclose their seed sets and internal processes when using predictive coding.
The court in Ruiz-Bueno rejected the idea that sharing information about the processes used to search for discoverable ESI would result in the disclosure of privileged communications. “Simply put, discussing how to go about searching for and producing ESI does not ordinarily or necessarily entail revealing confidential client communications.” That court also noted, however, that “not every case will justify directing counsel or a party to provide ‘discovery about discovery.’” Another court questioned placing a burden on the responding party to demonstrate the propriety of its search: “Though courts have on occasion ordered the responding party to provide evidence of a proper search where there was reason to question whether a proper search was conducted, the Court finds questionable the contention that the law places an independent burden on the responding party to provide such evidence in the regular course of discovery.” Brown v. West Corporation, No. 11-284 (D. Neb. Dec. 4, 2013)
When presented with disputes over search parameters, some courts ruled on which search terms the parties should use. In Saliga v. Chemitura Corp., No. 12-832 (D. Conn. Nov. 25, 2013), for example, the court stepped in and resolved the parties’ dispute regarding what search terms should be used because the parties had not cooperated and had spent almost a year just trying to agree on how the defendant should search and produce its emails.
Other courts declined to do so and ordered the parties to reach a compromise. In Uelian de Abadia-Peixoto, the court declined the defendants’ request to review the search parameters in camera because it was “not in a position to accurately assess the reasonableness of the search conducted nor advise on alternative search terms, sources, or custodians.” Similarly, in Fort Worth Employees’ Retirement Fund v. J.P. Morgan Chase & Co., No. 09-3701 (S.D.N.Y. Dec. 16, 2013), the court found that a “court-ordered middle ground” was “impractical” and “inappropriate” given the nature of the request and the complexities of crafting a search protocol. Instead, the court urged the parties to reexamine their positions and work together to create a mutually acceptable ESI search protocol; otherwise, it would “appoint a special master” to recommend a search protocol.
In AMEC Environment & Infrastructure, Inc. v. Geosyntec Consultants Inc., No. 12-2973 (N.D. Cal. July 26, 2013), the court made observations about how one should stage ESI at a hearing regarding the parties’ dispute over search terms and custodians, but it commented that the “[t]he parties of course know their discovery better” and encouraged the parties to meet and confer and try to work things out.
In the context of predictive coding, the court in In Re: Biomet M2A Magnum Hip Implant Products Liability Litigation, MDL No. 2391 (N.D. Aug. 21, 2013), rejected the plaintiffs’ argument that the defendant had an obligation under the Civil Rules to disclose what documents it used in the seed set to train the predictive coding software. The court said: “The [Plaintiffs’] Steering Committee wants to know, not whether a document exists or where it is, but rather how [defendant] used certain documents before disclosing them. Rule 26(b)(1) doesn’t make such information disclosable.” While the defendant “is right that it doesn’t have to identify the seed set,” the plaintiffs are “right that [the defendant]’s cooperation falls below what the Sedona Conference endorses. An unexplained lack of cooperation in discovery can lead a court to question why the uncooperative party is hiding something, and such questions can affect the exercise of discretion.”
In Gordon v. Kaleida Health, No. 08-378 (W.D.N.Y. May 21, 2013), the plaintiffs contended that where a party intends to use predictive coding to assist in the review and production of ESI, it is necessary for the parties to negotiate a protocol to guide the use of predictive coding software for the case. The defendants disputed the plaintiffs’ position based on the general rule that ESI production is within the “sound discretion” of the producing party. Because the defendants agreed to meet and confer with the plaintiffs and their consultants regarding the predictive coding protocol, the court did not address the merits of the plaintiffs’ motion.
These cases show that courts expect some level of transparency into the methods the parties use to find responsive ESI and that courts appear willing to order a party to disclose these methods when there is some showing that the party failed to produce relevant ESI and refused to cooperate with the other side. It remains to be seen how much transparency and cooperation courts will require with predictive coding.
5. Courts awarded e-discovery expenses under cost-shifting statutory provisions and denied these expenses as taxable costs under § 1920(4). The Fourth and Federal Circuits joined the Third Circuit’s reasoning in Race Tires Am., Inc. v. Hoosier Racing Tire Corp., 674 F.3d 158 (3d Cir. 2012), to limit the types of e-discovery costs that a prevailing party can recover under 28 U.S.C. § 1920(4) to “those costs necessary to duplicate an electronic document.” CBT Flint Partners, LLC v. Return Path, Inc., No. 2013-1036 (Fed. Cir. Dec. 13, 2013); Country Vintner of N. Carolina, LLC v. E. & J. Gallo Winery, Inc., 718 F.3d 249 (4th Cir. 2013). Under this reasoning, e-discovery costs incurred to do such things as organizing electronic documents into a database and indexing, de-duplicating, filtering, analyzing, searching, and reviewing these documents are not recoverable under § 1920(4).
The Federal Circuit said that “larger-scale shifting of litigation expenses” must be “addressed under other statutory provisions that set particular standards for particular types of cases to implement context-specific policies.” Consistent with the Federal Circuit’s approach, there were a number of cases last year where courts awarded non-taxable e-discovery costs based on a cost-shifting provision in a substantive statute.
In Scentsy, Inc. v. B.R. Chase, LLC, No. 11-CV-249 (D. Idaho Aug., 26, 2013), the court awarded e-discovery costs to the prevailing party in a trademark and copyright infringement case because “[t]he Lanham Act and the Copyright Act allow recovery of reasonable costs that are otherwise non-taxable under 28 U.S.C. § 1920(4).” Moreover, in United States ex rel. Becker v. Tools & Metals, Inc., No. 05-627 (N.D. Tex. Mar. 31, 2013), the court awarded the plaintiff $174,395.97 for uploading ESI and creating a search index and $271,110.23 for hosting costs under 31 U.S.C. § 3730(d) of the False Claims Act because the act does not limit the types of expenses that are recoverable.
In Gabriel Technologies Corp., discussed above, the court awarded not only the costs the prevailing party incurred for using predictive coding but for using contract attorneys to review documents. The court awarded these costs pursuant to 35 U.S.C. § 285, which permits a court to award “reasonable attorney fees to the prevailing party” in “exceptional cases” involving patent claims, and pursuant to Section 3426.4 of California’s Uniform Trade Secret Act, which permits a court to award reasonable attorneys’ fees to the prevailing party if “a claim of misappropriation is made in bad faith.” Cal. Civ. Code § 3426.4.
Interestingly, not long after its decision interpreting the scope of § 1920(4), the Federal Circuit issued a decision making it easier for defendants to recover fees under 35 U.S.C. § 285 for “exceptional cases.” The court said that a defendant does not need to prove that a plaintiff actually knew that its patent infringement claims were baseless. A defendant need only prove reckless conduct. The court also said that subjective bad faith may be shown in light of the totality of the circumstances with particular attention paid to the objective merits of the claims and other objective evidence indicative of bad faith. Kilopass Technology, Inc. v. Sidense Corp., No. 2013-1193 (Fed. Cir. Dec. 26, 2013). The U.S. Supreme Court will have more to say on this issue as it has granted certiorari in two cases involving the determination of whether a case is “exceptional” under § 285. Highmark Inc. v. Allcare Management Systems, Inc., Docket No. 12-1163; Octane Fitness, LLC v. Icon Health & Fitness, Inc., Docket No. 12-1184.
Stay tuned for part 2 of the “Top 10 e-discovery developments and trends in 2013” — scheduled for publication on Porter Wright’s Technology Law Source blog tomorrow, Jan. 15.