Big data yields big data breaches, and potentially produces large class sizes, making such lawsuits attractive to plaintiffs’ lawyers. Companies that store or process personal information face an increasing risk of class action lawsuits based not only on the company’s use of that information, but also on the theft or misuse of that personal information due to data breach. Many states, such as California and Delaware, which have liberal data breach laws that allow private rights of action for security incidents regardless of a likelihood of injury, have facilitated class action lawsuits. In one such case, plaintiffs sought damages of $5,000 per customer from the defendant, which could have resulted in possible damages totaling in the tens of billions of dollars – far more than the defendant company was worth. A recent survey of data breach litigation found that the average settlement award in these cases was approximately $2,500 per plaintiff, with mean attorneys’ fees reaching $1.2 million.1 In spite of these risks, companies may be able to avoid class certification if the plaintiffs fail to establish standing to bring suit on behalf of a class. A pivotal question for standing is establishing injury-in-fact, which has successfully prevented certification of many purported data breach class actions. Recent cases, however, have been breaking down the court’s resistance to class certifications, raising the stakes in data breach and privacy cases.
Data Breach Class Actions
Data breaches – the theft or loss of secure information – have become an all-too-common event; most companies should anticipate having a data breach. One report identifies the loss of 174 million data records in 855 separate incidents in 2011 alone.2 Another security study found that 90 percent of the companies and organizations surveyed had at least one data breach.3 Further, the advent of “big data” and cloud computing can mean enormous losses of data from a single breach, which can equal very large classes of potential plaintiffs. The plaintiff’s bar is watching data breaches very closely. A lawsuit was filed within 24 hours of Zappos.com reporting a data breach. In spite of these daunting statistics, it may be possible for companies to dismiss class action suits if there has been no injury-in-fact resulting from the data breach.
There is a disagreement between circuits regarding the need for actual harm to establish standing on the part of the plaintiff representing a class. For example, the Ninth Circuit has held that the theft of personal information that has not been misused can be sufficient to grant plaintiffs standing if they can show a “credible threat of real and immediate harm stemming from the theft of [their data].” Krottner v. Starbucks Corp., 628 F.3d 1139, (9th Cir. 2010). In Krottner, a laptop containing the plaintiffs’ unencrypted personal data had been stolen, yet the plaintiffs did not claim misuse of the data, but only alleged an “increased risk of future identity theft.” The Krottner court held that that this risk constituted an injury-in-fact, and was enough to confer standing on plaintiffs.
The First Circuit has also held that the cost of plaintiffs’ efforts to mitigate the risk of identity theft following a data breach were foreseeable damages in their claims of negligence and implied breach of contract. Anderson v. Hannaford Bros. Co., 659 F.3d 151 (1st Cir. 2011). In Anderson, a New England supermarket chain was hit by a sophisticated attack that directly targeted their customers’ data, resulting in the theft of 4.2 million customer credit card numbers. The customers filed suit, alleging that they had experienced more than 1,800 unauthorized charges to their accounts, and had incurred other losses, including fees for card replacement, overdrafts, and the cost of purchasing identity theft insurance. The First Circuit reversed the lower court’s dismissal of the suit, reasoning that since this breach involved a sophisticated attack aimed directly at customer data, and since that data was actually used to commit fraud, plaintiffs’ mitigation efforts were reasonable, and were therefore foreseeable damages.
These cases do not give class action plaintiffs a free pass, however – plaintiffs must demonstrate actual injury-in-fact. In Whitaker v. Health Net of Cal., Inc., an Eastern District of California court, acknowledging Krottner, held that the plaintiffs lacked standing to bring their class action data breach claims, as they failed to assert that they suffered “an injury in fact.” 2012 U.S. Dist. LEXIS 6545, *9 (E.D. Cal. 2012). In Whitaker, the defendants “lost nine Health Net server drives consisting of more than 800,000 California residents’ personal and medical information, including plaintiffs’ information.” The plaintiffs argued that there was difference between the theft of personal data and its loss. The court disagreed, observing that plaintiffs “[did] not explain how the [data] loss here has actually harmed them or threatens to harm them, or that third parties have accessed their data.”
The Third Circuit has also distinguished Krottner, holding that the plaintiffs’ “allegations of hypothetical, future injury” were not sufficient to establish standing. Reilly v. Ceridian Corp., 664 F.3d 38 (3d Cir. 2011). The Reilly court reasoned that, in the present case, there was “no evidence that the intrusion was intentional or malicious,” and that plaintiffs “alleged no misuse [of the stolen data], and therefore, no injury.”
Most recently, the Eleventh Circuit, breaking through barriers to class actions, has held that allegations of identity theft that cause monetary damages are an injury-in-fact sufficient to give plaintiffs standing in a putative class action. Curry v. AvMed, Inc., __ F.3d ___ (11th Cir. September 5, 2012). In Curry, two laptops containing unencrypted sensitive information for over one million customers were stolen from a health plan provider. Within fourteen months, two of the health plan’s customers – the named plaintiffs in the class action – suffered identity theft. Based on the plaintiffs’ claims that they guarded their private information carefully, they had never had their personal data compromised in the past, and only suffered identity theft shortly after the defendant’s laptops were stolen, the court held that the plaintiffs’ injury was enough to confer standing. The court noted, however, that “[h]ad Plaintiffs alleged fewer facts, [the court would] doubt whether the Complaint could have survived a motion to dismiss.”
Unusual complications can arise in privacy-related class action lawsuits. For example, it may be difficult to find judges or jurors without conflicts of interest in cases in which a large segment of the population may be potential members of a plaintiff class. This situation may be more common than not – consider how many of us may have protected personal data residing with large banks, hospital networks, and health plans, not to mention social media and search companies such as Facebook and Google. In a class action suit against Sutter Health in Sacramento, CA, where a health plan computer containing data for 4.24 million patients was stolen, there was a delay in scheduling while a judge was found who did not have his or her own data stolen in this theft.
Privacy Class Actions
In December 2009, a class action was brought against Netflix in the Northern District of California, claiming that Netflix violated fair trade laws and federal privacy law protecting video rental records when it disclosed insufficiently anonymized information about their customers as part of its recommendation system. Doe v. Netflix, Inc., (N.D. Cal. 2009). The complaint sought over $2,500 in damages for each of Netflix’s over two million customers.
The Netflix complaint stemmed from the company's efforts to improve its movie recommendation algorithm by offering a prize to anyone who could come up with a system that provided recommendations that were 10 percent better than Netflix’s current system. As part of this contest, Netflix provided contestants with enormous data sets, which included over 100 million movie ratings along with a unique ID number for the subscriber who gave the rating. The plaintiffs alleged that by sharing this insufficiently anonymized data, Netflix’s actions amounted to a “voluntary privacy breach,” and were illegal under the Video Privacy Protection Act.
Following discussions between Netflix and the Federal Trade Commission, Netflix settled the case in March 2010. Under the terms of the settlement, Netflix paid $9 million into a fund to benefit privacy groups, and agreed to change its data retention policies to no longer link customers to their rental histories for more than one year after customer cancellation. Netflix’s revised privacy policy took effect in July 2012.
In September 2011, plaintiffs brought a class action suit against the online music streaming company Pandora, alleging first that Pandora’s disclosures of their subscribers’ personal information violated Michigan’s Video Rental Privacy Act (VRPA), and second, that Pandora violated the Michigan Consumer Protection Act (MCPA) by disclosing subscriber personal information through its integration with Facebook. Deacon v. Pandora Media, Inc., (N.D. Cal. September 2011). As relief, plaintiffs sought statutory damages under the VRPA of $5,000 per class member, which could have potentially resulted in total damages of tens of billions of dollars.
The Pandora court dismissed the lawsuit on September 27, 2012, however, stating that the plaintiffs’ claims under the VRPA were insufficient, and that the lack of any “actual injury” in this case prohibited a class action under the MCPA. The court pointed out that even though plaintiffs only sought injunctive relief under their MCPA claim and not money damages, there was “no provision in the [Michigan law] permitting a class action solely for injunctive relief by a person who has not suffered actual loss. Had the Michigan Legislature intended to provide such a remedy, it plainly could have done so.”
The greater a company’s use or retention of protected customer data, the greater the risk of exposure to these class action suits. For example, in February and March of 2012, two separate privacy-related class action complaints were filed against Google. The first complaint, filed in the U.S. District Court for Delaware on February 17, 2012, claims that Google willfully violated the Federal Wiretap Act, the Stored Electronic Communication Act, and the Federal Computer Fraud and Abuse Act by circumventing certain privacy features of the Safari web browser. Soble v. Google, Civil Action No. 1:2012cv00200 (D. Del. Feb 17, 2012). The second class action complaint, filed on March 20, 2012 in the Northern District of California, claims that Google violated several federal privacy laws when it merged its separate privacy policies for multiple products and services into one. DeMars v. Google, Civil Action No. 5:2012cv01382 (N.D. Cal. March 20, 2012). In August 2012, Google moved the DeMars court to dismiss the claim in that court, stating that, inter alia, plaintiffs have not sufficiently pled any actual economic harm. On October 9, 2012, a magistrate court heard arguments from Google in support of their motion, with an order scheduled to follow. No doubt Google will be the target of continued scrutiny.
Conclusion
The likelihood of a data breach or privacy issue occurring in any business has become a virtual certainty. Class action lawsuits stemming from such incidents have upped the ante with the potential of millions of dollars of attorneys’ fees if not damage recoveries. All companies would be prudent to increase their risk mitigation efforts to beef up administrative, technical, and physical security to prevent data breaches coupled with enforcing security and privacy policies and procedures and strengthened indemnification provisions with third parties who have access to a company’s data. Such measures may also serve to convince a court that no likelihood of actual damages from an actual injury-in-fact exists upon which a class action lawsuit can be based. Companies should also evaluate their insurance coverage, and confirm that they have a liability policy in place that specifically covers the costs associated with data breaches and related incidents.
Endnotes
1 Sasha Romanosky et al., Empirical Analysis of Data Breach Litigation, Temple University Beasley School of Law, Legal Studies Research Paper No. 2012-29 (2012). Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1986461.
2 Verizon RISK Team, 2012 Data Breach Investigations Report (2012). Available at http://www.verizonbusiness.com/resources/reports/rp_data-breach-investigations-report-2012_en_xg.pdf.
3 Ponemon Institute, Perceptions About Network Security (2011). Available at http://www.juniper.net/us/en/local/pdf/additional-resources/ponemon-perceptions-network-security.pdf.
