Editor: Tell us about your practice.
Austrian: I represent companies in complex civil cases, where there is a very heavy emphasis on locating, organizing and presenting complex information to a judge or jury. Over the years, I have handled patent, environmental, product liability, toxic tort, commercial and other disputes. I am an adjunct professor at Washington College of Law, American University, teaching courses in e-discovery and advanced trial advocacy in the high-tech courtroom, and a member of the Georgetown Law School E-Discovery Advisory Board and of the Sedona Conference Working Group on Electronic Document Retention and Production.
Editor: Why is the cost of e-discovery of such great concern to corporate counsel?
Austrian: Virtually all of a corporation’s information is now created and retained in an electronic format (electronically stored information or “ESI”) of one type or another. The amount of ESI has increased exponentially over the years, and the methods by which ESI is stored and organized constantly changes over time as new technologies, such as cloud computing, are introduced. ESI is often dispersed throughout the company. Relevant information may be held by numerous custodians, stored in different systems and devices, and created in different formats. This combination of size, evolving technology and dispersal makes it difficult and expensive for large multinational corporations to locate, preserve, analyze and produce relevant information for a particular lawsuit, especially where the ESI is stored on older legacy systems.
This information must also be carefully reviewed for relevance and privilege. A recent Rand study reported that over 70 percent of the e-discovery costs are for attorney review. Therefore, the larger the amount of information that is initially collected for attorney review, the higher the cost. Companies can only lower this cost by implementing procedures and processes that reduce the amount of information initially collected for a particular case.
Adding to these costs are issues relating to cloud computing where the ESI is held by a third-party provider and is not actually under the company’s direct control. However, the courts have held that in this situation the company has possession, custody or control of this ESI for litigation purposes. ESI stored outside of the U.S. adds additional complexity because privacy restrictions often make it difficult to transfer relevant information to the U.S. and to adverse parties.
Editor: Would it be fair to say that the most significant developments contributing to imaginative thinking about reducing e-discovery costs are predictive coding and the Iqbal and Twombly cases?
Austrian: The Supreme Court’s decisions in Iqbal and Twombly were certainly influenced by the high cost of e-discovery and sought to limit cases where the plaintiff could not present sufficient facts to justify ongoing litigation. However, the much more significant issue relates to reducing costs through technologically assisted review, i.e. predictive coding. This process uses algorithms in place of attorneys to identify and review relevant ESI.
Predictive coding is a general, descriptive term for different approaches created and promoted by different vendors. The publicity surrounding predictive coding has gotten the attention of judges, as we saw from the Delaware Chancery Court’s recent decision in EORHB, Inc. v. HOA Holdings LLC, where on his own, Vice Chancellor Laster required the parties to apply predictive coding and employ a joint vendor. Over the next five to ten years, predictive coding will probably be the most important technological development for reducing e-discovery costs.
Editor: Please describe the amendments recently adopted by the District of Delaware to its Default Standard for Discovery and their implications.
Austrian: They were called default standards for a good reason. They did not require that the parties follow a particular process but present a pretty good indicator of how the court will most likely rule if the parties fail to reach agreement. These standards cover, for example, the types of ESI that need not be produced, the number of custodians contacted, production formats, search methodologies, preservation, privilege and privilege logs, and many other areas that are often subject to an agreement between the parties.
Patent cases are often situations where exchanging ESI can be especially costly and time consuming, and may involve information that goes back 15 or 20 years. This is one of the few districts with special requirements directed to managing patent cases early in the discovery process. For example, the standard covers the early production of claim charts and requires the early identification of infringement and contentions of invalidity. The Federal Circuit also has been very active in this area in creating e-discovery rules with respect to patent cases.
Editor: To what extent can the action of the District of Delaware be viewed as reflective of a more general effort by the federal courts to reduce the burdens of e-discovery through self-help rather than awaiting formal amendments to the FRCP?
Austrian: The district courts are well aware of the large debate between companies and judges that want to keep the federal rules broad and those who want more specificity, especially as they related to sanctions. Judges are concerned about making the rules so specific that they are overburdening counsel in the 95 percent of the cases that don’t raise significant e-discovery issues. It is highly unlikely that, beyond the changes implemented in 2006, the federal rules will be amended to reflect the specific requirements now contained in the local rules. There are now more than 40 federal district courts with local rules. These were enacted not only to reduce the costs of e-discovery but also to provide judges with some direction on how to deal with the complex technological, legal and practical issues they are being asked to decide.
Some districts, like the Southern District of New York, have pilot projects with very detailed processes, including requirements that the parties fill out forms before they even get to a conference with the court. One of the real concerns of many judges in the Southern District is that the proposed rules are making these cases more, not less, expensive to handle.
There are also serious concerns in the practicing bar about inconsistencies in these local rules. However, there is great reluctance to amend the federal rules because it might reduce the parties’ incentive to create solutions that may be unique to their individual cases. So while the local courts are continuing to add to their local rules, I think it’s unlikely that we’ll get significant changes to the federal rules, beyond those changes in 2006, in the near future.
Editor: Do counsel draw upon the principles that are being forged in the various courts?
Austrian: Yes. The local rules provide an illustration of the agreements that can be reached. Almost all the federal judges want to see that the concepts of cooperation and proportionality embedded in these local rules are implemented. That is quite different from the way things were before ESI became so important.
In terms of cooperation, these rules give examples of how cooperation can and ought to work. They are also a direct reminder to lawyers that if they don’t get involved in negotiating and coming up with realistic solutions, judges will impose them – and they may not like what the judges do. Companies and their clients are incentivized to think seriously about the consequences of not reaching agreement before asking the court to rule.
Editor: Do you think clients are sensitive to the fact that e-discovery burdens may be eased by appropriate advocacy?
Austrian: Yes. Corporations are demanding that counsel pay close attention to e-discovery developments so that when there are discussions with opposing counsel or the court, they have sufficient knowledge of the company’s ESI processes and procedures to effectively negotiate reductions in the amount of data that must be produced, and apply technology, like predictive coding, to reduce cost. This is an area that requires creative solutions, and counsel without experience cannot propose and implement them.
Editor: Overpreservation is a grave concern.
Austrian: This is one of the major areas of concern where the fear of spoliation and sanctions has created havoc. Companies have gotten into trouble by waiting for litigation to start before they actually start planning for e-discovery and its consequences. This includes creating internal teams with IT, legal and management representation to develop written document retention plans, litigation hold policies and internal processes to respond to e-discovery. Companies must decide whether to conduct the processing of ESI in-house or hire an outside vendor. This cannot be done at the last minute when, on a parallel track, the company is considering and developing its legal defenses. The team should also document the processes for analyzing and preserving relevant information.
Certainly the fear of spoliation leads to extensive overpreservation. But I think that companies are recognizing that it’s the policies and procedures they put in place to manage their data that will protect them from sanctions, allow them to reduce the amount of data preserved and, while spoliation claims can always be made, convince the courts that they should not be imposed. Courts do not like to impose sanctions when the company has acted reasonably, even when mistakes have been made.
Editor: The bulk of litigation occurs at a state level and not in the federal courts. What’s happening at a state level? Are the same trends becoming evident?
Austrian: Your observation is very appropriate. One of the risks, especially in state courts, is that of imposing detailed e-discovery rules, which make relatively small cases expensive. In federal courts, roughly 95 percent of all cases would be considered small cases. I’m sure that number is significantly higher in state cases. Many states have put into place local e-discovery rules (the last count was about 19 of them), but those rules tend to be significantly less detailed than those we’ve talked about at the federal level. State judges tend not to have the same opportunity to learn about the technological issues and are even less likely to get heavily involved in them.
Editor: I assume that U.S. judges get hot under the collar if for privacy or other reasons access cannot be gained to ESI in other countries.
Austrian: Increasingly, due to cloud computing, information may be stored by third-party providers in a foreign country. These countries, especially those in the EU, consider that since the information is maintained there, its privacy laws govern and you can’t simply transfer ESI to the U.S. without insuring that its laws are complied with. This is true even if the information may not necessarily have been generated in that foreign country. In cloud computing, for example, third-party providers may store information in servers located in the cheapest locations they can find. Companies don’t necessarily know where those servers are located. Thus, costs can increase when information is either stored or generated in a foreign location.
The Supreme Court has ruled that discovery cannot be denied simply because it is located in an uncooperative foreign jurisdiction. U.S. counsel, with the help of foreign counsel, must present to the court ways to get that information, for example, by conducting the discovery in phases and entering protective orders that will protect the privacy of information coming from abroad. With protective orders in place, counsel can go to the data protection officials in a foreign country and demonstrate to them that their privacy concerns will be satisfied.
What we are seeing is much more creative solutions being agreed to by the privacy data officials in the EU. When you have demonstrated to the judge that you have worked hard to get information residing in foreign jurisdictions, even if you are not successful, there is greater likelihood of judicial cooperation and a greater focus on whether this information is really necessary.