Special Masters and the Future of e-Discovery

A new article in the Cardozo Law Review lays the groundwork for greater use of special masters in e-discovery. This was not written by a student, nor a professor (professors still remain largely mute on e-discovery), but by Judge Shira A. Scheindlin and Jonathan M. Redgrave. The article is entitled Special Masters and e-Discovery: the Intersection of Two Recent Revisions to the Federal Rules of Civil Procedure. The authors and law review have allowed me to post a copy of it here. 

This is an important subject because Special Masters represent a practical way out of the competency morass the legal profession now finds itself in. Eventually, the Academy will wake up and professors will begin teaching students to practice law in the Twenty-First Century, instead of pretending like we still live in a paper world. Eventually trial lawyers will stop doing the same. But as things stand now, most professors and trial lawyers seem to be in cahoots to try to perpetuate the legal practice of Abraham Lincoln. (As a student of Lincoln, this seems especially ironic to me, since he was known for embracing the new technologies of his day.)

When the legal profession finally catches up with the information explosion and learns to embrace technology like Lincoln did, not shun it, then Specials Masters may become obsolete. But in the meantime, say for the next ten to twenty years, they will be the Jedi Knights of the trial world, sworn to protect litigants from the Dark Lords of extorted settlements, smoke and mirror cons, and outrageous e-discovery costs.

The Role of Special Masters in Civil Litigation

Special Masters are appointed by the courts under Rule 53, Federal Rules of Civil Procedure, to act as assistants to the court for all types of issues, including the supervision of e-discovery related issues. In this capacity, Special Masters act as guardians for e-discovery processes by filling the competency void created by one or all of the other actors in the dispute resolution process: lawyers, litigants, and judges. They bring special expertise to facilitate the adjudicative process in a number of ways.

The Special Masters article, at page 374, identifies four different roles for e-Discovery Special Masters:

1. facilitating the electronic discovery process; 
2. monitoring discovery compliance related to ESI; 
3. adjudicating legal disputes related to ESI; and 
4. adjudicating technical disputes and assisting with compliance on technical matters, such as conducting computer/system inspections.

The authors then explain each of these roles in a clear and concise manner with very detailed citations to case law. The article also includes a detailed discussion of three cases where a special master was employed. Hohider v. United Parcel Service, Inc., No. 04-363 (W.D. Pa. Dec. 19, 2007); In re Seroquel Products Liability Litigation, No. 6:06-md-1769-Orl-22DAB (M.D. Fla. Sept. 27, 2007); Medtronic Sofamor Danek, Inc. v. Michelson, 229 F.R.D. 550, 552 (W.D. Tenn. 2003).

Facilitating Process

The authors’ research, assisted by Judge Scheindlin’s law clerk, Rachel Spector, found that Special Masters have been appointed to facilitate the e-discovery process in a number of ways, but especially the following:

1. assisting with the Rule 26(f) conference discussions; 
2. developing preservation protocols; 
3. developing processes to identify locations and sources of potentially relevant documents and ESI; 
4. assisting the parties to develop protocols for the identification and depositions of knowledgeable witnesses regarding ESI issues (including guidelines for the scope of pre-trial examinations); 
5. developing protective orders to address privilege and privacy protection concerns; 
6. addressing search and retrieval issues (such as negotiating search terms); and 
7. agreeing upon form of production issues.

Id. The article points out how a special master appointed to facilitate disputes may also be provided with authority by the court to resolve disputes and issue orders, which then may, or may not if the parties agree in advance,  be appealable to the district court judge.  The primary example used to illustrate this kind of facilitation role is Board of Commissioners of the Port of New Orleans v. Lexington Insurance Co., No. 06 Civ. 8101 (E.D. La. Feb. 19, 2008) (Order Appointing Special Master for Electronically Stored Information). This action seeks insurance proceeds related to the destruction of a dock during Hurricane Katrina. The special master appointed was the best known of them all, Craig Ball. In fact, Craig Ball is the master appointed in several of the cases featured in the article, including the well known class action case In Re Seroquel Products Liability Litigation, 244 F.R.D. 650 (M.D. Fla. Aug. 21, 2007). 

Craig probably has more experience in this area than anyone. I have quizzed Craig closely on how he exercises his special master powers. He tries to avoid issuing orders, even when granted such authority by the court. Instead, Craig uses constructive suggestions and settlement techniques. As is well known among Jedi, the actual violent use of the force is a last resort.

Judge Scheindlin and Jonathan Redgrave opine at page 383 of their article, and I strongly agree, that this will be a fast growing area. They go on to provide a good explanation of the quasi Legal-Technical services special masters can provide:

We expect to see an expanding role for specials masters that facilitate ESI discovery. Such a role will have both legal and technical components. The legal component largely applies the procedural rules to the world of ESI. Lawyers with a good background in e-discovery issues and best practices guides, such as The Sedona Principles, should be able to effectively and efficiently steer the parties toward agreed-upon protocols and procedures. Assisting the lawyers to agree on the scope of preservation, disclosure and production and methods for resolving privilege claims are good examples of collaborative legal assistance.

The technical component focuses on the application of specialized knowledge and skills in information technology and/or computer forensics to assist the parties in handling discovery issues. This may involve providing technical specifications for discovery or actually conducting discovery investigations and searches in a neutral capacity. Specially trained lawyers and non-lawyers should be able to serve in this capacity. Such appointments will advance the ultimate resolution of the case by allowing the parties and the court to focus on the merits of a case as informed by any evidence derived from the electronic sources.

A decision by Judge Scheindlin issued shortly after this article was published shows how a judge can use the possible appointment of a special master to cajole parties to reach an agreement on their own. Securities and Exchange Commission v. Collins & Aikman Corp., 2009 WL 94311 (S.D.N.Y., Jan. 13, 2009). In this interesting case the SEC responded to e-discovery requests by making an electronic document dump of largely irrelevant information. The SEC produced on disks the paper equivalent of 10.6 million pages of ESI, with no organization to the production. The defendants could not afford to review such a massive pile of computer files. The SEC refused to perform a real search of their electronic records themselves, or even to speak with the defendants about it. 

Not surprisingly, Judge Scheindlin did not accept that approach and ordered the SEC to search and produce according to the specific issues requested by defendants. Judge Scheindlin wisely did not attempt to design the specifics of the search and production. The plan would have to meet the requesting parties need for evidence to defend themselves and the producing parties need to keep the production economical in accord with Rule 26(b)(2)(C) (something another federal agency completely failed to do recently in the In Re Fannie Mae Litigation case). As anyone who has ever attempted to formulate such a search plan knows, this is a monumental undertaking, which requires considerable skill and many, many hours of work. Instead, Judge Scheindlin ordered to parties to meet in good faith to work it out themselves:

While the SEC has raised legitimate concerns about the burdens imposed by particular requests, it cannot unilaterally determine that those burdens outweigh defendants’ need for discovery.  At the very least, the SEC must engage in a good faith effort to negotiate with its adversaries and craft a search protocol designed to retrieve responsive information without incurring an unduly burdensome expense disproportionate to the size and needs of the case. The parties are therefore directed to engage in a cooperative effort to resolve the scope and design of a search with respect to the rebate issues and a search of e-mail created and maintained by the SEC.  

Judge Scheindlin gave them a deadline to reach agreement and raised the specter of  a special master should they fail to agree.

If the parties remain at an impasse, the Court will be prepared to resolve further disputes and will consider the appointment of a Special Master to supervise the remaining discovery in this case.

I expect you will see this kind of “threat” to hire a special master used more and more frequently by judges. There is more going on here than meets the eye. Some recalcitrant parties will stubbornly refuse to cooperate because they know the judge assigned to the case does not have the time or skills needed to resolve the dispute. They take advantage of that and give mere lip service to the judge’s admonitions; always trying instead to make it look like the problems are caused by the other guy, not them. Mutual finger pointing is, after all, the norm in most courtrooms and so this tactic often works. The attorneys know that the judge will never take the time to figure out who is really right. In fact, the judge might be persuaded by mere puffing and convincing pointing to rule their way, or more likely, simply “split the baby.” Either way, it is a win for the offending party and a loss for justice.

Now the mere threat of a special master can defeat this kind of unethical “smoke and mirrors” ploy. The special masters threat is no paper tiger. Special masters have the skills and time to figure out what is really going on. When they are hired the parties will have to pay for it in more ways than one. This is why honest practitioners cheer the arrival of special masters and why I think of them as a new kind of legal Jedi.

Monitoring Compliance

On many occasions special masters have been used in a different capacity to monitor and bulldog compliance with prior court orders. In these situations, the parties have usually already made quite a mess of discovery and one side (but sometimes both) has been ordered to provide discovery. This is typically in very complex situations where it is not so easy to determine whether or not the enjoined party is complying in good faith with the order. That is where the special master typically comes in. The appointment of the master in these situations is part of the sanction and indicates the court’s disgust with all of the time and judicial resources the parties have already consumed with the e-discovery dispute. Typically, the losing party pays for all of the special master’s fees and expenses, which is unlike the friendly facilitator role where the masters fees are usually divided between the parties.

The main case which the article uses to illustrate this situation is the disaster ERISA litigation case, which I have previously written about, Wachtel v. Health Net, Inc., 239 F.R.D. 81 (D.N.J. 2006). The special master was appointed in Wachtel after repeated violations of discovery orders by the insurer. The court found the insurer’s actions had placed an “extraordinary drain on the Court’s resources,” in light of which “the need is clear for help in the form of a separate Special Master to monitor discovery compliance to ensure that all documents ordered to be produced have been produced and that all of the Court’s discovery Orders have been complied with.” Needless to say, this is never a position you want to find yourself in.

Adjudicating Discovery Disputes

The next role of the special master is a kind of “rent-a-judge” situation where the parties, or the court, delegate the resolution of certain defined issues to the special master to take evidence and make a decision. Typically, the decision is in the form of a report and recommended order, and the parties can then object with the district court judge. But if they want, typically in situations where time is of the essence, the parties can also agree in advance to waive or at least limit that right. The idea is to choose a person to resolve the dispute who has special expertise and knowledge, not to mention time, to provide a prompt and fair resolution. Typically a court will try to have the parties agree on the special master, but if not, one is selected by the court for them.

There is, of course, a considerable cost involved in the payment of the special master fees, which is typically done on an hourly basis. Still, most observers of the process, myself included, believe that assigning the dispute to a specialist will allow for a much quicker, and overall less expensive resolution of the dispute. It should also result in a significant improvement in the quality of the adjudication. I mean no slight to judges by the comment, but it is not reasonable to expect a judge to handle hundreds of different kinds of cases, from criminal law, to anti-trust, to civil-rights, and then move onto a highly technical e-discovery dispute involving large, arcane computer systems, and handle them as well, or as quickly, as a person who only works in e-discovery.  Sure, they can do it, but the odds are it will take much longer, and they are more likely to be confused by competing expert testimony and reach the wrong result. A special master may sometimes require expert testimony too, but often times not, or at least not as much. Most importantly, their ability to understand the testimony is likely to be much better than a generalist trial judge or magistrate.

The article discusses a number of different cases where special masters have been appointed to resolve disputes, sometimes very narrow and specific, but also sometimes fairly broad in scope. They also note that disputes concerning privilege particularly lend themselves to this procedure and special masters have been used many times to resolve complex or voluminous privilege disputes. In e-discovery, the number of documents at issue in a privilege dispute can be astronomical. For instance, In re Vioxx Products Liability Litigation, 501 F. Supp. 2d 789 (E.D. La. 2007) involved a log of over 30,000 documents withheld by the defendant, Merck. The court’s in camera review of confidential documents was the only way to resolve these privilege disputes, but no judge has the kind of time necessary to perform this function. In this situation, there was no question the judges has the necessary expertise to do the privilege review, but not the time; therefore, appointment of a special master was the best solution.

The article, at pages 384-385, elaborates on the adjudicative type services that  special masters can provide:

The supervision of discovery disputes is very time consuming. A district judge rarely has the time to provide the hands-on supervision such disputes often require. For example, particularly sensitive or contentious depositions may need on-site supervision in order to quickly resolve disputes. On-site inspections of computer systems by opposing counsel or experts often require careful protocols and supervision in order to protect confidentiality and privacy. Ruling on objections to interrogatories and document requests and resolving disputes regarding the scope of non-party discovery might also require a significant investment of judicial resources. Similarly, an in camera review for privilege can require a judge to carefully read thousands of documents. Finally, many discovery issues that arise in the context of e-discovery require a court to develop a sophisticated understanding of the accessibility of ESI and the costs and burdens of preserving that data and/or retrieving it and translating it into a useable format. There is also the question of whether a party should be sanctioned for its failure to produce ESI. It is these tasks that trial judges often assign to a judicial adjunct, be it a magistrate judge or a special master.

Adjudicating Technical Disputes and Assisting with Discovery Compliance

This last role for a special master is all about their skills; sometimes purely technical skills having little to do with the law. In fact, you will sometimes see courts appoint forensic engineers and other technical specialists with no legal background to resolve purely technical disputes. In this situation, they are sometimes called technical advisors, not special masters, and they provide expert reports to the court.

Conclusion

The law review article by Judge Shira Scheindlin and Jonathan Redgrave,  Special Masters and e-Discovery: the Intersection of Two Recent Revisions to the Federal Rules of Civil Procedure, is well worth reading, not only for information about special masters, but also for the excellent review of general e-discovery law and the new rules. Section Two of the article is a handy reference for all of the key issues in e-discovery, including most of the leading cases. For that reason alone this article will likely be often cited. But its key role is to lay the legal groundwork for the expanded use of special masters in state and federal courts around the country. The article opens the door for the few prepared experts in the area to assist the courts and help clean up the mess we are now in. Here is the excellent final paragraph in the article, which will, I hope, entice you to download and read the entire piece:

Generalist judges are not and cannot be experts on electronic hardware and software that enable people to create, store, retrieve, and search ESI. In order to manage discovery in cases involving a substantial amount of ESI, and assuming the stakes at risk warrant the cost, court adjuncts with specialized knowledge may become more the rule than the exception. Thankfully, revised Rule 53, which is no longer tethered to the concept of “exceptional condition,” now permits such appointments whenever the parties consent to an appointment or when the court determines that a pre-trial or post-trial matter “cannot be effectively and timely addressed by an available district judge or magistrate judge of the district.”  While we surely do not suggest that such appointments are needed in every case, or even the majority of cases, there will be a subset of cases—often (but not always) involving many parties and/or non-parties, voluminous records, or complex or technical issues—that will benefit from the availability of these adjuncts. The appointment of a special master may make it possible to accomplish in days what would otherwise consume months of litigation and require both sides to incur substantial costs. The key is finding a person who understands his or her role, has the skills to fulfill it, and is accorded the appropriate level of authority and respect to get the job done. If such appointments ultimately reduce the costs of litigation by resolving disputes over the scope of discovery (accessibility), the form of production, and a rational approach to search and retrieval, then the appointment will have been successful. Another measure of success will be a marked decrease in sanctions motions, which many in the legal community fear has become a prime goal of discovery—i.e. not to find documents but to learn that documents cannot be found and as a result someone should be punished. These benefits counsel in favor of the expanded use of special masters to assist in e-discovery matters in appropriate cases. (footnote omitted)

 

Did You See Judge John Facciola’s Christmas Present?

JOHN M. FACCIOLA, U.S. Magistrate Judge for the District Court in Washington D.C., worked late Christmas eve to give us all a present; he wrote yet another memorable e-discovery opinion: Covad Communications Co. v. Revonet, Inc., 2008 WL 5377698 (D.D.C. Dec. 24, 2008). The opinion showcases the archaic pre-digital Requests for Production of documents still used by many, perhaps most attorneys in America. The use of such Twentieth Century practices in the Twenty-First Century, along with the failure of counsel to communicate and cooperate as required by Rule 26(f), are the main reasons Judge Facciola had to work late the night before Christmas to resolve this discovery dispute.

Here plaintiff’s counsel used a form Request for Production (“RFP”) that was obviously designed for paper production, and then complained when the responding party produced in paper format the information responsive to the RFP, which, of course, was email. The plaintiff moved to compel the production of the email again, but this time in its native format. The defendant refused to produce for a second time the 35,000 pages of emails. Defendant did offer to produce the emails electronically, but in TIFF format, and only if the plaintiff would pay the fees the defendant would have to incur to delete and electronically redact privileged and irrelevant emails from the TIFF production. Plaintiff refused, contending that native production was required under the rules. Thus we have the dispute presented to Judge Facciola for resolution on Christmas eve.

The Terms of the Antique RFP

The first thing to do to try to resolve a discovery dispute is look at the discovery request itself. The RFP in question here was served on April 27, 2007. It instructed the defendant to:

[p]roduce all documents in [its] possession, custody or control, as they are kept in the ordinary course of business, including with all staples and clips attached and with all associated file folders, dividers and labels.

Nice to see such a detailed instruction on how to arrange the staples and paper clips in a production. Yes, this was an effective form in its day, which was at least 40 years ago! This same instruction could have been used (and probably was) in the Nineteenth Century, referring as it does to the paper clip, invented in 1867, and the staple fastener, invented in 1877. The only thing missing was an instruction for Bates stamping, patented in 1893, which you will still often find in many RFPs.  

Still, it gets worse. The  RFP goes on the define “documents” as:

[A]ny tangible thing upon which any expression, communication, representation or data has been recorded by any means including, but not limited to, handwriting, printing, photostating, photographing, on a computer, instant messages, magnetic impulse, or mechanical or electronic recording and any non-identical copies (whether different from the original because of notes made on such copies, because of indications that said copies were sent to different individuals than were the originals, or because of any other reason), including but not limited to working papers, preliminary, intermediate or final drafts, correspondence, memoranda, charts, notes, records of any sort of meetings, invoices, financial statements, financial calculations, diaries, reports of telephone or other oral conversations, desk calendars, appointment books, audio or video tape recordings, microfilm, microfiche, computer tape, computer disk, computer printout, computer card, and all other writings and recordings of every kind that are in your actual or constructive possession, custody or control.

Those of us who have been practicing law a long time instantly recognize this as an old form RFP, and I mean really old. Most of it was obviously prepared in the 1970s before the advent of personal computers, but spiced up by someone recently who added two words, “instant messages,” as an example of a “tangible thing.” So much for modernization! The RFP does not, in anyway, mention electronically stored information (ESI), the desired form of the production of ESI, for example, native format; nor does it breathe a word about metadata. In fact, even though someone added “instant messages” to the paper laundry list, it nowhere mentions emails. Instead, it requests “magnetic impulses,” “computer tape,” and “computer cards.” That is the kind of thing you might use for giant, old, IBM computers in the 1950s. Adding these words to the “document” definition made you look quite modern in the middle of the last century, but are just short of absurd today.

If you think that this is a rare exception of big firm blundering, you would be wrong. I wager that every AmLaw 100 firm has some trial lawyers who still use paper RFPs that are not too different from this one. Trial lawyers are an independent bunch; getting them to follow law firm best practice suggestions is, as the saying goes, “like herding cats.” It seems like the last thing that they and their obedient associates will part with are their old forms. No, instead they will spice them up from time to time by throwing in newfangled terms like “instant messages” and delude themselves into thinking that is good enough.

Although Judge Facciola points out the archaic language used, neither he, nor I, mean to cast stones on any particular law firm. In fact, the law firms in this case are very good by anyone’s standards, especially the plaintiff’s firms who used this form. No, we only take to task the individual lawyers in every firm who stubbornly cling to a contentious paper world that has long since passed them by. Their refusal to change and conduct discovery in a proper manner has consequences, none of them good, as Covad Communications Co. v. Revonet, Inc. demonstrates.

Judge Facciola’s Reaction to Ancient Boilerplate

Getting back to Judge Facciola’s Christmas eve labors, here is how he explains the predicament the parties presented to him for resolution:

Thus, I am supposed to determine by examining ancient boilerplate – designed for discovery in a paper universe – such nice questions as whether an e-mail, existing in a computer’s memory is a “tangible thing” and how e-mails are “maintained in the ordinary course of business.” While I have considered a similar provision in depth once before, FN1 See D’Onofrio v. SFX Sports Group, Inc., 247 F.R.D. 43, 47 (D.D.C.2008), I see no need to repeat that metaphysical exercise here because it is a waste of judicial resources to continue to split hairs on an issue that should disappear when lawyers start abiding by their obligations under the amended Federal Rules and talk to each other about the form of production. I would much prefer to carry out my duties in accordance with Rule 1, which provides that the rules “should be construed and administered to secure the just, speedy, and inexpensive determination of every action and proceeding.” Fed.R.Civ.P. 1.

The New Rules

Rule 34 of the Federal Rules of Civil Procedure states that: (1) the requesting party may designate the form in which the ESI should be produced; and, (2) if the request does not so specify, then it should be produced in a form in which it is ordinarily maintained, or in a reasonably usable form. The form in which ESI is “ordinarily maintained” is its native format. So when an RFP does not specify a form of production, you must produce in native or in a ”reasonably usable form.” This typically means an electronic form that is indexed and so can be searched by computer.

The plaintiff thus argued that the new rules require electronic production of email, either in native format or searchable image format, such as the TIFF offered by defendant. The defendant argued that the RFP asked for paper, and that is what they got. I assume they also argued that Rule 34 states that “a party need not produce the same electronically stored information in more than one form.” Here is how Judge Facciola deals with this quagmire of the parties own making:

More importantly, I do not need to parse words because no one is pretending that Revonet prints all of its e-mails or converts them to TIFF files on a daily basis no matter how ephemeral, meaningless or trivial their content.FN2 Therefore, though Covad’s instruction is hopelessly imprecise and Revonet could colorably argue that it should be interpreted to include several different formats, no reasonable person can honestly believe that hard copy is one of them. For hard copy to be an acceptable format, one would have to believe that Revonet, in its day to day operations, keeps all of its electronic communications on paper. There is no evidence in the record that Revonet operates in this manner, and no suggestion that such a practice would be anything but incredible. Therefore, even though I can’t say I know what Covad has asked for, I can say what they have not asked for, and that is what they got.

Footnote two in the above quote is worth reading in full, as it is a astonishing estimate of email volume that is far higher than previous estimates I have seen:

FN2. Note the following remarkable estimate:

Statistics, extrapolations and counting by Radicati Group from August 2008 estimate the number of emails sent per day (in 2008) to be around 210 billion. 183 billion messages per day means more than 2 million emails are sent every second. About 70% to 72% of them might be spam and viruses. The genuine emails are sent by around 1.3 billion email users.

Heinz Tschabitscher, How Many Emails are Sent Every Day? (last visited December 22, 2008).

The Christmas Eve “Stop Sign Crash” Case

Judge Facciola then focuses on the bottom line, how to best resolve this dispute in a situation where there is plenty of blame to go around. As usual, he does so with style and colorful language, which is bound to have us all remembering this decision as the “stop sign crash” case.

A much more fruitful use of the broad powers I unquestionably have to supervise discovery is to focus on the quickest and cheapest solution to the problem presented. I will assume that the e-mails at issue exist in what I have called their native format and can be copied onto a CD with a couple of keystrokes. Obviously, printing them or converting them to TIFF files probably (and ironically) costs more so Revonet is hard pressed to claim that producing them now in their native format is unfairly burdensome.

There is one burden, however: the cost of deleting privileged information from the CD to be produced. At one point, Revonet agreed to produce the e-mails in TIFF for-mat but insisted that Covad pay for a paralegal to “manually delete the non-responsive and privileged TIFF’s that have already been redacted from the paper document production.” Exhibit D to Pls. Mot. at 6. Revonet estimated that this would take the paralegal 5 to 10 hours at a rate of $178.50 for a maximum cost of $1,785.00. Id.

Since both parties went through the same stop sign, it appears to me that they both should pay for the crash. I will require them to share the cost of the paralegal removing the privileged e-mails, as I have described it, to a cost of no greater than $4,000, i.e., $2,000 each. I expect Revonet to keep a careful record of the time spent and to alert me if there is any risk that the cost will exceed $4,000. At that point (which I hope will not be reached) I will ask Revonet to estimate what it will cost to finish the job and seek the views of counsel as how to cover it.

Conclusion

Judge Facciola then concludes by making the moral of this Christmas eve story very clear:

Finally, I would hope that my decision will have a didactic purpose. This whole controversy could have been eliminated had Covad asked for the data in native format in the first place or had Revonet asked Covad in what format it wanted the data before it presumed that it was not native. Two thousand dollars is not a bad price for the lesson that the courts have reached the limits of their patience with having to resolve electronic discovery controversies that are expensive, time consuming and so easily avoided by the lawyers’ conferring with each other on such a fundamental question as the format of their productions of electronically stored information.

Thanks for the Christmas present Judge Facciola; we all appreciate it!

D.C. Appeals Court Affirms Order Requiring a Non-Party to Spend $6 Million, 9% of its Total Annual Budget, to Comply with an e-Discovery Subpoena

Appellate courts do not often weigh in on e-discovery issues, but when they do, it is a big deal. The United States Court of Appeals for the District of Columbia did so on January 6, 2009, when it issued an opinion on e-discovery and sanctions. In re Fannie Mae Securities Litigation, _ F.3d _, 2009 WL 215282009, U.S. App. LEXIS 9 (D.C. App. Jan. 6, 2009). Typically, I would be glad to have a Circuit Court opinion on e-discovery. Not so here.

Unfortunately, In re: Fannie Mae Securities Litigation sets a troubling precedent in favor of enforcing exorbitant e-discovery costs. In this case, the Office of Federal Housing Enterprise Oversight (“OFHEO”), was required to spend six million dollars, representing nine percent of its total annual budget, just to comply with a subpoena for electronic documents. Although OFHEO clearly had relevant information to the multidistrict litigation against the Federal National Mortgage Assn. (“Fannie Mae”) and the Federal Home Loan Mortgage Corp. (“Freddie Mac”), they were not a party to the litigation. This fact, coupled with the high expense involved in an over-broad e-discovery request, did not seem to concern the court, which is why this decision is troubling.

Why Fannie Mae Is Dangerous Precedent

The Appellate Court in the opinion written by Circuit Judge David S. Tatel rejected with little discussion the argument that the respondent should not have been compelled to comply with the subpoenas without considering the costs of compliance, cost shifting, narrowing the scope of production, or showing good cause to retrieve inaccessible data. Id. at *7. The Circuit Court did not even address the merits of these arguments, but, instead, basically said it was too late, you “agreed” to it, so now you are stuck with it regardless of the costs and consequences. For this reason, Fannie Mae should be strictly limited to situations where the parties have knowingly consented to an order they later fail to follow.  Here is the language of the Court:

Federal Rule of Civil Procedure 45 requires courts to safeguard non-party subpoena recipients from significant expense resulting from compliance. See Watts v. SEC, 482 F.3d 501, 509 (D.C. Cir. 2007). According to OFHEO, the district court violated Rule 45 by compelling compliance without considering cost-shifting, narrowing the scope of the requests, or “find[ing] that defendants demonstrated good cause for forcing OFHEO to retrieve its inaccessible data.” Appellant’s Opening Br. 31-32

Whatever the merits of these claims, OFHEO abandoned them by entering into the stipulated order. Indeed, OFHEO’s trial counsel agreed to the stipulation in the middle of a hearing scheduled for the very purpose of considering OFHEO’s objections to the subpoenas. Had OFHEO wanted review of the district court’s initial order to compel compliance with the subpoenas, it could have completed the hearing and attempted to convince the court to re-consider. Failing that, it could have defied the adverse ruling and appealed any ensuing contempt finding. See U.S. Catholic Conference, 487 U.S. at 76. Instead, it chose to sign the stipulated order, which ended the hearing and unquestionably settled the discovery dispute. Having stipulated to a schedule for complying with the subpoenas, OFHEO can hardly complain now about being held to its agreement.

The opinion should be limited to the “unusual” facts of the case. But, since the decision itself did not take pains to do so, the door is open for Fannie Mae to be exploited and cited to try to require parties and non-parties alike to incur excessive e-discovery expenses. I do not think that is what the appeals court had in mind, they were obviously just trying to uphold the lower court’s discretion, but that is the net result. As a consequence, other courts will have to deal with this precedent for years. I expect they will do so by carefully distinguishing and limiting the holding to the facts of the case.

Electronic discovery costs are a runaway train that must be stopped. This new opinion sets dangerous precedent making that task even more difficult than it was before. Certainly courts will distinguish and limit the case, but it is still a problem because many parties requesting e-discovery may try to inflate its importance to support outrageous demands.

Fannie Mae Shows Zero Tolerance For Common “Rookie” e-Discovery Errors

Even when limited to its facts, Fannie Mae is still dangerous because, like it or not, the facts presented are not all that unusual. When it comes to e-discovery, most lawyers are still struggling beyond their depth. It is not that uncommon for trial counsel to enter into agreements for search and production of electronic documents without first understanding its possible costs and burden. They also rarely understand how long it takes to review and produce gigabytes of data, nor do they know much if anything about modern search technologies. It is also not uncommon for generalist litigators to later make things worse with an “old-school” type fight-everything attitude and a series of  too little, too late, productions.

This is just what happened here. In the midst of a hearing, trial counsel agreed to restore backup tapes, search them using plaintiffs’ terms, and then produce all email and attachments that were not privileged. Obviously the attorney did not intend by this agreement to assume a $6 Million Dollar burden, nor did the client authorize their attorney to do so. How can you have a six million dollar agreement without a “meeting of the minds?” Yet, the district court keeps coming back to that agreement, made in the midst of a hearing, as the justification for requiring a non-party to spend 9% of their total annual budget to comply with a subpoena. The Circuit Court then upheld the decision as within the very broad discretion allowed to a district court to manage discovery. They did so without any comment or reaction as to the injustice of imposing this kind of e-discovery cost and burden.

This case sends a very strong message to the Trial Bar to be very careful about what they agree to in a midst of a hearing. It shows that they should never agree to anything concerning e-discovery without first consulting with an e-discovery specialist, preferably one who is part of a larger tech/law e-discovery team. In fact, this case clearly demonstrates that an e-discovery lawyer (if you can find one) should attend every discovery hearing.

Trial lawyers, no matter how smart and skilled they may be, are not qualified to swim in these waters alone. Fannie Mae shows just how dangerous these waters can be. You may think something can be done fairly easily and inexpensively, and later be shocked to learn that it will cost millions, and take months, or even years to do. Apparently the trial judge did not understand how common a mistake this is. I am confident that if the government lawyers for OFHEO had had an experienced e-discovery lawyer with them at the first hearing, they would not have stipulated to the order they did, and all of the disasters that followed could have been avoided. But they did not, and as a result, they were bushwhacked.

The requesting parties here were allowed to specify 400 keywords, which, not too surprisingly (to us anyway), returned over 660,000 emails and attachments. By the way, does anyone seriously think that the plaintiffs suing Fannie Mae and Freddie Mac will now actually review all of these emails? No way! They will cull it down and only review a small percentage, which is exactly what the government should have done. I am willing to bet that most of the 660,000 emails produced will have little or no probative value. Like most cases, it will probably come down to a few hundred key documents, and maybe a few thousand more of some interest (to the attorneys). At the end of the trial, the jury will only remember four or five. This kind of forced production of 660,000 emails is irresponsible over-kill on a vast scale. It is infuriating because we, the American taxpayers, are forced to underwrite this fiasco.

It gets worse. The obviously over-broad search terms retrieved approximately 80% of all of the agencies email. Yet, this did not seem to trouble the district court, nor the appeals court, who actually stated that the 80% return “figure may simply indicate that most of the emails actually bear some relevance, or at least include language captured by reasonable search terms.” Id. at *6.  Do they really think that 80% of the agency’s total email is relevant to the case? Obviously they have never read a large email collection before. The statement is incredible. As to the later explanation – “or at least include language captured by reasonable search terms” – the logic is meaningless and circular. Of course all of the emails retrieved contained one or more of the 400 search terms. That is why they were retrieved! How does that make them “reasonable” search terms?

In re: Fannie Mae Securities Litigation arguably stands for the proposition that if you agree to an e-discovery production, you may well be bound to your agreement, even if the agreement later proves to make no sense and the client has to go broke in the process. This is dangerous precedent. No one should be required to spend outrageous sums of money to comply with a subpoena, no matter what their lawyer may have said or done. Six Million Dollars, representing 9% of a governmental agency’s budget, is too high a burden to place on a party to litigation, much less a non-party. The principles of proportionality and Rule 26(b)(2)(C) should trump and prevail over any agreements inadvertently made by legal counsel.  No wonder the government is going broke, just like many of the homeowners that Fannie Mae and OFHEO were supposed to protect.  

Six Million Dollars

The district court later interpreted the stipulation that OFHEO’s attorney had made  in the middle of a hearing to allow the requesting parties to specify any search terms they wanted; so they did. They specified 400 keyword terms which returned over 660,000 documents. Even though the government lawyers clearly did not understand what they had agreed to, they went ahead with the search and spent $6,000,000 in the process. Another big mistake. They should have tried to get relief first before spending the money. If that was refused, they still should have refused to spend $6,000,000. Instead, as the appellate court later recommended, they should have “defied the adverse ruling and appealed any ensuing contempt finding.” The bottom line is, once you spend the money on e-discovery, it is very hard to get it back. Unless you have specifically reserved the right for cost shifting, and sometimes even when you do, there is no money back guarantee.

You may well wonder how they managed to spend $6,000,000 to comply with a subpoena. As is typical, the main costs were for privilege review. Here is the language of the appellate court: 

OFHEO undertook extensive efforts to comply with the stipulated order, hiring 50 contract attorneys solely for that purpose. The total amount OFHEO spent on the individual defendants’ discovery requests eventually reached over $6 million, more than 9 percent of the agency’s entire annual budget.

One would think at that point the courts would give the agency a break. But no, the government lawyers had made a series of promises to the court that they did not keep, mainly as to when the review would be completed and the emails produced. There are consequences for such failures. Unless you have done big projects like this before, it is unbelievable how long review can take and how many things can, and do, go wrong. Delays are inevitable and should be anticipated. They should enter into the calculus as to when a project can be completed and production made. 

Clearly the government lawyers were once again making promises and representations to the court without the assistance of an experienced e-discovery lawyer. They made promises they could not keep and did not give adequate explanations for the delays. They also failed to cooperate and make rolling productions. Instead, they produced nothing at all while the contract lawyer review expenses kept going through the roof. To the district court judge, who almost certainly was never involved in a large e-discovery project before appointment to the bench, it looks like a series of broken promises and downright contempt. I do not know what happened here, but it was probably a series of mistakes and miscalculations, not intentional defiance of the court.

Regardless of the true cause, some of the the courts in Washington have obviously grown tired of lawyers making promises and commitments that they do not keep. They wanted to send a message that their actions would not be tolerated, and that is exactly what they did. Here is the language of the district court explaining the situation at *3: 

*3 On November 29, 2007, the day before an interim deadline for production of several categories of material, OFHEO informed the district court that it would be unable to meet that deadline and moved for an extension until December 21, assuring the court that it could meet that extended deadline. The court granted the motion, but two days before the extended deadline, OFHEO informed the court not only that its previous assurances had been based on insufficient data, but also that it had only recently hired the necessary number of contract attorneys. OFHEO told the court that it would be unable to comply with the extended interim deadline, and that although it could produce all non-privileged documents by the ultimate January 4, 2008 deadline, it would be unable to produce all the required privilege logs until February 29.

The individual defendants renewed their motions to hold OFHEO in contempt. On January 22, the district court granted the motions. The court recognized OFHEO’s efforts at compliance, but deemed them “not only legally insufficient, but too little too late,” stating:

[T]he Court is cognizant of the large number of attorneys, contract attorneys, and OFHEO personnel working to comply with the subpoenas and the resulting costs of this compliance. Nevertheless, OFHEO has treated its Court-ordered deadlines as movable goal posts and has repeatedly miscalculated the efforts required for compliance and sought thereafter to move them.

OFHEO was held in contempt of court, as punishment they would not only have to pay for everything, with no hope of cost-shifting, but they would be required to produce some 20,000 emails without advance privilege review. The production was to be made on an “attorneys-eyes-only” basis, and a chance to later assert the unique kind of government privilege at issue here, the “deliberative process” privilege, but still, opposing counsel will be allowed to see all.

 Here is the appellate court’s explanation:

[T]he district court directed the office to provide the actual documents withheld on the basis of the deliberative process privilege and not logged by the deadline. The district court described the sanction as “designed to move the [d]iscovery process forward and to allow for [a] more targeted, and therefore more truncated, privilege litigation process.”Hr’g Tr. at 26 (Jan. 22, 2008). The district court therefore specified that the compulsory disclosure would not waive the privilege with respect to further disclosure; directed that the documents be provided only to individual defendants’ counsel; and created a mechanism for OFHEO to recover documents found to be privileged.

I am not troubled by the privilege sanction. It stops short of a waiver and it appears that OFHEO deserved this counter-measure to make up for its many prior delays.

The part of this decision that astounds and disturbs me is the affirmance of the propriety of requiring a non-party to spend 9% of its total budget to comply with a subpoena, simply because their attorneys blindly agreed to comply without considering the cost of compliance. Here is how the Appeals Court explains its holding at *8:

*8 District judges must have authority to manage their dockets, especially during massive litigation such as this, and we owe deference to their decisions whether and how to enforce the deadlines they impose. See Berry v. District of Columbia, 833 F .2d 1031, 1037 n. 24 (D.C.Cir.1987). Though we recognize OFHEO’s strenuous efforts to comply, the district court found them to be “too little too late,” Hr’g Tr. at 19 (Jan. 22, 2008), and determined that the office’s compliance was inadequate, id. at 21. In making this assessment, the court placed great weight on the long history of the discovery dispute and on OFHEO’s repeated requests for extensions, ultimately concluding that OFHEO had requested one extension too many and that strict enforcement of its deadline was warranted. Given the district court’s intimate familiarity with the details of the discovery dispute, the scale of the production requested, and the progress of the multidistrict litigation as a whole, we are ill-positioned to second-guess that assessment. Were we on this record to overturn the district court’s fact-bound conclusion that OFHEO dragged its feet until the eleventh hour, we would risk undermining the authority of district courts to enforce the deadlines they impose.

Conclusion

I understand the importance of district court discretion, enforcing deadlines, and not second-guessing discovery calls. I am in favor of that. Also, I have often called for judges to get tough and stop allowing all of the “hide the ball” games. On the whole, I think judges have been far too tolerant of attorney misconduct. But, this case goes too far and punishes the wrong people. A government agency has been unfairly punished for conduct of their legal counsel. In the process, the second of the fourteen Sedona Principles has clearly been breached:

2. When balancing the cost, burden, and need for electronically stored information, courts and parties should apply the proportionality standard embodied in Fed. R. Civ. P. 26(b)(2)(C) and its state equivalents, which require consideration of the technological feasibility and realistic costs of preserving, retrieving, reviewing, and producing electronically stored information, as well as the nature of the litigation and the amount in controversy.

A non-party should never be required to spend $6 Million to comply with a subpoena. Moreover, by requiring OFHEO to restore and search its disaster recovery backup tapes, the requesting plaintiffs and the courts also violated the eighth Sedona Principle:

8. The primary source of electronically stored information for production should be active data and information. Resort to disaster recovery backup tapes and other sources of electronically stored information that are not reasonably accessible requires the requesting party to demonstrate need and relevance that outweigh the costs and burdens of retrieving and processing the electronically stored information from such sources, including the disruption of business and information management activities.

This case demonstrates what can happen when these fundamental precepts are not followed. It cannot be fair and just to allow a lawyer’s actions to drain 9% of an agency’s total annual budget, a budget paid for by taxpayer funds. Congress did not approve OFHEO’s budget last year thinking that 9% of it would go to help a group of plaintiffs sue Fannie Mae. The appellate court, and the district court before it, should have made an economic analysis of the situation in reaching its decision and considered the gross inefficiencies of its ruling.

I understand the need to affirm trial court discretion, but the appeals court could have upheld the authority of the lower court and still mitigated the economic impact of this ruling in some way. For instance, it could have allowed for possible cost-shifting. At the very least, the appeals court should have carefully limited its holding and expressed some abhorrence of the burden imposed. It should have considered the possible impact of this ruling on the future jurisprudence of e-discovery. It did not. It went too far, just as the pendulum was beginning to swing the other way. The economy and fair minded judges everywhere will no longer tolerate this kind of excess. For this reason, I predict that Fannie Mae will come to be seen as the high-point of out-of-control e-discovery spending. It will be all downhill from here, especially if we can muster up the competence and cooperation needed to make it happen.