Discovery of Electronically Stored Information

Clayton L. Barker1	Philip W. Goodin1

This article describes the amendments to the Federal Rules of Civil Procedure related to discovery of electronically stored information, discusses how Missouri courts may address these issues, and outlines the associated duties of counsel.

I. Introduction

Discovery issues centering on electronically stored information (ESI) have been around for years, but as technology races ahead and fundamentally changes the way people communicate and store information, the legal system finds itself struggling to keep up. ESI discovery is not an issue relegated to large, sophisticated parties in immense federal lawsuits. It is a fact of life for all courts, at every level, and in all types of legal actions.

The legal community’s attention focused on ESI discovery when amendments to the Federal Rules of Civil Procedure concerning ESI went into effect on December 1, 2006. In the months that followed, courts and commentators analyzed, applied, and construed the changes. This article summarizes the federal changes and examines how Missouri courts, under the current Supreme Court Rules,² would likely address the most common ESI discovery issues. Although ESI disputes are not yet a standard feature in most Missouri courts, they inevitably will be.

A. What is Electronically Stored Information?

ESI includes any information stored in an electronic – as opposed to paper – form and includes a virtually limitless list including e-mails, presentation files (e.g., Power Point), spreadsheets (e.g., Excel), word processing files (e.g., MS Word, Word Perfect), databases, audio files, picture files, and web pages. ESI can be stored in a vast array of media, including hard drives, servers, diskettes, CD-ROMs, cell phones, thumb drives, and handheld wireless devices.

B. Why Amend the Federal Discovery Rules?

Numerous cases applied the pre-amendment federal rules to ESI discovery disputes.³ These courts determined that ESI, although not a perfect fit, came within the term “document,” as used in pre-amendment Federal Rule 34(a), which defined “document” to include “data compilations.”⁴ Rule 58.01(a)(1) more clearly defines “documents” to include “electronic records, and other data compilations from which information can be obtained. . . .”

Specific amendments to the Federal Rules of Civil Procedure, however, were necessary because existing case law did not adequately address essential differences between ESI and paper documents. The federal amendments sought to apply basic discovery concepts to ESI’s unique characteristics and thus seek to be more evolutionary than revolutionary.

There are several fundamental differences between paper and ESI discovery. First, the traditional paper-focused discovery process involved producing boxes of paper documents for inspection or copying, whereas discovery of ESI involves producing electronic files. Second, ESI discovery can be far more expansive than traditional discovery. Informal communications that once occurred as ephemeral conversations are now recorded for posterity as e-mails. Electronic files are notoriously difficult to erase or delete. ESI can be stored in numerous locations ranging from centralized servers to personal cell phones. Additionally, every electronic file contains not only the file’s substance but background information about the file, usually called “meta” or “embedded” data.⁵ Last, almost all computer systems contain discoverable “system data.”⁶

Third, ESI discovery can be far more burdensome and costly than traditional discovery. Many courts,⁷ including the United States Supreme Court,⁸ have raised concern on the corrosive effects that the excessive costs of discovery are having on the administration of justice. Several factors account for the increased cost of ESI discovery. ESI is stored in vastly greater volumes (a phenomena sometimes called “information inflation”). The time, cost, and effort to review ESI for relevance, responsiveness, and privilege increase in direct proportion to the volume. Lawyers must often use specialists to assist in ESI discovery to obtain, organize, and sort through the potentially responsive information. Erased data, data stored on obsolete systems, and data stored only for disaster recovery can be transformed into useable forms, but often only at great cost.

Fourth, ESI discovery presents unique technical issues not associated with paper discovery. ESI can be produced in different forms. It can be printed on paper, scanned to a disk, or the electronic file itself can be produced. Some ESI, such as databases, cannot be reduced to paper form because they were designed only to create reports based on criteria from the operator. Concepts applicable to paper documents, such as “original” and “copy” do not apply to ESI. Electronic files, unlike paper documents, can easily be organized and searched based on their content, allowing parties to run automatic searches for key words, phrases, or names as well as creation dates.

ESI also presents unique issues concerning preservation of evidence. Computer storage systems work by overwriting and deleting information without the operator’s specific direction or knowledge. Moreover, some ESI was intended only as short-term information and is difficult to preserve.

C. What Were The Goals of the Federal Rule Amendments?

Given ESI’s unique properties, the federal rule amendments sought to address these differences in a manner consistent with the spirit set forth by the mandate of Federal Rule of Civil Procedure 1 for “the just, speedy, and inexpensive determination of every action.” Missouri has a similar guiding provision, Rule 41.03, which directs that the rules “shall be construed to secure the just, speedy and inexpensive determination of every action.” The federal amendments address seven goals:

• Designate a third category of discoverable materials – “documents,” “tangible things,” and now “ESI” – and amend existing rules with ESI-specific provisions.⁹

• Compel the parties and the court to address ESI discovery issues early in the litigation process to resolve costly potential disputes.

• Lower the cost and expansiveness of ESI discovery by creating a presumption that ESI not reasonably accessible due to excess cost or burden is not discoverable.

• Specifying acceptable forms for ESI production.

• Lower the cost and burden of reviewing ESI by clarifying that inadvertent disclosure of privileged ESI to another party usually does not constitute a waiver of privilege.

• Address ESI’s unique storage features by creating a safe harbor from sanctions for good faith, routine destruction of ESI.

• Modify the rules governing subpoenas to track the changes to other discovery rules.

II. The Attorney’s New Roles

The principles that have guided discovery for decades remain firmly in place; ESI has merely created new issues to which the old principles apply. Discovery still seeks information on designated subjects; ESI merely expands the number of potential sources. The attorney’s role in the process has not changed, but in some cases the combination of ESI’s unique nature and the new discovery rules result in new roles for attorneys. If ESI discovery will play an important role in particular litigation, the attorney must become knowledgeable of the various aspects of ESI discovery so that the attorney can competently perform the discovery process.

A. The Need to Understand ESI Discovery

As an initial matter, a lawyer’s basic duties remain unchanged. A lawyer must possess the “legal knowledge, skill, thoroughness and preparation reasonably necessary” to undertake ESI discovery tasks.¹⁰ A lawyer must communicate with and properly advise the client on preservation and discovery obligations concerning ESI.¹¹ Last, a lawyer must ensure that representations concerning ESI made to the court are correct.¹²

When ESI discovery issues are important to a case, both the in-house and outside counsel must be knowledgeable of the client’s ESI systems, the client’s ESI retention, storage, and destruction procedures, and the applicable discovery rules and case law.¹³ Federal Rule of Civil Procedure 26(g) requires that an attorney must certify that every discovery response or objection was formed after a reasonable inquiry, which creates an affirmative duty to fully investigate all discovery responses. In conducting this investigation, an attorney can rely on the client’s assertions but only to the extent it is reasonable to do so.

To satisfy this standard, the attorney may need to conduct an extensive conversation with knowledgeable employees. Having this type of conversation may initially be a daunting task to attorneys who are not familiar with the technical aspects of computer systems, but it is necessary to properly represent a client, to reduce costs, to comply with discovery rules, and to avoid sanctions for inadvertent spoliation of potentially responsive materials.

B. Negotiating ESI Protocols with Opposing Counsel

One of the most important functions the attorney plays in grappling with ESI issues is to negotiate a protocol for the preservation, collection, and production of ESI. Absent an explicit understanding with opposing counsel, attorneys may wind up having their efforts second-guessed by opposing counsel or the court—which could very likely lead to duplicative efforts, unnecessary costs, insufficient searches and production, and accusations of evidence spoliation.

The importance of this role is reflected in the new federal rules which now require opposing counsel to meet early and thoroughly discuss ESI issues to identify and resolve disputes before conducting costly, time-consuming searches and document productions. This requires counsel to be familiar with their respective client’s systems before they meet and to know what questions to ask of each other. Three federal rule changes addressed this goal.

New Federal Rules of Civil Procedure 26(f)(3) and (4) require the parties to discuss ESI issues. The list of topics is limitless but includes what ESI is relevant, how it is stored, preservation plans, cost sharing, privilege waiver, and forms of production. Excellent topic checklists are in the District of Maryland’s Suggested Protocol for Discovery of Electronically Stored Information¹⁴ and the District of Kansas’ Guidelines for the Discovery of Electronically Stored Information.¹⁵ It is a good practice to memorialize all agreements between the parties in a court order.

New Federal Rules of Civil Procedure 16(b)(5) and (6) require the court’s scheduling order to address designated ESI and privilege issues. Amended Federal Rule of Civil Procedure 26(a)(1)(B) requires each party to disclose ESI it may use to support its claims or defenses.

Missouri has no corresponding procedural rules because it does not require the structured discovery process used in the federal courts. Nevertheless, in cases in which ESI discovery is anticipated to play a material role, early discussion and agreement is the best course of action. A circuit court can always issue discovery or scheduling orders under its inherent power and Rule 56.01(f) permits parties to stipulate their own discovery procedures.

C. Supervising ESI Preservation, Collection, and Review

To avoid unintentional loss, counsel must advise and supervise the client’s efforts to preserve and collect all potentially relevant ESI. At this point in the process, if the ESI discovery effort is large or complex and if the client lacks sophisticated in-house technical resources, counsel should consider retaining outside expert assistance from vendors who specialize in ESI litigation support. These include specialists in preserving, reviewing, indexing, and producing ESI, as well as forensic experts who specialize in the recovery of deleted or other hard-to-get ESI.

To properly preserve and collect data, the attorney will need a “data map” or list of all possible sources of potentially responsive ESI. The list of where to look can be extensive and can include centralized servers, office computer hard drives, loose disks or CDs, portable wireless devices, home computers, voice mail files, external storage devices (thumb drives), back-up tapes,¹⁶ legacy data,¹⁷ and deleted data.¹⁸ The attorney will also need to advise on the best way to preserve this data.

Once the ESI is preserved, the attorney must supervise the review of the potentially responsive ESI to identify responsive and privileged ESI. If the universe of potentially responsive ESI is large enough, the initial review is conducted automatically by searching for ESI with designated terms and date parameters. Other additional automatic steps, such as de-duplication – removing duplicative copies – can also be implemented. Once the potentially relevant ESI is culled down by automatic procedures, the review’s final step is done the old-fashioned way – by humans reading each e-mail or document to determine its relevance, importance, and privilege status. Given the volume of ESI, this can be a costly and time-consuming endeavor.

Because the opposing party can discover the process used to preserve, review, and produce the ESI, it is common for the parties to agree beforehand on search terms and other process details to avoid protracted and costly post-production discovery battles.

III. Production Issues Unique To ESI

Under the old discovery paradigm, there was a duty to produce responsive, relevant, and non- privileged information, unless the court sustained an objection or the producing party received a protective order. Under new Federal Rule of Civil Procedure 26(b)(2)(B), however, there is no duty to produce otherwise responsive, relevant, and non-privileged ESI from sources that the party identifies as “not reasonably accessible because of undue burden or cost.”

A. Responding to a Request to Produce ESI

In response to a request for production, only reasonably accessible ESI that is otherwise responsive, relevant, and not privileged is produced. “Not reasonably accessible” ESI is presumed non-discoverable and the producing party must merely identify and describe the unreasonably accessible ESI it is not producing. The producing party should provide adequate detail to allow others to evaluate the cost, burden, and likelihood of finding responsive information.

This creates a two-tiered approach to discovery. In most cases, the first tier, comprised of easily accessible ESI, will provide all reasonably useful information. In rare circumstances, the parties may need and obtain discovery of the second tier that is not reasonably accessible.

B. Distinguishing Reasonably from Unreasonably Accessible ESI

The issue thus becomes the reasonableness of the ESI’s accessibility. Keep in mind that there is no such thing as “inaccessible” ESI. If this information exists, it is accessible. Constant technological changes, however, make a specific definition of reasonable accessibility unworkable.

The most common approach¹⁹ to determining the reasonableness of accessibility looks to which ESI is used in the ordinary course of business and thus presumably does not require costly or burdensome effort to be useable. Reasonably accessible ESI is, therefore, that which is used in the ordinary course of business or available with little time or expense and includes active data (stored on a hard drive, server, or on disks but subject to robotic retrieval) and offline storage or archives.²⁰ On the other hand, unreasonably accessible ESI includes legacy data (created by systems no longer used), deleted or fragmentary data, and disaster recovery back-up tapes that are not searchable.

Note that designating ESI as not reasonably accessible does not relieve a party of its duty to preserve it. Furthermore, absent a material change in search technology, what a party claims is not reasonably accessible in one case must be inaccessible in all cases.

C. Challenging an Opposing Party’s “Not Reasonably Accessible” Designation

The requesting party may dispute a designation of unreasonably accessible ESI by filing a motion to compel. The producing party could also move for a protective order to confirm its unreasonably accessible designation. As always, the parties should discuss the issue and seek a compromise before going to the court. The producing party bears the burden of demonstrating unreasonable accessibility with specific facts.

To challenge the unreasonably accessible characterization, the requesting party usually needs discovery to support its argument (a practice often referred to as “discovery about discovery”). Such discovery may include deposing those with system knowledge,²¹ depositions of experts, or a sampling of data to verify accessibility claims concerning cost, burden, and data type. Challenging accessibility is a fact-oriented dispute and needs specific estimates of time, cost, required efforts, technology system characteristics, and anticipated results. Expert affidavits are usually necessary.

D. Showing Good Cause to Produce Unreasonably Accessible ESI

Under new Federal Rule of Civil Procedure 26(b)(2)(B), if the court characterizes the ESI as not reasonably accessible, the requesting party could still obtain the ESI if it shows good cause for the discovery in light of Federal Rule of Civil Procedure 26(b)(2)(C)’s factors: (i) whether the ESI is “unreasonably cumulative or duplicative, or” is obtainable from a less burdensome source; (ii) whether the requestor has had the opportunity to seek the information by other means; and (iii) whether the “burden or expense … outweighs its likely benefit, [taking into account] the needs of the case, the amount in controversy, the parties’ resources, the importance of the issues,” and the ESI’s importance in resolving the issues.

The official notes to Federal Rule of Civil Procedure 26(b)(2) set out additional factors to consider in making the good cause determination: (1) the specificity of the request; (2) the quantity of information from more available sources; (3) the failure to produce relevant information that may have once existed but is no longer available on more easily accessible sources; (4) the likelihood of finding relevant information that cannot be obtained on more easily accessible sources; (5) the predictions on usefulness and importance of information; (6) the importance of issues at stake in litigation; and (7) the parties’ resources.

Good cause is inherently difficult to show because neither party will know specifically what is in the unreasonably accessible ESI at issue. If good cause for production is shown, the court can limit the discovery or specify conditions for production.

E. Shifting Production Costs to the Requestor

Cost shifting is one of the most litigated aspects of ESI discovery. Cost-shifting is not unique to ESI, although, in the ESI context, it is frequent and applies unique factors. In federal court, the presumption has always been that the producing party pays the costs of production, although in rare circumstances, cost shifts have been permissible under Federal Rule of Civil Procedure 26(c).²² The no-discovery presumption for unreasonably accessible ESI makes cost-shifting a more frequently granted condition for discovery.

There are numerous cost-shifting approaches. One approach is to flexibly apply Federal Rule of Civil Procedure 26(b)(2)(C)’s factors.²³ Another approach is a “marginal utility” analysis, which weighs the likelihood that a request would unearth important information against the cost of obtaining it.²⁴ The lower the likelihood of discovering important information, the more likely the costs will be shifted to the requestor.

Additionally, there are several multi-factor approaches. The Zubulake decisions applied seven weighted factors:

1. The extent to which the request is specifically tailored to discover relevant information;

2. The availability of such information from other sources;

3. The total cost of production, compared to the amount in controversy;

4. The total cost of production, compared to the resources available to each party;

5. The relative ability of each party to control costs and its incentive to do so;

6. The importance of the issues at stake in the litigation; and

7. The relative benefits to the parties of obtaining the information.²⁵

The American Bar Association, ²⁶ The Sedona Conference,²⁷ and the Conference of Chief Justices²⁸ all have recommended multi-factor approaches. A simpler approach has been adopted by Texas, which simply mandates that the requesting party always pays the extraordinary production costs of not reasonably accessible ESI.²⁹ Under this approach, if the requesting party truly believes that important information is to be found in the unreasonably accessible ESI, it will be willing to pay for it.

F. Addressing A Request To “Mirror Image” A Computer Hard Drive

One of the most frequently litigated ESI production issues is whether the requestor can “mirror image” – make an exact copy of – the producer’s entire computer hard drive. Mirror imaging captures all information on the hard drive, not just relevant information.

Luckily, the courts have applied a consistent approach to this issue. As an initial matter, a request for production does not automatically permit the requestor to search around a hard drive looking for interesting information, just as it does not permit a requestor to poke around in the producer’s paper file drawers. A request for production is not a search warrant.³⁰ Moreover, absent a basis for reasonable doubt, a court is bound to accept the producer’s representation that all relevant information has been produced.³¹

Courts, however, have approved mirror imaging in two situations:³² first, when the computer itself is at the heart of the litigation, that is, an instrumentality of the alleged wrongdoing, which often occurs in computer fraud and theft of trade secret cases; second, when there is a reasonable basis to believe that relevant information likely to be on the hard drive has not been produced. This often occurs when there is suspicion that ESI files have been deliberately deleted. Although an operator cannot retrieve deleted information, the information is often still present on the hard drive and can be retrieved by a computer forensic expert.

If a request for a mirror image is granted, the court usually approves a protocol to protect the producer’s privacy and confidential information.³³ Generally, a forensic computer expert will be tasked with copying the hard drive, searching the hard drive for certain key terms, and then listing responsive contents. The producer’s attorney then reviews the list and produces responsive materials and lists privileged materials on a privilege log.

G. How a Missouri Court May Address Requests for Unreasonably Accessible ESI

Missouri’s discovery rules are similar to the pre-amendment federal rules and have no specific provisions applicable to unreasonably accessible ESI. A Missouri court would likely start from the presumption that all relevant, non-privileged information is discoverable, but could impose limitations or conditions, including cost-shifting, under Rule 56.01(c), if discovery of unreasonably accessible ESI is sought. Rule 56.01(c) provides that where discovery rises to a level of undue burden or expense a court can enter a protective order for good cause shown. Missouri recognizes the need for burden-benefit balancing in assessing discovery requests.³⁴

IV. The Different Options To Produce ESI

Unlike paper documents, which are created and stored in a form intelligible to people, ESI is created and stored as electronic code and must be converted before people can use the information. This creates another legal issue unique to ESI production: determining the form in which the ESI will be produced.

Under new Federal Rule of Civil Procedure 34(b)(i), ESI may be produced in a form in which it is ordinarily maintained, in a reasonably useable form, or in any other form agreed to by the parties. Under Federal Rule of Civil Procedure 34(b)(ii), the requesting party can request a specific form in its production request. The responding party can object and counter-designate a production form or, if no specific form was requested, can designate a form for its production. A party need not produce the same ESI in more than one form. If the parties cannot agree on the appropriate production form, the court may be called on to determine the form in which the ESI will be produced.³⁵

A. The Different Forms for ESI Production

As a general proposition, ESI can be produced in three forms: paper, imaged (quasi-paper), or native forms. Each has unique attributes.

1. Producing ESI in Paper Form

ESI is produced in paper form by printing the file, such as e-mails or documents. In many cases printing the relevant e-mails and other ESI is a perfectly satisfactory production method. Paper is a form familiar to all lawyers and it is easy to track by affixing tracking numbers. Paper form, however, is difficult to search for specific terms, difficult to use with litigation management software, burdensome to search and manage when produced in large quantities, and, because it is not in the form the ESI was maintained in the usual course of business, does not contain meta or embedded data.

2. Producing ESI as a Scanned Image

ESI is imaged when a picture of the file is scanned onto a CD. Common image forms are *.tif or *.pdf. Imaging has the advantage of mimicking paper production, since tracking numbers can be affixed. Additionally, the image cannot be altered, imaging is compatible with discovery management software, and, for voluminous productions, imaging takes up little space compared to paper. Depending on the imaging method, the images can be automatically searched for specific terms. A drawback to images is that they do not contain meta or embedded data and are usually not the form in which the files were maintained “in the usual course of business.”³⁶ A frequent problem with large productions of imaged ESI is linking e-mails to files attached to those e-mails.

3. Producing ESI in its Native Form

ESI is produced in native form when the entire electronic file itself is produced. For instance, document files are produced as a *.doc file on a computer disk. The advantage of producing ESI in its native form is that the files are produced as they are maintained in the usual course of business, the files can be searched for specific terms, and the files contain meta and embedded data. Native formatting has several drawbacks. Affixing tracking numbers and redacting privileged information can be difficult. Reviewing meta and embedded data can be costly and time consuming. Native files can be unintentionally altered by the requestor. Last, databases can be used to “create” exhibits unfamiliar to a witness if information is displayed in an unexpected arrangement.

B. The Production of Metadata

A recurring issue is the requestor’s need for meta or embedded data. Although meta and embedded data are an inherent part of an electronic file, the substance of the file can and often is produced without the metadata. Production of the electronic file with metadata often adds greatly to the time and cost of production. As a general proposition, if a court orders or a party agrees to produce ESI as it is maintained in the ordinary course of business, the ESI should be produced with the metadata intact.³⁷ The requestor, however, has the burden to demonstrate the relevance and the need for the metadata to justify the burden and cost of production and review.³⁸

C. How a Missouri Court Might Address Disputes Regarding the Form for Producing ESI

Missouri has no rules specifically addressing the form for producing ESI. Rule 58.01(c)(4) is different than the analogous Federal Rule of Civil Procedure 34(b)(i) in that it provides that a “party who produces documents for inspection shall produce them as they are kept in the usual course of business or shall organize and label them to correspond with the categories in the request.” Although this rule was presumably written with organization of paper documents in mind, it may be the basis to challenge a production form different from the way the ESI was stored “in the usual course of business.”³⁹

The parties should discuss production forms early and agreement can be addressed in a Rule 56.01(f) stipulation. If not resolved early, the requestor could specify the desired production form in the Rule 58.01 request for production and the producer could seek a different form by objecting and moving for a protective order directing production on different terms or conditions.

C. Evidentiary Admissibility of ESI

Although beyond the scope of this article and not affected by the amendments to the federal rules, admissibility of ESI as evidence will present challenges to lawyers and courts. During production, counsel must take proper steps to ensure that ESI is admissible as evidence for summary judgment and at trial.⁴⁰

V. Asserting Privilege After Inadvertent Production

The potential harm from waiving the attorney-client privilege by inadvertently disclosing privileged ESI and the detailed review effort necessary to avoid waiver add substantial cost and delay to producing ESI. The cost to review a large production can be staggering and take months because certain information does not appear on the face of a document, but only in its embedded data or metadata (e.g., draft language, comments, and deletions).

To shorten the review process and reduce cost, new Federal Rule of Civil Procedure 26(b)(5)(B) provides as a default that the producing party may designate material as privileged after it has been produced, and that the disclosure would not constitute a waiver or privilege. The rule does not affect the substantive legal issue of privilege waiver. The changes also acknowledge that the parties may agree between themselves to specific procedures to deal with inadvertent disclosure (“clawback” or “quick-peek” agreements). These agreements may be incorporated into a court order. Despite these rules, a negotiated non-waiver ESI production agreement does not excuse failure to undertake reasonable pre-production privilege review.⁴¹ Moreover, these agreements, even if incorporated into a court order, do not bind third parties. Future amendments to Federal Rule of Evidence 502 may make these agreements binding on third parties and possibly also on state courts.

Federal and state courts⁴² have adopted one of three general approaches regarding the level of inadvertence necessary for an inadvertent production to constitute a privilege waiver: “(1) the lenient approach” (only a knowing production is a waiver), (2) the strict approach (any production is a waiver), and (3) the “middle-of-the-road” approach (undertaking reasonable steps prevents waiver). New Federal Rule of Civil Procedure 26(b)(5)(B) applies the “middle of the road” approach, which is the approach applied by the Eighth Circuit and, presumably, by Missouri courts.⁴³

VI. Sanctions For Loss Or Destruction Of ESI

If there is one issue linked in the public’s perception to ESI, it is the potential for crippling sanctions. The common law creates a duty for all parties to preserve potential evidence. Breach of this duty, termed spoliation, is punishable by sanctions. Given the automatic destruction and overwriting of ESI and the ease with which operators can alter or delete ESI, it is not surprising that spoliation has become the most vexatious and dangerous aspect of ESI discovery.

It is now well-established in the federal courts that counsel has the duty to monitor compliance with the client’s preservation efforts so that all sources of discoverable ESI are identified and searched; and, having identified all sources of potentially relevant ESI, a party and its counsel are under a duty to retain that ESI and produce ESI responsive to the opposing party’s requests.

A. Imposing Sanctions

The loss of potentially relevant information can constitute spoliation if three elements are met: (1) a duty to preserve the information existed at the time it was destroyed; (2) a culpable state of mind; and (3) the evidence was relevant to the litigation.

A federal court has three sources for its power to sanction breaches of the duty to preserve: Federal Rules of Civil Procedure 26(g), 37(b)(2), and 37(c)(1); 28 U.S.C. §1927; and the court’s inherent power. Generally, only a court’s inherent power permits sanctions for conduct occurring before the initiation of litigation. The court has various types of sanctions at its disposal, including imposing fines, attorney’s fees, or costs; default; dismissal; adverse jury instructions; witness exclusion; and fact establishment. Although not a traditional sanction, a client who uses legal advice to develop or implement a record destruction policy later found to constitute spoliation can lose the protections of the attorney-client privilege under the crime-fraud exception.⁴⁴

Missouri courts recognize similar sanction options, including adverse evidentiary inferences and cost sanctions under Rules 55.03, 61.01(d), attorney discipline under Rules 4-3.3 and 4-3.4, and criminal liability for tampering with physical evidence under § 575.100, RSMo.⁴⁵

B. The Duty to Preserve Potentially Relevant ESI

Both state and federal common law impose a duty to preserve relevant information whenever litigation is reasonably foreseeable, which often occurs before a lawsuit is served. The duty equally applies to plaintiffs and defendants. Determining precisely when the duty starts and what material it covers are difficult issues and case law has failed to provide consistent benchmarks to guide litigators. Indeed, one company received opposite rulings from two different federal courts that examined the identical behavior.⁴⁶

1. When the Duty to Preserve Starts

The duty to preserve commences when litigation is likely or probable, not when litigation is merely possible.⁴⁷ Although a simple threshold in theory, it has proven difficult to apply in practice. So, while receipt of a demand letter triggers the duty to preserve,⁴⁸ receipt of a letter merely addressing a dispute without threatening litigation may not.⁴⁹ Informal complaints, however, especially in employment cases, can trigger the duty to preserve evidence.⁵⁰ Justifying a failure to preserve evidence by arguing that threats of litigation are a common negotiating tactic is ill-advised.⁵¹

An open question exists as to whether the duty to preserve applies when there is no specific litigation threat, but future litigation in a particular area is probable. Several Eighth Circuit courts have found a duty to preserve if “lawsuits or complaints have been filed frequently concerning [the] type of records at issue” and questioned the reasonableness of applying routine destruction policies to those records, even in the absence of a specific litigation threat.⁵² This concept, however, seems at odds with the Supreme Court’s dicta that “[u]nder ordinary circumstances, it is not wrongful for a manager to instruct his employees to comply with a valid document retention policy, even though the policy, in part, is created to keep certain information from others, including the government.”⁵³

2. What Must Be Preserved

The duty to preserve requires preservation of all information relevant to the potential litigation. This, however, can be a difficult line to draw. Generally, it requires preservation of relevant e-mails, databases, spreadsheets, hard drive data, and data on a server, including all metadata. On the other hand “a party need not” “preserve every shred of paper, every e-mail or electronic document, and every backup tape,” nor does the preservation obligation require freezing of all electronic documents and data, including e-mail.⁵⁴

Moreover, there is open disagreement on the duty to preserve’s contours. Although most courts do not require the preservation of disaster recovery back-up tapes, in some circumstances they do.⁵⁵ Once the duty to preserve is triggered, some courts see nothing wrong with moving active data to a back-up storage media, so long as nothing is lost,⁵⁶ while other courts find this sanctionable conduct.⁵⁷

Additionally, modern information systems have created questions that currently lack judicial guidance for preservation decisions. For instance, how does the duty to preserve apply to information that is only intended for brief temporary storage, such as voice mail? How does the duty to preserve apply to systems that normally do not preserve information but could be reconfigured to do so? How does the duty to preserve apply to systems, such as dynamic databases, that are designed to reflect only the current information status and which inherently undergo continual changes? Does the duty to preserve require preservation of materials that would not be kept in the ordinary course of business, such as drafts and notes?

C. A Culpable State of Mind

The most severe sanctions-adverse jury instructions, default, or dismissal—are usually imposed only for intentional, bad faith, or grossly negligent acts. Intentional bad faith destruction of relevant information always warrants the most severe sanctions. Almost as blameworthy is the willful indifference or reckless disregard of the duty to preserve that leads to the loss of information. Negligent conduct that leads to the loss of information, usually due to ineffective preservation efforts, usually receives less severe sanctions. Some federal circuits, however, permit harsh sanctions for merely negligent conduct.⁵⁸ Before imposing sanctions for spoliation, Missouri courts require evidence of intentional destruction indicating fraud, bad faith, and a desire to suppress the truth.⁵⁹

D. Federal Rule 37’s Safe Harbor From Spoliation Sanctions

Federal Rule of Civil Procedure 37(e) adopted a partial safe harbor from discovery sanctions: “Absent exceptional circumstances, a court may not impose sanctions under these rules for failing to produce [ESI] lost as a result of the routine, good-faith operation of an electronic information system.” This rule did not alter state law governing preservation of evidence or spoliation. Three phrases limit the rule’s protections.

1. “Routine . . . Operation”

A routine operation includes recycling of back-up tapes kept only for brief periods; automatic overwriting of information that has been deleted; programs that change metadata to reflect the latest access to particular information; and programs that automatically delete information not accessed within a defined period or that exceeds a certain age.

2. “Good Faith Operation”

Good faith implicates the level of effort to preserve potentially relevant information and requires interrupting routine destruction operations when the duty to preserve triggers.

3. “Under These Rules”

The safe harbor applies only to sanctions for loss after the initiation of litigation because the rules only apply once litigation commences. The safe harbor does not apply to sanctions brought under the court’s inherent power or under 28 U.S.C. § 1927.

D. How to Protect Your Client from Spoliation Sanctions

1. Create Your Own Safe Harbor

The easiest way to avoid sanctions is for the parties to create their own safe harbors by agreeing to specific preservation obligations and discovery protocols. This sets expectations and avoids costly disputes. The parties should reduce these agreements to formal preservation orders entered by the court.

2. Create and Enforce a Valid Record Retention / Destruction Program

An effective way to minimize the risk of sanctions is to have and consistently enforce a valid record retention policy. These policies identify what information needs to be retained as a record and for how long, and authorize destruction of all other information. Courts respect the need to manage information storage and to purge information that no longer serves a business purpose. The most common approach is age-based purge rules, especially for e-mails. Under these rules, all e-mail over a certain number of days old is automatically eliminated unless the operator takes affirmative action to preserve the e-mail.

3. Implement a Litigation Hold

Once a party believes that litigation is likely, it must issue what has come to be known as a “litigation hold.” A litigation hold’s purpose is to ensure preservation of all relevant information and requires the immediate suspension of any automatic purging of potentially relevant ESI. A documented, reasonable, and good faith preservation effort can provide protection from spoliation sanctions in the event of unintentional ESI destruction.

A litigation hold is usually an official directive sent to all individuals who may possess or control relevant information. Both fact witnesses and the computer department must be put on notice to preserve relevant information. The litigation hold should be hand-delivered, if possible, to ensure the recipients understand its importance. It should describe the types of information that must be preserved, contain clear instructions of how to preserve the different types of information, e.g., what to do with e-mails, electronic documents, paper documents, or voice mails, and it should require each recipient to reply and identify the information found and what was done with it. Additionally, follow-ups and reminders must be sent periodically, especially if relevant information continues to be created.

VI. Conclusion

ESI discovery presents many unique issues for lawyers and courts. These issues – cost, time, production, privilege, spoliation, and preservation – are not unique to ESI, but ESI presents unique problems. As courts and lawyers address these issues, the existing procedural rules must be applied flexibly to obtain a prompt, inexpensive resolution based on the merits, not one driven by excessive cost or burden. Moreover, the changes regarding ESI create new duties for all counsel, both outside and in-house. Counsel must also become fully versed in both the basics of the technology and with developing legal standards.

Footnotes

¹ Clayton L. Barker is engaged in private practice in Leawood, Kansas, focusing on electronic discovery, and is a 1997 graduate of the University of Kansas School of Law.

2 Although it is attentive to issues presented by ESI discovery, the Supreme Court of Missouri has no imminent plans to amend or supplement the current rules with ESI specific provisions.

3 See, e.g., Zubulake v. UBS Warburg LLC, 217 F.R.D. 309, 317 and n.38 (“Zubulake I”). Judge Scheindlin authored seven opinions in this case; four are frequently cited regarding ESI. The opinions, however, pre-date the new federal rules.

4 Fed. R. Civ. P. 34(a).

5 Metadata is information about a particular file which describes how, when, and by whom the file was created, accessed, or modified; its size; and how it is formatted. Embedded data include commands that control data, such as formulas in spreadsheets.

6 System data is information about the computer system, as opposed to a particular file, that includes when people logged on and off a computer, the programs and passwords they used, what files they accessed, and what websites they visited.

7 Zubulake I, 217 F.R.D. at 311 (“[D]iscovery is not just about uncovering the truth, but also about how much of the truth the parties can afford to disinter.”).

8 Bell Atlantic Corp. v. Twombly, 127 S. Ct. 1955, 1967 (2007).

9 Fed R.Civ. P. 34.

10 See Rule 4-1.1.

11 See Rules 4-1.4 and 4-2.1.

12 See Rule 4-3.3.

13 Cache La Poudre Feeds, LLC v. Land O’Lakes, Inc., 244 F.R.D. 614, 624-30 (D. Colo 2007).

14 Suggested Protocol for Discovery of Electronically Stored Information (“ESI”), United States District Court for the District of Maryland, Local Rules, available at http://www.mdd.uscourts.gov/localrules/localrules.html (last visited January 3, 2008).

15 Guidelines for Discovery of Electronically Stored Information, United States District Court for the District of Kansas, Guidelines, available at http://www.ksd.uscourts.gov (click on “Guidelines”) (last visited January 3, 2008).

16 Backup Tapes refer to exact copy of an entire computer system that permit a recreation of the entire system in the event of a system disaster. Generally, the tapes cannot be searched to identify, access, or retrieve individual files.

17 Legacy Data is information created or stored with software or hardware that has become obsolete or replaced, and thus it may be costly to restore or reconstruct.

18 Deleted Data is formerly active data that has been marked as available for overwriting and is no longer accessible by the operator even though all or some of the data remains.

19 See Zubulake I, 217 F.R.D. at 318-19.

20 Archival data is information that is intentionally maintained in long-term storage, but not immediately accessible to a computer system user. It is typically stored in an organized way to help identify, access, or retrieve individual records or files.

21 Federal Rule of Civil Procedure 30(b)(6) permits deposing a person knowledgeable of a party’s computer system. Heartland Surgical Specialty Hosp. LLC v. Midwest Div., Inc., Slip Copy No. 05-2164-MLB-DWB, 2007 WL 1054279 (D. Kan., Apr. 9, 2007) provides guidance on the deposition notice’s contents and what the deponent should know.

22 Oppenheimer Fund, Inc., v. Sanders, 437 U.S. 340, 358 (1978).

23 See e.g., Thompson v. U.S. Dep’t of Hous. & Urban Dev., 219 F.R.D. 93, 98 (D. Md. 2003) (Grimm, J.); Analog Devices Inc. v. Michalski, No. 01 CVS 10614, 2006 WL 3287382 (N.C. Super. Nov. 1, 2006).

24See, e.g., McPeek v. Ashcroft, 202 F.R.D. 31, 34 (D. D.C. 2001).

25 Zubulake I, 217 F.R.D. at 322 (describing cost-shift analysis); Zubulake v. UBS Warburg LLC, 216 F.R.D. 280, 284 (S.D. N.Y. 2003) (Zubulake III) (applying cost-shift analysis).

26 Section of Litigation, American Bar Association, Civil Discovery Standards, Civil Discovery Standards*, American Bar Association, 2004 Amendments, § 29(b)(iv) (2004), available at http://www.abanet.org/litigation/discoverystandards/2004civildiscoverystandards.pdf (last visited January 3, 2008.)

27 The Sedona Principles Addressing Electronic Document Production (The Sedona Conference, July 2005) available at http://www.thesedonaconference.org. See Working Group Series, WG1: Electronic Document Retention and Production.

28 Richard Van Duizend, Conference of Chief Justices, Guidelines for State Trial Courts Regarding Discovery of Electronically-Stored Information (NCSC Publications 2006) available at http://www.ncsconline.org/WC/Publications/CS_ElDiscCCJGuidelines.pdf.

29 Texas Rule of Civil Procedure 196.4 (“If the court orders the responding party to comply with the request [for unreasonably accessible ESI], the court must also order that the requesting party pay the reasonable expenses of any extraordinary steps required to retrieve and produce the information.”)

30 Balfour Beatty Rail Inc. v. Vaccarello, No. 3:06-cv-551-J-20MCR, 2007 WL 169628 at *3 (M.D. Fla., Jan 18, 2007).

31 Curdy Group LLC v. American Biomedical Group, Inc., 9 Fed. Appx. 822, 831 (10th Cir. 2001) (skepticism and mere desire to check that opposition has been forthright in its discovery responses is not a basis for production of hard drives.)

32 See, e.g., Ameriwood Indus. Inc. v. Liberman, No. 4:06CV524-DJS, 2006 WL 3825291 at *4 (E.D. Mo., Dec. 27, 2006).

33 Id. at *5-6.

34 See, e.g., State ex rel. Blue Cross and Blue Shield of Mo. v. Anderson, 897 S.W.2d 167, 169 (Mo. App. S.D. 1995); State ex rel. Anheuser v. Nolan, 692 S.W.2d 325, 328 (Mo. App. E.D. 1985).

35 See, e.g., Goss Int’l Americas Inc. v. Graphic Mgmt Assocs. Inc., No. 05 C 5622, 2007 WL 161684 at *3 (N.D. Ill Jan. 11, 2007); Hagenbuch v. 3B6 Sistemi Elettronici Industriali S.R.L., No. 04 C 3109, 2006 WL 665005 (N.D. Ill. Mar. 8, 2006).

36 Fed. R. Civ. P. 26(b)(2)(E)(i).

37 Williams v. Sprint/United Mgmt. Co., 230 F.R.D. 640, 652 (D. Kan. 2005) (Waxse, J.) (required production of metadata associated with certain spreadsheets)

38 Williams v. Sprint/United Mgmt Co., Civil Action No. 03-2200-JWL-DJW, 2006 WL 3691604, at *7 (D. Kan., Dec. 12, 2006) (Waxse, J.) (denied the request for metadata associated with certain e-mails).

39 Fed. R. Civ. P. 26(b)(2)(E)(i).

40 The leading case on ESI admissibility is Lorraine v. Markel Am. Ins. Co., 241 F.R.D. 534 (D. Md. 2007) (Grimm, J.) (denying cross motions for summary judgment for failure of both parties to present admissible ESI evidence).

41 Hopson v. Mayor and City Council of Baltimore, 232 F.R.D. 228, 244 n.39 (D. Md. 2005).

42 Gray v. Bicknell, 86 F.3d 1472, 1483-84 (8th Cir. 1996).

43 Id.; State ex rel. Tracy v. Dandurand, 30 S.W.3d 831, 835 (Mo. banc 2000).

44 A.G. Rambus Inc. v. Infineon Techs. A.G., 220 F.R.D. 264, 279 (E.D. Va. 2004).

45See Pikey v. Bryant, 203 S.W.3d 817, 825 (Mo. App. S.D. 2006).

46 Compare Rambus Inc. v. Infineon Techs. A.G., 220 F.R.D. 264 (E.D. Va. 2004) (destruction unreasonable) with Hynix Semiconductor Inc. v. Rambus Inc., No. C-00-20905 RMW, 2006 WL 565893 (N.D. Cal. Jan. 5, 2006) (destruction reasonable).

47 Cache La Poudre, 644 F.R.D. at 621 (litigation is always possible).

48 Oxford House, Inc. v. City of Topeka, Kansas, Slip Copy, 2007 WL 1246200, at *3 (D. Kan., Apr. 27, 2007); Consol. Aluminum Corp. v. Alcoa, Inc., 244 F.R.D. 335, 2006 WL 2583308 at *3, n.8 (M.D. La. 2006).

49 Cache La Poudre, 644 F.R.D. at 622.

50Broccoli v. Echostar Communications Corp., 229 F.R.D. 506, 513 (D. Md. 2005); Quinby v. WestLB AG, 245 F.R.D. 94, 2006 WL 2597900 at * 10 (S.D. N.Y 2006) (Quinby II).

51 In re Napster, Inc. Copyright Litig., 462 F.Supp.2d 1060, 1069-70 (N.D. Cal. 2006).

52 E*Trade Securities LLC v. Deutsche Bank AG, 230 F.R.D. 582, 589 (D. Minn. 2005); Lewy v. Remington Arms Co., 836 F.2d 1104, 1112 (8th Cir. 1988).

53 Arthur Andersen LLP v. United States, 544 U.S. 696, 704 (2005).

54 Oxford House, Slip Copy, 2007 WL 1246200 at *4; Zubulake v. UBS Warburg LLC, 220 F.R.D. 212, 217-18 (S.D. N.Y. 2003) (Zubulake IV)

55 Keir v. Unumprovident Corp., No. 02 Civ. 8781 (DLC), 2003 WL 21997747 at *1 (S.D. N.Y. Aug. 22, 2003).

56 Quinby v. WestLB AG, 245 F.R.D. 94, 2006 WL 2597900 at *9, nn.12, 13 (S.D. N.Y. 2006) (Quinby II); Quinby v. WestLB AG, No. 04Civ.7406 (WHP)(HBP), 2005 WL 3453908, at *8 n.10 (S.D. N.Y., Dec. 15, 2005) (Quinby I).

57 Treppel v. Biovail Corp., 233 F.R.D. 363, 372 and n.4 (S.D. N.Y. 2006).

58 Compare Greyhound Lines, Inc. v. Wade, 485 F.3d 1032, 1035 (8th Cir. 2007) (sanctions require bad faith ) with In re NTL, Inc. Securities Litigation, 244 F.R.D. 179 at *20 (S.D. N.Y. 2007) (only negligence).

59 Baldridge v. Dir. of Revenue, State of Mo., 82 S.W.3d 212, 223 (Mo. App. W.D. 2002).