UNITED STATES TAX COURT



     DYNAMO HOLDINGS LIMITED PARTNERSHIP,
  DYNAMO, GP, INC., TAX MATTERS PARTNER, Petitioner v.
   COMMISSIONER OF INTERNAL REVENUE, Respondent

          BEEKMAN VISTA, INC., Petitioner v.
    COMMISSIONER OF INTERNAL REVENUE, Respondent



Docket Nos. 2685-11, 8393-12.                Filed September 17, 2014.



       R requests that Ps produce electronically stored information
contained on two backup storage tapes or, alternatively, the tapes
themselves (or copies thereof). Ps acknowledge that the tapes contain
tax-related information but assert that the tapes also contain
privileged information that Ps have a right or duty to protect. Ps
assert that they must review the responsive information on the tapes
before giving the information to R to ensure that privileged or
confidential information is not disclosed. Ps request that the Court let
them use “predictive coding”, a technique prevalent in the technology
industry but not yet formally sanctioned by this Court, to help identify
the information that is responsive to R’s request.

      Held: Ps may use predictive coding in responding to R’s
request.
                                         -2-

      Martin R. Press, Edward A. Marod, Lu-Ann Mancini Dominguez, and Alan

Stuart Lederman, for petitioners.

      David B. Flassing and Lisa Goldberg, for respondent.



                                      OPINION


      BUCH, Judge: These consolidated cases are before the Court on

respondent’s motion to compel production of documents.1 The cases concern

various transfers from Beekman Vista, Inc. (Beekman), to a related entity,

Dynamo Holdings Limited Partnership (Dynamo). Respondent determined that

the transfers are disguised gifts to Dynamo’s owners. Petitioners assert that the

transfers are loans.

      Respondent requests that petitioners produce the electronically stored

information (ESI) contained on two specified backup storage tapes or,

alternatively, that they produce the tapes themselves (or copies thereof).

Petitioners assert that it will take many months and cost at least $450,000 to fulfill

respondent’s request because they would need to review each document on the

tapes to identify what is responsive and then withhold privileged or confidential

      1
       Respondent also moved to compel interrogatories. We will separately
address that motion in an order.
                                        -3-

information. Petitioners request that the Court deny respondent’s motion as a

“fishing expedition” in search of new issues that could be raised in these or other

cases. Alternatively, petitioners request that the Court let them use predictive

coding, a technique prevalent in the technological industry but not yet formally

sanctioned by this Court, to efficiently and economically identify the

nonprivileged information responsive to respondent’s discovery request.

      Respondent counters that he wants the backup tapes to review the ESI’s

metadata and verify the dates on which certain documents were created.

Respondent states that he also wants the backup tapes to ascertain all transfers

relevant to this proceeding. Respondent opposes petitioners’ request to use

predictive coding because, he states, predictive coding is an “unproven

technology”. Respondent adds that petitioners need not devote their claimed time

or expense to this matter because they can simply give him access to all data on

the two tapes and preserve the right (through a “clawback agreement”) to later

claim that some or all of the data is privileged information not subject to

discovery.2




      2
        We understand respondent’s use of the term “clawback agreement” to mean
that the disclosure of any privileged information on the tapes would not be a
waiver of any privilege that would otherwise apply to that information.
                                         -4-

      The Court held an evidentiary hearing on respondent’s motion. We will

grant respondent’s motion to the limited extent stated herein. Specifically, we

hold that petitioners must respond to respondent’s discovery request but that they

may use predictive coding in doing so.

                                    Background

I.    Relevant Entities

      A.     Beekman

      Beekman is a corporation wholly owned by a Canadian entity which is

controlled by Delia Moog. Beekman’s mailing address was in Florida when its

petition was filed.

      B.     Dynamo

      Dynamo is a limited partnership owned by a corporation and two trusts that

were established for Ms. Moog’s daughter and nephew. Dynamo’s tax matters

partner is Dynamo GP, Inc. Dynamo, through its tax matters partner, alleges that

its principal place of business was in Delaware when its petition was filed.

Respondent alleges that Dynamo’s principal place of business was in Florida at

that time.
                                        -5-

II.   Backup Tapes

      Dynamo backs up onto tapes its entire exchange server (inclusive of emails,

operating system, and configuration information). Dynamo performs this backup

work every four weeks and at the end of every month. Dynamo generally retains

its backup tapes for one year.

      Respondent seeks two of the backup tapes, specifically, the “Month End

August 2010 ORANGE” and the “Month End Jan 08 ORANGE”. These tapes

contain data backed up from (1) an exchange server and (2) a domain controller

and file server (KSH-DC). The exchange server database has approximately 200

mailboxes ranging in size from 500 megabytes to 1 gigabyte each. The KSH-DC

has a common group and a user group. The common group has shares where

assigned users may store data to be shared with other assigned users. The

common group has approximately 50 common top-level file shares and an

undetermined number of subfolders, and ownership of these files may not be

limited to the authors of the documents. The user group is in a section of the

network assigned to a specific individual and has approximately 200 user share

folders.
                                         -6-

III.   Petitioners’ Request To Use Predictive Coding

       Petitioners acknowledge that the two requested backup tapes contain

tax-related information but assert that the tapes also contain “personal

identification information, health insurance information, HIPAA protected

information and other confidential information that Petitioners have a duty to

protect.”3 Petitioners assert that if they must respond to respondent’s discovery

request, they must review the documents on the backup tapes to ensure that no

privileged or confidential information is disclosed before giving any information

to respondent. Petitioners ask the Court to let them use predictive coding to

efficiently and economically help identify the nonprivileged information that is

responsive to respondent’s discovery request. More specifically, petitioners want

to implement the following procedure to respond to the request:

             1. Restore some or all of the data from the tapes.

               2. Qualify the restored data; i.e., remove NIST files, system
       files, etc.[4]


       3
       The Health Insurance Portability and Accountability Act of 1996 (HIPAA),
Pub. L. No. 104-191, secs. 261-264, 110 Stat. at 2021-2033, contains privacy rules
and gave rise to privacy regulations relating to individually identifiable health
information.
       4
      The National Institute of Standards and Technology (NIST), which is an
agency of the U.S. Department of Commerce, maintains a database of hash values
                                                                  (continued...)
                                          -7-

            3. Index and load the qualified restored data into a review
      environment.

           4. Apply criteria to the loaded data to remove duplicate
      messages and other nonrelevant information.

             5. Through the implementation of predictive coding, review
      the remaining data using search criteria that the parties agree upon to
      ascertain, on the one hand, information that is relevant to the matter,
      and on the other hand, potentially relevant information that should be
      withheld as privileged or confidential information.

             6. Produce the relevant nonprivileged information and a
      privilege log that sets forth the claimed privileged documents and
      sufficient information supporting that claim.

                                      Discussion

I.    Discovery in General

      A party in this Court generally may obtain discovery of documents and ESI

to the extent that the information contained therein is not privileged and is relevant

to the subject matter of the case. See Rule 70(a)(1) and (b);5 see also Rule 72(a).6


      4
        (...continued)
of files that typically are part of an operating system or a piece of software. A
hash value, which is essentially a fingerprint of a file, is a numeric computation of
a file’s content which is used to identify the file. Two files with the same hash
values are exact copies of each other.
      5
          Rule references are to the Tax Court Rules of Practice and Procedure.
      6
          Rule 72(a) provides:

                                                                        (continued...)
                                        -8-

In this context, documents and ESI include “writings, drawings, graphs, charts,

photographs, sound recordings, images, and other data compilations stored in any

medium from which information can be obtained, either directly or translated, if

necessary, by the responding party into a reasonably usable form”.7 Rule 72(a)(1).



      6
          (...continued)
                     RULE 72. PRODUCTION OF DOCUMENTS,
           ELECTRONICALLY STORED INFORMATION, AND THINGS

            (a) Scope: Any party may, without leave of Court, serve on
      any other party a request to:

             (1) Produce and permit the party making the request, or
      someone acting on such party’s behalf, to inspect and copy, test, or
      sample any designated documents or electronically stored information
      (including writings, drawings, graphs, charts, photographs, sound
      recordings, images, and other data compilations stored in any medium
      from which information can be obtained, either directly or translated,
      if necessary, by the responding party into a reasonably usable form),
      or to inspect and copy, test, or sample any tangible thing, to the extent
      that any of the foregoing items are in the possession, custody, or
      control of the party on whom the request is served; * * *
      7
        Literature on electronic data storage has characterized electronically stored
data as falling within five categories. See Zubulake v. UBS Warburg LLC, 

, 318 (S.D.N.Y. 2003). These categories are active, online data (e.g.,
hard drives); near-line data (e.g., optical disks); offline storage/archives (i.e.,
removable optical disk or magnetic tape media); backup tapes (i.e., a device that
reads data from and writes it onto a tape); and fragmented, erased, or damaged
data (fragmented data consists of files that are broken up and placed randomly
throughout the disk). See 

The first three categories are generally
considered accessible, while the remaining categories are generally considered
inaccessible. See 

                                         -9-

And a party is generally required to produce documents or electronically stored

information in the form in which they are maintained. Rule 72(b)(3). A party,

however, is not required to provide discovery of ESI from sources that the party

establishes are not reasonably accessible because of undue burden or cost unless

the Court concludes that the requesting party has shown good cause for the

discovery.8 See Rule 70(c)(2). These Rules are all similar to corresponding

provisions found in the Federal Rules of Civil Procedure. See Fed. R. Civ. P.

34(a)(1)(A), (b)(2)(E), and 26(b)(2)(B).

II.   Respondent’s Request

      Respondent requests access to petitioners’ ESI. Petitioners resist this

request, primarily because of cost and of concern that privileged or confidential

information will be improperly disclosed. Respondent essentially responds that he

can alleviate both concerns if petitioners give him all of the requested information,

with a condition that he will allow them to later claim that some or all of that

information should not be disclosed further because it is privileged. Petitioners

remain mindful of their need to protect their privileged or confidential

information, as well as the projected cost of protecting that information, and ask


      8
       Petitioners do not claim that, if they use predictive coding, the requested
ESI is not reasonably accessible because of undue burden or cost.
                                        - 10 -

the Court to allow them to use predictive coding in responding to respondent’s

request.

       In this respect, we note that this request is somewhat unusual. Our Rules

are clear that “the Court expects the parties to attempt to attain the objectives of

discovery through informal consultation or communication” before resorting to

formal discovery procedures. Rule 70(a)(1). And although it is a proper role of

the Court to supervise the discovery process and intervene when it is abused by

the parties, the Court is not normally in the business of dictating to parties the

process that they should use when responding to discovery. If our focus were on

paper discovery, we would not (for example) be dictating to a party the manner in

which it should review documents for responsiveness or privilege, such as whether

that review should be done by a paralegal, a junior attorney, or a senior attorney.

Yet that is, in essence, what the parties are asking the Court to consider--whether

document review should be done by humans or with the assistance of computers.

Respondent fears an incomplete response to his discovery. If respondent believes

that the ultimate discovery response is incomplete and can support that belief, he

can file another motion to compel at that time. Nonetheless, because we have not

previously addressed the issue of computer-assisted review tools, we will address

it here.
                                         - 11 -

III.    Expert Witnesses

        Each party called a witness to testify at the evidentiary hearing as an expert.

Petitioners’ witness was James R. Scarazzo. Respondent’s witness was Michael L.

Wudke. The Court recognized the witnesses as experts on the subject matter at

hand.

        We may accept or reject the findings and conclusions of the experts,

according to our own judgment. See Chapman Glen, Ltd. v. Commissioner,

, 329 (2013). We also may be selective in deciding what parts (if

any) of their opinions to accept. See 

        The Court applies the standard of relevancy liberally when it comes to

matters of discovery, see, e.g., Zaentz v. Commissioner, 

, 471 (1979),

and a party challenging the requested production of a document (including ESI)

has the burden of establishing that the document is not discoverable, see Rutter v.

Commissioner, 

, 948 (1983); Branerton Corp. v. Commissioner, 

, 192-193 (1975).

        We believe that respondent’s request for the ESI is within the bounds of our

Rules, and petitioners do not appear to contest this point. At the same time,

however, we are faced with the competing interests of the parties. On one hand,
                                         - 12 -

we do not consider it appropriate to order petitioners to give all of their ESI to

respondent, subject to a right to later claim that some or all of the information that

he has reviewed is privileged or confidential information and thus outside the

bounds of discovery. Although the use of a clawback agreement may be an option

to which the parties might consent, petitioners reasonably resist entering into any

such agreement as part of a plan under which they would voluntarily allow

respondent to see all of the privileged or confidential information on the requested

tapes. On the other hand, given the time and expense involved with petitioners’

review of all the ESI to identify any privileged or confidential information, we

likewise do not consider it appropriate to order petitioners to go to that extreme

either.

          We find a potential happy medium in petitioners’ proposed use of predictive

coding. Predictive coding is an expedited and efficient form of computer-assisted

review that allows parties in litigation to avoid the time and costs associated with

the traditional, manual review of large volumes of documents. Through the

coding of a relatively small sample of documents, computers can predict the

relevance of documents to a discovery request and then identify which documents

are and are not responsive. The parties (typically through their counsel or experts)

select a sample of documents from the universe of those documents to be searched
                                        - 13 -

by using search criteria that may, for example, consist of keywords, dates,

custodians, and document types, and the selected documents become the primary

data used to cause the predictive coding software to recognize patterns of

relevance in the universe of documents under review. The software distinguishes

what is relevant, and each iteration produces a smaller relevant subset and a larger

set of irrelevant documents that can be used to verify the integrity of the results.

Through the use of predictive coding, a party responding to discovery is left with a

smaller set of documents to review for privileged information, resulting in a

savings both in time and in expense. The party responding to the discovery

request also is able to give the other party a log detailing the records that were

withheld and the reasons they were withheld.

      Magistrate Judge Andrew Peck published a leading, oft-cited article on

predictive coding which is helpful to our understanding of that method. See

Andrew Peck, “Search, Forward: Will Manual Document Review and Keyboard

Searches be Replaced by Computer-Assisted Coding?”, L. Tech. News (Oct.

2011). The article generally discusses the mechanics of predictive coding and the

shortcomings of manual review and of keyword searches. The article explains that

predictive coding is a form of “computed-assisted coding”, which in turn means

“tools * * * that use sophisticated algorithms to enable the computer to determine
                                       - 14 -

relevance, based on interaction with (i.e., training by) a human reviewer.” 

The article explains that

             Unlike manual review, where the review is done by the most
      junior staff, computer-assisted coding involves a senior partner (or
      team) who review and code a “seed set” of documents. The computer
      identifies properties of those documents that it uses to code other
      documents. As the senior reviewer continues to code more sample
      documents, the computer predicts the reviewer’s coding. (Or, the
      computer codes some documents and asks the senior reviewer for
      feedback.)

             When the system’s predictions and the reviewer’s coding
      sufficiently coincide, the system has learned enough to make
      confident predictions for the remaining documents. Typically, the
      senior lawyer (or team) needs to review only a few thousand
      documents to train the computer.

            Some systems produce a simple yes/no as to relevance, while
      others give a relevance score (say, on a 0 to 100 basis) that counsel
      can use to prioritize review. For example, a score above 50 may
      produce 97% of the relevant documents, but constitutes only 20% of
      the entire document set.

             Counsel may decide, after sampling and quality control tests,
      that documents with a score of below 15 are so highly likely to be
      irrelevant that no further human review is necessary. Counsel can
      also decide the cost-benefit of manual review of the documents with
      scores of 15-50.

             [Id.]

The substance of the article was eventually adopted in an opinion that states:

“This judicial opinion now recognizes that computer-assisted review is an
                                        - 15 -

acceptable way to search for relevant ESI in appropriate cases.” Moore v. Publicis

Groupe, 

, 183 (S.D.N.Y. 2012), adopted sub nom. Moore v.

Publicis Groupe SA, No. 11 Civ. 1279 (ALC)(AJP), 

(S.D.N.Y.

Apr. 26, 2012).

      Respondent asserts that predictive coding should not be used in these cases

because it is an “unproven technology”. We disagree. Although predictive coding

is a relatively new technique, and a technique that has yet to be sanctioned (let

alone mentioned) by this Court in a published Opinion, the understanding of

e-discovery9 and electronic media has advanced significantly in the last few years,

thus making predictive coding more acceptable in the technology industry than it

may have previously been. In fact, we understand that the technology industry

now considers predictive coding to be widely accepted for limiting e-discovery to

relevant documents and effecting discovery of ESI without an undue burden.10

See Progressive Cas. Ins. Co. v. Delaney, No. 2:11-cv-00678-LRH-PAL, 

       We use the term “e-discovery” to refer to “electronic discovery”, which in
turn means the obtaining of ESI in the discovery phase of litigation.
      10
         Predictive coding is so commonplace in the home and at work in that most
(if not all) individuals with an email program use predictive coding to filter out
spam email. See Moore v. Publicis Groupe, 

, n.2 (S.D.N.Y. 2012),
adopted sub nom. Moore v. Publicis Groupe SA, No. 11 Civ. 1279 (ALC)(AJP),

(S.D.N.Y. Apr. 26, 2012).
                                        - 16 -

3563467, at *8 (D. Nev. July 18, 2014) (stating with citations of articles that

predictive coding has proved to be an accurate way to comply with a discovery

request for ESI and that studies show it is more accurate than human review or

keyword searches); F.D.I.C. v. Bowden, No. CV413-245, 

, at

*13 (S.D. Ga. June 6, 2014) (directing that the parties consider the use of

predictive coding). See generally Nicholas Barry, “Man Versus Machine Review:

The Showdown between Hordes of Discovery Lawyers and a Computer-Utilizing

Predictive-Coding Technology”, 15 Vand. J. Ent. & Tech. L. 343 (2013); Lisa C.

Wood, “Predictive Coding Has Arrived”, 28 ABA Antitrust J. 93 (2013). The use

of predictive coding also is not unprecedented in Federal litigation. See, e.g.,

Hinterberger v. Catholic Health Sys., Inc., No. 08-CV-3805(F), 

(W.D.N.Y. May 21, 2013); In Re Actos, No. 6:11-md-2299, 

(W.D. La. July 27, 2012); Moore, 

. Where, as here, petitioners

reasonably request to use predictive coding to conserve time and expense, and

represent to the Court that they will retain electronic discovery experts to meet

with respondent’s counsel or his experts to conduct a search acceptable to

respondent, we see no reason petitioners should not be allowed to use predictive

coding to respond to respondent’s discovery request. Cf. Progressive Cas. Ins.

Co., 

, at *10-*12 (declining to allow the use of predictive
                                        - 17 -

coding where the record lacked the necessary transparency and cooperation among

counsel in the review and production of ESI responsive to the discovery request).

      Mr. Scarazzo’s expert testimony supports our opinion.11 He testified that

discovery of ESI essentially involves a two-step process. First, the universe of

data is narrowed to data that is potentially responsive to a discovery request.

Second, the potentially responsive data is narrowed down to what is in fact

responsive. He also testified that he was familiar with both predictive coding and

keyword searching, two of the techniques commonly employed in the first step of

the two-step discovery process, and he compared those techniques by stating:

      [K]ey word searching is, as the name implies, is a list of terms or
      terminologies that are used that are run against documents in a
      method of determining or identifying those documents to be
      reviewed. What predictive coding does is it takes the type of
      documents, the layout, maybe the whispets of the documents, the
      format of the documents, and it uses a computer model to predict
      which documents out of the whole set might contain relevant
      information to be reviewed.

             So one of the things that it does is, by using technology, it
      eliminates or minimizes some of the human error that might be
      associated with it. Sometimes there’s inefficiencies with key word
      searching in that it may include or exclude documents, whereas
      training the model to go back and predict this, we can look at it and
      use statistics and other sampling information to pull back the


      11
       Mr. Wudke did not persuasively say anything to erode or otherwise
undercut Mr. Scarazzo’s testimony.
                                        - 18 -

      information and feel more confident that the information that’s being
      reviewed is the universe of potentially responsive data.

He concluded that the trend was in favor of predictive coding because it eliminates

human error and expedites review.

      In addition, Mr. Scarazzo opined credibly and without contradiction that

petitioners’ approach to responding to respondent’s discovery request is the most

reasonable way for petitioners to comply with that request. Petitioners asked Mr.

Scarazzo to analyze and to compare the parties’ dueling approaches in the setting

of the data to be restored from Dynamo’s backup tapes and to opine on which of

the approaches is the most reasonable way for petitioners to comply with

respondent’s request. Mr. Scarazzo assumed as to petitioners’ approach that the

restored data would be searched using specific criteria, that the resulting

information would be reviewed for privilege, and that petitioners would produce

the nonprivileged information to respondent. He assumed as to respondent’s

approach that the restored data would be searched for privileged information

without using specific search criteria, that the resulting privileged information

would be removed, and that petitioners would then produce the remaining data to

respondent. As to both approaches, he examined certain details of Dynamo’s

backup tapes, interviewed the person most knowledgeable on Dynamo’s backup
                                       - 19 -

process and the contents of its backup tapes (Dynamo’s director of information

technology), and performed certain cost calculations.

      Mr. Scarazzo concluded that petitioners’ approach would reduce the

universe of information on the tapes using criteria set by the parties to minimize

review time and expense and ultimately result in a focused set of information

germane to the matter. He estimated that 200,000 to 400,000 documents would be

subject to review under petitioners’ approach at a cost of $80,000 to $85,000,

while 3.5 million to 7 million documents would be subject to review under

respondent’s approach at a cost of $500,000 to $550,000.

      Our Rules, including our discovery Rules, are to “be construed to secure the

just, speedy, and inexpensive determination of every case.” Rule 1(d). Petitioners

may use predictive coding in responding to respondent’s discovery request. If,

after reviewing the results, respondent believes that the response to the discovery

request is incomplete, he may file a motion to compel at that time. See Rule

104(b), (d).

      Accordingly,


                                                      An appropriate order will be

                                                issued.