*1 Before the Special Master is an unprecedented volume of discovery disputes, measured not only by the sheer number of disputed issues—a 101 page Redfern chart listing 70 issues in dispute—but also by the nearly one thousand pages of materials submitted by the parties in support of their discovery positions. (See Apr. 22, 2019 Joint Redfern for Discovery Issues (“Redfern”); Apr. 22, 2019 Letter from Novelis to Special Master (“Novelis Letter”); Apr. 22, 2019 Letter from Arconic to Special Master (“Arconic Letter”); Apr. 25, 2019 Reply Letter from Novelis to Special Master (“Novelis Reply”); Apr. 25, 2019 Reply Letter from Arconic to Special Master (“Arconic Reply”).) The Special Master's role is, of course, to resolve all disputes between the parties, but the parties' inability to resolve even the most basic discovery issues constitutes a substantial burden and expense—not least on the Court, who will have to review this Report & Recommendation, study each of 70 issues, and opine as to the correct result. Having now read and studied all of the disputed discovery issues, and having heard oral argument for many hours, the Special Master enters this Report & Recommendation to the Court.

During the June 17, 2019 oral argument (“Hearing”), the Special Master often asked questions that led to agreements between the parties resolving some of the disputes. Those agreements are memorialized below. On all other disputed discovery issues, rulings are recommended both on the disputes that were orally argued, as well as those that the Special Master concluded could be decided on the papers. As a predicate to the rulings, the Special Master reviews certain fundamental issues, which appear to have been a major reason for the extent of impasse reached by the parties in negotiating discovery on Novelis' Antitrust Counterclaims.

I. History of Parties' Antitrust Discovery Dispute

The Case Management Order regarding discovery on the Antitrust Counterclaims (“CMO”), which has been adopted by the Court, stated the following:

(CMO at 6.) Notwithstanding this clear directive, Arconic did serve “many broad requests” that were plainly cumulative of discovery that had been served and negotiated previously. The premise of the CMO is not that the prior discovery was in all respects sufficient, given the new issues raised by the Antitrust Counterclaims; rather that gaps in prior discovery should be specifically identified, rather than requesting “all documents” in areas that were already discovered, at least in substantial part.

*2 A prime example of this is discovery relating to purported alternatives to A951 treated aluminum, an issue that Arconic is understandably focused upon as it relates to Novelis' alleged market definition for the antitrust counterclaim. There is no dispute that this issue also had relevance to the original claims and counterclaims and that some discovery had already been conducted. Indeed, at least some of the possible alternative pretreatments, such as Titanium Zirconium, were included in the original search terms that yielded the documents already produced. Yet, despite the prior discovery and the clear directive of the CMO, Arconic served, by way of example, Request 40, which sought “[a]ll documents concerning substitutes for A951-treated aluminum in automotive applications.”[1]

At the Hearing, counsel for Arconic conceded that this request was cumulative of prior discovery and could not articulate what it did to review prior discovery to avoid being cumulative. (Hr'g Tr. at 19:14-20.) Yet Arconic insisted that the broad and cumulative request was “targeted.” Arconic defined “targeted” to mean “relevant and proportionate,” despite the clear directive of the CMO to avoid discovery on the antitrust counterclaim that plows much of the same terrain that discovery in the original counterclaim plowed. (Id. at 28:21-23.)

It is without question that the word “targeted” in the CMO meant something different from the basic standard of “relevant and proportional,” and no lawyer could read the Court's Order and conclude otherwise. This exemplifies the difficulties that the Special Master has worked through in preparing this Report & Recommendation to the Court. It has taken far more time than should have been necessary, caused delay and caused expense. Nonetheless, it is the Special Master's job to diligently work through each and every one of these disputed issues and make a recommendation to the Court, and that is what has been done.

The CMO was entered to manage discovery on the antitrust counterclaims, which started late because those counterclaims themselves were filed only after Arconic produced the Chemetall Agreement, after an unjustified five-month delay in its production of this document. At this late stage of discovery, it is not the generic “relevant and proportionate” standard that governs discovery. Rather, the applicable standard articulated explicitly in the CMO is that the discovery must be “targeted” to avoid overlapping the prior discovery on the other counterclaims made at the outset of the case.[2]

*3 Arconic chose not to comply with the order articulated in the CMO in drafting its discovery requests. There are virtually endless examples throughout the disputed document requests where Arconic sought “all documents” related to topics that have already been the subject of some discovery, rather than request the specific categories of additional information it needs to defend the antitrust counterclaims. Arconic is entitled to discovery in order to defend the antitrust counterclaims. That right is squarely set forth in the CMO. However, because the antitrust counterclaims were filed after the other counterclaims due to Arconic's improper delay in producing the key contract that begat the antitrust counterclaims, the CMO expressly required the parties to propound discovery demands that were targeted to information not already requested by it in connection with the other claims and counterclaims in the case that preceded the “new” antitrust counterclaims. By demanding “all documents,” and then stating during the meet and confer that it did not intend to re-request previously produced materials, Arconic shifted the burden to Novelis to tailor what would be produced to Arconic—precisely what the CMO explicitly barred. Arconic's non-compliance with the CMO was a significant cause of the volume of disputes that the parties were unable to resolve.

The above issues were then compounded by the parties' meet-and-confer process. During Hearing, Arconic acknowledged that it neither proposed, nor agreed to, any narrowing compromise on 80% of its document demands during the meet-and-confer conferences between the parties. At the Hearing, Arconic stated that it expressed a willingness to narrow 7 of its 35 disputed document requests. (Hr'g Tr. at 66:2-14.) And in its briefing papers to the Special Master, Arconic set forth no narrowing proposal on any of the 35 disputed document demands, except for two requests[3], making it an exceedingly time-consuming task for the Special Master to assess the parties' relative positions and determine the appropriate scope of discovery.

Specifically, for 22 of the disputed requests, Novelis' Redfern Chart set forth a list of categories of information responsive to Arconic's requests that Novelis believed responsive and proposed to produce. Rather than argue why Novelis' proposal was insufficient, or identify any gaps therein, Arconic simply complains repeatedly that Novelis is improperly attempting to “cherry pick” which documents it will produce, without saying what additional documents Arconic needs, thereby shifting the burden to the Special Master to ascertain what else Arconic needs. (See, e.g., Arconic Letter at 5 (citing Ashton Woods Holdings, LLC v. USG Group., 2016 WL 4414640, at *7 (E.D. Pa. Aug. 18, 2016)).) The CMO clearly states that Arconic is entitled to discovery to defend the antitrust counterclaims. But Arconic was ordered to state what those documents were, rather than serve overly broad requests and then issue repetitious, generic complaints about “cherry picking” when Novelis attempted to respond to such requests. What is left on that cherry tree is never stated by Arconic in its submissions to the Special Master.

Unlike Ashton, the case upon which Arconic bases it “cherry picking” complaint, Novelis is not proposing to produce only a sampling of documents individually selected by Novelis, but rather entire categories of responsive material.[4] For example, Arconic Request No. 38, demanded “[a]ll documents concerning any alleged harm, injury, damage or loss, without limitation, to Novelis caused by Arconic's alleged anticompetitive actions.” It is hard to imagine a broader document request. In an effort to respond nevertheless, Novelis' proposed to produce (1) sales data; (2) data sufficient to show the price Novelis paid Chemetall for A951 and other pretreatments; (3) documents concerning profitability analyses of A951 and other pretreatments; (4) documents concerning pricing analyses related to A951 and other pretreatments; and (5) documents concerning communications with Chemetall regarding the price of A951 and other pretreatments. (Redfern 8-10.)

*4 Those categories certainly seem quite comprehensive. Yet, Arconic did not say at the meet-and-confer what, if anything, was missing from those categories that it needed to defend the antitrust counterclaims. And its Redfern and letter briefs to the Special Master were equally opaque and silent. Arconic similarly made no argument as to what was missing from any of Novelis' narrowing proposals.

It simply does not do to lodge a complaint about being denied sufficient discovery, but then default when given the opportunity to say what else a party needs to discover. The meet-and-confer process is set forth in the rules for precisely this reason: to require the parties to state what information they really require, particularly when extremely broad discovery requests have been served. Here, that is especially true, because the CMO forbade the typical broad discovery requests for the reasons set forth above. In the meet-and-confer process, the recipient of discovery requests makes a proposal to the party seeking the information of what is feasible and proper to produce, and the counter party responds with what else it needs and believes that it is entitled to receive. Based on the Redfern, and confirmed at the Hearing, it appears that Arconic essentially declined to engage in this exercise at all, apparently preferring to rest on a generic complaint about “cherry-picking,” without stating what significant additional documents it desired, or to make any reasonable proposal of its own. This shifts the burden to the Special Master to infer from a blank slate what else Arconic might have proposed at the meet-and-confer or in its briefing papers. That is the task that this Report & Recommendation undertakes.

Arconic's explanation at the Hearing that it “would have made such a proposal in a place of ignorance” (Hr'g Tr. at 64:10-15) is nothing short of absurd. Arconic has sophisticated counsel, well experienced in antitrust law, and highly knowledgeable about the documents that large corporations maintain. Moreover, counsel has gained the knowledge from the millions of documents exchanged already in the case to know what types of documents exist. When corporations on both sides have tens of millions of documents, a party cannot simply demand “all documents,” and then go mute when asked to define any gaps in the counterparty's production proposal. Good faith participation in the meet-and-confer process is designed to prevent these disputes from requiring court intervention to the maximum extent possible, and it demands more than this. Even if Arconic did not know precisely what nomenclature Novelis used internally, it could have, and certainly should have, identified legally relevant categories of information that were not covered by Novelis' proposals. The inference is inescapable that Arconic simply chose not to do so for some unstated strategic reason.

Arconic had three opportunities to identify categories of information it needs, and defaulted each time: (i) in serving overbroad and cumulative requests in violation of the CMO, (ii) in refusing to participate in the meet-and-confer process in good faith, and (iii) in failing to make any argument in the Redfern and letter briefs submitted to the Special Master as to how any of Novelis' proposed responses to Arconic's requests were insufficient.

“Necessarily ... a party moving to compel another party to respond to a discovery request must ... inform the Court of how the response received to that request—if any—is deficient.” Parks, LLC v. Tyson Foods, Inc., 2015 WL 5042918, at *7 (E.D. Pa. Aug. 26, 2015) (“[T]he Court must know which request is at issue, how the response to that request is allegedly deficient, and why the producing party's response runs afoul of the discovery rules. Without this information, the Court cannot assess whether Plaintiff's alleged failure to provide.”); see also Royster v. Corizon, 2015 WL 853788, at *2 (M.D. Pa. Feb. 26, 2015) (moving party “must supply the court with ... [non-moving party's] response to [the] discovery request, and advise the court why the response is inadequate or otherwise improper” because “[w]ithout this information the court cannot accurately assess [the party's] motion to compel.”).

*5 Arconic's failure to make a showing of why Novelis' proposed responses are insufficient constitutes a waiver of its entitlement to relief under its motion to compel. “A party waives an argument if they fail to raise it in their opening brief.” Laborers' Intern. Union of North America, ALF-CIO v. Foster Wheeler Energy Corp., 26 F.3d 375, 398 (3d Cir. 1994) (failure to raise issue of arbitrability until the reply brief constituted waiver). Likewise, a party's failure to dispute an argument in its opposition brief constitutes waiver of the right to relief. Leisure Pass North America, LLC v. Leisure Pass Group, Ltd., 2013 WL 4517841, at *4 (Aug. 26, 2013) (“Plaintiff has waived its opposition to this argument by failing to respond to it.”); Duran v. Equifirst Corp., 2010 WL 936199, at *3 (D.N.J. Mar. 12, 2010) (“The absence of argument constitutes waiver in regard to the issue left unaddressed....”) Both in the Redfern and the Arconic Letter, Arconic failed to argue why Novelis' proposed response was insufficient on any of the 22 requests for which Novelis made a narrowing proposal. And even in the Arconic Reply, having had an opportunity to review Novelis' Letter and Redfern—which listed each proposed category in detail—Arconic again did not make any such argument. By its action, Arconic has waived any objection to Novelis' proposals of what it will produce in response to those 22 document requests.

It is thus far too late to ask for a fourth try to do what was ordered in the CMO. Three times is enough, and the job is now for the Special Master to do. Therefore, the Special Master will use her own substantial experience in presiding over a great many antitrust cases—including antitrust MDLs—to decide the discovery disputes in the deliberate absence of any counterproposals from Arconic. It is essential that discovery proceed forthwith, and not be delayed any further by one party's choice not to meaningfully participate in the meet-and-confer process. The Special Master will therefore issue its recommendations on these document requests, where Arconic has waived its opportunity to object, as stated below.

The parties also dispute whether it is appropriate, in certain circumstances, for Arconic to be ordered to produce “mirror image” discovery, even for categories of documents that Novelis did not request in its discovery requests served pursuant to the CMO's deadline to exchange substantially all requests. Arconic asserts that it would be unfair to allow Novelis to receive discovery that it did not seek before the deadline set forth in the CMO to propound antitrust discovery. (Arconic Reply at 4-5.) Novelis contends that while Arconic served numerous broad, cumulative requests in violation of the CMO, Novelis served discrete, non-cumulative requests, and thus did not ask for nearly as much as Arconic did, and should not now be punished for its effort to comply with the CMO. (Hr'g Tr. at 117:7-118:24.) Thus, Novelis argues, Arconic should not benefit from its violation of the CMO, while at the same time punish Novelis for its compliance with the CMO by holding Novelis to its more narrowly tailored requests.

The Special Master will not deny to Novelis reciprocal discovery on the ground that the request was not made, if it is apparent that Novelis was thereby attempting to comply with the CMO. However, the Special Master will not grant to Novelis mirror-image discovery in all instances; instead, each category will be considered on its own merits as to the basis for reciprocal discovery, based on the issues in dispute in the antitrust counterclaim.

At the Hearing, Novelis specifically requested reciprocal discovery on certain of Arconic's document requests, insofar as the Special Master recommends that Novelis be ordered to respond to them. (Id. at 120:17-122:22.) The party seeking reciprocal discovery must make a showing of the relevance of that discovery to its own case. The Special Master will undertake this analysis regarding the specific “mirror image” requests noted by Novelis in the discussion below as it relates to each category of documents.

II. Document Requests to Novelis by Arconic

*6 A series of Arconic requests seek “[a]ll documents concerning Novelis' allegation that ...” followed by various quotes of portions of the Amended Counterclaims. Novelis proposed categories of information to produce in response to these requests and contends that Arconic “refused ... to identify any gaps” in Novelis' proposals. (Novelis Letter at 7-8.)

The CMO did not authorize generic allegation document demands. First, they are anything but “targeted.” Second, it is extremely difficult to know which documents constitute the entirety of a request for “all documents concerning or relating to” a particular allegation in a complaint because it requires the producer to guess what the recipient thinks is a document that “concerns” or “relates to” a particular phrase. Thus this type of document demand is typically negotiated into categories of responsive material to be produced in the meet-and-confer session.

Where the Special Master has determined that some responsive documents should be produced, notwithstanding the improper demand, it so states below:

Topic 3/Request 37 seeks “all documents concerning” Arconic's alleged anticompetitive “scheme to raise the costs of Novelis and other rivals for sales of A951-treated aluminum by controlling and inflating prices.” It is, in essence, a request for every document that Novelis possesses “concerning” Novelis' antitrust counterclaims, including five sub-categories that read more like contention interrogatories than the “targeted” document discovery authorized by the CMO.

What this request, at its essence, appears to seek is the price charged to Novelis and “other rivals” for the A951 Chemicals sold by Chemetall. This topic was already the subject of some discovery conducted about Novelis' original counterclaims for price manipulation and breach of contract. Novelis has proposed to produce its pricing data and documents reflecting its pretreatment costs. (Redfern at 6-7.) Arconic does not argue in its papers why this proposed response to Request 37 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument.

Having reviewed the request and relevant law, the Special Master recommends that Novelis' proposed response addresses the subject of the Arconic's alleged scheme to raise Novelis' costs, and that it be ordered by the Court, without objection. The one issue not addressed in Novelis' proposed response is sub-part (a) regarding the identity of “any ‘rivals’ of Arconic that Novelis contends may have been affected by Arconic's alleged “scheme.” The Special Master recommends that Novelis be ordered to proffer whom it contends these “rivals” are, and whether it has any documents reflecting their costs for A951 chemicals sold by Chemetall.

Topic 7/Request 41 seeks “[a]ll documents relating to Arconic's alleged knowledge or purpose concerning the Chemetall Agreement's alleged effects on competition” and specifies as sub-categories documents related to three allegations in the Amended Counterclaims, such as “(a) Novelis' allegation that Arconic possessed an ‘intent to conceal the price differential’ that Novelis alleges the Chemetall Agreement reflects.” As written, the request is convoluted, vague and ambiguous. No party could ever verify that it has produced “all documents” “related to” the “knowledge,” intentions or motivations of its adversary.

*7 Novelis has proposed to produce “documents concerning communications with Chemetall regarding the price of A951.” (Redfern at 16.) Arconic does not argue in its papers why this proposed response to Request 41 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument. Nonetheless, because Novelis' proposal excludes internal Novelis documents, and communications with other parties, the Special Master recommends that, in response to Request 41, Novelis also produce any non-privileged Novelis internal memorandum or communication, and any non-privileged communication with another person or entity, regarding Arconic's alleged (i) concealment of the price differential for the Chemetall chemical, (ii) understanding that the inflated prices to Novelis would be passed on to customers, and (iii) knowledge that the inflated costs for the A951 chemicals would lessen the intensity of competition for A951 treated aluminum.

Topic 8/Request 42 seeks “all documents relating to” the alleged “substantial anticompetitive effect” of Arconic's alleged “scheme,” with sub-parts referencing various allegations by Novelis. The request is, again, overbroad, vague and ambiguous, requires substantial subjective interpretations, and not tailored to identified documents. Further, sub-parts (b) and (c) relate to an issue that has been in the case since the original counterclaims—Novelis' allegation that it paid inflated pricing for the A951 chemical, rather than pricing determined by “market forces” as provided in the Arconic-Novelis license. (See, e.g., Def. Mot. for Leave to File Am. Countercl., Ex. 2 (“Am. Countercl. Redline”) ¶ 268.) This topic has already been the subject of discovery, and Request 42(b) and (c) violate the CMO for seeking “all documents” related thereto.

To the extent that any documents are ordered to be produced in response to Request 42 prior to the exchange of expert reports, Novelis has proposed a production of its sales, cost and pricing data. (Redfern at 17.) Arconic does not argue in its papers why this proposed response to Request 42 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument.

The Special Master recommends that Request 42 be denied as overbroad and for violating the CMO. The Special Master recommends that Novelis produce its sales and pricing data, as set forth in the Novelis proposal, and that it be ordered by the Court, without objection. Further, upon exchange of expert reports, both parties shall on the same date produce the documents relied upon and considered by their respective experts that have not already been produced. This shall be a reciprocal requirement. In this way, both sides' experts will be able to submit rebuttal expert reports, with full knowledge of the documents relied upon by the counterparty's expert.

Topic 29/Request 69 seeks “all documents concerning” an allegation made in the Amended Counterclaims that Novelis' “market knowledge in supplying pre-treated aluminum coils to other customers” informed it that the correct “price for the A951 Chemicals should be much lower” than what Novelis was being charged under the Chemetall Agreement (citing Am. Countercl. ¶ 240). The substance of this allegation appeared in the original counterclaims. (See Redline ¶ 264.) Accordingly, this request for “all documents” is denied as cumulative of discovery previously sought in violation of the CMO. Further, Novelis proposed to produce pricing data, and to update certain prior productions. (Redfern at 46-47.) Arconic does not argue in its papers why this proposed response to Request 69 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument.

The Special Master recommends that Novelis produce the documents it proposed, and that it be ordered by the Court, without objection. Further, Novelis shall produce additional documents, if any, that it contends support its allegation that its market knowledge, based upon its sale of pre-treated aluminum coils to other customers, reflects what Novelis believes is “the correct price for the A951 Chemicals.” After expert reports are exchanged, Novelis shall supplement this production with any additional documents that are relied upon by its expert that support this allegation.

*8 Topic 30/Request 70 seeks “[a]ll documents concerning Novelis' allegation that the Target Price in the Chemetall Agreement includes ... consideration for expired patents” (citing Am. Countercl. ¶¶ 364-66). Novelis proposed that it respond by producing “[d]ocuments relating to analyses ‘to assign value to the expired patents or to apportion the value of the unpatented Technology and Know-How from the value associated with the expired patents,’ ” and by updating prior productions. (Redfern at 48.) Arconic does not argue in its papers why this proposed response to Request 70 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument. Furthermore, Novelis' proposed response directly responds to the information sought in the request. The Special Master recommends that Novelis' proposed production be deemed sufficient, and that it be ordered by the Court, without objection.

Topic 31/Request 71 seeks “[a]ll documents concerning Novelis' allegations that Arconic has prevented Chemetall from adjusting [ ] the price for the A951 Chemicals ... to reflect expiration of the patents.” (Citing Am. Countercl. ¶ 365). Novelis proposed to produce in response to Request 71 pricing analyses and “[d]ocuments concerning communications with Chemetall regarding the price of A951 and other automotive pretreatments.” (Redfern at 49.) Arconic does not argue in its papers why this proposed response to Request 71 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument.

Notwithstanding such waiver, in addition to the documents Novelis proposed to produce in response to Request 71, the Special Master recommends that Novelis be ordered to produce non-privileged communications with any party regarding the allegation that “Arconic has prevented Chemetall from adjusting [ ] the price for the A951 Chemicals ... to reflect expiration of the patents.”

Arconic shall likewise produce any documents in its possession or control that discuss or concern the price charged, or to be charged, by Chemetall for A951 chemical for the time period after expiration of the patent.

Topic 32/Request 72 seeks “[a]ll documents concerning Novelis' allegation that “inflated prices to Novelis (and other competitors) would be ‘passed on’ to customers such as Ford, at least in part” including “all evidence (including documents by Bates number) concerning any change in price to Ford or to other customers of A951 pretreated aluminum by Novelis or by other competitors.” Novelis offered to produce, in response to Request 72, sales and pricing data for A951, as well as communications with Ford and other auto-OEMs regarding the price of A951. (Redfern at 50-51.) Arconic does not argue in its papers why this proposed response to Request 72 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument. The Special Master recommends that the Court require Novelis to produce the documents it proposed, without objection.

Topic 35/Request 76 seeks “[a]ll documents concerning [Novelis'] Supplemental Damages Proffer” including documents related to various statements made by Novelis therein. Novelis did not offer to produce any new documents in response to this request, nor did Arconic offer any narrowing proposal. At the Hearing, upon questioning by the Special Master, Arconic clarified that it seeks the documents that support Novelis' damages theory and that would allow Arconic to test the veracity of the allegations. (Hr'g Tr. at 205:11-15, 208:11-17.) The parties reached an agreement that Novelis would produce documents in its possession, custody or control that it intends to rely on to prove its damages, in addition to the sales and pricing data it has previously agreed to produce (see, e.g., Redfern at 17), which experts would rely on in conducting a damages analysis. (Hr'g Tr. at 212:10-14, 214:11-17.) The parties noted that expert reports have not yet been drafted, and so there may be future documents that need to be identified and/or produced to the extent experts rely on them in the future. (Id. at 208:18-24.) Therefore, the documents to be produced for this category may be supplemented in the future, after the expert reports have been exchanged. Because the parties reached an agreement at the Hearing, the Special Master need not make a recommendation concerning this request.

*9 Topic 4/Request 38 seeks “[a]ll documents concerning any alleged harm, injury, damage or loss, without limitation, to Novelis caused by Arconic's alleged anticompetitive actions.” Novelis proposed to produce sales data, pricing data, pricing analyses and communications with Chemetall in response to Request 38. (Redfern at 8-10.) Arconic does not argue in its papers why this proposed response to Request 38 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument.

At the Hearing, Arconic agreed that Novelis' proposed response to this request and the parties' agreement with respect to Topic 35/Request 76 were sufficient to respond to Request 38. (Hr'g Tr. at 215:17-23.) Because the parties reached an agreement at the Hearing, the Special Master need not make a recommendation concerning this request.

Many Arconic requests relate to the relevant antitrust market, which Arconic asserts is the “heart of” its antitrust defense. (See, e.g., Redfern at 6.) Novelis counters that “the pricing of A951, alternatives (or lack thereof) to A951, and dealings with Chemetall and Ford” were central to the trade secret claims, and that Arconic already “aggressively took” discovery related to those issues. (Novelis Letter at 5.) Novelis asserts that Arconic's requests related to this topic were thus overbroad and cumulative in violation of the CMO because they broadly seek “all documents” and were not drafted around discovery already conducted, as the CMO required. (Id.)

The central question in the relevant market inquiry is the “reasonable interchangeability of use ... [which] implies that one product is roughly equivalent to another for the use to which it is put.” Queen City Pizza, Inc. v. Domino's Pizza, Inc., 124 F.3d 430, 437 (3d Cir. 1997). Another framing of the market definition is whether there is the “cross-elasticity of demand, in other words, the rise in the price of a good within a relevant product market would tend to create a greater demand for other like goods in that market.” Tunis Brothers Co., Inc. v. Ford Motor Co., 952 F.2d 715, 22 (3d Cir. 1991). Thus the two central discoverable issues regarding the relevant market are interchangeable substitutes for the use to which a product is put, and cross-elasticity of demand, in response to pricing.

The parties vigorously dispute the contours of the relevant market. The Special Master does not determine the correct market definition at this time, and instead will grant both parties the discovery needed to develop their respective theories, to the extent such discovery is not cumulative of prior discovery already taken, and otherwise complies with the CMO.

Novelis argues that A951's unique strength and durability was chosen by Ford to meet Ford's requirements for its F150 truck, which “is the largest-selling vehicle in America,” and that this has created a high premium price for A951, which is inelastic relative to other automotive materials due to Ford's clout as a huge customer of this pre-treated aluminum set in Ford's specifications for purchase. (Hr'g Tr. at 132:4-8, 134:21-22.) Arconic counters that “the market encompasses any materials that are used for structural applications in vehicles,” which is “a very competitive market in which a number of these materials are competing ... advantages and disadvantages.” (Id. at 138:20, 139:14-17.)

Novelis argues that prior assertions made by Arconic—that A951 was uniquely effective and therefore had no substitutes—forecloses Arconic from disputing Novelis' alleged relevant market—A951—and from seeking discovery related to that alleged market. (Novelis Letter at 20-21.) Arconic responds that “[s]aying something has ‘no performance rival’ is not the same as saying it has no market substitutes or alternatives.” (Arconic Reply at 8 n.3). The Special Master finds that prior statements by Arconic do not foreclose it from seeking discovery related to the relevant market for antitrust purposes. If the case moves to the merits, Arconic may have to deal with its prior statements, but this case is now only at the discovery stage.

*10 Topic 5/Request 39 seeks “[a]ll documents concerning the market for A951-treated aluminum alleged in the Counterclaims” and specifies twelve sub-categories. Novelis proposes several categories of documents to produce, including communications with Ford and Chemetall regarding A951 pricing, communications with other OEMs regarding pricing and selection of aluminum pretreatments, and various analyses, business plans and comparisons of A951 and other pretreatments. (Redfern at 11-12.) Arconic does not argue in its papers why this proposed response to Request 39 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument.

Further, the Special Master finds that the demand in Request 39 for “all documents” is overbroad, not particularized and cumulative of discovery about A951-treated aluminum that has already been conducted on the prior claims in the case. (For example, “all documents concerning whether any firm possesses market power in the alleged antitrust market” (Request 39(j)) is not a proper document request, to put it mildly). Accordingly, the Special Master recommends that Novelis produce its proposed responsive documents, as set forth in the Novelis Redfern proposal, and that it be ordered by the Court, without objection.

Additionally, as agreed at the Hearing, Novelis will produce any documents it intends to rely on in support of its case, including as to market definition, with the right to supplement after expert reports are completed and thereafter if additional documents are to be relied upon for this issue. (Hr'g Tr. at 208:25-209:6.)

Topic 6/Request 40 seeks “[a]ll documents concerning substitutes for A951-treated aluminum in automotive applications.” Novelis asserts that Arconic's desired discovery regarding the “internal chemical and process development of ‘alternatives’ ” was the subject of trade secret discovery that need not be repeated now. (Novelis Reply at 4.)[5] Thus an “all documents” request does not adhere to the Court-ordered mandate of the CMO to draft requests that are not cumulative of discovery that has already been sought.

Novelis states that it offered to produce several categories of information in response to this request. (See Redfern at 14.) Arconic does not argue in its papers why this proposed response to Request 40 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument.

Notwithstanding this waiver, when the Special Master asked at the Hearing for Arconic to identify any category of responsive information missing from Novelis' proposal, Arconic identified one category—documents relating to “what the other OEMs have discussed in terms of price” of automotive pretreated aluminum. (Hr'g Tr. at 20:14-19.) Arconic also took issue with the phrasing of “selection” in Novelis' proposal as possibly excluding communications with OEMs that Arconic would deem relevant. (Id. at 31:7-13.) Novelis clarified its use of “selection” referred broadly to Novelis' proposals, any OEM's feedback on such proposals, and communications relating to the pricing of A951 or its alternatives. (Id. at 45:7-12.) At bottom, Arconic stated that Request 40 seeks “documents that talk about what I can sell as an alternative to A951 aluminum for automotive applications,” including “competitive bids,” “internal[ ] evaluations,” and “guidance from industry consultants.” (Id. at 36:8-12, 39:17-40:7.)

*11 Based on Arconic's stated need, the Special Master recommends that Novelis be ordered to produce the categories of materials it offered (Redfern at 14), and that it be ordered by the Court, without objection, as well as (i) communications with automotive OEMs relating to pricing of any aluminum pretreatment for an automotive application; (ii) internal analyses or guidance from industry consultants regarding the suitability, functionality or interchangeability of any aluminum pretreatment for automotive applications; and (iii) communications with automotive OEMs regarding proposals, or feedback on proposals, or the OEMs' evaluation of, any aluminum pretreatment for an automotive application.

Novelis identified discovery related to the relevant market as an area it believes reciprocal production should be made by Arconic. Because pricing and interchangeability of use are the hallmarks of the market definition, the three categories of information identified above have equal relevance if produced from Arconic's files. For example, discussions between Arconic and OEM customers comparing the function of A951 with a pretreatment alternative would be relevant. Accordingly, the Special Master recommends that Arconic also be ordered to produce the three categories of information listed above.

Topic 10/Request 44 seeks “[a]ll documents concerning any automotive company's choice of materials to use for structural components in any vehicle and their reasons for such choice” for a specified date range. Novelis proposed to produce communications with Ford and other OEMs regarding the selection of automotive pretreatment and the price of A951, as well as various analyses and business plans regarding pretreated aluminum. (Redfern at 21-22.) Arconic does not argue in its papers why this proposed response to Request 44 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument.

At the Hearing, Arconic identified two ways in which Novelis' proposed response to this request was insufficient: that it (i) excluded steel, and (ii) failed to account for A951 alternatives that “are in developmental consideration” but not yet fully commercialized. (Hr'g Tr. at 174:7-14.)

In determining the appropriate scope of discovery on this important issue, the Special Master makes no finding about the correct market definition and instead seeks to provide both parties with the discovery needed to pursue any colorable and plausible theories recognized by Third Circuit antitrust law. Based on the colloquy at the Hearing, the unique aspect of Request 44 is that Arconic would include steel used for structural components in the scope of discovery. Though Novelis disagrees with Arconic's characterization that steel is an interchangeable product, and though Novelis does not manufacture steel, Arconic may attempt to pursue its theory in discovery. No decision is made at this time about whether steel is interchangeable with pretreated aluminum for its intended use.

Accordingly, the Special Master recommends that, in addition to the categories of documents in Novelis' proposal (Redfern at 21-22), Novelis be ordered to produce any analysis or communication with an OEM that compares the function or suitability of A951 pretreated aluminum with steel for structural components in an automotive application.

As to communications related to an alternative product that is not yet commercialized, and thus is not included on the list of known alternative products being used as search terms, the Special Master finds that a product must be already commercialized or pending immediate commercialization, in order to be discoverable. Products that are still at the R&D phase are too premature to have relevance to the market definition discovery dispute. Further, to require production of communications regarding developing products that are not yet on the market would allow Arconic to probe its competitor's R&D efforts for products that may never come to fruition and is too far attenuated from actual commercialization to have probative value. And the same would be true if reciprocal discovery were sought by Novelis to probe Arconic's R&D efforts.

*12 Novelis identified Request 44 as one for which it believed reciprocal production from Arconic was appropriate. (Hr'g Tr. at 122:6-15.) To the extent Arconic is in possession of communications with OEM customers comparing the function or suitability of steel and A951, those would be relevant for the same reason Arconic asserts such communications in Novelis' possession are relevant—customers' feedback and analysis of the function of products that are within Arconic's definition of the relevant market. Accordingly, the Special Master recommends that Arconic also be ordered to produce communications with OEM customers comparing the function or suitability of steel and A951.

Topic 11/Request 45 seeks “all documents” related to proposals by other entities for Ford's “bond durability tests.” The referenced paragraph in the Amended Counterclaims appeared almost verbatim in the original Counterclaims, and therefore this request is cumulative of prior discovery in violation of the CMO. (Am. Countercl. Redline ¶ 202.) Further, Novelis proposed to produce documents related to “Ford's selection of pretreatments for its P552 Program.” (Redfern at 22.) Arconic does not argue in its papers why this proposed response to Request 45 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument. The Special Master recommends that Novelis produce its proposed responsive documents, as set forth in the Novelis Redfern proposal, and that it be ordered by the Court, without objection.

Topic 15/Request 52 seeks “[a]ll documents concerning your purchase of 951 Chemicals (A951 Chemicals) or generic or substitute chemicals for pretreating aluminum from Chemetall, Arconic, or any other entity.” The issue of Novelis' purchase of A951 chemicals from Chemetall was in the case as part of the original Counterclaims (if not sooner.) It was the subject of several prior document requests (see Redfern at 27). Therefore, a broad “all documents” request violates the CMO. Further, Novelis proposed to produce pricing data and to update its prior production of communications with Chemetall. (Id. at 26-27.) Arconic does not argue in its papers why this proposed response to Request 52 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument.

The Special Master recommends that Novelis produce the documents set forth in the Novelis proposal, and that it be ordered by the Court, without objection.

Topic 18/Request 55 seeks “[a]ll documents concerning your efforts to market and sell pretreated aluminum” including A951 pretreated aluminum as well as a list of known alternatives. Novelis asserts that a properly targeted request would focus on “negotiations with OEMs relating to pretreatment price and selection of alternatives” and proposed to produce various categories of documents, including sales data, and to update prior productions, including communications with Ford and other OEMs regarding pretreatment selection and various business plans and analyses of pretreated aluminum. (Redfern at 31-32.) Arconic does not argue in its papers why this proposed response to Request 55 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument.

Notwithstanding such waiver, at the Hearing, the Special Master inquired of Arconic what other relevant information it sought beyond what was proposed by Novelis. Arconic identified one category: documents related to Novelis' efforts to market and sell pretreated aluminum to “downstream participants,” meaning not OEMs but also chemical manufacturers. (Hr'g Tr. at 183:8-184:7.) Novelis clarified that its production would include sales data for all of its customers, including sales quantity and price and that if responsive documents are located relating to some other type of customer, Novelis will include such documents in its production. (Id. at 185:8-19.) Because the parties reached an agreement at the Hearing, the Special Master need not make a recommendation concerning this request.

*13 Topic 28/Request 67 seeks “[a]ll documents concerning any marketing, advertising, proposals, or other statements or communications about pretreated aluminum or Novelis' ability or capacity to supply pretreated aluminum.” Novelis proposed to produce marketing and sales analyses regarding the sale of pretreated aluminum and communications with Ford and other OEMs regarding the selection of pretreatments. (Redfern at 45-46.) Arconic does not argue in its papers why this proposed response to Request 67 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument.

Further, the subject of communications with OEMs about pretreated aluminum is addressed by the Special Master's recommendation as to what Novelis shall produce in response to Requests 40 and 44. To the extent this request seeks communications related to Novelis' capacity to supply pretreated aluminum, the topics of capacity and supply are addressed in separate requests, discussed immediately below. Accordingly, the Special Master recommends that Novelis produce its sales and pricing data, as set forth in the Novelis Redfern proposal, and that it be ordered by the Court, without objection.

Arconic requests certain information about the production capacity and supply of A951 treated aluminum that it contends support its theory that the Chemetall Agreement, far from being anti-competitive as contended by Novelis, in fact has pro-competitive effects. Re-treading arguments made during briefing on Novelis' motion to amend its counterclaims to add the antitrust claims, Arconic argues that the question is whether the Chemetall Agreement as a whole promotes competition, whereas Novelis argues that Arconic must defend the price restraint itself under the rule of reason. (Arconic Reply at 10-11; Novelis Letter at 11-12.)

Topic 9/Request 43, Topic 21/Request 60 and Topic 25/Request 64 are all variations on a request for “all documents” regarding Novelis' capacity to produce A951 treated aluminum. Novelis objects to each outright. At the Hearing, Arconic asserted the relevance of production capacity as follows: “[T]he question is what is the but-for world without the [Chemetall] agreement?” Because Arconic maintains that it had no “obligation to enter into an agreement with Chemetall” they were free to “enter[ ] into agreement under certain terms.... It's not anticompetitive because we entered into this agreement. We did it based on these pricing terms. And you're way better off as a result. And competition is way better off as a result,” including because Novelis has increased its capacity to produce A951 treated aluminum. (Hr'g Tr. at 111:18-112:19.)

There are several points in this argument that raise questions the potential relevance of capacity data. First, it is an undisputed fact of the case that Ford required Arconic to license its technology to Novelis, and that the Arconic-Novelis license in turn obligated Arconic to contract with a third-party chemical compounder to supply the necessary A951 chemicals. Without licensing to Novelis and Chemetall, Arconic would have no sales to Ford. Thus, Arconic's theory that it was free to deal with whomever it chose, does not fit the known facts of this case. Second, the claimed pro-competitive benefits on which Arconic's argument is based derived from the Arconic-Novelis license, and not the Chemetall Agreement; yet it is the pricing terms of the Chemetall Agreement that give rise to the Antitrust Counterclaims. Thus, Novelis argues that its capacity data has no relevance in the rule of reason analysis of the pricing terms at issue in this case.

*14 The potential relevance of the requested capacity data depends upon a complex question of antitrust law, which will be decided in the future. It need not be decided now because Novelis has agreed to a mutual exchange of high-level capacity data. (Hr'g Tr. at 116:8-11.) Though its relevance may be limited, the Special Master recommends this mutual exchange so that Arconic can pursue its theory. A mutual exchange of data by both parties is appropriate because, if Arconic is entitled to pursue its theory that the Chemetall Agreement had pro-competitive effects, including increasing production capacity of A951, Novelis should be entitled to dispute the existence or causation of any such increase. Arconic's capacity data might suggest, for example, that broader market forces, rather than the Chemetall Agreement, caused relevant changes to Novelis' capacity. If either party's expert in the future needs additional capacity data, the party may request such data only insofar as the request is tied to principles of antitrust law in the Third Circuit.

As to the portion of Topic 21/Request 60 relating to issues other than capacity—“all documents showing the design, schematics, [and] operations ... for any of your production lines”—that subject is relevant to the trade secret claims—i.e. the mechanics of the production lines implementing Arconic's A951 technology and Novelis' improvements based on that technology. The request is overbroad and cumulative in violation of the CMO. Accordingly, the Special Master recommends that Request 60 be denied at this time, without prejudice to Arconic's right to raise the issue in the future at such time as discovery on trade secret topics is renewed.

Topic 16/Request 53 seeks “[a]ll documents concerning your supply of A951-treated aluminum to Ford and/or any other entity, including all documents showing quantities and pricing of A951 aluminum.” Novelis proposed to produce sales data, as well as update prior productions, including communications with Ford and other OEMs regarding the price and selection of pretreatments. (Redfern at 28-29.) Arconic does not argue in its papers why this proposed response to Request 53 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument.

At the Hearing, the Special Master asked Arconic to identify what additional categories of relevant information it sought. Arconic identified communications related to the “quantity of production and supply and capacity of production and supply” of A951. (Hr'g Tr. at 162:10-17.)

To the extent Novelis is producing data reflecting its sales and production capacity, communications about this information are not relevant to Arconic's stated need. Either output and capacity increased or it did not, and from any such increase, Arconic can make its argument regarding the effects of the Chemetall Agreement. Requesting that Novelis produce every communication about production and capacity is overbroad and burdensome without yielding any new relevant information. Accordingly, Novelis' proposed response shall be deemed sufficient. The remaining portions of this request—sales and production capacity data—are being produced by Novelis pursuant to other requests and need not be addressed separately here.

Topic 17/Request 54 seeks “all documents” relating to any supply agreements for A951. Novelis has agreed to produce “the final Ford multiyear agreements covering platforms that use A951” and communications with other OEMs regarding the price of A951. (Redfern at 30.) Arconic does not argue in its papers why this proposed response to Request 54 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument.

At the Hearing, Arconic clarified that, beyond what Novelis has agreed to produce, Arconic seeks any agreements with OEMs other than Ford to whom Novelis supplies A951 and documents sufficient to show the “negotiation and implementation” of any such agreements. (Hr'g Tr. at 165:16-20, 166:6-11.) Novelis countered that it would be amenable to a reciprocal exchange of such agreements and to communications reflecting the negotiation of the agreements, only to the extent they relate to price and volume (and not negotiations of contractual terms irrelevant to antitrust issues). (Id. at 167:11-20.) Novelis asserts that Arconic's documents reflecting such agreements and negotiations are relevant because “the sales price and negotiation goes to the premium they are charging for A951 ... [which is] relevant to whether you have market power.” (Id. at 168:6-14.)

*15 To the extent any Novelis supply agreements are purportedly relevant to the alleged antitrust market—as asserted by Arconic (Redfern at 29)—it stands to reason that any such Arconic agreements would also be relevant. Accordingly, the Special Master recommends that Novelis be ordered to produce the documents it proposed and that, pursuant to the parties' agreement, both parties be ordered to produce (i) any agreements with any automotive OEM relating to the supply of A951 pretreated aluminum and (ii) documents reflecting the negotiation of the terms of those agreements relating to pricing and volume of the supply.

Topic 12/Request 46 seeks all Novelis license, manufacturing or sale agreements related to any technology dating back to 2010. Novelis objects to this request outright and has not proposed any production, other than noting its prior production of agreements related to A951. (Redfern at 23.) Arconic argues that this request is relevant because “agreements like the Chemetall Agreement are not uncommon in this industry, and in fact serve procompetitive purposes.” (Id. at 23-24.) At the Hearing, Arconic conceded that this request could encompass agreements related to any technology, no matter how far removed from the automotive pretreated aluminum context of this case, even including a hypothetical license regarding mousetraps. (Hr'g Tr. at 69:21-70:6.) Arconic asserted that under the common industry practice line of cases, such agreements would be relevant. See Houser v. Fox Theater Mgmt., 845 F.2d 1225 (3d Cir. 1988). There, the Third Circuit observed that the alleged anticompetitive practice—a movie theater overbooking films—was common in the industry, undermining the plaintiff's claim of willful monopolization. Id. at 1231.

Arconic was unable to articulate the indicia of a contract that would be similar to the Chemetall Agreement and thus relevant to this analysis, or proffer whether the structure of the Chemetall Agreement was common in the industry. (Hr'g Tr. at 84:25-85:11, 86:13-87:20.) Further, the request is not limited to agreements in which Novelis entered into a similar pricing mechanism, but rather seeks “[a]ll agreements to which Novelis is a party” without limitation. Arconic offers no rationale for requiring every agreement relating to licensing, manufacturing and sale relating to any technology for a nine-year period, nor any narrowing principle based on some common industry contractual arrangement that the request could be based upon. The Special Master recommends that this request be denied as overbroad. A more targeted set of agreements to be produced will be discussed below relating to Topic 19/Request 56.

Topic 13/Request 47 even more broadly seeks “[a]ll documents related to any agreement produced in response to Request 46,” including “any reason Novelis entered into any such agreement, any negotiations ..., and the pricing arrangement in any such agreement.” Arconic's explained need for this information is the same as Request 46 for the underlying agreements. (Redfern at 24.)

For the reasons stated above regarding Request 46, the Special Master recommends that this request be denied as overbroad, unduly burdensome and disproportionate to the needs of the case.

Topic 14/Request 48 seeks “documents sufficient to show” the production volume and revenue “[f]or any ... agreement ... produced in response to Request for Production 46.” Arconic's explained need for this information is the same as Request 46 for the underlying agreements. (Redfern at 25.) Arconic makes no effort to explain why it is entitled to know the production volume and revenues related to agreements and technologies that have nothing to do with A951 and the issues in this case. For the reasons stated above regarding Requests 46 and 47, the Special Master recommends that this request be denied.

*16 Topic 19/Request 56 seeks all documents related to “any Novelis agreements granting or obtaining licenses to aluminum pretreatment process” including A951 and a list of known pretreatment alternatives “and including but not limited to Chemetall U.S., Inc. and Arconic. At the Hearing, the parties agreed to a mutual exchange of license agreements related to pretreated aluminum in automotive applications on an outside counsel's eyes-only basis. (Hr'g Tr. at 91:19-92:17.) Because the parties reached an agreement at the Hearing, the Special Master need not make a recommendation concerning this request.

Topic 20/Request 57 seeks “[a]ll documents concerning any effort to research, develop, market, manufacture, and/or sell pretreated aluminum for use in automotive applications using chemicals not sold by Chemetall and/or Arconic since 2010.” Arconic states that this request is “relevant to rebut Novelis' market definition.” Novelis proposed to produce sales data and to update prior productions, including communications with Ford and other OEMs regarding pretreatment selection and various business plans and analyses regarding pretreated aluminum alternatives. (Redfern at 34-35.) Arconic does not argue in its papers why this proposed response to Request 57 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument. Further, Request, 57 goes well beyond the legally relevant inquiry—interchangeability of use—seeking any R&D, marketing or sales efforts for alternatives to Arconic's A951 technology for automotive applications, regardless of whether such efforts culminated in any actual, non-hypothetical product, let alone whether such documents relate to the interchangeability of such a product with A951.

The Special Master recommends that Request 57 be denied as overbroad and for violating the CMO. The Special Master recommends that Novelis produce the documents it proposed, as set forth in the Novelis Redfern proposal, and that it be ordered by the Court, without objection.

Topic 33/Request 73 seeks all documents relating to Novelis' “efforts to research or develop a process and/or chemical to compete with Arconic's A951 process, including without limitation any communications with Rhodia, Solvay, Henkel, PPG or any third-party chemical manufacturer or supplier, or any submission of any process and/or chemical to Ford or to any other automaker for approval or qualification of pretreating aluminum for automotive applications.”[6] Novelis offered to produce product comparisons and communications with OEMs regarding the selection of automotive pretreatments. (Redfern at 52.) Arconic does not argue in its papers why this proposed response to Request 73 is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument. Further, this request is overbroad and cumulative in violation of the CMO because it broadly seeks all documents related to R&D efforts, regardless of where in the process from conception to commercialization such a hypothetical product is, or whether it ever came to fruition. The Special Master recommends that Request 73 be denied as overbroad and for violating the CMO. The Special Master recommends that Novelis produce the documents it proposed, as set forth in the Novelis Redfern proposal, and that it be ordered by the Court, without objection.

*17 Arconic makes four document requests that are identical to requests served during prior discovery. Specifically, Topic 22/Request 61, Topic 23/Request 62, Topic 24/Request 63 and Topic 26/Request 65 were all previously served, negotiated and responded to. Ignoring this, and the CMO, Arconic simply re-served them anyway. To the extent there are areas additional discovery, Arconic had the opportunity to specifically seek it but did not do so. Novelis nonetheless proposed to make or update certain productions in response to Requests 61, 62 and 65. (Redfern at 37-40, 43.) Arconic does not argue in its papers why this proposed response to these requests is insufficient, and for the reasons discussed in Section I(c), supra, this failure constitutes a waiver to make such an argument. The Special Master recommends that Requests 61, 62, 63 and 65 be denied for violating the CMO. The Special Master recommends that Novelis produce the documents set forth in its proposals, and that it be ordered by the Court, without objection.

Topic 27/Request 66 requests “[a]ll communications between you and any third party about this lawsuit.” Arconic asserts that “[s]uch communications, if they exist, may be relevant to Novelis' claims that it has been harmed by Arconic's conduct, that competition has been harmed, and that consumers (i.e., automakers) are being harmed by Arconic's alleged ongoing scheme.” (Redfern at 44.)

The request as written is denied as entirely irrelevant.

Topic 34/Request 75 broadly seeks various analyses, forecasts, agreements and other documents concerning a not-yet consummated transaction between Novelis and another aluminum manufacturer, Aleris. According to Arconic, Aleris is “one of the other main suppliers if A951 to Ford and a supplier of alternative pretreatments such as Ti-Zr.” (Redfern at 53.) Arconic asserts that “[M]erger-related documents are likely to discuss A951, competing products, substitute products, competitors, market share, and the impact of the merger on supply or price.” (Id.) At the Hearing, Arconic added that transaction documents “are especially helpful to experts, because when Novelis is evaluating potential acquisition or merger with another producer of aluminum ... it will take a step back and look at the market [and] ... the advantage or disadvantages of combining forces with this company.” (Hr'g Tr. at 195:11-20.)

This request seeks broad discovery into a pending transaction based on the speculative statement that such materials are “likely to” contain information relevant to the Antitrust Counterclaims. The purposes Arconic articulates for this request—competing or substitute products and market share—these subjects are the focus of a broad array of analyses and business plans Novelis has agreed to produce, including those that specifically relate to market share. (See, e.g., Redfern at 11.) Arconic has not articulated a special need to specifically target merger-related documents for discussions of this subject. The potential prejudicial impact of this probe exceeds any as-yet-unarticulated reason to demand that it be produced to a business competitor of Novelis.[7]

III. Document Requests to Arconic

*18 Topic 36/Request 141 seeks all documents concerning various analyses “relating to the sale of aluminum for automotive uses, including any analyses conducted by or at Arconic's request.” The dispute arose out of whether, in addition to producing data and analyses regarding pretreated aluminum, Arconic should also produce the same materials with regard to non-pretreated aluminum (referred to as mill grade aluminum).

At the Hearing, Novelis stated that it was not seeking production of Arconic documents that relate solely to mill grade aluminum. Rather, Novelis clarified that it seeks Arconic (i) documents that discuss both pretreated and non-pretreated aluminum and (ii) high level analyses and business plans that relate to automotive aluminum more generally without specifically referencing pretreated aluminum, but implicitly include it. (Hr'g Tr. at 217:13-219:17.) Arconic agreed to produce such documents to the extent otherwise responsive to Topic 36/Request 141. (Id. at 219:20-21.) Accordingly, the Special Master need not resolve this issue.

Topic 37/Request 142 has been resolved pursuant to the mutual exchange of agreements agreed to by the parties regarding Topic 17/Request 54. (Id. at 220:14-20.)

Topic 38/Request 144 seeks “Arconic's sales data for non-pretreated aluminum for automotive uses” and specifies various criteria for this production. Arconic objects on the grounds that non-pretreated aluminum has no relevance to this case. (Redfern at 60.) At the Hearing, the parties agreed to a mutual exchange of documents sufficient to show the price charged to each pretreated aluminum OEM customer for pretreatment. (Hr'g Tr. at 232:18-23.) Because the parties reached an agreement at the Hearing, the Special Master need not make a recommendation concerning this request.

IV. Interrogatories to Novelis

Arconic Interrogatory No. 21 asks Novelis Corp. to identify “who you contend was responsible, under the terms of the Novelis License (Article 3.5.1), to identify the ‘market forces’ and determine what they ‘[would] dictate’ with respect to the price charged by any third-party chemical compounder as referenced in that provision” (citing Am. Countercl. ¶¶ 259, 265). This interrogatory is barred by the CMO because it relates to a Novelis allegation regarding its preexisting claim for breach of contract—that the Novelis-Arconic License required “market forces” to determine the price of the A951 Chemical. (See Am. Countercl. Redline, Third Claim for Relief (“Contrary to the promise in the License, Arconic, rather than market forces, is dictating the price charged by Chemetall for access to the 951 Chemicals....”).)[8] Accordingly, the Special Master recommends that Arconic Interrogatory No. 21 be denied.

Arconic Interrogatory No. 34 asks Novelis Inc. to identify “all bases for your allegation that Arconic is “wrongly attempting to interfere with Novelis' supply agreements with automakers.” The interrogatory requests Novelis' “bases” for an allegation that appeared verbatim in the original Counterclaims. (See Am. Countercl. Redline ¶ 243.) Accordingly, the Special Master recommends that Arconic Interrogatory No. 34 be denied.

V. Disputed Custodians

*19 Eight additional custodians are proffered by Arconic as having knowledge relevant to alternatives for A951. Arconic asserts that this issue is at the heart of its antitrust defense. (Arconic Reply at 8) (citing Mylan Pharm. Inc. v. Warner Chilcott Pub. Ltd., 838 F.3d 421, 436-37, 441 (3d Cir. 2016)). Novelis counters that Mylan instructs that the relevant inquiry is “substitutability of products from the ‘point of view of buyers,’ ” and thus that Novelis' internal analyses of possible substitutes are not probative of the relevant inquiry. (Novelis Reply at 4) (quoting Mylan, 838 F.3d at 437.) Novelis' internal analyses are not themselves relevant to whether a product is reasonably substitutable, unless the analysis is of a product that is in the final testing phases of pending commercialization, and has been discussed with potential buyers such that it might be considered “by buyers” as a possible alternative product. A Novelis custodian who communicated with potential automotive OEM buyers about Novelis' analysis of other products, either already commercialized or pending immediate commercialization, is a potential custodian, provided that such custodian is not cumulative of other similarly situated custodians.

Novelis further argues that alternatives to A951 was a topic of discovery on the original claims and counterclaims, and thus that adding custodians, let alone eight custodians, on this subject is cumulative and not warranted. Arconic responds that “Novelis has not stated that it has produced all responsive documents and identified those relevant documents, as required if the custodians are truly cumulative.” (Redfern at 64.) As discussed in Section I(a), supra, Arconic misstates the standard of cumulativeness, particularly in light of the Court-ordered CMO, which prohibits discovery that is cumulative of topics previously discovered, such as alternatives to A951.[9]

There are already thirteen Novelis custodians whose documents will be included in the search conducted for discovery on the antitrust counterclaims. (Novelis Letter at 18.) In analyzing the suitability of the proffered additional custodians, the Special Master will determine whether Arconic has met its burden to demonstrate that any proposed custodian possesses unique information not covered by the other agreed-to custodians. See Enslin v. Coca-Cola Co., 2016 WL 7013508, *1 n.2 (E.D. Pa. May 13, 2016) (it is not sufficient to show “these additional individuals may have some connection to the events at issue in this action;” the party advancing their inclusion must “articulate [a] basis to believe that ... they would be in possession of additional non-cumulative responsive information.”).

Topic 46/Rainer Kossak and Topic 47/Peter Redmond: Of the proposed custodians with knowledge related to A951 alternatives, Arconic identified Rainer Kossak and Peter Redmond as most central because these employees, whom Novelis describes as merely “technical,” are uniquely knowledgeable about the viability of certain A951 alternatives. (Hr'g Tr. at 240:12-241:9.) Arconic articulated the information these custodians are likely to have—that they are “fluent on technical aspects of these A951 alternatives,” and “were running lines of them and sending them ... to OEM customers. (Id. at 241:10-242:9.)

Messrs. Kossak and Redmond appear to have had substantive, and unique responsibilities related to pretreatment alternatives and communicating with OEMs regarding such alternatives. For example, emails cited by Arconic involving Mr. Kossak reflect substantial involvement in communicating with, and analyzing pretreatment alternatives for, specific OEMs, including Ford and FCA. (E.g., Arconic Letter, Ex. 38, 40, 41.) Similarly, Mr. Redmond analyzed the suitability of B2, among other pretreatments, for FCA, facilitated trials of B2, internally coordinated pretreatment development plans for Ford, and analyzed costs and benefits of pretreatment development models. (E.g., Arconic Letter, Exs. 44, 46, 48). The Special Master notes that the fact that Arconic has already received production of such emails via trade secret discovery supports Novelis' contention that substantial discovery of pretreatment alternatives was already conducted. Nonetheless, of the eight alternatives custodians proffered by Arconic, Messrs. Kossak and Redmond appear, by far, to be the most substantively involved in the issues Arconic argues it needs for antitrust discovery. Therefore, the Special Master recommends that they be added as custodians. However, by permitting these individuals to be custodians, the Special Master does not change its ruling regarding R&D efforts, if those efforts are not close to commercialization and the ability to offer them to buyers as alternatives to A951.

*20 Topic 41: Duane Bendzinski was a trade secret custodian and is now proffered by Arconic as an Antitrust Counterclaims custodian because “has worked with Ford and other OEMs to develop aluminum applications, including non-A951 pretreatments for U.S. customers, and has been a point person in Novelis' efforts to Ford-qualify an A951-alternative.” (Redfern at 65.) As Novelis notes, “two documents Arconic cites as evidence of Mr. Bendzinski's ‘unique’ information ... were transmitted to Mr. Bendzinski as attachments to calendar invitations from custodian Todd Summe for meetings at which Mr. Bendzinski was not a required attendee.” (Novelis Letter at 23-24.) Arconic also relies on an email chain reflecting that Mr. Bendzinski prepared a “competitive product map” analyzing competing products, including steel. (Arconic Letter, Ex. 13.) This email was collected during trade secret discovery, where Mr. Bendzinski was a custodian, along with his supervisor, Todd Summe, to whom Mr. Bendzinski submitted the product map. This email in fact demonstrates that at least some of the information Arconic seeks now—Novelis' internal analysis of the pretreated aluminum (and steel) product landscape—was already produced during trade secret discovery. While Mr. Bendzinski was tasked with preparing that analysis, that alone is not sufficient to establish that he possesses unique, non-cumulative knowledge because he was working at the direction of Mr. Summe, and there is no showing why Mr. Summe is unlikely to possess in his files the information most relevant to Arconic's stated need. Accordingly, the Special Master recommends that Mr. Bendzinski be denied as an additional custodian because he is cumulative to Mr. Summe.

Topic 42: Corrado Bassi is advanced by Arconic as having “led discussions on seeking price rebates for A951 from Chemetall.” (Redfern at 66.) Beyond the fact that pricing from Chemetall has been an issue in the case since Novelis' original counterclaims, Novelis asserts that the issue of Novelis' “negotiations with Chemetall on price,” is sufficiently covered by Novelis custodians Thiago Simoes, Jose Gonzalez, Derek Prichett, and Dan Bergdahl, and therefore that Mr. Bassi's inclusion is cumulative. (Novelis Letter at 24.) Further, Novelis notes that Mr. Bassi's inclusion presents an extra burden because he (and his data) are located in Europe, necessitating compliance with European data privacy laws. (Id. at 24.) Arconic also asserts that Mr. Bassi had involvement in A951 alternatives, such as having “discussed” PT3 (a potential A951 alternative) and “encouraged pursuit of a Henkel alternative to A951 that Novelis was testing,” but these claims are simply too vague to establish that Mr. Bassi had any unique, non-cumulative knowledge of these issues. Accordingly, the Special Master recommends that Mr. Bassi be denied as a custodian.

Topic 48: John Doerschuk is proffered by Arconic as being “involved” in discussions about pretreatments and testing of B2 (an A951 alternative), and “complain[ing] internally at Novelis about the lack of sufficient resources to run B2 testing and trials to address early failures and qualify it with Ford.” (Redfern at 72.) The documents cited by Arconic (Arconic Letter, Exs. 49-53) do not evidence deep—let alone unique—knowledge of the functionality of alternatives. Rather, the emails reflect rather granular involvement in testing products, such as submitting a request for a “trial coil order” (Arconic Letter, Ex. 50), securing staffing for testing (id., Ex. 52), and testing procedures (id., Ex. 53). Further, the emails in which Mr. Doerschuk appears include numerous participants, including Theresa Warrender (nee, MacFarlane) and Michael Bull, who are antitrust custodians. The Special Master recommends that Arconic has not met its burden to include Mr. Doerschuk be included as a custodian.

Topic 49: John Nickell “interfaces with internal Novelis teams, including commercial and production, to ensure fulfillment of customer technical requirements,” such as Ford's APGE standards, according to Arconic, and therefore “is likely to have unique documents on the technical requirements set by Novelis' pretreated aluminum customers.” (Redfern at 72.) Novelis describes Mr. Nickell's involvement as “wholly technical” (Novelis Letter at 25), and Arconic's evidence of Mr. Nickell's role is sparse. He appears to have been involved in a discussion of the “feasibility for Toyota” of certain pretreatment products, but the two email chains and one calendar invitation, again, show that numerous other individuals were involved in this issue, including Theresa MacFarlane, as well as Peter Redmond and Rainer Kossak, whom the Special Master has recommended be added. Accordingly, the Special Master recommends that Arconic has not met its burden to include Mr. Nickell be included as an additional custodian.

*21 Topic 50: Greg Courval was a trade secret custodian based on his knowledge of “Arconic-Novelis technical interactions,” “A951 process and/or implementation of said process,” “Alternative pretreatment process and/or implementation of said process, and Novelis and Alcoa Technology and Know-How.” (Novelis Letter at 25 (quoting Novelis Initial Disclosures).) Arconic vaguely asserts that Mr. Courval “has information about alternatives to A951 pursued by Novelis.” (Redfern at 73.) This contention is not sufficient to meet Arconic's burden of demonstrating the unique information Mr. Courval is likely to possess. Similarly, Arconic's assertion that Mr. Courval “was the point person in discussions with Chemetall about Chemetall's prior efforts to test TiZr with Ford for use in North America” (id.) is unsupported by the sole document cited by Arconic related to Chemetall (Arconic Letter, Ex. 57), in which Mr. Courval relays information he learned from someone else regarding Chemetall's work with Ford to Michael Bull (an antitrust custodian). Accordingly, the Special Master recommends that Arconic has not met its burden to include Mr. Courval be included as an additional custodian.

Topic 51: David Custers was identified in “multiple Novelis interrogatory responses as having responsibilities related to A951 commercialization under the NDA,” according to Arconic. (Redfern at 74.) Novelis counters that those interrogatory responses did not indicate that Mr. Custers' responsibilities included A951 commercialization, only negotiations of the NDA, which does not have relevance to the Antitrust Counterclaims. (Novelis Letter at 25.) According to Arconic, Mr. Custers “was involved” in developing, commercializing and testing B2, an A951 alternative and design plans for a line to produce TiZr, another alternative product. (Redfern at 74.) As with other proffered custodians, Arconic cites several emails showing that Mr. Custers had some involvement related to A951 alternatives—here, testing of B2—but those same discussions show more substantive involvement of current custodians, including Michael Bull and Todd Summe, and it does not appear that Mr. Custers was central in these efforts, was a key decisionmaker, or otherwise possesses unique information. (E.g., Arconic Letter, Exs. 59, 61.) Accordingly, the Special Master recommends that Arconic has not met its burden to include Mr. Custers be included as an additional custodian.

Freddie Hughes is Novelis' Intellectual Property Manager. Arconic asserts that “he would likely have insight into Novelis' management of intellectual property surrounding aluminum pretreatments that Novelis has offered to the market or is planning to offer to complete with A951.” (Redfern at 67.) This statement is entirely speculative and is unsupported by the two emails cited by Arconic (Arconic Letter, Exs. 23 and 24), in which the text involving Dr. Hughes is almost entirely redacted, and the remaining portions of the emails are not addressed to him. Further, “insight into Novelis' management of intellectual property” is not relevant to the Antitrust Counterclaims but instead has more bearing on the trade secret claims. Indeed, Dr. Hughes was a trade secret custodian and therefore his data has already been searched and reviewed to the extent A951 alternatives were a subject of prior discovery. The Special Master recommends that Mr. Hughes is not an appropriate custodian for antitrust counterclaim discovery; to the extent further information is requested from him in connection with the trade secret claims, that decision is deferred until resolution of the summary judgment motion on those claims.

Arconic asserts that Chris Smith “has authority over broader pre-treated aluminum sales for Novelis and therefore would have information relevant to the way that Novelis pitches A951 and other pretreatment options to automotive customers. (Redfern at 68.) At the Hearing, Arconic claimed that Mr. Smith opined that “Ford's standards were too severe and that Novelis' existing pretreatments are viable to A951 that are ready for use,” which according to Arconic, is why technical Novelis employees like Mr. Smith should be added as custodians. (Hr'g Tr. at 242:20-16.) In fact, the document referred to by Arconic (Arconic Letter, Ex. 28) was authored by a current Novelis custodian, Brad Soultz, apparently memorializing a meeting that three current custodians (Mr. Soultz, Todd Summe and Jack Presutti) attended, along with Mr. Smith. Far from demonstrating that Mr. Smith has unique knowledge, this email confirms that Mr. Smith is cumulative of other custodians, as his involvement in discussions about the development of an A951 alternative was coextensive with numerous other custodians, and he did not write the statements counsel for Arconic attributed to him at the Hearing.

*22 Arconic's other stated rationale—that “Mr. Smith [had] information related to Arconic negotiations under the NDA and thus A951 pricing negotiations”—also fails. (Redfern at 68 (emphasis added).) The NDA—executed between Arconic and Novelis prior to the parties' licensing agreement—is separate from the Chemetall Agreement; thus Mr. Smith's purported involvement in NDA negotiations has no relevance to A951 pricing. Therefore, the Special Master recommends that he not be added as an additional custodian at this time.

At the Hearing, Arconic identified Pierre Labat first when asked to state the additional custodians it considers most central. (Hr'g Tr. at 233:3-10.) While not dispositive, this claim is somewhat undermined by the fact that Arconic failed to identify him during the negotiation process, only adding him after the deadline for the parties to declare impasse on discovery negotiations. (Novelis Letter at 27.)

As for the documents relied on by Arconic, they fail to establish that Mr. Labat played a unique role in the topics contended by Arconic. For example, Arconic claims Mr. Labat “directed Novelis' efforts to pitch A951 pretreatment alternatives to other OEMs like Jaguar,” (Refern at 69), but the cited email chain shows current custodian, Todd Summe, initiating the contact with Jaguar. (Arconic Letter, Ex. 35.) Another email notes that Mr. Labat is Tom Boney's “counterpart in Europe.” (Arconic Letter, Ex. 31.) Mr. Boney is an existing custodian, and there is no showing of what unique involvement Mr. Labat had in the issues relevant to the Antitrust Counterclaims. Arconic further notes that Mr. Labat “was on all these e-mails where they are talking about meeting with Henkel” (Hr'g Tr. at 237:17-19)—with whom Novelis was working to develop a pretreatment alternative—but the emails cited by Arconic (i) do not involve discussions relating to the development or functionality of such alternative products and (ii) do not demonstrate that Mr. Labat played an active role in discussions; rather, current custodian, Mr. Summe wrote most of the substance of these email chains, with Mr. Labat a silent, or minimal, participant. Arconic has not met its burden to show Mr. Labat possesses unique knowledge based on this evidence.

Finally, the public statements by Mr. Labat referenced by Arconic (that “be believed demand for automotive aluminum is set to more than double by 2024” and that “Novelis will continue to add new products in the years to come” (Redfern at 69)) are vague, not tied to antitrust issues (including A951 and its alternatives), and not evidence of deep involvement in the development of such alternatives. Therefore, the Special Master recommends that he not be added as a custodian.

Frederick Tenney is an agreed-to Arconic custodian, but the parties dispute the scope of the production Arconic should make of his files. Specifically, Mr. Tenney is an in-house lawyer, who, according to Novelis, was a “lead negotiator with Chemetall regarding” the Chemetall Agreement, which is central to Novelis' Antitrust Counterclaims. (Redfern at 75.) The parties apparently agreed previously that Arconic's production of his files would be limited to “communications with Chemetall,” and not internal Arconic communications. (Id.) Novelis now seeks a fuller production of Mr. Tenney's files, non-privileged “documents and communications related to A951 and Chemetall” regardless of who such communications are with. (Id.) Arconic objects because of the burden of reviewing a lawyer's emails, given the high volume of privilege determinations this review will require. (Id.)

*23 The Special Master recommends that Arconic be required to produce Mr. Tenney's non-privileged documents and communications related to A951 and Chemetall, whether those communications are with Chemetall, internal to Arconic, or with any other party. Novelis has agreed to narrow the scope of Mr. Tenney's communications that would be relevant—only A951 and Chemetall—and Arconic's role in bringing about the pricing terms in the Chemetall Agreement is highly relevant to this dispute. Though the privilege review of a lawyer's files constitutes some burden, in the context of this matter, the breadth of discovery Arconic has sought, and the narrow category of responsive material Novelis seeks, the Special Master finds that one lawyer custodian does not pose an unjustified burden. The privilege review and privilege log will be completed according to the agreed protocol. (See Dkt. 358-1.)

VI. Disputed Search Terms

The parties have been unable to agree to appropriate search terms for use in collecting data for discovery on the Antitrust Counterclaims. Below, the Special Master recommends search terms that address Arconic's stated needs and Novelis' objections. Should Novelis determine that any term recommended by the Special Master produce an overly burdensome yield when run, Novelis shall bring evidence of the search results to Arconic and a proposal for how to further narrow the term.

Before addressing the individual disputed search terms, the Special Master will address several issues that affect multiple terms.

An issue that cuts across many of the search strings is whether it is appropriate to include variations on aluminum (through the search term “alum*”) as a search term. Many of the strings proposed by Arconic would search for this term connected with various other phrases and terms. (See, e.g., Topic 54 (Redfern at 76-77), search string (“MSC” or “Material Science* Corp*” OR matsci or materialsciencorp) AND (pretreated OR alum* OR Arconic OR Alcoa OR 951 OR A951 OR 951).) Arconic's stated need for the term alum* is that searching for only known alternatives may fail to capture an unknown alternative, which would have relevance to Arconic's challenge of Novelis' alleged relevant market.

However, “alum*” is an overly-broad search term to use to search for alternative pretreatments. Novelis is an aluminum company. It manufactures a wide variety of aluminum products, for a range of sectors. It is unreasonable to ask an aluminum manufacturer to search for all discussions of aluminum in a case relating to a specific sub-set of that company's operations. Arconic's stated rationale makes it clear that it does not need all discussions of aluminum, but rather, discussions of pretreated aluminum. (See, e.g., Redfern at 77 (“Arconic's proposed search seeks to capture document referring ... [to] any alternative pretreatments to A951.”); id. at 82 (“searching the documents of Novelis custodians known to interact with Ford in connection with aluminum pretreatment alternatives should produce reasonably targeted results.”); id. at 88 (“the purpose of this search [is] to gather documents relating to Novelis' communications about pretreated aluminum”) (emphasis added in all).)

Novelis' offer to run all known alternatives to A951 is a partial solution to Arconic's need for this information. The Special Master notes that, while Arconic's proposed search strings included “alum*” and “pretreat*” in connection with other terms (e.g., that only documents hitting on “alum*” and Chemetall would be reviewed), Novelis offers to run the known alternatives as standalone terms (i.e. that a document containing TiZr would be gathered regardless of whether it contained another keyword).

To account for unknown alternatives, the more generic term “pretreat*” is also available. According to Novelis' proposed revisions to Arconic's search strings, Novelis is willing to run these terms as-is, so long as the variation including “alum*” is excluded. Thus, Novelis is willing to search for “pretreat*,” which in the opinion of the Special Master, is reasonably tailored to capture the results Arconic states it needs—documents about pretreated aluminum. Arconic's statement that the term “pretreat*” is “unlikely to capture all such discussions” (see, e.g., Redfern at 79) is too vague and unsupported to justify the broad and burdensome term for which it advocates.

*24 Accordingly, for all search strings containing “alum*” proposed by Arconic, that portion of the string shall be denied. Any other modifications needed to those terms will be addressed separately below. There was additionally some discussion in the submissions regarding searching for the abbreviation of aluminum, “AL,” and possible technical difficulties with searching for that text (e.g., that it could capture words containing the letter “al”). Because the reasoning above also applies to the term “AL” (which is potentially even more problematic for its higher likelihood of false hits), that term is also excluded.

The parties have agreed to run variations of 951 as standalone search terms. (See May 3, 2019 Email from Mark Ford (listing 951 or A951 or “Alcoa 951” or R951 or “Arconic 951”) as agreed-to search terms). Thus, any document that contains any such variation on 951 will be retrieved in a document collection. Many of the disputed search strings contain these same 951 variations as connectors, meaning terms that must appear in combination with other terms to be collected. For example, Topic 55 is for the disputed search term Chemetall AND (pretreat* OR alum* OR Arconic OR Alcoa OR 951 OR A951 OR R951 OR “4595”). Thus, a document containing Chemetall and 951 would be collected using this search string. However, a document containing 951 alone would be collected in the first instance by using that agreed-to standalone term. Thus, using 951 (and the other agreed-to variations) as connectors adds nothing; no new documents will be gathered. For simplicity and clarity, the search strings set forth in this Report & Recommendation will have such redundant connector terms, which separately appear as standalone terms, removed.

The parties appear to be in agreement as to the list of known A951 alternatives. Between Novelis' proposed revised term (Redfern at 84-85), and the parties' agreed-to terms (May 3, 2019 email from Mark Ford), the parties agree to the following terms: B2 OR B3 OR BONDERITE OR “M-NT 8453” OR PT2 OR PT3 OR TiZr OR “Titanium Zirconium” OR Ti-Zr OR Oxsilan OR Oxisilane OR Oxisilan OR Gardobond OR xb4595 OR “xb 4595” OR Henkel OR “P552” OR F150 OR “F-150” OR “P-552.” Each of these known alternatives will be run as a standalone term. Therefore, they will not be repeated as a connector in any of the other search terms discussed below.

Topic 54 is a disputed search term related to Material Science Corporation (“MSC”), which, according to Arconic, “licenses A951 from Chemetall to toll aluminum for both Arconic and Novelis.” (Redfern at 76-78.) The crux of the parties' dispute is whether the term should be limited to variations on MSC and “951,” or whether the term should be broader to encompass alternatives to A951.

The Special Master recommends that this search string be: ((“MSC” or “Material Science* Corp*” OR matsci or materialsciencorp) AND (pretreat* OR Arconic OR Alcoa)). Arconic's stated need for this search string is for “documents referring both to MSC and Alcoa or to MSC and any alternative pretreatments to A951.” (Redfern at 77.) This revised search string accomplishes this goal, along with the agreed-to standalone terms for A951 alternatives.

The only dispute regarding Topic 55 is whether alum* should be a possible connector term. For the reasons set forth in Section V(a), supra, alum* is overbroad. The Special Master recommends that this term be run as follows: Chemetall AND (pretreat* OR Arconic OR Alcoa OR “4595”).

*25 Arconic's proposed search that is the subject of Topic 56 is aimed at capturing “Ford, DDMP and Metokote's consideration, testing, purchase and sale of alternatives to A951.” (Redfern at 81.) DDMP and Metokoke “license A951 from Chemetall”. (Id.) The proposed search string contains highly generic terms as connectors: perform, test, purchase, sale and request. Thus, for example, any document containing both the terms “Ford” and “sale” would be captured. This is over-broad and not tailored to capture what Arconic's stated need is—documents related to “alternatives to A951.” (Id.)

The Special Master recommends that this search string be: (Ford OR DDMP OR Metokot*) AND (pretreat* OR Arconic OR Alcoa). This coupled with the known alternatives as standalone terms will capture documents related to Arconic's stated need.

Topic 57 is the following disputed search term: (Alternativ* OR substit* or gener* or interchang*) w/20 (Arconic OR Alcoa OR 951 OR A951 OR R951 OR alum* OR pretreat*). Arconic again justifies this term based on its need to discovery documents related to alternatives for A951. (Redfern at 83.) The Special Master recommends that this term be run as: (Alternativ* OR substit* OR gener* OR interchang*) w/20 (Arconic OR Alcoa OR alum* OR pretreat*). Because Arconic requested these terms within 20 words of each other, using alum* here appears to pose less of a risk of overbreadth.

The disputed term of Topic 58 involves known alternatives to A951, as well as chemical companies—PPG, Solvay and Rhodia—which, according to Arconic, “are known to be associated with these chemicals.” (Redfern at 84.) The dispute is focused on the chemical companies. Novelis contends that it does not purchase pretreatments from those companies, and thus that there is a high risk of false positives from using those terms, “especially when run against the ESI of four employees with procurement and sourcing roles.” (Id.) Arconic's assertion that the companies “are known to be associated with these chemicals” is not sufficiently specific to link those terms to pretreatment alternatives and to Novelis to require that the terms be run outright. To the extent Novelis communicated with any of these entities about a known alternative, those communications would be gathered by using the known alternative search terms. Running PPG, Solvay and Rhodia as standalone terms is not justified because Novelis likely had communications with them about a variety of subjects.

Accordingly, the Special Master recommends that Novelis run the term (PPG OR Solvay OR Rhodia) w/20 (pretreat* OR “treat* alum*”).

Topic 59 involves a disputed search string aimed at “discussions of pretreated aluminum, including A951 pretreated aluminum and alternatives, relating to automakers other than Ford.” (Redfern at 85.) In addition to listing OEMs, it also includes generic intellectual property terms such as patent, royalty and trade secret. Arconic states that the search “should also capture discussions regarding whether A951 or alternative aluminum pretreatments are protected by intellectual property.” (Redfern at 85.) Arconic does not explain how such documents—if they exist—would have any relevance to the antitrust counterclaims. The search also includes even broader terms like confidential, proprietary and fee, which are likely to yield tremendous volume without any stated need.

The Special Master recommends that this term need not be run. Documents related to alternatives to A951—Arconic's stated need—are sufficiently captured by the search terms for those alternatives, which will be gathered regardless of whether they are internal communications, or communications with Ford or other OEMs.

*26 Topic 60 relates to a search for documents regarding “the automotive market share of A951 and its alternatives and the automotive market share of Arconic and its competitors (including Novelis, Constellium, and Aleris).” (Redfern at 87.) To this end, Arconic seeks the search term (Market w/3 share) AND auto* AND (pretreat* OR alum* OR Arconic OR Alcoa OR 951 OR A951 OR R951 OR Novelis OR Constel* OR Aleris).

Novelis cannot be required to search its custodians' files for the term Novelis. This potentially could collect every single email in a custodian's inbox (e.g., Novelis will appear in their email address, signature, and many if not most of their correspondence and attachments).

Novelis proposes the following term: (Market w/3 share) AND auto* AND pretreat*. This term appears sufficient to capture Arconic's focus—document that discuss market share of pretreated aluminum. To the extent Novelis', Constellium's or Aleris' market share are discussed, this term would capture those documents.

Topic 61 relates to a search for “Novelis' communications about pretreated aluminum forecasts.” (Redfern at 88-89.) Novelis states that the parties “agreed to produce forecasts based on targeted collections rather than broad ESI searches, making the term redundant.” (Redfern at 89.) As with the high-level capacity data both parties are producing, Arconic may use the forecast data the parties have agreed to produce to “rebut arguments that Novelis is being harmed by Arconic's allegedly anticompetitive behavior.” (Redfern at 89.) Arconic has not established that it is entitled to communications about that data in order to make such arguments. Accordingly, the Special Master recommends that this term need not be run.

Topic 62 involves a term Arconic states is related to Novelis' allegation “that Arconic has manipulated the price and cost of pretreated aluminum through the technology access fee in the Chemetall Agreement.” (Redfern at 90.)

The Special Master recommends that Novelis run the term (value* OR pric* OR apportion* OR cost) w/20 (Arconic OR Alcoa OR pretreat* OR “treated alum*” OR (technology w/10 fee) OR patent). By adding a limitation of “within 20 words” this will narrow the results while still capturing the discussions Arconic states that it needs.

Topic 63 is the search term Chemetall w/10 (cost* OR pric*). Arconic contends the term is necessary because “the prices to Novelis and Arconic, and the costs of A951, are central issues in Novelis' antitrust counterclaims.” (Redfern at 91.) The Special Master recommends that Novelis should run the term to assess the yield, which should be limited somewhat by the “within 10 words” limitation. Should the yield of this test search be unreasonably large, Novelis shall bring evidence of the search results to Arconic and a proposal for how to further narrow the term.

Topic 64 is a disputed term, which Arconic asserts is relevant to requests seeking information about substitutes for A951, Novelis' purchase of a generic or substitute for A951 and efforts to find a competing product. (Redfern at 92.) The Special Master recommends that the string be modified to be: (Entr* or compet* or rival* OR market*) w/20 (pretreat* OR Arconic OR Alcoa). Though the terms entr*, compet* and market* are broad, limiting it with a “within 20 words” requirement will limit the scope of the search.

*27 Topic 65 is (Suppl* OR buyer* or customer* or player* or purchaser*) w/20 (pretreat* OR alum* OR Arconic OR Alcoa OR 951 OR A951 OR R951). Arconic states that it relates to “actual and potential market participants for pretreated aluminum, including A951.” (Redfern at 93.) Novelis objects that the term contains broad terms likely to yield false positives. (Id.) The Special Master recommends that this term need not be run. The generic terms suggested by Arconic are not especially likely to gather the documents it states it needs for this request. Indeed, the document requests for which Arconic states this term is needed all relate to alternatives to A951, which are being run as standalone terms, so this term is not needed. It is therefore denied.

Topic 66 is a search string that Arconic contends will “refute Novelis' allegations regarding the Chemetall Agreement's anticompetitive effects, by showing that the agreement in fact increased output” of A951 pretreated aluminum. (Redfern at 95.) The term proposed by Arconic is (Capacit* OR product* OR volume*) w/25 (pretreat* OR alum* OR Arconic OR Alcoa OR 951 OR A951 OR R951).

Because 951 and variations thereof are being run as standalone terms, the Special Master recommends that this term need not be run. Arconic states that the production of “A951 pretreated aluminum” is what it seeks to know and does not justify why this additional term is needed or what additional it would capture. Nor does the search string seem reasonably tailored to reach the desired results. For example, a document with “product” within 25 words of “Alcoa” does not seem especially likely to be relevant to Arconic's stated purposes. As discussed above, the parties have agreed to a mutual-exchange of capacity data, which addresses Arconic's stated need.

Topic 67 involves the search string (revenue* or sale* or profit*) w/10 (pretreat* OR alum* OR Arconic OR Alcoa OR 951 OR A951 OR R951). Arconic asserts it is “targeted to communications, analyses, or the like on the revenues, sale, or profits for pretreated aluminum or A951 treated aluminum, information that is relevant to whether substitute products constrain the pricing of A951 treated aluminum and Novelis' analysis of competing products.” (Redfern at 96.)

Because A951 and the known alternatives are being run as standalone terms, any communication that relates to sales or profits associated with those products would be captured. Additionally, Novelis will be producing pricing and sales data. The Special Master does not find that this term is either necessary or productive given the other searches and productions being conducted. It is therefore denied.

Topics 68 and 69 are two search strings proposed by Arconic relating to Novelis' ongoing acquisition of Aleris. Because the Special Master has denied the document request related to this transaction, search terms for this subject is not warranted.

Topic 70 relates to the Arizona Proving Ground Exposure (“APGE”) standards Ford established to determine candidates for the F-150 truck. (Redfern at 100.) Arconic states that the enlarged search string is relevant to document requests, including concerning A951 substitutes, as well as proposals for Ford's F-150 program and the P-552 program.

This disputed term is duplicative of agreed-to terms: the known A951 alternatives are being run as standalone terms, along with variations on pretreat*; and F-150, F150, P-552 and P552 are all standalone terms. The Special Master recommends that this term need not be run.

VII. Deadline for Completion of Fact Discovery

The CMO previously set the close of fact discovery for June 21, 2019. The Special Master set this date with the aim of moving forward productively with discovery on the antitrust counterclaims pending the resolution of the summary judgment decision on Arconic's trade secret claims. However, given the enormous quantity of the parties' disputes regarding the discovery served, that deadline could not be met. The Special Master cautions the parties that there shall never be a repetition of the non-productive meet-and-confer sessions that marked this dispute, and prolonged the work of the Special Master to an unprecedented extent, even for this hotly contested case.

*28 Novelis asserts that the new deadline should be extended by two and a half months. Arconic prefers five months for the completion of fact discovery. The Special Master recommends that the deadline for completion of fact discovery be extended to October 31, 2019. This time period will allow the parties ample time to conduct discovery consistent with this Report & Recommendation and to serve any needed follow-up requests, consistent with the directives of the CMO. The Special Master expects the parties to proceed diligently to avoid any unnecessary delays that would warrant a further extension of this deadline. The work necessary to comply with this Report & Recommendation shall commence forthwith.

For the foregoing reasons, the Special Master recommends that the Court order that:

A proposed Form of Order will be filed by the Special Master as Report & Recommendation #30.