Searching over 5,500,000 cases.

Buy This Entire Record For $7.95

Official citation and/or docket number and footnotes (if any) for this case available with purchase.

Learn more about what you receive with purchase of this case.


United States District Court, S.D. New York

November 14, 2005.


The opinion of the court was delivered by: ANDREW PECK, Magistrate Judge


Plaintiffs have moved to compel the production of documents from non-parties McGraw-Hill Companies and Intelligence Press, Inc. (collectively, "the Publications") pursuant to Fed.R.Civ.P. 45. (Dkt. No. 310: Pls. Notice of Motion.) For the reasons set forth below, plaintiffs' motion is GRANTED in part and DENIED in part.


  The Underlying Action

  The facts pertaining to the underlying action have been set forth in Judge Marrero's opinions denying defendants' motion to dismiss and granting plaintiffs' motion to certify the class. In re Natural Gas Commodity Litig. ("Natural Gas I"), 337 F. Supp. 2d 498 (S.D.N.Y. 2004); In re Natural Gas Commodities Litig. ("Natural Gas II"), 03 Civ. 6186, ___ F.R.D. ___, 2005 WL 2414449 (S.D.N.Y. Sept. 30, 2005). Only the facts necessary to this motion are discussed below.

  Plaintiffs allege that the defendants manipulated the prices of natural gas futures and option contracts by falsely reporting trade data — price and volume — to the Publications. (See, e.g., Dkt. No. 98: Consolidated Class Action Complaint ("CCAC") ¶¶ 59-62, 66; Dkt. No. 311: Pls. Br. at 10.) See also Natural Gas II, 2005 WL 2414449 at *1; Natural Gas I, 337 F. Supp. 2d at 501-02. This false information, allegedly, led to erroneous price indices reported by the Publications and relied upon by traders, thereby skewing the futures market. (See, e.g., CCAC ¶¶ 67-72; Pls. Br. at 10-11.) See also Natural Gas II, 2005 WL 2414449 at *1; Natural Gas I, 337 F. Supp. 2d at 503. The false trades were purportedly made between January 1, 2000 and December 31, 2002 (The "Class Period"). Natural Gas I, 337 F. Supp. 2d at 501. The NYMEX (N.Y. Mercantile Exchange) futures contracts are for delivery at the Henry hub only. Id. at 502. (See also Dkt. No. 335: Kovner Aff. Ex. L: Harris Class Cert. Aff. ¶ 8; 11/9/05 Oral Arg. Tr. at 33-34, 80.)

  The Publications


  Platts is a division of McGraw-Hill and is "the world's largest and most reliable source of news and information about the energy industry," publishing over 50 newsletters and magazines on the energy industry. (Dkt. No. 217: 1st Foster Aff. ¶ 2.) Two particular Platts publications, Inside FERC's Gas Market Report ("Inside FERC"), with indices published monthly, and Gas Daily, published daily, cover the natural gas industry. (Id. ¶ 3; Dkt. No. 333: 3d Foster Aff. ¶ 3.) They both publish price indices covering "numerous pricing points across the United States" using information obtained "by Platts reporters from actual buyers and sellers in the marketplace." (1st Foster Aff. ¶ 4.) Platts does not rely on other publicly available price surveys, but rather conducts its assessments and prepares its indices solely on the information gained through original reporting. (Id.) A sample Gas Daily is Exhibit B to the 1st Foster Aff.; a sample Inside FERC is Exhibit A to the 1st Foster Aff. During 2000-2002, Inside FERC published a price range for "approximately 70 monthly pricing points," adding up to thousands of transactions considered per month; Gas Daily used approximately 102-116 pricing points. (3d Foster Aff. ¶ 4; see 1st Foster Aff. ¶ 5.) During the Class Period, most energy companies submitted monthly price reports to Platts via email, although some sent reports via fax or telephone. (3d Foster Aff. ¶ 6.) For Inside FERC, all trade data reported by the energy companies were made during "bid week" — a five-day period at the end of each month. (Id.) Those trades reported by the companies "provided transaction-specific data, including price, volume, hub, and, in some cases, counter-party." (Id.) Contrariwise, for the daily trade reports to Gas Daily, the energy companies were not required to submit transaction-specific data, but rather were only "asked for: a low price, a high price, a weighted average and a total volume" for the daily trades. (3d Foster Aff. ¶ 7.) According to Platts, "Gas Daily would not have the specific transaction data that would enable plaintiffs to identify this as an example of a false trade report by the [energy] company." (Id.)

  Platts' reporters and editors analyzed the data, using "their experience, judgment and analytical tools to identify and question outlying data" which, depending on Platts' judgment after confirmatory calls to the source, may be excluded from the aggregate price index. (1st Foster Aff. ¶ 6.) The actual underlying data is not published, but rather serves as the basis for the published indices (id. ¶ 4) which are an amalgamation of quantitative analysis and editorial judgment (id. ¶ 6; see also 3d Foster Aff. ¶ 13).*fn1 Consequently, according to Platts, "the transaction data used to calculate any given index does not necessarily match the data reported by the energy company traders, and there is no one document or set of documents that reveal, on their face, `the data used' in the creation of the indices." (3d Foster Aff. ¶ 14.)

  Before Platts acquired Gas Daily in September 2001, "the publication routinely discarded the information submitted by the energy company traders"; once Platts took over, it told Gas Daily reporters of its preference for retaining the submissions, but at least until "well into 2002" reporters continued to discard the daily trade submissions. (3d Foster Aff. ¶ 8.) Gas Daily has "accessible electronic records of the type sought only from November 4, 2002 forward." (3d Foster Aff. ¶ 9; see also 2d Foster Aff. ¶ 8.)

  Platts' sources, the traders, provided their information on the understanding that their identity would be kept confidential and the price information would not be attributed to a particular company or trader. (1st Foster Aff. ¶ 7.) According to Platts, confidentiality is a crucial means through which Platts maintains its reliability as a transparent and comprehensive news source. (1st Foster Aff. ¶¶ 7-8; 2d Foster Aff. ¶ 2; 3d Foster Aff. ¶ 10.) "Both Inside FERC and Gas Daily had (and still have) many sources for their indices who are not party to the instant litigation." (3d Foster Aff. ¶ 5.) Also according to Platts, a number of natural gas sources stopped providing their trading information to Platts out of fear for a loss of confidentiality, after Platts received subpoenas seeking trade data. (3d Foster Aff. ¶ 10.)

  Platts has "spent countless hours" responding to subpoenas it has received from private litigants and government agencies. (3d Foster Aff. ¶ 11; see also 2d Foster Aff. ¶ 4; 1st Foster Aff. ¶¶ 10-11; Dkt. No. 335: Kovner Aff. ¶ 4.) "To date, Platts personnel, including [the Global Editorial Director for Platts, Larry Foster] and . . . the Chief Editor of Inside FERC, have expended more than 1,000 hours in gathering and reviewing documents called for by these many subpoenas." (3d Foster Aff. ¶ 11; see 11/9/05 Oral Arg. Tr. a 50.)

  Intelligence Press

  Intelligence Press is a "small news publishing company," with a "total staff of only thirteen full-time employees," that specializes in reporting on the natural gas industry. (Dkt. No. 218: Steis Aff. ¶¶ 2, 9; Dkt. No. 316: Steis Supp. Aff. ¶ 7.) Intelligence Press "conducts confidential surveys of natural gas prices paid in the spot wholesale market" and, based on the information it receives from individual traders, publishes aggregate price indices. (Steis Aff. ¶ 2.) It looks at the high and low trades and the averages of all trades reported at approximately 90 to 100 trading points nationwide. (Id.) The indices are published daily and then aggregated for weekly publication. (Id.) "Monthly surveys of 30-day transactions are conducted during the last five business days of the month for the ensuing month and are published [by Intelligence Press] as `Bidweek' surveys." (Id.)

  Intelligence Press' affidavits state that the published indices encompass not only numerical data but editorial decisions such as which transactions should be assigned to which pricing points and which outlying transactions are unreliable and consequently excluded. (Steis Aff. ¶ 10; see also Steis Supp. Aff. ¶ 8.) Nevertheless, Intelligence Press' published methodology states that it uses a specific formula (using the "volumetric weighted average method") to calculate published prices. (Dkt. No. 315: Siegel Aff. Ex. 1 ("Price Methodology") at 3-4 & Ex. 2 ("Price Survey Methodology") at 4.) Intelligence Press' published methodology does not indicate that any editorial process is involved in producing the indices other than the exclusion of "outliers" (typically prices that "fall more than two . . . standard deviations outside the sample mean") from the price calculations. (Siegel Aff. Ex. 1 at 3 & Ex. 2 at 4.)

  Intelligence Press' published price indices "are widely used in the industry in contracts that set the prices to be paid for natural gas." (Steis Aff. ¶ 2.) According to Intelligence Press, in order to obtain the "proprietary information" of trade prices from energy companies, Intelligence Press pledges confidentiality — of the source and the information — which is the key to building the source base it has built over the years. (Steis Aff. ¶ 5.) Moreover, Intelligence Press asserts that its spreadsheets are confidential and the information therein proprietary. (Steis Supp. Aff. ¶ 11.) Intelligence Press' published methodology "pledges" confidentiality to its sources: "Intelligence Press will not reveal the source of any price information, nor will we reveal the parties involved in any transaction to any outside organization. . . . Anonymity of the survey participants will always be preserved." (Siegel Aff. Ex. 1 at 1 & Ex. 2 at 2.)

  According to Intelligence Press, "there are potentially more than 800 spreadsheets that would be responsive" to plaintiffs' current subpoena. (Steis Supp. Aff. ¶ 3.) Intelligence Press did not regularly preserve its spreadsheets. (Id. ¶ 5.) Intelligence Press believes that they "would have to search for old spreadsheets . . . stored on multiple and antiquated individual computers used by individual Intelligence Press employees at the time." (Id.; see also Steis Aff. ¶ 8.) For the daily published indices, Intelligence Press' general practice was to create a "temp" file to make all editorial changes and calculations. (Steis Supp. Aff. ¶ 9.) This file would be separate from the daily spreadsheets filled with the trade data. (Id.; see also 11/9/05 Oral Arg. Tr. at 56.) Each day's "temp" file overrode that of the previous day. (Steis Supp. Aff. ¶ 9.) For the monthly indices, the editors generally saved "multiple drafts" as they went through the editorial process but did not make a final version reflecting the published indices and sometimes may not have even documented their editorial changes. (Id. ¶¶ 8, 10.)

  In order to comply with the subpoena, Intelligence Press anticipates that it would "likely take many hundreds of hours or more" to locate the old spreadsheets, and even then, the search would not likely locate more than half of the spreadsheets used. (Id. ¶ 5.) Redacting the names of non-defendants alone could take one hour per spreadsheet. (Id. ¶ 6.) Intelligence Press believes that even with a massive search, it would only locate "perhaps about 5% of the total [spreadsheets] Plaintiffs seek." (Id.)

  The First Round of Subpoenas

  Plaintiffs served their original subpoenas duces tecum on the Publications on November 8, 2004 (Dkt. No. 230: Persky Aff. ¶ 6 & Ex. B; Dkt. No. 335: Kovner Aff. Ex. I) and revised subpoenas on December 7, 2004 (Persky Aff. ¶ 7 & Ex. C).*fn2 On December 8, 2004, not to be left out, certain defendants served a subpoena duces tecum on McGraw-Hill (but not Intelligence Press). (See Kovner Aff. ¶ 19 & Ex. J; see also 11/9/05 Oral Arg. Tr. at 58.) Defendants' subpoena was similar to plaintiffs' subpoenas in terms of the information sought, but broader in scope. (See Kovner Aff. ¶ 19 & Ex. J at 4: "Documents to be Produced"; see also 11/9/05 Oral Arg. Tr. at 69-72.)

  On November 22, 2004, the Publications filed a motion to quash plaintiffs' original subpoenas, asserting that: the documents and information sought were protected by the qualified journalist's privilege (Dkt. No. 215: Platts Br. at 7-17; Dkt. No. 219: Intelligence Press Br.); the subpoenas were unduly burdensome pursuant to Fed.R.Civ.P. Rule 45(c)(3)(A) (Platts Br. at 17-19; see Intelligence Press Br.); and plaintiffs could not show substantial need warranting revelation of confidential information and trade secrets under Rule 45(c)(3)(B) (Platts Br. at 19; see Intelligence Press Br.). On January 7, 2005, the Publications filed a motion to quash Defendants' subpoena, asserting similar grounds as they had in their motion to quash plaintiffs' subpoena. (Dkt. No. 238: Platts Br.) This Court's Prior Decision to Quash the 2004 Subpoenas Without Prejudice

  On January 14, 2005, after hearing oral argument on the motions, this Court granted the Publications' motions to quash plaintiffs' and defendants' subpoenas, without prejudice to the parties later serving new subpoenas. (Dkt. No. 312: Briskin Aff. Ex. 15: 1/14/05 Peck Conf. Tr. at 4-10, 13-17.) This Court ruled that in light of the qualified reporter's privilege, plaintiffs and defendants had to seek the information from "other available sources," including, at a minimum, seeking to obtain information "through discovery from the defendants and nonparties" that would serve to narrow any possible future production requests to the Publications. (Id. at 4-5, 14, 16-17.) The Court opined that:

[B]ecause of the qualified press privilege it seems to [this Court] that if you get the false trade dates and locations th[e]n the amount of the information that you are going to need from the publications will be considerably less and may obviate their confidential source problems as well.
(Id. at 5.)

  Plaintiffs' Subsequent Discovery From Defendants

  Plaintiffs' discovery of defendants has revealed that none of the defendants have a complete set of "reported trades" (i.e., what their traders reported to the Publications, as opposed to actual transactions) for the Class Period. (See, e.g., 11/9/05 Oral Arg. Tr. at 3-5; Dkt. No. 311: Pls. Br. at 2, 15-21; Dkt. No. 312: Briskin Aff. Exs. 5, 8, 9, 12, 13, 19, 20.)*fn3 Indeed, at oral argument, defense counsel conceded that the defendants did not have anywhere near a complete set of reported trade information. (11/9/05 Oral Arg. Tr. at 4-5; see also id. at 35: the Publications' counsel conceded that the Publications "accept that it is not complete.")

  Plaintiffs' Expert Michael Harris

  Plaintiffs' expert, Dr. Michael Harris, a senior economist with EconOne Research Inc. in Los Angeles, has explained that, in order to compile a damage analysis, he will create a set of "but-for" indices comprised of price indices "that would have been reported by the trade publications but-for the actions of the Defendants." (Dkt. No. 312: Briskin Aff. Ex. 2: Harris Aff. ¶¶ 1, 4.) As Dr. Harris explained, "[t]he most straightforward way to derive the but-for indices is to use the data relied on by the trade publications and `recalculate' or `reconstruct' the index by purging the false data submitted by the Defendants." (Harris Aff. ¶ 4.) This analysis, according to Dr. Harris, requires both the information produced by defendants and the data from the trade publications. (Id.) While plaintiffs may be able to obtain from the defendants some (but not all — see n. 3 above) of the information supplied to the Publications, only the publications know which trades were actually used — after elimination of outliers and anomalies and editorial decisions — to comprise the reported index. (Harris Aff. ¶ 5.) It is this reported trade data actually used by the Publications that plaintiffs seek. (Id. ¶¶ 5-6.)

  In terms of actually reconstructing the indices, Dr. Harris opines that the Publications' production of the spreadsheets containing the reported trades data would be useful because the archival dates attached to those spreadsheets would be able to be matched up to the date an index was published. (Dkt. No. 319: Briskin Reply Aff. Ex. 23: Harris Reply Aff. ¶¶ 4-5.) The spreadsheets "contain the underlying data from which the indices were constructed." (Harris Reply Aff. ¶ 4.) Thus, the spreadsheets can be used to identify the defendants' missing trade reports. (Id. ¶ 6.) According to Dr. Harris, Intelligence Press' concern that editorial changes were not reflected in the underlying data, "unless `editorial changes' involved arbitrarily adding data that does not exist, the use of temp files is not a problem" since Intelligence Press' published methodologies refer to editorial decisions only in the context of outliers, which could be pinpointed given the underlying data. (Id. ¶ 4.) Additionally, Dr. Harris reiterates that defendants have not produced a complete set of reported trade reports, either because no paper record was made or no paper record was preserved. (Harris Aff. ¶ 7; see also 11/9/05 Oral Arg. Tr. at 4-5, where defendants' counsel conceded as much.) Thus, obtaining the reported trade data used by the Publications would "play an integral role in identifying the false reports." (Harris Aff. ¶ 9; see also id. ¶¶ 8, 10; Harris Reply Aff. ¶ 6.) Due to the incomplete information produced by defendants, plaintiffs allege that they are unable to limit the information they seek from the Publications by date or location. (Harris Aff. ¶ 9.)

  Finally, Dr. Harris opines that the subpoenas should not be limited to "major" hubs data because he cannot now say what indices influenced NYMEX (N.Y. Mercantile Exchange) prices, the "ultimate goal of the damage analysis." (Harris Aff. ¶ 10.)

  Other Investigations and Litigations Involving Subpoenas on McGraw-Hill*fn4

  In connection with a grand jury investigation, McGraw-Hill gave the U.S. Attorneys' Office in San Francisco over a thousand spreadsheets pursuant to court order. (Dkt. No. 312: Briskin Aff. Ex. 16: 3/25/04 S.D. Tex. Conf. Tr. at 23-24.) Those spreadsheets were not the "internal resource documents" but rather the documents that Williams (a defendant here) had provided to Inside FERC. (Id. at 25.) The California court did not recognize the federal First Amendment qualified privilege. (Id.) The spreadsheets McGraw-Hill produced in California redacted the identity of every confidential source other than the gas company under investigation. (Id.) In Texas, McGraw-Hill turned over to the Government unredacted internal spreadsheets, in original format, on a disk. (Briskin Aff. Ex. 17: 6/16/04 S.D. Tex. Conf. Tr. at 3.)

  In the civil context, in E&J Gallo Winery v. Encana Energy Servies, Inc., 33 Media L. Rptr. 1413 (S.D.N.Y. Jan. 12, 2005) (contained in Dkt. No. 335: Kovner Aff. Ex. D), Judge Preska quashed plaintiff Gallo's subpoena to non-party McGraw-Hill seeking trade publication documents in connection with plaintiffs' action against an energy company. Judge Preska held that the plaintiffs had failed to overcome the qualified reporters privilege which protected McGraw-Hill. Id., 33 Media L. Rptr. at 1414.

  In Foley v. American Elec. Power Cos., No. M8-85 (S.D.N.Y. Sept. 21, 2004) (contained in Kovner Aff. Ex. E), Judge Kaplan quashed a subpoena seeking discovery from a McGraw-Hill trade publication in a suit against an energy company, by memo endorsement of a motion that was unopposed by the subpoenaing party. (Kovner Aff. ¶ 8 & Ex. E.)

  In In re Application to Enforce Administrative Subpoena of the United States Commodity Futures Trading Comm'n v. McGraw-Hill Cos., No. 05-235, 2005 WL 2431262 (D.D.C. Oct. 4, 2005), the district court for the District of Columbia enforced a subpoena to McGraw-Hill seeking similar documents as here. In that litigation, the CFTC is claiming that the defendant energy company violated the Commodities Exchange Act. Id., 2005 WL 2431262 at *1.*fn5 Plaintiffs' Instant Subpoenas

  Despite having engaged in "significant discovery of Defendants' trade reports, [plaintiffs] determined that this information was not sufficient to reliably determine whether, where and to what extent Defendants had reported false trades, and what effect they had on the indices." (Dkt. No. 312: Briskin Aff. ¶ 4.) Therefore, on June 14, 2005, plaintiffs again served subpoenas on the Publications. (Briskin Aff. ¶ 4 & Ex. 3; see Dkt. No. 335: Kovner Aff. ¶ 20.) The new subpoenas, inter alia, requested:


All data used for constructing your published prices and indices for natural gas during the period January 1, 2000 to December 31, 2002 ("Class Period"). These data should include price, volume, date of each trade, and should identify any party or counterparty to each trade who is also a defendant in this action. Data produced in electronic format need not be produced in any other format.
(Briskin Aff. Ex. 3: Pls. Subpoena Att. A ¶ 1.) Plaintiffs make clear that they seek production of "purely numerical electronic price and volume information," and have agreed that the identity of sources other than defendants need not be produced. (Dkt. No. 311: Pls. Br. at 21.) The Publications objected to these subpoenas on the same grounds raised in their earlier motions to quash the 2004 subpoenas. (Briskin Aff. Ex. 4: 6/27/05 Intelligence Press Objection Letter & 7/8/05 McGraw-Hill Objection Letter; Kovner Aff. ¶ 20.)

  In a telephonic meet and confer with defendants and the Publications, plaintiffs explained that they were not willing to narrow the scope of the data sought in the subpoenas because, through discovery, plaintiffs had learned that "[d]efendants' false reporting was much more widespread tha[n plaintiffs] had initially thought." (Briskin Aff. ¶ 8.) Plaintiffs agreed that they would not need information compiled from any site and/or date where none of the defendants reported trades. (Id.) Counsel for McGraw-Hill acknowledged that the Publications possessed reported trade data that no one else had, but maintained that they would not voluntarily disclose the reported trade data from defendants or non-defendant natural gas traders. (Id. ¶ 9.)

  Defendants' Second Set of Subpoenas

  In August 2005, defendants served their second subpoena on McGraw-Hill (but not Intelligence Press). (Dkt. No. 356: 9/2/05 Defs. Letter Br. at 2; Dkt. No. 335: Kovner Aff. ¶ 21 & Ex. K: 8/12/05 Subpoena; see 11/9/05 Oral Arg. Tr. at 88.) This subpoena is broader in scope than plaintiffs'. (9/2/05 Defs. Letter Br. at 2; Kovner Aff. Ex. K: 8/12/05 Subpoena at 4: Documents to Be Produced; see 11/9/05 Oral Arg. Tr. at 69-72.) Specifically, defendants seek all data submitted by the defendants, not just the "actual data used" by the Publications, which is what plaintiffs seek. (Kovner Aff. Ex. K at 4.) Defendants' subpoenas also seek "[a]ll documents related to [the Trade Publications'] determination and publication of natural gas prices and indices . . .," including source identification of non-defendants. (Id.) Defendants have not moved to compel production from the Publications but have submitted a letter brief opposing plaintiffs' motion to compel. (9/2/05 Defs. Letter Br.)*fn6 Plaintiffs' Present Motion to Compel Production

  On August 1, 2005, plaintiffs moved to compel production of the reported trade data from the Publications (Dkt. No. 310: Pls. Motion to Compel) on the grounds that: (1) plaintiffs have overcome the qualified reporter's privilege because the documents sought are "highly material" and "critical" to their claim (Dkt. No. 311: Pls. Br. at 13-15) and are not available from other sources (id. at 15-21); (2) plaintiffs do not seek disclosure of confidential sources (id. at 21-23); and (3) plaintiffs' request is not unduly burdensome on the Publications (id. at 23-25).

  The Court heard oral argument on the motion on November 9, 2005. (See generally 11/9/05 Oral Arg. Tr.)



  A. Standard of Review on a Motion to Compel

  Federal Rule of Civil Procedure 45(c)(2)(B) provides, in part:

If objection has been made, the party serving the subpoena may, upon notice to the person commanded to produce, move at any time for an order to compel the production. Such an order to compel production shall protect any person who is not a party or an officer of a party from significant expense resulting from the inspection and copying commanded.
Motions to compel production of documents pursuant to a subpoena are "`entrusted to the sound discretion of the court.'" In re Fitch, Inc., 330 F.3d 104, 108 (2d Cir. 2003). The general rules of discovery apply to subpoenas issued under Rule 45. Rule 26(b)(1) provides that "[p]arties may obtain discovery regarding any matter, not privileged, that is relevant to the claim or defense of any party. . . ."*fn7

  The serving party bears the burden of showing the appropriateness of a subpoena served on a nonparty. In re Subpoena Duces Tecum Served on Bell Commc'n Research, Inc., No. MA-85, 1997 WL 10919 at *2 (S.D.N.Y. Jan. 13, 1997), modified on other grounds, 1997 WL 16747 (S.D.N.Y. Jan. 17, 1997). "In addition, where, as here, the discovery is sought from a non party, the Court should be particularly sensitive to weighing the probative value of the information sought against the burden of production on the non party." Fears v. Wilhelmina Model Agency, Inc., 02 Civ. 4911, 2004 WL 719185 at *1 (S.D.N.Y. Apr. 1, 2004); see, e.g., In re Application of Dow Jones &Co., No. 98 Misc. 8-85, 1998 WL 883299 at *6 (S.D.N.Y. Dec. 17, 1998); Concord Boat Corp. v. Brunswick Corp., 169 F.R.D. 44, 49 (S.D.N.Y. 1996) ("[T]he status of a witness as a non-party to the underlying litigation entitles [the witness] to consideration regarding expense and inconvenience.'"); Fed.R.Civ.P. 45(c)(2)(B) ("[A]n order to compel production shall protect any person who is not a party or an officer of a party from significant expense resulting from the inspection and copying commanded.").

  B. The Qualified Reporter's Privilege

  It is well-settled in the Second Circuit that reporters are protected by a qualified privilege against compelled disclosure of confidential sources and information prepared or obtained in connection with a news story. See, e.g., In re Fitch, Inc., 330 F.3d 104, 108-09 (2d Cir. 2003) ("[B]oth federal and New York courts recognize a special privilege that can be asserted by journalists resisting a subpoena."); United States v. Sanders, 211 F.3d 711, 719-20 (2d Cir.) (noting that the qualified journalist's privilege was first recognized by the Second Circuit in Baker v. F&F Inv., 470 F.2d 778, 785 (2d Cir. 1972), cert. denied, 411 U.S. 966, 93 S. Ct. 2147 (1973)), cert. denied, 531 U.S. 1015, 121 S. Ct. 574 (2000); Gonzales v. Nat'l Broad. Co., 194 F.3d 29, 32 (2d Cir. 1999) ("This circuit has long recognized the existence of a qualified privilege for journalistic information."); United States v. Burke, 700 F.2d 70, 76 (2d Cir.) ("When a litigant seeks to subpoena documents that have been prepared by a reporter in connection with a news story, this Circuit's standard of review . . . is well-settled"), cert. denied, 464 U.S. 816, 104 S. Ct. 72 (1983); In re Petroleum Prods. Antitrust Litig., 680 F.2d 5, 7 (2d Cir.) ("The law in this Circuit is clear that to protect the important interests of reporters and the public in preserving the confidentiality of journalists' sources, disclosure may be ordered only upon a clear and specific showing that the information" meets the three prong test described in the next paragraph.), cert. denied, 459 U.S. 909, 103 S. Ct. 215 (1982).*fn8

  The qualified privilege may be overcome only upon a "clear and specific showing" that the information is: "`[1] highly material and relevant, [2] necessary or critical to the maintenance of the claim, and [3] not obtainable from other available sources.'"*fn9 United States v. Cutler, 6 F.3d 67, 71 (2d Cir. 1993) (quoting In re Petroleum Prods. Antitrust Litig., 680 F.2d at 7); accord, eg., Gonzales v. Nat'l Broad. Co., 194 F.3d at 33 (Referring to the In re Petroleum Prods. Antitrust Litig. language as "language that has since been characterized as the Second Circuit's `test' for overcoming the journalist's privilege . . ."); United States v. Burke, 700 F.2d at 76-77; Baker v. F&F Inv., 470 F.2d at 783-85; New York Times Co. v. Gonzales, 382 F. Supp. 2d 457, 490 (S.D.N.Y. 2005); Persky v. Yeshiva Univ., 2002 WL 31769704 at *3 (referring to the test as a "`stringent test.'").*fn10 "This demanding burden has been imposed by the courts to `reflect a paramount public interest in the maintenance of a vigorous, aggressive and independent press capable of participating in robust, unfettered debate over controversial matters, an interest which has always been a principal concern of the First Amendment. . . .'" United States v. Burke, 700 F.2d at 77 (quoting Baker v. F&F Inv., 470 F.2d at 782); accord, e.g., Gonzales v. Nat'l Broad. Co., 194 F.3d at 33, 35; Von Bulow v. Von Bulow, 811 F.2d at 142, 144. The qualified reporter's privilege applies to non-confidential, as well as to confidential, information.*fn11 Gonzales v. Nat'l Broad. Co., 194 F.3d at 33, 36; see also, e.g., Von Bulow v. Von Bulow, 811 F.2d at 142-43; Pugh v. Avis Rent A Car Sys., Inc., 1997 WL 669876 at *5; In re Behar, 779 F. Supp. at 275. Where non-confidential information is sought, however, the nature of the press interest protected by the privilege is narrower, and therefore the privilege "should be more easily overcome." Gonzales v. Nat'l Broad. Co., 194 F.3d at 36. "Litigants seeking to subpoena nonconfidential press materials need make a less demanding showing than those who seek confidential press resources." Id. at 32; accord id. at 36. Accordingly, a litigant seeking non-confidential materials from a nonparty press entity "is entitled to the requested discovery . . . if he can show that the materials at issue are of likely relevance to a significant issue in the case, and are not reasonably obtainable from other available sources." Id. at 36.

  C. The Qualified Reporter's Privilege Applies to The Publications

  Before the Court applies the test to the facts of a particular case, it "must first determine whether the [party claiming privilege] has met his threshold burden of establishing the elements of a privileged relationship." SEC v. Seahawk Deep Ocean Tech., 166 F.R.D. 268, 270 (D. Conn. 1996); see Von Bulow v. Von Bulow, 811 F.2d 136, 144 (2d Cir.) ("It is axiomatic that the burden is on a party claiming the protection of a privilege to establish those facts that are the essential elements of the privileged relationship."), cert. denied, 481 U.S. 1015, 107 S. Ct. 1891 (1987); Persky v. Yeshiva Univ., 01 Civ. 5278, 2002 WL 31769704 at *2 (S.D.N.Y. Dec. 10, 2002) (quoting Von Bulow); Application of Waldholz, 87 Civ. 4296, 1996 WL 389261 at *2 (S.D.N.Y. July 11, 1996).*fn12

  Here, the Publications have asserted protection for their documents under the qualified reporter's privilege The financial press is included among the "reporters" protected by the qualified privilege. See, e.g., In re Application to Enforce Administrative Subpoena of the United States Commodity Futures Trading Comm'n v. McGraw-Hill Cos., No. 05-235, 2005 WL 2431262 at *3 (D.D.C. Oct. 4, 2005) (Journalist's qualified privilege applies to Platts' natural gas price indices); E&J Gallo Winery v. Encana Energy Servs., Inc., 33 Media L. Rptr. 1413, 1414 (S.D.N.Y. Jan. 12, 2005) (Dkt. No. 335: Kovner Aff. Ex. D) (recognizing that journalist's privilege applied to Platts); Stephens v. American Home Assurance Co., 91 Civ. 2898, 1995 WL 230333 at *6 (S.D.N.Y. Apr. 17, 1995) ("Publishers of corporate financial information have been held to be members of the press for purposes of the reporter's privilege.") (finding Best's insurance company ratings publication to be the "press"); In re Pan Am Corp., 161 B.R. 577, 584 (S.D.N.Y. 1993) (financial press, including Standard & Poor's rating services, "fully shielded by the umbrella of the First Amendment. Cases in this Circuit and elsewhere have addressed the specific question of the applicability of the journalist's privilege to financial publications and concluded that the privilege does indeed cover such works.") (collecting cases); see generally, e.g., In re Fitch, Inc., 330 F.3d 104 (2d Cir. 2003) (recognizing that the qualified journalist's privileged can apply to Fitch rating agency); In re Petroleum Prods. Antitrust Litig., 680 F.2d 5 (2d Cir. 1982) (recognizing that the qualified journalist's privilege applied to Platts), cert. denied, 459 U.S. 909, 103 S. Ct. 215 (1982).

  Plaintiffs do not dispute that the privilege applies, at least at some level, to the Publications. (See Dkt. No. 311: Pls. Br. at 11, 13; 11/9/05 Oral Arg. Tr. at 8.)

  However, while plaintiffs concede that the qualified reporter's privilege applies to the Publications, they argue that since they do not seek confidential sources other than when the source is a defendant, and only seek numerical information, the subpoenaed material should be considered non-confidential, and thus the more flexible test would apply. (Pls. Br. at 11, 13; 11/9/05 Oral Arg. Tr. at 9-10, 13-15.) See Gonzales v. Nat'l Broad. Co., 194 F.3d 29, 35-36 (2d Cir. 1999), discussed at page 21 above.

  On the other hand, the Publications assert both that defendants themselves were confidential sources and, in any event, it is not only the sources that are confidential, but the underlying price information as well, and finally, the privilege is the reporters' as well as the sources'. (Dkt. No. 316: Steis Supp. Aff. ¶ 11; Dkt. No. 218: Steis Aff. ¶ 5 (The source base that Intelligence Press has built "would face real jeopardy if Intelligence Press were forced to disclose the identities of its survey participants or the confidential information they have submitted."); Dkt. No. 333: 3d Foster Aff. ¶ 10; Dkt. No. 217: 1st Foster Aff. ¶ 7; see Dkt. No. 230: Persky Aff. Ex. D: Platts Methodology & Specifications Guide at 4, "Security and Confidentiality"; Dkt. No. 315: Siegel Aff. Ex. 1 at 1 & Ex. 2 at 2; see generally 11/9/05 Oral Arg. Tr. at 85.)

  The fact that defendants themselves have subpoenaed the Publications for all data defendants and others submitted to the Publications could be construed as a waiver of any confidentiality understandings they once had with the Publications. (See 11/9/05 Oral Arg. Tr. at 9-12.) Indeed, at oral argument, defense counsel conceded that defendants have no objection to the Publications revealing their "source" data, i.e., are waiving their confidential source status. (11/9/05 Oral Arg. Tr. at 71.) The Publications, however, continue to insist that they are not waiving their own interest in protecting their sources, because of concerns about their ability to obtain voluntary price reports in the future. (11/9/05 Oral Arg. Tr. at 85.)

  The Court believes that defendants' concession that they are waiving their source confidentiality protection significantly weakens the Publications' argument. Nevertheless, the Court need not decide whether the subpoenaed information is confidential or non-confidential. As discussed in depth below, plaintiffs have made a showing sufficient to overcome even the high burden of the three prong test applicable to confidential sources. II. APPLICATION OF THE SECOND CIRCUIT'S THREE PART TEST TO THE SUBPOENAS

  A. The Material Sought Is Highly Material and Relevant

  The first step in analyzing whether plaintiffs have overcome the qualified reporter's privilege is whether they made a "clear and specific showing" that the subpoenaed information is "highly material and relevant." (See page 19 above.)

  The complaint alleges that defendants submitted false trade reports to the Publications to thereby artificially alter the price for NYMEX futures contracts. As Judge Marrero summarized the complaint in denying defendants' motion to dismiss:

Plaintiffs allege that during the Class Period, Defendants "manipulated the price and volume information of their natural gas trades to trade publishers." (CCAC ¶ 59.) Plaintiffs claim that Defendants intentionally submitted false price and volume information on spot trades to several of the gas industry publications that collect that information with the goal of artificially skewing the published reports on natural gas trades in the physical market, and thereby artificially altering the futures market for natural gas, which is determined at least in part by those published reports. In the words of the CCAC, the Defendants "planned and executed a scheme designed to cause price instability and increase volatility in spot prices and thereby manipulate the price of natural gas futures and options traded on the NYMEX to artificial levels." (CCAC ¶ 12.)
See In re Natural Gas Commodity Litig., 337 F. Supp. 2d 498, 502-03 (S.D.N.Y. 2004).*fn13


As plaintiffs' brief contends: Integral to Plaintiffs' claims is that Defendants used the Publications as the instrumentality of a scheme to manipulate the prices of natural gas and natural gas futures. Indeed, it is now widely acknowledged that throughout the Class Period, false reporting was rampant in the natural gas industry. . . . As the Publications themselves and this Court has acknowledged they are the only repository of the trade data that Defendants falsified in order to manipulate prices, and the only means by which Plaintiffs can fully reconstruct the published indices, in order to show the effect of the Defendants' rampant false reporting on natural gas futures prices.
(Dkt. No. 311: Pls. Br. at 1.) The Court agrees, at least with respect to the Henry hub (see pages 29-31 below), that plaintiffs have shown that the subpoenaed information is highly material and relevant.

  The Publications claim that the subpoenaed information is not highly material and relevant because of the Court's prior dicta, requiring plaintiffs to identify false trade reports from defendants before seeking information from the Publications. (See Dkt. No. 332: Platts Br. at 1; see also 11/9/05 Oral Arg. Tr. at 45.)*fn14 The Publications thus argue that "[w]hen the plaintiffs last made this request, this Court made unequivocally clear that `because of the way the qualified press privilege works, until plaintiffs have identified on what days and from what defendants there were arguably false trades, [they] don't have the initial bit of information that [they] need to show that there was an effect on the indexes. . . .'" (Platts Br. at 1, quoting 1/14/05 Conf. Tr. at 6; see also Platts Br. at 18-19; 11/9/05 Oral Arg. Tr. at 37, 39, 48.) This argument, however, goes more to the third prong of the Second Circuit test, whether plaintiffs have shown the information was not available from non-press sources. As discussed further in Point II.C below, the Court is satisfied that plaintiffs have shown that the false reported trade information is not obtainable from defendants.

  The Publications also argue that "to the extent that identification of false trade reports depends on comparison between actual trades and reported trades, such identification is impossible without complete records of defendants' actual trades," which the Publications do not have. (Platts Br. at 3.) Plaintiffs, however, have obtained defendants' actual trading data. (Dkt. No. 318: Pls. Reply Br. at 7; Dkt. No. 312: Briskin Aff. Ex. 2: Harris Aff. ¶¶ 7, 8.) Defendants confirmed this at oral argument. (11/9/05 Oral Arg. Tr. at 21.) Thus, by matching defendants' actual trades with their reported trades to the Publications, plaintiffs can identify false reports. (Briskin Aff. Ex. 2: Harris Aff. ¶ 8.)

  The Publications also implicitly argue that since the CFTC and FERC were able to obtain settlements with certain energy companies without this data, plaintiffs do not need the Publications' data. (Platts Br. at 22-23; Dkt. No. 335: Kovner Aff. ¶¶ 12-17; 11/9/05 Oral Arg. Tr. at 46-48.) As plaintiffs correctly point out, the administrative agencies need merely prove some false reports and need not prove damages; plaintiffs, however, need to show the false reports, causation and damages. (See Pls. Reply Br. at 3; see also Dkt. No. 351: 10/24/05 Persky Letter to Court at 2.) Cf. Natural Gas II, 2005 WL 2414449 at *4. According to plaintiffs' expert, Dr. Michael Harris, in order to prove damages, plaintiffs must construct "but for" indices using the actual trade transaction data — already obtained from defendants — and the reported trade data reported to and used by the Publications — available only from the Publications. (See Pls. Reply Br. at 1: "By comparing the trade reports actually received and used by the Publications with the natural gas trades reflected in each defendant's books and records, plaintiffs and their expert can both identify which trade reports were false, and identify the impact these false reports had on the published indices."; see also pages 10-11 above.) Additionally, to prove causation, plaintiffs must show that the false reported trade data used by the Publications had an impact on the published indices. (See pages 25-26 & n. 13 above.)

  Indeed, in In re Application to Enforce Administrative Subpoena of the United States Commodity Futures Trading Comm'n v. McGraw-Hill Cos., No. 05-235, 2005 WL 2431262 (D.D.C. Oct. 4, 2005), an action brought by the CFTC pursuant to the Commodities Exchange Act, the District Court for the District of Columbia compelled production from McGraw-Hill of documents similar to those subpoenaed by plaintiffs here.

  The Court held, in language equally applicable here, that the subpoenaed information was "crucial to the plaintiff's proof":

In its investigation of false reporting, the CFTC seeks to prove that Energy Company submitted false data on several occasions, so it needs evidence of incorrect data having been submitted. A request for the data Platts received from Energy Company clearly goes to the heart of proving that the data was false.
The second charge under investigation is whether Energy Company, having submitted false data, did so in an attempt to affect energy prices. The CFTC suspects that Energy Company attempted to manipulate prices by providing false data to Platts, knowing that it would be published and that readers, in turn, would rely upon it to set prices going forward. To determine whether Energy Company's false reports actually did affect prices, or had the potential to do so, the CFTC needs to know how Platts transformed the raw data from companies such as Energy Company into its published indices. The CFTC indicated at the Hearing that it intends to recalculate the prices using correct data to see what effect the false reporting might have had. Thus, Platts' formulas are crucial for determining whether Energy Company's false reporting did, or could have, had an effect on prices.
Id., 2005 WL 2431262 at *4. The same analysis and conclusion applies here.*fn15

  The Court cannot say, however, that plaintiffs have met their burden of showing that data from all trading "hubs" is "highly material and relevant." As noted above, plaintiffs' claim is that NYMEX (N.Y. Mercantile Exchange) prices for natural gas futures contracts were affected by defendants' (allegedly) false reported trades used in the Publications' indices. (See pages 2, 25 above.) NYMEX futures contracts are for future delivery of natural gas at the Henry hub only. Natural Gas I, 337 F. Supp. 2d at 502. (See also 11/9/05 Oral Arg. Tr. at 33-34, 80; Kovner Aff. Ex. L: Harris Aff. ¶ 8; Dkt. No. 356: Defs. 9/2/05 Letter Br. at 2 & n. 1; Platts Br. at 19-20.) Plaintiffs' expert, Dr. Harris, concedes that he cannot now determine which hubs affect the NYMEX prices:

10. In addition, it is not possible to limit the data requested from the trade publications by reference to whether a reporting location is a "major" location. Generally, one might describe a reporting location as major based on some measure of the volume of natural gas transacted at that location. Thus, major trading points would be large, liquid, heavily transacted locations. However, imposing such a limitation assumes that the underlying damage analysis can similarly be limited to only major reporting locations. The ultimate goal of the damage analysis used in the calculation of class-wide damages is to determine the amount of artificiality in the prices of natural gas on the New York Mercantile Exchange ("NYMEX"). The NYMEX prices are a function of the reported index prices. However, which set of reported index prices is an empirical question that has yet to be determined. It may very well be that NYMEX prices are more heavily influenced by index prices at major reporting locations. However, at the same time, it may also be the case that index prices at smaller less liquid reporting locations also have a significant explanatory effect.
(Briskin Aff. Ex. 2: Harris Aff. ¶ 10) (fn. omitted).*fn16

  The Court is no economist, but it seems that reported trade prices for the Henry hub would have the most direct effect on NYMEX future prices for natural gas to be delivered at that very hub. It stands to reason that defendants also might have to report trades at "similar" prices at other hubs, so the Publications (and industry) would not view the reporting as to the Henry hub as being out of line with other hubs. But it would be the false trades at the Henry hub that would be of most immediate impact. (See 11/9/05 Oral Arg. Tr. at 31, 80-81.) While the Court's analysis may be incorrect, Dr. Harris' opinion that what hub matters is an "empirical question that has yet to be determined" (Harris Aff. ¶ 10) is insufficient to meet plaintiffs' burden of showing that reported trade information at other than the Henry hub is "highly material and relevant" to plaintiffs' case. (See also Defs. 9/2/05 Letter Br. to Court at 2 n. 1: "NYMEX prices are based on the price for natural gas deliverable on the Sabine Pipeline Co. at Henry Hub in Erath, Louisiana, when the current contract month expires. Plaintiffs have no theory as to how any of the other index prices affected those prices."; see also Platts Br. at 19-20.) Accordingly, the Court quashes the subpoenas to the extent they seek information from the numerous hubs other than the Henry hub.*fn17

  With that (significant) limitation, the Court finds the subpoenaed information to be highly material and relevant to plaintiffs' case, satisfying the first prong of the Second Circuit's test. B. The Material Sought is "Necessary or Critical" to Plaintiffs' Claim

  The second step in the Circuit's three prong test is to determine whether the material sought is "necessary or critical" to the case. To do so there must be a finding that the claim for which the information is to be used "`virtually rises or falls with the admission or exclusion of the proffered evidence.'" In re Application to Quash Subpoena to Nat'l Broad. Co., 79 F.3d 346, 351 (2d Cir. 1996) (applying the identical test under N.Y. Civil Rights Law § 79-h(c)); Zerilli v. Smith, 656 F.2d 705, 713 (D.C. Cir. 1981) ("If the information sought goes to `the heart of the matter,' that is, if it is crucial to [the party's] case, then the argument in favor of compelled disclosure may be relatively strong. . . . On the other hand, if the information sought is only marginally relevant, disclosure may be very difficult to justify.") (citations & fn. omitted); Persky v. Yeshiva Univ., 01 Civ. 5278, 2002 WL 31769704 at *3 (S.D.N.Y. Dec. 10, 2002); Pugh v. Avis Rent A Car Sys., Inc., No. M8-85, 1997 WL 669876 at *4 (S.D.N.Y. Oct. 28, 1997) (unsupported speculation of what the subpoenaed materials might contain is not enough to make a clear and specific showing that the information sought is highly material and relevant nor necessary or critical to the party's claim).

  Information sought that would be cumulative of other evidence cannot be "necessary or critical" under the test. In re Application of Behar, 779 F. Supp. 273, 275 (S.D.N.Y. 1991); see also, e.g., United States v. Burke, 700 F.2d 70, 78 (2d Cir.) (affirming the district court's finding that party failed to overcome the privilege because the evidence sought would be cumulative of other impeachment evidence), cert. denied, 464 U.S. 816, 104 S. Ct. 72 (1983). Plaintiffs assert that the trade reports used by the Publications not only is highly material and relevant but is necessary and critical to their claims because it is "the only source that can directly show the quantitative effect of each defendant's false reporting on the published indices." (Dkt. No. 311: Pls. Br. at 14.) Plaintiffs argue that without knowing which false trade reports were made to and actually used by the Publications, there is no way for plaintiffs to prove that the false trade reports impacted the market or that plaintiffs were damaged by it. (Dkt. No. 318: Pls. Reply Br. at 3.) Moreover, proof of false trades is necessary for liability as well as damages. (See 11/9/05 Oral Arg. Tr. at 17, 22, 76.)

  The Publications argue that the reported trade data used is not necessary or critical to plaintiffs' claim because plaintiffs have not identified which trade reports were false, which "leaves plaintiffs at the same place they were when" this Court quashed plaintiffs' first set of subpoenas without prejudice. (Dkt. No. 332: Platts Br. at 18-19.) Plaintiffs have demonstrated to the Court that they were unable to get the false trade report information from defendants because defendants have incomplete records of their reported trades. Indeed, defendants have conceded that they do not have anywhere near complete records of their reported trades. (11/9/05 Oral Arg. Tr. at 4-5.) According to Dr. Harris, a comparison of the defendants' actual trades (information that defendants have produced in full to plaintiffs, see 11/9/05 Oral Art. Tr. at 21) with the data defendants reported to the Publications will demonstrate not only that false reports were made (needed to prove liability) and when they were made, but also the effect they had on the published indices (i.e., causation and damages). (See pages 10-12 above.) Thus, only the actual trade data received from defendants combined with the reported trade data possessed by the Publications will reveal the full extent of, and effect of, the false trades. (Pls. Reply Br. at 7.) The Court agrees. Moreover, the argument really relates to the third prong — availability of the subpoenaed information from non-press sources.

  Second, the Publications argue that plaintiffs could use regression techniques to determine "but-for" prices, which does not require the reported trade data compiled by the Publications, as an alternative to direct reconstruction of the indices. (Platts Br. at 19.) Plaintiffs' expert Dr. Harris explained these two procedures in an affidavit submitted in support of plaintiffs' motion for class certification; Dr. Harris opined that the direct reconstruction approach was preferable but that the regression technique could be used if the information from the Publications was not available. (Dkt. No. 335: Kovner Aff. Ex. L: Harris Aff. ¶¶ 32-37.) Dr. Harris' affidavit made clear that using the reported trade report information obtained from defendants and the Publications was the "preferred" approach, preferable to the "alternative regression approach" (id.), the latter of which likely would lead to methodological challenges by defendants. The Publications have cited no case, and the Court has not found any, defining "necessary or critical" as requiring uniqueness. While the regression model is an available alternative, it is not as good (according to Dr. Harris). Indeed, defendants have stated that they doubt the regression analysis can be done, and already have asserted that they will challenge any such regression analysis as violative of Daubert. (11/9/05 Oral Arg. Tr. at 73-75.) Thus, obtaining the reported trade information from the Publications is sufficiently "necessary" to satisfy the Second Circuit's standard.*fn18 Third, the Publications argue that plaintiffs only subpoenaed the reported trade data used by the Publications, not the reported trade data submitted by the defendants, and thus if plaintiffs need the data from the Publications because of defendants' incomplete records, they have not subpoenaed such documents. (Platts Br. at 19 n. 17; see 11/9/05 Oral Arg. Tr. at 23.) However, as plaintiffs pointed out in reply, the reported trade data actually used by the Publications is a subset of the reported trade data submitted by defendants (removing "outliers" or other anomalous data), and was a result of plaintiffs' attempt to narrow their subpoenas to the Publications. Only those false reported trades that actually made it into the published indices could have affected market prices. (See Pls. Reply Br. at 8.) Nevertheless, plaintiffs made clear that the "raw" reported trade data is acceptable if the Publications prefer. (Id.; see 11/9/05 Oral Arg. Tr. at 23.)

  Fourth, the Publications argue that plaintiffs have not shown a need for reported trade data from any location other than the Henry hub. (See Platts Br. at 20.) The Court has already limited the subpoenas to reported trade data from only the Henry hub, in this Court's discussion of the "highly material and relevant" prong of the Second Circuit test. (See Point II. A above.)

  C. The Information Subpoenaed From The Publications Is Not Obtainable From Other Available Sources

  As to the third prong of the Second Circuit test, the courts in this Circuit abide by a strict standard requiring a showing by the party seeking disclosure "that it has exhausted all other available non-privileged sources for the information." In re Pan Am Corp., 161 B.R. 577, 585 (S.D.N.Y. 1993) (deposition of two representatives fell short of the effort required to show unavailability); see In re Application to Quash Subpoena to Nat'l Broad. Co., 79 F.3d 346, 353 (2d Cir. 1996) (party "failed to make the requisite clear and specific showing of unavailability" because she "simply failed to exhaust all other available sources of information."); In re Petroleum Prods. Antitrust Litig., 680 F.2d 5, 8-9 (2d Cir.) (considering subpoena for Platts' documents, overturning order requiring disclosure of journalist's notes despite the fact that hundreds of depositions had been taken, since deponents were not asked if they had communicated pricing information to Platts), cert. denied, 459 U.S. 909, 103 S. Ct. 215 (1982); Pugh v. Avis Rent A Car Sys., Inc., No. M8-85, 1997 WL 669876 at *4 (S.D.N.Y. Oct. 28, 1997) (claiming that deposing other individuals who may possess the information sought would be futile is not enough); In re Application of Behar, 779 F. Supp. 273, 276 (S.D.N.Y. Oct. 18, 1991).

  As discussed above, this Court quashed plaintiffs' first set of subpoenas to the Publications because, at that time, plaintiffs had not shown whether they could obtain the false reported trade information from defendants. (See page 9 above.) The Court suggested that plaintiffs first identify the false reported trades by discovery from defendants, and then seek only the indices for those days from the Publications. (See page 9 above.) The Publications argue that plaintiffs have obtained some false reported trade information from defendants, and that plaintiffs should be required to make further efforts to obtain additional data, including review of thousands of hours of audiotaped conversations. (Dkt. No. 332: Platts Br. at 25-29.) If plaintiffs spent the thousands of hours necessary to listen to all the audiotapes, plaintiffs would have additional examples of false reported trades, but by no means a complete or nearly complete set of data. (See Dkt. No. 311: Pls. Br. at 15; Dkt. No. 318: Pls. Reply Br. at 6-7.) The simple fact is that defendants did not keep anything like complete records of their reported trades to the Publications. (See pages 9-10 & n. 3 above.) Defendants have conceded as much. (11/9/05 Oral Arg. Tr. at 4-5.) Plaintiffs now have made the necessary efforts to try to obtain these records from defendants. (See pages 9-10 & n. 3 above.) Indeed, in light of defendants' agreement at oral argument that defendants do not have anywhere near complete reported trade data, the exhaustion requirement clearly has been met. No matter what efforts plaintiffs go through, defendants do not have all or even most of this data.

  The third prong of the Second Circuit test refers to information being obtainable from "available" sources; it does not require that every theoretical source be exhausted. The analysis of the District Court for the District of Columbia is right on point:

Of equal importance in the balancing test is the extent to which the party seeking disclosure has attempted to obtain the information from other possible sources. "Even when the information is crucial to a litigant's case, reporters should be compelled to disclose their sources only after the litigant has shown that he has exhausted every reasonable alternative source of information." Zerilli, 656 F.2d at 713. Thus, before seeking the information from McGraw-Hill, the CFTC must show that it has attempted to obtain the data from other sources, if any. This requirement reflects the principle that the privilege should be abrogated only as a last resort. Carey, 492 F.2d at 638. This does not mean, however, that a litigant must engage in an "onerous," "wide-ranging" or "ill-lighted" discovery burden before seeking to obtain the information from a news gatherer. Id. at 639. Energy Company itself would appear to be the first source for the data it reported. The CFTC did obtain from Energy Company some relevant records, but the data provided by Energy Company was incomplete. Only after a review of the data obtained from Energy Company revealed that it was inadequate did the CFTC subpoena McGraw-Hill. This Court has not been directed to any other plausible source of the transaction data Energy Company submitted to Platts.
McGraw-Hill argued at the Hearing that the CFTC should be required first to seek the data from other energy companies who might have inadvertently received and retained the data. This overstates the CFTC's obligation: the CFTC must have exhausted only those alternative sources that are reasonably available, not every other conceivable source. Zerilli, 656 F.2d at 714 (discussing Carey, 492 F.2d at 639, for the proposition that while the obligation is "very substantial," "there are some limits to the obligation to pursue alternative sources"). This Court finds that data that may or may not reside in the files of various other unrelated companies is not reasonably available to the CFTC. Accordingly, as there exists no other reasonably available source for this information, the CFTC has sufficiently exhausted alternative sources.
In re Application to Enforce Administrative Subpoena of the United States Commodity Futures Trading Comm'n v. McGraw-Hill Cos., No. 05-235, 2005 WL 2431262 at *5 (D.D.C. Oct. 4, 2005); see also, e.g., Lee v. Dep't of Justice, 413 F.3d 53, 61 (D.C. Cir. 2005) ("While [plaintiff] did not depose every individual who conceivably could have leaked the information, . . . As the Court stated in Carey, `litigants [need not] be made to carry wide-ranging and onerous discovery burdens where the path is . . . ill lighted.' [Plaintiff] has done far more to exhaust alternatives than the plaintiff in Zerelli who did not meet his burden, and at least as much as the successful plaintiffs in Garland and Carey.") (citations omitted).*fn19 Here, as in the CFTC decision, it is clear to this Court that "none of the defendants has a complete set of its trade reports, many of which were not retained, were submitted on unrecorded telephone lines, or are in many hundreds of thousands of hours of raw audiotape." (Dkt. No. 351: 10/24/05 Persky Letter to Court at 2; see pages 9-10 & n. 3 above; see also Pls. Br. at 2: "no Defendant in this case has produced all of the trade reports it provided to the Publications, either because it did not retain the data, it destroyed the data, or it communicated the data in a form that was not preserved.")

  While the Publications argue that the D.D.C. decision is distinguishable because CFTC should stand in a better position than a private litigant (Dkt. No. 352: Kovner 10/21/05 Letter to Court at 1-2; 11/9/05 Oral Arg. Tr. at 51), the Court notes that the present case is a class action, against numerous defendants, who have waived their "source" privilege. (See 11/9/05 Oral Arg. Tr. at 71.) Moreover, this is a situation, according to plaintiffs' complaint (based on CFTC and FERC investigations and reports), in which the Publications (allegedly) were used by defendants as instrumentalities in defendants' scheme to manipulate natural gas futures prices. (See 11/9/05 Oral Arg. Tr. at 17-19; see also page 26 above.) The Publications' innocent and unwitting involvement in defendants' scheme, the class action nature of this case, the defendants' waiver of their "source" privilege, and the fact that the information involved is purely numeric (price) information and of an outdated, historical nature, all make it more appropriate than it might be in other cases for the subpoenas to the Publications to be enforced. (See 11/9/05 Oral Arg. Tr. at 17, where plaintiffs' counsel noted that "[w]e have a rather sui generis situation here."; cf. Platts Br. at 7-8: "[T]o the extent the allegations of false reporting are true, Platts is a victim of the conduct under investigation, since it was the direct, intended recipient of any misrepresentations.") Frankly, the Court is somewhat surprised that Platts, the victim/instrumentality, does not wish to help the government agencies and the class plaintiffs to remedy the defendants' (alleged) wrongs.*fn20 (See 11/9/05 Oral Arg. Tr. at 40-41.)*fn21 * * *

  For these reasons, the Court finds that the Publications are entitled to claim the qualified reporter's privilege, but that plaintiffs have satisfied the three prongs of the Second Circuit test to abrogate that qualified privilege and enforce the subpoenas, limited to information for the Henry hub. The Court now turns to the separate issue of burdensomeness.

  D. The Court Modifies The Subpoenas To Some Extent To Obviate Any Burden On The Publications

  Although the Court has found that plaintiffs have made a sufficient showing to abrogate the Publications' qualified reporter's privilege, that does not end the inquiry. The Publications also oppose plaintiffs' motion to compel on the ground that the subpoenas are burdensome. (Dkt. No. 332: Platts Br. at 30-32; Dkt. No. 317: Intelligence Press Br. at 1-12.)*fn22 Federal Rule of Civil Procedure 45(c)(3)(A)(iv) provides that the court shall quash or modify a subpoena that "subjects a person to undue burden."

  The Court already has modified the subpoenas so as to call only for data as to the Henry hub, not all the other hubs. (See pages 29-31 above.) That significantly reduces any burden on the Publications.

  As to Platts' Gas Daily and its daily price indices, the undisputed evidence is that it discarded reported trade data before November 2002. (Platts Br. at 6: "For all daily indices published before November 2002, Gas Daily simply does not have the records that would enable plaintiffs — or anyone else — to retroactively match the data used to create the index with its sources.") (citing 3d Foster Aff. ¶ 7). Because the Class Period is 2000-2002, enforcement of the subpoena as to Gas Daily would provide plaintiffs with information for less than two months of the thirty-six month Class Period. While plaintiffs' counsel asserted at oral argument that Dr. Harris could use that limited data to make assumptions about the rest of the class period, the Publications' counsel responded that there already was publicity about the false reporting by that time, so the last two months is different than the rest of the period. (Compare 11/9/05 Oral Arg. Tr. at 26-29 with id. at 52.) Plaintiffs' counsel's assertion is not supported by any affidavit from Dr. Harris, and the Court is dubious. In any event, balancing the burden on Platts with the limited utility of the information for such a short period, the Court quashes the subpoena as to Gas Daily.

  As to the other Platts publication, Inside FERC, Platts' burden argument is not well developed or supported, and appears largely based on the large number of indices because of the large number of hubs. (See Platts Br. at 31.) The Court believes that limiting the subpoena to the Henry hub should obviate that burden. Indeed, the subpoenaed data is kept in electronic (computerized) format; and Platts has already produced both redacted and unredacted spreadsheets to the government in connection with criminal proceedings in California and Texas. (Dkt. No. 311: Pls. Br. at 23-24: "The information Plaintiffs request here is electronic data that already have been entered on Excel spreadsheets . . ." (citing to Platts' methodology); see also Dkt. No. 318: Pls. Reply Br. at 5.) While those spreadsheets themselves will not satisfy plaintiffs' subpoenas here, they do show the Publications can produce this information in electronic format and still redact information about companies other than the target sources, here defendants. (See 11/9/05 Oral Arg. Tr. at 24.) Moreover, during negotiations over the subpoenas, the parties "learned that McGraw-Hill has an outside document management firm that would oversee this [document] production, and thus this would not impinge on the newsgathering responsibilities of Platts employees." (Pls. Reply Br. at 5, citing Dkt. No. 312: Briskin Aff. Ex. 17: 6/16/04 S.D. Tex. Conf. Tr. at 30; see also 11/9/05 Oral Arg. Tr. at 49.) Thus, McGraw-Hill has not shown that plaintiffs' subpoena (as limited to the Henry hub) would be unduly burdensome.*fn23

  Intelligence Press seems to argue that as to its daily indices, the "temp" file for each day containing the editorially-altered data was used to create the next day's "temp" file. (See page 7 above.) Thus, none of the temp files should exist, all having been overwritten.*fn24 The non-edited spreadsheets for each day, however, exist to some extent on old computers (see 11/9/05 Oral Arg. Tr. at 56-57) and at least reveal reported trade data whether or not "used" by Intelligence Press.

  Intelligence Press acknowledges that spreadsheets for its monthly indices exist, but in the form of "multiple drafts" without record of what is the "final version." (See Intelligence Press Br. at 6-7, 10.) Whatever editorial judgment may be "missing" from those spreadsheets (id.), Intelligence Press concedes that the spreadsheets would contain the raw reported trade data (id. at 11), which is what plaintiffs need.

  The Court is cognizant of the fact that Intelligence Press is a small organization. (See page 5 above.) Nevertheless, Intelligence Press has indicated that the historical data plaintiffs seek is located on "old" and "antiquated individual computers" (see Intelligence Press Br. at 2, 7; Dkt. No. 218: Steis Aff. ¶ 8) — which presumably could be searched by specially retained outside personnel (at plaintiffs' cost) without significantly interfering with Intelligence Press' ongoing operations. To the extent that Intelligence Press complains that the necessary redactions would "alone take hundreds of hours or more" (Intelligence Press Br. at 7-8, 12), plaintiffs have offered to pay for technical and clerical staff to be hired by Intelligence Press to perform this ministerial job (see Pls. Reply Br. at 10; 11/9/05 Oral Arg. Tr. at 23-24). Thus, any burden is alleviated by plaintiffs paying the cost of hired staff for Intelligence Press.

  Nevertheless, the Court requires plaintiffs and Intelligence Press to negotiate a reasonable "sample" protocol — perhaps searching one of the "old" computers, with leave to re-visit the burden vs. utility question based on information from that process. (See 11/9/05 Oral Arg. Tr. at 90.) Plaintiffs and Platts also should discuss utilizing a search protocol (e.g., searching for one year's worth of data, instead of three, to start) to expedite production, limit the burden on the Publications, and perhaps develop information to return to Court to refine the Court's ruling. (See 11/9/05 Oral Arg. Tr. at 90-92.) This discussion process should occur even while objections are pending before Judge Marrero. Finally, when the Publications redact names of non-defendants, they should do so by utilizing a unique code name or number for the non-defendants. Counsel for the Publications shall retain the identification key. In case any information about a non-defendant becomes necessary in this case, the Publications thus would not have to repeat their search and production, but could identify the code for that particular entity.

  The Court notes that there is a Confidentiality Stipulation and Order in this case, which will protect the Publications' confidentiality interests. (See Pls. Br. at 7-8.)


  For the reasons set forth above, plaintiffs' motion to compel the Publications to comply with the subpoenas is GRANTED limited to the reported trade data for the Henry hub, and subject to the additional limitations discussed above. Based on the parties' agreement at oral argument (11/9/05 Oral Arg. Tr. at 97-98), the Court stays implementation of this Opinion and Order for the ten (10) day period allowed for the filing of objections; if objections are filed, by agreement of the parties, the stay will remain in effect until the objections are decided by Judge Marrero (or until further Court order).


   Pursuant to 28 U.S.C. § 636(b)(1) and Rule 72(b) of the Federal Rules of Civil Procedure, the parties shall have ten (10) days from service of this Opinion to file written objections. See also Fed.R.Civ.P. 6. Such objections (and any responses to objections) shall be filed with the Clerk of the Court, with courtesy copies delivered to the chambers of the Honorable Victor Marrero, 40 Centre Street, Room 414, and to my chambers, 500 Pearl Street, Room 1370. Any requests for an extension of time for filing objections must be directed to Judge Marrero. Failure to file objections will result in a waiver of those objections for purposes of appeal. Thomas v. Arn, 474 U.S. 140, 106 S.Ct. 466 (1985); IUE AFL-CIO Pension Fund v. Herrmann, 9 F.3d 1049, 1054 (2d Cir. 1993), cert. denied, 513 U.S. 822, 115 S.Ct. 86 (1994); Roldan v. Racette, 984 F.2d 85, 89 (2d Cir. 1993); Frank v. Johnson, 968 F.2d 298, 300 (2d Cir.), cert. denied, 506 U.S. 1038, 113 S.Ct. 825 (1992); Small v. Secretary of Health & Human Servs., 892 F.2d 15, 16 (2d Cir. 1989); Wesolek v. Canadair Ltd., 838 F.2d 55, 57-59 (2d Cir. 1988); McCarthy v. Manson, 714 F.2d 234, 237-38 (2d Cir. 1983); 28 U.S.C. § 636(b)(1); Fed.R.Civ.P. 72, 6(a), 6(e).



© 1992-2005 VersusLaw Inc.

Buy This Entire Record For $7.95

Official citation and/or docket number and footnotes (if any) for this case available with purchase.

Learn more about what you receive with purchase of this case.