The opinion of the court was delivered by: Samuel Der-yeghiayan, District Judge
This matter is before the court on Defendant Goldman Sachs & Company's (Goldman) renewed motion for summary judgment and on the parties' motions to strike. For the reasons stated below, we deny the renewed motion for summary judgment and grant in part and deny in part the motions to strike.
Plaintiff Premium Partners, L.P. (Premium) alleges that it held substantial short positions in 30-Year Treasury Options at 9:25 a.m. on October 31, 2001. Premium contends that Goldman and Defendant Massachusetts Financial Services Company (MFS) paid Defendant Peter J. Davis (Davis) to funnel to them nonpublic information that Davis discovered at a confidential United States Department of Treasury (Treasury Department) quarterly refunding meeting (Meeting), including a conference that took place between 9:00 a.m. and 9:25 a.m. on October 31, 2001. Davis allegedly learned during the Meeting that the Treasury Department would suspend the 30-Year Treasury Bond. Premium contends that when the information was released, the demand for 30-Year Treasury Bonds increased and, in turn, the costs increased for investors that had to cover short positions in 30-Year Treasury Futures and 30-Year Treasury Bond Options.
According to Premium, after the Meeting at 9:35 a.m., Davis called Defendant John M. Youngdahl (Youngdahl), who worked for Goldman, and at approximately 9:38 a.m. Davis called Defendant Steven E. Northern (Northern), who worked for MFS. Davis allegedly informed Youngdahl and Northern that the Treasury Department was going to suspend the 30-Year Treasury Bond. Goldman then allegedly immediately purchased $84 million in 30-Year Treasury Bonds and significant amounts of 30-Year Treasury Futures. MFS also allegedly immediately purchased $65 million in 30-Year Treasury Bonds before public disclosures. At 9:43 a.m. on October 31, 2001, a Treasury announcement was posted as a press release on the Treasury website and there was a formal announcement reported by Reuters at 9:57 a.m., Premium claims that Defendants manipulated the 30-Year Treasury Bond market, artificially influencing the price of 30-Year Treasury Bonds, Futures, and Options. Premium contends that Defendants' manipulation in turn required investors such as Premium to pay additional costs to cover short positions in 30-Year Treasury Futures and Options.
Premium includes in its complaint Commodity Exchange Act (CEA), 7 U.S.C. § 1 et seq., claims brought against Goldman and Youngdahl (Count I), CEA claims brought against MFS and Northern (Count II), a CEA claim brought against Davis (Count III), Illinois Consumer Fraud and Deceptive Business Practices Act, 815 ILCS 505/1 et seq., claims brought against all Defendants (Count IV), civil conspiracy claims brought against Goldman, Youngdahl, and Davis (Count V), civil conspiracy claims brought against MFS, Northern, and Davis (Count VI), Sherman Antitrust Act claims brought against Goldman, Youngdahl, and Davis (Count VII), and Sherman Antitrust Act claims brought against MFS, Northern, and Davis (Count VIII).
The prior judge in this case granted in part and denied in part Defendants' motions to dismiss, dismissing all claims brought against Northern, and all state law claims (Counts IV-VI). The prior judge also dismissed all Sherman Antitrust Act claims without prejudice. Premium then indicated that the only remaining claims that it was pursuing were the CEA claims brought against Youngdahl, Goldman, and MFS. (4/11/08 Mot. Reinst. 4). Goldman and MFS each moved for summary judgment on the remaining claims pending against them. Premium also filed a motion for leave to conduct additional discovery and motions to strike. On July 30, 2008, we granted MFS's motion for summary judgment in its entirety and we granted Goldman's motion for summary judgment to the extent that the CEA claim brought against Goldman is based on the trades of 30-Year Treasury Bonds. We denied without prejudice Goldman's motion for summary judgment to the extent that the CEA claim is based upon alleged purchases of 30-Year Treasury Futures. We also granted Premium's motion for leave to conduct appropriate discovery and denied without prejudice Premium's motions to strike. Goldman has now filed a renewed motion for summary judgment and the parties have filed motions to strike.
Summary judgment is appropriate when the record, viewed in the light most favorable to the non-moving party, reveals that there is no genuine issue as to any material fact and the moving party is entitled to judgment as a matter of law. Fed. R. Civ. P. 56(c). In seeking a grant of summary judgment, the moving party must identify "those portions of 'the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any,' which it believes demonstrate the absence of a genuine issue of material fact." Celotex Corp. v. Catrett, 477 U.S. 317, 323 (1986)(quoting Fed. R. Civ. P. 56(c)). This initial burden may be satisfied by presenting specific evidence on a particular issue or by pointing out "an absence of evidence to support the non-moving party's case." Id. at 325. Once the movant has met this burden, the non-moving party cannot simply rest on the allegations in the pleadings, but, "by affidavits or as otherwise provided for in [Rule 56], must set forth specific facts showing that there is a genuine issue for trial." Fed. R. Civ. P. 56(e). A "genuine issue" in the context of a motion for summary judgment is not simply a "metaphysical doubt as to the material facts." Matsushita Elec. Indus. Co., Ltd. v. Zenith Radio Corp., 475 U.S. 574, 586 (1986). Rather, a genuine issue of material fact exists when "the evidence is such that a reasonable jury could return a verdict for the nonmoving party." Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986); Insolia v. Philip Morris, Inc., 216 F.3d 596, 599 (7th Cir. 2000). The court must consider the record as a whole, in a light most favorable to the non-moving party, and draw all reasonable inferences that favor the non-moving party. Anderson, 477 U.S. at 255; Bay v. Cassens Transport Co., 212 F.3d 969, 972 (7th Cir. 2000).
I. Goldman's Motion to Strike Reports and Testimony of Glen Donaldson
Goldman moves to strike the reports and testimony of Glen Donaldson (Donaldson), an expert hired by Premium for this case. Pursuant to Rule 702 and the analysis provided in Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993), in order for an expert opinion to be admissible: (1) "the witness must be qualified 'as an expert by knowledge, skill, experience, training, or education,'" (2) "the expert's reasoning or methodology underlying the testimony must be scientifically reliable," and (3) "the testimony must assist the trier of fact to understand the evidence or to determine a fact in issue." Ervin v. Johnson & Johnson, Inc., 492 F.3d 901, 904 (7th Cir. 2007)(quoting in part Fed. R. Evid. 702)(stating that the opinion must be "both relevant and reliable"). In assessing the reliability of an expert opinion, a court must consider factors such as "(1) whether the scientific theory can be or has been tested," "(2) whether the theory has been subjected to peer review and publication," and "(3) whether the theory has been generally accepted in the scientific community." Id.; Sheehan v. Daily Racing Form, Inc., 104 F.3d 940, 942 (7th Cir. 1997)(stating that an "expert's failure to make any adjustment for variables bearing on the decision" at issue and "his equating a simple statistical correlation to a causal relation . . . indicates a failure to exercise the degree of care that a statistician would use in his scientific work, outside of the context of litigation"); People Who Care v. Rockford Bd. of Educ., School Dist. No. 205, 111 F.3d 528, 537-38 (7th Cir. 1997)(stating that "[a] statistical study is not inadmissible merely because it is unable to exclude all possible causal factors other than the one of interest," but "a statistical study that fails to correct for salient explanatory variables, or even to make the most elementary comparisons, has no value as causal explanation and is therefore inadmissible in a federal court"); Sanner v. Board of Trade of City of Chicago, 2001 WL 1155277, at *2-*6 (N.D. Ill. 2001)(noting that the parties agreed that statistical regression and event studies were accepted methods for market studies)(stating also that "[a]s the Seventh Circuit has stated more colorfully, '[t]he principle of Daubert is merely that if an expert witness is to offer an opinion based on science, it must be real science, not junk science'")(quoting in part Tuf Racing Products, Inc. v. American Suzuki Motor Corp., 223 F.3d 585, 591 (7th Cir. 2000)). see also Walker v. Soo Line R. Co., 208 F.3d 581, 589 (7th Cir. 2000)(stating "[t]hat two different experts reach opposing conclusions from the same information does not render their opinions inadmissible").
A. Summary of Donaldson's Reports and Opinions
Donaldson indicates in his November 26, 2008 report (Donaldson's November 2008 Report) that he performed several studies to arrive at his conclusions. Donaldson explains that 30-Year Treasury Futures and 30-Year Treasury Options are derivative instruments, meaning that their values are tied to the price behavior of other underlying instruments. (Don. Nov. R. 4). Donaldson contends that an analysis of the price fluctuations of other instruments such as 30-Year Treasury Bonds and 30-Year Treasury Bond Repurchase Agreements is valuable to understanding fluctuations as to 30-Year Treasury Futures and 30-Year Treasury Options. (Don. Nov. R. 4). Donaldson also contends that an analysis of fluctuations in the prices of closely related instruments such as 10-Year and 5-Year Treasury instruments is valuable in understanding movements in the prices of 30-Year Treasury Futures and 30-Year Treasury Options. (Don. Nov. R. 4).
1. Cancellation/Suspension Comparative Studies
In Donaldson's November 2008 Report, Donaldson compares the price behavior of 30-Year Treasury Bonds around the time period of the cancellation with the price behavior of other treasury instruments around the time of their cancellation. (Don. Nov. R. 6). Donaldson presents a chart to show that there was a significant rise and fall in the closing price of the 30-Year Treasury Bond around the period when the bond was cancelled on October 31, 2001. (Don. Nov. R. 7). Donaldson then presents other charts to show that there was not such a rise and fall for other instruments during time periods when other instruments were cancelled or suspended, such as with the 3-Year Treasury Note that was suspended in 1998 and again in 2007, the 4-Year Treasury Note that was canceled in 1990, and the 7-Year Treasury Note that was cancelled in 1993. (Don. Nov. R. 7-8). Donaldson concludes that since there was no significant rise and fall in the prices for those other related instruments during the period around their cancellation or suspension, it "suggests that something unusual is likely occurring when, and for some time after, it was announced on October 31, 2001, that the 30-Year T-Bond would be cancelled." (Don. Nov. R. 9)(emphasis added).
2. Minute-by-Minute Analyses of October 31 Prices
Donaldson also presents minute-by-minute analyses of the closing prices for 30-Year Treasury Bonds and Futures on October 31, 2001. (Don. Nov. R. 10). Goldman allegedly obtained information concerning the cancellation of the 30-Year Treasury Bond at 9:35 a.m. on October 31, 2001. Donaldson presents a chart showing that starting at 8:00 a.m. on October 31, 2001, the price remained stable until 9:57 a.m. when the official full announcement of the cancellation was made. (Don. Nov. R. 10). The chart shows that the price gradually rose after 9:57 a.m. with the initial peak some time between 10:15 a.m. and 10:20 a.m., and then rose to a higher peak around 1:00 p.m. (Don. Nov. R. 10). Donaldson indicates that it cannot be presumed that the price rise was the result of the cancellation of the 30-Year Treasury Bond. (Don. Nov. R. 11). He contends that the bond market is not instantaneously responsive to such announcements and that "even after the official public release, at 9:57 am, of the news that the 30-Year Treasury Bond would be cancelled, there could have still been private information left in the market that might or might not have been fully reflected in the price. . . ." (Don. Nov. R. 11).
Donaldson then presents charts showing the minute-by-minute prices for Treasury Call Options and 30-Year Treasury Put Options on the morning of October 31, 2001. Donaldson utilizes certain calculations to compare the price fluctuations during the 8-minute window between 9:35 a.m. and 9:43 a.m. (8-Minute Window) when Goldman allegedly possessed special information. (Don. Nov. R. 13). Donaldson also evaluates the 22-minute window between 9:35 a.m. and 9:57 a.m. Donaldson concludes that certain price movements were "unusual." (Don. Nov. R. 13-14). Donaldson concludes that "it is difficult to rule out the possibility that Goldman's behavior contributed to the upward move in market prices" and that, based on Goldman's percentage of the trading at the time, "it seems possible that Goldman's behavior may have had the effect of moving market prices for on-the-run 30-yr T-Bonds away from their 'normal' values." (Don. Nov. R. 14).
3. Event Study Tests - Regression Analyses
Donaldson also performed certain event study tests, using regression analyses to assess whether Goldman's trading during the 8-Minute Window affected prices. Donaldson presents certain charts to show that Goldman began purchasing sizable amounts of 30-Year Treasury Bonds and Futures positions starting at about 9:35 a.m. (Don. Nov. R. 17-18). Donaldson explains how he prepared a multiple regression model using a mathematical formula to assess whether price fluctuations were "abnormal." (Don. Nov. R. 15, 20-22). In the model, Donaldson assigned certain probability values to variables and he uses a "dummy variable" for what he deemed the effect at 9:57 a.m., when the official Treasury announcement was made (Announcement Variable). (Don. Nov. R. 21). Donaldson also uses a "dummy variable" for what Donaldson deems was the "effects of Goldman" during the "event window" (Goldman Variable). (Don. Nov. R. 21). Donaldson concludes after his calculations that the "Goldman Variable does indeed help explain 30-Year Treasury futures Prices." (Don. Nov. R. 21). Donaldson contends his finding is "robust." (Don. Nov. R. 25). Finally, Donaldson presents some other models to show that Goldman's activities may have affected the prices in question. (Don. Nov. R. 26-30).
4. Donaldson's December 2008 Report
Donaldson also prepared a report dated December 16, 2008 (Donaldson's December 2008 Report). In Donaldson's December 2008 Report he explains why he believes that the conclusions of Goldman's expert, Bradford Cornell (Cornell), are incorrect. Donaldson contends that Cornell's conclusions are contrary to the data and are not based on proper formulas or procedures. (Don. Dec. R. 2). Donaldson goes into great detail explaining why he believes that Cornell did not apply the correct formulas and that Cornell's conclusions are not correct. For example, Donaldson contends that Cornell's statement that prices began to rise after 9:43 a.m. when Goldman was "either absent from the market or largely a net seller in the relevant securities," is contrary to Cornell's own data. (Don. Dec. R. 3). Donaldson contends that Cornell's data shows that Goldman was in fact a relevant buyer during that time period. (Don. Dec. R. 3-4). Donaldson also disagrees with Cornell's conclusion that Goldman was a small trader in the market, a conclusion that Cornell reached based on his findings that Goldman only traded approximately two thousand instruments during the 8-Minute Window, as compared to the more than half a million instruments traded on that same date of October 31, 2001. (Don. Dec. R. 6). Donaldson contends that the proper comparison should have been on the percentage of trades executed by Goldman during the 8-Minute Window causing price rises during that period and that it is not appropriate to compare Goldman's trades during the 8-Minute Window with the total trades that were made on the market during the entire day. (Don. Dec. R. 6). Donaldson contends that his calculations show that Goldman performed over of the market trades "in the on-the-run 30yr Treasury Bond between 9:35am and 9:43am. . . ." (Don. Dec. R. 6). Donaldson also contends that Cornell improperly concludes that prices are mainly influenced by "widespread macroeconomic forces," such as the cancellation of the 30-Year Treasury Bond. (Don. Dec. R. 6). Donaldson argues that prices can be influenced by individual traders and cites to a study published in a financial journal to support his position. (Don. Dec. R. 7). Donaldson also contends that one article co-authored by Cornell himself recognizes that the "buying of stock by insider traders can push up stock prices." (Don. Dec. R. 9). Donaldson also explains that he used the data from Cornell's November 2008 Report to make new calculations using Donaldson's formulas and that Donaldson's positions actually are stronger using Cornell's own data. (Don. Dec. R. 10). Donaldson also argues that Cornell improperly selected an alternative hypothesis for his hypothesis test and improperly used a "two-tailed test" for his hypothesis test. (Don. Dec. R. 16). Donaldson contends that Cornell should have used a "one tailed approach" for the hypothesis test. (Don. Dec. R. 176). Donaldson also argues that Cornell focused on the wrong data and "did not appropriately estimate the test statistics themselves. . . ." (Don. Dec. R. 16-23).
B. Donaldson's Regression Analyses
Goldman argues that Donaldson's regression analyses are unreliable and biased and that the court should strike Donaldson's reports and testimony relating to such analyses.
1. Lack of Analysis of Options
Goldman contends that Donaldson's regression analyses are flawed since Donaldson did not evaluate 30-Year Options even though Premium's sole trading at issue in this case involved such options. Premium, however, points to Donaldson's explanation that an analysis of bonds and futures is instructive for this case since the prices of bonds and futures directly affected the options prices at issue in this case. (Ans. Str. Don. 17). Premium also asserts that Donaldson "analyzed the specification contracts for 30yr CBOT options and futures and observed that the features of these contracts made it likely that any artificiality in the futures would be translated into the options." (Ans. Str. Don. 17)(Ans. Str. Don. 18 n. 13). Premium has provided sufficient justification for why Donaldson did not specifically analyze options. Thus, Goldman has not shown Donaldson's opinions and reports to be inadmissible simply because Donaldson did not specifically analyze options.
Goldman contends that Donaldson's regression analyses are also flawed because Donaldson did not use the actual trading data. Goldman argues that when such actual trading data is available there is no reason to construct dummy variables as Donaldson did. However, even if Goldman is correct that the use of the actual trading data would be a better way to analyze the prices, that does not necessarily mean that such is the only way to statistically analyze price fluctuations. It does not mean that regression analyses employing dummy variables, such as the methods utilized by Donaldson, can be of no assistance to the trier of fact. Premium, in fact, contends that the regression analyses Donaldson employed are standard in financial market cases and the industry, and that Donaldson used such analyses to find statistical significance in the price changes. (Ans. Str. Don. 9-10). Premium also points out how Donaldson, in forming his opinions, did take into consideration the trading summaries provided by Goldman. (Ans. Str. Don. 5). Goldman also takes issue with an academic source relied upon by Donaldson, contending that Donaldson misread and distorted the source. (Str. Don. 9 n.6). Goldman can question Donaldson on cross examination at trial, if he testifies, about that academic source or Goldman could offer testimony by Goldman's own expert on that point. However, a difference in opinions does not necessarily show that Donaldson's opinions are unreliable and inadmissible. Thus, Goldman has not shown that Donaldson's opinions and reports are inadmissible due to the alleged fact that Donaldson failed to rely on actual trading data.
3. Impact and Length of Impact
Goldman argues that Donaldson's regression analyses are also unreliable because the analyses fail to take into consideration how long it took Goldman's actions to impact the market and how long the impact lasted. Goldman's arguments however, merely offer opinions as to how Donaldson's analyses could, in Goldman's opinion, have been better constructed. Goldman has not shown that the analyses performed by Donaldson are inherently unreliable in light of such arguments. As indicated above, while Goldman can attempt to rebut Donaldson's opinions at trial, the reasons advanced by Goldman do not provide a basis for striking Donaldson's opinions or reports.
4. Formulation of Dummy Variables
Goldman argues that Donaldson's regression analyses also are unreliable and biased because Donaldson did not properly formulate the dummy variables.
Goldman offers extensive arguments why it would have formulated the variables differently, thus making for better regression analyses. Goldman also contends that its own expert has found that Donaldson's dummy variables are improperly formulated. Premium, in response, has offered ample justifications for Donaldson's determinations and Goldman has not shown that the dummy variables are so defined as to render Donaldson's regression analyses unreliable. Goldman also argues that Donaldson's regression analyses are unreliable since they fail to "differentiate between the impact of the allegedly unlawful and lawful conduct." (Str. Don. 14). There remains genuinely disputed facts on this issue that should be considered by the trier of fact. Therefore, as to the formulation of dummy variables, Goldman's motion to strike Donaldson's opinions is denied.
Goldman argues that Donaldson's regression analyses also are unreliable because they employ a one-tail test. Premium in turn offers justifications showing why, in Donaldson's opinion, the two-tail test proposed by Goldman would be inappropriate in this case and that the one-tail test is the proper method for the analyses. (Ans. Str. Don. 21). The difference of opinion by Goldman on this ...