Summary
holding that regression analysis using proxy was not a fit issue at the Daubert stage
Summary of this case from In re Avandia Mktg.Opinion
No. 08–md–2002.
01-26-2015
In re PROCESSED EGG PRODUCTS ANTITRUST LITIGATION. This Document Applies to: All Direct Purchaser Actions.
Processed Egg Products Antitrust Litigation, pro se.
Processed Egg Products Antitrust Litigation, pro se.
MEMORANDUM
GENE E.K. PRATTER, District Judge.
Several of the nation's largest egg producers allegedly conspired to control and limit the supply of eggs and egg products, resulting in artificially inflated prices during the period of 2000–2008. Direct Purchaser Plaintiffs, a putative class of entities and individuals that purchased eggs or egg products directly from Defendants, are seeking class certification and have offered the testimony of Dr. Gordon Rausser, who holds a Ph.D. degree in economics, in support of their Motion for Class Certification. Defendants have filed a motion seeking to exclude entirely Dr. Rausser's declaration, opinions, and testimony. Although there are a number of challenging issues that bear critical analysis and certainly merited the defense and the Court's attention, the Court ultimately concludes that the Motion to Exclude should be denied.
I. Preliminary Legal Background
a. The Daubert Standard and Factors
Federal Rule of Evidence 702 governs the admissibility of expert testimony and provides:
A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if:
(a) the expert's scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue;
(b) the testimony is based on sufficient facts or data;
(c) the testimony is the product of reliable principles and methods; and
(d) the expert has reliably applied the principles and methods to the facts of the case.The Supreme Court, in Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), imposed a gatekeeper role upon district courts by charging them to “ensure that any and all scientific evidence is not only relevant, but reliable.” Daubert, 509 U.S. at 589, 113 S.Ct. 2786. When “faced with a proffer of expert scientific testimony ... the trial judge must determine at the outset, pursuant to Rule 104(a), whether the expert is proposing to testify to (1) scientific knowledge that (2) will assist the trier of fact to understand and determine a fact in issue.” Id. at 592, 113 S.Ct. 2786. This gatekeeping function of the district court extends beyond scientific testimony to “testimony based on ‘technical’ and ‘other specialized’ knowledge.” Kumho Tire Co. v. Carmichael, 526 U.S. 137, 141, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999).
Federal Rule of Evidence 702 provides “three distinct substantive restrictions on the admission of expert testimony: qualifications, reliability and fit.” Elcock v. Kmart Corp., 233 F.3d 734, 741 (3d Cir.2000). The party offering the expert testimony has the burden of establishing that the proffered testimony meets each of the three requirements by a preponderance of the evidence. See Padillas v. Stork–Gamco, Inc., 186 F.3d 412, 418 (3d Cir.1999).
b. Daubert at the Class Certification Stage
A threshold question is whether, and to what extent, Daubert applies at the class certification stage. Although there is no definitive Third Circuit precedent on point, the general consensus appears to be that the Court should subject expert witnesses to Daubert scrutiny at the class certification stage of the litigation. See, e.g., Romero v. Allstate Ins. Co., No. 01–3894, 52 F.Supp.3d 715, 720–24, 2014 WL 4966147, at *3–5 (E.D.Pa. Oct. 6, 2014) (applying Daubert to expert testimony at the class certification stage); In re Chocolate Confectionary Antitrust Litig., 289 F.R.D. 200, 207–08 (M.D.Pa.2012) (“Despite the paucity of relevant precedent in the Third Circuit and the discordant views percolating in the circuits, the court finds that a thorough Daubert analysis is appropriate at the class certification stage ....”); In re Flonase Antitrust Litig., 284 F.R.D. 207, 235 (E.D.Pa.2012) (suggesting that expert testimony must satisfy Daubert at the class certification stage); McLaughlin on Class Actions § 3:14 (11th ed.) (“The way courts apply Daubert in the class certification context has evolved toward near universal acceptance that the requirements of Daubert and Rule 702 apply with full force at the class certification stage.”). The Supreme Court has suggested as much in dicta, see Wal–Mart Stores, Inc. v. Dukes, ––– U.S. ––––, 131 S.Ct. 2541, 2553–54, 180 L.Ed.2d 374 (2011) (“The District Court concluded that Daubert did not apply to expert testimony at the certification stage of class-action proceedings. We doubt that is so ....” (citation omitted)), and the other circuit courts appear to agree that Daubert applies to expert witnesses at class certification, see Newberg on Class Actions § 7:24 (5th ed.) (discussing the precedents in various circuits and concluding that all but the Sixth and Tenth Circuits endorse engaging in some form of Daubert inquiry at the class certification stage). The parties have not disputed this approach in their briefing.
There are two potential complications, however, relating to the scope of the Daubert inquiry. The first potential complication is that the question might arise as to whether the Daubert analysis is limited to expert testimony relating to class certification, meaning the analysis does not extend to expert testimony regarding the merits. See id. (“The problem courts have confronted is that in many cases an expert who will testify at trial also testifies as to one of the prongs of the class certification inquiry.... Because the expert is testifying as to a prong of the class certification standard, the Court is tempted to ensure the testimony meets the Daubert test and must determine whether it is convincing as to the certification standard, yet because the same expert will be proffered at trial for a related point, a court is tempted to not prejudge the testimony before the discovery phase of the lawsuit enables a full development of the case's facts and of the expert testimony.”). This is a particularly unsettled and confounding issue which the Court does not necessarily need to address, as Dr. Rausser's entire testimony could well be relevant in some form at class certification (he is, after all, Direct Purchaser Plaintiffs' “class certification expert,” see Case Management Order No. 21 at 4). But, to the extent this question arises, the Court's view is that it is wiser and more useful to err on the side of a more rigorous Daubert inquiry. That is, so long as the testimony could plausibly be relevant to the class certification analysis, the Court will conduct a full Daubert inquiry on that testimony. After all, the class certification analysis will “frequently entail ‘overlap with the merits of the plaintiff's underlying claim,’ ” but this overlap does not relieve a court of its duty to conduct a full, rigorous certification analysis. See Comcast Corp. v. Behrend, ––– U.S. ––––, 133 S.Ct. 1426, 1432, 185 L.Ed.2d 515 (2013) (quoting Wal–Mart, 131 S.Ct. at 2551 ). Similarly, an expert's testimony will often be relevant both to class certification and to the underlying merits should the case proceed to trial. This overlap does not, however, relieve a court of its duty to conduct a full, rigorous analysis of the testimony for class certification purposes. A significant step in this analysis is ensuring that the Daubert standard is met.
The second potential complication is that this Daubert inquiry might overlap significantly with the questions to be resolved at the class certification hearing, which might invite the notion that the issue is better resolved at the more intensive class certification hearing than at the Daubert hearing. The Court embraces the standard of Daubert: “Proponents of expert testimony do not ‘have to prove their case twice—they do not have to demonstrate to the judge by a preponderance of the evidence that the assessments of their experts are correct, they only have to demonstrate by a preponderance of evidence that their opinions are reliable.’ ” In re DVI, Inc. Sec. Litig., No. 03–5336, 2014 WL 4634301, at *5 (E.D.Pa. Sept. 15, 2014) (quoting In re Paoli R.R. Yard PCB Litigation, 35 F.3d 717, 744 (3d Cir.1994) (emphasis in original)).
To be sure, the Court recognizes that in the class certification context, the line between Daubert and the ultimate issues might prove somewhat illusory. That is because the reliability of the means of proving classwide impact frequently factors into the predominance determination in antitrust class actions. See, e.g., In re Rail Freight Fuel Surcharge Antitrust Litig., 725 F.3d 244, 252–53 (D.C.Cir.2013) (“Common questions of fact cannot predominate where there exists no reliable means of proving classwide injury in fact.” (emphasis added)). Similarly, the Daubert analysis requires the Court to determine whether the expert testimony is reliable. See Daubert, 509 U.S. at 589, 113 S.Ct. 2786 (“[T]he trial judge must ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable.”). How, then, should the Court rule upon this Daubert motion without deciding the issues tied in with class certification? For one, the focus of the two inquiries is subtly different. Daubert “focuses on principles and methodology, not on the conclusions generated by principles and methodology.” In re TMI Litig., 193 F.3d 613, 670 (3d Cir.1999). At class certification, on the other hand, the Court will rule upon the conclusions generated by the principles and methodology. Further, the issues of class certification require the Court to consider questions beyond the reliability of Dr. Rausser's testimony, such as whether further evidence can support the notion that common issues predominate over individual issues. That is, the Court need not find that Dr. Rausser's methods are, by themselves, sufficient to show, say, a common impact or that there is a reliable means of proving damages on a classwide basis—only that his methods are reliable and useful to the questions to be addressed at class certification. Therefore, though some extent of overlap is inevitable, deeming Dr. Rausser's expert testimony admissible under Daubert does not preclude the Court from denying class certification (an issue that the Court has not yet addressed).
This is consistent with the language of the Third Circuit Court of Appeals in In re Hydrogen Peroxide Antitrust Litigation, 552 F.3d 305, 323 (3d Cir.2008), in which the court said, “It follows that opinion testimony should not be uncritically accepted as establishing a Rule 23 requirement merely because the court holds the testimony should not be excluded, under Daubert or for any other reason.” This quote implies not only that Daubert analysis is appropriate at the class certification stage, see Behrend v. Comcast Corp., 655 F.3d 182, 215 n. 18 (3d Cir.2011) (Jordan, J. concurring in the judgment and dissenting), rev'd on other grounds ––– U.S. ––––, 133 S.Ct. 1426, 185 L.Ed.2d 515 (2013), but also that the Court can refuse to exclude testimony under Daubert but nonetheless reject the import of that testimony at class certification.
c. A Quick Word about Credibility
The Daubert hearing involved a somewhat odd presentation of the testimony of Dr. Rausser. Dr. Rausser's credibility has been questioned in the ongoing Rail Freight litigation because of his ties to a company that is involved with investments in potential class action claims, including, it seems, claims in that case. This credibility issue led to some preliminary discussion among the parties and with the Court as to whether Dr. Rausser's credibility was a proper issue at the Daubert stage. Although Defendants hardly mentioned Dr. Rausser's recent credibility issues in their briefing, see Reply in Supp. of Mot. to Exclude 2 (Doc. No. 1101) (mentioning only that “new evidence that Dr. Rausser has been a stakeholder in, and long-time consultant to, companies that depend on recoveries by certified classes underscores the importance of the prohibition of experts exploiting their credentials as vehicles for factual narrative,” (quotation marks and citation omitted)), at the hearing Plaintiffs did call Dr. Rausser “to address briefly the ‘credibility’ issue Defendants have raised,” see December 8, 2014 Fax from Steven A. Asher, Esq.
The parties dispute the extent to which credibility is relevant during a Daubert inquiry. The Court follows the standard set out in Elcock:
We do not hold, however, that a district court can never consider an experts witness's credibility in assessing the reliability of that expert's methodology under Rule 702. Such a general prohibition would be foreclosed by the language of
Rule 104(a), which delineates the district court's fact-finding responsibilities in the context of an in limine hearing on the Daubert reliability issue. Indeed, consider a case in which an expert witness, during a Daubert hearing, claims to have looked at the key data that informed his proffered methodology, while the opponent offers testimony suggesting that the expert had not in fact conducted such an examination. Under such a scenario, a district court would necessarily have to address and resolve the credibility issue raised by the conflicting testimony in order to arrive at a conclusion regarding the reliability of the methodology at issue. We therefore recognize that, under certain circumstances, a district court, in order to discharge its fact-finding responsibility under Rule 104(a), may need to evaluate an expert's general credibility as part of the Rule 702 reliability inquiry.
...
[A]n expert's prior dishonesty or misconduct should not qualify as an appropriate factor in assessing methodological reliability when the acts are wholly unrelated to the expert's use of a particular methodology, but that a court should take such dishonesty or misconduct into account when the nexus between the acts and the expert's methodology is more direct—e.g., when the prior dishonest acts involve fraud committed in connection with the earlier phases of a research project that serves as the foundation for the expert's proffered opinion. Under this approach, for instance, the fact that an expert witness falsely reported his salary on an income tax return has little if any bearing on the reliability of a diagnostic test he frequently employs, but the fact that the expert lied about whether his methodology had been subjected to peer review, or intentionally understated the test's known rates of error, is a different matter entirely.
233 F.3d at 751 n. 8 (citation omitted).
Hence, the Court now turns to the substance of the Motion to Exclude Dr. Rausser.
II. Issues Raised in Daubert Motion
The issues the Court must consider in a Daubert hearing are qualifications, reliability, and fit. The first of these considerations—qualifications—was not raised in the initial Motion to Exclude Dr. Rausser. To be “qualified” to render expert testimony under Daubert, Dr. Rausser must merely “possess specialized expertise,” and this requirement is interpreted “liberally.” Pineda v. Ford Motor Co., 520 F.3d 237, 244 (3d Cir.2008) (citation omitted). In their Memorandum in Opposition to Defendants' Motion to Exclude (Doc. No. 1058), Direct Purchaser Plaintiffs detail Dr. Rausser's qualifications and experience as recounted in his curriculum vita. The Court concludes that Direct Purchaser Plaintiffs have demonstrated by a preponderance of the evidence that Dr. Rausser is qualified in the field of economics.
a. Whether Dr. Rausser Properly Considered and Discussed the Factual Record in Sections IV and V of his Declaration
The first challenge raised by Defendants to the reliability of Dr. Rausser's expert testimony is that Dr. Rausser's testimony relies not on his economic expertise but merely upon his cursory consideration of the evidence. That is, in his analysis of whether the “Egg Industry is Conducive to Price Manipulation through Output Restriction” (Section IV) and whether “Conspiracy Period Behavior and Pricing is Consistent with Collusion” (Section V), Dr. Rausser does not express opinions involving the application of his “scientific, technical, or specialized knowledge. Instead, Dr. Rausser merely lends his expert credentials in an effort to bolster Plaintiffs' legal advocacy by providing a narrative based on these documents most to Plaintiffs' liking.” Mem. in Supp. of Mot. to Exclude 4–5 (Doc. No. 1032). Such testimony, Defendants argue, is neither sufficiently based in specialized knowledge to qualify as “expert testimony” nor helpful to the finder of fact. Defendants also argue that Dr. Rausser's testimony is flawed because he did not analyze certain pieces of data, such as publicly available data to determine whether supply was reduced. See id. at 7.
By way of example, in their Reply, Defendants point to the following aspects of Dr. Rausser's testimony as offering improper interpretations of the record:
• Dr. Rausser's conclusion that the UEP animal welfare guidelines were not developed with animal welfare as their objective and that “the prospect of higher egg prices through supply reductions” was “an incentive for producers to participate” in the program. See Rausser Decl. 67–68 (Doc. No. 978–2).
• Dr. Rausser's statements in his Declaration that UEP's influence was “pervasive” and UEP certification standards “became de facto requirements for doing business in the egg industry,” among other similar conclusions about UEP's influence. See id. at 25, 32, 40, 43, 47, 58.
• Dr. Rausser's statements in his Declaration that “the overwhelming evidence from the record is that UEP explicitly told producers not to build more cages,” and that the Defendants “understood that the increased cage space requirements would lead to reduced supply absent any new capital investment in the construction of cages.” Rausser Reply Decl. 10, 46 (Doc. No. 1059).
Plaintiffs counter that Dr. Rausser's use of the factual record is wholly appropriate and necessary for him to offer his expert opinion. They claim that this testimony about the facts is necessary for the development and testing of a hypothesis. They further argue that Dr. Rausser's opinions are based on his economic expertise, particularly because “whether certain conduct is consistent or inconsistent with one's economic self-interest, and whether it is consistent or inconsistent with collusion, are proper subjects on which a professional economist may opine.” Mem. in Opp. of Mot. to Exclude 5 (Doc. No. 1058). Finally, Plaintiffs argue that whether Dr. Rausser “should have considered additional documents or testimony goes to the weight, not the admissibility, of his testimony.” Id. at 6.
Federal Rule of Evidence 704 provides that expert testimony is not objectionable merely because it embraces the ultimate issue in a case. However, expert testimony must still meet the requirements of Rule 702. Under Rule 702, an expert's testimony may be excluded if it falls outside his or her area of expertise or if it is unhelpful to the jury. See Fed.R.Evid. 702 (“A witness who is qualified as an expert ... may testify in the form of an opinion or otherwise if: (a) the expert's scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue ....”). However, “expert testimony that usurps the role of either the jury or the courts is not admissible.” Romero, 52 F.Supp.3d at 720–24, 2014 WL 4966147, at *3–5 ; see also Berckeley Inv. Group, Ltd. v. Colkitt, 455 F.3d 195, 217 (3d Cir.2006) (“[A]n expert witness is prohibited from rendering a legal opinion. Such testimony is prohibited because it would usurp the District Court's pivotal role in explaining the law to the jury.” (citation omitted)).
Defendants cite to several cases in support of their position that an expert cannot opine on whether the factual record provides evidence of collusion. One such citation is to the Eleventh Circuit's City of Tuscaloosa v. Harcros Chemicals, Inc., 158 F.3d 548 (11th Cir.1998). In Tuscaloosa, the Eleventh Circuit reviewed the district court's rulings on various aspects of a case in which a set of chemical companies were alleged to have colluded in violation of the antitrust laws. Id. at 553. Most pertinent to the present case is the Eleventh Circuit's affirmation of the district court's decision to exclude a statistician's “characterizations of documentary evidence as reflective of collusion ... because the trier of fact is entirely capable of determining whether or not to draw such conclusions without any technical assistance from McClave or other experts.” Id. at 565. Plaintiffs argue that this opinion is distinguishable because Dr. Rausser is an economist, not a statistician like the expert in Tuscaloosa. Defendants, however, counter that the Eleventh Circuit's opinion applies to expert testimony on evidence of collusions generally. In a footnote, the Eleventh Circuit does consider in passing dicta the testimony of an economist and concludes that the economist's “characterizations of pieces of documentary evidence as tending to show collusion” are “for the court to make at summary judgment.” Id. at 567 n. 27.
There is little Third Circuit precedent on point. The closest scenario to the present one came in a pre-Daubert opinion that was later reversed on other grounds. The Third Circuit analyzed a similar expert report and concluded:
The trial court found that DePodwin did not use economic expertise in reaching the opinion that the defendants participated in a Japanese television cartel.... As a result, the court also held the opinions to be unhelpful to the factfinder. What the court in effect did was to eliminate all parts of the report in which the expert economist, after describing the conditions in the respective markets, the opportunities for collusion, the evidence pointing to collusion, the terms of certain undisputed agreements, and the market behavior, expressed the opinion that there was concert of action consistent with plaintiffs' conspiracy theory. Considering the complexity of the economic issues involved, it simply cannot be said that such an opinion would not help the trier of fact to understand the evidence or determine that fact in issue.
In re Japanese Elec. Products Antitrust Litig., 723 F.2d 238, 280 (3d Cir.1983), rev'd on other grounds, sub nom. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986).
However, the Court does not find the Eleventh Circuit's passing dicta in Tuscaloosa as outlining a convincing approach that would control the outcome here. The more convincing and widely followed approach allows economic experts to testify about whether certain conduct is indicative of collusion. See, e.g., In re Titanium Dioxide Antitrust Litig., No. 10–0318, 2013 WL 1855980 (D.Md. May 1, 2013). In Titanium Dioxide, the court considered whether to admit the testimony of various proposed experts. Like the Eleventh Circuit in Tuscaloosa, the court excluded the testimony of an non-economist expert who sought to testify as to whether the “conduct of the Defendants was more consistent with collusion than with competition.” Id. at *7. However, the court did permit the economist experts to testify on precisely this issue. See id. at *4 (stating that “Courts regularly admit expert testimony regarding whether conduct is indicative of collusion,” and distinguishing Tuscaloosa as “hing[ing] not on a violation of Rule 704 but on a determination that the expert's assertions were outside his competence as a statistician.” (quotation marks and citations omitted)).
The parties each argue that Judge Easterbrook's decision in Mid–State Fertilizer Co. v. Exchange National Bank, 877 F.2d 1333 (7th Cir.1989), supports their position. In Mid–State Fertilizer, an economist submitted a brief affidavit (no more than nine sentences) setting out legal conclusions, such as, “In my opinion the handling of the loan arrangement and use of the locked box or blocked account ... was not an appropriate traditional banking practice.” Id. at 1339. The Seventh Circuit ruled that this affidavit was inadmissible. Defendants cite the case for Judge Easterbrook's admonishment of the expert for having merely “examined materials produced in discovery and [drawn] inferences from the record, speaking in legal rather than economic terms.” Id. at 1340. Plaintiffs counter that the case turned on the expert's lack of analysis of the factual record—the fact that the expert “presented nothing but conclusions—no facts, no hint of an inferential process, no discussion of hypotheses considered and rejected.” Id. at 1339.
Defendants also cite the Northern District of Illinois case of JamSports and Entertainment, LLC v. Paradama Productions, Inc., No. 02–2298, 2005 WL 14917, at *10 (N.D.Ill. Jan. 3, 2005). In JamSports, the Court excluded portions of an economist's report, finding that:
Baade's discussion of the question of anticompetitive conduct consists, in large part, of his interpretation of correspondence and other evidence. Jamsports has failed to persuade the Court that this is proper testimony by an expert in economics. There is nothing in Baade's expertise that suggests that he is any more competent than the average juror in interpreting these communications or in divining from them the intent of Clear Channel. Such insights, as Judge Easterbrook noted in a similar context, “are no part of the economist's armamentarium.” Mid–State Fertilizer Co. v. Exchange Nat'l Bank of Chicago, 877 F.2d 1333, 1340 (7th Cir.1989). In conducting this analysis, Baade has simply “examined materials produced in discovery and [drawn] inferences from the record.” Id. Such testimony, in this context at least, does not constitute the application of “scientific, technical, or other specialized knowledge [that] will assist the trier of fact.” Fed.R.Evid. 702.
JamSports, 2005 WL 14917, at *10. However, JamSports is not as conclusive as Defendants suggest, as the Court did find the economist competent to testify as to findings that “have a grounding in economics” such as statements regarding the effect of certain acts on competition and the well-being of consumers. Id.
While these cases may be tension with one another, they are not necessarily diametrically opposed. The cases all appear to agree that an economist is capable of testifying as to whether an industry's market structure makes it particularly susceptible to collusion. On the flip side, the cases are clear that an economist's testimony is not admissible where he or she simply reads and interprets evidence of collusion as any juror might, or where an economist infers intent to collude from mere documentary evidence, unrelated to his or her economic expertise. The Court balances these principles and examines Dr. Rausser's report to determine whether his conclusions are based on his expertise as an economist, or merely on his lay examination of the evidence. See, e.g., In re Urethane Antitrust Litig., No. 04–1616, 2012 WL 6681783, at *3 (D.Kan. Dec. 21, 2012) (permitting an expert to testify that “particular events, assuming they occurred, are consistent with a conspiracy. Certainly, he may not give any opinion concerning the credibility of witnesses or whether a particular event actually occurred.... [T]he Court concludes that it would be helpful to a jury for an expert to put events in an economic context.”).
Here, Section IV of Dr. Rausser's Report is titled “The U.S. Egg Industry is Conducive to Price Manipulation through Output Restriction.” This Section of his report is admissible. The field of economics contains a field of study on the features of markets that make them particularly conducive to collusion, and Dr. Rausser appears to be drawing from that field in his report. He states that:
Economists consider the following industry characteristics that make it easier for producers to manipulate prices and therefore make the occurrence of collusive price fixing both rational and more likely to have occurred: (a) the participants control a large share of the market and therefore do not risk being seriously undermined by non-participant competitors (e.g. other egg producers who might increase their own output, thus capturing foregone sales and driving down prices); (b) barriers to entry are high and/or participants have the ability to incentivize any would-be competitors to join the conspiracy; (c) the demand for the product is inelastic, so that price increases will not result in commensurate volume losses which would make the manipulation unprofitable; (d) there are methods of communication and coordination available to facilitate the scheme (e.g. participating in trade associations); and (e) participants have the ability to monitor and instill discipline, through punishment or chastisement of those who deviate from the collusive agreement. All of these characteristics are presents in the U.S. egg industry and would have made it feasible and rational for the Defendants to have reduced supply on a market-wide basis, as alleged, without Class members having the ability to escape the resulting price elevation.
Rausser Decl. 10–11.
Dr. Rausser then continues to examine these characteristics of the egg market in detail, examining whether substitutes to eggs exist, what the geographic scope of the eggs market is, what the interchangeability among certain types of eggs and eggs products is, the market share of the largest firms, barriers to entry, whether the demand for eggs is inelastic, and whether there exist methods of communication, coordination, and monitoring. These subjects are appropriate for an economist to opine on, given that economists regularly study the conditions that give rise to collusion. Further, this testimony would be helpful to the finder of fact, since the conditions of a market that make collusion possible are outside the layperson's body of knowledge and expertise. Defendants have not pointed to a case holding that an economist is unable to opine on these subjects and, as Plaintiffs point out, an economist not only should, but must, examine the factual record to arrive at his opinion.
See generally, e.g., Robert C. Marshall and Leslie M. Marx, The Economics of Collusion (2012).
Section V of Dr. Rausser's Declaration is a closer question. Nevertheless, the Court concludes that Section V is admissible. Section V is titled “Conspiracy Period Behavior and Pricing is Consistent with Collusion.” In this Section, Dr. Rausser goes through the evidence of collusion in the record. When read in isolation, it certainly appears that Dr. Rausser's conclusions are based on nothing more than his reading the record and finding evidence in support of Plaintiffs' position. But this is too narrow a reading of Section V. In Section IV, Dr. Rausser went through the factors that would make a market conducive to collusion. In Section V, Dr. Rausser is applying those principles to the factual record and rendering an opinion—Dr. Rausser's expertise is in knowing what evidence to look for. Further, Dr. Rausser is, as experts regularly do, relating his factual findings so that in subsequent sections of his report, he can test whether the data are consistent with that theory of the case.
Dr. Rausser in Section IV described the opportunities for collusion and in Section V seeks to provide his opinion on whether the evidence is consistent with the conditions of collusion. In Section V.A, Dr. Rausser begins by declaring that the “discovery record in this litigation is rich with evidence of coordinated action by Defendants through supply management programs.” Rausser Decl. at 39. He then discusses the history of the UEP Animal–Welfare Certified Program “and its role in reducing supply.” He observes that “Defendants have acknowledged that the UEP Animal Welfare Program was a de facto supply adjustment program and that it led to higher profits.” Id. at 44. He cites emails and newsletters for this proposition. He discusses how the audits and compliance programs of UEP were used to ensure the animal welfare guidelines were met.
This Subsection is admissible. Dr. Rausser in this Subsection is tying the evidence of the case to the economic theory of collusion as explained in Section IV. In particular, he is describing evidence consistent with “(c) the demand for the product is inelastic, so that price increases will not result in commensurate volume losses which would make the manipulation unprofitable; (d) there are methods of communication and coordination available to facilitate the scheme (e.g. participating in trade associations); and (e) participants have the ability to monitor and instill discipline, through punishment or chastisement of those who deviate from the collusive agreement.” Id. at 10–11. His economic expertise allows him to opine on whether the evidence supports that these conditions for collusion were being met. Here, he is attempting to demonstrate that Defendants were operating on the assumption that the demand for eggs was inelastic (e.g. he notes that “Defendants have acknowledged that the UEP Animal Welfare Program was a de facto supply adjustment program and that it led to higher profits ,” Id. at 44 (emphasis added)), that Defendants were communicating and coordinating in implementing the scheme (e.g. he notes that “UEP members were made aware of the effects the Animal Welfare Program had on supply,” Id. at 45), and that Defendants had enforcement mechanisms in place (e.g. he notes that “UEP implemented a system of audits and monthly compliance reports required of all Certified members.”).
In Section V.B, Dr. Rausser posits that “documents indicate that USEM did actively seek opportunities to export large orders of eggs for the express purpose of increasing domestic egg prices.”Id. at 51. He then goes through “evidence of coordinated export prices below cost.” He analyzes communications within the egg industry, such as a letter from United States Egg Marketers to its members that states, “The recent export of 90 containers is now completed and we believe it accomplished its goal of improving domestic prices at a time when shell egg producers needed it so desperately.” Id. at 52 (citing CM00245529–530, at 528). He then discusses the “Timeline of Supply Adjustment Programs,” again relying primarily on communications and statements from Defendants' representatives. Id. at 58. Finally, he opines that “Industry Reports and the Discovery Record Confirm the Causes of Price Increases During the Conspiracy Period.” Id. at 66. He then goes through a series of statements as evidence for this view.
Subsection V.B is again offered as evidence consistent with Dr. Rausser's findings that the eggs market was conducive to collusion. In particular, this Subsection supports the argument that collusion is more likely to arise when “the demand for the product is inelastic, so that price increases will not result in commensurate volume losses which would make the manipulation unprofitable.” Id. at 10. Dr. Rausser is opining on whether the evidence supports this condition, and he finds that it does because the documentary record suggests that Defendants, at the very least, operated on the assumption that a reduction in supply would lead to an increase in profits. This analysis is appropriate for an economist to offer and is, therefore, admissible.
In Subsection V.C, Dr. Rausser discusses whether the actions of Defendants can be explained by rational, self-interested behavior absent collusion. Here, Dr. Rausser reviews the evidence and argues that the animal welfare program would not have been in Defendants' self-interest unless implemented in a collusive manner. His primary support is that the “adoption of the [animal welfare] program was inconsistent with [Defendants'] unilateral self-interest because egg producers themselves were unsure if many consumers were demanding such changes. Further, the measures they adopted were not even considered by UEP's own Scientific Committee or animal welfare groups to be in the best interests of improved animal husbandry ....” Rausser Decl., 67–68. He also opines that the exports below cost were not in Defendants' best interests. See id. at 73 (“Each producer participating in the Defendants' export program was foregoing the opportunity to sell eggs at the higher domestic prices that prevailed at the time, relative to export prices. This failure to pursue unilateral self-interest is only rational in the context of a collusive agreement to coordinate increased exports in order to collectively benefit from higher domestic prices.”). Dr. Rausser's testimony in Subsection V.C is admissible because his analysis here is said to be based on his economic expertise: in particular his expertise as to whether actions can be explained by rational self-interest. This is precisely the type of inquiry an economist can be expected to make using his expertise. See Mid–State Fertilizer Co., 877 F.2d at 1340 (“Economics is the study of rational, self-interested accommodation to scarcity”).
In sum, the Court finds that Dr. Rausser's opinion testimony as outlined in Sections IV and V of his Declaration is admissible. An economic expert may permissibly testify as to whether certain conduct is consistent with collusion or an entity or individual's self-interest, and, moreover, it is consistent with sound economic practice to review the factual record and formulate a hypothesis that can then be tested using economic theory—the examination of the factual record is necessary to determine which tests to run and to confirm that the stories drawn from the data and from the factual record are consistent.See Urethane, 2012 WL 6681783, at *3 (“[S]tructure-conduct-performance analysis by an economist is well-accepted in this field, and the Court concludes that it would be helpful to a jury for an expert to put events into an economic context.”). Dr. Rausser does explain that he relies heavily on the factual record, but that is because he is opining on whether the evidence is consistent with orthodox economic theory as to when collusion can and does arise. His testimony could be helpful to the trier of fact because antitrust legal theory is inextricably linked with economic theory and Dr. Rausser's testimony can help explain the economic theory and how the evidence connects to it. Moreover, Sections IV and V are necessary for the formulation of a hypothesis—Dr. Rausser is examining the factual record with the tools of an economist so as to formulate a theory that he then tests with econometric and statistical evidence.
The Court recognizes that Section V is cast in definitive and conclusive language, and the Court will not allow this language to usurp the Court's own role at the class certification hearing. Dr. Rausser will not be permitted to testify as to the ultimate issue of whether Defendants engaged in an illegal conspiracy to fix prices, or on the credibility of witnesses, or even on whether certain events did or did not occur, but, in the context of an economic report, and especially considering that whether Dr. Rausser considered the evidentiary record is a potential subject for cross-examination, this Section is consistent with the role of an expert.
Finally, the Court rejects Defendants' contention that Dr. Rausser's testimony should be excluded because he failed to consider certain aspects of the record. The briefing and arguments have not revealed that Dr. Rausser's failure to include certain facts in his analysis was so egregious as to make his methodology unreliable. This contention goes to the weight of Dr. Rausser's testimony, not its admissibility. The record in this case is voluminous, to say the least, and no expert will be able to include every document in his analysis.
b. Whether Dr. Rausser's Co–Movement Analysis in Support of Classwide Impact is Reliable
Defendants next ask the Court to conclude that Dr. Rausser's “analyses are so fundamentally flawed that they should be excluded.” Mem. in Supp. of Mot. to Exclude 9. On contrast, Plaintiffs use Section VI of Dr. Rausser's Declaration to try to meet Rule 23(b)(3)'s predominance requirement by showing that damages are capable of proof through evidence common to the class.
Dr. Rausser's “Common Impact” analysis takes the following form: (1) the pressures of supply and demand are common to the entire class, as the market for eggs is nationwide; and (2) all shell eggs and egg products are related no matter their form, meaning that purchasers of any type of egg product or shell eggs were commonly impacted. To prove this, Dr. Rausser opines that (1) the prices of the various types of egg products move together; (2) these prices are affected by a set of “common factors” (to show this, Dr. Rausser relies on a “Common Factors” Regression, discussed in Parts II.c–II.e below); and (3) these prices moved upward due to the alleged conspiracy (again, discussed in Parts II.c–II.e below).
Defendants challenge each of Dr. Rausser's opinions, including his methodology for showing that the prices of various egg products move together. Defendants heavily criticize this section in their Motion for Exclusion and they “incorporate by reference” their criticisms of Dr. Rausser's findings in their Memorandum in Opposition to Class Certification. Mem. in Supp. of Mot. to Exclude 8. They raise three arguments against Dr. Rausser's methodology in his co-movement analysis: First, they criticize his analysis for relying on “visual inspection” of graphs of prices, as opposed to a more rigorous methodology. Id. at 9. However, this criticism appears to have been abandoned due to Dr. Rausser's Reply Declaration's use of statistical analysis to confirm the visual inspection's conclusions. See Rausser Reply Decl. 33–34. Second, they criticize his use of averages, because averages “by their very nature mask individual differences between purchasers and preclude the ability to determine any impact on any of the plaintiffs.” Mem. in Supp. of Mot. to Exclude at 9. Third, they criticize his co-movement analysis because it is “meaningless.” They argue that prices can co-move even if not commonly affected by an event, and that the comparisons Dr. Rausser makes (such as comparing prices based on geography, type, and so forth) are meaningless because they do not “fit” the theory of the case.
Plaintiffs counter that “Dr. Rausser is not offering co-movement as a means, by itself, to show impact on all or virtually all class members. Rather the finding of co-movement is one of several bases ... that support a finding of common impact because they tend to show that there is no subgroup of class members that would have been able to systematically avoid the impact of the conspiracy.” Mem. in Opp. of Mot. to Exclude 12. Dr. Rausser describes the utility of his co-movement analysis as “support[ing] my conclusion regarding a nationwide market for eggs and egg products, as the prices of eggs and egg products are related to each other, irrespective of differences in grade, size, color, egg product type, and so on. Second, the analysis supports that if the Defendants had market power in such a market ... then their actions to restrict supply would have caused common price increases, again irrespective of differences in grade, size, color, egg product type, and so on.” Rausser Reply Decl. 27.
The Court need not conclude at this juncture that Dr. Rausser's co-movement analysis is alone capable of proving common impact. Rather, the Court need only conclude that his co-movement analysis is a reliable analytical form and is “relevant for the purposes of the case and must assist the trier of fact.” Schneider ex rel. Estate of Schneider v. Fried, 320 F.3d 396, 404 (3d Cir.2003). That is, the Court could conclude that Dr. Rausser's testimony is at least relevant to showing a common impact—perhaps because the co-movement analysis rules out some counterargument against common impact—without deciding that it alone is sufficient to show common impact.
i. Whether Dr. Rausser's use of averages makes his co-movement analysis unreliable for showing common impact
Defendants argue that Dr. Rausser's co-movement analysis, in which he attempts to demonstrate that the prices of various types of eggs and egg products in various markets move together, is flawed because it relies on the average prices of these egg products. This, Defendants argue, does not “fit” the issues of the case because it is incapable of demonstrating a common impact on every Plaintiff, as the case law requires. See Mem. in Supp. of Mot. to Exclude 16 (citing Hydrogen Peroxide, 552 F.3d at 311 ).
Defendants offer the Declaration of their own expert, Dr. William C. Myslinski, who explains that using averages can “hide substantial variation across individual cases, which may be key to determining whether there is common impact.” Myslinski Decl. 25. Dr. Myslinski then goes on to demonstrate how these averages hide wide variations in the prices actually paid in individual transactions. Id. at 26–27. By averaging these transactions to arrive at an average price, the data could be hiding the true cause of the co-movement—suppose that, say, only large eggs were changing in price nationwide and that the prices of small and medium eggs were unchanged. Using Dr. Rausser's analysis, that could make it look like the prices of all eggs, including small and medium eggs, co-move in different regions of the country. This could, Dr. Myslinski argues, hide that the prices paid by individual buyers do not display co-movement. Id. at 28. Dr. Myslinski then demonstrates how some sets of prices do not exhibit co-movement, such as between the prices of eggs sold by Michael Foods and those sold by Daybreak. Id. at 25–39.
Plaintiffs counter that the critics of averaging miss the essential point of the co-movement analysis: that prices across the eggs industry move together, such that no subset of the egg industry would have been unaffected by the alleged scheme. They contend that Defendants are merely speculating about how averaging hides differences among individual buyers. Dr. Rausser contends that Dr. Myslinski's hypotheticals of how averages can hide co-movement are not grounded in reality and, further, that Dr. Myslinski's examples of a lack of co-movement can be explained by the small sizes of the samples tested. By including the data of multiple defendants, contends Dr. Rausser, the co-movement again reappears.
For example, Dr. Rausser criticizes the hypothetical that the change in the price of large eggs would mask the fact that the prices of small and medium eggs have not changed. He argues that the data and evidence (including Dr. Myslinski's own deposition testimony) show that consumers will quickly switch to small or medium eggs if the price of large eggs jumps up, which means that a significant disequilibrium in price among various sized eggs is impossible. See Rausser Reply Decl. 32.
The use of averages in a common impact analysis is controversial, and courts have come down on both sides of the issue at the class certification stage. There is no binding Third Circuit precedent on point, but there are a number of district court opinions discussing the issue, albeit difficult to reconcile. Even acknowledging the controversy marking the use of averaging (or even the use of Dr. Rausser himself), the Court finds that Dr. Rausser's methodology meets the requirements of Daubert, since it is reliable for demonstrating, to some extent, that the prices of eggs across various product types and geographic locations are related and “driven by the same supply and demand factors.” Rausser Reply Decl. 27. This is relevant to class certification, even if it does not alone demonstrate common impact (which the Court need not address here—Defendants will have the opportunity to show why the use of averages defeats Plaintiffs' attempts to show common impact at the class certification hearing). For example, if the averages hide what is really driving the co-movement, Defendants are certainly capable of demonstrating what they contend is behind the co-movement.
The parties cite many cases discussing the use of averages, but they primarily dispute the significance of two particular cases: Reed v. Advocate Health Care, 268 F.R.D. 573 (N.D.Ill.2009), and In re Flonase Antitrust Litigation, 284 F.R.D. 207 (E.D.Pa.2012).
In Reed, a class of nurses working for a healthcare network sought certification. The Court's analysis turned on the issue of whether the nurses had offered a reliable method of proving impact. See 268 F.R.D. at 581–82. The nurses relied on the expert testimony of the same Dr. Rausser whose analysis and testimony is at issue here. Dr. Rausser performed an analysis in which he compared the “average effective competitive wage”—what nurses should have been paid absent any collusion—and the “actual average effective wage paid.” Id. at 585. Defendant Advocate Health Care offered their own expert who criticized Dr. Rausser's use of averages, arguing that “the model proposed by Dr. Rausser fails to answer the relevant question for purposes of class certification, which is whether the alleged suppression had an impact on virtually every class member. Averages mask what individual nurses were actually paid; virtually all nurses were paid more or less than average.” Id. at 586.
The court in Reed agreed that this was a fundamental flaw, id. at 590–91, and accepted the defense expert position that “[e]ven if one assumes the average wage was reduced by the alleged conspiracy, that would not mean that all members of the proposed class suffered a reduced wage or that any reduction for an individual nurse could be calculated in a formulaic way by common proof.” Id. at 591.
In Flonase, the court considered whether to certify a class of indirect purchasers of Flonase, who had allegedly suffered an injury because of GlaxoSmithKline's alleged pay-for-delay scheme. 284 F.R.D. at 210–11. In support of their motion for class certification, the plaintiffs offered the expert testimony of, again, Dr. Rausser. The court noted that the “task for plaintiffs at class certification is to demonstrate that the element of antitrust impact is capable of proof at trial through evidence common to the class rather than individual to its members.” Id. at 220 (quoting Hydrogen Peroxide, 552 F.3d at 311–12 ). Dr. Rausser again offered an analysis of how the average price of paid for Flonase was artificially inflated. The Flonase court accepted Dr. Rausser's analysis. The court noted the potential flaw in using averages, but concluded that “[I]t is critical to note that this case does not involve several different products or different markets of supply and demand.” Id. at 229. The Court also distinguished Reed in a footnote, concluding that
[T]his case and Reed are simply inapposite. Reed involved allegations of wage-suppression by a class of hospital-based registered nurses. The court noted that numerous factors could affect the wages of an individual nurse—age, nurse performance and merit, sign-on or retention bonuses, and non-wage compensation (i.e. employee benefits, overtime). Defendants presented evidence that, because of these various factors, the wages paid varied greatly and, more importantly, certain nurses actually received pay increases during the class period. Although Dr. Rausser's methodology failed to control for all the relevant factors in Reed, I find Rausser's methodology to be sufficient for the facts of this case.
Id. at 228 n. 25.
Defendants here seek to distinguish Flonase by arguing that the court's analysis turned on the small amount of variance in the individual data, whereas here the prices paid for eggs can vary dramatically. This Court does not read Flonase as turning solely on the amount of variance in the data, but, nevertheless, notes that Dr. Rausser's Reply Declaration shows that when the volume of egg transactions is considered, the variance in egg sales is not as dramatic as Defendants suggest. See Rausser Reply Decl. 31.
Essentially, the case law seems to compel the court to view averages as at least somewhat suspect, but not as fatally flawed so long as (1) the differentiation among the data being averaged is not so great as to make the use of averages misleading; and (2) there are other indicia that the averages are not concealing the true story of the data.Although it arguably is a close call, the Court finds that Dr. Rausser's analysis here manages to meet Daubert. True, averages are suspect, but this case is not like Reed. The market for eggs is not like the market for nurses. Rather, it is more like the market for Flonase. Eggs are commodity products, and Dr. Rausser's co-movement analysis is supported by his analysis of the market structure and the common factors affecting price. Moreover, Dr. Rausser's co-movement analysis is a fairly straightforward economic analysis that can easily be tested (such as by Dr. Myslinski, who seeks to show what Dr. Rausser's analysis supposedly obscures). The variance in the price of eggs is a concern, but does not render this methodology inexorably unreliable or unhelpful for Daubert purposes. That said, Defendants are certainly not precluded from arguing at class certification that this analysis is insufficient to reliably show that damages can be proven on a classwide basis. Whether the analysis is convincing enough to support class certification is not for the Court to decide now in resolving this Daubert motion. Rather, the Court merely finds that the methodology is reliable and fits the issues of the case.
Defendants in their Memorandum in Opposition to Class Certification raise the argument that Dr. Rausser's use of a single average overcharge suffers from many of the same flaws as the use of averages in his co-movement analysis. Mem. in Opp. of Class Cert. 24. However, this position was not discussed in the briefing for the Daubert motion, nor was it discussed at the Daubert hearing. If the Court had to reach this question, it would likely reach the same conclusion with respect to averages as in the co-movement context, discussed above.
ii. Whether Dr. Rausser's co-movement analysis does not fit the theory of the case
Defendants next assert that Dr. Rausser's co-movement analysis is unhelpful because it is poorly tied to the issue he is seeking to address. Defendants assert that a proper method would have been to “take[ ] putative class members' purchases from Defendants' transaction data and compare[ ] their purchase data to see if the prices paid by individual class members in fact ‘co-moved.’ ” Reply in Supp. of Mot. to Exclude 9.
This point is raised for the first time in Defendants' Reply in Support of the Motion to Exclude and is not particularly well-developed. That said, to the extent there is a legitimate argument to make here, it is better addressed at class certification. That is, this issue goes to the weight of Dr. Rausser's testimony, not to its admissibility. Again, the co-movement analysis is relevant to understanding the eggs market and its uniformity or lack thereof, which is relevant to the question of common impact. His methods are ones that any expert in statistics or economics can understand and test. They will reliably arrive at their results. As to whether those results are convincing—that is a question for class certification.
c. Whether Dr. Rausser's Common Factors Regression in Support of Common Impact is Reliable
Defendants next challenge Section VLB of Dr. Rausser's Declaration, in which he seeks to demonstrate that he can create a regression that shows the price of eggs is determined by a set of common factors. In other words, Dr. Rausser examined all the sales data and compared how the prices were affected by, say, the change in the price of chicken feed. He took the factors he found to affect the prices and created a sort of price-predicting model. This model could explain 65% of the variations in the prices of shell eggs and 66% of the variations in the prices of egg products (i.e. they had an “R-squared” of .65 and .66). The purpose of this section of Dr. Rausser's report is to show common impact. In Section VII he uses his regression to show that the prices of eggs during the conspiracy period were higher than the regression suggests they should have been absent any collusion.
Defendants argue that Dr. Rausser's regression is flawed because it fails to consider many important factors that impact the price of eggs and is an implausible model of the eggs market. In particular, Defendants note that Dr. Rausser's regression model includes only one demand-side variable: population. That is, according to Dr. Rausser's model, the only factor reliably affecting the demand for eggs is the size of the population. Oddly, population appears to have a negative impact on price—when the population increases, the price for eggs decreases. Defendants argue that Dr. Rausser should have included factors like dietary trends, income, avian diseases, and legislation on animal welfare.
Dr. Rausser's Reply Declaration addresses this criticism by showing that inclusion of income, animal welfare legislation, dietary trends, and avian disease would not affect his model. For income, he runs a new regression including income and finds it has a slight impact on the price of eggs, but not enough to undermine his results. Rausser Reply Decl. 89. He concludes that animal welfare legislation was not likely to impact his model because the impact would not be felt until after the conspiracy period (the legislation in California was implemented on January 1, 2015, and the legislation in Michigan is not to be implemented until 2019). Id. at 91–92. He argues that dietary trends are insignificant, since they tend to be short-term. Also, he notes that Dr. Myslinski has not demonstrated that including dietary trends would undermine his findings. Id. at 94. Finally, he concludes that the impact of avian disease is too minimal to include.
California recently implemented a new law requiring egg producers meet certain animal welfare standards. See generally Missouri v. Harris, No. 2:14–0341, 2014 WL 4961473 (E.D.Cal. Oct. 2, 2014). However, these reforms, passed in 2008 and 2010, are essentially the first of their kind in the nation and did not take effect until January 1, 2015—far after the period in question in this case.
Defendants cite various cases to demonstrate that failing to include these certain important variables does “not simply go to the weight accorded to the expert's opinion, but to its very admissibility.” Mem. in Opp. of Class Cert. 28 n. 18 (Doc. No. 1033). The Court must determine whether the absent factors in Dr. Rausser's common factors analysis are so important as to render his analysis unreliable. See, e.g., Multimatic, Inc. v. Faurecia Interior Sys. USA, Inc., 358 Fed.Appx. 643, 655 (6th Cir.2009) (“Perceived flaws in an expert's opinion go to weight only if they fall within the accepted norms of the discipline and have a non-speculative basis in fact.”); Blue Cross & Blue Shield United of Wisconsin v. Marshfield Clinic, 152 F.3d 588, 593 (7th Cir.1998) (“Any nonconspiratorial factors likely to have made the prices charged by the Marshfield Clinic higher than the prices charged by other health-care providers had to be taken into account in order to make a responsible estimate of the prices that Blue Cross would have paid had it not been for the conspiracy.” (emphasis added)); Tagatz v. Marquette Univ., 861 F.2d 1040, 1045 (7th Cir.1988) ; In re Wireless Tel. Servs. Antitrust Litig., 385 F.Supp.2d 403, 428 (S.D.N.Y.2005) (“[Expert's] failure to test for these obvious and significant alternative explanations renders [expert's] analysis essentially worthless.” (emphasis added) (quotation marks and citation omitted)).
The Court finds that Defendants have not identified any factor that Dr. Rausser failed to account for that is so significant as to render his analysis unreliable, even despite the curious correlation between population and prices. The absent factors Defendants fault Dr. Rausser for not including in his regression are far too speculative to render his regression inadmissible—they are not analogous to the missing factors in the rejected regression in the cases Defendants cite. For example, in Multimatic, the expert's regression sought to estimate profits Chrysler would make on a certain automobile program 4–8 years in the future. 358 Fed.Appx. at 654. The expert made a number of “utterly speculative” assumptions, such as speculation on the plaintiff's “profit margins on yet-to-be-signed contracts years into the future.” Id. (“Experts may not assume facts without some support for those assumptions in their expert report or elsewhere in the record.”). And in Tagatz, an expert sought to demonstrate that Catholic professors and non-Catholic professors had experienced a disparity in pay. 861 F.2d at 1042. However, the expert's “failure to control for differences in rank, like his failure to control for differences in scholarly productivity and teaching evaluations, made his [analysis] essentially worthless—and such variables can be controlled for.” Id. at 1045 (emphasis in original).
The seemingly incoherent relationship between prices and population is explained by the extent to which supply-side variables are tied to population. Dr. Rausser states in a footnote of his declaration, “The log of population size enters into the regression with a positive coefficient if entered without the addition of supply cost variables, suggesting that its negative coefficient here is merely due to a correlation with those supply variables.” Rausser Decl. 85 n. 308.
In Blue Cross, the experts sought to demonstrate that health care costs were higher because of an alleged market division by major health care providers. 152 F.3d at 592–93. However, one expert made “no correction for any other factor except differences in the treatment mix,” and the other “compared the prices charged by the Marshfield Clinic with the prices charged by other providers on the basis of price per patient (correcting for differences in the treatment mix), rather than price per service. The result of this procedure was that if Marshfield Clinic provided on average more treatment per patient than other providers, the difference was attributed to the division of markets.” Id. at 593–94.
In Wireless Telephone, the expert sought to demonstrate that the price of wireless handset telephones had been inflated by the illegal restraint of trade of the defendants (specifically by tying the purchase of a telephone plan to the purchase of a certain telephone). 385 F.Supp.2d at 428. The expert's regression, however, did not include any independent variables, including obvious factors that may have affected the cost of wireless handsets such as the “shift from analog to digital technology and the related advances in handset features, including caller ID, email, and web browsers.” Id.
In each of these cases, the absent factors were crucial to creating a reliable regression. The absent factors here—namely, dietary trends, legislation on animal welfare, and avian diseases —do not rise to the same level as the absent factors in the above cases. On behalf of the defense, Dr. Myslinski does not provide any data showing how these factors could have been controlled for. Indeed, controlling for dietary trends or animal welfare legislation would appear extraordinarily speculative in a statistical model and would be unlikely to make the model more accurate. This is not to say that Defendants cannot argue that these factors make Dr. Rausser's model less persuasive, but they do not make it inadmissible.
d. Whether Dr. Rausser's Damage Model Provides a Reliable Method for Calculating Class-wide Damages
In Section VII of his Declaration, Dr. Rausser applies his regression model to show damages. He shows that if you apply his model to the alleged conspiracy period, the price of eggs should have been lower than it actually was. He concludes that there was an 18.5% overcharge for shell eggs and a 12.7% overcharge for egg products. Rausser Reply Decl. 89.
By using the overcharge Dr. Rausser arrived at after using income in his regression—without income in his regression he arrived at 19.3% for shell eggs and 13.8% for egg products—the overcharges are calculated as set forth in the text.
Defendants raise a number of challenges to this portion of Dr. Rausser's analysis. Defendants assert that Dr. Rausser's model is flawed because “(a) Dr. Rausser's regressions imply injury for putative class members that could not have been injured, (b) his model is unstable and unreliable, (c) his benchmark period, against which he measures prices during the ‘conspiracy’ period, does not allow Dr. Rausser to estimate any alleged ‘overcharge,’ and (d) Dr. Rausser makes faulty assumptions about cage space requirements in the but-for world.” Mem. in Opp. of Class Cert. 21.
The Daubert motions make only a passing mention of each of these objections. They are, however, discussed at length in the Memorandum in Opposition to Class Certification, which Defendants incorporated by reference into their Motion to Exclude Dr. Rausser.
--------
i. Whether Dr. Rausser's model is flawed because it calculates an overcharge where none could exist
Defendants argue that Dr. Rausser's model is fatally flawed, much like the model in Rail Freight, because the single average overcharge percentage would apply even to putative class members with long-term contracts, insulating them from the alleged conspiracy. 725 F.3d at 244. In particular, Defendants assert that a large amount of eggs were purchased under long-term contract tied to the price of grain, which would insulate the contracts from a reduction in the supply of eggs. Defendants argue that this is the same flaw that caused the D.C. Circuit to reject Dr. Rausser's model in Rail Freight.
Plaintiffs counter that the “small number of contracts cited by Defendants would be impacted by the conspiracy.” Reply in Supp. of Class Cert. 27 (Doc. No. 1060). Plaintiffs argue that, unlike the contracts in Rail Freight, here, the contracts were entered into during the conspiracy, “meaning they embody prices that were already inflated.” Id. In other words, argue Plaintiffs, these contracts would be affected because the conspiracy “set an artificially high baseline for price negotiations.” In re Titanium Dioxide Antitrust Litig., 284 F.R.D. 328, 346 (D.Md.2012). Also, Plaintiffs contend, the grain-based contracts were, in fact, “set with regard to market conditions.” Reply in Supp. of Class Cert. 27. Finally, say Plaintiffs, even those defendants with these ostensibly immunized contracts also purchased eggs outside these contracts, meaning they would have been injured by the conspiracy regardless.
Dr. Rausser's Reply Declaration provides support for Plaintiffs' counterarguments. See Rausser Reply Decl. 99–102. Dr. Rausser notes that all but one of these challenged contracts were set during the conspiracy period and that even those' set based on grain prices were affected by market prices. He contends that the negotiations themselves would have been affected by a higher market price (which makes intuitive sense—if the non-discounted price is higher than it otherwise would be, the negotiated discounted price is also likely to be higher than it otherwise would be). Second, many of these contracts either factored in the market price or gave the seller the option to alter the price due to market conditions.
In Rail Freight, the D.C. Circuit Court of Appeals considered a class action against rail freight shippers who had allegedly conspired to add a “fuel surcharge,” violating the antitrust price-fixing prohibition. 725 F.3d at 247. Dr. Rausser provided a regression estimating an average overcharge and seeking to demonstrate that the class could prove common injury through common proof. Id. at 250. Defendants there argued that Dr. Rausser's model was flawed because it would have ascribed an injury even to those Plaintiffs who had negotiated legacy contracts with Defendants before the alleged conspiracy period. Id. The Court of Appeals vacated the District Court's certification of the class, finding that the District Court had not adequately considered this flaw. Id. at 255.
Dr. Rausser's model here does not suffer from the same flaws as his model in Rail Freight. For one, the contracts here were set during the conspiracy period, unlike the contracts in Rail Freight, which were set prior to the conspiracy period, making injury impossible. Cf. In re Urethane Antitrust Litig., 768 F.3d 1245, 1267 (10th Cir.2014) (“Second, Dr. McClave's model does not suffer from the same flaw identified in Rail Freight. There, the appeals court could not credit the expert's opinion because his methodology yielded damages for a time period in which prices had been freely set. Thus, the expert found damages for plaintiffs who could not possibly have suffered injury. Here, by contrast, Dow has not identified a single class member for whom injury was impossible.”). This is a significant distinction because the artificially inflated price would have served as the baseline market price for the negotiations. See In re Polyurethane Foam Antitrust Litig., No. 10–2196, 2014 WL 6461355, at *33 (N.D.Ohio Nov. 17, 2014) (“Contract negotiations thus took place in the context of artificially inflated baseline pricing, effects which likely became ‘baked into’ the contracts.”). Dr. Rausser's model is admissible under Daubert, though how these issues affect its weight can and should be revisited at class certification.
ii. Whether Dr. Rausser's model is unstable and unreliable
Defendants next argue that Dr. Rausser's model is hopelessly flawed because their expert, Dr. Myslinski, used Dr. Rausser's regression model on various subsets of the data and found inconsistent results. Dr. Myslinski tested this regression by attempting to apply it to just one certain defendant's transactions. So, as Defendants note, even though Dr. Rausser's regression implies that an increase in gasoline will increase the price of eggs (which is to be expected), when by testing this regression using just Daybreak's transactions, the model suggests an increase in gasoline prices would actually decrease the price of eggs.
Plaintiffs counter that these curious results demonstrated by Dr. Myslinski are the product of inappropriate “data mining,” which “involves applying a model to arbitrary subsets of the transactional data without an economic theory for selecting such subsets.” Reply in Supp. Class Cert. 30. As Dr. Rausser explains, “traditional statistical tests become unreliable under relentless searching to find patterns in data that may merely be the product of chance. Because data mining is unable to distinguish between whether a pattern in the data is the result of pure chance or a hypothesis that is relevant to the case, it becomes irrelevant that any results found are statistically significant.” Rausser Reply Decl. 96. Further, argues Dr. Rausser, by narrowing the amount of data, Dr. Myslinski necessarily makes the regression more unstable and unreliable.
The Court finds Dr. Rausser's regression sufficiently reliable for purposes of Daubert. Direct Purchaser Plaintiffs persuasively note that Dr. Myslinski's methods for testing Dr. Rausser's regression model hardly demonstrate that the model is, as a whole, unreliable for demonstrating an expected overcharge. It is intuitive that what would adequately describe the data as a whole might not accurately describe a certain subset of the data—and curious results like those noted by Defendants are to be expected. The model is easily testable, as Dr. Myslinski as demonstrated, and is reliable enough for purposes of Daubert. That said, just because the Court has found Dr. Rausser's regression model reliable enough for Daubert purposes does not mean that Defendants cannot argue that the curious results uncovered by Dr. Myslinski make the regression model unconvincing for purposes of class certification, especially if upon more in-depth “mining” on wider sampling, the “curious results” become less a curiosity and more a norm.
iii. Whether Dr. Rausser's but-for causation model is flawed because his benchmark assumptions and data are flawed
Defendants argue that Dr. Rausser's regression is flawed because the shell eggs benchmark data is based on the transaction data from only four defendants, but the data for the conspiracy period comes from eleven defendants. For egg products, the data is based on three defendants in the benchmark period and ten in the conspiracy period. Defendants assert that Dr. Rausser has presented no evidence to show that the few defendants from which he derives his benchmark data are representative of the whole. Defendants note that during the alleged conspiracy period, the “average shell egg price of a Defendant without benchmark-data was 14% higher than the average price of shell eggs belonging to a Defendant with benchmark-data.” Mem. in Opp. of Class Cert. 32. One might expect that the average shell egg price of a defendant without benchmark data would also be higher during the benchmark period, mitigating, if not eliminating, any demonstrated price increase during the conspiracy period.
Plaintiffs again return to the argument that Dr. Rausser's benchmark data need not be perfect but only reasonably similar. Reply in Supp. of Class Cert. 32–33 (citing various cases in support). Dr. Rausser defends his model in two ways. First, he contends that Dr. Myslinski's critique is inconsistent because he first claims the model underestimates the damages (with the assumption that there was no anticompetitive conduct in the benchmark period) and then argues that it overestimates them (by not including certain defendants in the benchmark data). Rausser Reply Decl. 105. But this defense is not the most convincing—an inaccurate model is an inaccurate model, even if the inaccuracies pull in opposing directions. The second defense is more convincing. Dr. Rausser explains that he tested his regression by including a set of variables accounting for the defendant-specific differences in pricing. That is, he used a regression where any potential bias created by certain defendants just having higher prices than other defendants would be negated. He found that his overcharge estimates were confirmed.
Dr. Rausser's Reply Declaration adequately accounts for the flaw that Defendants point to, and the Court finds that this critique of Dr. Rausser's model does not make it inadmissible under Daubert.
iv. Whether Dr. Rausser's but-for causation model is flawed because it does not contemplate that cage-space would have increased regardless
This criticism echoes the above criticism of Dr. Rausser's common factors regression. Essentially, Defendants argue that Dr. Rausser failed to account in his regression for the extent to which cage space for the hens would have increased regardless of any anticompetitive activity. Defendants point to several pieces of evidence showing that there was a consumer push for animal welfare standards (including ballot initiatives and requirements from retailers). They also argue that the model fails to account for the costs of expanding the cages to accommodate an increase in the bird flock (had there been no conspiracy).
The Court rejects Defendants' arguments on this front. The absence of these factors does not make the regression so unreliable as to be inadmissible. These factors go to the weight, not the admissibility of his model. Although Defendants could get some traction on the absence of any of these factors at the class certification stage, they do not make Dr. Rausser's opinion so unreliable as to make it inadmissible.
e. Whether Dr. Rausser's Declaration does not fit the theory of the case because it does not consider whether there was a reduction in supply
Finally, Defendants argue that Dr. Rausser's regression is fundamentally flawed because it does not fit the case. Defendants argue that Dr. Rausser's regression, which models price, not supply or flock size, does not fit the Plaintiffs' theory that Defendants conspired to reduce the supply of egg-laying hens. Defendants argue that in light of the Supreme Court's decision in Comcast, the failure to closely tie the model exhibiting damages to the theory of the injury makes the model unreliable.Cf. 133 S.Ct. at 1433 (“There is no question that the model failed to measure damages resulting from the particular antitrust injury on which petitioners' liability in this action is premised”). Defendants further argue that the supply of eggs per capita actually increased during the conspiracy period and that by applying Dr. Rausser's regression to the supply levels, one sees that the supply was higher during the conspiracy period than Dr. Rausser's model suggests it would have been in a world without a conspiracy. Dr. Rausser's graph of the supply of eggs per capita, found in his Reply Declaration, does not demonstrate a decrease in the supply of eggs per capita until 2006, six years into the alleged conspiracy. This failure to tie the damages to the alleged industry is, according to Defendants, a fatal flaw.
Plaintiffs argue that Defendants are over-reading Comcast. In Comcast, the plaintiffs posited four possible theories for the cause of their antitrust damages. Id. at 1430. Their class was only certified on the basis of one of these theories. Id. at 1431. However, their expert's model estimating damages was based on all four possible theories and could not select for the injuries caused by the single theory for which the class was certified. Id. at 1431. Plaintiffs contend that here, where all the theories of liability remain in the case, Comcast does not apply.Dr. Rausser defends his methods by arguing that even though there was an overall increase in supply, the increase would have been much larger but for the anticompetitive conspiracy. He asserts that he did properly account for a reduction in supply due to the conspiracy. Dr. Rausser contends that his sifting through the discovery record (see Parts IV and V of his Declaration) allowed him to conclude that a conspiracy to reduce the supply of eggs had taken place. In other words, Dr. Rausser is saying that the evidentiary record shows a constraint on the growth of supply, so he does not need a model to demonstrate it—his model just measures the extent to which it caused prices to go up as a way of showing classwide impact. See Rausser Reply Decl. 66 (“Moreover, such an analysis of supply is unnecessary, if one can reliably analyze pricing, in conjuncture with the detailed discovery evidence that the Defendants did act to reduce supply and the industry characteristics ... which determine that such actions would have been successful and would have increased prices. Analyzing prices is sufficient because ... price and quantity are co-determined by supply and demand.”). Finally, Dr. Rausser attacks Dr. Myslinski's estimates of the increase in supply during the alleged conspiracy period. He argues that Dr. Myslinski failed to account for certain important events and factors, that his use of the price regression on quantities is problematic, and that his small number of data points (203 data points compared to Dr. Rausser's millions of data points) makes his results unreliable.
Comcast does not defeat the admissibility of Dr. Rausser's model. The case law would appear to endorse the admissibility of this type of regression. For example, in Polyurethane Foam, the Northern District of Ohio found that a single-overcharge regression, much like the one in this case, met the standard of Comcast. 2014 WL 6461355 at *56. In In re High–Tech Employee, 289 F.R.D. 555, 582 (N.D.Cal.2013), the court considered a regression model (similar to the one here) that demonstrated that during the period when the defendant corporations were allegedly agreeing not to poach one another's employees, the employees of the corporations were underpaid. The court found this regression model adequate under Comcast. See id.; see also In re Cox Enters., Inc. Set–Top Cable Television Box Antitrust Litig., No. 12–2048, 2014 WL 104964, at *13 (W.D.Okla. Jan. 9, 2014) (distinguishing Comcast and finding an overcharge model adequate); In re Cathode Ray Tube Antitrust Litig., No. 07–5944, 2013 WL 5391159, at *6–7 (N.D.Cal. Sep. 19, 2013) (reading Comcast narrowly).
Without making any pronouncements about the ultimate ability of Direct Purchaser Plaintiffs to clear the Comcast bar, the Court notes simply that, for Daubert purposes at least, the model proposed by Dr. Rausser is reliable and fits the case. Dr. Rausser's damages model “bakes in,” at least to an extent, the effect of supply on the price—price is determined by supply and demand. Dr. Rausser has put forth evidence that the demand for eggs is inelastic, meaning that the price is primarily determined by the supply of eggs. As the theory goes, if the price of eggs went up, and we have controlled for the legitimate changes in supply, then the price increase can fairly be attributed to an illegitimate restriction of supply. Perhaps at class certification Defendants will be able to show that the connection between the price increase and any reduction in supply is too remote under Comcast. But for Daubert purposes, his testimony is relevant and reliable as a fairly straightforward way of estimating the increase in price attributable to the conspiracy. See
In re TFT–LCD (Flat Panel) Antitrust Litig., No. 07–1827, 2012 WL 555090, at *5 (N.D.Cal. Feb. 21, 2012) (“Even if regression models are not enough, standing alone, to establish classwide impact, they may nevertheless be relevant to the issue. A large average overcharge, for example, might make it more likely that every direct purchaser was overcharged to some degree. And, of course, if Defendants are correct that plaintiffs acknowledge the limits of their experts' models, they will have little trouble establishing those points on cross-examination.”).
III. Conclusion
For the reasons stated above, the Court denies Defendants' Motion to Exclude the Opinions and Testimony of Dr. Rausser. An appropriate order follows.
ORDER
AND NOW, this 26th day of January, 2015, upon consideration of Defendants' Motion to Exclude the Opinions and Testimony of Gordon Rausser, Ph.D. (Doc. No. 1031), it is hereby ORDERED that the Motion is DENIED for the reasons set forth in the accompanying Memorandum.