Publications: Disaggregation of Economist Liability Testimony in Section One Litigation

This information or any portion thereof may not be copied or disseminated in any form or by any means or downloaded or stored in an electronic database or retrieval system without the express written consent of the American Bar Association.

Introduction

Expert economic testimony has become a common method for plaintiffs to attempt to establish the existence of a price-fixing conspiracy. Such opinions are frequently offered as a laundry list of factors that, in the opinion of the economist, make the behavior of the industry “consistent with collusion.” Expert testimony concerning industry conditions is often based on the economist’s interpretation of disputed evidence that the jury, as the finder of fact, may not ultimately accept. Nonetheless, “[t]he interpretation of those facts through the lens of economics has become the center ring of antitrust litigation.”¹

Despite the fact that economist opinions of consistency with collusion rely on an interpretation of a range of disputed evidence, such opinions are often offered without any effort to prioritize which factor(s) were most important to the economist in reaching his or her opinion. It is not difficult to imagine a jury deliberating what to do with an economist’s opinion of consistency with collusion when the jury finds that the facts support several of the identified industry factors but not others. The jury is left without any guidance to determine whether the industry behavior overall is nonetheless consistent with collusion. It is the conclusion of this article that such "check the box" opinions should be excluded pursuant to the Supreme Court's rulings in Daubert² and Kumho Tire³ because such opinions are distinctly unhelpful to, and are likely to confuse, the jury. In addition, this article uses as a case study this author’s recent experience in trying to a verdict the price-fixing allegations in In re High Pressure Laminates Antitrust Litigation, which illustrates the potential for jury confusion.⁴

The Present State of Expert Liability Testimony in Section One Cases

Much has been written about the substance of expert testimony in Section One cases, which typically involves economists presenting opinions of collusion based on traditional “plus factors.”⁵ Greg Werden recently complained that “the present state of the law [concerning such testimony] is unsatisfactory because neither the courts nor testifying economists have explicitly grounded their analyses in modern oligopoly theory, which is the only rational basis for evaluating economic evidence on the existence of collusion.”6 Such a broad critique is beyond the scope of this article. Rather, this article focuses on ensuring that economists offering such structural opinions give the jury certain tools which ensure that their opinions are helpful to the jury and not likely to cause confusion.

Plus factors identified in numerous texts, articles and legal opinions include industry concentration, entry, cross-ownership, regularity of product orders, buyer power, demand elasticity, demand trends, product homogeneity, market share stability, opportunity for communication or contacts, excess capacity, transparency, actions against interest and production cost trends.⁷ Typical of the literature, however, is the lack of significance that is ascribed to any of the delineated factors. This agnosticism may be understandable in the abstract because one would have to apply these plus factors to the specifics of a particular industry in order to reach a conclusion as to what was important in explaining the behavior of that industry.

But in the context of litigation, after a detailed industry-specific study, “Plaintiffs’ expert economists . . . often testify that industry conditions are conducive to collusion . . . .”⁸ However, all too often, these opinions are presented in an aggregated bundle that fails to prioritize the observed factors making use of the industry-specific data available to the testifying economist.

In re High Fructose Corn Syrup Antitrust Litigation (“HCFS”),⁹ is an excellent example of an economist’s use of this “check the box” approach to plus factors. In his HFCS opinion, Judge Posner observed that the “plaintiffs’ economic expert opined in his report . . . that the structure of the HFCS market, far from being inimical to secret price fixing, is favorable to it.” The specific factors observed by the economist were few sellers, homogenous product, pricing transparency and excess capacity.¹⁰ In addition, the plaintiff’s economist found behavior consistent with collusion based on evidence of market share stability, ability to price-discriminate and pricing behavior following certain communications.¹¹ Although much of the evidence supporting these conclusions was in dispute, Judge Posner found the facts sufficient to overcome summary judgment.¹²

Indeed, at summary judgment, disputed facts supporting expert opinions will usually be resolved in favor of the plaintiff, which allows the court at that stage to ignore the potential pitfalls of an aggregated economist liability opinion.¹³ But the instant the plaintiff-friendly evidentiary presumptions of the summary judgment standard are removed, the problem of the failure to disaggregate immediately becomes an issue of concern.¹⁴ The HFCS case settled before trial, thus avoiding the need for a jury to sort through the aggregated economist liability opinion.

Disaggregation of Expert Liability Opinions Should be Required for Admissibility of Expert Testimony

Federal Rule of Evidence 702 assigns courts a “gatekeeping” function with regard to expert testimony. The Supreme Court in Daubert requires a court in its “gatekeeper” role to “make a twofold inquiry: whether the expert’s testimony is based on a reliable methodology, and whether the testimony will assist the fact finder in understanding the evidence or in determining a fact in issue.”¹⁵ It is this second inquiry that is all too often given short shrift by trial judges. There is not a single reported decision in an antitrust case where an expert liability opinion, while based on a reliable methodology, was nonetheless excluded because the opinion would not assist the trier of fact.

In contrast, many courts have not hesitated to exclude damages opinions in antitrust cases that present a total damages figure without disaggregating which amounts are attributable to particular acts of wrong doing. In Intimate Bookshop, Inc. v. Barnes & Noble, Inc., the court excluded a damages model that did not disaggregate whether a plaintiffs' alleged lost sales were “caused by defendants' alleged violation of the Act, as opposed to other intervening market factors.”¹⁶ This decision is noteworthy in that it is not predicated on whether the economist used accepted methods that would meet peer review. Rather, exclusion occurred because, in the opinion of the court, the expert did not provide the jury tools it could use to unscramble the bases for the excluded opinion, creating a risk of a confused outcome.

The necessity for disaggregating damages was made clear by the facts in MCI Communications Corp. v. AT&T,¹⁷ where MCI sued AT&T for twenty-two discrete anticompetitive acts. MCI provided a single damage calculation that covered all of the allegedly improper acts. The district court granted summary judgment to AT&T on seven of the twenty-two acts; the jury found in favor of AT&T on five additional acts, and the Seventh Circuit set aside the jury’s finding of liability as to several other acts. In considering the jury’s damages determination, the Seventh Circuit concluded that it was necessary to reject MCI’s damages theory and to remand for a new determination on damages as MCI’s damages theory failed to provide an “adjustment of the damages award to reflect findings of non-predation.”¹⁸

Thus even though the expert’s damages model may use sound economic methods, it can still be excluded if it could require the jury, in parsing the damages calculation, to rely on “speculation or guesswork.” The exact same concerns should apply to expert liability opinions where the expert fails to give the jury a reasonable and principled basis to give weight to the liability opinion if it found some but not all of the bases for the opinion to be supported by the evidence. Otherwise juries are left to speculate as to whether industry conditions support collusion even if the jury finds pricing opacity or capacity constraints. It is not defensible for an economist to claim all industry conditions bear equal weight in the opinion, or that their opinion would still have the same validity if a majority of the plus factor conditions are found by the jury to be inconsistent with the evidence.

A Case Study of the Need to Disaggregate Economist Liability Opinions

In re High Pressure Laminates Antitrust Litigation involved allegations of a seven year conspiracy by the four manufacturers of high pressure laminates (“HPL”) to set a “range” of prices. Plaintiffs presented the testimony of Dr. Gordon Rausser, who offered two opinions regarding competition in the HPL industry. Dr. Rausser first opined that “the economic conditions of the industry were consistent with defendants’ ability to form and maintain an agreement to fix prices” (“Industry Conditions Opinion”) based on nine factors he believed to be supported by the factual record: homogenous product; high concentration; low price elasticity of demand; mobility barriers for non-defendant suppliers; entry barriers; excess capacity; absence of buyer power; possibility of strategic behavior on the part of incumbents toward new entrants; and information on new entrants.

Similarly, Dr. Rausser opined that the HPL manufacturers’ “behavior was consistent with collusion” (“Collusion Opinion”) based on ten factors he believed were supported by the factual record:

• Although [defendant] Wilsonart had excess capacity, it did not take advantage of [defendant] Formica’s weakness; 
• [Defendant] Formica led three price increases in the 1990s, despite its weakness; 
• Price increases were announced well ahead of their effective date, and defendants often had advance knowledge of each other’s price moves; 
• The manufacturers shared additional confidential business information; 
• HPL market shares were stable; 
• Gross margins improved;
• Defendants adhered to a strategy of meeting but not beating competitor pricing;
• Customer prices were unusually transparent to competitors; 
• Throughout the 1990s, defendants received timely quarterly updates on their HPL market shares from [trade association] NEMA and additional market share information from CPMA;
• Frequent meetings at trade shows and industry associations.

Many of these factors presented factual assessments of the industry that were within the jury’s understanding.¹⁹ Consequently, the jury was required to assess the facts supporting Dr. Rausser’s factors as well as the conclusions Dr. Rausser based on those facts.
If the jury disagreed with Dr. Rausser’s assessment of just one of the factors underlying either opinion, Dr. Rausser provided the jury with no methodological basis to assess the validity of that particular opinion because he had not explained the significance of any of the factors or how changes in those factors would impact his opinions.²⁰ This left the HPL jury in the difficult position of having to speculate as to the continuing validity of Dr. Rausser’s overall conclusions if it did not accept all the individual factors that led to those conclusions.

Excess capacity provides an excellent example of the way Dr. Rausser’s approach invited jury speculation and guesswork. Dr. Rausser determined that Defendant Wilsonart in particular, and the HPL industry generally, had excess capacity by applying a formula he called “practical capacity,” which was not a capacity measurement metric used by any defendant. Evidence from contemporaneous Wilsonart documents and executives at Wilsonart contradicted Dr. Rausser’s determination that Wilsonart had excess capacity. For example, Wilsonart’s internal operating plans showed that Wilsonart was operating at 93% capacity in 1993 when the collusion supposedly began and at 101.4% capacity in 1994.

The jury thus had to evaluate whether the evidence supported Dr. Rausser’s conclusion that there was excess capacity, and then apply that conclusion to Dr. Rausser's ultimate opinion as to whether industry conditions supported collusion. If the jury viewed this evidence differently than did Dr. Rausser, it had no means of determining whether Dr. Rausser would still conclude that “the economic conditions of the industry were consistent with defendants’ ability to form and maintain an agreement to fix prices.”²¹

Implications for the Practitioner

In the HPL case, Judge Brieant denied Wilsonart’s Daubert motion and admitted Dr. Rausser’s opinions without any explanation beyond the following sentence: “these issues concern the weight that the jury may choose to give Dr. Rausser’s testimony.”²²
With all respect to Judge Brieant’s decision, the matter of aggregated liability opinions can hardly be said to be definitively resolved. The very problem presented by aggregated liability opinions is that they give a jury no accepted method for assigning the opinions weight unless they accept every fact and subcomponent of the opinion. Absent disaggregation, the jury’s determinations as to weight are inherently left to the vagaries of speculation and guesswork unmoored from any accepted scientific method.

Defense counsel should continue to push plaintiff economists to give some prioritization of significance to the facts and conditions they observe. The precision of an ordered ranking is probably not necessary nor is a descent into a Sudoku-like series of permutations of possible industry condition combinations required to gain admissibility. Rather, economists should be able to identify the few facts they found most significant, and be able to identify where they would start losing certainty of the validity of their conclusions. Defense lawyers should continue to try to exclude the opinions of those recalcitrant economists who refuse to assign significance to one or more factors or pieces of evidence. At a minimum, an interesting appellate issue will be preserved.

¹ A. Gavil, After Daubert: Discerning the Increasingly Fine Line Between the Admissibility and Sufficiency of Expert Testimony in Antitrust Litigation, 65 ANTITRUST L. J. 663 (1997).
² Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).
³ Kumho Tire Co. v. Carmichael, 627 U.S.137 (1999).
⁴ I say potential because the jury ultimately was able to deliver a complete defense verdict in less than five hours of deliberations, partly because it did not focus on the morass of economic testimony in its deliberations.
⁵ “Economists often explain whether conduct is indicative of collusion . . . courts have held that an expert is permitted to testify that the ‘climate’ of a specific market was consistent with a conspiracy.” U.S. Information Sys., Inc. v. Int’l Bhd. of Elec. Workers No. 3, 313 F. Supp. 2d 213, 240 (S.D.N.Y. 2004).
⁶ G. Werden, Economic Evidence of Collusion, 71 ANTITRUST L. J. 719 (2004).
⁷ Typical of the scholarly presentation of plus factors is Massimo Motta’s discussion in his text COMPETITION POLICY: THEORY AND PRACTICE 142 (Cambridge Univ. Press 2004). Dr. Motta describes plus factors as follows: “if a given factor relaxes the incentive constraints of the firms, then it facilitates collusion; if it makes it more binding, it hinders it, if the effect is ambiguous, then the factor does not have a clear impact on collusion.” Dr. Motta then identifies factors that “facilitate collusion” which include concentration, entry, cross-ownership, regularity of product orders, buyer power, demand elasticity, demand trends, product homogeneity, opportunity for communication or contacts, excess capacity, transparency, and production costs.
⁸ Werden, supra note, 6 at 789.
⁹ 295 F.3d 651, 656 (7th Cir. 2002).
¹⁰ Id. at 656-57.
¹¹ Id. at 656-62.
¹² Id. at 660.
¹³ “The standards–and judicial methods–for granting summary judgment thus stand in stark contrast to those used to determine questions of admissibility of expert testimony. . . .questions of admissibility of expert testimony are far more focused and less deferential to the non-movant.” Gavil, supra note 1 at 701-02.
¹⁴ Judge Posner (unwittingly perhaps) previewed this potential in his discussion of excess capacity, observing that “the significance of excess capacity depends on the ratio of fixed to variable costs. . . . This is so far an unexplored issue in the litigation.” High Fructose Corn Syrup Antitrust Litig., 295 F.3d at 657.
¹⁵ Industrial Enclosure Corp. v. Northern Ins. Co. of New York, 2000 WL 1029192, at *2 (N.D. Ill. July 26, 2000).
¹⁶ 2003 WL 22251312, at *6 (S.D.N.Y. 2003). See also In re Independent Serv. Orgs. Antitrust Litig., 85 F. Supp. 2d 1130, 1153 (D. Kansas 2000) ("To meet its burden to disaggregate, plaintiff must provide the jury a reasonable basis upon which to estimate the amount of losses caused by lawful factors.").
¹⁷ 708 F.2d 1081, 1163 (7th Cir. 1983).

¹⁸ Id. at 1164. See also Image Tech. Servs., Inc., 125 F.3d at 1224 (remanding for a new trial on damages because of failure to disaggregate); Universal Amusements Co. v. General Cinema Corp. of Texas, Inc., 635 F. Supp. 1505, 1526 (S.D. Tex. 1985) (“Since Plaintiffs' model provided the jury no basis to determine the effect on damages if it found one and none of the challenged practices innoxious, the model could not support a jury award.”).                                             ¹⁹ See, e.g., Brooke Group, 509 U.S. at 242 (“Expert testimony is useful as a guide to interpreting market facts, but it is not a substitute for them.”); Virgin Atlantic Airways Ltd. v. British Airways PLC, 69 F. Supp. 2d 571, 580 (S.D.N.Y. 1999) (“An expert’s opinion is not a substitute for plaintiff’s obligation to provide evidence of facts that support the applicability of the expert’s opinion to the case.”).
²⁰ Dr. Rausser was asked at his deposition whether he had, in effect, considered the question of disaggregation. Dr. Rausser stated as follows: “[Y]ou’re asking about robustness analysis to look at all the different behaviors and conduct and the doing of a stand alone analysis to see whether or not if you back out one sequence of actions that your results would still be robust . . . . that’s not an analysis that I’ve conducted . . . . you certainly could’ve conducted such an analysis.” (Rausser Dep. Tr. at 81:8-82:25).
²¹ See also id. at 85:2-9:
Q. But as you sit here today, you’re not offering any explicit guidance to a jury as to how to evaluate this industry as a matter of economics if in fact there was no excess capacity during the period you studied; is that correct?
A. No. I’m here to offer the opinions about the analysis that I’ve conducted for the two questions I was asked to assess.
²² 2006 WL 931692, at *2 (S.D.N.Y Apr. 17, 2006).