Summary
noting that Daubert factors should be applied flexibly in determining admissibility of economic and statistical expert testimony
Summary of this case from City of Tuscaloosa v. Harcros Chemicals, Inc.Opinion
No. C-1-93-553.
March 11, 1996.
William J. Brown, Robert G. Cohen, Emens, Kegler, Brown, Hill Ritter, Columbus, OH, Colleen K. Nissl, Senior Litigation Counsel, Borden, Inc., Columbus, OH, Thomas F. Ryan, Sidley Austin, Chicago, IL, for Borden, Inc., and Meadow Gold, Inc.
G. Jack Donson Jr., Michael R. Rickman, Taft, Stettinius Hollister, Cincinnati, OH, for Louis Trauth Dairy, Inc. and David Trauth.
Robert M. Stephenson, Cotsirilos, Stephenson, Tighe Streicker, Chicago, IL, for David Trauth and Daniel G. Smith.
ORDER DENYING DEFENDANTS' MOTION TO EXCLUDE EXPERT TESTIMONY
MOTIONS CONSIDERED
This matter is before the Court on the Defendants' Motion to Exclude Expert Testimony (doc. 354), Plaintiff State of Ohio's response (doc. 385), Defendant Trauth Dairy's reply (doc. 407) and Defendants' Joint reply (doc. 409).
Ohio filed a supplemental memorandum in opposition to the motion to exclude (doc. 430), to which Borden (doc. 439) and Trauth (doc. 447) filed motions to strike, Ohio responded (doc. 456) and Borden replied (doc. 484). Borden also filed a motion for leave to file an additional memorandum in support of the motion to exclude (doc. 483) to which Ohio responded (doc. 464).
The Court finds that the original arguments and pleadings are sufficient for the Court to determine the question in issue. Accordingly, the Court GRANTS Defendants' motions to strike and also DENIES Borden's motion for leave to file supplemental memorandum.
BACKGROUND
The State of Ohio filed suit against thirteen dairies alleging a conspiracy in violation of § 1 of the Sherman Antitrust Act and the Valentine Act. Ohio instituted this lawsuit on behalf of 451 school districts in eastern and southwestern Ohio. Ohio accuses the Defendants of conspiring to set prices and allocate territories in the sale of milk to school districts in violation of federal and state antitrust statutes. Ohio alleges a series of overlapping conspiracies covering the Southwest, Northeast and Southeast regions of Ohio.
Fifteen dairies were originally in this action. Three of the dairies, Hillside Dairy, Meyer Dairy and Coors Dairy, have settled with the Plaintiff and were dismissed from the complaint. All of the remaining Defendants joined in this motion. The Court, however, recently severed the alleged eastern conspiracies. Accordingly, only Borden, Inc.; Meadow Gold Dairies, Inc. (through its parent, Borden, Inc.); Louis Trauth Dairy, Inc.; and David Trauth remain as Defendants in this action. The Court transferred the severed parties to the Northern District of Ohio. See Document 527.
The United States indicted three dairies in southwestern Ohio after an investigation of illegal behavior in school milk procurement programs. Two dairies, Coors and Meyer, plead guilty and agreed to testify against their alleged co-conspirator, Trauth Dairy. Several principals of Meyer testified in the criminal trial that they participated with Trauth Dairy in bid-rigging school milk auctions. This testimony was in direct conflict with their earlier testimony to the grand jury. A jury acquitted Trauth Dairy and its president David Trauth in those criminal proceedings.
The evidence against the remaining Defendants is circumstantial. In fact, much of Ohio's evidence consists of statistical and econometric analyses of the Ohio milk market and the bidding practices of the Defendants.
Defendants seek to exclude the testimony of three of the State of Ohio's expert witnesses under Rules 104(a), 403, and 702 of the Federal Rules of Evidence. The Court held a hearing on this matter January 5, 1996. After reviewing the pleadings and relevant law, we conclude that the State's experts will be allowed to testify.
The Experts
Dr. James McClave has a Ph.D in statistics from the University of Florida. Dr. McClave has taught a number of undergraduate and graduate level courses in statistics and econometrics. Dr. McClave has authored numerous books and articles on statistics and econometrics. He has authored several textbooks including Statistics for Business and Economics and A First Course in Business Statistics.
Dr. McClave founded Info Tech in 1977. Info Tech provides litigation consulting and software development. He has offered testimony or provided consultation in over fifty antitrust cases. Dr. McClave wrote two articles on statistical analysis of competitive markets. He has also presented numerous seminars and courses on detection of collusion through the use of computer analysis.
Dr. Robert Porter has a Ph.D in economics from Princeton University. He presently teaches economics at Northwestern University. Dr. Porter has received several grants relating to research concerning auctions and bidding. He has written numerous articles in various economic and econometric journals, including several on collusive bidding.
Dr. J. Douglas Zona is an economist with National Economic Research Associates ("NERA"), an economic consulting firm that provides testimony and litigation support services. Dr. Zona has a Ph.D from the State University of New York Stony Brook. His dissertation analyzed bid-rigging in the construction industry. Dr. Zona performed his doctoral research under Dr. Porter. Drs. Zona and Porter wrote an article together discussing statistical methods for determining bid rigging.
The Methodologies
Dr. McClave and Info Tech created a data base in order to analyze the school milk market. The data base contains four types of data about school milk contract auctions: 1) information about the district conducting the auction; 2) information pertaining to the district's specifications (i.e. type of milk, whether coolers were required to be provided); 3) information regarding the vendors; and 4) the bids and specifications of the vendors. Dr. McClave performed five types of analyses: 1) market share; 2) incumbency analysis; 3) bid sequence analysis; 4) price-distance analysis; and 5) econometric price comparison.
Market share analysis compares the market shares of the Defendants over time. For example, Dr. McClave's data shows that, Trauth, Meyer and Coors maintained relatively stable market shares from 1984 to 1988 in the districts near Cincinnati. Dr. McClave opines that this is evidence of the dairies protecting each other "incumbencies." On the other hand, Meyer's and Trauth's market shares fluctuated in 1989 when an alleged "price war" broke out.
Incumbency analysis is a study of the turnover in winners of bids. Incumbency rates are calculated by "computing the percentage of the districts won by the same vendor from one year to the next." McClave Report at 10. McClave claims that "[m]arket allocation schemes are usually accompanied by high incumbency rates." Id.
The bid sequence analysis is a graph created using bidding patterns over a bidding season (May-August). This analysis studies the coordination of bidding behavior of two or more dairies. Dr. McClave compares the bid levels and margins between winning and losing bids to look for evidence of coordination among Defendants.
Price-distance analysis compares a vendor's price to the distance from the school district. Dr. McClave opines that in a competitive market prices should increase as a dairy bids farther from its plant. For example, in southwest Ohio in 1983 Meyer and Trauth Dairies' bids increase from 10 cents within thirty miles of their plants to closer to 13 cents at 100 miles from their plants. David Meyer has testified that their agreement with Trauth Dairy fell apart in 1983. On the other hand, both Meyer's and Trauth's prices decrease with distance in 1986 at the height of the alleged conspiracy.
Finally, Dr. McClave performed an econometric price comparison of the alleged conspiratory markets with a baseline area in North-Central Ohio. The study controlled for differences in raw milk prices, distance, specifications, and demand. Dr. McClave found that prices in the Southwestern market (excluding Columbus) were on average 10% higher than in the North-Central market. Prices in Columbus were on average 6% higher.
Econometrics is "the application of statistical and mathematical techniques in solving problems as well as in testing and demonstrating theories." Random House Dictionary Unabridged Edition 1979.
Doctors Zona Porter
Drs. Porter and Zona identified a number of market factors that facilitate collusion. They found several of these factors operative in the school milk market. For example, the school milk market is considered inelastic since the demand for milk is not affected by price increases. Furthermore, there are limited numbers of bidders in any particular auction all of whom have similar price structures and only compete as to price. Finally, schools conduct separate publicly announced auctions and the Defendants have numerous opportunities for interaction through trade and other associations.
Drs. Porter and Zona compared the bidding practices of the Defendants to a control group. They examined the bid submission decision (i.e. whether a particular firm chose to bid) and the bid prices of each Defendant. Drs. Porter and Zona then compared the observed behavior of the Defendants to that of the control group. They created a graph of predicted behavior based upon the control group and compared the Defendants conduct to that prediction.
Drs. Porter and Zona concluded that several of the Defendants bid more often and at higher prices than expected close to their plant. In their opinion, this behavior indicates territory allocation or complimentary bidding.
STANDARD OF REVIEW
The rules governing the admissibility of expert testimony have recently undergone dramatic change. In Daubert v. Merrell Dow Pharmaceuticals Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), the Supreme Court overruled the long-standing "general acceptance" test of Frye v. United States, 293 F. 1013, 1014 (D.C. Cir. 1923). The Court found that the Federal Rules of Evidence displaced the Frye Test. Id. at 587-88, 113 S.Ct. at 2794. First, Frye has been superseded by Rule 702 which governs the admissibility of expert testimony.
Furthermore, the Supreme Court concluded that Frye's rigid formulation is "at odds with the `liberal thrust' of the Federal Rules and their `general approach of relaxing the traditional barriers to opinion testimony.'" Daubert, 509 U.S. at 588, 113 S.Ct. at 2794. The basic rule of admissibility is "[a]ll relevant evidence is admissible. . . ." Fed.R.Evid. 402. "'Relevant Evidence' means any evidence having any tendency to make the existence of any fact . . . more probable or less probable than it would without the evidence." Fed.R.Evid.401.
The Federal Rules of Evidence do provide guidelines on the admissibility of expert testimony. Rule 702 provides that "[i]f scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise." As with all other types of evidence, the rules require that expert testimony must satisfy two threshold questions before it is admissible: relevancy and reliability. Daubert, at 589, 113 S.Ct. at 2795 ("[T]he trial judge must ensure that any and all scientific testimony or evidence is not only relevant, but reliable.").
"Relevancy," in the context of expert testimony, is testimony that will "'assist the trier of fact to understand the evidence or determine a fact in issue.'" Id. (quoting Rule 702). The testimony must be helpful or "fit" with the issues to be resolved in the case. "Rule 702's 'helpfulness' standard requires a valid scientific connection to the pertinent inquiry as a precondition to admissibility." Id. at 591-92, 113 S.Ct. at 2796.
In the context of scientific evidence, "reliability" also takes on special meaning. Rule 702 requires that expert testimony relate to "scientific knowledge." The Supreme Court stated that scientific knowledge is that which is "derived from the scientific method." Id. at 590, 113 S.Ct. at 2795. In other words, the reliability of scientific evidence is "based upon scientific validity." Id. at 590 n. 9, 113 S.Ct. at 2795 n. 9.
We note that the Supreme Court limited the Daubert analysis to "scientific" knowledge. The Sixth Circuit has applied the Daubert framework to technical and other specialized knowledge as well. See Cook v. American S.S. Co., 53 F.3d 733, 738 (6th Cir. 1995).
Accordingly, the test for the admissibility of scientific evidence is "whether the reasoning or methodology underlying the testimony is scientifically valid and whether that reasoning or methodology properly can be applied to the facts in issue." Id. at 592-93, 113 S.Ct. at 2796. The Court stressed that this inquiry is flexible and that the focus should be on the "methodology, not the conclusions they generate." Id. at 595, 113 S.Ct. at 2797. In other words, the focus is "not on what the experts say, but what basis they have for saying it." Daubert v. Merrell Dow Pharmaceuticals, Inc., 43 F.3d 1311, 1316 (9th Cir.), cert. denied, ___ U.S. ___, 116 S.Ct. 189, 133 L.Ed.2d 126 (1995).
The Supreme Court identified some of the factors that courts should consider in making this inquiry:
1) whether the theory or technique can be tested;
2) whether the technique or theory, has been subject to peer review and publication;
3) the known or potential rate of error and standards controlling the technique's operations; and
4) whether the theory or technique has been generally accepted by the scientific community.
The Court emphasized that these factors are not a definitive list but only a partial list of what courts should consider. Daubert, 509 U.S. at 592-93, 113 S.Ct. at 2796.
DISCUSSION
Defendants' arguments against admissibility fall into three categories: 1) claims that Plaintiff's proffered experts are not qualified to testify; 2) the methodologies used by the experts are fatally flawed; 3) even if the methodologies are valid, the conclusions of the analysis cannot distinguish between legal and illegal parallelism and are therefore of no assistance to the trier of fact. The over-arching rationale for exclusion is the claim that Plaintiff's experts cannot distinguish between legal and illegal collusion.
An initial question we must address is whether Daubert applies to the testimony offered by the State's experts. The Supreme Court limited the Daubert analysis to "scientific" knowledge. Rule 702, however, allows experts to testify regarding technical or other specialized knowledge.
Drs. Zona and Porter are economists and Dr. McClave is a statistician. All three experts evaluate economic data using multiple regression analysis of the milk industry, called econometrics. Econometrics is "primarily statistical with an emphasis on methods and problems most important to the relative substantive discipline." FEDERAL JUDICIAL CENTER REFERENCE MANUAL ON SCIENTIFIC EVIDENCE, David Kaye and David A. Freedman, Reference Guide to Statistics, 336 [hereinafter REFERENCE MANUAL]. Neither economics or statistics seems to completely qualify as "scientific knowledge."
"Multiple regression analysis is a statistical tool for under standing the relationship between two or more variables." REFERENCE MANUAL, Daniel Rubinfield, 419.
Although the proffered experts' testimony may not qualify as "scientific knowledge," the reasoning of Daubert still applies. The Sixth Circuit has applied the Daubert framework to technical and other specialized knowledge. Cook v. American S.S. Co., 53 F.3d 733, 738 (6th Cir. 1995). Accordingly, the general framework of Daubert applies to all expert testimony. See American College of Trial Lawyers, Standards and Procedures for Determining Admissibility of Expert Testimony after Daubert, reprinted in 157 F.R.D. 571 (1994) (arguing Daubert principles should be used to analyze nonscientific expert testimony). The Court, however, should apply the Daubert factors with a clear understanding that they were devised specifically for scientific testimony. In other words, the Daubert analysis should be modified in the case of social science or other nonscientific expertise.
In this case the inquiry is whether the experts are testifying to economic, statistical or econometric knowledge that will assist the trier of fact to understand a fact in issue. See In re Aluminum Phosphide Antitrust Litigation, 893 F. Supp. 1497, 1506 (D.Kan. 1995) (reformulating the Daubert Test in context of proffered economics expert). The Court must decide if the proffered testimony is based upon valid economic, statistical or econometric methodologies and reasoning that can properly be applied to the facts of this case.
Reliability
Econometric and regression analyses are generally considered reliable disciplines. Petruzzi's IGA Supermarkets v. Darling-Delaware Co., 998 F.2d 1224, 1238 (3d Cir.), cert. denied, 510 U.S. 994, 114 S.Ct. 554, 126 L.Ed.2d 455 (1993) (finding use of multiple regression analysis reliable under Rule 702); see also, Daniel L. Rubinfield, Econometrics in the Courtroom, 85 COL.L.REV. 1084, (1985). Furthermore, defense expert Dr. Myslinski conceded that Dr. McClave's regression analysis is testable, generally accepted and reproducible. Myslinski Report at 57. Regression and statistical analysis have been admitted in antitrust cases to prove injury and to determine damages. State of Colorado v. Goodell Brothers Inc., 1987 WL 6771 (D.Colo. Feb. 17, 1987) (admitting one and excluding one of Dr. McClave's econometric models estimating damages). Other courts have admitted similar analysis on the issue of liability. Commonwealth of Kentucky ex rel. Chris Gorman v. Trauth Dairy, No. 92-50, slip op. at 5 (W.D.Ky. Oct. 12, 1995) [hereinafter Trauth Dairy, Kentucky Case]; but cf., City of Tuscaloosa v. Harcros Chemicals, Inc., 877 F. Supp. 1504 (N.D.Ala. 1995) (excluding Dr. McClave and his colleague for failing analysis under Daubert).
Defendants rely on Harcros as authority that Dr. McClave's analysis is inadequate under Daubert. We disagree with the court's decision in Harcros. First, the Harcros court applied an inappropriate standard. The court stated that "[s]ince there is no direct evidence of conspiracy, expert testimony must be carefully scrutinized." Id. at 1512. There is no basis in case law or the Federal Rules of Evidence for such a conclusion. Instead, the Supreme Court's ruling in Daubert liberalizes the standard for the admission of expert testimony. Nothing in Daubert even implies that the standard is raised in absence of "direct evidence."
Second, the Harcros court tied the Daubert analysis to the experts' conclusions. See Trauth Dairy, [Kentucky Case], slip op. at 5 (concluding that Harcros applied Daubert to the expert's conclusions not methodologies). Under Daubert, the court must focus its inquiry on the "methodology, not the conclusions they generate." Daubert, 509 U.S. at 595, 113 S.Ct. at 2797. The Harcros court, however, found that the economist's conclusion that market shares were stable was incongruent. Harcros, at 1527. The court reviewed the data for itself and concluded that the it indicated instability. Such inquiry is inappropriate to determine whether expert testimony is admissible.
Defendants further argue that the experts' testimony should be excluded because Plaintiff's experts' data base is invalid. Specifically, Defendants assert that Plaintiff's experts' control group is flawed and prejudices the analysis. Problems in selection of a sample bear on the weight of the testimony, not its admissibility. Berry v. City of Detroit, 25 F.3d 1342, 1352-52 n. 11 (6th Cir. 1994), cert. denied, ___ U.S. ___, 115 S.Ct. 902, 130 L.Ed.2d 786 (1995). In Berry, plaintiff's attorney chose 187 incidents of police shootings for his expert to examine out of a possible 636 incidents. The Sixth Circuit determined that such a choice did not in itself make the study inadmissible. Id. Relevancy
Relevance is the "fit" requirement of the Daubert framework. Under this prong of Daubert, State must demonstrate that its expert testimony will be helpful to the jury in resolving this case. Defendants argue that the State's experts should be excluded because they cannot prove the existence of illegal collusion.
The Sherman Antitrust Act only makes certain types of "non-competitive" behavior illegal. For example, super-competitive prices resulting from tacit collusion (i.e. "ogopolistic price coordination or conscious parallelism") are not illegal. Brooke Group v. Brown Williamson Tobacco Corp., 509 U.S. 209, 227, 113 S.Ct. 2578, 2590, 125 L.Ed.2d 168 (1993). In order to show a violation of § 1 of the Sherman Act, the plaintiff "must present evidence that tends to exclude the possibility that the alleged conspirators acted independently. . . ." Matsushita Electric Industrial Co. v. Zenith Radio Corp., 475 U.S. 574, 588, 106 S.Ct. 1348, 1356, 89 L.Ed.2d 538 (1986).
Section 1 states in pertinent part: "Every contract combination in form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several states . . . is declared illegal. . . ." 15 U.S.C. § 1.
Essentially, illegal collusion requires proof of an agreement. At least one commentator has stated that regression analysis cannot directly test for antitrust violations. Daniel L. Rubinfield, Econometrics in the Courtroom, 85 COL.L.REV. 1084, 1089 (1985). Statistical analysis cannot directly demonstrate the existence of an agreement. Such a conclusion does not necessarily mean that regression analysis is unhelpful in determining whether illegal collusion has occurred.
Circumstantial evidence derived from the business practices of the defendants can be used as an inference of illegal action. See TV Communications Network, Inc. v. ESPN, Inc., 767 F. Supp. 1062 (D.Colo. 1991), aff'd, 964 F.2d 1022 (10th Cir.), cert. denied, 506 U.S. 999, 113 S.Ct. 601, 121 L.Ed.2d 537 (1992) (finding that a claim of conspiratorial conduct under Sherman Act for price fixing may be established through circumstantial evidence); Reserve Supply Corp. v. Owens-Corning Fiberglas Corp., 799 F. Supp. 840 (N.D.Ill. 1990), aff'd, 971 F.2d 37 (7th Cir. 1992) (Parallel business behavior is circumstantial evidence which a court may rely upon to infer collusion in restraint of trade.). Plaintiff's experts provide analytical data on the business activity of the Defendants. Accordingly, the experts' statistical analysis of the Defendants business practices is admissible. See Commonwealth of Kentucky ex rel. Chris Gorman v. Trauth Dairy, No. 92-50, slip op. at 7-8 (W.D.Ky. Oct. 12, 1995) (allowing testimony as to bidding practices to support or contradict other evidence of collusion). Defendants argue that it is doubtful whether an antitrust plaintiff could survive summary judgment on the basis of statistical analysis alone. This fact, however, does not make the evidence inadmissible. Whether evidence is sufficient to meet the burden of proof is not pertinent to a discussion of admissibility of expert testimony.
The Third Circuit concluded that the regression analysis in an antitrust case probably would not be adequate to survive summary judgment. Nonetheless, the Third Circuit found that the economic analysis coupled with statements by the defendants "tends to exclude the possibility that the market as it exists today came about [as] a result of independent action." Petruzzi's IGA, 998 F.2d at 1241. Likewise, Judge Bertelsman admitted similar experts' economic analysis to support or contradict the other evidence of alleged conspiracy. Trauth Dairy, [Kentucky Case] slip op. at 6.
Furthermore, the experts' analysis will assist the jury assimilate complicated economic data. As the Third Circuit pointed out, "the testimony clearly will assist the trier of fact because it makes sense out of pricing data through the use of multiple regression analysis. The data standing alone are meaningless to a lay person. In fact, they are meaningless to an economist until a statistical study is run." Petruzzi's IGA, 998 F.2d at 1238.
Consequently, the Court will allow the Plaintiff's experts to testify regarding the business activities of the Defendants as demonstrated by the statistical analysis. While Plaintiff's experts' statistical analysis is admissible under Daubert, their conclusions are not admissible because they embrace a legal conclusion. The Sixth Circuit limited the admissibility of expert opinions on ultimate issues to factual conclusions. Berry, 25 F.3d at 1353. "Although an expert's opinion may `embrace an ultimate issue to be decided by the trier of fact[,]' the issue embrace must be a factual one." Id. (internal citation omitted). In other words, the State's experts may not express an opinion in the form of a legal conclusion regarding the existence of an illegal conspiracy. Instead, these experts may testify how their analyses are consistent with the other evidence of conspiracy. Trauth Dairy, [Kentucky Case] slip op. at 7-8.
CONCLUSION
Accordingly, the Defendants' Motion to Exclude Expert Testimony is hereby DENIED. The Plaintiff's experts Drs. Porter, Zona, and McClave will be allowed to testify, however they will not be allowed make a legal conclusion regarding the existence of an illegal conspiracy. This decision is limited to the alleged southwestern conspiracy. The transferee judge will decide whether these experts may testify in the eastern conspiracies.
SO ORDERED.