United States District Court, E.D. Wisconsin
FOND DU LAC BUMPER EXCHANGE, INC., on behalf of itself and others similarly situated, Plaintiff,
JUI LI ENTERPRISE COMPANY, LTD., et al., Defendants. DZIDRA FULLER, et al, Plaintiffs,
JUI LI ENTERPRISE COMPANY, LTD., et al., Defendants. FIREMAN’S FUND INSURANCE, CO., on behalf of itself and others similarly situated, Plaintiff,
JUI LI ENTERPRISE COMPANY, LTD., et al., Defendants. NATIONAL TRUCKING FINANCIAL RECLAMATION SERVICES LLC, on behalf of itself and others similarly situated, Plaintiff,
JUI LI ENTERPRISE COMPANY, LTD., et al., Defendants.
DECISION AND ORDER
LYNN ADELMAN, District Judge.
This antitrust litigation involves two groups of plaintiffs, direct purchasers and indirect purchasers of aftermarket sheet metal products. Direct purchaser plaintiffs (“DPPs”) have filed a motion for class certification, and defendants have filed a motion to exclude their expert’s testimony. I now address defendants’ motion to exclude.
DPPs rely heavily on the expert opinion of Dr. Russell Lamb in proving the elements necessary for class certification. Accordingly, I must rule on defendant’s challenge to Dr. Lamb’s opinion before deciding class certification. Am. Honda Motor Co., Inc. v. Allen, 600 F.3d 813, 815-16 (7th Cir. 2010). Federal Rule of Evidence 702 governs expert witness evidence and states that such evidence is only admissible if (1) the witness is qualified by knowledge, skill, experience, training, or education; (2) the witness’s specialized knowledge will help the factfinder understand evidence or determine a fact issue; (3) the opinion is based on sufficient facts or data; and (4) the expert has reliably applied principles and methods to the facts of the case. See also Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137 (1999); Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579 (1993). The key inquiry is “the validity of the methodology employed by an expert, not the quality of data used in applying the methodology or the conclusions produced.” Manpower, Inc. v. Ins. Co. of Pa., 732 F.3d 796, 806 (7th Cir. 2013).
Dr. Lamb is an economist with over 20 years of experience consulting on the economics of markets and prices. Defendants do not challenge Dr. Lamb’s qualifications but rather the reliability of his methodologies. DPPs asked Dr. Lamb to analyze (1) whether evidence common to the DPP class can be used to demonstrate that all members of the class would have been injured as result of a price-fixing conspiracy for aftermarket sheet metal parts and (2) whether it is possible to measure damages suffered by the DPP class on a classwide, rather than individual, basis. With regard to the first question, Dr. Lamb concluded that common evidence can demonstrate the alleged anti-competitive conduct would have resulted in artificially-inflated prices and that these artificially-inflated prices would have been paid by all or nearly all of the proposed class members. He based this conclusion on factors about the aftermarket sheet metal market, his conclusion that a price structure exists in the aftermarket sheet metal market, and a multiple regression analysis he conducted. With regard to the second question, Dr. Lamb concluded that his multiple regression analysis can be used to calculate damages on a classwide basis by measuring the percent by which class members were overcharged.
Multiple regression analysis is a statistical tool used to understand the relationship between two or more variables, Daniel L. Rubinfeld, Reference Guide on Multiple Regression, in Reference Manual of Scientific Evidence 303, 305 (3d ed. 2011), and is a widely accepted economic methodology, Manpower, Inc., 732 F.3d at 808. Generally, it involves a single dependent variable, which is the variable to be explained, and several explanatory variables, which are the factors thought to produce or be associated with changes to the dependent variable. Rubinfeld, supra, at 305. In this case, the dependent variable is the price of aftermarket sheet metal parts and the explanatory variables are various factors that could affect that price, for example demand, price of component parts, or anti-competitive conduct. By factoring out explanatory variables other than anticompetitive conduct, an economist can observe whether or not there is a correlation between price and the alleged anti-competitive conduct.
Dr. Lamb used multiple regression analysis to determine and then compare prices consumers paid during the alleged conspiracy with prices they would have paid in a free market absent anti-competitive conduct, known as the “but for” price. First Lamb Report at 38 (ECF No. 712-1). To determine the but for price, Dr. Lamb compared prices from the alleged period of time during which the anti-competitive conduct occurred, January 1, 2003 through September 4, 2009,  with a benchmark period, a period during which the anticompetitive behavior did not occur, here September 5, 2009 through December 31, 2011. Id. at 38-39. Dr. Lamb then calculated an overcharge percentage based on the difference between the actual prices paid and the but-for prices,  determining that prices were approximately 17.54% higher during the period of the alleged conspiracy than they would have been absent the alleged misconduct. Id. at 47.
Defendants argue that Dr. Lamb made several flawed assumptions which render his opinion unreliable. First, Dr. Lamb assumed that the misconduct alleged in the complaint occurred. Second, Dr. Lamb compared the alleged class period, January 1, 2003 through September 4, 2009, with a benchmark period, September 5, 2009 through December 31, 2011, and assumed that no anti-competitive activity occurred during this benchmark period. It is appropriate for experts to make assumptions in reaching their opinions as long as their assumptions are based on the expert’s field of study or supported by evidence. In re Ready-Mixed Concrete Antitrust Litig., 261 F.R.D. 154, 165 (S.D. Ind. 2009).
Dr. Lamb’s assumptions do not render his report inadmissible at the class certification stage because they are based on economics. With regard to Dr. Lamb’s assumption that the alleged misconduct existed, he explained that this assumption was necessary to reach proper conclusions regarding common impact. Second Lamb Report at 11 (ECF No. 813-1). The questions he was asked to answer, whether there was common proof of impact to all class members and whether damages can be calculated on a classwide basis, required him to assume that the alleged misconduct existed. See Michelle M. Burtis & Darwin V. Neher, Correlations and Regression Analysis in Antitrust Class Certification, 77 Antitrust L.J. 495, 500 (2011) (“At the class certification stage, a key question is whether there is common proof of impact to direct purchasers under the assumption that the alleged anticompetitive behavior took place.”); First Ordover Report at 3 (ECF No. 786-1) (defendants’ expert also assuming “that the DPP class is correct in its allegation that Defendant manufacturers . . . jointly agreed . . . to elevate prices”). For purposes of class certification, Dr. Lamb was not asked to determine whether defendants engaged in anti-competitive conduct, and thus Dr. Lamb did not assume the question he was asked to prove, as defendants assert.
With regard to Dr. Lamb’s benchmark period selection, again, he worked off of DPP’s theory of the case, assuming the allegations to be true, including the timing of the misconduct, as he should at the class certification stage. See Burtis & Neher, supra at 500. Further, Dr. Lamb explained that even if his benchmark period contains some anticompetitive conduct, it would only render his overcharge estimate conservative. This makes sense; if anti-competitive conduct occurred during the benchmark period, then the but for price Dr. Lamb calculated using the benchmark period would presumably be higher. Comparing a higher but for price with the actual price would result in a lower overcharge percentage, and thus even if Dr. Lamb’s benchmark period is incorrect, it would work against DPPs, not for them.
Next, defendants argue that Dr. Lamb did not control for all non-competitive factors that could explain a price correlation, pointing specifically to the fact that Dr. Lamb did not account for or explain a gradual decrease in prices from 2004 to 2009 that defendants’ expert found. According to defendants’ expert, from 2004 to 2009, prices of aftermarket sheet metal gradually decreased, and Dr. Lamb should have accounted for this decrease using a trend variable. An economist need not account for every possible variable in order for a regression analysis to be admissible. Bazemore v. Friday, 478 U.S. 385, 400 (1986) (Brennan, J., concurring) (“[I]t is clear that a regression analysis that includes less than ‘all measurable variables’ may serve to prove a plaintiff’s case.”). “While the omission of variables from a regression analysis may render the analysis less probative than it otherwise might be, it can hardly be said, absent some other infirmity, that an analysis which accounts for the major factors must be considered unacceptable as evidence.” Id. “Normally, failure to include variables will affect the analysis’ probativeness, not its admissibility” unless the regression analysis is “so incomplete as to be inadmissible as irrelevant.” Id. at 400 & n.10; see also Manpower, Inc., 732 F.3d at 808.
Here, Dr. Lamb’s failure to account for the alleged price decrease does not render his regression analysis so incomplete as to render it inadmissible. First, Dr. Lamb did account for other major factors in his regression analysis, and thus his regression analysis appears to be substantially complete. See First Lamb Report at 43-46 (listing the numerous variables applied to Dr. Lamb’s regression model). Additionally, Dr. Lamb gives a reason, based on his experience as an economist, as to why he did not account for such a decrease in his regression analysis, namely he does not believe that the alleged gradual decrease is supported by evidence. Second Lamb Report at 35. Thus, whether there was such a decrease and whether that decrease required the application of a trend variable is a fact question and does not bear on admissibility. Smith v. Ford Motor Co., 215 F.3d 713, 718 (7th Cir. 2000) (“The soundness of the factual underpinnings of the expert’s analysis and the correctness of the expert’s conclusions based on that analysis are factual matters to be determined by the trier of fact.”).
Defendants also attack Dr. Lamb’s reliance on qualitative market characteristics to support his conclusion that common evidence can be used to prove a classwide impact of price-fixing because the use of qualitative market characteristics is subjective and unreliable. “[A]ntitrust impact based on a simple description of general market characteristics cannot be presumed.” Burtis & Neher, supra, at 501. However, in this case, Dr. Lamb did not rely solely on market characteristics to reach his conclusion that there is common evidence that can prove a classwide impact. Rather, Dr. Lamb’s conclusions were based on market characteristics, his regression analysis, and his conclusion that a price structure existed in the aftermarket sheet metal industry. Together, this analysis is reliable enough to pass the Daubert standard. To the extent that defendants’ expert disagrees with Dr. Lamb’s conclusions regarding market characteristics, those are fact issues best left to the factfinder. Daubert, 509 U.S. at 595 (“The focus, of course, must be solely on principles of methodology, not on the conclusions that they generate.”).
Finally, defendants argue that the methodology Dr. Lamb used to conclude that a price structure exists in the aftermarket sheet metal industry is flawed because his analysis only shows that the price of single parts moved together. This is not enough, defendants argue, to infer that all class members were impacted because the analysis does not show that the prices of different parts also moved together, which is needed to show that virtually all class members were impacted by such a price structure. Defendants expert, analyzing the same data as Dr. Lamb, concludes that in fact no price structure existed because the price of different parts frequently moved in opposite directions. First Ordover Report at 7-8. Here, defendants do not appear to be attacking Dr. Lamb’s methodology as much as they are attacking his conclusion that a price structure exists. Whether that is true is a fact issue and not appropriately addressed via a Daubert motion.
For these reasons, I conclude that defendants have failed to show that Dr. Lamb’s methodologies are inadmissible. Dr. Lamb has adequately accounted for obvious alternative explanations, and it appears that he has exercised a great deal of care in conducting his analysis. Fed.R.Evid. 702 Advisory Committee’s Notes (2000 Amends.) (listing benchmarks for gauging reliability). Thus, his opinions are relevant and will assist the trier of fact. Whether his opinion ...