Opinion
No. 02-4478.
Submitted Under Third Circuit LAR 34.1(a) December 12, 2003.
Decided January 9, 2004.
Appeal from the United States District Court for the District of New Jersey. (D.C. Civil Action No. 01-cv-02100). District Judge: Honorable Joel A. Pisano.
Thomas S. Howard, Kirsch, Gartenberg Howard, Hackensack, NJ, for Appellant.
Michael K. Furey, Sigrid S. Franzblau, Riker, Danzig, Scherer, Hyland, Perretti, Morristown, NJ, Thomas L. Weisenbeck, Bressler, Amery Ross, Florham Park, NJ, Peter J. Meyer, Gardner, Carton Douglas, Chicago, IL, William W. Robertson, Robertson, Freilich, Bruno Cohen, Newark, NJ, for Appellee.
Before AMBRO, FUENTES and GARTH, Circuit Judges.
OPINION
Joseph Weisfeld appeals the decision of the District Court denying his motion for class certification. Weisfeld filed a civil antitrust lawsuit alleging that companies in the ink printing industry illegally restrained the labor market in that industry by entering into a series of "no hire" agreements. In October 2002 the District Court denied Weisfeld's motion for class certification, stating he failed to satisfy the requirements of Rule 23(b)(3) of the Federal Rules of Civil Procedure ("FRCP"). Thereafter, Weisfeld timely filed a petition pursuant to Rule 5(a) of the Federal Rules of Appellate Procedure and Rule 23(f) of the FRCP for leave to appeal the District Court's decision. We granted Weisfeld permission to appeal and have jurisdiction pursuant to 28 U.S.C. § 1292(e). For the reasons that follow, we affirm the District Court's decision.
I.
Weisfeld was, until May 31, 2000, an employee of Sun Chemical Corporation ("Sun"). In his First Amended Complaint, Weisfeld brings suit on behalf of himself and
all persons who were employed by defendants, or any predecessor, affiliate or subsidiary of any defendant, at any time during the period beginning at least as early as May 1, 1997 and continuing through May 1, 2001 inclusive (the "Class Period"), and who suffered damages as a result of defendants illegal conspiracy and violation of the antitrust laws.
Despite failing to revise his complaint, Weisfeld sought to narrow the definition of the class in his motion for class certification. The new proposed class was defined as
personnel who provide technical services and who possess specialized knowledge and skills in the manufacture, distribution and sale of printing inks who were employed by defendants, or any predecessor, affiliate or subsidiary of any defendant, at any time during the period beginning at least as early as May 1, 1997 and continuing through May 1, 2001 (the "Class Period"), and who suffered damages as a result of defendants' illegal conspiracy and violation of the antitrust laws.
The District Court considered this revised class definition in its analysis, and we will do the same. See, e.g., Robidoux v. Celani, 987 F.2d 931, 937 (2d Cir.1993) (stating that a court "is not bound by the class definition proposed in the complaint").
Given the District Court's well-written summation of the facts underlying Weisfeld's complaint, we repeat them only briefly here. See Weisfeld v. Sun Chemical Corp., 210 F.R.D. 136 (D.N.J.2002). Weisfeld alleges that defendants, who control a dominant market share in the manufacturing of printing inks, entered into a conspiracy in violation of Section One of the Sherman Act, 15 U.S.C. § 1, to restrain the labor market for technical employees in that industry. Specifically, Weisfeld alleges that defendants implemented "no hire" policies by which each agreed not to hire the employees of their competitor printing ink manufacturers. According to Weisfeld, these agreements adversely affected the national market for employees with "specialized knowledge and skills" in the ink printing industry, resulting in lower salaries and lost opportunities to seek more gainful employment within the industry.
II.
We review a decision granting or denying class certification for abuse of discretion. In re LifeUSA Holding Inc., 242 F.3d 136, 143 (3d Cir.2001). The District Court abused its discretion if its decision "rests upon a clearly erroneous finding of fact, an errant conclusion of law or an improper application of law to fact." In re General Motors Corp. Pick Up Truck Fuel Tank Prods. Liab. Litig., 55 F.3d 768, 783 (3d Cir.1995) (citations and quotations omitted). Although a court should be cautious in making a class certification decision based on its own impression of a case's merits, some inquiry into the factual and legal issues underlying a plaintiffs causes of action is usually necessary. See Newton v. Merrill Lynch, Pierce, Fenner Smith, Inc., 259 F.3d 154, 166-168 (3d Cir.2001).
III.
In denying class certification in this case, the District Court concluded that Weisfeld failed to meet both the predominance and superiority requirements of Rule 23(b)(3). On appeal, he alleges three errors: (1) the District Court analyzed the no hire agreements under a rule of reason standard rather than as a per se violation; (2) the Court misinterpreted our decision in In re Linerboard Antitrust Litig., 305 F.3d 145 (3d Cir.2002), as increasing the amount of proof required to show predominance at the class certification stage; and (3) it concluded that a class action was not the superior method for resolving the dispute because the determination of detrimental impact would involve too many individual issues. To an extent, all three issues involve whether Weisfeld has sufficiently demonstrated that antitrust injury, or impact, is susceptible to common proof. Although we disagree with certain statements made by the District Court as detailed below, we agree with its overall analysis and conclusions.
Rule 23(b)(3) of the FRCP provides that an action may be maintained as a class action if, in addition to meeting the requirements of subsection (a), "the court finds that the questions of law or fact common to the members of the class predominate over any questions affecting only individual members, and that a class action is superior to other available methods for the fair and efficient adjudication of the controversy."
Section One of the Sherman Act states: "Every contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States, or with foreign nations, is hereby declared to be illegal." 15 U.S.C. § 1. Courts have interpreted this language to prohibit two types of unreasonable restraints of trade. They are conduct that is unreasonable per se and conduct that violates the "rule of reason." See FTC v. Indiana Fed'n of Dentists, 476 U.S. 447, 106 S.Ct. 2009, 90 L.Ed.2d 445 (1986). "Under the per se test, 'agreements whose nature and necessary effect are so plainly anti-competitive that no elaborate study of the industry is needed to establish their illegality' are found to be antitrust violations.'" Eichorn v . AT T Corp., 248 F.3d 131, 138 (3d Cir.2001) (quoting Nat'l Soc'y of Prof. Eng'rs v. United States, 435 U.S. 679, 682, 98 S.Ct. 1355, 55 L.Ed.2d 637 (1978)). Conduct that does not constitute a per se violation is analyzed under the rule of reason test. Eichorn, 248 F.3d at 138. Based largely upon our decision in Eichorn, the District Court concluded that the no hire agreements did not constitute a per se antitrust violation.
While Weisfeld argues a per se violation occurred in this case, we conclude that the issue is irrelevant to whether the requirements of Rule 23(b)(3) are met for class certification purposes. It appears the confusion stems from the following statement made by the District Court: "Plaintiff argues that when a defendant has been accused of price fixing, impact should be presumed as a matter of law. Antitrust injury may be presumed when a plaintiff alleges a per se violation of the antitrust laws." Weisfeld, 210 F.R.D. at 142. This statement goes further than does the Supreme Court, for it has specifically rejected the contention that antitrust injury may be presumed in the case of a per se antitrust violation. Atlantic Richfield Co. v. USA Petroleum Co., 495 U.S. 328, 341, 110 S.Ct. 1884, 109 L.Ed.2d 333 (1990).
As we have recognized, the "Supreme Court has been cautious in extending the per se approach to claims that fall outside certain previously enumerated categories of liability." Eichorn, 248 F.3d at 143 (citations omitted). In Eichorn, we held that the no hire agreement at issue did not constitute a per se violation and, in fact, did not constitute an unreasonable restraint of trade at all. Id. at 144, 145-146. But the facts in Eichorn are different than those here. In Eichorn, AT T adopted a policy not to allow employees of Paradyne Corp., an AT T affiliate, to transler to other divisions of AT T. The purpose of the policy was to make Paradyne more attractive to potential buyers. Shortly thereafter, Paradyne was sold, and AT T entered into a post-closing agreement in which it agreed not to hire, solicit or rehire any Paradyne employee or consultant whose compensation exceeded $50,000 for a period of 245 days (8 months). Id. at 136-37. In this case, however, Weisfeld alleges that the three largest manufacturers in the ink printing industry conspired not to hire each others' employees for a period of four years.
"Per se and rule-of-reason analysis are but two methods of determining whether a restraint is 'unreasonable,' i.e., whether its anticompetitive effects outweigh its procompetitive effects.
. . .
The purpose of the antitrust injury requirement is different. It ensures that the harm claimed by the plaintiff corresponds to the rationale for finding a violation of the antitrust laws in the first place. . . .
Id. (citation omitted).
It is true that In re Linerboard used the phrase "presumption of impact" in discussing the "Bogosian short cut." In re Linerboard, 305 F.3d at 151; see also Bogosian v. Gulf Oil Corp., 561 F.2d 434 (3d Cir. 1977). In Bogosian, we stated that "when an antitrust violation impacts upon a class of persons who do have standing, there is no reason in doctrine why proof of the impact cannot be made on a common basis so long as the common proof adequately demonstrates some damage to each individual." 561 F.2d at 454. Properly understood, therefore, the presumption (or "short cut") in Bogosian is that, even if issues requiring individualized proof are present, evidence of a common antitrust injury to every class member flowing from an alleged antitrust violation can, in certain circumstances, satisfy the requirements of Rule 23(b)(3). See Newton, 259 F.3d at 180 n. 21 (citing Bogosian for the proposition that antitrust injury "may be presumed when it is clear the [antitrust] violation results in harm to the entire class") (emphasis added). The "Bogosian short cut," however, does not allow a court to presume antitrust injury because a per se violation occurred.
Weisfeld's second claim of error is that the District Court misinterpreted our decision in In re Linerboard as to the type and amount of evidence required to demonstrate whether antitrust injury is susceptible to common proof for class certification purposes. In support of his motion for class certification, Weisfeld provided a three-page expert declaration. The expert states he planned to use two approaches, multiple regression and yardstick analysis, as common proof that the no hire agreements artificially depressed the compensation levels of class members. The District Court stated that it was "not convinced" that Weisfeld's expert "will be able to prove impact on a classwide basis." Weisfeld, 210 F.R.D. at 144. Contrary to the support presented by the plaintiffs' experts in In re Linerboard, the District Court found telling that Weisfeld's expert provided only "naked conclusions" that common proof would demonstrate injury to class members. Id. at 144-45. Weisfeld alleges the District Court improvidently rejected his generalized proof of common impact, in effect requiring specific proof of classwide impact as if the case were ready for trial.
Generally speaking, regression analysis seeks to define statistically the relationship between a dependent variable ( i.e., salary) and one or more independent variables (e.g., education or work experience). Multiple regression analysis allows one to control for other independent variables so as to isolate and identify the effect of a single independent variable ( i.e., the alleged no hire agreements) on the dependent variable.
A yardstick analysis seeks to identify the effect of an independent variable on a dependent variable ( i.e., the effect of the alleged no hire agreements on salary) by comparing the dependent variable in one industry ( i.e., salary in the ink printing industry) with that in another (or yardstick) industry. Selection of the yardstick industry is based on its similarity-for example, in terms of structure, product demand, capital, labor and materials input trends and product technology — to the industry being investigated.
We conclude, however, that the District Court did not misinterpret our prior decisions. Weisfeld correctly points out that multiple regression and yardstick analyses have been widely accepted in this Circuit. As discussed in greater detail below, however, Weisfeld's expert provides no independent analysis or evidence to support his conclusions of common impact. In In re Linerboard, we recognized that, in granting class certification, "the district court used a belt and suspenders rationale to support its conclusion that the putative class had met its burden of showing impact. In addition to relying on the Bogosian short cut, it credited the testimony of plaintiffs' experts. . . ." 305 F.3d at 153. We found it significant, however, that those experts engaged in "extensive empirical investigation" and economic analysis. Id. at 513. Further, the experts "effectively utilized supporting data, including charts and exhibits, to authenticate their professional opinions." Id. at 155. The In re Linerboard plaintiffs, therefore, did not rely "solely on presumed impact and damages" and adequately demonstrated "that they could establish [antitrust] injury on a class-wide basis." Id. While the District Court should not have decided (and, in fact, did not decide) at the class certification stage whether Weisfeld could prove that the alleged no hire agreements had a common impact on the class members, it was entirely proper for the Court to decide whether Weisfeld's attempt to prove antitrust injury would predominately involve common legal and factual questions.
Further, the cases Weisfeld cites in arguing the District Court misapplied In re Linerboard belie his assertion that a court may not disregard the unsupported conclusions of a expert. The expert in In re Mercedes-Benz Antitrust Litig., 213 F.R.D. 180, 189-90 (D.N.J.2003), sampled prices charged and dealer gross profits in the relevant market area to support his conclusion that common proof could demonstrate generalized antitrust injury within a certain range. In addition, In re Microcrystalline Cellulose Antitrust Litig., 218 F.R.D. 79, 89-93 (E.D.Pa.2003), engaged in an extensive analysis of the opinions and supporting evidence provided by plaintiffs' various experts before concluding the predominance requirement of Rule 23(b)(3) had been met. The Court in that case stated it "considered not only the pleadings and memoranda, but also the expert declarations and factual support on record." Id. at 83 (emphasis added); see also Nichols v. SmithKline Beecham Corp., Civ. A. No. 00-6222, 2003 WL 302352 at *4 (E.D.Pa. Jan.29, 2003) (stating that, at the class certification stage, a court should examine whether an expert "has identified a generally accepted methodology for determining impact which is applicable to the class, whether this methodology uses evidence common to all class members and whether his opinion has probative value") (emphasis added) (citation omitted).
We must decide at this stage only whether the District Court abused its discretion in determining that (1) common issues did not predominate in the antitrust injury inquiry and (2) a class action was not a superior means to adjudicate this case. The predominance requirement of Rule 23(b)(3) ensures that a proposed class is "sufficiently cohesive to warrant certification." In re Newton, 259 F.3d at 187 (citing Amchem Products, Inc. v. Windsor, 521 U.S. 591, 623, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997)). "Unlike commonality, predominance is significantly more demanding, requiring more than a common claim." In re Newton, 259 F.3d at 187 (citing Amchem, 521 U.S. at 623-24, 117 S.Ct. 2231). As the District Court correctly realized, however, Rule 23(b) does not require the absence of individual issues, only that they have less overall significance than issues common to the class. Some individual issues will almost always be present in a class action. Further, we are well aware antitrust defendants "routinely argue" that the complexities of their particular case prevent common proof from predominating on the issue of anti-trust impact. In re Mercedes-Benz Anti-trust Litig., 213 F.R.D. at 187 (citations omitted). We conclude, however, that the District Court did not abuse its discretion in this case.
The superior means requirement is closely related to the predominance requirement because it is generally true that "only where predominance exists do the economics of scale justify aggregating claims in a class action." In re Mercedes-Benz Antitrust Litig., 213 F.R.D. at 186. This is such a case. Therefore, because we conclude the District Court did not abuse its discretion in determining that common issues did not predominate, we need not engage in the largely duplicative superior means analysis.
In analyzing whether common issues will predominate in the antitrust injury inquiry, we follow the teachings of Bogosian and its progeny. In Bogosian, we relied on the fact that the alleged anticompetitive conduct had predictable effects.
If the price structure in the industry is such that nationwide the conspiratorially affected prices at the wholesale level fluctuated within a range which, though different in different regions, was higher in all regions than the range which would have existed in all regions under competitive conditions, it would be clear that all members of the class suffered some damage, notwithstanding that there would be variations among all dealers as to the extent of their damages.
Bogosian, 561 F.2d at 455 (emphasis added). Since neither the parties nor the District Court in Bogosian focused on these issues, we remanded the case for further consideration. Id. In In re Liner-board, we found "significant" the findings of plaintiffs' expert that "[d]espite any variations in particular boxes or customers, prices for all corrugated containers would have responded over time to linerboard price increases in a similar manner." 305 F.3d at 153 (emphasis added) (quotations omitted). Even the cases Weisfeld cites recognize a plaintiff must present some evidence that the antitrust injury suffered by class members is within a manageable range. In In re Mercedes-Benz, plaintiffs' expert found that car prices remained "within a small range for the seven dealers sampled." 213 F.R.D. at 189. In such a situation, the Court concluded other variables could be controlled for using benchmark and multiple regression analysis. Id. The expert's analysis in In re Microcrystalline Cellulose showed that, even though the alleged antitrust activity affected numerous products, the prices of all products in this industry still moved closely together. 218 F.R.D. at 92.
Although rejecting the notion that antitrust injury in an employee boycott or no hire context can never be proven by common evidence, we affirm the District Court's conclusion that common issues do not predominate in this case. In addition to decreased salaries, Weisfeld alleges the deprivation of new job opportunities. Yet Weisfeld testified that he neither sought employment outside of Sun nor responded to the inquiries of "head hunters." This demonstrates that, at least in regard to lost employment opportunities, an individualized inquiry may be required for each class member. In regard to salary levels, the proposed class includes individuals employed in positions that vary widely in terms of skill requirements and responsibilities. Indicative of this is that, at Sun alone, these positions have more than 20 compensation grades. The District Court correctly notes that both the decreased salary and deprivation of opportunities inquiries would require considering numerous individual factors. These include: "whether a covenant not to compete was included in a particular employee's contract; the employee's salary history, educational and other qualifications; the employer's place of business; the employee's willingness to relocate to a distant competitor; and [employees'] ability to seek employment in other industries in which their skills could be utilized ( e.g., pharmaceuticals, cosmetics)." Weisfeld, 210 F.R.D. at 144.
Finally, the declaration proffered by Weisfeld's expert provides no guidance. For class certification purposes, an expert's opinion need not be supported by the type of specific evidence that would be used at trial (or even to survive summary judgment). But s/he must present some support. Weisfeld's expert claims he conducted a preliminary analysis. Yet there is no actual analysis in the declaration itself and no discussion of the evidence on which his analysis was based. There are no studies or findings attached to the declaration. In addition, the declaration is silent as to whether the salaries for all members of the proposed class exhibit any sort of correlation over time. The declaration contains only bare conclusions and a statement that the expert "proposes" to use a multiple regression model (which may not take into account all relevant variables). Such support is insufficient to satisfy the predominance requirement of Rule 23(b)(3) in this case.
IV.
In this context, we affirm the District Court's decision denying Weisfeld's motion for class certification.