Opinion
No. 90 C 6400
March 16, 2001
MEMORANDUM OPINION AND ORDER
Before the court is a motion by defendants to modify the class period set forth in this court's May 14, 1999 memorandum opinion and order certifying a class action. Panache Broadcasting of Pennsylvania, Inc. v. Richardson Electronics, Ltd., 1999 WL 342392 (N.D. Ill. May 14, 1999). Defendants argue that the ending date of the class period was inadvertently and unjustifiably extended more than six years longer than the date originally sought by plaintiffs in their motion for certification. Defendants point out that, at the time of the briefing on class certification, the parties only addressed the issue of whether the class should be certified and did not address the length of the class period. Defendants thus believe that it is appropriate to now consider this issue for the first time. Also before the court are plaintiffs' objections to the Magistrate Judge's discovery ruling.
BACKGROUND
The following facts are relevant to the motion to modify the class period. On March 17, 1993, plaintiffs filed their Second Amended Complaint (the "complaint") seeking treble damages and injunctive relief under Sections 1 and 2 of the Sherman Act and Section 7 of the Clayton Act. Plaintiffs generally allege that Richardson Electronics, Ltd. ("Richardson" or "REL") and Varian Associates, Inc. ("Varian") conspired to eliminate competition in the market for electron power tubes. Plaintiffs specifically allege that the conspiracy was effectuated through a joint venture known as the Varian Supply Company ("VASCO"). VASCO was created pursuant to a series of written agreements entered into by the defendants in 1986. Under these agreements, known as the VASCO agreements, the defendants allegedly agreed to control prices, exchange information about customers, and to acquire competitors. Shortly after VASCO was formed, the prices of electron tubes allegedly increased sharply to artificially high levels.
A more complete summary of the background facts is set forth in this court's certification ruling, see Panache Broadcasting of Pennsylvania, Inc. v. Richardson Electronics, Ltd., 1999 WL 342392 (N.D. Ill. May 14, 1999), and in the Magistrate Judge's 9/27/95 Report and Recommendation.
VASCO was created on February 26, 1986. Sometime around 1989, the Justice Department began investigating VASCO for possible antitrust violations. As a result of this investigation, in 1991, Richardson and Varian each pleaded guilty to a felony violation of Section 1 of the Sherman Act relating to certain activities of VASCO. See United States v. Richardson Elecs., Ltd., No. 91 CR 0791 (N.D. Ill. Oct. 7, 1991) (Judge Marovich); United States v. Varian Assocs., Inc., No. 91 CR 0946 (N.D. Ill. Dec. 4, 1991) (Judge Lindberg). On April 1, 1992, the defendants entered into a consent decree, which (among other things) required that VASCO be dissolved effective April 1, 1992. Although the complaint focuses primarily on the allegations relating to the VASCO joint venture, the complaint also contains allegations that the defendants engaged in anticompetitive conduct both before VASCO was created and after it was dissolved. In addition, VASCO did not involve all of the electron tubes manufactured and distributed by the defendants.
The decree was entered by Judge Lindberg and remains in effect until April 1, 2002.
On January 13, 1995, plaintiffs filed a motion for class certification. They requested certification of the following class:
Technically, this motion for class certification is an "amended" motion. The original certification motion was filed on October 15, 1990, while the case was still pending in the Eastern District of Pennsylvania.
All persons and business entities who purchased electron power tubes from one or more of the defendant corporations at any time during the period from February 26, 1986 through March 16, 1993,. . . .
On September 27, 1995, (then) Magistrate Judge Pallmeyer issued a 51-page Report and Recommendation ("RR"), in which she rejected certification of the broader class proposed by plaintiffs and instead recommended certification of a narrower class limited to purchasers of tubes related to VASCO and the VASCO agreements. She defined the class as consisting of
all persons and business entities who purchased electron tubes that were (1) sold through VASCO, (2) related to the dud tube collection program instituted by Defendants, or (3) related to acquisitions by the Defendants arising out of the VASCO agreements, from one or more of the defendants at any time between February 26, 1986 and the present.
(RR at 50.)
The parties then filed objections to the RR. With only minor revisions, we adopted the RR in our May 14, 1999 opinion and certified the class recommended by the Magistrate Judge. We described the certified class as
all persons and business entities who purchased electron tubes that were (1) encompassed in the VASCO agreements; (2) related to the dud tube collection program instituted by Defendants, or (3) related to acquisitions by Defendants arising out of the VASCO agreements, from one or more of the defendants at any time between February 26, 1986 and May 13, 1999.Panache, 1999 WL 342392 at *10.
In response to plaintiffs' objections, we agreed to substitute the phrase "encompassed in the VASCO agreements" in place of the phrase "sold through VASCO" in clause (1) of the class definition.
In the briefing before both the Magistrate Judge and this court, the parties focused exclusively on the question of whether a class should be certified and not on the length of the class period. As described below, although the issue of the length of the class period was never briefed or analyzed, the end date of the class period was gradually extended from March 16, 1993 to May 13, 1999.
In their 1993 complaint, plaintiffs defined the class period as running from February 26, 1986 to "the present." At that time, "the present" was March 16, 1993 — the date of the complaint. In their 1995 motion for class certification, plaintiffs did not use "the present" language but instead inserted the date of the earlier-filed complaint — March 16, 1993. In her 1995 RR, the Magistrate Judge inserted "the present" language from the complaint. This meant that the end date of the class period had thus been extended from March 16, 1993 to September 27, 1995, which is the date of the RR. In a similar manner, we further extended the class period to May 13, 1999. Without analyzing the issue, we set the end date of the class period simply by taking "the present" language in the RR and converting it to a specific date — May 13, 1999, the day before the opinion was issued. In sum, over time, but without notice or comment by anyone, the end date of the class period was extended more than six years beyond the date originally sought by plaintiffs.
Defendants state that this issue was dormant and only surfaced when plaintiffs made a document request for current customer lists in connection with the mailing of the class certification notice. As a result, on June 15, 2000, defendants filed their present motion to modify the order certifying the class. They believe that the motion is akin to a motion to correct a technical defect. As to the substantive question raised by the motion, defendants assert that the class period should end on April 1, 1992, which is the date that VASCO was dissolved pursuant to the consent decree, or that at the latest, it should end on March 16, 1993, which is the date plaintiffs originally requested in their 1995 certification motion. Anything beyond March 16, 1993 — according to defendants — would be an unjustified windfall for plaintiffs. This court set a briefing schedule and subsequently allowed each side to file two additional surreply briefs.
Specifically, we reviewed the following: (1) defendants' 6/15/00 Motion to Modify the Order Certifying the Class; (2) plaintiffs' 7/25/00 memorandum in opposition; (3) defendants' 8/15/00 reply; (4) plaintiffs' 9/05/00 surreply; (5) defendants' 9/21/00 sur-surreply; (6) plaintiffs' 10/17/00 supplemental memorandum; and (7) defendants' 2/26/01 response to plaintiffs' supplemental memorandum.
DISCUSSION
I. The Motion To Modify Class Period.The first issue, which we will address only briefly, is whether it is appropriate at this point to consider modifying our class certification order issued on May 14, 1999. The short answer is yes. Under Rule 23(c)(1), this court clearly has the authority to alter or amend its class certification order at any time before a decision on the merits. Moreover, as explained above, it is clear that we never considered the issue of whether the class should be extended beyond the original March 16, 1993 date. This was not the fault of the defendants because plaintiffs' 1995 motion only sought to certify a class up until that point. The issue of extending the class period for an additional six years — an extension that nearly doubles the length of the class period — is obviously an important issue that should be decided on the merits if possible. For these reasons, it is appropriate to now analyze whether the ending date of the class period should be shortened from the current date of May 13, 1999 to March 16, 1993 or possibly to some intermediate date in between.
At the same time, absent convincing reasons, we will not re-visit issues that were previously considered or issues that fairly could have been raised by the parties. Thus, we will not consider shortening the date to April 1, 1992 as requested by defendants. Defendants were aware at the time of the briefing on the original motion that plaintiffs were seeking a class period that extended at least until March 16, 1993, and defendants did not object at that time. This holding is not really material as defendants have now stated that they are willing to accept an ending date of March 16, 1993, although they clearly believe that the earlier April 1, 1992 date is even more appropriate.
We do not believe, as defendants argue, that plaintiffs should be held to the March 16, 1993 date they included in their 1995 certification motion. Plaintiffs state that this one reference to a specific date was inadvertent and that they otherwise always used the phrase "to the present" and always intended for this phrase to refer to the latest possible point in time.
We now turn to the substantive issue; namely, at what point should the class period end? Defendants argue that the class period should end soon after VASCO ended because the entire basis for class certification was the common proof linked to the VASCO joint venture. Defendants state — and plaintiffs do not appear to dispute — that VASCO effectively ceased operation in late 1991 and was dissolved on April 1, 1992 pursuant to the consent decree. Defendants argue that it is fair to presume that the consent decree effectively ended the alleged conspiracy. Moreover, they point out that Varian exited the electron power tube market altogether in 1995.
Plaintiffs, who have the burden of establishing that the requirements for class certification have been met (RR at 13), offer two arguments in response. First, they argue that, despite the existence of the consent decree, defendants continued their illegal conduct after VASCO was dissolved. Second, they argue that the effects of the VASCO activity continued long after VASCO was dissolved and in fact are still continuing today. Plaintiffs thus contend that, even though this issue was never addressed, there is in fact ample justification for leaving the current date of May 13, 1999 in place. For the reasons explained below, we are not persuaded by either of these two arguments and therefore conclude that the class period should be shortened back to March 16, 1993.
In fact, in their briefs, plaintiffs suggest that the date really should extend beyond May 13, 1999 and should continue to the present. However, plaintiffs have not made a formal request to extend the date in this manner.
A. Post-VASCO Misconduct.
The complaint, which is 36-pages long, contains only one allegation concerning post-VASCO conduct. It states:
On information and belief, despite the formal dissolution of VASCO in or about April 1992, and despite the formal elimination of exclusivity from REL's distribution contract with Varian, both pursuant to the Consent Decree, REL and Varian have continued by express or tacit agreement to refrain from competing significantly with each other.
(¶ 45). Defendants initially complained that the above allegation — that the conspiracy continued "by express or tacit agreement" — is conclusory and is pled on information and belief, in fairness to plaintiffs, the complaint was filed on March 16, 1993 and therefore could not really speak to the post-1993 activities. In their October 17, 2000 supplemental memorandum, plaintiffs submitted affidavits and other materials in an attempt provide some specificity to their post-VASCO allegations. Among other things, plaintiffs have submitted affidavits of two individuals who claim that the prices of certain electron tubes have remained at the same allegedly high levels after VASCO was dissolved that existed while VASCO was in operation. Plaintiffs also point out that Richardson entered into a new distributor agreement with Varian sometime after the consent decree.
Defendants argue that the above allegations and evidence are sketchy and are insufficient to overcome a presumption that the 1992 consent decree ended the alleged conspiracy and restored competition to the market. Defendants emphasize that the consent decree comprehensively regulates the parties' conduct in the post-VASCO period. For example, in addition to mandating the dissolution of VASCO, the consent decree prohibits or severely limits the dud tube collection program, acquisitions of competitors, exclusive distribution arrangements between the parties, profit sharing, price fixing and resale price maintenance. Also, the decree gives the Justice Department the right to inspect documents and interview employees to make sure the parties are abiding by these restrictions. Likewise, the district court retained jurisdiction for 10 years to aid in the enforcement of the decree. In light of all these restrictions, and in the absence of specific evidence to the contrary, defendants argue that it is fair to assume that the anticompetitive conduct ended on April 1, 1992.
In response, plaintiffs state that, "[w]hile it may be typical for the issuance of a consent decree to bring about a cessation of the conduct covered by the decree, no consent decree addresses all conduct that is violative of the antitrust laws, and there are (unfortunately) not infrequent examples of parties who fail to fully comply with decrees." (Pls. 9/5/00 Surreply at 3.) Plaintiffs believe that they have come forward with enough evidence to show that, in this particular case, the consent decree did not end the anticompetitive activity. At a minimum, they assert that the question is a factual one that should be resolved at trial or on a motion for summary judgment.
We agree with the above two points made by plaintiffs. First, plaintiffs are correct that allegations in support of a motion for certification are generally presumed. Second, we have not found any clear directive in the case law stating that a consent decree creates a presumption that all antitrust activity ends on the date the decree was entered into. However, the above two points only take plaintiffs so far.
Of course, it is also true, as the Magistrate Judge noted in her RR, that a litigant may not rest on mere conclusory allegations, and must make a minimal factual showing that the conditions for certification exist. (RR at 13.)
The question before us is whether to certify a class action covering the alleged post-VASCO misconduct. Therefore, it is not enough to simply come forward with allegations of anticompetitive activity in the post-VASCO period. Instead, plaintiffs must meet the class certification requirements of Rule 23 and in particular the predominance requirement of subsection (b)(3). As discussed below, even assuming plaintiffs' allegations are true and that the defendants have engaged in post-VASCO misconduct, plaintiffs have not met their burden of showing that common questions will predominate over individual questions as to this post-VASCO misconduct.
To explain why, it will be helpful to summarize the Magistrate Judge's analysis of this same general question. Defendants argued to the Magistrate Judge that the original class proposed by plaintiffs was too unwieldy because it lumped together purchasers of many different types of electron tubes. There are over 15,000 different tube types in the electron power industry, and these tubes vary in both size and price and generally cannot be substituted for one another. (RR at 5-6.) Each tube has unique applications and purchasers, and thus each tube has a unique market. ( Id. at 17.) Defendants argued that a violation of the antitrust laws relating to one tube type would not necessarily lead to the conclusion that there was a violation relating to all tube types. ( Id. at 28.) Defendants thus argued that plaintiffs could not show that common questions would predominate over individual questions. The Magistrate Judge agreed with this argument and refused to recommend certification of the broad-ranging class sought by plaintiffs because it did not satisfy the predominance requirement of Rule 23(b)(3).
Although she rejected the class proposed by plaintiffs, the Magistrate Judge concluded that a "more restricted" class could satisfy the predominance requirement of Rule 23(b)(3). Specifically, she concluded that a class limited to purchasers of products related to the VASCO agreements could satisfy the predominance requirement of Rule 23(b)(3). Even though this narrower group of VASCO-related purchasers still comprised a fairly diverse segment of purchasers that raised concerns about manageability of the class, the group was linked by the fact that all the claims were "in some way tied to the evidence" relating to the VASCO agreements. ( Id. at 32-33.) She noted that most of the "specific allegations" in the complaint related to the VASCO agreements. In contrast, she found that there was "meager evidence" of how plaintiffs intended to establish the conspiracy relating to the non-VASCO related purchasers and that those purchasers would have to establish a conspiracy "in some other fashion." ( Id. at 32.) She recognized that there might be "a few questions" of fact or law common to both the VASCO-related and non-VASCO-related purchasers, but concluded that such questions would not predominate if the class included both groups of purchasers. ( Id.) We agreed with this analysis and noted that the non-VASCO-related purchasers would need "individual proof" of a "separate" anticompetitive conspiracy. Panache, 1999 WL 342392 at *5.
In sum, all along, there has been a concern that the diverse nature of the electron tube industry was not suitable for a class action. The VASCO agreements provided the common link to overcome this problem. When the conspiracy extends in time beyond the existence of VASCO, it is not clear that this common link still exists. Although plaintiffs allege that the VASCO conspiracy continued by "by express or tacit agreement," they do not provide any specific description of how that conspiracy operated. In their supplemental memorandum, plaintiffs refer to the fact that there was a post-VASCO distributor agreement between the defendants, but they have not shown that this agreement functioned in the same manner as the prior arrangement involving VASCO.
By contrast, the complaint describes in greater detail how the how the conspiracy operated when VASCO was in existence. Specifically, the complaint alleges that the defendants operated VASCO through the JV Management Committee, which was composed of senior executives of Varian and Richardson. It further alleges that the members of this committee met "regularly" and that they exchanged "information regarding prices, costs, customers and customer allocation, business plans, potential acquisitions of other competitors' businesses or assets, and activities directed against tube rebuilders." (Cmplt. ¶ 43.)
Plaintiffs think that it is reasonable to assume that the VASCO conspiracy continued in operation after VASCO was dissolved. In light of the consent decree, however, it is not reasonable to assume that the conspiracy simply continued operating in the same way as it did before the consent decree. VASCO was dissolved in 1992. The consent decree imposed numerous restrictions on the defendants' post-VASCO activities. The parties were well aware that the Justice Department had the power to examine their operations and could quickly bring any violations to the district court's attention. In sum, although it may not be fair to presume that the consent decree in fact ended all anticompetitive conduct, it is reasonable to presume that the consent decree changed the nature of that conduct. For example, it is very likely that the consent decree forced the conspiracy to operate in a more covert manner. Plaintiffs have alleged that, before the consent decree, senior executives met in formal and regular meetings and operated pursuant to formal agreements. There has been no allegation that such meetings continued in the same manner and frequency.
Ultimately, we are unable to conclude that the alleged post-VASCO conspiracy meets the predominance requirement of Rule 23(b)(3) largely because plaintiffs have provided no real explanation of how this conspiracy operated, other than simply declaring that VASCO conspiracy "continued" by "express or tacit" agreement. Based on the existence of the consent decree, we will not assume that the post-VASCO conduct was the same conduct as before, even if there are some similarities between the two conspiracies. As the Magistrate Judge noted, "common issues of fact will not predominate if plaintiffs' intent is to link multiple conspiracies into one overall conspiracy." (RR at 32.) Defendants raise an additional valid point — even if there are some common elements between the VASCO and post-VASCO claims, the latter claims likely will implicate distinct additional elements such as the intervening technological and market developments. In her RR, written in 1995, the Magistrate Judge noted that electron power tube industry is on the decline due to the rise of the solid state industry and that increasing numbers of electron tubes have therefore become obsolete. ( Id. at 7.) We note that the post-VASCO period (from 1993 to 1999) is almost as long as the VASCO period (1986 to 1993).
The decision to exclude post-VASCO misconduct from the class is consistent with the decision to exclude pre-VASCO conduct — a decision apparently made by plaintiffs on their own. Although the complaint contains allegations of anticompetitive conduct beginning as early as 1981 ( i.e. well before VASCO was formed), the 1995 motion for certification sought a class period that began at the time VASCO was formed. In fact, plaintiffs chose to define the class period as beginning on the precise day — February 26, 1986 — that VASCO was formed. This is some indication that plaintiffs themselves viewed the class as coinciding with the operation and life of VASCO.
B. Continuing Effects From The VASCO Misconduct.
In their second argument, plaintiffs assert that the extension of the class period until 1999 can be justified on the separate theory that the VASCO-related conduct had continuing economic effects after VASCO went out of existence. As plaintiffs articulate the issue in their briefs, the termination of the wrongdoer's conduct does not end a wrongdoer's exposure where past misconduct produces continuing economic effects. Plaintiffs argue that, notwithstanding the consent decree, the supracompetitive pricing and lack of competition in the market continued unabated after defendants dissolved VASCO. In particular, plaintiffs state that the Justice Department's settlement, as embodied in the consent decree, did not require the defendants to roll back prices or to divest any of the operations they acquired during the life of VASCO.
Defendants make a number of arguments in response, many of them the same arguments made above. First, they argue that there is not a single allegation in the current complaint that the VASCO conduct had continuing economic effects. Second, they argue that plaintiffs have failed to offer any concrete evidence of such effects. Defendants point out that the evidence relating to the high prices of electron tubes is not clearly linked to the electron tubes associated with the VASCO agreements. Moreover, defendants maintain that the fact that the prices remained at the same levels as before is an unremarkable fact. Finally, defendants claim that the reason there was no price rollback or divestiture is because the Justice Department did not believe that such actions were needed to eliminate any negative effects remaining in the market.
As noted above, at this stage of the litigation, we cannot resolve disputed factual questions. Plaintiffs claim that there were continuing economic effects from the VASCO conduct, based largely on the fact that the prices remained high in the post-VASCO period and on the assertion that the VASCO conduct structurally altered the competitive structure of the industry such that meaningful competition has not been restored, even up until today. We also do not find any reason to dispute the general proposition advanced by plaintiffs that an antitrust violation in theory could have continuing effects. See, e.g., Law v. NCAA, 5 F. Supp.2d 921, 933 (D. Kan. 1998).
Despite these points, we do not believe that this rationale justifies extending the class period beyond March 16, 1993. First, by allowing the class period to extend through March 16, 1993, we are extending the class period almost a year beyond the dissolution of VASCO and even farther beyond the date when defendants claim VASCO effectively ended its operations.
Second, plaintiffs' continuing effects argument is undercut by its first argument regarding post-VASCO misconduct. Specifically, plaintiffs are claiming that defendants continued to engage in various illegal activities after VASCO. Plaintiffs further claim that these activities had an effect on the prices of electron tubes. Even though we have concluded above that plaintiffs have not met their burden of showing that this post-VASCO activity satisfies the predominance requirement of Rule 23(b)(3), we have assumed those allegations were true for the purposes of this ruling. Given that plaintiffs believe that there was continuing illegal activity in the post-VASCO period, it is impossible to know whether the allegedly high prices that existed after April 1, 1992 were simply the "continuing effects" of the VASCO conspiracy or whether they were the effect of the newer post-VASCO conspiracy. For the above two reasons, we reject plaintiffs' continuing effects argument.
II. Objections To The Magistrate Judge's Discovery Ruling.
Also before the court are plaintiffs objections' to the December 19, 2000 and the January 2, 2001 orders of Magistrate Judge Levin, as well as defendants' response to those objections. On December 19, 2000, Magistrate Judge Levin denied plaintiffs' motion to compel Varian to respond to plaintiffs' fourth set of interrogatories insofar as the motion sought answers for time periods after 1993. On January 2, 2001, the Magistrate Judge denied plaintiffs' motion for reconsideration of the December 19th Order. The Magistrate Judge concluded that the information after 1993 was not relevant and that it would be too burdensome to produce. He further noted that he would revisit the issue later if warranted. The Magistrate Judge's ruling was made while the motion to modify the class period was pending.
The parties agree that, in order to modify the Magistrate Judge's discovery ruling, we must conclude that it was "clearly erroneous or contrary to law." Fed.R.Civ.P. 72(a). Plaintiffs argue that post-1993 discovery is relevant based on the fact that the class period extends until 1999. In light of our ruling above, shortening the class period back to 1993, this objection is no longer applicable. In their objections, plaintiffs have advanced no other argument for why the information is relevant. For this reason, this court denies plaintiffs' request to vacate the Magistrate Judge's two discovery orders.
To the extent that plaintiffs believe that there are other arguments for extending discovery beyond 1993, they should raise those arguments with the Magistrate Judge in the first instance.
CONCLUSION
For the above reasons, the motion to modify the class period is granted and the class period set forth in this court's May 14, 1999 memorandum opinion is modified by changing the end date of the class period from May 13, 1999 to March 16, 1993. Plaintiffs' request to vacate the Magistrate Judge's December 19, 2000 and January 2, 2001 orders is denied.