Bifolck v. Philip Morris USA, Inc.

• Court: District Court, D. Connecticut
• Decided: May 12, 2020
• Docket: 3:06-cv-01768
• Status: Unknown

Opinion

UNITED STATES DISTRICT COURT DISTRICT OF CONNECTICUT

VINCENT J. BIFOLCK, AS EXECUTOR OF THE ESTATE OF JEANETTE D. No. 3:06-cv-1768 (SRU) BIFOLCK, AND INDIVIDUALLY, Plaintiff,

v.

PHILIP MORRIS USA INC., Defendant.

ORDER

In 2000, at the age of 42, Jeanette Bifolck died from lung cancer. She had been addicted to Marlboro cigarettes (“Marlboros”) from the early 1970s to 1997 and to Marlboro Lights (“Lights”) from 1997 until her death. In 2006, Jeanette’s husband, Vincent Bifolck (“Bifolck”), sued Philip Morris USA Inc. (“PM USA”) as executor of his late wife’s estate and individually and asserted that PM USA had negligently designed Marlboros and Lights in violation of the Connecticut Product Liability Act (“CPLA”). On the eve of trial, Bifolck asked me to give preclusive effect to six issues that were purportedly decided against PM USA in a federal RICO case against numerous cigarette manufacturers in the District of Columbia that concluded in 2006. See United States v. Philip Morris USA, Inc., et al., 449 F. Supp. 2d 1 (D.D.C. 2006) (“DOJ”). The DOJ case was a bench trial that lasted nine months and resulted in a nearly 1,000- page opinion. I declined to give those six issues preclusive effect because I felt that the issues in the two cases were not identical and that it would be impossible to say that any of the particular issues had been “necessary” to the DOJ court’s determination. So, this case went to trial in fall 2017, and the jury reached a defense verdict. On appeal, the Second Circuit disagreed with my decision not to give preclusive effect to one particular issue because the issue was identical to an issue resolved in DOJ, and the issue was necessary to the DOJ court’s judgment. But the Second Circuit did not vacate the judgment. It only remanded for me to consider an additional factor that I had not yet adequately considered: whether it would be unfair to apply nonmutual

offensive issue preclusion. See Bifolck v. Philip Morris USA Inc., 936 F.3d 74, 86 (2d Cir. 2019). The parties have briefed the issue, and I held a hearing on January 17, 2020. The parties then submitted post-hearing briefing. In my view, it would have been unfair to apply nonmutual offensive issue preclusion, and so I adhere to my prior ruling for that reason. I. Standard of Review In attempting to invoke nonmutual offensive issue preclusion (“NMOIP”), a plaintiff seeks to “preclude a defendant from relitigating an issue the defendant has previously litigated and lost to another plaintiff.” Faulkner v. Nat’l Geographic Enters., Inc., 409 F.3d 26, 37 (2d Cir. 2005) (citing Parklane Hosiery Co. v. Shore, 439 U.S. 322, 329 (1979)). Courts are “war[y]” of applying NMOIP because doing so is “detailed, difficult, and potentially dangerous.” Bifolck, 936 F.3d at 84 (citing Jack Faucett Assocs. v. Am. Tel. & Tel. Co., 744 F.2d 118, 124

(D.C. Cir. 1984) (internal quotation marks omitted)). Before a court may apply NMOIP, a court must make a two-part inquiry. In part one, a court must satisfy itself that the plaintiff has shown that the issue on which he or she seeks preclusive effect satisfies four elements. See id. at 79–80. Those four elements are: - Identicality: the issues in both proceedings must be identical. - Actually litigated and decided: the issue in the prior proceeding must have been actually litigated and actually decided. - Full and fair opportunity: there must have been a full and fair opportunity for litigation in the prior proceeding. - Necessity: the issue previously litigated must have been necessary to support a valid and final judgment on the merits. Id. The Second Circuit reviews de novo a district judge’s determination on the first part of the NMOIP test. See id. at 80 (citing Faulkner, 409 F.3d at 34). If part one of the NMOIP test is satisfied, the court then must engage in part two of the analysis: a “fairness” inquiry—that is, whether applying NMOIP is either fair or unfair in the particular circumstance. See id. The goal of the fairness inquiry is to “blunt the fear that [applying NMOIP] may be unfair to a defendant or fail to promote judicial economy.” Id. at 84 (citing Parklane Hosiery, 439 U.S. at 331). There is no mandatory list of factors that a district court must consider when answering this inquiry, but the Second Circuit has noted several that are important:

- Efficiency: whether applying NMOIP would increase the efficiency of the proceedings. - Inconsistent prior judgments: whether the judgment relied upon as a basis for estoppel is itself inconsistent with one or more previous judgments in favor of the defendant. - Procedural opportunities: whether the second action affords the defendant procedural opportunities unavailable in the first action that could readily cause a different result. - Incentive to raise: whether the defendant had little-to-no incentive to raise the issue in the earlier action. - Type of trial: whether the underlying case was tried before a jury or a judge sitting as a trier of fact. - Scope and complexity: whether the cases differ in scope or complexity and involve different causes of action.

See id. “The principal virtue” of NMOIP is that “it promotes judicial economy.” S.E.C. v. Monarch Funding Corp., 192 F.3d 295, 303 (2d Cir. 1999). “When the efficiency rationale for collateral estoppel fails . . . courts have understandably declined to apply the doctrine.” Id. at 304. Trial courts have “broad discretion to determine when” NMOIP “should be applied.” Parklane Hosiery, 439 U.S. at 331. The Second Circuit reviews the district judge’s determination of the fairness inquiry for abuse of discretion. See Bifolck, 936 F.3d at 80 (citing Remington Rand Corp. v. Amsterdam-Rotterdam Bank, N.V., 68 F.3d 1478, 1486 (2d Cir. 1995)). II. Background

A. Underlying Action – the DOJ litigation In August 2006—following seven years of litigation and a nine-month bench trial— Judge Gladys Kessler of the U.S. District Court for the District of Columbia published a nearly 1,000-page decision in DOJ. The DOJ court found that PM USA (and numerous other cigarette manufacturers) had violated the federal RICO laws by “engaging in a lengthy, unlawful conspiracy to deceive the American public about . . . the addictiveness of nicotine . . . and their manipulation of the design and composition of cigarettes in order to sustain nicotine addiction.” DOJ, 449 F. Supp. 2d at 26–27. For over 50 years, Judge Kessler found, the tobacco companies “denied to the public what they recognized internally beginning as early as the 1950s: people smoke primarily because of the pharmacological effects of the drug nicotine.” Id. at 858. Still, the tobacco companies “designed their cigarettes with a central overriding objective—to ensure that smokers obtain enough nicotine to create and sustain addiction.” Id. at 859.

The DOJ court, limited to forward-looking injunctive remedies, ordered both general injunctive relief and that the defendants make corrective statements to the public. See id. at 938– 41. In particular, the DOJ court ordered that the defendants were “permanently enjoined from making, or causing to be made in any way, any material false, misleading, or deceptive statement or representation, or engaging in any public relations or marketing endeavor that is disseminated to the United States public and that misrepresents or suppresses information concerning cigarettes.” Id. at 938 ¶ 3. And the DOJ court ordered PM USA (among other defendants) to file a corrective statement concerning “Defendants’ manipulation of cigarette design and composition to ensure optimum nicotine delivery.” Id. at 938–39. It took nearly ten years to finalize the precise language in the corrective statements. Judge Kessler began drafting the corrective statements in 2012. See United States v. Philip Morris USA, Inc., et al., 907 F. Supp. 2d 1, 8–9 (D.D.C. 2012). At that point, the

corrective statements regarding the tobacco companies’ manipulation of cigarette design read, in relevant part: • “Defendant tobacco companies intentionally designed cigarettes to make them more addictive,” and • “Cigarette companies control the impact and delivery of nicotine in many ways, including designing filters and selecting cigarette paper to maximize the ingestion of nicotine, adding ammonia to make the cigarette taste less harsh, and controlling the physical and chemical make-up of the tobacco blend.”

Id. at 9. The D.C. Circuit upheld that language in May 2015 (based on waiver). See United States v. Philip Morris USA Inc., et al., 801 F.3d 250, 258–61 (D.C. Cir. 2015). In late 2017— two weeks after the trial in this matter concluded—Judge Kessler finalized the corrective statements. See United States v. Philip Morris USA Inc., 257 F. Supp. 3d 1, 4–5 (D.D.C. 2017). The only difference between the language quoted above from 2012 and the final version from 2017 is that the first bullet point ultimately read: • “Altria, R.J. Reynolds Tobacco, Lorillard, and Philip Morris USA intentionally designed cigarettes to make them more addictive.”

Id. at 5.

B. Proceedings in This Case In November 2006, Bifolck filed this case and alleged, as relevant here, that PM USA had negligently designed Marlboros and Lights within the meaning of the CPLA and that that negligent design caused his wife’s death. See Compl., Doc. No. 1, at ¶¶ 40–43; Second Am. Compl., Doc. No. 218, ¶¶ 18–36 (filed Feb. 2, 2017) (operative complaint). The case eventually was tried before a jury from October 23 to November 9, 2017. The jury rendered a defense verdict, finding that Bifolck had failed to prove by a preponderance of the evidence that PM USA had negligently designed Marlboros and Lights within the meaning of the CPLA. See Verdict Form, Doc. No. 456.

Between 2006 and 2017, the parties engaged in extensive discovery, and I even certified a question of law to the Connecticut Supreme Court in 2014 that halted this case for over two years. Despite the pendency of the case for over a decade, it was not until three weeks before trial, on September 27, 2017, that Bifolck first made a motion to give preclusive effect to six issues purportedly resolved against PM USA in DOJ: (1) Philip Morris knew and falsely denied to the public, both before and during the time that Mrs. Bifolck smoked, that smoking cigarettes causes lung cancer.

(2) Philip Morris knew and falsely denied to the public, both before and during the time that Mrs. Bifolck smoked, that the nicotine in cigarette smoke is addictive.

(3) Both before and during the time that Mrs. Bifolck smoked, Philip Morris manipulated cigarette design and composition to assure nicotine delivery levels which create and sustain addiction.

(4) Both before and during the time that Mrs. Bifolck smoked Marlboro Lights, Philip Morris falsely asserted that Marlboro Lights presented fewer health risks than full-flavor cigarettes when it knew that was not the case.

(5) Both before and at the time that Mrs. Bifolck began to smoke, Philip Morris deliberately marketed to underage smokers, while falsely denying such marketing.

(6) Both before and during the time that Mrs. Bifolck smoked, Philip Morris suppressed research results which, if published, would have shown that the tars in cigarette smoke are carcinogenic and that the nicotine in cigarette smoke is addictive.

See Pl.’s Mem. in Supp. Mot. to Give Preclusive Effect (the “Preclusion Motion”), Doc. No. 363, at 1–2. Bifolck’s Preclusion Motion was the first motion regarding issue preclusion presented to me in this case. I requested at the post-remand hearing in this matter that the parties submit further briefing regarding the history of Bifolck’s request: that is, even though Bifolck did not make his Preclusion Motion until September 27, 2017, had Bifolck flagged the issue before then? The parties’ submissions and my review of the history of this case reveal that Bifolck did not

meaningfully flag the issue before September 27. Although Bifolck did not unearth it in his post-hearing briefing, the first mention of the DOJ case of which I am aware occurred at a status conference on January 12, 2017. There, Bifolck mentioned the possibility that he would “seek to offer the judicial findings that were made against Philip Morris in the government’s RICO action.” See Status Conf. Tr., Doc. No. 216, at 15:6–7. Of course, that formulation does not mention issue preclusion or suggest that Bifolck would seek to give preclusive effect to any issues from the DOJ case. Bifolck’s counsel argues that, throughout the spring and summer of 2017, the preclusion issue was on his mind and that he brought it to my attention. Bifolck’s argument centers on the deposition and potential expert testimony of a PM USA employee, Richard Jupe.1 Bifolck points

out that, beginning in February 2017, he sought to conduct a Rule 30(b)(6) deposition—from Jupe or another PM USA employee—to obtain testimony on PM USA’s cigarette design practices. See Pl.’s Post-Hr’g Br., Doc. No. 515, at 3; Conf. Tr., Ex. B to Decl. of D. Golub, Doc. No. 516-1, at 73–76. Although Jupe ultimately was not deposed as a Rule 30(b)(6) witness, the parties agreed that PM USA would not object to Jupe’s testimony on the grounds that “Jupe is not authorized to speak on PM USA’s behalf.” Email Stipulation, Ex. F to Decl. of D. Golub, Doc. No. 516-1, at 113.

1 Mr. Jupe is an expert in cigarette design and manufacturing who worked at Philip Morris in the 1990s and 2000s. He was called as an expert on cigarette design, testing, manufacturing, and processing. See Trial Tr., Doc. No. 468, at 1993:6–17; 2029:9–14. Jupe’s deposition was repeatedly delayed and took place in two parts—on August 15, 2017 and September 6, 2017. See Pl.’s Post-Hr’g Br., Doc. No. 515, at 6–7. During the delay (in June), the parties wrote to my law clerk and explained that Bifolck was “willing to wait until [mid-August] for the deposition, but will be unable to file motions in limine addressed to the

scope of Jupe’s testimony or the exhibits defendant seeks to offer until after the deposition.” See Email, Ex. R to Decl. of D. Golub, Doc. No. 516-1, at 179. I granted the parties’ request. See Conf. Mem. and Order, Ex. S to Decl. of D. Golub, Doc. No. 516-1, at 181. On July 10, 2017, PM USA filed a motion in limine to preclude any mention of the DOJ case. See PM USA’s Mot. in Limine, Doc. No. 301. On July 28, 2017, Bifolck filed a partial reply to PM USA’s motion and asked me to defer ruling on that motion in limine until two weeks after the completion of Jupe’s deposition. See Pl.’s Omnibus Opp’n, Doc. No. 314, at 1–2. Nowhere in that opposition did Bifolck mention the potential preclusive effect of issues purportedly determined in the DOJ case: preclusion is not mentioned even once. On August 24, 2017, Bifolck submitted a supplemental opposition memorandum. See Pl.’s Supp. Opp’n, Doc.

No. 344. In that memorandum, Bifolck argued that he should be able to offer numerous DOJ factual findings (1) to impeach Jupe’s testimony, see id. at 4–8; 12–13, (2) as a basis for his own expert’s testimony, see id. at 9–11, and (3) as admissible hearsay, see id. at 11–12. Again, Bifolck never mentioned giving preclusive effect to issues purportedly decided in the DOJ case. On September 14, 2017, I held a hearing to address numerous pending motions in limine. There, following a lengthy discussion, I granted PM USA’s motion to exclude reference to the DOJ case. See Min. Entry, Doc. No. 354; Hr’g Tr., Doc. No. 359, at 45–58. At the end of my ruling, the following colloquy ensued: GOLUB: Well, there’s also an issue of issue preclusion, Judge. COURT: There may be, but I haven’t gotten that motion yet. GOLUB: We’re offering [certain DOJ findings] as substantive evidence. We’re claiming that there’s issue preclusion on the findings that [Judge Kessler] made that are relevant to this case, and that there’s a full and fair opportunity to –

COURT: My recollection is that your footnote2 said we might make a motion with respect to issue preclusion. I don’t remember seeing a motion on that. I think you reserved your right, but I don’t think I’ve seen it yet. And so when you offer it as evidence, it’s hearsay, and for the reasons I’ve already expressed, it doesn’t come in. If you want to say this issue is no longer in the case, then that’s a different motion. It seems to me I need to get an opposition to that because I haven’t.

GOLUB: Well, let me go back and look at what our footnote said. COURT: All right. That’s just my recollection. GOLUB: All right. COURT: All right, so I’m granting 301. Hr’g Tr., Doc. No. 359, at 57:14–58:12. On September 18, Bifolck filed a motion to preclude Jupe from testifying, in part, about PM USA’s past actions and knowledge. See Mot. in Limine, Doc. No. 355, at 8, 16–20. However, again, nowhere in that motion did Bifolck mention relying on the DOJ case for purposes of issue preclusion. On September 27, 2017, Bifolck filed the Preclusion Motion, which asked me to give preclusive effect to six issues resolved against PM USA in DOJ. See Preclusion Mot., Doc. No. 363, at 1–2. On October 9, PM USA filed an opposition. See PM USA’s Opp’n to Preclusion Mot., Doc. No. 380. At a hearing on October 11, 2017, I denied Bifolck’s Preclusion Motion. I explained: I am not going to grant preclusive effect to the findings of Judge Kessler for numerous reasons. One, the issues are not identical. Probably most importantly,

2 I do not remember what footnote I was referencing. Neither has Bifolck identified it in his post-hearing briefing. no finding, and certainly not all the findings that the plaintiff seeks to give preclusive effect to, are necessary to the Court’s ruling in the DOJ case, and if I had to reach it, I probably would find that there are equitable considerations here that would make it inappropriate to use them in any event.

I think the first two are the more important ones. We have a massive RICO case versus a Connecticut product liability case. We have all the cigarette manufacturers being sued at the same time, along with some others, versus one cigarette manufacturer, and because there were so many hundreds of findings, the idea that any one, two, 20, 100 of them were necessary to the Court’s ruling becomes literally impossible to figure out. And so this is not taking a simple case and applying it to another simple case. It’s taking one of the most complex cases in decades and trying to apply it to a relatively complex case raising different causes of action, and I just don’t think it’s feasible or appropriate either legally or kind of in terms of the amount of effort it would take, and in the end we’d be left with something that I think the jury would find very confusing.

Hr’g Tr., Doc. No. 388, at 29:12–30:11.

Trial began on October 23, 2017 and concluded on November 9, 2017, when the jury returned a defense verdict. See Verdict, Doc. No. 456. C. Appeal and Second Circuit Mandate Bifolck appealed my decision not to give preclusive effect to one issue: the fact that “Philip Morris manipulated cigarette design and composition to assure nicotine delivery levels which create and sustain addiction.” See Notice of Appeal, Doc. No. 478; Bifolck, 936 F.3d at 79–81. (I will refer to this as the “Manipulation Issue.”) After reviewing the transcript of my oral ruling, the Second Circuit concluded that I had denied Bifolck’s Preclusion Motion based on (1) the lack of identicality between the issues from the DOJ case and this case, and (2) the fact that none of the issues identified by Bifolck had been necessary to support the DOJ judgment. Id. at 78–79. The Second Circuit disagreed with both of those conclusions. Regarding identicality, the Second Circuit found that the Manipulation Issue was identical to a “certain and material point” resolved in DOJ because “[s]everal factual findings from DOJ touch on this issue.” See Bifolck, 936 F.3d at 81. And, even though the Manipulation Issue was quite general (i.e., not limited temporally or by cigarette brand), the Second Circuit still held that the Manipulation Issue would have “supported Bifolck’s theory at trial.” Id. at 82. The Second Circuit also explained that—although relevant to the fairness inquiry—differences in scope and type between this case and the DOJ case did not affect the identicality inquiry. See

id. Regarding necessity, the Second Circuit agreed that “[t]he conclusion in DOJ that Philip Morris violated RICO did not turn on any of the factual findings to which Bifolck sought to give preclusive effect” and that it would be “virtually impossible to determine which, if any, factual findings were necessary to the finding of liability.” Id. at 82–83. However, the Second Circuit held that the Manipulation Issue was necessary to the judgment in DOJ because the judgment included the remedy imposed. Id. at 83. The remedy in DOJ included the corrective statements, which were, in part, “directly supported by, and dependent upon” the Manipulation Issue. See id. (citing the portion of the corrective statement requiring PM USA to say that it “intentionally design[ed] cigarettes to make them more addictive”). Id. Thus, the Second Circuit held that

Bifolck had satisfied the necessity requirement. Despite holding that I incorrectly applied the first part of the NMOIP test, the Second Circuit remanded (without vacating the judgment), because it held that I had not yet reached the second part of the NMOIP test: the fairness analysis. See id. at 77–78. Recall that in my oral ruling, I had said: “[I]f I had to reach it, I probably would find that there are equitable considerations here that would make it inappropriate to” give preclusive effect to the issues Bifolck identified based on the DOJ case. Hr’g Tr., Doc. No. 388, at 29:18–20. The Second Circuit explained that it was not clear what factors I relied on to make that determination “or if [I] would maintain this view on remand during a full analysis of the fairness considerations.” Bifolck, 936 F.3d at 85. Thus, the Second Circuit remanded for me “to review, in [my] sound judgment, whether application of nonmutual offensive collateral estoppel would be unfair.” Id. The Second Circuit noted that “[o]ne fairness consideration weighs particularly heav[ily] here: whether applying collateral estoppel would have increased the efficiency of the

proceedings.” Id. at 84. Indeed, the Second Circuit pointed out, the issue on which Bifolck sought preclusion “is no silver bullet” because “Bifolck’s formulation of the issue makes no mention of Marlboros [or Lights] or the time period during which” Mrs. Bifolck smoked (the early 1970s to 2000). Id. Thus, the Second Circuit reasoned: Nonmutual offensive collateral estoppel could have failed to increase efficiency in several ways: by confusing the jury on the differences between what was decided in DOJ and what Bifolck ultimately needed to prove, by causing the production of duplicative evidence, or because the general nature of the preclusion may be too inconsequential to Bifolck’s case. On the other hand, a baseline factual finding about nicotine manipulation could have increased efficiency by shortening and focusing the expert testimony and streamlining jury deliberation.

See id. at 84–85. The Second Circuit also noted that I could consider whether the timing of Bifolck’s motion—just three weeks before trial—counseled against applying NMOIP. See id. at 84 n.9. Finally, the Second Circuit held that my error was not harmless. If the Manipulation Issue had had preclusive effect in this case, Bifolck could have more easily established that PM USA “knew that it could manipulate nicotine levels,” which would “support[] the conclusion that it made Marlboros ‘unnecessarily addictive’ by virtue of a design choice not to lower nicotine levels and, therefore, made them ‘unnecessarily dangerous’ under the CPLA.” Id. at 85. Instead, Bifolck sought to establish that point through Jupe’s cross-examination. See id. Rather than confirming the Manipulation Issue, the Second Circuit noted, Jupe “made claims inconsistent with the findings in DOJ, painting a cloudy picture of what the company could or could not do and did or did not know.” Id. III. Discussion A. The Proper Vantage Point As I expressed to the parties at oral argument,3 I find one aspect of the Second Circuit’s

decision not entirely clear. Specifically, the Second Circuit’s mandate does not indicate from what vantage point I should undertake my fairness inquiry. Recall that the Second Circuit instructed me to determine whether it “would be” unfair to give the Manipulation Issue preclusive effect. See Bifolck, 936 F.3d at 78, 85–86. That is a present tense formulation that suggests I should conduct the inquiry today, in 2020. My analysis today would be different than my analysis would have been on October 11, 2017. Specifically, today, I could consider events that have occurred since October 11, 2017, including the multi-week jury trial that occurred in this case and which PM USA won. In my view, though, the Second Circuit’s mandate—because it upheld the judgment pending my

review—suggests that I should conduct the inquiry as if I went back in time to the hearing on October 11, 2017, properly concluded that the first part of the NMOIP test had been satisfied, and thus reached the fairness inquiry. Luckily, the vantage point from which I conduct the fairness inquiry is not dispositive, because I hold that giving preclusive effect to the Manipulation Issue would have been unfair on October 11, 2017 and would be unfair today. First, I explain why I ruled the way I did on October 11, 2017. Much of that discussion regards the timing of the Preclusion Motion and the manner in which it was presented to me. Second, I explain why on October 11, 2017—taking

3 See, e.g., Hr’g Tr., Doc. No. 512, at 3:20–4:12; id. at 27:19–31:20. into account the parties’ present arguments (insofar as they do not rely on post-October 11, 2017 facts) and limiting my analysis to the Manipulation Issue—it would have been unfair to give preclusive effect to the Manipulation Issue. Finally, I explain why my decision would be the same if made knowing what I know today.

B. October 11, 2017: How It Happened

In this section, I assume the vantage point of October 11, 2017 and explain further why on that day I thought that, “if I had to reach [the fairness inquiry], I probably would find that there are equitable considerations here that would make it inappropriate to” give preclusive effect to the six issues on which Bifolck sought preclusion. See Hr’g Tr., Doc. No. 388, at 29:18–20. 1. The Timing and Manner of the Preclusion Motion Both PM USA (in 2017) and the Second Circuit (in 2019) have noted the possibility that Bifolck could have made his Preclusion Motion—which was filed just three weeks before trial— much earlier. See PM USA’s Opp’n to Preclusion Mot., Doc. No. 380, at 4; Bifolck, 936 F.3d at 84 n.9 (explaining that as part of my fairness analysis, I was “well suited to determine whether the timing of [Bifolck’s] motion counsels against applying” NMOIP). At the post-remand oral argument in this case in January 2020, I asked the parties for post-hearing briefing regarding whether—and, if so, when—Bifolck had raised the prospect of filing the Preclusion Motion before September 27, 2017. As the procedural history detailed above shows, Bifolck had never indicated his intent to file a Preclusion Motion before filing it on September 27, 2017. To be sure, Bifolck had mentioned the DOJ case as early as January 2017. See Status Conf. Tr., Doc. No. 216, at 15:6–7 (Bifolck’s lawyer mentioning that he would “seek to offer the judicial findings that were made against Philip Morris in the government’s RICO action”). Based on Bifolck’s representations, PM USA made a motion in limine to “preclude testimony, argument, and evidence” regarding the DOJ case. See PM USA’s Mot. in Limine, Doc. No. 301, at 1. When Bifolck responded in full on August 24, 2017, he explained that he “intended to use,” by my count, over 500 specific

findings of fact from the DOJ case. See Specific Findings of Fact, Ex. A to Pl.’s Supp. Opp’n, Doc. No. 344-1, at 2. Bifolck planned to use those findings of fact both to cross-examine Jupe (if Jupe testified inconsistent with them) and as affirmative evidence, either through his own expert witness or otherwise. See Pl.’s Supp. Opp’n, Doc. No. 344, at 4–13. Nowhere in his opposition did Bifolck mention the quite different proposition that certain issues should be taken out of the case by giving the DOJ findings preclusive effect. Bifolck argues that his motions in limine regarding Jupe were equivalent to the Preclusion Motion. See Hr’g Tr., Doc. No. 512, at 10:24–11:5. That is an unconvincing argument. There is certainly some connection between Jupe’s testimony and Bifolck’s Preclusion Motion. The parties understood that Jupe would testify, in part, about PM USA’s

company history and policies. And Bifolck wanted to convince the jury that PM USA knew— from the 1970s to 2000—that its design of Marlboros and Lights was unreasonably dangerous. But the legal arguments regarding Jupe’s testimony and the Preclusion Motion are quite dissimilar. In the first, Bifolck sought to cabin Jupe’s testimony based on certain findings from the DOJ case and to introduce other findings from the DOJ case as affirmative evidence; in the second, Bifolck asked me to instruct the jury that six major issues regarding PM USA’s company history had been definitively determined as a matter of law.4 There was no reason that Bifolck

4 Indeed, in his Preclusion Motion, Bifolck mentions Jupe not once. See Preclusion Mot., Doc. No. 363. And in his motion in limine to limit Jupe’s testimony, Bifolck references the DOJ case, but never mentions issue preclusion. See Mot. in Limine, Doc. No. 355. needed to wait for Jupe’s deposition to make his Preclusion Motion.5 Indeed, if Bifolck had made his Preclusion Motion earlier and prevailed, it would have eliminated the risk that Jupe would testify in a manner that Bifolck disliked. The timing of Bifolck’s motion was a strategic decision.

The timeline of this litigation and the DOJ case reaffirm that Bifolck could have brought his Preclusion Motion years earlier. As PM USA points out, Bifolck had been on notice since 2006 that PM USA would be contesting the Manipulation Issue. See PM USA’s Post-Hr’g Br., Doc. No. 519, at 3–6 (citing paragraphs in PM USA’s answers to Bifolck’s original complaint (2007), first amended complaint (2014), and second amended complaint (2017) in which PM USA denied that it negligently manufactured and designed Marlboros and Lights at the time periods relevant to this case). Bifolck could have sought preclusive effect of issues purportedly resolved in the DOJ case possibly as early as 2009 and certainly by May 2015. In 2009, the D.C. Circuit upheld the final judgment in the DOJ case and the concept of corrective statements as remedies. See United

States v. Philip Morris USA Inc., 566 F.3d 1095, 1150 (D.C. Cir. 2009). In 2015, the D.C. Circuit affirmed nearly all the precise wording of the corrective statements relevant to the Manipulation Issue. See United States v. Philip Morris USA Inc., 801 F.3d 250, 259 (D.C. Cir.

5 Bifolck argues that the Preclusion Motion was unnecessary until Bifolck knew how Jupe would testify because— if Jupe testified consistent with the Manipulation Issue (and other issues)—the Preclusion Motion would have been unnecessary. See, e.g., Pl.’s Mem., Doc. No. 495, at 44 n.38.; Hr’g Tr., Doc. No. 512, at 23:1–3 (“[T]he reason that the collateral estoppel motion couldn’t be made was we didn’t know what Jupe was going to say.”); Pl.’s Post-Hr’g Br., Doc. No. 515, at 5 & nn.14–15; Exs. I and J to Decl. of D. Golub, Doc. No. 516-1, at 141–48 (Jupe’s testimony from other Engle-progeny trials). But Bifolck articulates a strategic decision—the content of Jupe’s testimony was simply not a necessary condition to Bifolck’s bringing the Preclusion Motion. Additionally, as PM USA points out, Bifolck was well aware (from PM USA’s answers to Bifolck’s complaint and subsequent amended complaints) that PM USA was contesting, among other issues, the Manipulation Issue. See PM USA’s Post-Hr’g Br., Doc. No. 519, at 3–6. 2015).6 Indeed, by 2013, numerous plaintiffs had attempted to give preclusive effect to certain issues purportedly decided in DOJ. See PM USA’s Opp’n to Preclusion Mot., Doc. No. 380, at 1 n.1 (citing eleven such cases from 2006 through 2013). Bifolck’s waiting until three weeks before trial to make the Preclusion Motion was unfair

in two ways. First, it was significantly unfair to PM USA. It is true that PM USA filed a robust response to Bifolck’s Preclusion Motion, no doubt in part because it had defended such motions in prior litigation. Still, had I granted the Preclusion Motion, its timing would have meant that PM USA had engaged in needless discovery, motion practice, expert witness preparation, and other trial preparation. See Def.’s Mem., Doc. No. 497, at 14. It simply was not fair to PM USA to file such a major motion—ripe for years—just three weeks before trial in a case that had been pending since 2006. Second, it was unfair to me. To the extent possible,7 major motions that will define the contours of a case should be made early so that the important issues can receive the briefing and attention they merit. Instead of receiving the Preclusion Motion far in advance, though, I

received it three weeks before trial and had to decide it at the same time that the parties had filed numerous other pre-trial motions. I gave it all the attention that I could at the time. 2. My Main Consideration: Fear of Juror Confusion and Unreasonable Prejudice

6 Although the language of the corrective statements figured prominently in the Second Circuit’s decision on appeal, see Bifolck, 936 F.3d at 78, Bifolck did not bring to my attention the “interim” language. So far as I can tell, Bifolck mentioned corrective statements’ language only once and said that “[t]he defendants have continued to contest the precise wording of the corrective statements for the last 11 years.” See Preclusion Mot., Doc. No. 363, at 10 n.5. Bifolck’s only arguments to me regarding the corrective statements regarded the general fact that the defendants in the DOJ case would be required to issue a corrective statement regarding their “manipulation of cigarette design and composition to ensure optimum nicotine delivery.” See id. at 8–10; DOJ, 449 F. Supp. at 939. 7 It would be one thing if Bifolck represented that he did not have the idea for the Preclusion Motion until September 2017. But that is not what he says. Instead, as described above, Bifolck argues that he had long intended to make the motion if Jupe testified a certain way at his deposition. Although I erroneously relied upon the necessity and identicality prongs of the first part of the NMOIP test in denying the Preclusion Motion, the Second Circuit noted that part of my identicality analysis was relevant to the fairness inquiry. See Bifolck, 936 F.3d at 82. In particular, I had noted differences in scope, trial type, and cause of action between the DOJ case

and this case. I said: We have a massive RICO case versus a Connecticut product liability case. We have all the cigarette manufacturers being sued at the same time, along with some others, versus one cigarette manufacturer . . . . And so this is not taking a simple case and applying it to another simple case. It’s taking one of the most complex cases in decades and trying to apply it to a relatively complex case raising different causes of action, and I just don’t think it’s feasible or appropriate either legally or kind of in terms of the amount of effort it would take, and in the end we’d be left with something that I think the jury would find very confusing.

Hr’g Tr., Doc. No. 388, at 29:22–30:11. On October 11, 2017, I was thinking about those differences. And, especially because of those differences, I was concerned that granting the Preclusion Motion would cause the jury to reflexively find for Bifolck. As explained more fully below, courts routinely rely on those considerations—which courts often discuss alongside concerns of “efficiency”—in denying a plaintiff’s request to apply NMOIP. C. October 11, 2017: A Full Analysis Had I reached the fairness inquiry on October 11, 2017—having been fully briefed and limiting my analysis to the Manipulation Issue—I would have held that giving the Manipulation Issue preclusive effect would be unfair for several reasons. First, doing so would not have provided significant gains in efficiency and would have led to a high risk of juror confusion. Second, the numerous verdicts in favor of PM USA (and the tobacco companies generally) in cases relatively similar to DOJ would have given me further pause. 1. Most Important: Efficiency, Risk of Juror Confusion, and Differences Between the Two Cases “The principal virtue” of NMOIP is that “it promotes judicial economy.” Monarch Funding, 192 F.3d at 303. “When the efficiency rationale for collateral estoppel fails . . . courts have understandably declined to apply the doctrine.” Id. at 304. Indeed, the Second Circuit in this case noted that whether giving the Manipulation Issue preclusive effect would increase the

efficiency of proceedings was a consideration that “weighs particularly heav[ily] here.” Bifolck, 936 F.3d at 84. As the Second Circuit further noted, granting the Manipulation Issue preclusive effect could have failed to increase “efficiency” by, for instance, “confusing the jury on the differences between what was decided in DOJ and what Bifolck ultimately needed to prove, by causing the production of duplicative evidence, or because the general nature of the preclusion may be too inconsequential to Bifolck’s case.” Id. i. Bifolck’s arguments

Bifolck first points out the importance of the Manipulation Issue and how not giving it preclusive effect would disadvantage him. Bifolck argues that giving the Manipulation Issue preclusive effect would establish that (1) PM USA had the ability to manipulate the nicotine delivery levels of cigarettes; (2) it actually did so; and (3) its motivation was to create and sustain nicotine addiction. See Pl.’s Mem., Doc. No. 495, at 37. Bifolck admits that he would still need to show that PM USA engaged in such manipulation between the 1970s and 2000 with respect to Marlboros and Lights. See id. Still, Bifolck notes that giving the Manipulation Issue preclusive effect would be “extremely important” to his case. Id. For instance, Bifolck argues that doing so would preclude Jupe from testifying falsely regarding the Manipulation Issue. Id. at 38. In addition, Bifolck argues that deciding not to give the Manipulation Issue preclusive effect would

disadvantage him because he has neither the resources nor the physical ability—because some of the unavailability of certain witnesses—to recreate the evidence that supported the findings undergirding the Manipulation Issue in DOJ. See id. at 39–40 & n.29. Beyond its import to his case, Bifolck argues that giving the Manipulation Finding preclusive effect would streamline the presentation of evidence. That is because much of Jupe’s

testimony would likely be unnecessary and also because Bifolck’s own experts would not need to testify about topics pertaining to the Manipulation Issue. See Pl.’s Reply, Doc. No. 502, at 4– 5. Along the same lines, Bifolck argues that giving the Manipulation Issue preclusive effect would simplify the case for the jury. Instead of a trial focused on the Manipulation Issue, Bifolck argues that the case could be “laser-focused on Philip Morris’s design of Marlboro and Marlboro Lights, and why it failed to make those brands safer.” Id. at 6. Bifolck points to the Engle progeny cases to suggest that the jury would not be confused about giving the Manipulation Issue preclusive effect. See Pl.’s Mem., Doc. No. 495, at 22–24; see generally J.B. Harris, Engle v. Liggett: Has Big Tobacco Finally Met Its Match?, 86-NOV Fla. B.J. 16 (Nov. 2012). Engle began in 1994, when a class of smokers sued cigarette

manufacturers in Florida state court. The class was certified, and the trial was broken into phases, the first of which was a year-long trial to determine factual issues common to the class. The jury made numerous findings, including that “the defendants placed on the market cigarettes that were defective and unreasonably dangerous.” See Harris, Engle v. Liggett, at 17. On appeal, the Florida Supreme Court decertified the class but allowed individual former class members to bring cases against the defendants in which the phase one jury’s factual determinations would be given res judicata effect. See Engle v. Liggett Group, Inc., 945 So. 2d 1246, 1277 (Fla. 2006). Those subsequent individual trials are the Engle progeny. Bifolck explains that the experience in Engle progeny trials shows definitively that juries are not confused when they are instructed to accept general facts regarding cigarette company liability—even “with respect to a very broad range of unlawful conduct”—and apply them to specific, individual cases. See Pl.’s Mem., Doc. No. 495, at 23. More particularly, Bifolck

argues that Engle progeny juries do not always find for the plaintiff. See id. (only 64.1 percent plaintiffs’ verdicts); see also Decl. of D. Golub, Doc. No. 496, at ¶ 8.8 To Bifolck, that outcome demonstrates that a proper instruction on the Manipulation Issue’s preclusive effect would not confuse the jury or unfairly prejudice the defendant. Bifolck further brushes aside the entirely different postures of the DOJ case and this case. Indeed, Bifolck explains that the Manipulation Issue “is a completely straightforward factual determination” that “has nothing to do with the RICO context in which the finding was made, and its relevance in the product liability context of this case is both direct and self-evident.” Pl.’s Mem., Doc. No. 495, at 35–36. ii. PM USA’s arguments

PM USA argues that giving the Manipulation Issue preclusive effect would be inefficient because it would not reduce the trial’s duration and because it would not meaningfully streamline the presentation of evidence. PM USA points out that—even if I gave the Manipulation Issue preclusive effect—Bifolck would still have to prove the same issue, just with more specificity. In particular, Bifolck would still have to elicit testimony concerning the design of Marlboros and Lights between the 1970s and 2000, including how PM USA allegedly manipulated nicotine in

8 Bifolck’s statistic apparently takes into account results in Engle progeny cases through fall 2019. Of course, whatever statistic Bifolck submitted to me in October 2017 could not have incorporated results from cases that would conclude in the future. Still, I consider Bifolck’s statistic at face value because: (1) there is no reason to believe that the general point (or even the specific percentages) has changed since October 11, 2017, and (2) both sides rely on the experience in the Engle progeny trials, which is relevant and important. those products; show that reasonable alternative designs were available that would have avoided unreasonable danger; and demonstrate how PM USA was negligent in not implementing an acceptable alternative design. See Def.’s Mem., Doc. No. 497, at 12; Def.’s Reply, Doc. No. 501, at 11. PM USA further argues that—even were the Manipulation Issue given preclusive

effect—it would have to present evidence overlapping with the Manipulation Issue to support defenses such as comparative responsibility and to limit its exposure to punitive damages.9 See Def.’s Mem., Doc. No. 497, at 13. PM USA also views the lesson from the Engle progeny cases quite differently from Bifolck and argues that the experience of Engle progeny trials emphasizes the inefficiency that would accompany giving the Manipulation Issue preclusive effect. PM USA points out that although Engle progeny juries are told not to consider whether the cigarettes at issue are defective, Engle progeny cases still on average last between 13 and 14 trial days from voir dire through verdict. See Decl. of G. Michael, Ex. 1 to Def.’s Reply, Doc. No. 501-1, at ¶ 3.10 Further, despite the various preclusive effects, three experts whom Bifolck planned on calling to

testify (and who did in fact testify)—Drs. Kenneth Cummings, William Farone, and Neil Grunberg, see Witness List, Doc. No. 472—have testified in many Engle progeny cases about “the history of cigarette design and other issues.” See id. at ¶ 4.

9 PM USA explains that it is particularly worried that juror confusion regarding how to apply the Manipulation Issue instruction might lead to an unconstitutional punitive damages penalty. In Philip Morris USA v. Williams, the Supreme Court held that a jury award of punitive damages that is based, in part, on the jury’s desire to punish the defendant for injury inflicted on a non-party is an unconstitutional taking of that defendant’s property. 549 U.S. 346, 349, 353 (2007); see also In re Light Cigarettes, 691 F. Supp. 2d 239, 251 (D. Me. 2010) (finding that if issue preclusion from the DOJ case were imposed, “much of the Defendants’ underlying liability would be based in part on actions that inflicted injuries upon nonparties. Despite instructions to compartmentalize certain factual findings, the jury could be confused about what facts may or may not be considered when determining punitive damages”); Grisham v. Philip Morris, Inc., 670 F. Supp. 2d 1014, 1037 (C.D. Cal. 2009) (same). PM USA argues that that concern would be present here, too. 10 This timeline takes into account 315 Engle progeny trials from 2009 through October 2019. I have already explained why—even though the statistic includes events taking place after October 11, 2017—I still consider the more general point. See supra n.8. PM USA also points out that Engle progeny juries are given a much simpler task than would a jury in this case, were I to give the Manipulation Issue preclusive effect. In particular, in Engle progeny cases, the jury is told not to consider general causation and only to focus on individual causation in the current case. See Def.’s Reply, Doc. No. 501, at 12–13. Here, a jury

would be asked to apply a finding from a completely different federal RICO case to a specific state law product liability case. Id. at 13. The difficulty in asking the jury to apply such a finding from a disparate context might “result [i]n a mini-trial on the DOJ decision itself.” Id. Whatever jury instruction was decided upon would be complicated because it would have to explain how the Manipulation Issue—which is overbroad as applied here—would relate to the different CPLA standard. See Def.’s Mem., Doc. No. 497, at 18. Finally, and importantly, PM USA points out that almost all other courts that have considered whether to give preclusive effect to certain issues purportedly decided in DOJ have declined to do so, often based on lack of efficiency gains. See Def.’s Mem., Doc. No. 497, at 3 n.2 (citing eleven cases for that proposition); id. at 14 (citing Schwab v. Philip Morris USA Inc.,

449 F. Supp. 2d 992, 1079 (E.D.N.Y. 2006), rev’d on other ground sub nom. McLaughlin v. Am. Tobacco Co., 522 F.3d 215 (2d Cir. 2008) (noting that “little efficiency would be gained” because “plaintiffs’ proof of reliance and damages would almost certainly—as a matter of legal burden and persuasive strategy—require presentation of all evidence available to them of defendants’ alleged scheme”); Grisham v. Philip Morris, Inc., 670 F. Supp. 2d 1014, 1037 (C.D. Cal. 2009) (pointing out that, to avoid running afoul of Supreme Court precedent that forbids punitive damages awards based on harm to non-parties, the plaintiff would still need to “introduce voluminous evidence regarding the specific harms that Defendants’ wrongful actions caused her” which evidence “largely duplicates the DOJ case’s basic findings, thus negating the efficiency-related benefits provided by issue preclusion”); In re Light Cigarettes, 691 F. Supp. 2d 239, 251 (D. Me. 2010)).11 iii. Discussion In my view, giving the Manipulation Issue preclusive effect would not produce many (if

any) gains in efficiency, would have significant potential to confuse the jury, and would be fundamentally unfair because of the differences in the scope, type, and complexity between the DOJ case and this case. Bifolck and PM USA agree that giving the Manipulation Issue preclusive effect would establish only a quite helpful fact for Bifolck, who would still need to prove that PM USA negligently designed Marlboros and Lights within the meaning of the CPLA between the 1970s and 2000. Thus, PM USA would be allowed to contest the Manipulation Issue as applied to the particular products and times at issue. The upshot is that almost all the evidence that would have been required absent giving the Manipulation Issue preclusive effect would still be required if I gave the Manipulation Issue preclusive effect. PM USA also convincingly argues—and other courts have noted—that it would need to introduce such

evidence to assert affirmative defenses—e.g., comparative negligence—and to defend itself against, for instance, an unlawful award of punitive damages. Giving the Manipulation Issue preclusive effect would not clearly shorten the trial or streamline testimony. The experience in Engle progeny trials—in which the plaintiff need not prove the defendant’s general liability (a broader preclusive effect than sought here)—proves the point: plaintiffs call some of the same expert witnesses that Bifolck plans to call here, those

11 Bifolck attempts to distinguish all these cases by arguing that in each the plaintiff sought broader preclusion— e.g., Grisham, 670 F. Supp. 2d at 1032 (2,600 DOJ findings); In re Light Cigarettes, 691 F. Supp. 2d at 250 (1,083 DOJ findings); and Schwab, 449 F. Supp. 2d at 1077–78—than Bifolck seeks here, which Bifolck characterizes as just “a single, purely factual issue.” Pl.’s Reply, Doc. No. 502, at 6–7. However, as discussed above, Bifolck seeks preclusion on the Manipulation Issue, which, by Bifolck’s own briefing, implicates at least 54 factual findings from the DOJ case. See Pl.’s Supp. Opp’n, Doc. No. 344, at 13. experts testify about many background facts, and the cases last, on average, 13 to 14 trial days. That is no different than my estimate for this trial if I did not give the Manipulation Issue preclusive effect.12 Were I to give preclusive effect to the Manipulation Issue, whatever instruction I gave the

jury would be difficult to understand and potentially prejudicial to PM USA. I would instruct the jury that, in general, they must accept as fact that PM USA manipulated cigarette design and composition to assure nicotine delivery levels that create and sustain addiction. However, I would continue, that does not mean that PM USA manipulated Marlboros and Lights any time between the 1970s and 2000. The average juror would be confused by that instruction, and would likely jump to the conclusion that PM USA was liable for negligent design under the CPLA. Although the DOJ case was monumental and important, it does not represent the universe of tobacco litigation. The differences between the DOJ case and this case are many. In the DOJ case, which concluded in 2006, the United States sued PM USA and many similar companies

alleging that they acted together to defraud the American public. A judge found that over the course of 50 years, that group of defendants knew that nicotine caused addiction, lied about it, and made their cigarettes with enough nicotine to ensure that smokers would get and stay addicted. In this case, a single plaintiff sued a single tobacco company alleging that that company negligently designed two particular products between the 1970s and 2000 within the meaning of the CPLA. Importantly, too, this case involved a jury trial whereas DOJ was a bench

12 Before addressing the Preclusion Motion (and thus envisioning a trial in which the Manipulation Issue would not be given preclusive effect), I estimated that this case would last 13 trial days. See Tr. of Jury Selection, Doc. No. 453, at 19:23–20:2. Although jury selection was held on October 12, 2017—the day after the hearing on the Preclusion Motion—the estimated trial dates had been decided well before then. trial. Although courts may give preclusive effect to findings from non-jury trials, see, e.g., Parklane Hosiery, 439 U.S. at 333–37, many courts decline to give preclusive effect to issues purportedly decided in DOJ in part based on the potential unfairness of depriving the defendant a jury trial. See, e.g., Grisham, 670 F. Supp. 2d at 1036; In re Light Cigarettes, 691 F. Supp. 2d at

251; Shaffer v. R.J. Reynolds Tobacco Co., 860 F. Supp. 2d 991, 998 (D. Ariz. 2012); Curtis v. Altria Grp., Inc., 2009 WL 5820516, aff’d, 792 N.W. 2d 836, 853–55 (Minn. Ct. App. 2010), rev’d on other grounds, 813 N.W. 2d 891 (Minn. 2012); Pooshs v. Philip Morris USA, Inc., 904 F. Supp. 2d 1009, 1034 (N.D. Cal. 2012). Finally, as discussed above, the timing of Bifolck’s motion also suggests that giving the Manipulation Issue preclusive effect would be unfair. 2. Inconsistent Prior Verdicts In the fairness inquiry, I may consider whether “the judgment relied upon as a basis for estoppel is itself inconsistent with one or more previous judgments in favor of the defendant.” Parklane Hosiery, 439 U.S. at 330; see also Bifolck, 936 F.3d at 84. The parties spend many

pages focusing on whether the prior verdicts that PM USA identifies as inconsistent with the judgment in the DOJ case are, in fact, inconsistent. I examine the arguments in some detail and explain why, when examined at a low level of generality, both parties have decent arguments on this point. Still, when considered at a high level of generality—which I believe is appropriate when undertaking a broad fairness analysis—the sheer number of defense verdicts for tobacco companies in cases substantially similar to DOJ gives me further hesitation and counsels slightly against granting the Manipulation Issue preclusive effect. Cf. Schwab, 449 F. Supp. 2d at 1079 (“[D]efendants have won so many of the tobacco cases that applying the rule of the Restatement according conclusive effect to the last of the series of litigations is inappropriate.”); City of St. Louis v. Am. Tobacco Co., Inc., 2010 WL 2917188, at *7 n.11 (Mo. Circ. Ct. June 2, 2010) (“[T]he vast multitude of prior inconsistent verdicts the Court would have to ignore” to give preclusive effect to DOJ findings “is simply too great to try to rationalize or explain away”). i. The parties’ arguments

PM USA argues that granting the Manipulation Issue preclusive effect would be unfair given that PM USA has won defense verdicts in over 20 cases in which the plaintiff “rais[ed] allegations related to nicotine addiction and manipulation similar to those raised in the DOJ case.” See Def.’s Mem., Doc. No. 497, at 2; see also App’x A of Def.’s Mem., Doc. No. 497-1 (chart of those cases and relevant allegations and arguments in each); Ex. A to App’x A (DVD with Exs. A1 to A109, which are .pdfs of complaints, jury opening statements and closing arguments, and verdict forms in App’x A cases). Nine of the cases that PM USA cites as “inconsistent” with the Manipulation Issue were decided by general verdict. See Pl.’s Mem., Doc. No. 495, at 25 n.22; App’x A of Def’s Mem., Doc. No. 497-1.13 One of those cases did not result in a jury verdict; instead, it ended in a hung jury.14 (Although PM USA argues that I can

consider a hung jury as an “inconsistent judgment” for purposes of this fairness analysis, I do not

13 Those are: (1) Blue Cross & Blue Shield of N.J., Inc. v. Philip Morris, Inc., 98 Civ. 3287 (E.D.N.Y. Aug. 11, 2000) (Ex. A1 to App’x A of Def.’s Mem, Doc. No. 497-1); (2) Iron Workers Local Union v. Philip Morris, Inc., No. 1:97 Civ. 1422 (N.D. Ohio Sept. 15, 1997) (Ex. A5); (3) Larsen v. Philip Morris USA Inc., 002-00406-02 (Mo. Cir. Ct. Nov. 1, 2007) (Ex. A9); (4) Vandenburg v. Brown & Williamson Tobacco Corp., et al., 03CV237238 (Mo. Cir. Ct. Mar. 21, 2003) (Ex. A21); (5) Welch v. Brown & Williamson Tobacco Co., et al., Case No. 00-CV-209292 (Mo. Cir. Ct.) (Ex. A41); (6) Lucier v. Philip Morris, Inc., No. 02AS01909 (Cal. Super. Ct. 2000) (Ex. A53); (7) Tompkin v. Am. Tobacco Co., Case No. 5:94 CV 1302 (N.D. Ohio Jun. 9, 1995) (Ex. A75); (8) Dunn v. RJR Nabisco Holdings Corp., et al., No. 18C01-9305-CT-06 (Ind. Super. Ct. Mar. 19, 1993) (Ex. A90); (9) Rogers v. R.J. Reynolds Tobacco Co., No. C87-2106 (Ind. Cir. Ct. Mar. 24, 1988) (Ex. A94). 14 See Falise v. Am. Tobacco Co., No. 99-cv-7392 (E.D.N.Y.). rely on that case.15) And sixteen were either special verdicts or general verdicts with interrogatories.16 PM USA relies most heavily on two civil RICO cases—one federal and one state—in which juries reached defense verdicts. PM USA argues that the experience in those cases alone

indicates that juries in cases extremely similar to the DOJ case have reached verdicts inconsistent with the judgment in DOJ. See Def.’s Mem., Doc. No. 497, at 8. The first is Blue Cross & Blue Shield of N.J., Inc. v. Philip Morris, Inc., 98 Civ. 3287 (E.D.N.Y. Aug. 11, 2000). In Blue Cross, the plaintiffs asserted RICO, antitrust, and various state law (statutory and common law) claims.17 The jury returned a general verdict rejecting, in relevant part, the plaintiffs’ RICO claims.18 PM USA points out that Blue Cross was quite similar to the DOJ case: The complaint in Blue Cross pled allegations regarding nicotine manipulation, and the plaintiffs in Blue Cross called at least four of the same expert witnesses who testified in DOJ. See Def.’s Mem., Doc.

15 PM USA cites Setter v. A.H. Robins Co., 748 F.2d 1328, 1330 (8th Cir. 1984) for the proposition that “courts have relied on hung juries as inconsistent verdicts.” Def.’s Mem., Doc. No. 497, at 7. I do not think Setter stands for that proposition. In Setter, the Eighth Circuit affirmed a trial court’s decision not to apply NMOIP because of inconsistent verdicts in prior product-liability cases involving the same defendant when, of 21 prior cases, 12 had ended in verdicts favorable to the defendants, 8 had ended in verdicts favorable to the plaintiffs, and one had ended in a hung jury. See Setter, 748 F.2d at 1330. In making its determination, the trial court had plainly relied much more on the nearly even split of prior verdicts than the single hung jury. In any event, because PM USA relies on just one case that ended in a hung jury, it is not an essential part of its argument. 16 Those are: (1) Grill v. Philip Morris USA Inc., No. 05-CV-9174 (S.D.N.Y. 2005) (Ex. A13 to App’x A of Def.’s Mem., Doc. No. 497-1); (2) Haglund v. Philip Morris USA Inc., No. 01-2367 (Mass. Super. Ct. 2002); (Ex. A17); (3) Beckum v. Philip Morris USA Inc., No. 02-01836 (Fla. Cir. Ct. Feb. 6, 2003) (Ex. A25); (4) Longden v. Philip Morris USA, Inc., No. 00-C-442 (N.H. Super. Ct. Aug. 3, 2001) (Ex. A33); (5) Coolidge v. Philip Morris Inc., No. RIC-361063 (Cal. Super. Ct. Jul. 11, 2001) (Ex. A29); (6) Reller v. Philip Morris Inc., No. BC 261796 (Cal. Super. Ct. Nov. 14, 2001) (Ex. A37); (7) Allen v. R.J. Reynolds Tobacco Co., Case No. 01-4319 (S.D. Fla. 2001) (Ex. A45); (8) Inzerilla v. The Am. Tobacco Co., No. 11754/96 (N.Y. Super. Ct. 1996) (Ex. A49); (9) Carter v. Philip Morris Corp., No. 4567 (Pa. Ct. Comm. Pl. 1999) (Ex. A57); (10) Tune v. Philip Morris Inc., No. 97-4678-CI-11 (Fla. Cir. Ct. Mar. 10, 1998) (Ex. A63); (11) Hyde v. Philip Morris, Inc., No. 97-359ML (D.R.I.) (Ex. A67); (12) In re Tobacco Litig., No. 00-C-6000 (S.D.W. Va. Jan. 31, 1997) (Ex. A71); (13) Mehlman v. Philip Morris Inc., No. L- 1141-00 (N.J. Super. Ct. Feb. 4, 1999) (Ex. A79); (14) Apostolou v. The Am. Tobacco Co., No. 13162/97 (N.Y. Sup. Ct.) (Ex. A83); (15) Butler v. Philip Morris, Inc., No. 94-5-53 (Miss. Cir. Ct.) (Ex. A87); (16) Cipollone v. Liggett Group, Inc., No. 83-2864 (D.N.J.) (Ex. A98). 17 See, e.g., Blue Cross and Blue Shield of N.J., Inc. v. Philip Morris, Inc., 36 F. Supp. 2d 560, 566–88 (E.D.N.Y. 1999). 18 The jury rendered a plaintiff’s verdict on a state law claim, but even that was overturned on appeal for reasons irrelevant here. See Empire Healthchoice, Inc. v. Philip Morris USA, Inc., 393 F.3d 312, 315 (2d Cir. 2004). No. 497, at 9 n.7. Similarly, in Iron Workers Local Union No. 17 Ins. Fund v. Philip Morris, Inc., a jury returned a general defense verdict against plaintiffs who had brought a state-law equivalent of a federal RICO claim. No. 1:97 Civ. 1422 (N.D. Ohio Sept. 15, 1997); see also Iron Workers Local Union No. 17 Ins. Fund v. Philip Morris, Inc., 186 F.R.D. 453, 454–55

(N.D. Ohio 1999) (describing jury’s verdict). In Iron Workers, too, PM USA points out, the plaintiffs argued that the defendants wrongfully manipulated the nicotine in their cigarettes. See Def.’s Mem., Doc. No. 497, at 10. PM USA argues that those two verdicts alone establish that giving the Manipulation Issue preclusive effect would be unfair. Bifolck attacks PM USA’s reliance on Blue Cross and Iron Workers on a few grounds. First, Bifolck argues that general jury verdicts—both Blue Cross and Iron Workers were general verdicts—cannot be “inconsistent” with the judgment in the DOJ case because it is “impossible to determine the specific basis for” the jury’s determination. Pl.’s Mem., Doc. No. 495, at 25. That is so, Bifolck argues, even when plaintiffs specifically alleged and argued, among other theories, that the defendants manipulated nicotine levels to make cigarettes more addictive. Pl.’s

Reply, Doc. No. 502, at 1. Typically, tobacco companies “contest[] every element of every cause of action asserted”; in this very case, PM USA asserted 33 affirmative defenses. Pl.’s Mem., Doc. No. 495, at 26. Bifolck asserts (incorrectly, as I explain below) that Second Circuit precedent indicates that “general verdicts are inherently ambiguous, and should be given no weight in determining whether collateral estoppel should apply.” See Pl.’s Supp. Mem., Doc. No. 507 at 2 (citing Tucker v. Arthur Andersen & Co., 646 F.2d 721, 729 (2d Cir. 1981)). Further, Bifolck pokes holes in the two out-of-circuit cases that PM USA cites to support the proposition that general jury verdicts can be inconsistent verdicts for the purposes of preclusion. See Pl.’s Supp. Mem., Doc. No. 507, at 2 & n.2 (explaining that in both Hardy v. Johns-Manville Sales Corp., 681 F.2d 334, 346 (5th Cir. 1982); and State Farm Fire & Cas. Co. v. Century Home Components, Inc., 550 P.2d 1185, 1192 (Or. 1976), the court satisfied itself that the jury had found a particular fact). Relying on Hardy and State Farm, PM USA argues that courts may rely on

“inconsistent” general jury verdicts in determining whether it would be fair to give preclusive effect to a particular issue decided in a prior bench trial. Even if that were not so, though, PM USA points to numerous special verdicts and general verdicts with interrogatories that are “inconsistent” with the judgment in DOJ. In PM USA’s view, some of those leave no doubt that the jury made a specific conclusion inconsistent with the Manipulation Issue. See generally supra n.16; see also Def.’s Reply, Doc. No. 501, at 6 (citing, e.g., Verdict Form, Inzerilla v. The Am. Tobacco Co., No. 11754/96 (N.Y. Super. Ct. 1996), Ex. A49 to App’x A of Def.’s Mem., Doc. No. 497, at 9 (finding that the plaintiff had not established “that the cigarettes made by Philip Morris and smoked by Mrs. Inzerilla were negligently designed or manufactured”); Verdict Form, In re Tobacco Litig., No. 00-C-6000 (S.D.W. Va. Jan. 31, 1997), Ex. A71 to

App’x A of Def.’s Mem., Doc. No. 497, at 2 (answering “no” to the question “[a]re the cigarettes which were manufactured, designed and sold by the defendant Tobacco Companies defective, in the sense of being not reasonably safe for their intended use?”)). Bifolck also attacks PM USA’s reliance on Blue Cross and Iron Workers on the basis that other courts have voiced various doubts about those cases. For instance, in Grisham, the court— although ultimately refusing, on other grounds, to give preclusive effect to issues purportedly decided in the DOJ case—held that Blue Cross and Iron Workers were not “inconsistent” with the DOJ case because they were different in too many respects. 670 F. Supp. 2d at 1034–35. The Grisham court discounted Blue Cross because it had been a split verdict: The Blue Cross jury had returned both a general defense verdict on the federal RICO claim and also a plaintiff’s verdict on state-law fraud and misrepresentation claims. Id. at 1034. Regarding Iron Workers, the Grisham court held that “the nature of the jury instructions” made it “unclear whether the jury decided that the defendants did or did not engage in fraudulent activities.” Id. at 1035.

Thus, in the Grisham court’s view, no previous case “appears to include an ultimate finding of fact absolving tobacco companies of liability on the basis that they did not engage in fraudulent activities.” Id. at 1035.19 Bifolck also points out that—in a different context—Judge Kessler herself noted in dicta that no cases “contain[ed] findings that contradict the findings in this case.” See Pl.’s Mem., Doc. No. 495, at 28; United States v. Philip Morris USA, Inc., 907 F. Supp. 2d 1, 20 (D.D.C. 2012). PM USA responds that Bifolck’s reliance on Grisham and DOJ is weak. With respect to Grisham, PM USA emphasizes that the court’s language regarding prior inconsistent verdicts was (1) anomalous and (2) dicta because the Grisham court ultimately decided that it would be unfair to grant preclusive effect to certain issues purportedly decided in DOJ. See Def.’s Reply,

Doc. No. 501, at 5. Regarding DOJ, PM USA argues that Bifolck’s reliance is entirely misplaced because the DOJ court was undertaking a different analysis in a different posture—a First Amendment analysis regarding whether the proposed corrective statements were “controversial” under Zauderer v. Office of Disciplinary Counsel of Supreme Court of Ohio, 471 U.S. 626 (1985)—having nothing to do with collateral estoppel. See id.

19 Bifolck also points to Hunter v. Philip Morris USA Inc., in which an Alaska trial court adopted the Grisham court’s reasoning and granted special weight to DOJ findings because of how much evidence had been presented and how vigorously the case had been defended. No. 4BE-06-0040CI, at 5 (Ak. Super. Ct. Oct. 12, 2017) (Ex. G to Decl. of D. Golub, Doc. No. 496-7). Finally, Bifolck argues that I should discount or ignore the numerous verdicts that PM USA relies upon that were reached before DOJ was decided in 2006 because those plaintiffs faced enormous disadvantages due to the cigarette manufacturers’ long history of suppressing and concealing evidence. See Pl.’s Mem., Doc. No. 495, at 25, 29–31. In contrast, Bifolck

argues, the DOJ court had the resources to engage fully with the cigarette manufacturers’ wrongdoing and to document it. See id. at 29–31. Even still, because of the cigarette manufacturers’ wrongdoing, it is “virtually impossible to know what materials existed prior to their destruction.” See id. at 30 (quoting DOJ, 449 F. Supp. 2d at 801 (¶ 3864)). For that reason, and because the DOJ court was so clear about what exactly it found, Bifolck emphasizes that the DOJ decision was more reliable than pre-DOJ verdicts. See Pl.’s Reply, Doc. No. 502, at 3 (“No private litigant could hope to recreate the vast record that served as the basis for the DOJ decision on the Manipulation Issue, and no lay fact finder could hope to render a decision as well-informed and reliable.”). PM USA disagrees and emphasizes that judicial hesitance in applying NMOIP arises

from precisely what Bifolck asks me to do here: adhering to the last-in-time decision without proper consideration for earlier inconsistent judgments. See, e.g., Parklane Hosiery, 439 U.S. at 330 n.14. PM USA concedes that the DOJ court found that the defendants had used improper litigation tactics in the past; but that finding alone—although it may move the needle a bit—is not reason enough to disregard so many prior inconsistent verdicts. Along the same lines, PM USA points out that numerous plaintiffs before 2006 were not at any meaningful disadvantage. For instance, not only did Blue Cross retain Dewey Ballantine, but by the time of that litigation, Blue Cross had access to millions of tobacco company documents that were made available on the internet in 1998 as part of the resolution in a case brought by 40 state attorneys general. See Def.’s Reply, Doc. No. 501, at 8. ii. Discussion On balance, this factor weighs slightly in favor of PM USA. Viewed at a low level of

generality, each side distinguishes the other’s authority relatively persuasively. Given the inconclusiveness of that analysis, I look at the arguments from a higher level of generality. As I have already mentioned, the broad nature of my task—to determine whether applying NMOIP to the Manipulation Issue would be unfair—allows me to do so. As some other courts have remarked, the sheer number of defense verdicts in cases highly similar to the DOJ case indicates that some juries have found something inconsistent with the judgment in the DOJ case (and the findings underlying that judgment). See, e.g., Schwab, 449 F. Supp. 2d at 1079; City of St. Louis, 2010 WL 2917188, at *7 n.11. On one hand, some of Bifolck’s arguments are attractive. The DOJ case was, no doubt, monumental. The parties presented an enormous amount of evidence and zealously litigated the

case; each side expended huge amounts of time and energy; and the decision was immense, careful, and detailed. In addition, it seems clear, as a matter of fact, that PM USA—and other cigarette manufacturers—engaged in improper internal campaigns to obfuscate the effects of their cigarettes. For those reasons, one might conclude that the DOJ case signaled a turning point in tobacco litigation and that, perhaps, I should discount pre-2006 verdicts. On the other hand, Bifolck simply tries too hard to thread the needle. Several courts have—without searching analysis—concluded that the sheer number of inconsistent prior verdicts counsels against giving DOJ findings preclusive effect. See, e.g., Schwab, 449 F. Supp. 2d at 1079; City of St. Louis, 2010 WL 2917188, at *7 n.11. Even more courts have placed particular emphasis on Blue Cross (federal RICO) and Iron Workers (state-law RICO) in concluding that giving DOJ findings preclusive effect would be unfair. See, e.g., Shaffer, 860 F. Supp. 2d at 996–97 (citing Aspinall v. Philip Morris Cos., Inc., 2012 WL 1063342 (no page numbers) (Mass. Super. Ct. Mar. 13, 2012)); City of St. Louis, 2010 WL 2917188, at *8; Schwab,

449 F. Supp. 2d at 1079; Curtis, 2009 WL 5820516 (no page numbers)). I disagree with Bifolck’s claim that general jury verdicts should not figure into my analysis. It is true that when a party seeks to give a certain issue preclusive effect, that party normally cannot rely on a general jury verdict to establish that that issue has been conclusively determined. See, e.g., Tucker, 646 F.2d at 729; United Access Technologies, LLC v. Centurytel Broadband Servs. LLC, 778 F.3d 1327, 1331 (Fed. Cir. 2015) (“It is well established that a general jury verdict can give rise to collateral estoppel only if it is clear that the jury necessarily decided a particular issue in the course of reaching its verdict.”) (citing Ashe v. Swenson, 397 U.S. 436, 444 (1970)). But that does not describe the situation here. Bifolck—not PM USA—is the party seeking to give the Manipulation Issue preclusive effect on the ground that that Issue

was determined in the DOJ case. In contrast, PM USA argues against the application of NMOIP and merely relies on general jury verdicts as relevant data points in the fairness inquiry. Furthermore, numerous inconsistent verdicts that PM USA identifies were either special verdicts or general verdicts with interrogatories, in which the jury was “specifically asked in the verdict form whether PM USA manufactured cigarettes that were defectively designed” and answered in the negative. See Def.’s Reply, Doc. No. 501, at 6 (citing Verdict Form, Reller v. Philip Morris Inc., No. BC 261796 (Cal. Super. Ct. Nov. 14, 2001), Ex. A37 to App’x A of Def.’s Mem., Doc. No. 497, at 2 (finding by a 10-to-2 vote that there was no defect in design); Verdict Form, Inzerilla, No. 11754/96, Ex. A49 to App’x A of Def.’s Mem., Doc. No. 497, at 9; Verdict Form, In re Tobacco Litig., No. 00-C-6000, Ex. A71 to App’x A of Def.’s Mem., Doc. No. 497, at 2). There simply is too much for Bifolck to explain away. D. May 2020: The Analysis is the Same. I have already explained why I believe the correct method for my analysis is to take

myself back in time to October 11, 2017 and imagine that I had properly reached the fairness inquiry with respect to the Manipulation Issue. That is what I have attempted to do above, and, in my view, my inquiry could end there. In the interest of completeness, though, I also examine whether it would be unfair from the vantage point of today to order a new trial and, in that trial, to give the Manipulation Issue preclusive effect. Doing the analysis from this vantage point does not change the outcome: it would still be unfair. Viewing the situation today, certain additional factors do weigh in Bifolck’s favor. For example, as the Second Circuit noted, Jupe’s trial testimony was not forthcoming—Jupe “paint[ed] a cloudy picture of what [PM USA] could or could not do and did or did not know”— and was inconsistent with the findings in DOJ that support the Manipulation Issue. See Bifolck,

936 F.3d at 85. More specifically, Jupe testified that although Marlboros and Lights “were dangerous” and “not safe,” they were neither “defective” nor “unreasonably dangerous” because they had “inherent risk.” See Trial Tr., Doc. No. 468, at 2165:14–25. Jupe explained that the design of Marlboros and Lights made them “no more danger[ous] than the inherent risk of inhaling burned tobacco.” Id. at 2166:21–23. When asked what PM USA scientists and employees knew about the scientific effects of nicotine in the 1970s and 80s, Jupe gave cagey responses. He acknowledged that scientists at PM USA understood that nicotine activated nicotine receptors but refused to answer whether he agreed that “there’s a physiological dependence that is caused by the nicotine addiction.” Id. at 2209:8–2210:3. Jupe testified that there was no “company position” on “the minimum effective dose necessary to addict someone on a daily basis.” Id. at 2210:9–21. Jupe refused to acknowledge that Philip Morris had an interest in getting smokers addicted. Id. at 2214:19–21. Jupe also seemed to say that PM USA did not care about or pay attention to the amount of nicotine in its Marlboros and Lights because

the amount of nicotine was merely “a nuance” that generally followed the amount of tar PM USA decided to put in a cigarette. Id. at 2217:16–25. Indeed, Jupe denied that PM USA had designed an addictive cigarette because “[c]igarettes are inherently addictive.” Id. at 2213:25– 2214:2. Further, Jupe’s testimony may have been important to the jury. I instructed the jury that PM USA’s design of Marlboros and Lights was “unreasonably dangerous” if the cigarettes were either “unnecessarily addictive” or “unnecessarily carcinogenic.” Trial Tr., Doc. No. 469, at 2306:22–2307:2. The jury asked, in a note, whether it was “permitted to find that a product was manufactured so as to be ‘reasonably dangerous?’” and, if so, what did “reasonably dangerous” mean? Trial Tr., Doc. No. 470, at 2422:19–22. I referred the jury back to the relevant portion of

the jury instructions. See id. at 2426:20–2427:2. Later, I also gave the jury an Allen charge when it indicated it was having trouble reaching unanimity. See Trial Tr., Doc. No. 471, at 2467:21–2469:17. About an hour after that, the jury returned a defense verdict. Bifolck argues that Jupe’s testimony “flatly contradicted” the Manipulation Issue and thus led to confusion at least as harmful as any confusion that would arise from giving the Manipulation Issue preclusive effect. See Pl.’s Mem., Doc. No. 495, at 36. Despite those arguments, viewing the situation from today does not change my mind for several reasons. First, ruling in Bifolck’s favor at this point would entail significantly more prejudice to PM USA than it would have in October 2017 because it would require vacatur of a favorable judgment, followed by a new trial.20 Further, the trial experience in this matter indicates that having given the Manipulation Issue preclusive effect would not have increased “efficiency” or meaningfully limited the evidence presented. Bifolck points to the testimony of Dr. Neil Grunberg,21 Dr. William Farone,22 Dr. Kenneth Cummings,23 and Richard Jupe24 and

argues that, had I given the Manipulation Issue preclusive effect, much of that testimony could have been significantly shortened. See Pl.’s Reply, Doc. No. 502, at 4–5 nn. 5–13. But the portions of trial transcript that Bifolck claims would be “unnecessary or greatly shortened” total only 293 pages—out of 2,478 pages. See id. Moreover, since October 11, 2017, PM USA has won at least two more jury verdicts that are—at least arguably—inconsistent with the DOJ case: the first iteration of this case, and another in Massachusetts. See Def.’s Notice of Supp. Auth., Doc. No. 510 (citing Main v. Philip Morris USA Inc., et al., No. 1684-cv-03883 (Mass. Tr. Ct. Dec. 23, 2019)). Bifolck’s arguments regarding Jupe’s testimony are not frivolous, but they assume too much. Although Jupe’s testimony was indeed unclear and contradicted findings in the DOJ case

underlying the Manipulation Issue, it is not possible to know how the jury interpreted that testimony. Bifolck apparently assumes that the jury took all of Jupe’s testimony at face value.

20 I understand that the parties disagree about whether—were I to hold that giving the Manipulation Issue preclusive effect would not be unfair—I would order a new trial. Compare Pl.’s Mem., Doc. No. 495, at 47 (arguing new trial is appropriate remedy in this situation) with Def.’s Reply, Doc. No. 501, at 18–19 (arguing that ordering new trial would exceed mandate from the Second Circuit). I need not address that dispute because I hold that giving the Manipulation Issue preclusive effect would be unfair from any perspective. 21 Dr. Grunberg was the scientific editor of the 1988 Surgeon General’s report and is an expert on nicotine and addiction. His testimony lasted almost two full trial days. See Trial Tr., Doc. Nos. 454 and 459, at 64:18–580:21. 22 Dr. Farone was a biochemist employed by Philip Morris in the 1970s and 1980s as director of applied research. His testimony lasted almost two full trial days. See Trial Tr., Doc. Nos. 460 and 461, at 613:20–927:19. 23 Dr. Cummings is a cancer and tobacco researcher who worked for 30 years on smoking issues at Roswell Park Cancer Institute. His testimony lasted more than one full trial day. See Trial Tr., Doc. Nos. 464 and 465, at 1260:20–1557:10. 24 Jupe’s testimony lasted almost one full trial day. See Trial Tr., Doc. No. 468, at 2013:7–2244:22. He was called as an expert on cigarette design, testing, manufacturing, and processing. See Trial Tr., Doc. No. 468, at 1993:6–17, 2029:9–14. But, in my view, it is just as likely that the jury viewed at least portions of Jupe’s cagey testimony with skepticism.25 Indeed, I note that this was a close case: the jury deliberated for two days and I had to give it an Allen charge. Relatedly, the jury was not necessarily “confused” when it asked me what it meant for a product to be “reasonably dangerous.” That is an

inherently complicated term that requires significant interpretation: any jury might struggle with it. Bifolck makes one further argument—for the first time26—on remand: that giving the Manipulation Issue preclusive effect must be fair because not doing so would allow PM USA to elicit testimony that violates the injunctive relief ordered in the DOJ case. See Pl.’s Mem., Doc. No. 495, at 15. In 2006, the DOJ court ordered that PM USA was “[1] permanently enjoined from making, or causing to be made in any way, any material, false, misleading, or deceptive statement or representation, or [2] engaging in any public relations or marketing endeavor that is disseminated to the United States public and that misrepresents or suppresses information concerning cigarettes.” DOJ, 449 F. Supp. 2d at 938. (I refer to this as the “DOJ Injunction.”)

Bifolck argues that the DOJ Injunction—by enjoining PM USA from causing misleading statements “to be made in any way”—bars PM USA from eliciting misleading testimony in separate, later proceedings. See Pl.’s Mem., Doc. No. 495, at 16–17. Despite the DOJ Injunction’s apparent focus on publicly disseminated statements, Bifolck argues that only clause [2] of the DOJ Injunction has a public dissemination element. See Pl.’s Reply, Doc. No. 502, at

25 Indeed, in his closing argument, Bifolck’s lawyer alternatively relied on Jupe’s testimony for certain points, see, e.g., Trial Tr., Doc. No. 469, at 2351:1–7, and denigrated its trustworthiness, see id. at 2342:19–24. 26 The parties disagree about whether Bifolck has waived this argument. PM USA argues that it is “black-letter law” that when a plaintiff has an opportunity and incentive to raise a particular issue on an initial appeal and does not “the issue is considered waived and the law of the case doctrine bars the district court on remand . . . from reopening such issues unless the mandate can reasonably be understood as permitting it to do so.” Def.’s Reply, Doc. No. 501, at 15 (quoting Call Ctr. Techs., Inc. v. Interline Travel & Tour, Inc., 622 F. App’x 73, 75 (2d Cir. 2015)). Bifolck counters that he has not waived the issue because the Second Circuit directed me to undertake “a full analysis of the fairness considerations.” Pl.’s Suppl. Mem., Doc. No. 507, at 4–5; Bifolck, 936 F.3d at 85. 9. In contrast, clause [1] by its unambiguous terms prohibits PM USA from eliciting testimony of the kind that it did in this trial from Jupe (noted above). Even if the public dissemination language from clause [2] is applicable to clause [1], Bifolck argues, witness testimony in a public trial satisfies that requirement. See id.

PM responds first that Bifolck has waived this argument. See supra n.26. Second, PM USA explains that Bifolck does not have standing to ask me to enforce the DOJ Injunction because (1) all such requests must be made to the issuing court, and (2) a third party cannot enforce an injunction entered in a case originally brought by the federal government. See Def.’s Reply, Doc. No. 501, at 17 n.4. (Bifolck concedes that he is not asking me to do so.27) Despite believing that I should not reach this argument, PM USA also responds to Bifolck on the merits and explains that the DOJ Injunction does not apply to PM USA’s efforts to defend itself at trial. See id. at 15. Rather, PM USA points out that the DOJ Injunction applies to public-facing statements. Id. at 16. Additionally, PM USA argues that curbing its ability to offer a vigorous trial defense would be unconstitutional by violating its right to due

process under the Fifth Amendment and—because the terms of the DOJ Injunction do not clearly prohibit PM USA’s speech in the form of eliciting testimony as a trial defense—under the First Amendment as a prior restraint on speech. See id. at 17–18. Relatedly—but far afield from the concerns central to this inquiry—the parties disagree whether the Noerr-Pennington doctrine28

27 Bifolck denies that he is asking me to “enforce” the DOJ Injunction; rather, Bifolck argues that the DOJ Injunction is simply a strong data point indicating that it would not be unfair to apply NMOIP in this case. Pl.’s Reply, Doc. No. 502, at 8; Hr’g Tr., Doc. No. 512, at 8:8–12 (“Now, we’re not saying here that you could have or that we could have enforced the injunctions here. What we’re saying is it can’t be unfair to prevent Philip Morris from doing something in court that it has been enjoined from doing by a federal court.”). 28 The Noerr-Pennington doctrine is “a doctrine, rooted in the Petition Clause of the First Amendment, that protects ‘an attempt to persuade the legislature or the executive to take particular action with respect to a law . . . .’” United States v. Philip Morris USA Inc., 566 F.3d 1095, 1123 (D.C. Cir. 2009) (quoting E. R.R. Presidents Conf. v. Noerr Motor Freight, Inc., 365 U.S. 127, 136 (1961)). PM USA argues that “the right to petition extends to all departments of the Government” and the “right of access to the courts is . . . one aspect of the right to petition.” applies to protect PM USA’s right to petition the government through in-court activities. Compare Pl.’s Reply, Doc. No. 502, at 9 (arguing that the DOJ court and the D.C. Circuit have concluded that the Noerr-Pennington doctrine did not protect PM USA’s right to petition the government by asserting deliberately false claims in judicial proceedings) with Def.’s Reply,

Doc. No. 501, at 18 & n.6 (disputing that the DOJ court or the D.C. Circuit made any finding that PM USA engaged in fraud during the defense of trials, and so PM USA may still rely on Noerr-Pennington in that context). Although Bifolck may indeed have waived this issue, even if I did consider it, it would not alter my conclusion that giving the Manipulation Issue preclusive effect would be unfair. That is largely because—as even Bifolck agrees I should—I consider the DOJ Injunction only as another relevant data point in my fairness inquiry. As another relevant data point, it does not move the needle much because the connection between the DOJ Injunction and this case is tenuous. An injunction must be specific and definite enough to “apprise those within its scope of

the conduct that is being proscribed.” City of N.Y. v. Mickalis Pawn Shop, LLC, 645 F.3d 114, 143 (2d Cir. 2011) (citing S.C. Johnson & Son, Inc. v. Clorox Co., 241 F.3d 232, 240–41 (2d Cir. 2001) (internal quotation marks omitted)). “Since an injunctive order prohibits conduct under threat of judicial punishment, basic fairness requires that those enjoined receive explicit notice of precisely what conduct is outlawed.” Id. (citing Schmidt v. Lessard, 414 U.S. 473, 476 (1974) (internal quotation marks omitted)). Although the threat of judicial punishment—i.e., holding PM USA in contempt for violating the DOJ Injunction—is not at issue here, the same considerations, in my view, limit the applicability of the DOJ Injunction to this fairness inquiry.

Def.’s Reply, Doc. No. 501, at 18 (quoting California Motor Trans. Co. v. Trucking Unlimited, 404 U.S. 508, 510 (1972)). Jupe’s testimony was confusing, evasive, and seemingly contradictory. But Bifolck was able to argue that point to the jury (and did, in part, as discussed above, see supra n.25). And although the DOJ court did remark upon the defendants’ improper litigation tactics, the case was not about fraud in courtroom proceedings. Further, by its basic terms, the DOJ Injunction does not seem to

reach courtroom testimony: Bifolck’s bifurcated reading is plausible, but it is strained. Thus, even were I to consider this argument, it would not change my mind. IV. Conclusion For the foregoing reasons, I hold that granting the Manipulation Issue preclusive effect in this case would have been—and would be—unfair.

So ordered. Dated at Bridgeport, Connecticut, this 12th day of May 2020.

/s/ STEFAN R. UNDERHILL Stefan R. Underhill United States District Judge