GAMCO v. Vivendi, No. 13-1194 (2d Cir. 2016)

Annotate this Case
Justia Opinion Summary

This appeal stems from the same set of underlying facts as those in In re Vivendi S.A. Securities Litigation, Nos. 15‐180‐cv(L), 15‐208‐cv(XAP), in which the court today issued a separate opinion. GAMCO, so-called "value investors," filed a securities fraud action against Vivendi under section 10(b) of the Securities Exchange Act of 1934 (Exchange Act), 15 U.S.C. 78j(b), as well as the Securities Exchange Commission’s (SEC) Rule 10b–5 (Rule 10b–5) promulgated thereunder, 17 C.F.R. 240.10b–5. The district court subsequently entered judgment for Vivendi. The court concluded that the record supports the district court’s conclusion that, if GAMCO had known of the liquidity problems and their concealment, GAMCO would still have believed Vivendi’s PMV to be “materially higher” than the public market price. The court also concluded that it was also not clearly erroneous for the district court to conclude that knowledge of Vivendi’s liquidity problems would not have changed GAMCO’s belief that a catalyst was likely — i.e., its belief that the market price would rise towards the PMV, if not immediately, then over the course of the next several years. In this case, the record at the trial simply does not establish that it was clearly erroneous for the district court to find that GAMCO, had it known of the liquidity problems at Vivendi, would have made the choice to buy the same securities it purchased. Accordingly, the court affirmed the judgment.

Download PDF
13 1194(L) GAMCO v. Vivendi UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT August Term 2015 (Argued: March 3, 2016 Decided: September 27, 2016) No. 13 1194(L), 13 1377(XAP) –––––––––––––––––––––––––––––––––––– GAMCO INVESTORS, INC., GAMCO GLOBAL SERIES FUNDS, INC., GABELLI CAPITAL ASSET FUND, THE GABELLI VALUE FUND, INC., THE GABELLI ASSET FUND, THE GABELLI GLOBAL MULTIMEDIA TRUST INC., THE GABELLI EQUITY TRUST, INC., Plaintiffs Appellants Cross Appellees, THE GAMCO MATHERS FUND, THE GABELLI CONVERTIBLE AND INCOME SECURITIES FUND, INC., GAMCO INTERNATIONAL GROWTH FUND, INC. Plaintiffs, v. VIVENDI UNIVERSAL, S.A., VIVENDI S.A., Defendants Appellees Cross Appellants. –––––––––––––––––––––––––––––––––––– Before: CABRANES, LIVINGSTON, and LYNCH, Circuit Judges. 1 Plaintiffs Appellants Cross Appellees, so called “value investors,” appeal from the judgment of the United States District Court for the Southern District of New York (Scheindlin, J.), granting judgment to the Defendants because they had rebutted the fraud on the market presumption of reliance invoked by the Plaintiffs as part of their claim under § 10(b) of the Securities Exchange Act of 1934, 15 U.S.C. § 78j(b). For the reasons stated below, we AFFIRM the judgment of the district court. ANDREW J. ENTWISTLE, Entwistle & Cappucci LLP, New York, N.Y. (Vincent R. Cappucci, Arthur V. Nealon, Jordan A. Cortez, Entwistle & Cappucci LLP, New York, N.Y., on the brief), for Plaintiffs Appellants Cross Appellees. MARK A. PERRY, Gibson, Dunn & Crutcher LLP, Washington, D.C. (Miguel A. Estrada, Lucas C. Townsend, Gibson, Dunn & Crutcher LLP, Washington, D.C.; Caitlin J. Halligan, Gabriel K. Gillett, Gibson, Dunn & Crutcher LLP, New York, N.Y.; James W. Quinn, Gregory Silbert, Weil, Gotshal & Manges LLP, New York, N.Y.; Daniel Slifkin, Timothy G. Cameron, Cravath, Swaine & Moore LLP, New York, N.Y., on the brief), for Defendants Appellees Cross Appellants. PER CURIAM: This appeal stems from the same set of underlying facts as those in In re Vivendi S.A. Securities Litigation, Nos. 15 180 cv(L), 15 208 cv(XAP), in which we have today issued a separate opinion. Plaintiffs Appellants Cross Appellees (collectively, “GAMCO”) are so called “value investors” who make their own estimation of the value of a publicly traded company’s securities and attempt to 2 buy such securities when the market price is lower than its own valuation, betting that the market price will rise over time. Between 2000 and 2002, GAMCO invested in the securities of Defendants Appellees Cross Appellants Vivendi Universal, S.A. and Vivendi S.A. (collectively, “Vivendi”). Thereafter, when the nature of Vivendi’s liquidity situation came to light, the price of those securities dropped dramatically. GAMCO subsequently brought a securities fraud action against Vivendi under § 10(b) of the Securities Exchange Act of 1934 (“Exchange Act”), 15 U.S.C. § 78j(b), as well as the Securities Exchange Commission’s (“SEC”) Rule 10b–5 (“Rule 10b–5”) promulgated thereunder, 17 C.F.R. § 240.10b–5. After a bench trial solely on the question whether Vivendi successfully rebutted the fraud on the market presumption of reliance invoked by GAMCO to satisfy the reliance element of its § 10(b) claim, see Basic Inc. v. Levinson, 485 U.S. 224, 247 (1988), the District Court for the Southern District of New York (Scheindlin, J.) entered judgment for Vivendi. See GAMCO Investors, Inc. v. Vivendi, S.A., 927 F. Supp. 2d 88, 91 (S.D.N.Y. 2013) (“GAMCO”). GAMCO now appeals.1 1 Vivendi also cross appeals from the district court’s determination that Vivendi was barred by collateral estoppel from contesting the other elements of GAMCO’s § 10(b) claim. Because we affirm the district court’s decision in favor of Vivendi, we need not reach this claim. 3 DISCUSSION “On appeal from a bench trial, the district court s findings of fact are reviewed for clear error and its conclusions of law are reviewed de novo.” Beck Chevrolet Co., Inc. v. Gen. Motors LLC, 787 F.3d 663, 672 (2d Cir. 2015) (quoting Mobil Shipping & Transp. Co. v. Wonsild Liquid Carriers Ltd., 190 F.3d 64, 67 (2d Cir. 1999)). To succeed on a § 10(b) and Rule 10b–5 claim, a plaintiff must prove “(1) a material misrepresentation or omission by the defendant; (2) scienter; (3) a connection between the misrepresentation or omission and the purchase or sale of a security; (4) reliance upon the misrepresentation or omission; (5) economic loss; and (6) loss causation.” Halliburton Co. v. Erica P. John Fund, Inc., 134 S. Ct. 2398, 2407 (2014) (quoting Amgen Inc. v. Conn. Retirement Plans and Tr. Funds, 133 S. Ct. 1184, 1192 (2013)). The traditional way a plaintiff demonstrates reliance is directly, i.e. “by showing that he was aware of a company’s statement and engaged in a relevant transaction . . . based on that specific misrepresentation.” Id. (quoting Amgen, 133 S. Ct. at 1192)). However, a plaintiff may, in some circumstances, invoke a “rebuttable presumption of reliance.” Id. at 2408. This presumption rests on the “‘fraud on the market’ theory” which states “that ‘the 4 market price of shares traded on well developed markets reflects all publicly available information, and, hence, any material misrepresentations.’” Id. (quoting Basic, 485 U.S. at 246). Because “the typical ‘investor who buys or sells stock at the price set by the market does so in reliance on the integrity of that price’—the belief that it reflects all public, material information— . . . his ‘reliance on any public material misrepresentations . . . may be presumed for purposes of a Rule 10b–5 action.’” Id. (quoting Basic, 485 U.S. at 247). But the presumption, as its name suggests, is rebuttable.2 “Any showing that severs the link between the alleged misrepresentation and either the price received (or paid) by the plaintiff, or his decision to trade at a fair market price, will be sufficient to rebut the presumption . . . .” Id. (quoting Basic, 485 U.S. at 248) (alteration omitted). As an initial matter, GAMCO argues that the District Court “erred in concluding that . . . because Vivendi demonstrated that Plaintiffs are value investors[,] Vivendi . . . proved [GAMCO] did not rely on the integrity of the To demonstrate successfully that the presumption should apply in a given case, “a plaintiff must [show] . . . (1) that the alleged misrepresentations were publicly known, (2) that they were material, (3) that the stock traded in an efficient market, and (4) that the plaintiff traded the stock between the time the misrepresentations were made and when the truth was revealed.” Id. It was not contested at the trial that the presumption itself applied, only whether it had been rebutted. 2 5 market in purchasing Vivendi [securities].” GAMCO Br. at 41. According to GAMCO, though it, like most value investors, does not believe that the market price necessarily equals, at any given time, the efficient value of a security (and thus makes its money by identifying under or over valued securities), Halliburton forecloses the conclusion that this fact alone suffices to rebut the presumption.3 Vivendi, for its part, suggests that such a showing indeed suffices to rebut the presumption. See Vivendi Br. at 33; cf. Teamsters Loc. 445 Freight Div. Pension Fund v. Bombardier Inc., 546 F.3d 196, 199 n.4 (2d Cir. 2008) (“The . . . fraud on the market theory involves [the] rebuttable presumption[] . . . ‘that . . . investors rely on the market price of securities as an accurate measure of their intrinsic value.’” (quoting Hevesi v. Citigroup Inc., 366 F.3d 70, 77 (2d Cir. 2004)). It is true that the district court reasonably found facts that demonstrate that GAMCO’s purchasing decisions relied in large part on an independent valuation See Halliburton, 134 S. Ct. at 2410 (“Debates about the precise degree to which stock prices accurately reflect public information are . . . largely beside the point.”); id. at 2411 (noting that “there is no reason to suppose that . . . the value investor[] is as indifferent to the integrity of market prices as Halliburton suggests”); but see id. (noting that “Basic never denied the existence of [value] investors,” as “Basic concluded only that ‘it is reasonable to presume that most investors . . . will rely on the security’s market price as an unbiased assessment of the security’s value in light of all public information.’” (quoting Amgen, 133 S. Ct. at 1192)). 3 6 of the worth of Vivendi’s securities that was separate and distinct from the market price. The district court found that GAMCO would begin by calculating a “Private Market Value[]” (“PMV”) for a given security — a value that approximated “the price that an informed industrialist would be willing to pay for [the company], if each of its segments were valued independently in a private market sale.” GAMCO, 927 F. Supp. 2d at 94–95. GAMCO would then compare this PMV to the public market price. If there was a sufficiently large “spread between [the] PMV and the market price,” and if there was evidence of the existence of a “catalyst,” id. at 94–96, or “some dynamic that [GAMCO thought was] going to help surface value over time,” J.A. 359, then GAMCO would generally elect to purchase the security in question. Employees at GAMCO often referred to its PMV as the “intrinsic value” of the stock, see, e.g., J.A. 338, 508, and frequently referred to the market price, in contrast, as reflecting the whims of the sometimes “irrational . . . Mr. Market,” J.A. 358. Though the district court indeed found that GAMCO does not necessarily consider the market price to be an efficient reflection of the objective value of a security at any given time, however, it explicitly disclaimed any reading of its opinion as suggesting that this fact was sufficient, on its own, to rebut the 7 presumption. See GAMCO, 927 F. Supp. 2d at 102 (“[O]ne can imagine a case where an investor . . . used the market price of a security merely as a comparator with a private method of valuation, but in which the fraud on the market presumption could not fairly be rebutted, because, but for the material misstatements, that investor would not have transacted in the securities at issue.”). We thus decline to explicate the contours of Halliburton here, further theorize on the presumption, or otherwise address the relevance of the typical value investing model to a rebuttal showing. The district court’s holding was, instead, based on a much narrower theory: that, given the facts in the record, Vivendi proved that GAMCO would have purchased Vivendi securities even had it known of Vivendi’s alleged fraud. See Halliburton, 134 S. Ct. at 2408 (observing that a defendant may rebut the presumption by demonstrating, inter alia, “that a plaintiff would have bought or sold the stock even had he been aware that the stock’s price was tainted by fraud”); Kline v. Wolf, 702 F.2d 400, 403 (2d Cir. 1983) (noting that the presumption may be rebutted through a showing “that plaintiffs did not significantly rely on the integrity of the market or that, even if they had known of the alleged misrepresentation, they would still have purchased the stock”).4 GAMCO argues that this factual conclusion was error, and asserts that 4 The district court found facts sufficient to establish that that it was more likely 8 “[i]t was clearly erroneous for the District Court to find that a sophisticated financial manager would purchase securities in advance of undisclosed adverse information which would have the predic[t]able impact of causing market prices than not that GAMCO would have purchased the relevant securities even if it had known that Vivendi’s statements were misrepresenting its liquidity. See GAMCO, 927 F. Supp. 2d at 97 (finding, inter alia, that “the liquidity crisis at Vivendi was irrelevant to Plaintiffs’ investment decisions”). Nevertheless, GAMCO takes issue with the district court’s approach to determining whether GAMCO would have purchased the security even if it had known there was a fraud. Specifically, GAMCO questions the district court’s reliance on the argument that GAMCO would have considered Vivendi’s securities a more attractive investment had Vivendi fully divulged its liquidity problems to the entire market prior to GAMCO’s purchase. See id. at 102 (where the district court observes that “in this counterfactual scenario [where Vivendi did not commit any fraud] . . . the price of Vivendi ADS’s would have been lower” as the market price would have fallen with revelation of the truth, and thus “Plaintiffs would have seen Vivendi as a more attractive investment”). GAMCO does not necessarily question the premise that, if Vivendi had fully disclosed its liquidity problems to the market, then the market would likely have internalized the value of those problems, the stock’s price would have fallen, and the spread between the PMV and the market price would thus have been larger. Nor does GAMCO deny that it may have purchased the security at this lower price had there been no fraud at all. Rather, GAMCO contests the relevance of the district court’s inquiry to what GAMCO believes is the central issue, namely whether GAMCO would have purchased shares “at the same price[]” had it been aware that many of Vivendi’s statements were fraudulent. Id. at 103 (emphasis added) (internal quotation marks omitted); see also Halliburton, 134 S. Ct. at 2408 (noting that a defendant may rebut the presumption by showing “that a plaintiff would have bought or sold the stock even had he been aware that the stock’s price was tainted by fraud”). Though we observe that GAMCO’s framing of the appropriate inquiry has much to recommend it, we need not conclusively endorse it here. Even adopting, arguendo, GAMCO’s preferred framing, see GAMCO Br. at 27 n.11 (acknowledging that the presumption may be rebutted by showing that a “plaintiff would have purchased the stock at the same price even if she had known of the fraud”), the district court did not clearly err in concluding, as detailed herein, that revelation of Vivendi’s liquidity problems would not have changed GAMCO’s purchasing decisions. 9 to decline.” GAMCO Reply Br. at 8. Our review of the record demonstrates that it was not “clearly erroneous” for the district court to so conclude as to GAMCO itself. Under that deferential standard of review, we must accept the district court’s findings of fact “unless we have a definite and firm conviction that a mistake has been committed.” Souratgar v. Lee, 720 F.3d 96, 103 (2d Cir. 2013) (internal quotation marks omitted). We therefore need not, and do not, decide whether we would have reached the same factual conclusions as the district court. First, though GAMCO repeatedly asserts on appeal that it is all but unthinkable that a sophisticated investor would ever buy a security if he was aware its market price might be tainted by fraud, the idea was not so ridiculous to the Chief Investment Officer of Gabelli & Co. Asked whether he would “buy a stock in a company . . . [where he] knew . . . the price was inflated as a result of a management fraud,” Mario Gabelli testified that such a purchase was “possible, but very unlikely,” and clarified that he would “assign a 90 10 probability” to such a scenario. J.A. 530–31. He went on to qualify this statement, noting “[t]here would have to be a lot of other circumstances.” Id. at 531. The district court largely dismissed Gabelli’s testimony as “motivated 10 hairsplitting by an interested witness.” GAMCO, 927 F. Supp. 2d at 94. It is thus highly significant that a prominent witness the district court believed was exaggerating the facts in favor of the plaintiffs admitted on the stand that, in the abstract, there was a ten percent possibility that, in a given case, he would buy stock inflated by fraud. Such testimony, at minimum, established — in contrast to GAMCO’s repeated suggestions — that GAMCO had no ironclad policy against purchasing a stock it believed might be inflated by fraud, provided that other circumstances made the deal one worth pursuing. To that end, the record in this case includes sufficient evidence supporting the district court’s finding that such circumstances were present here: in other words that, even if aware of Vivendi’s liquidity problems and its concealment of those problems, GAMCO would still have believed the deal it made was a good one. As already noted, the district court reasonably found that the core of GAMCO’s investment decisions could be reduced to two questions: first, was there a material spread between market price and PMV, and second, was there reason to believe a catalyst would cause the market to recognize this discrepancy in the next two to five years. Evidence in the record also supported the district court’s conclusion that, had GAMCO known of the liquidity problems at Vivendi, 11 such knowledge would not have altered the answers to either of these questions. As to the spread, Andrew Rittenberry, the analyst primarily responsible for calculating Vivendi’s PMV in the relevant period, testified that, once he learned of the liquidity crisis at Vivendi (as the fraud came to light), the knowledge of the liquidity problems had no “impact [on his] PMV calculation,” as he believed those problems constituted “a short term issue . . . . [s]omething that could be solved within a year or so.” J.A. 401 03. Rittenberry also testified that PMV was generally calculated to anticipate the value of the company over the long haul. This testimony further supports the district court’s factual finding that GAMCO would not have believed that a short term problem, especially one that it did not perceive to be as significant as the market ultimately came to believe it was, would affect its calculation of the spread. See J.A. 368 (noting that, in calculating the PMV, Rittenberry “projected the business performance of [the] differen[t] pieces of the business over a five year period”); J.A. 383 (“[T]he goal in buying a stock is to have it appreciate 50 percent over a two year period”); J.A. 527 (testimony of Mario Gabelli: “So the way we think is like a McKinsey thinking out over the next five or ten years”). Thus, the record supports the district court’s conclusion that, if GAMCO had known of the liquidity problems and 12 their concealment, GAMCO would still have believed Vivendi’s PMV to be “materially higher” than the public market price. J.A. 363. Second, it was also not clearly erroneous for the district court to conclude that knowledge of Vivendi’s liquidity problems would not have changed GAMCO’s belief that a catalyst was likely — i.e., its belief that the market price would rise towards the PMV, if not immediately, then over the course of the next several years. Such evidence came in various forms. First, GAMCO continued to buy Vivendi securities as the full extent of Vivendi’s alleged fraud came to light. GAMCO argues that this information is hardly dispositive of whether GAMCO would have purchased prior to the revelation of the fraud at the same price at which it purchased (given that the revelation caused the market price to fall as fraud induced inflation dissipated). That may indeed be so, but the evidence is specifically relevant to the narrower question whether knowledge of the liquidity problems would have affected GAMCO’s belief in the existence of a catalyst. That GAMCO continued to buy reasonably suggested to the district court that GAMCO still believed a catalyst to be on the horizon (and thus demonstrated that the liquidity problems had not altered, and would not have altered, this conclusion). Further, Gabelli testified that a “change of 13 management” was one of the most frequent catalysts on which GAMCO would rely in purchasing a stock. J.A. 482. This testimony further reinforced the district court’s factual findings: had GAMCO known of the liquidity problems at Vivendi (notwithstanding whether they had been fraudulently concealed), the potential for a change in management would have seemed even more likely. In short, evidence in the record supported the district court in concluding that, had GAMCO known of Vivendi’s liquidity problems, GAMCO would still have believed, first, that Vivendi’s securities were substantially undervalued by the market and second, that an event was likely to happen in the next few years that would awaken the market to that fact. Further, the record provided sufficient evidence to find that GAMCO would have believed that, even if the liquidity crisis in fact came to light, it would resolve within a short period of time — and thus, that even if the market price might go down in the short run, the potential for long term profit still remained. In other words, whereas one can imagine situations in the abstract where a sophisticated investor, apprised of a fraud, would necessarily conclude that a security was no longer a logical purchase, the district court did not clearly err in concluding, on this record, that in this case, and with regard to this particular fraud, GAMCO would still have 14 viewed Vivendi’s securities as a profitable investment — even if it might have been concerned about the hidden liquidity risks. GAMCO counters this extensive evidence by citing, almost exclusively, to testimony of Andrew Rittenberry, which GAMCO employs to argue that the district court’s factual findings were clearly erroneous. Asked whether, assuming the information “did not affect [his] PMV calculation, a liquidity crisis would similarly not affect [his] buy, sell, or hold recommendation,” Rittenberry replied, “[n]o, not necessarily.” J.A. 401. “The big driver,” he continued, “[is] PMV. But the discount to the PMV between the market price and the PMV is a key factor. If there is a liquidity crisis or some type of event like that, you would expect the discount to widen. I mean you would expect the stock to go down.” Id. When further asked whether he “would expect the public market price to go down,” Rittenberry answered, “[y]es. You wouldn’t want to buy it in front of that — if you thought it was going to go down, you would wait to buy it or buy more or recommend to buy more at that point.” Id. Pointing to Rittenberry’s testimony, GAMCO argues that even if it would not have believed the liquidity problems to affect Vivendi’s PMV, and even if it would still have believed it likely the investment would be profitable over the 15 long run, the record renders clearly erroneous the district court’s finding that it would have purchased Vivendi’s securities had it known of the fraud. This is because, GAMCO argues, even if it indeed still thought Vivendi securities were materially undervalued notwithstanding the crisis, it would have waited until the liquidity crisis came to light and then bought the stock. Such a delayed purchase not only would have mitigated the risk associated with the undisclosed liquidity problems (as fraud always creates some risk that the underlying calculations are erroneous), but also would have potentially netted GAMCO a more profitable investment. It was not clearly erroneous, however, for the district court to find that GAMCO would indeed have purchased Vivendi’s securities at the price quoted, rather than simply await the potential of a public liquidity crisis. First, both Rittenberry’s testimony and GAMCO’s argument rest on the assumption that Rittenberry (and GAMCO) would have known that the stock price would indeed go down upon revelation of the liquidity crisis. As already noted, however, the district court reasonably found facts demonstrating that Rittenberry, or GAMCO, apprised of the hidden liquidity risks, would not have been so certain. At the time GAMCO had to make its purchasing decision, even if it had known of the 16 fraud, it would not have been able to see the future. And the record evidence as to how GAMCO would have approached determining how likely it was that the hidden risks would be exposed in a way that would materially affect the market price does not support GAMCO’s position. Thus, Rittenberry testified that, when he did find out about the liquidity problems, he believed that they constituted a very short term concern with minimal impact on Vivendi’s value. It was therefore not clearly erroneous for the district court to infer that GAMCO, privy to this information prior to its relevant purchases, would have believed it possible that the crisis would pass without incident or public revelation, or at minimum be uncertain as to just how the market would value the crisis were it to come to light. Given that the record makes clear that GAMCO would, in at least some circumstances, have purchased a security even believing it was tainted by fraud, it is highly relevant that the record also provides sufficient evidence to find that GAMCO would have believed this particular fraud unlikely to become material. Further, against this uncertainty of a future, hypothetical better deal, the record provided evidence that GAMCO had reasons to believe the deal — which, as already noted, would have appeared to be a good one — could get worse if 17 GAMCO waited. As Rittenberry testified, the premise of GAMCO’s model was that it found a stock was “trad[ing] at a discount to what [GAMCO believed] it[] [was] worth temporarily,” and that “that discount [would] . . . narrow over time.” J.A. 357 (emphasis added). As a general matter, then, to wait, rather than buy the stock, always risked the possibility that the prophesied catalyst would come early, or that the market would recognize that the value of the stock was depreciated. Further, Gabelli testified that he believed the market, in the aggregate, to be “down sharply” by the beginning of 2001, J.A. 522, and in GAMCO’s December 2000 Annual Report to investors, it “forecast . . . an economic (and earnings) recovery” in “the second half of 2001” which would “help[] . . . renew investor confidence and revive the market,” J.A. 591. To wait for the hypothetical revelation of the liquidity problems, then, would risk not only that the market, specifically, would realize that Vivendi’s stock was undervalued, but also that the market, generally, would rebound and the market price would rise accordingly, sacrificing the spread. It was not clearly erroneous for the district court to conclude that GAMCO, aware of the fraud, would not have waited to buy the securities. In other words, the premise of Rittenberry’s testimony, and GAMCO’s 18 argument, assumes what it seeks to prove: that GAMCO would have known that the liquidity problems would manifest in the way that, in hindsight, they manifested, and would thus obviously have waited to buy if it purchased at all. But the district court’s rejection of that conclusion was not clearly erroneous. Further, GAMCO exaggerates the probative value of Rittenberry’s testimony in another way as well. Even on its own terms, Rittenberry’s testimony does not clearly suggest that GAMCO would not have bought if it had known of the hidden liquidity risks. Asked whether knowledge of such a hidden crisis would “not affect” his buy or sell recommendation, Rittenberry, not unlike Gabelli, answered equivocally: “no, not necessarily.” J.A. 401 (emphasis added). And, when describing what his recommendation would be, he did not say that he would recommend that GAMCO not buy; instead, he testified that “if [he] thought [the market price] was going to go down, [he] would wait to buy it or buy more or recommend to buy more at that point.” Id. (emphasis added). In other words, the district court did not have to read this testimony as GAMCO reads it. Instead, it could reasonably have read Rittenberry’s testimony as suggesting that GAMCO would buy and then wait and buy more if indeed the liquidity problems surfaced. Such an approach would potentially allow GAMCO to ensure it 19 captured a good deal when it saw one while hedging against the possibility of an uncertain future. It may seem unlikely, in the abstract, that an investor, aware of fraud, would opt to purchase a given security. Nevertheless, after a trial, we do not review a district court’s factual findings for whether they seem, in the abstract, correct — we review the record for sufficient evidence in support of those judgments. And here, the record at the trial simply does not establish that it was clearly erroneous for the district court to find that GAMCO, had it known of the liquidity problems at Vivendi, would have made the choice to buy the same securities it purchased.5 We also conclude that GAMCO’s challenge to the district court’s grant of summary judgment is unreviewable. GAMCO concedes that we do not have jurisdiction to hear an appeal from a denial of summary judgment after a trial, unless the “district court’s error was purely one of law.” Stampf v. Long Island R.R. Co., 761 F.3d 192, 201 n.2 (2d Cir. 2014); see GAMCO Reply Br. at 31–32. GAMCO further acknowledges that one ground on which the district court denied summary judgment was its finding that “GAMCO could possibly have uncovered non public corrective information which would rebut the presumption of reliance.” GAMCO Br. at 9; see also J.A. 92. GAMCO concedes that, as a legal matter, this scenario does indeed rebut the presumption of reliance. See GAMCO Br. at 27 n.11 (noting that one of the “recognized ways to rebut the . . . presumption [is when] plaintiff possessed non public information concerning the fraudulent misstatements.”). GAMCO argues that evidence in the record created no material question of fact, at summary judgment, whether GAMCO had such information. This argument misses the point. To the degree that GAMCO agrees with the legal standard the district court applied (non public corrective information can rebut the presumption of reliance) and argues only that the district court erred in finding a material question of fact pursuant to that 5 20 CONCLUSION Accordingly, and finding no merit in the Plaintiffs’ remaining arguments, we AFFIRM the judgment of the District Court. legal standard, at least one ground for the district court’s denial of summary judgment was not a pure error of law, but one of fact. Thus, the denial is not reviewable by this Court. See Stampf, 761 F.3d at 201 n.2. 21