IN RE SCIENTIFIC ATLANTA, INC. SECURITIES LITIGATION (N.D.Ga. 11-18-2010)

rill Lynch Co., 396 F.3d 161, 172-77 (2d Cir. 2005) (requiring that the disclosure reveal that the defendant's "misstatement or omission concealed something from the market that, when disclosed, negatively affected the value of the security"). In Lentell, the Second Circuit identified the event which revealed the previously-concealed information as the "materialization of the risk" concealed by earlier misrepresentations. See Lentell, 396 F.3d at 173. While such an event may take the form of a traditional corrective disclosure expressly admitting to an earlier fraud, it need not necessarily do so in order to satisfy loss causation requirements. See In re Vivendi Universal, S.A. Sec. Litig., 605 F. Supp. 2d 586, 598 (S.D.N.Y. 2009). External events, such as a ratings downgrade that reveals the risk of deteriorating liquidity, may also suffice. Id. The form of the event is immaterial; the relevant inquiry is whether the later event, by some mechanism, "disclose[s] part of the truth that was previously concealed by the fraud." Id. (finding genuine issue of material fact existed as to whether credit ratings downgrades and quick asset sales within eleven day period materialized the concealed risk of defendant's severe liquidity crisis).

In the face of such extensive case law undercutting their proposed rule, Defendants have cited to two authorities providing weak support for their position while failing to contradict the rule utilized in other jurisdictions. Defendants first cite In re Retek Inc. Securities Litigation, where the plaintiffs failed to establish that their loss resulted from misleading public statements about the defendant-company's financial condition. In re Retek Inc. Sec. Litig., 621 F. Supp. 2d 690, 708 (D. Minn. 2009). The Retek plaintiffs argued that a press release discussing the current financial condition of the defendant-company disclosed the fact that the defendants had engaged in improper revenue recognition practices in order to conceal the company's true financial outlook. Id. at 692. The court rejected this argument, finding that the purported corrective disclosure failed to provide any new information on the defendant-company's accounting practices, and ultimately granted summary judgment on the issue of loss causation. Id. at 708. The court did not suggest that the plaintiff's case failed for want of an express admission of fraud; indeed, the court noted that "the relevant truth [concealed by the defendant's misrepresentations] may be disclosed directly or indirectly." Id. at 699. Rather, what proved fatal to the plaintiff's case was the absence of evidence showing that the market recognized a link between the disclosure and the underlying fraud. See id. at 702-03. Therefore, Retek does not support Defendants' proposed rule.

Defendants' reliance on In re Omnicom Group, Inc. Securities Litigation is similarly misplaced. There, the plaintiffs argued that the defendant-company's spin-off of two parts of its business was a sham and that the company's subsequent announcements that it intended to buy back those parts of its business constituted disclosure of the alleged fraud. See In re Omnicom Grp., Inc. Sec. Litig., 541 F. Supp. 2d 546, 552-53 (S.D.N.Y. 2008). The court found that this announcement did not correct any earlier misrepresentation, as the mere fact that the defendant-company intended to buy back these businesses did not suggest any fraud with respect to its initial decision to spin them off. See id. at 553. Here again, the critical failing was the absence of any new information in the "corrective" disclosure rather than the absence of an express repudiation of earlier misstatements. While the Omnicom court found that the disclosure at issue was not corrective because it did not "reveal the falsity of the alleged misstatements," no language in the opinion suggests that only an express admission of fraud would have constituted an adequate disclosure. See id. at 552-53. As inRetek, the court noted that "a disclosure need not reflect every detail of an alleged fraud," although "it must reveal some aspect of it." Id. at 551. As such, neither of the cited authorities support Defendants' rigorous requirements in a corrective disclosure and are in fact consistent with the more permissive rule advocated by Plaintiffs.

Furthermore, as the Fifth Circuit noted in Flowserve, Defendants' proposed rule would allow wrongdoers to avoid liability by merely refusing to admit the falsity of prior misstatements. See Flowserve, 572 F.3d at 230. As one court has discussed:

A company that has, for example, booked revenue from non-existent contracts could simply issue some damaging announcement that appears on its face unrelated to any fraudulent scheme, e.g., a significant earnings warning citing order weakness, wait for its share price to plummet, and then disclose the wrongdoing once the damage has been done. The plaintiff would be unable to tie its loss, i.e., the share price decline, to the fraud, rather than to the apparently unrelated announcement.

Motorola 505 F. Supp. 2d at 544. Indeed, such a rule would bar any securities fraud action lacking an express admission of fraud on the part of repentant defendants.

Therefore, having reviewed the case law and the arguments of the parties, the court adopts the rule articulated in Flowserve that a corrective disclosure need not specifically address the falsity of prior statements. So long as a later disclosure reveals new information which an earlier omission fraudulently concealed, it is immaterial whether the disclosure contains an express admission of prior fraudulent conduct. See also In re Teco Energy, Inc. Sec. Litig., No. 8:04-CV-1948-T-27EAJ, 2006 U.S. Dist. LEXIS 73833, at *23-24 (M.D. Fla. Oct. 10, 2006).

In the present case, Plaintiffs cite three disclosures that revealed the weakness in demand for S-A's products that were concealed by earlier public statements. On July 19, 2001, S-A issued a press release disclosing that, although earnings and other financial metrics had exceeded its expectations, its total bookings for the fourth fiscal quarter had declined over the previous year's fourth quarter. (SMF ¶ 33a.) In its press release, S-A attributed the decline in bookings to "the uncertain economic climate and reduced digital marketing efforts by cable operators during the slower summer vacation period, in addition to customer inventory levels and the slower than expected deployment of interactive applications." (Id. (emphasis added).) On the same date, during the conference call discussing the press release, Defendant McDonald told investors that S-A's customers "probably moved early in the year to get the product" and, as the fourth quarter progressed, likely began absorbing inventory that had accumulated during the previous quarter. (Id. ¶ 33b.)

The third disclosure occurred on August 16, 2001. On that date, S-A filed its 2001 Form 10-K, disclosing in a press release that it anticipated adverse effects on its results in fiscal 2002 even though its customers had recently reaffirmed or increased their estimates of new digital subscribers to be added through the end of calendar year 2001. (SOF ¶ 231.) The press release attributed a decline in demand to customers' accumulation of inventories during calendar years 2000 and 2001, as well as to the declining economy.

Plaintiffs assert that these disclosures revealed to the market not only that demand was declining, but that such declines were attributable to excessive customer inventories resulting from S-A's channel stuffing practices. (Pl. Opp. at 94.) The new information regarding accumulated customer inventories allegedly revealed that these sales practices had concealed weakening demand for S-A's products in earlier public statements. (Id.)

In support of this contention, Plaintiffs cite several reports in the days following the July 19 and August 16 statements where analysts discussed the detrimental impact of excessive customer inventories on future demand for S-A's products.

In a report issued July 20, 2001, Gerard Klauer Mattison Co. stated, "MSOs lowered digital STB inventory; further field inventory correction expected. . . . Field inventory is now estimated at 1.3 million units, or 14 weeks of installations. We believe the desired range is 4-6 weeks. Therefore, we anticipate shipments will fall below install rates for the foreseeable future. . . ." (SMF ¶ 35d.)

A second analyst report from July 20 expressed similar sentiments: "What we didn't anticipate was how quickly and how sharply the growth would fall off for SFA. We attribute this sharp fall off primarily to over-ordering of settops by MSOs in the December and March quarters. As a result, . . . we expect a significant decline in [year-over-year] revenue and earnings." (SMF ¶ 35e.)

The corrective nature of these disclosures is most clearly evident in an August 16 report from analyst UBS Warburg (US):

Yesterday's 10-K filing by Scientific-Atlanta (S-A) calls into question two things. First, channel inventories, and second, end user demand. . . . So with inventories coming down 20 days in the second half of 2001, what could be wrong with this analysis? The simple answer is that the absolute level of inventories may still be too high to begin with. . . . There appears to be no reasonable case yet for an annualized slowdown of consumer demand, whether due to the recession or other secular saturation reasons. . . . With a healthy demand picture, in our view, any further weakness in set-top box shipments in the second half of 2001 must come from a greater-than-anticipated inventory correction.

(SMF ¶ 35g (emphasis added).)

These reports demonstrate that the July and August statements revealed the excessive inventories associated with Defendants' alleged undisclosed channel stuffing activities. Analysts following S-A clearly reacted with surprise to the new information regarding customer inventory levels, and clearly understood the import of S-A's suggestion that such excess inventories accounted, at least in part, for the decline in demand. Regardless of whether analysts understood that channel stuffing was the cause of such excessive inventories, these disclosures revealed at a minimum that Defendants' statements from January and April 2001 obscured the degree to which the apparent demand was bolstered by S-A's channel stuffing practices.

However, the Court agrees with Defendants that these disclosures fail to establish a causal link between Defendants' accounting violations and the drop in S-A's stock price. As inRetek, Plaintiffs' cited disclosures provided the market with no new information about S-A's accounting practices. Thus, Plaintiffs cannot demonstrate that the removal of any inflationary component in the stock price resulted from a disclosure of these practices. Accordingly, because Plaintiffs have failed to put forward evidence of loss causation as to Defendants' GAAP violations, the Court grants summary judgment to the extent that Plaintiffs' claims rely on such violations.

B. Disaggregation of Plaintiffs' Losses from Non-Fraud Related Events

The Court is ultimately left with the question of whether Plaintiffs have presented a genuine issue of fact as to whether a causal connection exists between the disclosures revealing the effects of Defendants' alleged channel stuffing activities and the subsequent drop in the price of S-A's stock. Defendants have argued that Plaintiffs, in order to prevail at summary judgment, must offer evidence disaggregating the detrimental effect of the alleged fraud from that of other, non-fraud-related events upon Plaintiffs' investment. Plaintiffs argue that they need not exclude other possible causes of the stock's decline in order to survive summary judgment.

The precise issue of what a plaintiff must show in order to survive summary judgment on the loss causation issue has not been addressed in this circuit. The leading authority is the Eleventh Circuit's decision in Robbins v. Koger Properties, Inc., reviewing the lower court's denial of a judgment as a matter of law on loss causation grounds. Robbins v. Koger Props., Inc., 116 F.3d 1441, 1446 (11th Cir. 1997). In addressing the loss causation issue, the Eleventh Circuit held that, due to the "many different, complex, and often unknowable factors" affecting the value of a stock, a plaintiff need not show that the defendant's fraud was "the sole and exclusive cause of the injury [the plaintiff] has suffered," but rather that the fraudulent conduct was a substantial (i.e., contributing) cause of the stock's decline in value. Id. at 1447.

In Robbins, the defendant-auditors allegedly misrepresented the defendant-company's cash flow, allowing for the payment of higher dividends to investors. The plaintiffs alleged that this misrepresentation of the company's cash flow and ability to pay dividends at a given rate created an artificially high price at which plaintiffs purchased the company's stock. Id. at 1445. The plaintiffs' stock declined in value after the defendant company announced a dividend cut; however, in its public statements announcing the dividend cut, the defendant denied any problems in cash flow, and therefore did not correct the earlier alleged misrepresentation. Id.

The court ultimately held that plaintiff failed to offer evidence that connected the defendant's misrepresentations to any decline during the class period and thus failed to prove that the misrepresentations were a substantial cause of the stock's decline. Id. at 1448. In so holding, the court focused on the plaintiffs' failure to show that any price inflation resulting from misleading statements regarding company cash flow levels had been removed from the market price during the class period. Id.

Therefore, in Robbins, the Eleventh Circuit's decision relied more on the absence of any correction of the fraud than on insufficient proof that the fraud was a substantial cause of the plaintiff's loss. The plaintiffs failed to demonstrate that defendant's misrepresentation played any role in the loss, let alone a substantial one, when no information revealing the misrepresentation came to light during the class period. See id. at 1448-49 (noting that the minutes of the defendant's board meeting attributed the dividend cut resulting in the stock's decline to factors other than the cash flow, which was the subject of the misrepresentation). As such, this decision, which implicitly relies on the absence of any corrective disclosure, provides little guidance in applying its rule to the current case, where Defendants argue that Plaintiffs' evidence fails to account for competing causes.

Only one district court in the Eleventh Circuit has citedRobbins in the context of summary judgment. In In re MIVA, Inc., the Middle District of Florida found that, in the absence of evidence attributing any portion of the inflated stock price to the defendant's misrepresentation, the plaintiff failed to show that these misrepresentations were a substantial or significant contributing cause of the plaintiff's loss. See In re MIVA, Inc., No. 2:05-cv-201-FtM-29DNF, 2009 WL 3821146, at *12 (M.D. Fla. Nov. 16, 2009). Here again, the court's decision hinged on the lack of evidence supporting the fraud-related cause of the loss rather than plaintiff's failure to discount non-fraud-related causes and is therefore of little assistance in the present case.

Other jurisdictions have split on the question of whether, in addition to providing affirmative evidence that the defendant's fraudulent conduct caused the plaintiff's loss, plaintiffs must also offer evidence at summary judgment which excludes competing, non-fraudulent causes. The Seventh Circuit has adopted the most plaintiff-friendly rule on this issue. In Caremark, Inc. v. Coram Healthcare Corp., the Seventh Circuit, in finding that the plaintiffs' complaint adequately pleaded loss causation, also held that at a future summary judgment stage, the defendant could win on loss causation only "by establishing that the decline in the value of the security is attributable in total to some other factor." Caremark, Inc. v. Coram Healthcare Corp., 113 F.3d 645, 649-50 (7th Cir. 1997) (emphasis added).

In contrast, other circuits have looked to the plaintiff at the summary judgment phase to exclude loss causation factors unrelated to the fraud. In In re Williams, the Tenth Circuit held that the plaintiffs failed to establish loss causation where the corrective disclosures contained both information revealing the alleged fraud and information related to non-fraud-related factors that would have affected the investment's value. In re Williams Sec. Litig.-WCG Subclass, 558 F.3d 1130, 1143 (10th Cir. 2009). The court found unpersuasive argument that the presence of multiple causes, some of which included fraud, created a genuine issue of fact precluding summary judgment. The court instead found that a jury's conclusion as to whether the loss resulted from the fraud as opposed to other factors "would be no less speculative and unreliable if reached by jurors than when reached by [plaintiffs' expert]." Id.

In so holding, the Tenth Circuit relied on several decisions from the Second Circuit. While the Second Circuit has not addressed the role of confounding news at the summary judgment stage, the court upheld a motion to dismiss a complaint where the plaintiffs failed to allege that a corrective disclosure was the proximate cause of the plaintiffs' loss, as opposed to other potentially damaging statements released the same day "that were much more consequential and numerous." Lattanzio v. Deloitte Touche LLP, 476 F.3d 147, 158 (2d Cir. 2007) (noting also that the complaint was deficient in failing to allege facts allowing a fact-finder to ascribe some rough portion of the loss to Defendants).

Moreover, the Southern District of New York has repeatedly required plaintiffs to disaggregate non-fraud factors when deciding loss causation issues at the summary judgment stage. The court found that plaintiffs had met this burden in In re Vivendi Universal, S.A. Securities Litigation, where the plaintiffs' expert conducted a regression analysis factoring out the effects of any other company-specific information released on days when the plaintiffs' investment experienced declines. In re Vivendi Universal, S.A. Sec. Litig., 605 F. Supp. 2d 586, 604-05 (S.D.N.Y. 2009). In so holding, the court found it dispositive that the defendants failed to point to an obvious competing cause for the plaintiffs' loss on the days identified. Id. at 605. The court also emphasized that a plaintiff need only disaggregate some rough percentage of the declines resulting from non-fraud-related events rather than the precise amount of loss attributable to the defendant's fraudulent conduct. Id. at 600.

The same court has also held that plaintiffs' loss causation analysis must discount negative characterizations of previously known information. See In re Omnicom Grp., Inc. Sec. Litig., 541 F. Supp. 2d 546 (S.D.N.Y. 2008). In Omnicom, the defendant company's stock dropped following the resignation of one of its directors and the onset of rumors of a negative Wall Street Journal article. See id. at 548. When the article was released the following week, it reported that the defendant-company avoided losses by shifting failing investments into a holding company and that the director's resignation occurred amid questions regarding the handling of this transaction. In addition, two accounting professors quoted in the article explicitly questioned the propriety of the transaction. This resulted in further declines in the defendant's stock. Id. at 548-50.

The court emphasized that when a corrective disclosure accompanies several pieces of new information released to the market simultaneously, plaintiffs must disaggregate the market's negative reaction to the corrective disclosure in particular from other confounding factors affecting stock price. See id. at 553-54. In order to do so, the Omnicom plaintiffs relied on an event study by Dr. Scott Hakala, Plaintiffs' loss causation expert in the present case. Dr. Hakala attempted to discount the article's effect by noting that, "[w]hile the article also reasserted concerns regarding Omnicom's organic growth, cash flow, put obligations and earn-outs, those issues had previously been identified and fully [sic] in analyst and news reports . . . well before [the WSJ Article]." The court rejected this event study as inadequate, noting that while the information contained within the article may not have been new, the negative characterization of previously known facts was an additional confounding factor that Hakala's analysis omitted altogether. Consequently, the court found that the Omnicom plaintiffs had failed to satisfy their burden with respect to loss causation. See id. at 554.

A footnote in Robbins suggests the Eleventh Circuit's approach to situations where multiple causes may have resulted in the stock's decline. In dicta, the court noted that while a plaintiff may establish loss causation without showing that a misrepresentation was the sole reason for the investment's decline in value, the plaintiff may only recover those damages actually caused by the misrepresentation. See Robbins, 116 F.3d at 1447 n. 5. The court further concluded that in order to establish the amount of recoverable damages, other contributing factors to the decline must be isolated and removed. Id. The Court emphasized that loss causation, which did not require the isolation of contributing forces, was a distinct inquiry from the determination of damages, which did require such isolation. Id.

Defendants contend that the Supreme Court has explicitly rejected the Robbins standard. In Dura, the Court held that in order to satisfy the PSLRA's pleading requirements with respect to loss causation, plaintiffs must allege that the defendants' conduct resulted in a fraudulently inflated stock price and that a decline in the stock's value resulted from the disclosure of facts underlying the fraud. See Dura Pharms., Inc. v. Broudo, 544 U.S. 336, 347, 125 S. Ct. 1627, 161 L. Ed. 2d 577 (2005). In so holding, the Court rejected the reasoning of the Ninth Circuit, which had held that plaintiffs must only show that a misrepresentation "touches upon" the reasons for an investment's decline in value. Broudo v. Dura Pharms., 339 F.3d 933, 937-38 (9th Cir. 2003). The Court noted that because a decline potentially reflects a "tangle of factors," — including "changed economic circumstances, changed investor expectations, new industry-specific or firm-specific facts, conditions, or other events" — an allegation that the fraudulent conduct merely "touches upon" a loss is insufficient to show that the misrepresentation caused the loss.Dura, 544 U.S. at 343.

The Eleventh Circuit, in Robbins, used the same "touches upon" phrase in the course of explicating its "substantial factor" rule. See Robbins, 116 F.3d at 1447. Defendants seize on this language to argue that Dura amounts to an express repudiation of the Robbins rule on loss causation. (Reply Memo. at 77-78.)

While the Court agrees that Dura requires plaintiffs to plead something more than the tenuous causal connection suggested by the "touch upon" language, it is reluctant to read Dura as requiring a wholesale rejection of the "substantial factor" standard. At least one district court has limited Dura's application in the context of summary judgment proceedings, noting that the Court "did not address [in Dura] what loss causation requires of a securities fraud plaintiff beyond the pleading stage." In re Motorola Sec. Litig., 505 F. Supp. 2d 501, 550 (N.D. Ill. 2007). The Court agrees that Dura, which centered on the adequacy of pleading, did not establish any clear standard with respect to summary judgment so as to supplantRobbins in its entirety.

The court likewise rejected the rule proposed by Defendants in the present case, finding that the Supreme Court's discussion of other factors was intended to "[provide] a rationale to support its observation that an inflated purchase price, standing alone, does not necessarily indicate that a later loss was caused by a misrepresentation," but did not require plaintiffs to "rule out all other factors that may have contributed to the loss following a disclosure." Motorola, 505 F. Supp. 2d at 549-50. As such, the court held that Dura did not disturb the Seventh Circuit'sCaremark rule that defendants bear the burden on summary judgment of excluding the misrepresentation as a cause of any loss. See id. at 550-51.

By the same token, while the Dura decision may not have invalidated the Robbins "substantial factor" test, the Supreme Court did signal that plaintiffs' allegations must satisfy some critical threshold above a mere showing that the defendants' fraudulent conduct had some amorphous, unquantified impact on the price of the stock. Although the Eleventh Circuit has clearly stated that fraudulent conduct need not be the sole cause of the plaintiff's loss, the requirement that the fraud play a "substantial" role in the loss is consistent with a rule that the plaintiff must provide some rough measure of the effect of the fraud versus other non-fraud-related factors. Dura reinforces and clarifies the substantiality requirement by foreclosing theories of causation which rely on the rejected "touch upon" standard.

Moreover, while the Court remains cognizant of the Eleventh Circuit's admonition that the loss causation and damages inquiries are separate and discrete, Dura teaches that the loss causation analysis is to some extent dependent upon the determination of recoverable damages. InDura, the Supreme Court treated proof of the plaintiff's actual economic loss as part and parcel of the loss causation inquiry. In its discussion of the PSLRA's history, the Court noted that misrepresentation actions, the common-law predecessor of securities fraud actions, required that a plaintiff "show not only that had he known the truth he would not have acted but also that he suffered actual economic loss." Dura, 544 U.S. at 343-44 (emphasis added). The Court went on to emphasize that the PSLRA "expressly imposes on plaintiffs `the burden of proving' that the defendant's misrepresentations `caused the loss for which the plaintiff seeks to recover,'" making clear Congress' intent to permit such recovery only where "plaintiffs adequately prove the traditional elements of causation and loss." Id. at 346 (emphases added and citation omitted). Therefore, the Court indicated that a plaintiff could not effectively establish loss causation without providing some rough measure of the loss itself. See id. at 348. As such, a determination of the substantiality of the fraudulent conduct's effect requires some measurement of the loss attributable to that conduct.

Having reviewed the jurisprudence of neighboring circuits, the Court finds persuasive the position of the Second Circuit that in order to defeat summary judgment, plaintiffs in a securities fraud case must present evidence disaggregating the fraud and non-fraud-related causes of the plaintiff's loss. This rule accords with recent decisions of this court utilizing the Second Circuit's disaggregation approach at the pleading stage. See Waterford Township General Employees Retirement System, Individually and on Behalf of All Others Similarly Situated v. Suntrust Banks, Inc., No. 1:09-CV-617-TWT, 2010 WL 3368922, at * 4 (N.D. Ga. Aug. 19, 2010) (finding that plaintiffs failed to apportion the loss among the multiple factors that ultimately destroyed plaintiffs' investment); In re HomeBanc Corp. Sec. Litig., 706 F. Supp. 2d 1336, 1361-62 (N.D. Ga. 2010) (finding insufficient a complaint that failed to make any allegations distinguishing losses caused by defendant's alleged misrepresentations from those caused by industry-wide factors).

This approach is also consistent with the Robbins rule as modified by Dura. Merely showing that multiple forces act upon an object is insufficient to show that a particular force was any more substantial than the others. Plaintiffs must provide the fact-finder with a basis for evaluating the relative effects of competing causes of a loss, thereby determining which factors were substantial and which were relatively minor or inconsequential. Accordingly, where several competing factors may have resulted in a decline in the plaintiff's loss, the plaintiff must provide sufficient evidence to apportion the loss between fraud-related and non-fraud-related causes. Such an analysis requires the plaintiff's expert to disentangle the effects of the alleged fraud from both industry-wide information and company-specific information unrelated to such fraud. See Vivendi, 605 F. Supp. 2d at 591.

Furthermore, an overly permissive reading of the substantiality requirement, allowing a plaintiff to establish loss causation without disaggregating other potential causes, would risk subjecting the parties to needless, protracted litigation only to reach the same result. The Fourth Circuit decision in Miller v Asensio provides a cautionary example. In that case, the jury concluded that the plaintiff proved that the defendant's fraud constituted a substantial cause of the plaintiff's injury under the loss causation inquiry, but that the plaintiff failed to provide any mechanism for discerning the portion of the injury solely attributable to the defendant's fraud under the damages inquiry. See Miller v. Asensio Co., Inc., 364 F.3d 223, 232-33 (4th Cir. 2004). On review, the Fourth Circuit upheld the verdict finding liability but awarding zero damages. See id. at 235. Nothing would be gained by permitting such an outcome in the present case, especially when considering the fact that the vast sums expended by both parties over several years of litigation are likely to pale in comparison to the expenses the parties would incur should this case proceed to trial.

Defendants assert that even if S-A's July 19 and August 16 statements constitute corrective disclosures, Plaintiffs have not disaggregated the information relating to the alleged fraud from other factors affecting the stock price on those days. In particular, Defendants point to two competing causes: (1) the impact of several industry-wide factors upon S-A and its competitors; (2) S-A's discussion of how such industry-wide factors would have specifically affected its operations.

As noted above, the July 19 and August 16 disclosures discussed several factors aside from excess customer inventories that contributed to S-A's fourth quarter performance and revisions to its guidance for FY 2002. These factors included the uncertain economic climate, reduced digital marketing efforts by cable operators during the summer, and the unexpectedly slow deployment of interactive applications associated with digital cable. In the days following the July 19 disclosure, several analysts discussed the impact of these confounding factors, attributing S-A's declines partially to the slowdown in the economy as well as the delayed deployment of interactive applications. (SOF ¶ 253.)

Plaintiffs have sufficiently accounted for the contributing industry-wide factors in their analysis. As Plaintiffs note, the existence of these confounding factors was not new information to the market on July 19. (Opp Br. at 103-11.) A number of analyst reports pre-dating S-A's corrective disclosures had discussed the risks to S-A's business posed by a softening economy:

• An April 20, 2001 report from Kaufman Brothers stated that "Scientific-Atlanta's results are certainly impressive in the wake of a communications equipment market that is seemingly screeching to a halt." (SMF ¶ 39c.)

• An article in Cable World from May 28, 2001 noted skepticism about whether S-A's Explorer 8000 would "become a must-have, especially during an economic slowdown." (SMF ¶ 39d.)

• A June 13, 2001 report from Josephthal Co. Inc. noted "the impact of a slowing economy on subscriber addition rates for digital TV service," while also praising S-A's strong earning performance as compared to other companies. (SMF Ex. 39e.)

Likewise, analysts had discussed the potential delays in the deployment of interactive applications, although the tone of these discussions suggests measured skepticism about the pace of such deployments rather than solid predictions concerning their impact upon S-A:

• An April 4, 2001 report from Deutsche Bane Alex. Brown Inc. noted that while there was "strong European support for interactive services, we remain concerned that US progress will be slow this year."

• A May 14 report from Wachovia Securities discussing a seminar on the cable television industry noted that "2001 deployment plans [for video-on-demand] are far from complete," while also reporting that the video-on-demand service "was discussed with enthusiasm by all panel members."

• A June 14 report from Thomas Weisel Partners noted that North American cable operators were "not yet focused on [interactive] applications" and that such applications would "wait until next year."

(SMF ¶ 41a, b; SMF Ex. 40d.) In addition, at least one July 17 report mentioned the effect of reduced digital marketing efforts during the summer prior to S-A's July 19 disclosures. (SMF ¶ 44a.) As such, Plaintiffs have presented an issue of fact as to whether the market was aware of the potential effect of these industry-wide factors prior to the July 19 and August 16 disclosures.

In a similar vein, Plaintiffs correctly argue that to the extent that these confounding factors affected the industry as a whole, Dr. Hakala's event study in the present case accurately accounts for these factors through the use of an industry index. (Pl. Opp. at 111-12.) In his analysis, Dr. Hakala employed a composite index comprised of reported information regarding eight companies with similar business models to S-A's, as well as a broader industry index. When calculating losses, Dr. Hakala determined the extent to which S-A's stock price differed from a "true value" price that was calculated using the composite index, which accounted for market-wide factors as a matter of course. In the days following S-A's July 19 disclosures, this composite index remained relatively steady, while S-A's share price dropped dramatically. (Hakala Report Ex. B-3: Scientific-Atlanta's Share Price Compared with Composite Index Over the Event Study Period.) This evidence indicates that S-A experienced a drop in share price as a result of its July 19 disclosure.

However, Plaintiffs fail to address Defendants' critical point that in the July 19 and August 16 statements, S-A confirmed that the pre-existing, industry-wide factors had impacted it specifically during the fourth quarter. Analyst reports prior to S-A's disclosures discussed the potential effect of the slowing economy, delays in the deployment of interactive services, and seasonality in MSO marketing efforts, and Dr. Hakala's analysis accounts for the effect of such reports on the industry as a whole as compared to S-A.

What was new to the market on July 19 was S-A's confirmation that these factors had in fact impacted S-A's business. Previous discussions of these industry-wide factors with respect to S-A tended to mix measured skepticism — as in the Cable World article mentioning the potential effects of an economic slowdown — with optimistic appraisal of S-A's resistance to these trends. The April 20 Kaufman Brothers report characterized S-A's sales as impressive in light of the downward trends in the overall market. The June 13 Josephthal Co. report likewise praised S-A's "stellar" performance as compared to other companies that were reporting "declining asset turnover."

These optimistic appraisals clashed sharply with S-A's July 19 disclosures, which mentioned at several points the impact of the slowing economy. The July 19 statements focused in particular on the lack of historical data that would have allowed S-A to gauge the "sensitivity of demand to changes in the economy," as well as the declining economy's adverse effect upon "consumer purchases of new digital services, and thus purchases of the company's digital products by the MSO's." (SOF ¶ 252.)

Moreover, in the days following the July 19 disclosures, analysts frequently cited to the effect of the weakened economy upon S-A's sales as frequently or more often than other factors affecting S-A's growth, including inventory issues:

• Kaufman Brothers, July 19: "Based on the slower growth prospects for digital subscriber additions among domestic MSOs, a heightened risk of industry consolidation that could curb spending in the near term, and softening economic conditions lead us to cut our estimates for FY02. . . ."

• ABN AMRO, July 20: "Subscriber bookings were down 6%. . . . Reasons for the slowdown include uncertain economic climate and order pushouts by MSOs due to installation delays."

• Merrill Lynch, July 20: "Subscriber product bookings were down 6%. . . . This decline is owing to the current macroeconomic environment, customer inventory levels, and slower-than-expected deployment of interactive applications. . . . [T]he company's overall performance in [transmission] was affected by a softening economy."

• Wells Fargo, July 20: "A slowdown in the rollout of digital cable services by cable operators — due to slowing economy, slow rollout of interactive services, inventory adjustments, and seasonal factors — began in the June quarter and is expected to continue into the next calendar year."

(SOF ¶ 253 (emphases added).)

Other analyst reports downplayed the role of customer inventory levels while focusing more heavily on other confounding factors. Although a report from Deutsche Bank Alex Brown discussed customer inventory levels (along with the softening economy and decreased demand for digital cable), the report characterized S-A's July 19 statements as attributing the disappointing fourth quarter to "global economic weakness and seasonal factors." (SOF Ex. 105.)

In addition, at least one report omitted discussions of inventory levels altogether, focusing more on S-A's discussion of how unexpectedly slow deployment of interactive applications impacted the demand for digital set-top boxes:

Reducing Rating to Buy from Strong Buy Due to Estimates Reduction, Uncertainty of Market Acceptance of Interactive Applications, and Above All, Likely Slowing Demand for Digital Cable in North America. . . . Our main reason for our downgrade is new information that the demand for digital cable is slowing in North America. . . . Management stated that they believed set top box shipments were down sequentially because cable MSOs had not deployed interactive applications, which would drive demand for digital set top boxes, and industry seasonality where people do not watch as much TV in summer months.

(SOF Ex. 109.)

Given the change in analysts' tone from cautious optimism about S-A's strength relative to the weakening economy prior to July 19 to definitive statements about the adverse effect of the economy and other industry-wide factors following the July 19 disclosure, the evidence suggests that S-A's July 19 discussion of these confounding factors played some role in the decline of S-A's stock. As in Omnicom, this new characterization of previously-known information presents a confounding factor, the effect of which Plaintiffs have failed to isolate.

While Plaintiffs have proffered evidence that the market reacted negatively to the July 19 disclosures, Dr. Hakala's analysis fails to disentangle the effect of the new information regarding customer inventory levels from S-A's new, negative characterization of how industry-wide trends were affecting it specifically. As noted inOmnicom, such a partial disaggregation of confounding factors is insufficient to establish that the alleged misrepresentations actually caused Plaintiffs' loss. See Omnicom, 541 F. Supp. 2d at 554. Furthermore, Plaintiffs' sole argument on this point echoes the failed position of the Omnicom plaintiffs, i.e., that the issues discussed had previously been identified in analyst and news reports. See id. As such, this failure to acknowledge the effect of S-A's public statements regarding the company-specific effect of market-wide trends is likewise fatal to Plaintiffs in the present case.

Therefore, the record evidence provides no method by which a jury can determine how much, if any, of Plaintiffs' loss is attributable to Defendants' failure to disclose its alleged channel stuffing activities. At best, a jury could determine that S-A's fraudulent conduct was a factor in Plaintiffs' loss, but would have no basis for concluding that such conduct was a substantial factor. As Plaintiffs are unable to satisfy theRobbins standard on loss causation with respect to their claims of channel stuffing, Defendants' motion for summary judgment is due to be granted.

Defendants also contend that Plaintiffs, by improperly including inflation caused by alleged misrepresentations prior to the start of the class period, fail to determine the portion of S-A's price inflation that is properly attributable to misrepresentations during the class period. However, Plaintiffs argue persuasively that the class period inflation includes not only the price increase in response to S-A's January 2001 statements, but also the pre-class period inflation that would have been removed from the stock price had S-A accurately provided information about demand for its products and how it was generating sales. Other courts considering this question have endorsed such a theory of recovery on 10b-5 claims. See In re Bristol-Myers Squibb Sec. Litig., No. Civ.A. 00-1990 (SRC), 2005 WL 2007004, at *17 (D.N.J. Aug. 17, 2005) ("Logic suggests that if a material omission serves to conceal information that would otherwise cause the stock price to fall, the very fact that the price does not change (until corrective disclosure) would evince the statement's materiality."); Swack v. Credit Suisse First Boston, 383 F. Supp. 2d 223, 240 (D. Mass. 2004) ("Defendants' conduct could have tempered a drop in price that would otherwise have occurred, or resulted in a greater increase than the stock would otherwise have enjoyed, absent the deceptive analyst reports."). As Dr. Hakala's report quantifies both the inflation added as well as the inflation maintained by Defendants' alleged misrepresentations, the inclusion of such pre-class period inflation provides no basis for a grant of summary judgment. (Pl. Opp. at 99; Hakala Report at 37.) Nonetheless, the point is moot in light of Plaintiffs' failure to otherwise establish loss causation.

Conclusion

For the foregoing reasons, Defendants' Motion for Summary Judgment [412, 420] is GRANTED. Defendants' Motion to Strike and Motion in Limine Concerning Inadmissible Evidence of Collateral Government Proceedings [439, 445] is DENIED as moot, with leave to re-file as necessary.

SO ORDERED.