Berson v. Applied Signal Technology, Inc.Berson v. Applied Signal Technology, Inc.
We consider whether plaintiffs adequately pled a claim of securities fraud— something that is much harder now than in days gone by.
Facts
Plaintiffs bought stock in Applied Signal Technology, Inc., during the six months before the company revealed that its revenue had fallen 25% from the preceding quarter. Immediately following this disclosure, the stock price dropped 16%. Plaintiffs sued the company and two of its officers under Securities Exchange Act § 10(b),
Applied Signal’s customers are almost all agencies of the federal government. Two civilian agencies together account for 80% of the company’s revenue; military agencies account for most of the rest. According to plaintiffs, these government customers can, at any time and for any reason, order the company to stop working on existing contracts for up to 90 days. Compl. ¶ 26;
see
The district court dismissed the complaint on several grounds and plaintiffs, naturally, appeal.
Analysis
We first consider whether plaintiffs have pled with sufficient particularity the existence, content and effect of the stop-work orders. We then discuss whether defendants’ alleged practice of counting stopped work as backlog was misleading, before weighing plaintiffs’ allegations of scienter and loss causation. Finally, we ponder whether the backlog reports are immune from liability as “forward-looking” statements.
Defendants claim that plaintiffs haven’t alleged sufficient facts to show that the company ever received three of the four stop-work orders, or that these orders halted any work that was later reported as backlog, and that plaintiffs therefore haven’t pled with “particularity” the “reasons” why the backlog figures were misleading.
But the complaint identifies four confidential witnesses who worked for Applied Signal and who allegedly will testify to the existence and effect of the stop-work orders. Defendants quibble that these witnesses weren’t in a position to see the stop-work orders first-hand because they were “engineers or technical editors” rather than managers. But any number of company employees would be in a position to infer the issuance of stop-work orders, which would have had the very obvious effect of putting numerous employees out of work. It’s entirely plausible that “engineers or technical editors” would know, or could reasonably deduce, that the company had suffered such setbacks.
See In re Daou Sys., Inc.,
The complaint also alleges with particularity that defendants counted stopped work as backlog. Defendants admitted as much in two conference calls with analysts (transcripts of which were included in the company’s SEC filings). See pp. 986-97 infra. And the complaint alleges that the stop-work orders were still in effect when defendants touted the company’s backlog; defendants’ arguments to the contrary simply misread the complaint. 1 Defendants suggest that the stop-work orders may have expired (and work may have resumed) before they announced the backlog, and they speculate that two of the orders may have actually been “renewals” of an earlier order. But plaintiffs’ confidential witnesses suggest otherwise, so these disputes must at least await discovery.
2. Ruminations on “misleading”
Defendants argue that, even if they did count stopped work as backlog, this couldn’t have misled reasonable investors, who would have understood that this was just what defendants were doing.
See Brody v. Transitional Hospitals Corp.,
Our backlog ... consists of anticipated revenues from the uncompleted portionsof existing contracts .... Anticipated revenues included in backlog may be realized over a multi-year period. We include a contract in backlog when the contract is signed by us and by our customer. We believe the backlog figures are firm, subject only to the cancellation and modification provisions contained in our contracts.... Because of possible future changes in delivery schedules and cancellations of orders, backlog at any particular date is not necessarily representative of actual sales to be expected for any succeeding period, and actual sales for the year may not meet or exceed the backlog represented. We may experience significant contract cancellations that were previously booked and included in backlog.
(Emphasis added.) According to defendants, reasonable investors would interpret the emphasized phrase to mean that backlog includes stopped work: Because a stop-work order doesn’t actually cancel the contract, the contract continues to “ex-istí ]”; and because stopped work is perforce “uncompleted,” it still counts as backlog, even though Applied Signal may never get to complete it.
While this is a conceivable interpretation of this paragraph, it is hardly the only — or even the most plausible — one. It is just as likely that investors would interpret the underlined phrase as limited to work still in progress or work yet to be started on ongoing contracts. And, though the paragraph refers to customers’ rights to “ean-eel[]” or “modiffy]” existing contracts, it says nothing about the right to simply stop work and thus immediately interrupt the company’s revenue stream. The passage, moreover, speaks entirely of as-yet-unrealized risks and contingencies. Nothing alerts the reader that some of these risks may already have come to fruition, and that what the company refers to as backlog includes work that is substantially delayed and at serious risk of being cancelled altogether.
In ruling otherwise, the district court thought it significant that in a conference call with analysts, 2 one defendant hinted that stopped work counted as backlog. (As discussed above, see p. 985 supra, this conversation supports plaintiffs’ allegation that Applied Signal did count stopped work as backlog.) The brief exchange went as follows:
Q. [0]ne more question for you ... please. Does the $143 million [of backlog] include — is that net of any potential debooking?
A. That includes the $12 million that has not been debooked.
Q. So it’s not net of any potential de-booking? Includes?
A. That’s right.
With the benefit of hindsight and some help from the briefs,
3
we can see how this exchange might be interpreted to communicate that defendants counted stopped work as backlog. The $12 million figure matches the amount of work halted by the first stop-work order (which the company had disclosed some three months earlier) and the company representative (who was
Defendants also argue that reasonable investors wouldn’t have been misled because federal regulations warned them that Applied Signal’s customers might issue stop-work orders.
See
None of this means that defendants had an affirmative duty to disclose the stop-work orders.
See Gallagher v. Abbott Labs., Inc.,
3. Mental states and strong inferences
Plaintiffs must “state with particularity facts giving rise to a strong inference” that defendants acted with the intent to deceive or with deliberate recklessness as to the possibility of misleading investors.
As described above, plaintiffs allege the existence and effect of the stop-work orders with particularity.
See
pp. 984-85
supra.
But plaintiffs allege no particular facts indicating that Yancey and Doyle actually knew about the stop-work orders.
4
Instead, plaintiffs
infer
that these high-level managers must have known about the orders because of their devastating effect on the corporation’s revenue. We approved a similar inference in
No. 84 Employer-Teamster Joint Council Pension Trust Fund v. America West,
Plaintiffs’ inference here is far stronger than the one we approved in America West. Defendants in America West were outside directors who did no more for the company than attend board meetings and serve on a board committee. Id. at 943. Here, by contrast, Yancey and Doyle were directly responsible for Applied Signal’s day-to-day operations, see Compl. ¶¶ 8-10, 13, so it is hard to believe that they would not have known about stop-work orders that allegedly halted tens of millions of dollars of the company’s work. 5 If plaintiffs in America West could rely on an inference that outside directors were aware of maintenance problems over which they had no direct management responsibility, then plaintiffs here are entitled to rely on a similar inference as to the four stop-work orders. .
Defendants argue that
America West
conflicts with our later decision in
In re Read-Rite Corp.,
Plaintiffs also rely upon Epstein v. Itron, Inc.,993 F.Supp. 1314 (E.D.Wash.1998), in which a district court held, “facts critical to a business’s core operations or an important transaction generally are so apparent that their knowledge may be attributed to the company and its key officers.” .... Epstein, however, predates Silicon Graphics, which clarified the requirement that a plaintiff must plead with particularity....
Plaintiffs may have established a “reasonable inference” that, based upon their job duties at Read-Rite, [defendant officers] “would be aware of thefalsity of some or all of the statements” regarding the [company’s] products. The existence of a “reasonable inference,” however, does not satisfy the PSLRA’s requirement that Plaintiffs allege particular facts....
Id. at 848-49 (citations omitted). Defendants misread Read-Rite. In that case, plaintiffs hadn’t even alleged that defendants’ optimistic statements were false or misleading when made. (Plaintiffs tried to show that these statements were inconsistent with defendants’ later admissions, but on close inspection, it turned out that there wasn’t any inconsistency at all — and therefore no support for the proposition that the earlier optimistic statements were false. See id. at 847-48.) So plaintiffs in Read-Rite were drawing a different inference from the kind at issue here and in America West. The Read-Rite plaintiffs were trying to show that defendants ought to have discovered as-yet-unalleged facts that would have convinced defendants that their optimism was unfounded. We held that this was an inference too far: Where defendants make cheerful predictions that do not come to pass, plaintiffs may not argue, based only on defendants’ prominent positions in the company, that they ought to have known better. Instead, the PSLRA requires “particular allegations which strongly imply Defendants’] contemporaneous knowledge that the statement was false when made.” Id. at 847.
Here, unlike
Read-Rite,
plaintiffs alleged particular facts (the stop-work orders) that support the inference that the backlog statements were misleading when made. These facts were prominent enough that it would be “absurd to suggest” that top management was unaware of them.
America West,
4. Particular causes of loss
Plaintiffs must allege that defendants’ misleading backlog statements caused them a “later economic loss.”
Dura Pharm., Inc. v. Broudo,
The parties dispute the standard for pleading loss causation. Plaintiffs claim that Rule 8(a)(2) applies, and that they need only provide a “short and plain statement” of their loss causation theory.
We need not decide the question here. Assuming — without deciding — that
5. Backlog doesn’t look forward
Defendants argue that the PSLRA bars liability for their statements about backlog because those statements were “forward-looking.”
REVERSED and REMANDED.
Notes
. The complaint alleges that in January 2005, defendants’ bacldog figure included work that “probably would not be realized as a result of Stop-Work Order No. 1.” Compl. ¶ 34. That is just another way of saying that the order was still in effect. The complaint also alleges that the third stop-work order removed “customer funding” from a “major project” that was "abandoned” several months later.
Id.
¶ 43. Again, this is a more detailed and powerful way of making the same point. And the complaint alleges that the fourth stop-work order was received in December 2004.
Id.
¶ 35(c). Because the default duration of stop-work orders is 90 days,
see
. The district court's memorandum cites two conference calls, but the second call occurred on the final day of the period in which plaintiffs bought stock, so this call cannot have had any effect on what a reasonable investor would have thought during that period.
. Plaintiffs cite this conversation as proof that defendants counted stopped work as backlog. See p. 985 supra. Defendants mention the conversation in passing, see Appellees' Br. at 8 n. 4, but do not rely on it to argue that their backlog statements weren't misleading, see id. at 33-37. We discuss the phone call here because the district court relied on it in finding that the backlog statements weren’t misleading.
. It’s undisputed that defendants knew about the first stop-work order by September 9, 2004 at the latest — that's the day they disclosed the order in a filing with the SEC. They also referred to it, rather obliquely, in a conference call with investors three months later. See p. 986-87 supra. The relevant question is whether defendants knew about this order on August 24, 2004, when they announced their backlog to investors in a conference call without disclosing the order.
. The first stop-work order halted between $10 and $15 million of work on the company's largest contract with one of its most important customers. Compl. ¶ 29. The size of the contract and the prominence of the client raise a strong inference that defendants would be aware of this order. And defendants clearly did know about it later, when they disclosed it in an SEC filing. That disclosure occurred just two weeks after defendants’ allegedly misleading backlog statement, and the "temporal proximity” of the misleading statement and the subsequent disclosure ”bolster[s]” the inference that defendants knew about the order when they made the statement.
Ronconi v. Larkin,
The second stop-work order halted $8 million of work. Compl. ¶ 30(b). According to a confidential witness, the order followed a "series” of client meetings where "management” tried unsuccessfully to "negotiate away” certain contract requirements. Id. Those alleged meetings raise a strong inference that defendants were aware of this client’s dissatisfaction and thus also aware of the stop-work order that resulted from it.
A different confidential witness will allegedly testify that the third stop-work order caused the company to reassign "50-75 employees,” with the result that one of the company’s facilities became a "ghost town.” Id. ¶43. The magnitude of this reallocation of personnel raises a strong inference that defendants were aware of it, and thus also aware of the stop-work order that caused it.
The fourth stop-work order came from a difficult client, an agency that required Applied Signal to “complete massive volumes of paperwork” to meet the agency’s reporting requirements. Id. ¶ 35(c). Applied Signal was "unable to comply” with the agency’s demands and, apparently as a result, the head of the project was “demoted.” Id. Such difficulties make it highly likely that defendants were aware of this customer’s problems; this customer in particular would be on defendants’ radar screen because it had previously canceled other large contracts with Applied Signal. Id.