BACHARACH, Circuit Judge.

Spirit AeroSystems, Inc. agreed to supply parts for three types of aircraft manufactured by Gulfstream Aerospace Corporation and The Boeing Company. These aircraft were the Gulfstream G280 and G650 and the Boeing 787. For these aircraft, Spirit managed production of the parts through three projects. Each project encountered production delays and cost overruns, and Spirit periodically reported to the public about the projects' progress. In these reports, Spirit acknowledged risks but expressed confidence about its ability to meet production deadlines and ultimately break even on the projects. Eventually, however, Spirit announced on October 25, 2012, that it expected to lose hundreds of millions of dollars on the three projects. Spirit's stock price fell roughly 30 percent following the announcement.

The plaintiffs brought this action on behalf of a class of individuals and organizations that had owned or obtained Spirit stock between November 3, 2011, and October 24, 2012. (We refer to this period as the "class period.") The named defendants are Spirit and four of its executives:

According to the plaintiffs, Spirit and these executives misrepresented and failed to disclose the projects' cost overruns and production delays, violating § 10(b) of the Securities Exchange Act of 1934 and the Securities and Exchange Commission's Rule 10b-5. 15 U.S.C. § 78j(b); 17 C.F.R. § 240.10b-5.2

The defendants moved to dismiss the complaint, arguing that the plaintiffs had failed to allege facts showing

The district court granted the motion to dismiss, concluding in part that the plaintiffs had failed to allege facts showing scienter.3

The plaintiffs appeal. We affirm because the plaintiffs have not alleged facts creating a cogent and compelling inference of scienter.

I. The Plaintiffs' Pleading Burden on Scienter

For the plaintiffs' claims under § 10(b) and Rule 10b-5, scienter is an essential element. In re Zagg, Inc. Sec. Litig., 797 F.3d 1194, 1200 (10th Cir. 2015).4 Scienter consists of "`a mental state embracing intent to deceive, manipulate, or defraud,' or recklessness." Adams v. Kinder-Morgan, Inc., 340 F.3d 1083, 1105 (10th Cir. 2003) (quoting City of Philadelphia v. Fleming Cos., 264 F.3d 1245, 1259 (10th Cir. 2001)). Conduct is considered reckless only if the defendants (1) acted in "an extreme departure from the standards of ordinary care" and (2) presented "a danger of misleading buyers or sellers" that was

In re Level 3 Commc'ns, Inc. Sec. Litig., 667 F.3d 1331, 1343 n. 12 (10th Cir. 2012) (quoting City of Philadelphia v. Fleming Cos., 264 F.3d 1245, 1260 (10th Cir. 2001)).

For scienter, the Private Securities Litigation Reform Act of 1995 creates a heightened duty for the plaintiffs to "state with particularity facts giving rise to a strong inference that the defendant[s] acted with the required state of mind." 15 U.S.C. § 78u-4(b)(2)(A); see also In re Zagg, 797 F.3d at 1201-02 (discussing the heightened duty).

We consider this statutory duty through de novo review of the dismissal. In re Gold Res. Corp. Sec. Litig., 776 F.3d 1103, 1108 (10th Cir. 2015). In conducting de novo review, we accept the complaint's factual allegations as true. Dronsejko v. Thornton, 632 F.3d 658, 666 (10th Cir. 2011). We then assess these allegations holistically and consider "whether all of the facts alleged, taken collectively, give rise to a strong inference of scienter, not whether any individual allegation, scrutinized in isolation, meets that standard." In re Zagg, 797 F.3d at 1201-02 (emphasis in original) (quoting Tellabs, Inc. v. Makor Issues & Rights, Ltd., 551 U.S. 308, 323, 127 S.Ct. 2499, 168 L.Ed.2d 179 (2007)).

To assess the strength of an inference of scienter, we compare the "inferences urged by the plaintiff[s]" with "competing inferences rationally drawn from the facts alleged." Tellabs, 551 U.S. at 314, 127 S.Ct. 2499. An inference of fraudulent intent must be more than "`reasonable' or `permissible' — it must be cogent and compelling." Id. at 324, 127 S.Ct. 2499. Thus, the complaint suffices "only if a reasonable person would deem the inference of scienter cogent and at least as compelling as any opposing inference one could draw from the facts alleged." Id.

With this standard in mind, we consider whether the plaintiffs adequately pleaded scienter. We conclude that they did not.

II. The Existence of Conflicting Inferences of Innocence and Scienter

The plaintiffs urge the following inference of the defendants' scienter:

The defendants argue that an innocent inference is more compelling even if we credit the allegations in the complaint:

Our task is to determine whether the inference arising from the plaintiffs' scienter allegations is cogent and compelling when compared to the defendants' suggested inference. In our view, the plaintiffs' scienter inference is not cogent and compelling.

Spirit's executives knew that Spirit had encountered problems in containing costs and meeting production deadlines. And we can assume (without deciding) that Spirit did not adequately communicate these problems to the public.

But why didn't Spirit adequately communicate these problems? The plaintiffs allege that the Spirit executives intentionally misrepresented or recklessly ignored economic realities. That is possible, but it is more probable that the Spirit executives were overly optimistic and failed to give adequate weight to financial red flags, for the plaintiffs supply little reason to suspect malevolence rather than benign optimism. See In re Level 3 Commc'ns, Inc. Sec. Litig., 667 F.3d 1331, 1345 (10th Cir. 2012) (holding that the plaintiffs' scienter allegations, which were based on inconsistency between the defendants' progress reports and internal reports, were inadequate because "the strongest inference we can draw is that defendants were negligent in failing to put together the pieces"); see also Plumbers & Pipefitters Local Union 719 Pension Fund v. Zimmer Holdings, Inc., 679 F.3d 952, 955 (7th Cir. 2012) ("Knowing of `problems,' which are common, differs from [specifically] knowing that a facility must be closed and some of its products recalled.").

To establish scienter, the plaintiffs rely on

Together, these allegations do not create a cogent, compelling inference of scienter.

III. The Existence of a Potential Motive to Lie

The plaintiffs do not argue that the defendants had a particularized motive for committing securities fraud. The district court viewed the absence of a particularized motive as a factor mitigating against an inference of scienter. The court's consideration of this factor was proper.

The plaintiffs need not show that the defendants acted with a particular motive. See Nakkhumpun v. Taylor, 782 F.3d 1142, 1152 (10th Cir. 2015) ("[S]cienter allegations may suffice even without a motive."). But "[t]he absence of a motive allegation... is `relevant.'" In re Level 3 Commc'ns, Inc. Sec. Litig., 667 F.3d 1331, 1347 (10th Cir. 2012).

On appeal, the plaintiffs argue that the defendants were "motivated to mislead in order to buy time in the hope that a difficult financial situation `would right itself.'" Appellants' Opening Br. at 69. This argument was not preserved and is invalid.

The plaintiffs did not raise this argument in district court, so this argument is forfeited. See Richison v. Ernest Grp., Inc., 634 F.3d 1123, 1128 (10th Cir. 2011) ("[I]f [a] theory ... wasn't raised before the district court, we usually hold it forfeited."). Ordinarily we can consider forfeited arguments under the plain-error standard. Id. But the plaintiffs do not argue plain error in the appeal. As a result, this argument has been waived. Id.

Even if this argument had been preserved, it would not have cured the plaintiffs' failure to adequately plead scienter. At most, the plaintiffs' argument suggests a generalized corporate motive; but that would not contribute to an inference of scienter. See Level 3 Commc'ns, 667 F.3d at 1346 ("[G]eneral motives for management to further the interests of the corporation fail to raise an inference of scienter."); City of Philadelphia v. Fleming Cos., 264 F.3d 1245, 1269 (10th Cir. 2001) ("[G]eneralized motives shared by all companies and which are not specifically and uniquely related to [the defendants] in particular, are unavailing."). Thus, the plaintiffs failed to allege facts indicating a motive for securities fraud, which mitigates an inference of scienter from the plaintiffs' other factual allegations.

IV. Information Furnished by Corroborating Witnesses

The plaintiffs contend that corroborating witness accounts show that the defendants knew long before October 2012 that the projects' costs would surpass revenues. We accept the truth of the witnesses' accounts as pleaded in the complaint and do not assess the witnesses' credibility. See Tellabs, Inc. v. Makor Issues & Rights, Ltd., 551 U.S. 308, 322, 127 S.Ct. 2499, 168 L.Ed.2d 179 (2007) (we accept factual allegations in the complaint as true); Nakkhumpun v. Taylor, 782 F.3d 1142, 1152 (10th Cir. 2015) (in reviewing the grant of a motion to dismiss, we do not assess witness credibility).

To raise an inference of scienter, the plaintiffs point to two types of information supplied by corroborating witnesses:

The generalized descriptions of internal meetings, cost reports, delays, and mismanagement do not contribute to a cogent, compelling inference of scienter. The factual allegations regarding Mr. George suggest that he had a culpable intent, but the plaintiffs do not tie Mr. George's potentially culpable state of mind to any public disclosures.

A. Generalized Descriptions of Internal Meetings, Cost Reports, Delays, and Mismanagement

The plaintiffs contend that Spirit's chief executive officer and chief financial officer must have known long before October 2012 that Spirit would lose hundreds of millions of dollars on the three projects. For this contention, the plaintiffs point to information supplied by corroborating witnesses about

The witnesses' accounts do not support a scienter inference.

First, the plaintiffs point to an internal meeting that Spirit's chief executive officer and chief financial officer led in mid-2011, when they allegedly admitted that one of the Gulfstream projects was "behind schedule and over budget." Appellants' Opening Br. at 55.

This admission indicates that the two executives knew in mid-2011 that production was behind schedule and that costs exceeded projections for one of the projects. But this meeting had taken place months before the class period began, as well as before the executives' public disclosures. The plaintiffs do not identify any particularized account showing that the two executives knew during the class period that their progress reports were inaccurate.6 See Wolfe v. Aspenbio Pharma, Inc., 587 Fed.Appx. 493, 498 (10th Cir. 2014) (unpublished) (concluding that a defendant's remarks about a company's problems, made months before a class period, would not establish scienter during the class period).7

Second, the plaintiffs point to the collective accounts of ten corroborating witnesses who furnished information about

But these accounts do not contribute to a cogent and compelling inference of scienter for three reasons:

First, Corroborating Witness #5 referred to a 2011 cost-study report prepared by an internal business group. This report had stated that Spirit could not meet cost-reduction goals. Appellants' App'x, vol. I at 122. In addition, Corroborating Witness #5 stated that the business group had regularly reported losses on the Gulfstream projects. Id. at 121-22. But the account by Corroborating Witness #5 does not establish the executives' knowledge because

Second, Corroborating Witness #10 contributed to the preparation of quarterly cost reports on the three projects that were ultimately furnished to Mr. Anderson and Mr. Turner. Id. at 134-35. To create an inference of scienter based on these quarterly cost reports, the plaintiffs must adequately describe the content of the reports provided to Mr. Anderson and Mr. Turner. See Lipton v. Pathogenesis Corp., 284 F.3d 1027, 1036 (9th Cir. 2002) ("[N]egative characterizations of reports relied on by insiders, without specific reference to the contents of those reports, are insufficient to meet the heightened pleading requirements of the [Private Securities Litigation Reform Act]."). The plaintiffs have not identified the content of the quarterly cost reports. Instead, the plaintiffs rely on Corroborating Witness #10's role in contributing data on procurement costs used to generate these reports. But the content of the resulting quarterly cost reports is unknown because (1) they went through numerous layers of revision before reaching Mr. Turner and Mr. Anderson, and (2) Corroborating Witness #10 did not purport to know how the procurement costs affected the cost forecasts for the Boeing 787 project.

Corroborating Witness #10 participated in the reporting of procurement costs, but these reports went through numerous layers of revision before any of the data would have been seen by Mr. Turner or Mr. Anderson. For example, Corroborating Witness #10 furnished reports on procurement costs to Supply Chain Management, which approved or disapproved the report and submitted a report on procurement costs to a Supply Chain Director. If approved, the report was sent to the Controller of Spirit's Tulsa Facility, who reported to vice presidents in the Tulsa Facility. The vice presidents provided the eventual report on procurement costs to the Finance Group, which incorporated forecasts on internal costs such as labor and processing. The Finance Group then provided the consolidated report to Tulsa's Vice President of Operations (Mr. Chris Collins). If approved, the report was forwarded to Wichita Management, which would "provide feedback" and in "some circumstances" request Corroborating Witness #10's group to provide supporting documentation for particular costs. Appellants' App'x, vol. I at 136. Once Wichita Management agreed with the report, it would be sent to Mr. Turner and Mr. Anderson.

Corroborating Witness #10 was able to see only a part of the cost reports' data, which purportedly showed "cost overruns" and "changes after changes" to engineering designs on the Boeing 787 project. See Appellants' App'x, vol. I at 136-37. This data involved procurement costs. But the quarterly cost reports contained another important set of data that Corroborating Witness #10 does not allege to have seen: Spirit's forecasts on internal costs. Id. at 135.

Spirit's Finance Group consolidated the information supplied by Corroborating Witness #10's group with forecasts on internal costs. There is nothing in the complaint to indicate that Corroborating Witness #10 would have had information about the consolidation process. Id.; see also id. at 135 (alleging that Corroborating Witness #10's group was "responsible for preparing the procurement costs"). Thus, the complaint sheds little light on what Mr. Turner and Mr. Anderson would have seen if they had read the quarterly cost reports.9

But even if we assume that Mr. Turner and Mr. Anderson were aware of overruns on procurement costs, we have no information from Corroborating Witness #10 on what the internal costs were or how the overruns on procurement costs would have affected the total project costs. In these circumstances, Corroborating Witness #10's alleged knowledge of overruns on procurement costs would not contribute to an inference of scienter.

Third, the plaintiffs rely on general reports of delay and mismanagement of the Gulfstream projects. Appellants' Opening Br. at 55-56. To the plaintiffs, these reports suggest that the Spirit executives must have long known that Spirit would need to take a substantial forward loss on the projects. But this inference does not follow from anything that the corroborating witnesses alleged.

Only one of the corroborating witnesses worked closely with any of the four Spirit executives, and this witness spoke only of general corporate mismanagement; the witness did not address whether the executives could have known that Spirit would be unable to meet long-term cost forecasts. See Appellants' App'x, vol. I at 112 (observing a general "lack of cohesive team environment, [lack of] consistent quality control standards and [lack of] updated technology"); id. (noting broad concerns that contractors would steal tools, "creat[ing] tool shortages" that lead to delays and additional costs); id. at 113 ("Because the Gulfstream program became so delayed, Spirit had to use a special Russian aircraft to fly the parts to the customer's destination at a cost of almost $1 million each time....").10

The other corroborating witnesses were further removed from the defendants. See, e.g., id. at 113, 116-17, 123, 127 (four corroborating witnesses who had no alleged reporting relationship to the defendants); id. at 129, 132 (two corroborating witnesses who were separated from the chief executive officer by at least four levels of management within Spirit's hierarchy).11 As relatively low-level employees, these witnesses could not provide evidence bearing on the executives' mental states. See Zucco Partners, LLC v. Digimarc Corp., 552 F.3d 981, 998 (9th Cir. 2009) ("[G]eneralized claims about corporate knowledge are not sufficient to create a strong inference of scienter, since they fail to establish that the witness reporting them has reliable personal knowledge of the defendants' mental state."); see also In re Cabletron Sys., Inc., 311 F.3d 11, 31 (1st Cir. 2002) (requiring a corroborating witness to provide a "level of specific detail" about what the defendants must have known to create an inference of scienter).

The accounts by corroborating witnesses do not contribute to an inference of scienter. These witnesses were too far removed from the four Spirit executives and did not provide sufficiently particularized accounts of what the Spirit executives must have known.

B. Information Furnished by Corroborating Witness #3 about Mr. George's Culpability on the Boeing 787 Project

For scienter, the plaintiffs rely in part on Corroborating Witness #3's account to establish that Mr. George (the vice president overseeing Spirit's Boeing 787 project) "personally directed the reporting of inaccurate projected costs [for the Boeing 787 project] ..., rejecting cost figures generated by Spirit's functional managers." Appellants' Opening Br. at 56. This account creates a potentially plausible inference of Mr. George's scienter, but one could also infer from Corroborating Witness #3's account that Mr. George was simply too optimistic about Spirit's ability to control costs on the Boeing 787 project. This inference of innocence is more compelling than an inference of Mr. George's scienter.

The plaintiffs point to five public statements that Mr. George made about the Boeing 787 project at an investor conference in March 2012:

Id. at 51, 55 (emphasis omitted). According to the plaintiffs, these statements were knowingly false and misleading based on Corroborating Witness #3's interactions with Mr. George.

Corroborating Witness #3 worked in Spirit's finance department and consulted with Mr. George about cost forecasts for the Boeing 787 project. Id. at 38 n. 5. Mr. George relied on Corroborating Witness #3 to forecast costs by collecting projections from various "Functional Managers." Id. at 115. After reviewing these forecasts, Mr. George allegedly said that the cost projections were too high and ordered Corroborating Witness #3 to predict the project's future costs at a level that Corroborating Witness #3 regarded as unachievable. Id. at 115-16. When Corroborating Witness #3 balked, Mr. George allegedly threatened to find managers who "could achieve the forecasts." Id. at 116. Later, as Spirit was about to fall short of Mr. George's required forecast, Spirit allegedly tried to "off-load some of the work to an offshore vendor." Id.

From Corroborating Witness #3's account, we can reasonably infer that Mr. George was too optimistic about Spirit's ability to reduce costs on the Boeing 787 project. But there is nothing in Corroborating Witness #3's account to suggest that Mr. George knew that his statements at the March 2012 investor conference were false or misleading.

Corroborating Witness #3's account has two fundamental gaps:

In light of these gaps, we cannot plausibly infer from Corroborating Witness #3's account that Mr. George knowingly or recklessly misled attendees at the March 2012 investor conference.13

V. The Executives' Alleged Involvement in Spirit's "Core Operations"

The plaintiffs allege that Mr. Turner, Mr. Anderson, Mr. Kummant, and Mr. George were "top executives" who "personally monitored" the progress of the three projects.14 Appellants' Opening Br. at 61. Based on this involvement, the plaintiffs ask us to infer that the four Spirit executives must have known that the projects would not meet long-run cost forecasts. The allegations in the complaint do not support this inference.

We cannot infer scienter based only on a defendant's position in a company or involvement with a particular project. See, e.g., City of Philadelphia v. Fleming Cos., 264 F.3d 1245, 1264 (10th Cir. 2001) (concluding that "the mere fact" that defendants "occupied senior positions in the company" was inadequate for an inference of scienter).

The executives' positions at Spirit would help establish whether they should have known that particular cost projections were unrealistic. But additional particularized facts are necessary for an inference of scienter. See S. Ferry LP, No. 2 v. Killinger, 542 F.3d 776, 784 (9th Cir. 2008) ("Where a complaint relies on allegations that management had an important role in the company but does not contain additional detailed allegations about the defendants' actual exposure to information, it will usually fall short of the [securities laws' pleading] standard.").

The plaintiffs do not allege additional particularized facts. According to the plaintiffs, the four Spirit executives

Appellants' Opening Br. at 61. For these characterizations, the plaintiffs rely on Paragraphs 54, 69, 70, 75, 76, 79, 80, 92, and 107 of the complaint. Id.; Appellants' Reply Br. at 21. Of these paragraphs, only Paragraph 69 refers to an executive's direct involvement in the projects: the chief financial officer, Mr. Anderson, said that he had spent a lot of time in Spirit's Tulsa facility, "making sure the changes [to production processes were] taking hold." Appellants' App'x, vol. I at 45-46. The plaintiffs do not cite any allegations in the complaint stating that one of the four Spirit executives actually reviewed the underlying cost data for the Gulfstream and Boeing projects.

We assume, for the sake of argument, that the plaintiffs' allegations create an inference that the four executives were briefed on the underlying cost data while attending meetings where the three projects were discussed. But mere attendance at meetings does not contribute to an inference of scienter. See MHC Mut. Conversion Fund, L.P. v. Sandler O'Neill & Partners, L.P., 761 F.3d 1109, 1122 (10th Cir. 2014) ("[N]othing in the defendants' routine attendance at ... meetings serves to suggest a `cogent' and `compelling' inference of scienter."); In re Level 3 Commc'ns, Inc. Sec. Litig., 667 F.3d 1331, 1344 (10th Cir. 2012) (concluding that the fact that the defendants "monitored [a contested program] through regular meetings and reports" was insufficient to draw a "strong inference" of the defendants' scienter).

Based on the plaintiffs' allegations of the defendants' involvement in Spirit's core operations, we can infer only that the four executives were overly optimistic about Spirit's ability to achieve the forecasted production schedules and cost reductions. The plaintiffs have not provided a good reason to believe that the executives knew that the projects were unlikely to meet forecasts.

VI. A Duty to Disclose Production Delays and Cost Overruns

The plaintiffs also allege scienter based on the defendants' purported duty to disclose production delays and cost overruns. In our view, however, this duty would not support an inference of scienter here.

Liability for securities fraud requires the making of a material misrepresentation. Geman v. SEC, 334 F.3d 1183, 1192 (10th Cir. 2003). This requirement may be satisfied when the defendant remains silent while owing a duty to investors to correct their misperceptions. Id. But even if a defendant incurs a duty to speak, liability under the securities laws requires scienter. Id. To establish scienter, the plaintiffs must establish that a scienter inference is more cogent and compelling than an inference that the defendants had "not believe[d] they had a continuing duty to disclose information." Minneapolis Firefighters' Relief Ass'n v. MEMC Elec. Materials, Inc., 641 F.3d 1023, 1030 (8th Cir. 2011); see also Dirks v. SEC, 463 U.S. 646, 663 n. 23, 103 S.Ct. 3255, 77 L.Ed.2d 911 (1983) (distinguishing fraudulent intent from fraudulent conduct).

The plaintiffs do not explain why the existence of a duty to disclose supports an inference that the defendants (1) knew that they needed to disclose more or (2) were reckless in failing to disclose more. As a result, the defendants' alleged duty to disclose does not support an inference of scienter in this case.

VII. The Plaintiffs' Remaining Factual Allegations

The plaintiffs also point to

The plaintiffs argue that in light of the magnitude of the loss that Spirit ultimately sustained, Spirit must have had scienter when each of these occurrences took place. These arguments amount to allegations of "fraud by hindsight," which does not constitute securities fraud. See In re Level 3 Commc'ns, Inc. Sec. Litig., 667 F.3d 1331, 1347 (10th Cir. 2012) ("[The] defendants' hindsight review.... contributes nothing to an inference of scienter.").

A. Spirit's July 2012 Recovery Plan

According to the plaintiffs, Spirit adopted a "recovery plan" in July 2012 to put the Boeing 787 project back on schedule. The plaintiffs argue that the existence of this plan shows that Spirit executives knew that their subsequent representations about the Boeing 787 project were false or misleading.

Between implementing the recovery plan in July 2012 and announcing a forward loss on the Boeing 787 project in October 2012, Spirit executives issued multiple public statements on the status on the project. The plaintiffs argue that these statements in August and September 2012, summarized below, materially misrepresented the project's progress:

August 2, 2012

Appellants' App'x, vol. I at 70-71.

August 7, 2012

September 12, 2012

Id. at 74.

September 13, 2012

Id. at 75.

The plaintiffs argue that the Spirit executives knew that these statements were false or misleading because a recovery plan was already underway to bring the delivery schedule up to date with initial forecasts for the Boeing 787 project. Id. at 79. Spirit later reported that it had learned, when implementing the recovery plan, that future costs would increase dramatically if Spirit were to adhere to the required delivery schedule.

We may assume, without deciding, that the statements in August and September 2012 were materially false or misleading. To infer scienter, however, we must compare the parties' dual explanations for Spirit's overly optimistic reports on the status of the Boeing 787 project. The plaintiffs attribute the statements to the executives' recklessness or intent to mislead investors; the defendants attribute the statements to an honest belief that Spirit's recovery plan would reduce costs and accelerate production. The plaintiffs provide little reason to question the executives' explanation.

The eventual announcement of a forward loss suggests that Spirit had placed too much confidence in the recovery plan. But the same can always be said when a company delays announcement of a forward loss based on remedial efforts to increase profitability or production.

We addressed an analogous issue in In re Zagg, Inc. Sec. Litig., 797 F.3d 1194 (10th Cir. 2015). There, a corporation allegedly failed to disclose that a chief executive officer had pledged approximately half of his shares in the corporation as collateral in a personal margin account. 797 F.3d at 1198-99. The chief executive officer was eventually forced to resign, and the corporation adopted a corrective policy restricting executives' pledges of corporate shares. Id. at 1199. The plaintiffs argued that adoption of the policy showed that the corporation had a fraudulent intent. Id. at 1203, 1205. We rejected this argument, concluding that the new policy showed only that the corporation had "identified a better way of doing things moving forward." Id. at 1205.

In re Zagg is instructive. Spirit's July 2012 recovery plan shows that Spirit identified an interim step to reduce costs and expedite production on the Boeing 787 project. The plaintiffs suggest that Spirit executives wanted to mislead investors. That is possible, but the plaintiffs' theory presupposes that Spirit executives knew that the recovery plan could not accomplish the plan's stated objectives. There is nothing in the complaint to support that logical leap. The stronger inference is that the Spirit executives thought that they had "identified a better way of doing things." Id. As a result, we regard the defendants' "innocent inference" as more cogent and compelling than the plaintiffs' "scienter inference."

B. The Chief Executive Officer's Statements Explaining the Forward Loss

After Spirit announced the forward loss, Spirit's chief executive officer explained the forward loss on the three projects by stating that

Appellants' App'x, vol. I at 81-82. The chief executive officer added that

Id. at 81-83.

The plaintiffs argue that these statements show that Spirit recklessly ignored production delays and cost overruns. This argument is not supported by what the chief executive officer said. He acknowledged that Spirit had mistakenly projected Spirit's ability to improve efficiency and that Spirit ultimately learned that it could not meet projections. But the chief executive officer did not suggest that Spirit executives had knowingly or recklessly misevaluated earlier data. His statements simply show that Spirit ultimately learned that it could not meet the company's forecasts.

The plaintiffs allege that

On August 7, 2012, the chief executive officer stated that Spirit was meeting its cost estimates for the Boeing 787 project and would continue to do so. Id. at 73-74. Over two months later, Spirit announced a forward loss on the Boeing 787 project, and the chief executive officer acknowledged that (1) Spirit could no longer meet its cost estimates and (2) Spirit had recently fallen behind on production for the wing components on the Boeing 787 project. Id. at 80-82.

The plaintiffs argue that the October acknowledgments prove that the August prediction had been designed to mislead investors. We disagree: The statements were made over two months apart, and the complaint does not allege that the chief executive officer was aware in August 2012 that Spirit was not meeting its cost estimates. See La. Sch. Emps.' Ret. Sys. v. Ernst & Young, LLP, 622 F.3d 471, 484-85 (6th Cir. 2010) (holding that statements made after the end of the class period, recognizing that "there was a problem," did not contribute to an inference of scienter because the statements did not provide a sufficient basis to assume that the defendant had known that events would turn out badly simply because they eventually did).

In our view, the chief executive officer's statements suggest an honest mistake in predicting Spirit's future production and costs, not an inference of scienter.

C. Spirit's Disclosures of Risks

Throughout the class period, Spirit warned investors about risks from the three projects:

November 2011

February 2012

May 2012

August 2012

The plaintiffs argue that these warnings suggest a culpable mental state because the four executives must have known that the risks had already materialized: "the more a defendant speaks about a topic, the likelier it is that [the defendant] knows about the topic." Appellants' Opening Br. at 65. We disagree.

Ordinarily, a defendant's warnings weaken an inference of scienter. Ezra Charitable Tr. v. Tyco Int'l, Ltd., 466 F.3d 1, 8 (1st Cir. 2006). The plaintiffs argue the opposite, relying solely on Reese v. Malone, 747 F.3d 557 (9th Cir. 2014). In Reese, a group of shareholders sued BP for securities fraud after one of BP's pipelines leaked 200,000 gallons of oil onto the Alaskan tundra. Id. at 563. The plaintiffs alleged that BP had made false and misleading statements about company practices in an annual report post-dating the spill. Id. The Ninth Circuit Court of Appeals held that the plaintiffs' allegations of scienter were sufficient, in part because the attention given to the spill in the annual report underscored the importance of the oil spill to BP management. Id. at 579-80. But the court did not suggest that the mere existence of warnings established knowledge by BP management that the statements were untrue. Instead, the court inferred that the company knew that the statements were untrue because (1) BP was not in compliance when publishing the false report and (2) the spill had been prominently discussed in the report. Id.

Throughout the class period, Spirit was working on multiple projects and warned of the risks that many of them posed. For example, during the class period, Spirit warned about a significant risk of additional forward-loss charges on the Sikorsky CH-53K helicopter project and a possibility of significant cost growth and technical problems for the Airbus A350 XWB project. These warnings would not have suggested that the Sikorsky or Airbus projects were particularly important to Spirit executives or that Spirit executives knew that the risks on these programs were certain to result in a forward loss. Instead, the warnings simply reflected an awareness of risks that led Spirit to discuss the risks in its public filings. The same is true of the filings expressing potential risks for the Gulfstream and Boeing projects involved here.

D. Alleged Accounting Violations

The plaintiffs argue that Spirit violated generally accepted accounting principles and internal accounting policies by delaying announcement of the forward loss. We disagree, for this argument is simply another variation of "fraud by hindsight."

Even if we assume that Spirit had violated established accounting principles or internal policies, these violations would not have been enough to state a valid cause of action. City of Philadelphia v. Fleming Cos., 264 F.3d 1245, 1261 (10th Cir. 2001). "Only where such allegations are coupled with evidence that the violations or irregularities were the result of the defendant's fraudulent intent to mislead investors may they be sufficient to state a claim." Id.

No such evidence exists here, and the plaintiffs do not identify particularized facts showing that the Spirit executives had known, before Spirit publicly announced the forward loss, that the three projects would fall behind forecasts on cost or production.

The defendants' purported accounting violations are based on "fraud by hindsight"; and even if we assume the existence of accounting violations, the plaintiffs have not alleged facts indicating the executives' knowledge, fraudulent intent, or recklessness.

E. The Magnitude of the Forward Loss

In October 2012, Spirit announced a forward loss of $434.6 million on the three projects.15 The plaintiffs argue that because the forward loss was so large, the executives must have known long before October 2012 that Spirit would incur a loss. But this argument relies on hindsight review based on the size of Spirit's eventual loss and does not include particularized allegations that the four executives knew before October 2012 that Spirit would lose money on the three projects.

The loss of $434.6 million was undoubtedly significant. And the three projects were presumably large enough to capture the attention of the four Spirit executives. No one suggests otherwise. For scienter, however, the question is whether these executives (1) had known that their public disclosures were too encouraging or (2) had recklessly failed to heed red flags from potentially problematic reports. The size of the loss does not suggest that the four executives knew or recklessly disregarded the risks that Spirit was eventually going to lose money on the three projects.

* * *

To create a cogent, compelling inference of scienter, the plaintiffs point to a motive to lie, corroborating witnesses' accounts, the defendants' involvement in Spirit's "core operations," a duty to disclose, the adoption of a recovery plan, statements by the chief executive officer, disclosures of risks, accounting violations, and the size of the eventual forward loss. But even when these allegations are considered together, they do not create a cogent, compelling inference of scienter.

As a result, we affirm the dismissal.

LUCERO, J. concurring in part and dissenting in part.

The majority analyzes whether defendants recklessly made false statements regarding Spirit's projected ability to meet cost targets. However, plaintiffs' argument does not rest solely on future-looking statements. Rather, they contend that the defendants made false statements about then-present facts. In my view, plaintiffs have adequately alleged scienter and materiality as to statements made by Jeffrey Turner and Philip Anderson regarding the existing performance of Spirit's 787 program. Accordingly, I concur in part and respectfully dissent in part.

To plead a claim for violation of the Securities Exchange Act of 1934 § 10(b), plaintiffs must allege five elements, of which the parties focus only on the first and third:

In re Zagg, Inc. Sec. Litig., 797 F.3d 1194, 1200 (10th Cir. 2015) (quotation and emphasis omitted). To apply these factors, I think it helpful to first determine which statements plaintiffs have adequately alleged were false before analyzing whether the defendants acted with scienter as to those statements.

To demonstrate that the defendant made an untrue or misleading statement of material fact, a plaintiff must allege specific facts that contradict the defendant's statements. See Adams v. Kinder-Morgan, Inc., 340 F.3d 1083, 1097 (10th Cir. 2003). Moreover, the statement must be shown to be false when it was made. See Grossman v. Novell, Inc., 120 F.3d 1112, 1124 (10th Cir. 1997). Plaintiffs may not rely on a subsequent event triggering a drop in stock price "to say that the later, sobering revelations make the earlier, cheerier statement a falsehood." Id. (quotation omitted). Relatedly, forward-looking statements coupled with meaningful cautionary statements are not actionable because a reasonable investor simply cannot suppose that the future is settled or is not subject to multiple possible outcomes. Id. at 1120.

The matter before us concerns two distinct categories of statements — those projecting future performance, and those reflecting past performance. Plaintiffs argue, in essence, that the defendants stated both "we will be on budget" and "we are on budget." These two statements differ markedly for purposes of a securities fraud claim. Yet the majority opinion focuses its scienter analysis exclusively on the former. For example, the majority agrees with defendants that their public statements reflected "an honest belief that Spirit's recovery plan would reduce costs and accelerate production." (Majority Op. 1248.) The majority thus attributes the executives' statements to "benign optimism," (id. at 1238), and "mistaken[] project[ions]," (id. at 1249), about future performance. Therefore, the majority concludes the fact that the company "ultimately learned that it could not meet the company's forecasts" did not demonstrate that the optimistic statements regarding those forecasts were recklessly false at the time they were made. (Id.)

I agree in large measure. To the extent the plaintiffs argue that the company knew or should have known it would be unable to meet future cost forecasts, these forward looking statements — although later proven overly optimistic — are not actionable because they reflected opinion rather than fact, and were not demonstrably false. See In re Level 3 Commc'ns, Inc. Sec. Litig., 667 F.3d 1331, 1340 (10th Cir. 2012). I also agree with the majority that plaintiffs have not stated a claim as to defendants Alexander Kummant and Terry George. Further, I concur in the majority's dismissal of the claims based on defendants' statements regarding two of the Spirit programs at issue: the Gulfstream G280 and G650.

But I must part ways with the majority with respect to the portion of plaintiffs' claim based on false statements made by Turner and Anderson regarding the then-current performance of Spirit's 787 project. Plaintiffs have alleged specific statements by these two individuals indicating that the 787 program was on budget, and alleged particular facts showing they knew such claims to be false at the time they were made. These contentions do not rely on any prediction as to future costs. Rather, they show that executives inaccurately depicted the project's performance in meeting cost forecasts up to the time the statements were made. I would hold that these statements are actionable.

As the complaint explains, Spirit adopted a contract accounting method with respect to the 787 program. It reported a profit rate for the program based on actual historical costs plus future estimated costs. Because the company predicted it would benefit from a "learning curve" as the project progressed, the average cost per unit was expected to decrease as the total number of units produced increased. Thus, a downward-sloping "cost curve" was used to describe the average cost per unit over the course of the program.

According to confidential witness number ten ("CW10"), project leaders prepared Estimate at Completion ("EAC") reports each quarter. The EAC reports compared procurement costs actually incurred against forecast procurement costs. In other words, these reports showed whether the project's external expenses were on the cost curve and thus meeting expectations to that point or above the cost curve and thus over budget. CW10 was responsible for the EAC reports for every product delivered from Spirit's Tulsa facility, including the 787. He described that the EACs he prepared for the 787 program during 2011 and 2012 showed cost overruns relative to forecasts because "costs were much higher than expected." CW10 also stated that the EAC reports were distributed to Anderson and Turner.1 At the pleading stage, CW10's statements are sufficient to demonstrate that: (1) Spirit reviewed actual versus forecast costs on a quarter-by-quarter basis; (2) the 787 program was not meeting quarterly forecasts; and (3) Turner and Anderson received the quarterly reports indicating the 787 program was not tracking the forecast cost curve.

Despite receiving this information, Turner and Anderson repeatedly informed investors that 787 program costs were declining as planned. On a November 3, 2011 conference call, Turner stated that Spirit was "tracking per plan to identify and implement cost improvements" on the 787 project. On the same call, Anderson stated that "the curves" for the project were "on plan." In March 2012, Anderson stated at an investor conference that the program was "very much on plan and where we expected to be." After releasing the company's first quarter 2012 results, Turner again told investors that the company "continue[s] to track per plan to identify and implement cost improvements" on the 787 project, and Anderson stated that the project was "on plan as of the end of the first quarter here in 2012." On August 2, 2012, Turner publicly stated that "we continue to improve our unit cost performance, as planned." And on August 7, 2012, Turner discussed with investors the 787's cost curve, stating that the company expected to "continue to track that curve" and the 787 project was "going to keep coming down the curve."

Plaintiffs have alleged specific facts demonstrating that these statements were false at the time they were made. Turner's final statement, in particular, cannot be dismissed as mere optimism. He specifically identified the 787 cost curve and stated that company expected to "continue" to track it. Thus, he assured investors that Spirit had been and was presently on the curve. Spirit's internal reports indicated that the 787 program was not in fact on the planned cost curve.2 And as revealed in subsequent Securities and Exchange Commission filings, Spirit implemented a cost recovery plan in late July 2012 to attempt to bring the 787 program back in line with forecasts. That the company had just implemented a program to attempt to regain the curve renders the statement that it was going to "continue to track that curve" demonstrably false.

Having established falsity, the plaintiffs must demonstrate materiality. A statement is material "if a reasonable investor would consider it important in determining whether to buy or sell stock." Level 3, 667 F.3d at 1339 (quotation omitted). We distinguish between statements that are "material and those that are mere puffing not capable of objective verification." Id. (quotation and ellipses omitted). Defendants do not contest that statements as to a project's progress along cost projections would have been considered important by traders. However, the district court determined that none of the statements alleged in the complaint are capable of objective verification. I disagree.

Through its EAC process, Spirit internally measured whether the projects were meeting quarter-by-quarter cost forecasts. Whether the projects were "on track" or "on plan" was objectively verifiable each quarter. And Spirit's argument that executives' statements did not communicate anything about timing, costs, or other objective metrics is unconvincing. Unlike the "broad claims ... regarding integration efforts and the customer experience" identified in Level 3 as immaterial, 667 F.3d at 1340, Turner and Anderson specifically identified the 787's cost curve and unit costs as being presently in line with projections. And analysts understood the statements to mean that the company was operating with costs "at or below projected levels." Read in context, this inference is even more compelling. In August 2012 Turner stated both that the 787 program would "continue to track that curve," and that the Gulfstream G280 and G650 projects were "very close to meeting the curves that we placed but are not quite meeting them yet." These statements are not vague assurances about probable performance, they are objectively verifiable claims about a very specific internal measurement. Accord id. (statements that a project was "ahead of plan" and "under budget" had a "basis in objective and verifiable fact").

Finally, the plaintiffs must show that "the defendant acted with scienter, that is, with intent to defraud or recklessness." In re Zagg, 797 F.3d at 1200 (quotation and emphasis omitted). To plead scienter, plaintiffs must allege particularized facts that give rise to a strong inference that defendants acted intentionally or recklessly to mislead investors. 15 U.S.C. § 78u-4(b)(2); Kinder-Morgan, 340 F.3d at 1105. Recklessness is "conduct that is an extreme departure from the standards of ordinary care, and which presents a danger of misleading buyers or sellers that is either known to the defendant or is so obvious that the actor must have been aware of it." City of Philadelphia v. Fleming Cos., 264 F.3d 1245, 1260 (10th Cir. 2001) (quotations omitted). "[D]ivergence between internal reports and external statements on the same subject and disregard of the most current factual information before making statements can be factors supporting scienter." Level 3, 667 F.3d at 1345 (quotations omitted). And plaintiffs must show that "a reasonable person would deem the inference of scienter cogent and at least as compelling as any opposing inference." Tellabs, 551 U.S. at 324, 127 S.Ct. 2499 (2007). But the inference that the defendant acted with scienter does not need to be irrefutable. Level 3, 667 F.3d at 1343.

CW10 stated that the quarterly cost reports for each project — which incorporated the EACs during 2011 and 2012 showing cost overruns for the 787 program — were distributed to Turner and Anderson. Thus, taking the facts in the complaint as true, these defendants received quarterly reports demonstrating that the 787 project was not meeting the forecast cost curve. They nevertheless repeatedly informed investors that the project was proceeding along the cost curve as planned. Even assuming Turner and Anderson failed to read those reports and did not have actual knowledge of the cost overruns, it would be an extreme departure from the standards of ordinary care to leave project reports unopened while simultaneously assuring investors that the projects were on plan. The far more likely inference is that Turner and Anderson did review the reports before making the relevant statements. In that case, they intentionally lied to investors about the status of the 787 project.

This inference is further bolstered by the fact that the 787 program was important to Spirit's overall bottom line. See Reese v. Malone, 747 F.3d 557, 575 (9th Cir. 2014) (alleged false statements concerning core operations contribute to an inference of scienter). Although Spirit maintains that the core of its business was its manufacturing operations for existing Boeing models, the company acknowledged in February of 2012 that a "significant" portion of future revenues would be derived from new development programs, "most notably the [Boeing] 787." That project was expected to provide twenty percent of total revenue by 2014.

Although not dispositive, several other factors contribute to the inference of scienter as well. We have previously explained that the "magnitude of the alleged falsity" strengthens the inference of scienter. Kinder-Morgan, 340 F.3d at 1106. Spirit's cost overruns on the 787 project resulted in forward-losses of $184 million, approximately one quarter of reported earnings across 2010-2011. Although the "desire to protect one's own position" is not alone a sufficient basis to infer scienter, it was certainly present here. See Pirraglia v. Novell, Inc., 339 F.3d 1182, 1191 (10th Cir. 2003). By misleading investors about the status of the 787 project, Anderson and Turner sought to obtain additional time to correct their failures on that project. Individuals frequently attempt to "conceal[] bad news in the hope that it will be overtaken by good news ... like embezzling in the hope that winning at the track will enable the embezzled funds to be replaced before they are discovered to be missing." Makor Issues & Rights, Ltd. v. Tellabs Inc., 513 F.3d 702, 710 (7th Cir. 2008).

Considering "all the allegations holistically," Tellabs, 551 U.S. at 326, 127 S.Ct. 2499, I conclude the plaintiffs have demonstrated falsity, materiality, and scienter as to statements made by Turner and Anderson regarding the progress of the 787 project. The parties do not argue the remaining prongs of the § 10(b) test, and the district court did not consider them. Thus, I would reverse in part the district court's dismissal for failure to state a claim, and remand for further proceedings.