United States District Court, Northern District of Illinois, Eastern Division
March 12, 2015
Fernando Rossy, Individually and on Behalf of all Others Similarly Situated Plaintiff,
Merge Healthcare Inc., Michael W. Ferro, Jr., Jeffery A. Surges, Justin Dearborn and Steven M. Oreskovich, Defendants.
MEMORANDUM OPINION AND ORDER
Elaine E. Bucklo, United States District Judge
Plaintiff is a shareholder of defendant Merge Healthcare Inc., alleging, on behalf of itself and a class of others similarly situated, that Merge and several of its current and former executives committed securities fraud by materially overstating the company’s “subscription backlog, ” a financial metric defendants told investors was a reliable measure of future revenue, and by misrepresenting the effectiveness of the company’s internal controls to ensure the accuracy of its statements. Two motions to dismiss are before me. The first is by defendants Merge, Ferro, Oreskovich, and Dearborn, and the second is by defendant Surges. For the reasons discussed below, the motions are granted.
The following facts are drawn from the allegations in plaintiffs’ Consolidated Amended Complaint (“CAC”), which I assume to be true for present purposes, Tellabs, Inc. v. Makor Issues & Rights, Ltd., 551 U.S. 308, 322 (2007) (“Tellabs I”), and, where noted, from other materials properly before me, including Merge’s SEC filings, press releases, analyst reports, and public statements referenced in the CAC.
Defendant Merge is a healthcare information technology company that develops medical imaging, interoperability, and clinical systems for providers and consumers in the healthcare market. The individual defendants are current or former officers or directors of Merge. In May of 2012, Merge announced the creation of two operating groups: Merge Healthcare, dedicated to developing, selling, and implementing medical imaging and clinical solutions for providers, and Merge Data & Analytics (“Merge DNA”), focused on consumer solutions including a subscription-based clinical trials platform called eClinical OS.
The press release announcing these changes informed investors that due to increased client demand for subscription-based offerings, both operating units would transition towards a subscription-based business model and away from a model based on the sale of perpetual license agreements. The company explained that in addition to reporting its financial results in accordance with GAAP, it would use non-GAAP financial measures to report “non-recurring revenue backlog” (which the parties generally refer to as “subscription backlog” or simply “backlog”), defined as “revenue that we anticipate recognizing in future periods from signed customer contracts as of the end of the period presented.” May 7, 2012, Press Release, Def.’s Mot., Exh. 1, at 8. Merge acknowledged that the transition would result in short-term volatility in Merge’s revenues but emphasized that the new model would maximize transparency, predictability and profitability.
The CAC alleges that Merge’s reported subscription revenue backlog “skyrocketed” over the six quarterly periods ending June 30, 2012, through September 30, 2013. But in January of 2014, Merge announced that it was revising the figures it had previously reported for those periods after discovering that a former sales employee in Merge DNA’s eClinical business had falsified the existence or amount of certain customer contracts.
In a press release attached to a Form 8-K Merge filed on January 8, 2014, the company stated that based on the results of management’s internal investigation, Merge’s Board of Directors authorized an independent investigation by outside legal counsel and a forensic accounting firm. That investigation concluded that the former employee had falsified contracts with an apparent value of approximately $5.8 million and $9.4 million in 2012 and 2013, respectively, to “achieve sales quotas and receive additional commissions totaling approximately $250, 000.” January 8, 2014 Press Release, Def.’s Mot., Exh. 9 at 4. According to the press release, the independent investigation “did not find any evidence that other company employees had participated in, or were aware of, this improper conduct.” Id.
According to the CAC, however, defendants committed securities fraud in violation of Sections 10(b) and/or 20(a) of the Securities Exchange Act of 1934 because they knew or should have known that the subscription revenue backlog figures they originally reported were overstated due to the sales employee’s misconduct. Specifically, plaintiff identifies seventeen SEC filings, press releases, and conference calls made during the putative Class Period (August 1, 2012 to January 7, 2014) and alleges that defendants knew or recklessly ignored the risk that each contained materially false and misleading statement about Merge’s subscription revenue backlog (the “Backlog Statements”) and about the adequacy of Merge’s internal controls (the “Controls Certifications”).
Section 10(b) of the Securities Exchange Act of 1934 prohibits the use of manipulation or deception in connection with the purchase or sale of a security, and SEC Rule 10b-5 forbids, among other things, that any person “make any untrue statement of a material fact” in that connection. City of Livonia Employees’ Retirement System and Local 295/Local 851 v. Boeing Co., 711 F.3d 754, 755-56 (7th Cir. 2013). A successful claim under these provisions requires plaintiffs to plead and prove “both the facts constituting the alleged violation, and the facts evidencing scienter, i.e., the defendant’s intention to deceive, manipulate, or defraud.” Tellabs I, 551 U.S. at 313 (internal quotation marks and citation omitted). Defendants’ challenge to the CAC focuses predominantly on the latter requirement.
To show scienter, plaintiffs must establish that the defendant “either knew the statement was false or was reckless in disregarding a substantial risk of its being false.” City of Livonia, 711 F.3d at 756. Recklessness, in this context, means “an extreme departure from the standards of ordinary care…to the extent that the danger was either known to the defendant or so obvious that the defendant must have been aware of it.” Id. (ellipses in original) (citing cases).
Section 20(a) of the Exchange Act provides that “[e]very person who, directly or indirectly, controls any person liable under any provision of this title or of any rule or regulation thereunder shall also be liable jointly and severally with and to the same extent as such controlled person….” 15 U.S.C.
§ 78t(a). To prevail on a section 20(a) claim, a plaintiff must establish a primary violation of section 10(b) and further show that the “control person” 1) actually exercised control over the operations of the entity principally liable, and 2) had the power or ability-regardless of whether it was exercised-to control the specific transaction or activity upon which the primary violation was predicated. Donohoe v. Consolidated Operating & Production Corp., 982 F.2d 1130, 1138 (7th Cir. 1992) (citing Harrison v. Dean Witter Reynolds, Inc., 974 F.2d 873, 880-81 (7th Cir. 1992)). A section 20(a) violation does not require proof of scienter. Jones v. Corus Bankshares, Inc., 701 F.Supp.2d 1014, 1030 (N.D. Ill. 2010). But because a primary violation under section 10(b) and Rule 10b-5 does require scienter, it follows that a plaintiff must establish scienter on the part of at least one controlled defendant to prevail on a claim under section 20(a).
At the pleading stage, plaintiffs must comply not only with the particularity requirements of Rule 9(b), but also with the “[e]xacting pleading requirements” of the Private Securities Litigation Reform Act, 15 U.S.C.
§ 78u-4 (“PSLRA”). Allegations supporting scienter thus must “state with particularity facts giving rise to a strong inference that the defendant acted with the required state of mind.” City of Livonia, 711 F.3d at 756-57 (original emphasis). This means that, when viewing the complaint as a whole, “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. at 756 (quoting Tellabs I 551 U.S. at 324)).
Defendants devote the bulk of their respective memoranda to the argument that the CAC fails to satisfy the PSLRA’s pleading standards with respect to scienter. Although they identify other plausible grounds for dismissal as well, my analysis begins and ends with this one because it is dispositive of each of the CAC’s claims.
At the outset, plaintiff concedes that it has not alleged any defendant’s actual knowledge of the falsity of any challenged statement, and that its theory of scienter is limited to recklessness. Pl.’s Opp. at 13. The parties’ dispute thus boils down to whether the CAC, taken as a whole, raises a cogent inference, at least as compelling as any opposing inference, that defendants deliberately ignored a substantial risk that any of the Backlog Statements or Controls Certifications was false.
Defendants assail plaintiff’s allegations with a battery of arguments on this front. They argue: 1) that the CAC’s repeated iteration of “boilerplate mantras” fails to specify what facts any individual defendant is alleged to have that would have alerted him to the risk that any of the Backlog Statements or Controls Certifications was false; 2) that recklessness cannot be inferred from the discovery and revelation of an employee’s misconduct, or from Merge’s subsequent improvements to its internal controls, since there is no “fraud by hindsight”; 3) that allegations that the individual defendants were among Merge’s “senior management” do not adequately plead recklessness because they do not assert these individuals’ personal involvement in, or oversight of, the employee’s fraudulent transactions that resulted in Merge’s false backlog statements; 4) that the allegations based on information obtained from confidential witnesses are not only inherently unreliable but also substantively insufficient to raise a compelling inference of recklessness; and 5) that the CAC fails to suggest any motive as to any defendant. Finally, defendants argue that the most plausible and compelling inference raised by plaintiff’s allegations is that defendants themselves were defrauded by a rogue employee, rather than that defendants participated in securities fraud.
These arguments, on the whole, are persuasive. Not all of them require lengthy discussion-indeed, plaintiff concedes some of defendants’ arguments, such as that neither the fact of the former employee’s misconduct, nor Merge’s subsequent changes to its internal controls, supports an inference of scienter. Accordingly, I focus on the contested arguments, emphasizing those that relate to the CAC’s central defects.
As defendants correctly observe, the CAC contains numerous “boilerplate” allegations set forth in a formulaic manner. For example, the section captioned “Defendants’ False and Misleading Statements” identifies seventeen dates on which defendants made one or more public statements about Merge’s subscription backlog. This section is organized by date, and in the three to five paragraphs devoted to each, identifies the documents filed or calls made on that date, then asserts that each contained statements that were materially false and misleading because:
Merge DNA’s publicly reported subscription revenue backlog figures were fraudulently and materially overstated throughout the Class Period. Specifically, because of the reckless lack of proper internal controls to monitor and oversee the input of new contracts and backlog figures into the Merge computer system, a former sales employee in its eClinical business was able to easily falsify “the existence or amount of certain customer contracts.”
CAC at ¶¶ 81, 85, 90, 94, 96, 101, 106, 108, 113, 116, 122, 125, 128, 131, 136, 139, and 143. The CAC likewise repeats verbatim allegations that each Controls Certification was false and misleading because:
Merge’s internal controls were grossly ineffective and routinely overridden by Defendants during the Class Period. Specifically, CW 1 stated that eClinical’s senior management pressured sales employees to engage in aggressive sales tactics, which led to commissions being paid on contracts that were ultimately cancelled. Similarly, CW 2 stated that Defendant Dearborn, as well as other senior management, instructed eClinical’s accounting personnel to disregard the Company’s internal guidelines regarding the recording of revenues and the payment of sales commissions. Finally, CW 3 stated that eClinical’s senior management failed to address known material weaknesses in eClinical’s processes for recording customer contracts and paying commissions, such that eClinical sales employees could collect commissions on unfulfilled and/or falsified contracts.
Id. at ¶ 87, 98, 110, 118, 133, and 145.
I agree with defendants that these allegations do not contribute to a strong inference of any defendant’s scienter. With the exception of defendant Dearborn (whom I address separately below), they attribute no specific knowledge or conduct to any defendant that suggests he deliberately ignored an obvious risk that any of his statements was false. “[R]ote conclusions” such as these, which fail to differentiate among multiple defendants, do not raise a compelling inference of scienter as to any of them, nor do they satisfy the particularity requirements of Rule 9(b). See DiLeo v. Ernst & Young, 901 F.3d 624, 629 (7th Cir. 1990) (“rote conclusions” and “[b]oilerplate” allegations insufficient to plead scienter); Last Atlantis Capital LLC v. Chicago Bd. Options Exchange, Inc., No. 04 C 397, 2005 WL 3763262, *4 (N.D. Il. Mar. 30, 2005) (“lumping all defendants into a single group” fails “to provide the ‘who’ required by Rule 9(b) and the PSLRA”).
Plaintiff concedes that the CAC “does, at times, refer to ‘Defendants’ collectively, ” but insists that its carbon-copy allegations nevertheless contribute to a strong inference of scienter “because the Court may scrutinize such allegations with respect to each of the Individual Defendants.” Pl.’s Opp. at 20. In this connection, plaintiff directs me to ¶¶ 22-65 of the CAC for “detailed, non-boilerplate allegations establishing Defendants’ recklessness.” Id. at n. 13. Yet, the section to which plaintiff points is likewise replete with generic allegations that fail to differentiate among the defendants. See, e.g. CAC at ¶ 54 (“Defendants were alerted to, and/or should have been aware of, material weaknesses in the Company’s internal control processes for recording customer contracts and paying sales commissions…”), ¶ 55 (“Defendants and other senior management in Merge’s eClinical division knew and/or should have known of material weaknesses in Merge’s internal controls, which severely undermined the reliability and predictability of [the subscription backlog] metric”), and ¶ 56 (“Throughout the Class Period, Defendants aggressively sought to increase bookings for Merge’s subscription-based offerings at any cost, including the Company’s internal control processes.”).
Moreover, while some of the cited paragraphs do include non-boilerplate details, none offers sufficient facts to support a strong inference that any defendant disregarded a substantial risk that his statements were false. For instance, paragraphs 60-62, which exemplify some of the CAC’s most detailed allegations, allege that CW 3 “noted critical flaws in eClinical’s internal control processes, ” one of which he brought to El-Assi’s attention:
Specifically, CW 3 explained that under eCOS’s “pay-as-you-go” pricing model, eClinical sales employees negotiated a Master Service Agreement (“MSA”) with the customer, which was usually signed electronically by the customer and then sent to Merge’s Contracts and Proposal Department and to El-Assi, for countersigning. However, the signing of the MSA had no financial implications for the customer. Instead, when the customer wanted to use eCOS for a clinical study, the customer completed and signed a “Configurator Quote, ” which would determine the pricing for the specific study. Unlike the MSA, the Configurator Quote was not sent to the Contracts and Proposal Department, nor was it counter-signed by El-Assi. Rather, the sales employee was the only individual at Merge who oversaw the completion and signing of the Configurator Quote.
The signing of the Configurator Quote would also allow the sales employee to collect commissions on his or her “sale, ” even though the customer had not yet been billed for any services. Only when the customer “went live” with the study did billing begin. Other than the Contract and Proposal Department’s review and El-Assi’s counter-signing of the MSA, there was no other oversight over the contracting process.
According to CW 3, the gaps in Merge’s internal controls could be manipulated to allow sales employees to collect unearned commissions in one of two possible ways. First, a sales employee could collect commissions by asking a customer to submit a Configurator Quote while assuring the customer that it would not be billed unless it ‘went live’ with the study. CW 3 noticed this glaring flaw within a few months of working at Merge and specifically brought it to El-Assi’s attention during the Class Period.
However, El-Assi’s response was dismissive. Second, a sales employee could falsify signed contracts by attaching the customer’s electronic signature, which he or she would have access to by virtue of the signed MSA, to an illegitimate Configurator Quote. Under this scenario, the customer would never become aware of the falsified Configurator Quote because it would only be invoiced when the study “went live, ” which would never occur. Nevertheless, under either scenario, the sales employee would earn a commission on the “sale, ” and the “sale” would be incorporated into Merge’s subscription revenue backlog.
Id. at ¶¶ 60-62.
These allegations do not raise a strong inference of scienter for several reasons. First, as the Seventh Circuit has repeatedly admonished, information obtained from confidential witnesses is generally subject to a “discount” that it has alternately characterized as “steep” and “heavy.” Higginbotham v. Baxter Intern., Inc., 495 F.3d 753, 756-57 (7th Cir. 2007) (discount afforded allegations based on confidential sources is “usually” steep); City of Livonia, 711 F.3d at 759 (“[a]llegations concerning…unnamed confidential sources of damaging information require a heavy discount.”). As the court explained in Higginbotham, “[i]t is hard to see how information from anonymous sources could be deemed ‘compelling’ or how we could take account of plausible opposing inferences. Perhaps these confidential sources have axes to grind. Perhaps they are lying. Perhaps they don’t even exist.” 495 F.3d at 757.
Furthermore, even assuming that CW 3’s statements are truthful and accurate, they reveal nothing about any defendant’s knowledge or state of mind. Nothing in the CAC suggests that the internal control flaws CW 3 described to El-Assi were ever communicated to any individual defendant. Indeed, by plaintiff’s own account, El-Assi was “dismissive” of CW 3’s concerns, undermining any inference that El-Assi thought enough of them to alert other members of Merge’s senior management. Nor do the CAC’s remaining allegations suggest that defendants were aware of the flaws CW 3 described. For at least these reasons, CW 3’s statements are unlike those the court found contributed to an inference of scienter in Tellabs II and Silverman v. Motorola, Inc., No. 07 C 4507, 2008 WL 4360648 (N.D. Ill. Sept. 23, 2008) (Moran, J.).
In Tellabs II, the issue was whether the defendants intended to deceive investors with statements that sales for its flagship product were “still going strong” with no “weakness at all in demand, ” while in reality, the market for the product was “evaporating, ” and the defendant corporation “had been flooding its customers with tens of millions of dollars of [the product] that the customers had not requested, in order to create an illusion of demand.” 513 F.3d at 706-07. The plaintiffs also challenged statements that the defendants’ customers were “embracing” their new product, for which the company was “satisfying a very strong demand, ” including through a multiyear, $100 million contract, when in fact, there had been no sales or shipments of the new product at all. Id. at 706, 709. The court acknowledged that “allegations based on anonymous informants are very difficult to assess, ” but found that the information obtained from numerous confidential sources who knew, first hand, information such as “that sales of [the flagship product] were dropping off a cliff while the company pretended that demand was strong, ” and that the new product “was still in the beta stage and failing performance tests conducted by prospective customers, ” corroborated other allegations tending to show that it was “exceedingly unlikely” that the defendants innocently, or merely carelessly, misrepresented the demand for the company’s two most important products. Id. at 711-12.
In Silverman, the challenged statements related to the defendant company’s ability to meet production and shipping deadlines for several new products. In September of 2006, the corporation’s CEO answered a question about the products’ availability, confirming that the products “will ship in Q4. So, I was told that all our products are going to ship.” Silverman, 2008 WL 43606488 at *4. He prefaced this answer by affirming, “every week I have a staff meeting, every week I have a call, how are we doing?” Id. According to multiple confidential witnesses, by the time the CEO made these statements, it was widely known within the company that the products were plagued with design defects creating “quality issues resulting in poor manufacturing yields, ” and that although defendant’s engineers “worked extremely long hours during the second half of 2006 to resolve the quality issues, ” the product would not be available “in time for critical fourth quarter sales.” Id. at *5. These witnesses also described multiple software and hardware problems delaying the products’ launch, while others stated that the company’s engineering directors held weekly meetings with “various company department vice-presidents to inform them about the status of the project.” Id. at *6. Citing Tellabs II, the court concluded that although confidential witness statements are generally disfavored, the detailed confidential witness statements reflecting widely known delays affecting the products’ launch date and corroborating the CEO’s own statement regarding “weekly” product updates, “could help generate a strong inference of scienter.” Id. at *14.
The allegations attributed to the four confidential witnesses in this case are far less compelling. Most are wholly lacking in detail, see, e.g., CAC at ¶¶ 98, 110, 118, 133, 145 (all alleging that CW 1 “stated that eClinical’s senior management pressured sales employees to engage in aggressive sales tactics, which led to commissions being paid on contracts that were ultimately cancelled”); ¶¶ 58, 64 (citing CW 2’s statements that “sales employees offered products to customers that were not ready to be delivered” and that “shady stuff” happened with respect to revenue recognition); ¶ 64 (alleging that CW 4 called Merge “a mess of an organization” and said El-Assi and Rooney “believed in closing a sale at all costs”), and do not reference any defendant or allege any specific information known to any of them. Nor do they meaningfully corroborate one another, except to the extent that several note aggressive sales practices by unnamed or non-party members of Merge’s senior management. As plaintiff’s own authority confirms, “general statements such as [these], by themselves, are not enough to create a ‘strong inference’ of scienter.” Silverman 2008 WL 4360648, at *14.
The only confidential witness whose statements expressly implicate any individual defendant is CW 2, described as a Senior Accountant in eClinical between October 2010 and August 2012. According to CW 2, despite eClinical’s “internal guidelines, ” pursuant to which “the accounting department was not supposed to recognize revenue or pay sales commissions until it received a signed contract, ” CW 2 would “often receive instructions either verbally or via e-mail directly from Defendant Dearborn or El-Assi to recognize revenue and pay sales commission before receiving a signed contract.” (Original emphasis) CAC at ¶ 64. In most such instances, Dearborn or El-Assi simply told CW 2, “don’t worry about it, ” and that “the contract would be signed.” Id.
These allegations, however, do not raise a strong inference that Dearborn recklessly disregarded a substantial risk that any statement defendants made during the Class Period was false. Even setting aside that CW 2’s employment with Merge ended during the first month of the Class Period, his statements suggest, at most, that on unspecified dates and with respect to unspecified contracts, Dearborn knowingly departed from an internal accounting guideline. CW 2 does not allege that Merge ultimately failed to receive a signed contract on any of these occasions, nor that Dearborn knew or suspected it might not. In fact, CW 2’s statement suggests the opposite-that Dearborn expected the contracts would ultimately be signed.
Accordingly, the conduct CW 2 attributes to Dearborn does not bear the indicia of “reckless behavior, ” the definition of which “should not be a liberal one, ” and “comes closer to being a lesser form of intent than merely a greater degree of ordinary negligence.” Sanders v. John Nuveen & Co., Inc., 554 F.2d 790, 793 (7th Cir. 1977). Even assuming that adherence to eClinical’s “internal guideline” might have averted the sales employee’s fraud, and that Dearborn was negligent-or even grossly negligent-for failing to anticipate that departure from this guideline opened a loophole for unscrupulous sales people, neither CW 2’s statement, nor anything else in the CAC, raises a strong inference that Dearborn intended that result. The same is true of the conduct CW 2 attributes to El-Assi and Rooney, eviscerating plaintiff’s argument that the putative scienter of these non-parties also supports a 10(b) claim against Merge.
In short, even plaintiff’s non-boilerplate allegations based on information obtained from confidential witnesses are insufficient, individually and collectively, to raise a strong inference that any defendant deliberately ignored a substantial risk that any of the Backlog Statements or Controls Certifications was false.
Once the confidential witness statements are discarded, little of substance remains of the CAC. Allegations, for example, that defendants were high-ranking officers and directors of Merge merit scant consideration. While plaintiff is correct that these facts are not “irrelevant” to the scienter analysis, even plaintiff’s authority acknowledges that “[a]n individual’s position as a corporate officer, without more, is generally insufficient to support a strong inference of scienter.” Jones v. Corus Bankshares, Inc., 701 F.Supp.2d 1014, 1029 (N.D. Ill. 2010). As discussed above, the CAC does not identify any red flags of which the individual defendants were aware, nor does it otherwise contain allegations of a kind that would make it “almost inconceivable that an individual defendant would be unaware of the matters at issue.” Id. (citation omitted).
The most plausible story that emerges from the CAC and other materials of record is that defendants themselves were hoodwinked by an unscrupulous employee who, unbeknownst to management, falsified customer invoices for the purpose of inflating his own sales commissions. While initially praising the employee for his sales results, El-Assi “became suspicious” by June of 2013 and sent an email to the employee “to inquire about a sale that had not closed.” CAC at ¶¶59, 65. By January 8, 2014, Merge announced that it had conducted both internal and independent investigations into the matter, and corrected its previously announced backlog figures. Id. at ¶ 77; Jan. 8 Press Release, Def.’s Mot., Exh. 9.
The natural inference from these allegations is not that Merge’s senior executives encouraged Merge’s sales employees to exploit a known weakness in the company’s internal controls in order to beef up the company’s backlog, but instead that management congratulated them for seemingly impressive sales results, but investigated after becoming “suspicious” that the results were not genuine and took measures to correct the problem. Nor does the six-month lapse plaintiff decries between El-Assi’s first inkling of the rogue employee’s shenanigans and Merge's restatement of its backlog totals detract from this story. "Knowing enough to launch an investigation...is a very great distance from convincing proof of intent to deceive." Higginbotham, 495 F.3d at 758. Even setting aside that the complaint does not allege when any defendant began to suspect the employee of wrongdoing, as in Higginbotham, the conduct plaintiff attributes to Merge's management more readily suggest "a pursuit of truth [rather] than reckless indifference to the truth." Id.
For the foregoing reasons, plaintiff's allegations of scienter do not meet the strict pleading standards of the PSLRA, dooming plaintiff's 10(b) claim and with it, plaintiff's 20(a) claim. See Donohoe v. Consolidated Operating & Production Corp., 982 F.2d 1130, 1138 (7th Cir. 1992) ("control person" liability attaches to person with power or ability to control specific transaction or activity on which primary violation was predicated). Accordingly, defendants' motions to dismiss are granted.