Staehr v. the Hartford Financial Services Group, Inc.

06-3877-cv Staehr v. The Hartford Financial Services Group, Inc. UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT August Term, 2007 _________________ (Argued: June 19, 2008 Decided: November 17, 2008) Docket No. 06-3877-cv _________________ STEVE STAEHR, on behalf of all others similarly Situated, Alaska Laborers Employers Retirement Fund and The Communication Workers of America Plan for Employees’ Pensions and Death Benefits, Plaintiffs-Appellants, Kevin Montoya, Individually and on Behalf of All Others Similarly Situated, Philip Stampfel, on behalf of Hartford Financial Services Group Inc., Consolidated-Plaintiffs, David Wexler, d/o/b Hartford Financial Services Group Inc., Consolidated-Plaintiff-Appellant, – v. – THE HARTFORD FINANCIAL SERVICES GROUP, INC., Thomas M. Marra, Ramani Ayer, David K. Zwiener and David M. Johnson, Defendants-Appellees, Edward J. Kelly III, Paul G. Kirk, Jr., Gail J. McGovern, Robert W. Selander, Charles B. Strauss, H. Patrick Swygert, Gordon I. Ulmer, Ronald E. Ferguson, Rand V. Araskog, Donald R. Frahm, and Robert J. Price, Consolidated-Defendants-Appellees. _____________________________________________________x William S. Lerach Eric Alan Isaac Tor Gronborg Debra Wyman Tami Falkenstein Hennick, LERACH COUGHLIN STOIA GELLER RUDMAN ROBBINS LLP Counsel for Appellants David R. Scott Erin Green Comite SCOTT & SCOTT LLP Counsel for Appellants Jack C. Auspitz, Jamie A. Levitt John W.R. Murray MORRISON & FORRESTER LLP Counsel for Appellees Timothy A. Diemand WIGGIN & DANA LLP Counsel for Appellees BEFORE: HON. PETER W. HALL, HON. DEBRA ANN LIVINGSTON, Circuit Judges, HON. COLLEEN MCMAHON, District Judge. ∗ ∗ The Honorable Colleen McMahon, United States District Court Judge for the Southern District of New York, sitting by designation. 2 1 MCMAHON, District Judge: 2 3 This case is one of many stemming from the so-called “contingent commission” 4 arrangements between insurers and brokers that were prevalent prior to October 2004. 5 “Contingent commissions” is a euphemism for kickbacks— insurance brokers would receive 6 payments from insurers for steering business their way. 7 Appellants purport to represent all persons who acquired common stock of The Hartford 8 Financial Services Group, Inc. (“The Hartford”) during the period of August 6, 2003 through 9 October 13, 2004 (the “Class Period”). (Joint Appendix (“J.A.”) at 2.) They bring this action 10 against The Hartford and its senior officers (collectively, “Appellees”), alleging that investors 11 acquired The Hartford’s stock at prices that were artificially inflated due to Appellees’ 12 omissions, misrepresentations, and fraudulent concealment regarding kickbacks, bid rigging, and 13 price manipulation schemes engaged in by insurers and brokers. 14 The District Court (Droney, J.) granted Appellees’ motion to dismiss the complaint as 15 barred by the statute of limitations. Taking judicial notice of materials that purportedly 16 constituted “storm warnings” about The Hartford’s alleged fraud, see, e.g., Dodds v. Cigna 17 Securities, Inc., 12 F.3d 346, 350 (2d Cir. 1993), the District Court concluded that Appellants 18 were on inquiry notice of the fraud no later than July 25, 2001. Since the two-year statute of 19 limitations began to run on that date, it expired in July 2003— more than a year before the suit 20 was filed. 21 We disagree with the District Court’s conclusion that Appellants were on inquiry notice 22 of their claims against The Hartford by July 25, 2001. Accordingly, we vacate the judgment and 23 remand the case to the District Court for further proceedings. 24 1 1 2 I. BACKGROUND 3 4 A 5 6 The Hartford is a large insurer in the property-casualty and life insurance industries. 7 Through its subsidiaries, The Hartford markets and sells investment products and insurance to 8 individuals and businesses. 9 Commercial insurance brokers act as intermediaries between insurance companies and 10 their policyholders. These brokers are hired by clients to assist them in soliciting price quotes, 11 recommending insurers, and purchasing insurance products. After a period of consolidation in 12 the insurance brokerage industry, Marsh, Inc. (“Marsh”) and Aon Corporation (“Aon”) emerged 13 as the dominant brokers, controlling more than seventy percent of the market by 2003. (J.A. 14 11.) 1 15 Appellants allege that The Hartford entered into contingent commission kickback 16 arrangements with insurance brokers in order to increase The Hartford’s market share and 17 artificially inflate its insurance prices. According to Appellants, The Hartford paid brokers fees 18 based on: the volume of the premiums the brokers steered to The Hartford; the growth and 19 retention of business; and the profitability of the products purchased by the brokers’ clients. (J.A. 20 11-14.) 21 The contingent commission/kickback arrangements, Appellants allege, were 22 memorialized in “placement service agreements” or “market service agreements” between The 23 Hartford and insurance brokers who steered business its way. These agreements allegedly bound 1 Much of the factual background is based upon allegations in the Complaint. For purposes of evaluating a motion to dismiss pursuant to Federal Rule of Civil Procedure 12(b)(6), this Court sets forth facts asserted in the Complaint and assumes their truth. Spool v. World Child Int’l Adoption Agency, 520 F.3d 178, 180 (2d Cir. 2008); Shah v. Meeker, 435 F.3d 244, 246 (2d Cir. 2006). 2 1 The Hartford to make payments that could amount to $150 million per year, creating a conflict of 2 interest by motivating brokers to serve the interests of the insurers – with whom they had 3 kickback arrangements – rather than to serve the interest of their clients. (J.A. 12-13.) 4 Insurers, including The Hartford, also allegedly engaged in “bid rigging” with insurance 5 brokers, to eliminate competition and artificially inflate the price of insurance products. 6 Appellants contend that throughout the Class Period, The Hartford and its senior officers 7 misled investors by failing to disclose The Hartford’s participation in the insurer-broker 8 contingent commission kickback and bid-rigging schemes. (J.A. 13-16.) 9 On October 14, 2004, the Office of the New York Attorney General (“NYAG”) filed a 10 lawsuit against Marsh detailing “the undisclosed commission pay-offs and bid-rigging schemes 11 that a cartel of large insurers and insurance brokers were using to prop up their businesses and 12 cause customers to purchase insurance at higher prices and less favorable terms than the market 13 otherwise would have dictated.” (J.A. 2-3.) The complaint cited The Hartford for its role in 14 paying undisclosed contingent commissions and providing inflated bids. (J.A. 3.) The NYAG 15 lawsuit, and The Hartford’s purported involvement in the practices the lawsuit alleged, were the 16 subject of widespread media attention; Eliot Spitzer, then Attorney General of New York, 17 commented publicly that The Hartford was “involved in the scheme” and that “the corruption is 18 remarkable.” (J.A. 32.) 19 After the NYAG lawsuit was filed, The Hartford disclosed that it had paid $145 million 20 in kickbacks to brokers in 2003 alone. It announced that it had ceased paying contingency 21 commissions in the wake of the suit. (J.A. 32-33.) 3 1 The day after the NYAG announcement, Appellant Steve Staehr filed his original 2 complaint in this action on behalf of a putative class of shareholders. Several similar complaints 3 followed. (J.A. 819-35.) 4 On February 10, 2005, the District Court issued an order consolidating those complaints 5 and appointing Alaska Laborers Employers Retirement Fund and the Communication Workers 6 of America Plan for Employees’ Pensions and Death Benefits as Co-Lead Plaintiffs. (J.A. 828.) 7 Appellants filed their Consolidated Amended Complaint on April 12, 2005 (the 8 “Complaint”). On June 10, 2005, Appellees moved to dismiss the Complaint for failure to state 9 a claim on multiple grounds, including (1) the statute of limitations had run; (2) Appellees’ 10 alleged omissions were immaterial; (3) loss causation could not be established; (4) the Complaint 11 failed to allege scienter; and (5) Appellants could not establish control person liability against the 12 individual defendants. Staehr v. Hartford Fin. Svcs. Group, Inc., et al, 460 F. Supp.2d 329 (D. 13 Conn. 2006.) Appellees submitted various materials along with their motion to dismiss— 14 materials that were not attached to or specifically referenced in the Complaint. These exhibits 15 include various reports from the mainstream media and from insurance industry newsletters, 16 complaints filed in four lawsuits, and filings made by The Hartford with state regulatory 17 agencies. Appellees argue that, when viewed together, these exhibits illustrated the public nature 18 of the alleged fraud and should have put Appellants on inquiry notice far earlier than two years 19 prior to the commencement of this action. 20 Appellants opposed the motion and cross-moved to strike exhibits that were neither 21 referenced in nor attached to the Complaint, or, alternatively, to convert the motion to dismiss 22 into a motion for summary judgment. (J.A. 613-15, 630-77.) 4 1 On July 13, 2006, the District Court issued an opinion and order granting Appellees’ 2 motion to dismiss. Staehr v. Hartford Fin. Svcs. Group, Inc., 460 F. Supp. 2d 329 (D. Conn. 3 2006). The District Court took judicial notice of the exhibits introduced into the record by 4 Appellees pursuant to Federal Rule of Evidence 201. Id. at 334. The District Court emphasized 5 that it did not take judicial notice of the documents for the truth of the matters asserted, and it 6 denied Appellants’ request to convert the motion to dismiss into a motion for summary 7 judgment. Id. at 335. Addressing the merits of the motion, the District Court held that 8 `Appellants’ suit was time-barred by the applicable two-year statute of limitations. Id. at 340. 9 The District Court found it unnecessary to rule on any of the other grounds for dismissal urged 10 by Appellees. Staehr, 460 F. Supp.2d at 340. 11 Appellants timely filed their Notice of Appeal on August 17, 2006. (J.A. 814.) 12 13 B 14 15 The crux of Appellants’ claims is that The Hartford’s shareholders were misled because 16 they believed they were investing in a company whose success (and concomitant high stock 17 price) was premised on the strength of its business, when it was in fact the product of kickbacks 18 and bid-rigging schemes that were not disclosed to investors. The Complaint alleges that 19 Appellees’ statements: 20 failed to disclose that defendants had guaranteed tens of millions of 21 dollars in commission-oriented kickbacks to insurance brokers . . . 22 in exchange for these brokers illicitly steering business to Hartford 23 and artificially inflating the premiums paid by insurance 24 customers. [D]efendants were able to report earned premium 25 growth and high premium renewal retention rates directly as a 26 result of Hartford’s involvement with the brokers in the 27 commission kickbacks and bid-rigging schemes. (J.A. 16-17 28 (internal citations omitted).) 29 5 1 In other words, The Hartford’s shareholders believed that The Hartford’s success was 2 premised on the company’s unique business model and favorable market conditions, not on the 3 company’s participation in a conspiracy by insurers and brokers to manipulate the insurance 4 industry and inflate the financial performance of its participants. 5 The bid-rigging scheme, according to Appellants, eliminated competition, artificially 6 inflated the price of insurance, guaranteed higher earned premiums for The Hartford, produced 7 larger kickbacks for the brokers, and resulted in high customer retention rates (which is an 8 important measure of an insurance company’s success). (J.A. 13-14.) The pleading alleged that 9 Appellees regularly released reports touting The Hartford’s growth and business success without 10 disclosing the role that kickbacks and bid rigging played in this success. Instead, Appellees 11 stated that The Hartford’s growth and increased income were the result of quality products and 12 “[f]avorable market conditions.” (J.A. 24.) Appellees told investors: “We continue to 13 consistently deliver great performance because of the strength of our business model,” and “our 14 execution has been great.” (J.A. 19, 30.) 15 Count 1 of the complaint alleges that The Hartford and the individual defendants violated 16 § 10(b) of the Securities Exchange Act of 1934 (“1934 Act” or “Exchange Act”) and Securities 17 Exchange Commission (“SEC”) Rule 10b-5 by engaging in undisclosed contingent commission 18 and bid-rigging schemes with insurance brokers for the purpose of maintaining artificially high 19 market prices for The Hartford’s publicly traded securities. (J.A. 48-51.) Count 2 alleges that 20 the individual defendants also violated § 20(a) of the 1934 Act because they acted as “controlling 21 persons” of The Hartford within the meaning of the Act. (J.A. 12-13, 51-52.) 22 23 C 24 6 1 Since July 30, 2002, a private federal action for securities fraud must be commenced 2 before the earlier of “2 years after the discovery of the facts constituting the violation” or “5 3 years after such violation.” 28 U.S.C. § 1658(b). See Sarbanes-Oxley Act of 2002, Pub. L. No. 4 107-204, § 804, 116 Stat. 745, 801 (2002) (codified at 28 U.S.C. § 1658 & note); Shah v. 5 Meeker, 435 F.3d 244, 248–49 & n.3 (2d Cir. 2006) (citing Exchange Act § 9(e), 15 U.S.C. § 6 78i(e)). 7 The two-year statute of limitations for securities fraud claims under the Exchange Act 8 begins to run only after the plaintiff “obtains actual knowledge of the facts giving rise to the 9 action or notice of the facts, which in the exercise of reasonable diligence, would have led to 10 actual knowledge.’” LC Capital Partners, LP v. Frontier Ins. Group, Inc., 318 F.3d 148, 154 (2d 11 Cir. 2003) (quoting Kahn v. Kohlberg, Kravis, Roberts & Co., 970 F.2d 1030, 1042 (2d 12 Cir.1992)) (emphasis added). When there is no actual knowledge, but “the circumstances would 13 suggest to an investor of ordinary intelligence the probability that she has been defrauded, a duty 14 of inquiry arises, and knowledge will be imputed to the investor who does not make such an 15 inquiry.” Dodds v. Cigna Sec., 12 F.3d 346, 350 (2d Cir. 1993). “Such circumstances are often 16 analogized to ‘storm warnings.’” Id. 17 We have recognized in the past that determining “whether a plaintiff had sufficient facts 18 to place it on inquiry notice is ‘often inappropriate for resolution on a motion to dismiss. . . .’” 19 LC Capital Partners, 318 F.3d at 156 (quoting Marks v. CDW Computer Ctrs., Inc., 122 F.3d 20 363, 367 (7th Cir. 1997)). However, we have also stated that courts can “readily resolve the 21 issue” of inquiry notice as a matter of law on a motion to dismiss – as has been done in “a vast 22 number of cases” in this circuit – where “the facts needed for determination of when a reasonable 23 investor of ordinary intelligence would have been aware of the existence of fraud can be gleaned 7 1 from the complaint and papers . . . integral to the complaint.” Lentell v. Merrill Lynch & Co., 396 2 F.3d 161, 168 (2d Cir. 2005) (internal quotation marks omitted); see also LC Capital Partners, 3 318 F.3d at 156; Dodds, 12 F.3d at 352 n.3 (citing In re Integrated Res. Real Estate Ltd. P’ships 4 Sec. Litig., 815 F. Supp. 620, 639 n.7 (S.D.N.Y. 1993) (citing cases)). 5 In granting Appellees’ motion, the District Court relied on the “storm warnings” doctrine 6 we have enunciated in recent years, beginning in Dodds, 12 F.3d 346. The plaintiff in Dodds 7 was an unsophisticated investor who made clear to the investment-adviser defendant that she 8 sought to pursue a conservative investment strategy. The defendant-adviser provided plaintiff 9 with a portfolio of prospectuses and other information about the financial products he 10 recommended. He assured the plaintiff that the investments he recommended (which included 11 several limited partnerships) were suitable. Almost a year later, the plaintiff met with an 12 accountant who told her that the investments in the limited partnerships were not suitable 13 investments for her. Plaintiff alleged that she was induced to invest in securities that were too 14 risky and illiquid for her. Id. at 347-50. 15 The district court dismissed the claims as time-barred, and we affirmed, stating that when 16 an investor was given prospectuses that disclosed the risky and illiquid character of certain 17 investments, she was placed on inquiry notice that such investments might be inappropriate for a 18 conservative portfolio. Id. at 347. The fact that the plaintiff failed to read the prospectuses was 19 irrelevant. The commissions, the risk, and the illiquidity of investments in the limited 20 partnerships were clearly disclosed in the prospectuses received by the plaintiff in Dodds. See 21 id. at 350. By receiving them, she had constructive notice of facts sufficient to create a duty to 22 inquire further into the appropriateness of the investments. “[W]hen the circumstances would 23 suggest to an investor of ordinary intelligence the probability that she has been defrauded, a duty 8 1 of inquiry arises, and knowledge will be imputed to the investor who does not make such an 2 inquiry.” Id.. 3 Since Dodds, we have addressed the issue of what constitutes sufficient “storm warnings” 4 to give rise to a duty to inquire in at least five published opinions. See LC Capital Partners 5 (Frontier II), 318 F.3d at 148; Newman v. Warnaco Group, Inc., 335 F.3d 187 (2d Cir. 2003); 6 Levitt v. Bear Stearns & Co., 340 F.3d 94 (2d Cir. 2003); Lentell, 396 F.3d at 161; Shah, 435 7 F.3d at 244. 8 The plaintiffs in LC Capital Partners alleged that defendant Frontier engaged in 9 negligent insurance practices, including under-reserving for claims. The defendants allegedly 10 concealed these practices by making false statements and material omissions in a series of 11 financial reports, press releases, and other public statements. LC Capital Partners, 318 F.3d at 12 150-52. 13 Frontier had taken several restructuring charges between 1994 and 1999. In press 14 releases, the company provided various benign explanations for these charges, such as business 15 expansion, or “adoption of a more conservative reserving policy.” See id. at 151. In December 16 1998, Frontier announced a $139 million charge against reserves that “paid the bill” for past 17 practices. Id. Soon after, Frontier issued a press release downplaying its financial troubles, and 18 its CEO made similar misleading statements. Six months later, National Underwriter published 19 an article discussing Frontier’s problems. In March 2000, Frontier’s rating was downgraded 20 because of “significant under-pricing,” and the next month Frontier issued a press release 21 indicating that there was “substantial doubt about the Company’s ability to continue as a going 22 concern.” Id. at 152. Plaintiffs filed suit in July 2000. At that time, the statute of limitations for 23 securities fraud claims was three years from the date of the fraud or one year from the discovery 9 1 of the fraud or of facts that placed an investor of ordinary intelligence on inquiry notice. See id. 2 at 154. The issue in LC Capital Partners was whether the lawsuit had been filed within a year 3 after there were sufficient “storm warnings” to place an investor of ordinary intelligence on 4 notice. 5 More than five years earlier, in November 1994, another class action against Frontier had 6 been filed in the Eastern District of New York. In re Frontier Ins. Group Sec. Litig., 172 F.R.D. 7 31 (E.D.N.Y. 1997)(“Frontier I ”). The complaint alleged that Frontier had fraudulently 8 concealed and misrepresented its knowledge of “grossly inadequate” reserves and insufficient 9 staffing. Id. at 153. In a report recommending denial of the plaintiff class’s motion to amend the 10 class period to extend through November 1999, the magistrate judge concluded that investors 11 were on inquiry notice of the fraud no later than December 21, 1998—the date of the National 12 Underwriter article discussing the massive $139 million reserve charge. Id. The district court 13 approved the report and recommendation. Expressly following the finding of Frontier I that 14 inquiry notice existed as of December 1998, the district court in LC Capital Partners dismissed 15 that complaint. Id. 16 We affirmed, finding that Frontier’s December 1998 announcement that it was taking a 17 substantial reserve charge constituted a sufficient “storm warning” to place investors on inquiry 18 notice of possible securities fraud. Id. at 155. This inquiry obligation was not excused by self- 19 serving statements by Frontier’s management that its reserve problems were behind it, because 20 the three “substantial reserve charges” taken in the brief period from 1994 to 1998 should have 21 alerted an ordinary investor about Frontier’s financial troubles. Id. Although the National 22 Underwriter article that specifically discussed Frontier’s reserve problems “[a]lso contribut[ed] 23 to a duty of inquiry” – as did the Frontier I litigation in the Eastern District – the multiple 10 1 reserve charges themselves were identified as the “principal warnings” in LC Capital Partners. 2 Id. 3 A few months later, we decided Newman 335 F.3d at 187. The plaintiffs were 4 stockholders of Warnaco, a clothing manufacturer. Id. at 189. Plaintiffs alleged that one of 5 Warnaco’s executives forced the company’s Forecasting and Planning Division knowingly to 6 increase its sales projections beyond its attainable or actual sales projections. Allegedly because 7 of this fraudulent practice, excess inventory had to be written down in value or written off 8 entirely. Id. Plaintiffs filed suit in August 2000, and the district court granted the individual 9 defendants’ motion to dismiss on statute of limitations grounds. Id. at 192. The court concluded 10 that inquiry notice was triggered in April 1999, when Warnaco revised its Form 10-K for Fiscal 11 Year 1998 to reveal “the dramatic reversals in financial data and the language of the 1998 Form 12 10-K.” Id. 13 We vacated the dismissal, concluding that notice of the filing of the revised 1998 Form 14 10-K did not put reasonable investors on notice of any fraud. Id. at 195. The 1998 Form 10-K 15 “did not contain any indication” that the cause of the inefficiency costs was inventory fraud, 16 which was the basis of plaintiff’s complaint. Id. at 194 (emphasis in original). It was 17 “reasonable for Plaintiffs not to inquire” after the April 1999 filing, because Warnaco had 18 provided the “seemingly benign explanation” that it had revised its financial information to 19 conform to changed accounting methods. Id. 20 The next month, in Levitt v. Bear Stearns & Co., we again vacated the district court’s 21 dismissal of a securities fraud suit, this time because there was a genuine issue of fact about 22 whether investors had sufficient “storm warnings” to commence the limitations period— an issue 23 that could not be decided on a motion to dismiss. 340 F.3d at 96, 104. Plaintiffs – investors who 11 1 used a brokerage firm to purchase stock of ML Direct, Inc. following an initial public offering of 2 its securities – claimed that defendant Bear Stearns had knowingly participated in a fraudulent 3 scheme perpetrated by the brokerage involving the IPO. Bear Stearns had acted as the clearing 4 agency for the brokerage. 5 The Levitt plaintiffs commenced their action in February 1999. Id. at 100. The district 6 court held that plaintiffs’ claim was time-barred, finding that there were sufficient “storm 7 warnings” so that a reasonable investor should have discovered Bear Stearns’ alleged role more 8 than one year (the then-applicable limitations period) before the complaint was filed. See id. at 9 101. These “storm warnings” included a March 1997 National Association of Securities Dealers 10 (“NASD”) arbitration proceeding against Bear Stearns, in which market manipulation with 11 regard to the IPO was alleged, as well as an action commenced in Texas federal court against the 12 brokerage and Bear Stearns in August 1997. Id. at 99-100. 13 Levitt differed from prior “storm warnings” cases, because it alleged liability against a 14 secondary wrongdoer: plaintiffs sued the clearing agent (Bear Stearns), not ML Direct, the 15 company whose stock price dropped. The Levitt plaintiffs conceded that their duty to inquire 16 about goings-on at ML Direct arose toward the end of 1996, when they learned of their 17 investment losses. Id. at 102. However, plaintiffs could not have “allege[d] a prima facie case 18 against Bear Stearns simply by examining ML Direct’s financial statements and media coverage 19 of the company.” Id. at 103. The issue in Levitt was whether plaintiffs had exercised reasonable 20 diligence in discovering the facts that established Bear Stearns’ participation in the brokerage’s 21 fraudulent scheme. Id. at 103. We vacated the district court’s dismissal of the complaint 22 because there were factual disputes about (1) the scope of the inquiry conducted by plaintiffs and 23 (2) whether reasonable inquiry could have revealed enough information to satisfy the strict 12 1 pleading requirements of the Private Securities Litigation Reform Act (“PSLRA”). We 2 concluded that “discovery should have been permitted on the question of what information was 3 realistically available to Plaintiffs and when it was available.” Id. at 104. 4 In Lentell v. Merrill Lynch & Co., we found that the “storm warnings” about defendants’ 5 conflict of interest were insufficient to give rise to inquiry notice. 396 F.3d at 170-71. The 6 plaintiffs in Lentell claimed that Merrill Lynch and its star analyst, Henry Blodget, issued 7 misleading favorable reports intended to artificially inflate the stock prices of companies that 8 were clients of the firm’s investment bank, in order to cultivate investment banking business 9 from those clients. Id. at 165. Plaintiffs did not own stock in defendant Merrill Lynch; rather, 10 they were investors of the companies that were the subject of Blodget’s allegedly misleading 11 reports. Id. at 164. 12 The Lentell lawsuits were filed shortly after the New York Attorney General announced 13 an investigation into Merrill’s practices. Id. The district court concluded that plaintiffs’ claims 14 were time-barred because articles that appeared in the financial press years earlier had placed 15 investors on inquiry notice of Merrill’s alleged misconduct. Id. at 168, 170. Although we 16 affirmed dismissal on the alternate ground that plaintiffs failed to state loss causation, id. at 176, 17 we found, as an antecedent matter, that investors were not put on inquiry notice of possible 18 securities fraud by “generic articles on the subject of structural conflicts [at financial firms]” that 19 appeared in the financial press. Id. at 170. We stated: 20 Conflicts of interest present opportunities for fraud, but they do 21 not, standing alone, evidence fraud—let alone furnish a basis 22 sufficiently particular to support a fraud complaint. Nor does the 23 existence of temptation trigger a duty of inquiry— at least, not by a 24 reasonable investor. Something more than conflicted interest is 25 required, no matter how well publicized the conflict may be. 26 13 1 Id. Emphasizing the circumstance-specific nature of the inquiry notice determination, we found 2 that the “articles relied upon to support that [the district court’s] finding fall well short of the 3 specificity required to prompt further inquiry by a reasonable investor.” Id. 4 In Shah v. Meeker, decided one year after Lentell, the alleged fraud was Morgan 5 Stanley’s representation to investors that its research analysts – including senior analyst Mary 6 Meeker – were unbiased and objective. 435 F.3d at 245. Mr. Shah, a Morgan Stanley 7 shareholder, claimed that the conflict of interest between the research and investment banking 8 groups at Morgan Stanley – together with the failure to disclose improper practices to 9 shareholders – artificially inflated the price of Morgan Stanley’s stock, and resulted in a loss to 10 him when Morgan Stanley’s conflict of interest came to light and the value of its stock 11 decreased. Id. Morgan Stanley’s stock dropped immediately after the NYAG announced an 12 investigation in April 2002; Shah brought his suit in July 2003. 435 F.3d at 247-48. 13 The “storm warnings” in Shah were: a 1996 Wall Street Journal article on page C1 14 generally describing the conflict of interest; another general article in the Journal in 1998; one 15 article about Mary Meeker in The New Yorker in 1999; and two articles in Fortune, including a 16 feature article about Mary Meeker in May 2001. Id. at 247. 17 In Shah, we affirmed the district court’s conclusion that the complaint was time-barred. 18 We noted that, “‘Storm warnings in the form of company-specific information probative of fraud 19 will trigger a duty to investigate,’” Id. (quoting Lentell, 396 F.3d at 169), and that, “Information 20 contained in articles in the financial press may trigger the duty to investigate.” Id. (citing, inter 21 alia, LC Capital Partners, 318 F.3d at 155). We concluded that inquiry notice was triggered no 22 later than May 2001, when Fortune magazine featured an article that specifically described the 23 practices of Morgan Stanley and Mary Meeker. Id. at 249, 251. In contrast to the press reports 14 1 that were found to be insufficient in other decisions (like Lentell, and Fogarazzo v. Lehman 2 Bros., Inc., 341 F. Supp. 2d 274, 300 (S.D.N.Y. 2004)), the Fortune article in Shah “did not 3 report the mere existence of a conflict of interest,” but “described the alleged conflicts of 4 interest” with a “degree of specificity” sufficient to put the plaintiffs on notice of a possible 5 cause of action. 435 F.3d at 250-51. We expressly declined to address whether the other articles 6 cited by defendants as “storm warnings” would have triggered a duty to inquire. Id. at 252. 7 Against the backdrop of this jurisprudence, we turn to the “storm warnings” relied on by 8 Appellees in this case. 9 10 D 11 12 To support their contention that the market was aware of the information underlying the 13 alleged fraud long before this action was filed, Appellees identified the following information 14 that was publicly available: complaints in four lawsuits that were filed in California and Illinois 15 state courts between 1999 and 2001; various articles from news sources and industry newsletters; 16 portions of regulatory filings filed by The Hartford with the SEC; and samples of insurance 17 filings that The Hartford files annually for each of its insurance subsidiaries in every state in 18 which the subsidiaries are domiciled (the “Insurance Expense Exhibits”). Staehr, 460 F. Supp. 19 2d at 335. Appellees also submitted three press releases that reported on the NYAG’s 20 enforcement actions pertaining to contingent commissions, but these releases were all issued in 21 or after June 2004 and so do not advance their argument. (J.A. 184, 210, 287-88.) 22 15 1 1. Lawsuits 2 3 a. Daniel v. Aon Corp. 4 5 Daniel v. Aon Corp. was filed in Illinois Circuit Court, Cook County, on August 19, 1999 6 by Alan Daniel, an Illinois insurance agent who had obtained an errors and omissions policy 7 through Aon. Daniel v. Aon Corp., et al., No. 1999-CH-11893 (Ill. Cir. Ct. 1999). Mr. Daniel 8 brought this class action against Aon Corporation and Aon Group, Inc., on behalf of purchasers 9 of commercial insurance through Aon’s brokerage services. 10 The complaint alleged that Aon and its affiliates received undisclosed financial incentives 11 from insurers that caused them to refer business to insurance companies even if the policy quoted 12 by the insurer was not in the best interest of the customer. The complaint also referenced a 1999 13 New York Times article (another exhibit submitted by Appellees) for the proposition that large 14 commercial brokers had an incentive to steer business to those insurers that offered the biggest 15 kickbacks, rather than to those that provided the best coverage and lowest prices to the insureds. 16 ( J.A. 161.) Amended complaints were filed on November 1, 1999, and February 9, 2001. The 17 Hartford was neither named as a party nor mentioned in any of the complaints. 18 The Daniel lawsuit was mentioned in an insurance industry newsletter, National 19 Underwriter, on October 2, 2000. None of the materials pertaining to the Daniel lawsuit 20 (including the National Underwriter article) mentions The Hartford. 21 22 b. Turner I 23 24 Turner v. Aon Corp. (“Turner I”) was filed on December 23, 1999, in California Superior 25 Court, County of Monterey. Turner v. Aon Corp., et al., No. M 47230 (Cal. Super. Ct. 1999). 26 Plaintiff Scott C. Turner – a California attorney who specializes in insurance law – sued major 16 1 brokers, including Aon, Marsh and Willis Corroon (and their affiliates), on behalf of the general 2 public. (J.A. 123-32.) 3 The Turner I amended complaint alleged that, since at least 1986, and without 4 policyholders’ knowledge, defendant brokers entered into “undisclosed agency contracts” with 5 insurance companies, pursuant to which the defendants were paid commissions based upon the 6 amount of business they produced for the insurance companies. (J.A. 128, ¶¶ 23-24.) These 7 undisclosed agency contracts allegedly created a conflict of interest between the financial 8 interests of the defendants and the interests of their policyholder clients. 9 Turner did not sue The Hartford, and the complaint mentioned the company only once, 10 when it alleged, “The insurance companies which entered into the undisclosed agency contracts 11 with the defendants include The Travelers Insurance Group, The Chubb Group of Insurance 12 Companies, The Hartford Insurance Group and, upon information and belief, other major 13 property and casualty insurance companies.” (J.A. 128 ¶ 24.) 14 The October 2000 National Underwriter article mentioned above also noted the Turner I 15 lawsuit, but as with the Daniel action it did not implicate The Hartford or any other insurer. The 16 allegation from the Turner I complaint that is quoted above did not appear in the article, which 17 stated only that the Turner I suit claimed that “brokers should have disclosed the fees to their 18 clients earlier and that the fees should have been passed onto their clients.” ( J.A. 224 (emphasis 19 added).) 20 21 c. Village of Orland Hills v. Arthur J. Gallagher & Co. 22 23 Village of Orland Hills, filed on September 22, 2000, in Illinois Circuit Court, Cook 24 County, was a class action alleging that Gallagher – a broker – breached its fiduciary duties “by 25 receiving undisclosed contingent commissions (kickbacks) from insurance companies for 17 1 insurance placed on behalf of its insureds.” Village of Orland Hills v. Arthur J. Gallagher & 2 Co., No. 00-CH-13855 (Ill. Cir. Ct. 2000). ( J.A. 171 ¶ 1.) 3 The complaint also alleged that Gallagher’s contingent commission agreements with 4 insurance companies created a conflict of interest, in breach of fiduciary duties, because the 5 broker puts its financial interests first, secretly profiting at the expense of its clients by placing 6 insurance policies with those insurance companies which pay contingent commissions. (J.A. 174 7 ¶ 15.) No insurer, including The Hartford, was named as a defendant. 8 The Village of Orland Hills lawsuit was discussed in a brief (167 word) “squib” that 9 appeared in the Chicago Tribune. The article succinctly summarized the allegations in the 10 complaint. (J.A. 227.) Neither The Hartford nor any other insurer was mentioned in the Tribune 11 article. 12 13 d. Turner II 14 15 Turner v. Hartford Fire Ins. Co., et al. (“Turner II”) was filed by Scott C. Turner on July 16 25, 2001, in California Superior Court, County of San Francisco. Turner v. Hartford Fire Ins. 17 Co., et al., No. 323192 (Cal. Super. Ct. 2001); (J.A. 134-53.) Among the defendants were a 18 number of insurance companies, including various Hartford subsidiaries, although not The 19 Hartford Financial Services Group, Inc., the Defendant in this action. 20 The Turner II complaint alleged that the defendants entered into “Undisclosed Agency 21 Contracts” with brokers, whereby the defendants paid brokers millions of dollars in undisclosed 22 fees and commissions based upon the amount of business produced for the defendants by the 23 brokers. (J.A. 140 ¶¶ 26-27.) According to the complaint, these fees and commissions constitute 24 “kickbacks to brokers of premiums paid to defendants by policyholders.” (Id. ¶ 27.) The 25 complaint also alleged that the defendants engaged in unfair business practices, and breached 18 1 their duty of good faith and fair dealing by failing to disclose the contingent commissions (and 2 the concomitant increased cost of premiums to cover these commissions), to their policyholders. 3 (J.A. 144.) In addition, the complaint alleged that the insurer defendants helped brokers prepare 4 statements to claimants that contained false or misleading information about the material fact 5 that brokers had a conflict of interest. (J.A. 148.) 6 The Turner II complaint contained many allegations that appear in the instant complaint 7 concerning the broker-insurer commission scheme (though no allegations about bid rigging or 8 price manipulation). However, this lawsuit apparently flew under both industry and press radar 9 screens. Filed in a California state court, it was not discussed in any of the press articles 10 submitted by Appellees. The lawsuit also was not mentioned in any of The Hartford’s public 11 filings. Based on the exhibits submitted by Appellees, the only way this lawsuit could have 12 come to the attention of the investing public was if someone had encountered it while examining 13 the docket of the Superior Court of San Francisco County. 14 15 2. Media Reports 16 17 The news articles that allegedly constituted part of the “storm warnings” can be broken 18 into two categories: (i) stories in mainstream publications, and (ii) stories in industry publications 19 or newsletters. The former includes one article that appeared in the New York Times, two 20 Financial Times articles, and a short piece in the Chicago Tribune. The other thirteen articles 21 appeared in three publications directed to the insurance industry: National Underwriter, Business 22 Insurance, and Canadian Underwriter. 23 The articles submitted to the court by Appellees focused primarily on brokers, not 24 insurance companies. None of the articles from mainstream publications so much as mentioned 25 The Hartford, and only one of the industry newsletters did. Moreover, many of the articles 19 1 appearing in 1999 or later suggested that concerns about disclosure had been resolved or, at the 2 very least, that brokerages were working toward a solution. 3 4 a. Mainstream Publications 5 6 i. New York Times 7 8 On February 14, 1999, the New York Times ran an article titled Insurers Pay the Brokers, 9 Making Customers Wary. (J.A. 229-31.) This article appeared on the fourth page of the business 10 section. It began by describing the controversy surrounding brokers’ receipt of contingent 11 commissions from insurance companies: 12 The rapid consolidation of the brokerage industry is making 13 contingency fees even more controversial. Two giant brokers, 14 [Marsh and Aon,] now dominate the field. And risk managers . . . 15 have become increasingly concerned that brokers, less restrained 16 by competition, could concentrate their business with insurers that 17 offer the biggest incentives rather than those that provide the best 18 service and lowest prices. 19 20 According to the article, these contingency fee arrangements were not generally known until 21 recently, in part because brokers downplayed their importance: 22 Many corporate executives said they had been unaware of 23 contingency fees until recently. State regulators in New York say 24 they, too, were in the dark. * * * The brokers acknowledge that 25 they have not called attention to contingency fees but say they 26 have always been willing to discuss them in general terms, if 27 asked. They say that the payments from the insurers are a small 28 part of their total revenue – no more than 5 percent – and that the 29 fees do not influence their choice of carriers. 30 31 The article also presented the insurance companies’ point of view regarding contingency 32 fees, implying that the insurers were at the mercy of the brokerage firms that had come to 33 dominate the industry: 34 But the insurance companies that pay the fees see them differently. 35 Though some insurers do not pay contingency fees, many others 20 1 consider them a necessary tool for making sure they get a slice of 2 the pie. 3 4 The article concluded by noting regulators’ emphasis on disclosure, but also mentioning 5 that there was no proof customers had suffered financially because of the practice. 6 [L]ast summer, after becoming aware of the controversy, the New 7 York State Department of Insurance issued a formal reminder that 8 according to state law insurance buyers should be informed of all 9 compensation arrangements between insurer and broker so they 10 can ‘understand the costs of coverage and the motivation of their 11 broker.’ . . . . The New York regulators and the risk managers 12 were unable to cite cases in which brokers had arranged coverage 13 that might have been bought for less or had steered customers to 14 insurers that failed to meet clients’ needs in other ways. 15 16 The Hartford was not mentioned in the article— a particularly important omission, since the 17 writer acknowledged that not all insurers paid contingency fees to get business. 18 19 ii. Financial Times 20 21 The Financial Times (“FT ”) ran articles on the contingent commissions story: Aon Set to 22 Disclose Fees, March 16, 1999; and Consolidation at Relentless Pace, April 28, 2000. (J.A. 23 233-36.) 24 The March 1999 article appeared at page thirty-three. (J.A. 233.) This article noted that 25 both Marsh and Aon took steps to disclose to clients their contingent commission revenue after 26 brokers had “come under fire” for receiving millions of dollars in fees and incentives from 27 insurance underwriters in addition to the payments they receive from clients. According to the 28 article, Marsh and Aon – which had become the dominant brokerages as a result of consolidation 29 in the global broking industry – were “clearly anxious to avoid criticism from regulators,” 30 especially after “the New York State Insurance Department [in 1998] said lack of transparency 31 over broker remuneration could be construed as a violation of the state’s insurance law.” 21 1 The April 2000 article – which appeared at page three – is similar to the FT article 2 discussed above, to the point of repeating verbatim many of the same phrases. The article also 3 referred to brokers’ receipt of “incentive payments” as “double dipping,” which, according to the 4 article, “called into question the principle that customers should receive ‘best advice’ and raised 5 doubts over conflict of interest.” (J.A. 235-36.) The Hartford was not mentioned in the article. 6 7 iii. Chicago Tribune 8 9 As noted already, on September 23, 2000, an article on page five of the Chicago Tribune 10 discussed the Village of Orland Hills lawsuit, which focused on the practices of broker Arthur J. 11 Gallagher & Co. The article did not mention The Hartford or any of its competitors. 12 13 b. Industry Publications 14 15 The industry newsletters submitted by Appellees date back to 1997. The articles that 16 appeared in these niche publications also focused on the allegations against brokers, not 17 insurance companies. 18 An October 13, 1997 article in Business Insurance discussed the possibility that 19 consolidation of insurance brokerages might result in brokers using their clout to extract larger 20 commissions. Though The Hartford was not mentioned, the article did name a broker (Marsh) 21 and an insurer (Chubb), noting that “the business placed with Chubb through [Marsh’s] Global 22 Broking Centers is subject to a ‘placement services agreement’ . . . [which] grants [Marsh] an 23 additional commission for the placement of large volumes of business through the centers.” The 24 article went on to say: “Other U.S. brokers have commission agreements with insurers providing 25 for the payment of contingent commissions based on the profitability of the business they place 26 with insurers.” (J.A. 238.) It did not identify those brokers or insurers. 22 1 An April 27, 1998 article in Business Insurance concentrated on the behavior of brokers 2 abroad: “Recent U.K. media reports have implied that large brokers, unbeknownst to their 3 policyholder clients, have been using size and market share to strong-arm insurers into entering 4 incentive commission arrangements.” Citing the potential conflict of interest for brokers, the 5 article noted that “U.K. risk managers echo concerns voiced by their U.S. counterparts over 6 whether some forms of contingent commissions could lead to brokers placing their business with 7 an insurer that is not necessarily providing the best coverage for their exposures.” (J.A. 242.) 8 The May 11, 1998 Business Insurance article submitted by Appellees highlighted some 9 of the issues that were discussed at the Risk and Insurance Management Society (“RIMS”) 10 annual conference. The article stated that the issue of contingent commissions had “come to a 11 head” with the consolidation of the brokerage industry and concomitant increased clout for the 12 largest brokers. But the article also quoted brokers who argued that placing a large volume of 13 business with a certain insurance company actually benefited policyholders, because it led to a 14 strong relationship between the brokerage and the insurance company, allowing brokers to solve 15 their policyholders’ problems. (J.A. 246-48.) 16 A June 15, 1998 National Underwriter article discussed the increased attention to 17 contingency fees in the U.K., noting a proposal which called for a code of conduct that would 18 make payment of incentives to brokers illegal in that country. But the article also included a 19 statement by the Chief Executive of the British Insurance and Investment Brokers Association, 20 who said: “There is no evidence, no suggestion, no clients coming forward saying, ‘We think 21 brokers are abusing their positions of power, their size, their leading position in the marketplace, 22 to our detriment.’” (J.A. 250.) 23 1 Several of the articles submitted by Appellees appeared in industry publications in the 2 first half of 1999. Although they discussed the controversy surrounding the contingent 3 commissions received by brokers, these articles also emphasized brokers’ efforts to respond to 4 calls for disclosure of their contingent commission arrangements. 5 The January 18, 1999 National Underwriter article stated that “recent industry attention 6 . . . has focused on the issue of disclosing [contingency fee] arrangements . . . . [T]he New York 7 State Department of Insurance issued a circular letter recommending that brokers allow clients 8 access to this information as well.” (J.A. 254-55.) The article discussed the possibility of a 9 market solution to contingency fee arrangements, noting that broker Willis Corroon had 10 negotiated “carve-out arrangements” with certain clients who asked to take their premium out of 11 the contingency fee equation. Id. 12 Appellees also submitted an article titled “RIMS, Broker in Deal to Reveal 13 Commissions,” which appeared in Business Insurance on February 1, 1999. This article, which 14 focused on Marsh but also mentioned brokers Aon and Willis, noted that contingent 15 commissions were not necessarily a problem, although they should be disclosed. According to 16 the article, “Risk managers have voiced concerns that the commissions, which are paid by 17 insurers to brokers based on volume targets and the profitability of the business, may create a 18 conflict of interest.” (J.A. 257-59.) 19 Another Business Insurance article, from February 22, 1999, discussed Aon’s efforts to 20 notify clients of the potential contingent commissions it received based on the volume of 21 business it placed with insurers. (J.A. 262.) 22 Likewise, the March 1999 Canadian Underwriter article submitted by Appellees noted 23 that “[Marsh had] agreed to reveal contingency commission arrangements to clients on request.” 24 1 The article quoted a senior vice president of Willis Corroon, who emphasized the brokerage’s 2 transparency with clients: “We don’t consider contingencies a huge issue, it accounts for less 3 than 2% of our total revenues. So we have nothing to hide from our clients and are willing to 4 negotiating [sic] unique deals.” According to the article, however, “Risk managers . . . [were] 5 disappointed it ha[d] taken the weight of the industry to compel disclosure.” (J.A. 265.) 6 Five industry newsletters from 2000 mentioned or alluded to the Turner I lawsuit. One of 7 those articles also referred to the Daniel suit. Not a single industry publication (at least, none 8 included in the record in this case) picked up on the Turner II lawsuit, in which The Hartford’s 9 subsidiaries were named as defendants and allegations approximating those asserted here were 10 made. 11 Only one of the so-called “storm warnings” from the industry/niche press mentioned The 12 Hartford: an article that appeared in National Underwriter on February 14, 2000. Referring to 13 the Turner I complaint, the article picked up on the mention of The Hartford in the complaint: 14 “The three top brokers are being sued by an attorney who alleges that the firms . . . deceived 15 policyholders in the state of California by their failure to disclose contingency fee arrangements. 16 . . . These contingency fee contracts were made with insurers including The Travelers Insurance 17 Group, The Chubb Group of Insurance Companies, The Hartford Insurance Group, and possibly 18 other major property-casualty insurers, according to the suit.” (J.A. 212.) 19 The same day, an article about Turner I appeared in a publication called Business 20 Insurance. (J.A. 213-15.) The Business Insurance article, however, did not mention The 21 Hartford or any other insurer, and focused entirely on the alleged misconduct by brokers. “In an 22 effort to compel brokers to disclose contingent commissions they earn, a law firm known for 23 battling commercial insurers on behalf of policyholders is suing the world’s three largest 25 1 brokers,” the article said. “The [Turner I] suit names Aon Corp., Marsh Inc., Willis Group Ltd., 2 their units, and other companies the brokers have acquired or merged with. The plaintiff is a 3 California attorney associated with Anderson, Kill.” The article summarized the Turner I 4 complaint’s allegations as follows: 5 The suit arises from some brokers’ practice of collecting 6 contingent commissions from insurers with which they place 7 business. These fees, which may be paid to the broker without a 8 policyholders’ knowledge, can be rewards for the volume of 9 business placed with an insurer, as well as compensation for 10 services the broker performs on the insurer’s behalf. 11 12 The article noted, however, that the lawsuit “is not supported . . . by [RIMS], the largest 13 organization of commercial insurance policyholders,” and that the president of RIMS was taken 14 aback by the lawsuit, saying that the issue of contingent fees was “not even on my radar screen 15 right now.” (J.A. 213.) 16 Like several of the other articles from industry newsletters, this article mentioned that 17 brokers had agreed to reveal some of their commission arrangements. It noted that RIMS 18 “focused attention” on the issue of contingent commissions at its annual meeting in 1998, 19 resulting in the negotiation of a “voluntary policy of disclosure with Marsh and later Aon.” 20 Again indicating that brokers, not insurers, were the target of the lawsuit, the article cited an 21 industry source as saying, “The brokers named in the complaint have united to defend against it.” 22 The Turner I suit was identified in the article as the first lawsuit over contingent fee 23 arrangements of which industry sources were aware, and the article quoted the senior litigation 24 counsel of Aon Corp. as saying, “Let everybody draw their own conclusions” from the fact that a 25 lawyer was the named plaintiff. (J.A. 214.) 26 1 Two weeks later, on February 28, 2000, another article appeared in National 2 Underwriter. This article, which was written in the style of an editorial, mentioned the legal 3 challenge to brokers’ contingency fee arrangements but questioned the viability of the lawsuit: 4 The question now is whether a legal challenge to the contingency- 5 fee system will hold up in court, and if it does, what impact this 6 will have on the brokerage community financially and 7 operationally. * * * [National Underwriter] has taken brokers to 8 task a number of times for the way in which contingency fees have 9 been kept from policyholders. . . . We applauded the fact that after 10 much arm-twisting, the major brokers seemed to come clean about 11 contingency fees. * * * However, while we stand squarely with 12 the consumer on this issue, we are frankly skeptical about this 13 lawsuit’s merits. * * * Bottom line, as outspoken as we were 14 about the need to disclose contingency-fee arrangements to clients, 15 we believe this is an ethical, rather than a legal issue – unless the 16 plaintiffs can actually prove individual instances of conscious 17 malpractice by brokers. 18 19 (J.A. 217-18.) 20 21 Similarly, a May 1, 2000 National Underwriter article expressed skepticism about the 22 Turner I lawsuit, describing it as “an attempt to revisit the [contingent commissions] issue,” and 23 noting that the lawsuit does not name specifically any of the defendant-broker’s clients who were 24 injured by the fee arrangements. The article also quoted the president of RIMS as saying that, 25 because the three major brokers had agreed to disclose their fee policies, the issue of contingent 26 commissions was “over” and lawsuits were a “waste of money and time.” (J.A. 220-22.) Again, 27 The Hartford was not mentioned in the article. 28 The most recent piece from a niche publication appeared in National Underwriter on 29 October 2, 2000. After discussing Turner I and Daniel (without mentioning The Hartford or 30 other insurers), it suggested that the contingent commissions issue had been resolved, noting that 31 the “storm appeared to be over” in early 1999. According to the article, representatives from 32 RIMS and the brokerage firms said that there had been “little talk” about contingency fees since 27 1 the disclosure policies were announced. Nevertheless, the article noted, “the announcements of 2 the disclosure agreements did not stop suits from being filed in California and Illinois. In both 3 cases . . . the suits claim that the brokers should have disclosed the fees to their clients earlier and 4 that the fees should have been passed onto their clients.” (J.A. 224-25.) 5 6 3. Regulatory Filings 7 8 Finally, Appellees submitted certain documents that are required to be filed each year by 9 The Hartford’s subsidiaries in states where they are domiciled (the Insurance Expense Exhibits). 10 See Appellees’ Br. at 30. The record includes exemplars of such filings for the years 1997-2002. 11 (J.A. 268-79.) Appellees also submitted portions of a Form 10-K filed by The Hartford in 2001. 12 (J.A. 281-82.) 13 The various operating expenses incurred by the filing subsidiaries are grouped into 14 various categories. There is a category called “Commission and Brokerage.” Under this heading 15 are various rows, one of which is headed “Acquisition, Field Supervision and Collection 16 Expenses.” A “contingent – direct,” “contingent-reinsurance assumed,” and “contingent – 17 reinsurance ceded” line items appear under this heading and row in the Insurance Expense 18 Exhibits. (J.A. 269, 271, 273, 275, 277, 279.) 19 The record also includes an excerpt from The Hartford’s Form 10-K for 2000, filed with 20 the SEC on March 23, 2001. At page 96 of this 166 page document the following language 21 appears: “Retrospective and contingent commissions and other related expenses are incurred and 22 recorded in the same period that the retrospective premiums are recorded or other contract 23 provisions are met.” (J.A. 281-82.) 24 28 1 E 2 Taking judicial notice of the materials described above, the District Court concluded that 3 the “lawsuits and press accounts” from 1999 to 2001 “clearly spelled out the existence of 4 contingent commission kickback schemes between insurance companies and brokers.” Staehr, 5 460 F. Supp. 2d at 338. Even though The Hartford was not mentioned in any of the mainstream 6 articles – and was mentioned but once, and fleetingly, in one article in an industry newsletter – 7 the District Court concluded that, “The obvious inference has to be that whatever practice the 8 brokerage companies employ as to some insurance companies (and here the relevant practice is 9 the contingent commissions) they employ as to most or all of the insurance companies.” Id. 10 Given The Hartford’s status as one of the largest insurers, the District Court reasoned, it would 11 have seemed “especially likely” to a reasonable investor that The Hartford was among the 12 insurers paying contingent commissions. Id. 13 The District Court also concluded that, while the “storm warnings” materials included no 14 reference to The Hartford’s bid-rigging schemes or the magnitude of the kickbacks it paid, there 15 was “clearly enough information available to place the plaintiffs on notice of the large majority 16 of the fraud they allege.” Id. at 339. The District Court reasoned that the “cumulative effect of 17 the publicly available documents should have suggested to a reasonable investor of ordinary 18 intelligence that The Hartford was likely one of the insurance companies paying contingent 19 commissions to the insurance brokers.” Id. (internal citation and quotation omitted). 20 For the date by which inquiry notice was triggered, the District Court selected July 25, 21 2001— the date the Turner II lawsuit was filed in California Superior Court. The court 22 emphasized the fact that the Turner II complaint named subsidiaries of The Hartford as 23 defendants and specifically alleged that they paid millions of dollars in undisclosed fees and 29 1 commissions to brokers. Id. at 338. The court did not discuss the fact that there was no evidence 2 in the record that the lawsuit had received any publicity. 3 Because Appellants did not claim to have undertaken any investigation with respect to 4 Appellees’ alleged non-disclosure, the District Court ruled that they were on constructive notice 5 of the fraud as of July 25, 2001, “the time [Turner II] was filed.” Id. at 339. The District Court 6 then held that this lawsuit, filed in 2004, was barred by the Exchange Act’s applicable two-year 7 statute of limitations. Id. at 340. 8 9 II. DISCUSSION 10 11 Standard of Review 12 13 We review the District Court’s grant of Appellees’ Rule 12(b)(6) motion to dismiss de 14 novo, accepting as true all factual allegations in the complaint and construing all reasonable 15 inferences in the non-movant’s favor. See, e.g., Vietnam Ass’n for Victims of Agent Orange v. 16 Dow Chemical Co., 517 F.3d 104, 115 (2d Cir. 2008); Shah, 435 F.3d at 248. We review the 17 District Court’s determination of whether to take judicial notice of facts for abuse of discretion. 18 See Island Software & Computer Serv., Inc. v. Microsoft Corp., 413 F.3d 257, 261 (2d Cir. 19 2005); United States v. Byrnes, 644 F.2d 107, 112 (2d Cir. 1981); accord In re NAHC, Inc. Sec. 20 Litig., 306 F.3d 1314, 1323 (3d Cir. 2002). 21 22 Analysis 23 This appeal raises two issues: whether the District Court’s decisions to take judicial 24 notice of documents not within the four corners of the complaint and to consider those judicially 25 noticed documents as part of the motion to dismiss were proper; and whether the materials 30 1 identified as “storm warnings” were sufficient to put an investor of ordinary intelligence on 2 inquiry notice of the frauds alleged in the complaint. 3 4 A. Judicial Notice 5 6 The District Court took judicial notice of, inter alia, media reports, state court 7 complaints, and regulatory filings. See Staehr, 460 F. Supp. 2d at 335; see also J.A. 122-82, 211- 8 82, 799. The court did “not take judicial notice of the documents for the truth of the matters 9 asserted in them, but rather to establish that the matters [had] been publicly asserted.” Staehr, 10 460 F. Supp. 2d at 335. 11 Appellants argue that, because none of these materials is cited or referred to in the 12 Complaint, the District Court should have granted their motion to strike these materials from the 13 record. They assert that the District Court impermissibly “reached far outside the Complaint” to 14 dismiss their claims, and erroneously used these materials to draw inferences in favor of 15 Appellees about what an ordinary investor would have known. See Appellants’ Br. at 14, 18. 16 Appellants cite Levitt for the proposition that it is error to determine that inquiry notice was 17 triggered without first permitting discovery on what information was “realistically available” to 18 Appellants and when it was available. See Levitt, 340 F.3d at 104. And they argue that the 19 District Court’s decision to take judicial notice of the materials submitted by Appellees conflicts 20 with our decisions in LC Capital Partners and Dodds, and so is grounds for reversal. 21 None of these contentions is persuasive. 22 We have previously held that it is proper to take judicial notice of the fact that press 23 coverage, prior lawsuits, or regulatory filings contained certain information, without regard to 24 the truth of their contents, in deciding whether so-called “storm warnings” were adequate to 25 trigger inquiry notice as well as other matters. See Global Network Commc’ns, Inc. v. City of 31 1 New York, 458 F.3d 150, 157 (2d Cir. 2006) (“A court may take judicial notice of a document 2 filed in another court not for the truth of the matters asserted in the other litigation but rather to 3 establish the fact of such litigation and related filings.” (citation omitted)); Kavowras v. N.Y. 4 Times Co., 328 F.3d 50, 57 (2d Cir. 2003) (SEC filings); LC Capital Partners, 318 F.3d at 155 5 (press coverage and prior litigation); Kramer v. Time Warner, Inc., 937 F.2d 767, 774 (2d Cir. 6 1991) (“[C]ourts routinely take judicial notice of documents filed in other courts . . . to establish 7 the fact of such litigation and related filings.”). In Kramer, we found that the district court could 8 take judicial notice of, inter alia, the publicized condition of the junk bond market during the 9 relevant time period. The reference was not relied on for its truth, and so was not “a ground for 10 decision and does not run afoul of the rule that a district court must confine itself to the four 11 corners of the complaint when deciding a motion to dismiss under Rule 12(b)(6).” 937 F.2d at 12 773; accord DeBenedictis v. Merrill Lynch & Co., Inc., 492 F.3d 209, 214 (3d Cir. 2007) (taking 13 judicial notice of USA Today and Time Magazine articles submitted by defendants). 14 Here, the District Court did not abuse its discretion in denying Appellants’ motion to 15 strike the materials submitted by Appellees for judicial notice purposes. The materials, like those 16 in Kramer, were offered to show that certain things were said in the press, and that assertions 17 were made in lawsuits and regulatory filings, which is all that is required to trigger inquiry 18 notice. None of those materials were offered for the truth of the matter asserted. 19 Appellants further argue that the District Court erred in considering judicially noticed 20 materials in ruling on the motion to dismiss without first converting the motion to one for 21 summary judgment. Although the general rule is that a district court may not look outside the 22 complaint and the documents attached thereto in ruling on a Rule 12(b) motion to dismiss, we 23 have acknowledged that the court “may also consider matters of which judicial notice may be 32 1 taken.” Kramer, 937 F.2d at 773. We have stated that, “Dismissal under FED. R. CIV. P. 2 12(b)(6) is appropriate when a defendant raises . . . [a statutory bar] as an affirmative defense and 3 it is clear from the face of the complaint, and matters of which the court may take judicial notice, 4 that the plaintiff’s claims are barred as a matter of law.” Conopco, Inc. v. Roll Int’l, 231 F.3d 82, 5 86 (2d Cir. 2000) (emphasis added). Appellee’s motion to dismiss pursuant to FED. R. CIV. P. 6 12(b)(6) was a proper means to plead that the claim should be dismissed as untimely. 7 The lapse of a limitations period is an affirmative defense that a defendant must plead 8 and prove. FED. R. CIV. P. 8(c)(1). However, a defendant may raise an affirmative defense in a 9 pre-answer Rule 12(b)(6) motion if the defense appears on the face of the complaint. McKenna 10 v. Wright, 386 F.3d 432, 436 (2d Cir. 2004); see also 5 Charles Alan Wright & Arthur R. Miller, 11 FEDERAL PRACTICE & PROCEDURE § 1226 (3d ed. 2004) (“[T]he current trend in the cases is to 12 allow [the statute of limitations defense] to be raised by a motion to dismiss under Rule 12(b)(6) 13 when the defect appears on the face of the complaint.”). Timeliness is “material when testing the 14 sufficiency of a pleading.” FED. R. CIV. P. 9(f). 15 Appellants’ argument that the District Court erred by considering “matters outside the 16 pleading” without converting Appellees’ motion to one for summary judgment is predicated on 17 the ground that a Rule 12(b)(6) motion to dismiss “shall be treated as one for summary 18 judgment” if matters outside the pleadings are presented to and not excluded by the court. See 19 FED. R. CIV. P. 12(b). However, matters judicially noticed by the District Court are not 20 considered matters outside the pleadings. 5 Wright & Miller, FEDERAL PRACTICE & PROCEDURE 21 § 1366 & n.33 (3d ed. 2004) (“[M]atters of which the district court can take judicial notice” are 22 “not considered matters outside the pleadings for purposes of conversion.”) Thus, the District 33 1 Court did not err in declining to convert Appellees’ Rule 12(b)(6) motion to a motion for 2 summary judgment. 3 4 Despite Appellants’ assertions to the contrary, the District Court simply took judicial 5 notice of Appellees’ exhibits for the purpose of establishing that the information in the various 6 documents was publicly available. Staehr, 460 F. Supp. 2d at 335. The court did not consider 7 the contents of those exhibits for their truth. The District Court’s finding that Appellants were 8 placed on inquiry notice in July 2001 was based on a determination that publicly available 9 materials (whether true or not) “should have suggested to a reasonable investor of ordinary 10 intelligence” that The Hartford was likely one of the insurance companies allegedly paying 11 contingent commissions to brokers, thus suggesting to investors the probability that they were 12 being defrauded. See id. at 339. 13 While we have no quarrel with the District Court’s decision to take judicial notice of 14 what information was “out there,” we do disagree with its conclusion that the total mix of 15 information before it was sufficient to rule, as a matter of law, that an investor of ordinary 16 intelligence was on inquiry notice of The Hartford’s allegedly fraudulent conduct by July 2001. 17 This decision does not foreclose the District Court from revisiting this issue at the summary 18 judgment stage or at trial. 19 20 B. The Exhibits Did Not Place the Appellants on Inquiry Notice by July 2001 21 22 The date on which one imputes knowledge to a reasonable investor for purposes of 23 “constructive” or “inquiry” notice varies, depending on what the investor does after being placed 24 on constructive notice: 25 If the investor makes no inquiry once the duty arises, knowledge 26 will be imputed as of the date the duty arose. However, if the 34 1 investor makes some inquiry once the duty arises, [this Court] will 2 impute knowledge of what an investor in the exercise of reasonable 3 diligence, should have discovered concerning the fraud, and in 4 such cases the limitations period begins to run from the date such 5 inquiry should have revealed the fraud. 6 7 LC Capital Partners, 318 F.3d at 154 (citations and internal quotation marks omitted). As we 8 said in Dodds, “Equitable tolling will stay the running of the statute of limitations only so long as 9 the plaintiff has exercised reasonable care and diligence in seeking to learn the facts which 10 would disclose fraud.” 12 F.3d at 350 (internal citations and quotation marks omitted). 11 Here, Appellants do not claim to have made any inquiry prior to the NYAG’s 12 announcement in 2004. Since well-settled law in this Circuit penalizes a plaintiff who fails to 13 make some inquiry after the “storm warnings” have given rise to constructive notice, the issue 14 before us is whether a duty of inquiry existed prior to October 15, 2002, two years before this 15 suit was filed. See 28 U.S.C. 1658(b). 16 For inquiry notice to exist, the triggering information must “relate[] directly to the 17 misrepresentations and omissions the Plaintiffs later allege in their action against the 18 defendants.” Newman, 335 F.3d at 193; see Levitt, 340 F.3d at 104. 19 Whether a plaintiff was placed on inquiry notice is analyzed under an objective standard. 20 See, e.g., Lentell, 396 F.3d at 168; Salinger v. Projectavision, Inc., 972 F. Supp. 222, 229 21 (S.D.N.Y. 1997). This objective determination can be resolved as a matter of law— it need not 22 be made by a trier of fact. See Dodds, 12 F.3d at 350 (rejecting the argument that “whether an 23 investor of ordinary intelligence would be on inquiry notice in the circumstances described in the 24 complaint is for a trier to determine” and could not be resolved as a matter of law). 25 “Storm warnings” need not detail every aspect of the alleged fraudulent scheme: “An 26 investor does not have to have notice of the entire fraud being perpetrated to be on inquiry 35 1 notice.” Dodds, 12 F.3d at 352. Rather, a totality-of-the-circumstances analysis applies. See 2 Shah, 435 F.3d at 249 (noting that a duty to inquire “arises ‘when the circumstances would 3 suggest to an investor of ordinary intelligence the probability’” of fraud (quoting Dodds, 12 F.3d 4 at 350)); see also de la Fuente v. DCI Telecomm., Inc., 206 F.R.D. 369, 383 (S.D.N.Y. 2002) 5 (“The issue . . . is whether the objective facts and circumstances, taken as a whole, provided 6 inquiry notice.”) (emphasis omitted). Inquiry notice may be found as a matter of law only when 7 uncontroverted evidence clearly demonstrates when the plaintiff should have discovered the 8 fraudulent conduct. see, e.g., Nivram Corp., 840 F. Supp. at 249. 9 Appellants contend that the exhibits submitted by Appellees are too vague and non- 10 specific to suggest to an investor of ordinary intelligence the probability of fraud by The 11 Hartford. Appellants also argue that the bulk of the judicially noticed information was not 12 reasonably accessible to the ordinary investor. Based on a review of the record in this case, we 13 find these arguments sufficiently compelling that the case cannot be decided on a motion to 14 dismiss. 15 16 1. Press Report 17 18 It is unremarkable that courts consider the extent of media coverage in deciding when 19 inquiry notice for securities fraud claims was triggered. See, e.g., In re Alcatel Sec. Litig., 382 F. 20 Supp. 2d 513, 523 (S.D.N.Y. 2005) (noting that determination of whether storm warnings existed 21 “depends on the quantity of information available to the plaintiffs”); cf. Rochelle v. Marine 22 Midland Grace Trust Co., 535 F.2d 523, 532 (9th Cir. 1976) (“Information in public records or 23 published by the news media may be so massive that investors will not be heard to say that they 24 remained ignorant of the financial plight of the corporation involved. . . .”) In this case, the 36 1 extent of media is seventeen articles—four from so-called “mainstream” publications and 2 thirteen from newsletters and publications addressed specifically to the insurance industry as 3 “storm warnings.” 4 The primary mainstream article relied upon by Appellees is the February 14, 1999 New 5 York Times article. We find that this article, and the remaining small amount of public 6 information available about contingent commissions at The Hartford, was not enough to create a 7 duty to inquire. On this record, we cannot say as a matter of law that an ordinary investor who 8 stumbled across this article would have inferred that The Hartford was involved at all. That is a 9 far cry from the District Court’s conclusion that it was “especially likely” that The Hartford was 10 implicated by the article. Staehr, 460 F. Supp. 2d at 338. 11 As we have noted, none of the other three mainstream press stories in the record mentions 12 The Hartford, either. Nor do twelve of the thirteen articles from industry newsletters. We stated 13 in Lentell that, “Storm warnings in the form of company-specific information probative of fraud 14 will trigger a duty to investigate.” 396 F.3d at 169. Because nearly all of the stories in the 15 record are devoid of company-specific information, the argument that they constitute “storm 16 warnings” is far from compelling. To be sure, we have expressly declined to limit a finding of 17 inquiry notice to company-specific storm warnings: “We do not mean to suggest that inquiry 18 notice could never be established on the basis of non-specific public pronouncements, but the 19 level of particularity in pleading required by the PSLRA is such that inquiry notice can be 20 established only where the triggering data ‘relates directly to the misrepresentations and 21 omissions’ alleged.” Lentell, 396 F.3d at 171 (quoting Newman, 335 F.3d at 193) (emphasis 22 omitted). However, the fact remains that the specificity of storm warnings bears directly on the 37 1 determination of whether, under the totality of the circumstances, a plaintiff should be charged 2 with a duty to inquire. 3 The “secondary wrongdoer” cases that have come before us are worth examining briefly, 4 because these cases involved storm warnings that were not specific to the company whose stock 5 price was allegedly affected by fraud. In Levitt, we acknowledged that a “case involving the 6 liability of a secondary wrongdoer” is “not a typical storm warnings case.” 340 F.3d at 103. In 7 that case, we held that the plaintiffs could not allege a prima facie case against Bear Stearns, a 8 secondary wrong-doer, simply by examining the financial statements of and media coverage 9 about, the issuer whose stock price fell. And in Lentell, we held that media reports about 10 conflicts within financial firms (the secondary wrongdoers) did not suggest to investors in other 11 companies that Merrill Lynch’s analyst reports falsified information about those companies. 12 Lentell, 396 F.3d at 170-72. 13 In Shah, we further explained (and contrasted) Lentell and its discussion of secondary 14 wrongdoers: 15 [I]n Lentell . . . articles cited by the district court [that] strongly 16 suggest grounds to believe that certain investment 17 recommendations were less than candid, but in which neither 18 company [recommended by defendants] is mentioned, cannot put 19 the holders of unmentioned companies on inquiry notice that those 20 companies were the subject of unobjective ratings. In the instant 21 suit, by contrast, the reports triggering a duty to investigate are not 22 limited to those mentioning specific corporations whose stocks 23 were overrated. An article specifically describing the business 24 practices at Morgan Stanley that serve as the basis of Shah’s 25 complaint as a holder of stock in Morgan Stanley must be regarded 26 as a “storm warning” of the fraud alleged. 435 F.3d at 251 27 (internal quotation marks omitted). 28 29 The case against The Hartford does not concern a secondary wrongdoer. Appellants’ 30 complaint alleges that Appellees’ misrepresentations and omissions artificially inflated The 38 1 Hartford’s own stock price. But as was true in Lentell, the publicly available information in the 2 record before us consists almost exclusively of generic articles on conflicts of interest for 3 insurance brokers, not insurers. We have described in detail the various media reports 4 principally to underscore the fact that there is no mention of The Hartford at all except in one 5 National Underwriter article. We conclude that, on the record before us, that article was not 6 sufficient to trigger inquiry notice as a matter of law.. 7 Appellees attempt to distinguish our decision in Lentell – and the nonspecific nature of 8 the news articles discussed therein – by limiting its holding to cases where “secondary 9 wrongdoers” are sued. We find this distinction unpersuasive. Although pleading a claim against 10 a secondary wrongdoer “may be especially difficult,” the critical factor for purposes of inquiry 11 notice in any case is “when the circumstances would suggest to an investor of ordinary 12 intelligence the probability that she has been defrauded.” Lentell, 396 F.3d at 168 (internal 13 quotation marks omitted) (quoting Levitt, 340 F.3d at 101). In this most important regard, this 14 case is like Lentell, not Shah, and the rationale behind Lentell cannot be set aside based on a 15 strained “secondary wrongdoer” distinction. 16 Because we have not created a categorical rule that inquiry notice can only be triggered 17 by public pronouncements containing company-specific information, we do not take issue with 18 the District Court’s statement, in the abstract, that “information that is not so company-specific 19 may also suffice for a court to find that the circumstances would suggest to an investor of 20 ordinary intelligence the probability that she has been defrauded.” Staehr, 460 F. Supp. 2d at 21 339. This statement may be true in some cases. However, the District Court’s application of this 22 principle to the facts of this case is erroneous. The news reports in the record were insufficient 23 to give a reasonable investor notice of a probability that they might have been defrauded. The 39 1 contingent commission schemes in this case were not publicized extensively until 2004— three 2 years after the date when constructive notice was triggered, according to the District Court. (See 3 J.A. 183-91, 196-98 (press releases, news articles, and filings from 2004 discussing the NYAG 4 action).) The pre-2004 press clippings in the record do not resemble those in cases like Shah, 5 where feature articles describing the subject of the suit (Morgan Stanley’s analysts) appeared 6 prominently in popular and widely read publications such as Fortune. The articles identified by 7 Appellees are much more similar to those that we found insufficient to trigger inquiry notice in 8 Lentell; they are general articles about structural conflicts in an industry, rather than being 9 specific to the company in which Appellants hold stock. Indeed, the press articles in this case 10 are even less indicative of the fraud alleged in the complaint than the articles we found deficient 11 in Lentell. 12 Appellees also would have us distinguish Lentell on the basis that the press reports in this 13 case clearly indicate that contingent commission schemes were “industry-wide.” The District 14 Court found that argument persuasive: 15 A number of the press accounts state that the brokerage industry 16 had been consolidating and that the large majority of the business 17 was controlled by Marsh and Aon. With two or three companies 18 controlling the brokerage industry, they effectively act as a funnel 19 through which much of the insurance business must pass. The 20 obvious inference has to be that whatever practice the brokerage 21 companies employ as to some insurance companies . . . they 22 employ as to most or all of the insurance companies. 23 24 Staehr, 460 F. Supp. 2d at 338 (emphasis added). However, to reach that conclusion, the district 25 court drew factual inferences that were not clearly demonstrated by the press accounts. Such 26 inferences are not appropriately drawn on a motion to dismiss. 27 The media reports cited by Appellees at most suggest the potential for conflicts of 28 interest involving some insurers, but not the probability of fraud by The Hartford. 40 1 This is underscored by the very article on which Appellees and the District Court rely most 2 heavily – the New York Times article – which specifically said that not all insurers pay contingent 3 commissions, without identifying either those companies that do or those that do not. The 4 references in the articles and newsletters to “contingent commissions” and brokerage firms’ 5 conflicts are not specific enough to provide an ordinary investor with indications of the 6 probability (not just the possibility, see Shah, 435 F.3d at 249; Lentell, 396 F.3d at 168; Newman, 7 335 F.3d at 193-9; Dodds, 12 F.3d at 350) of fraud by The Hartford. Focusing almost 8 exclusively on brokers’ practices, the press articles do not clearly suggest that any insurers 9 committed misconduct. The media reports in the record before us simply do not reveal the scope 10 of The Hartford’s schemes as alleged in the action now before the Court. 11 Our decision in Shah was influenced by the fact that the press reports about Morgan 12 Stanley’s analysts were so specific. When we distinguished Shah from Lentell, we stated that 13 inquiry notice was triggered “[b]ecause of the degree of specificity” with which the Fortune 14 magazine article described the conflict of interest alleged in plaintiff’s complaint. 435 F.3d at 15 251. The Fortune article, we noted, “stands in stark contrast to the ‘generic articles on the 16 subject of structural conflicts’ that we previously have held to be insufficient to trigger inquiry 17 notice.” Id. (quoting Lentell, 396 F.3d at 170). The Fortune magazine article’s specific 18 description of the business practices at the defendant company (Morgan Stanley), which served 19 as the basis of the plaintiff’s complaint against that company, was a sufficient “storm warning” 20 of the fraud alleged. Id. 21 Appellees argue that Shah compels us to affirm the District Court’s decision because the 22 exhibits “specifically describ[ed] the business practices” at The Hartford, and that those practices 23 “serve as the basis of” Appellants’ complaint. See Appellees’ Br. at 43 (internal quotation marks 41 1 omitted) (quoting Shah, 435 F.3d at 251). If anything, Shah supports reversal, because in that 2 case the media reports were so issuer-specific and so prominent. In this case, none of the stories 3 that appeared in the mainstream press contains any specific information about The Hartford’s 4 business practices— they do not even mention The Hartford by name. The same is true of 5 twelve of the thirteen stories from specialty publications. Just as in Lentell, these articles fall 6 short of the specificity required to prompt further inquiry by a reasonable investor. See 396 F.3d 7 at 170. 8 The thirteenth article – the February 14, 2000 story in the National Underwriter – does 9 not move this case significantly closer to Shah. While it mentions The Hartford (by quoting the 10 one sentence in the Turner I complaint that lists The Hartford as one of the insurers who paid 11 contingent commissions), it does so in passing, and in the context of an article concerning 12 practices by brokers— practices that would appear to victimize insurers. 13 Appellees correctly point out that we noted an article that had appeared in National 14 Underwriter when we affirmed the dismissal of the LC Capital Partners complaint. However, 15 the principal storm warnings that gave rise to the duty of inquiry in LC Capital Partners were 16 not press reports – let alone an article from a niche publication – but the repeated charges against 17 reserves by defendant Frontier. 318 F.3d at 155. Moreover, the National Underwriter article 18 mentioned in LC Capital Partners recited information about a lawsuit already filed against 19 Frontier. By contrast, the February 14, 2000 National Underwriter article in this case appeared a 20 year and a half before the filing of Turner II, the first lawsuit that named The Hartford as a 21 defendant. 42 1 In short, we cannot say either that this article would have come to the attention of a 2 reasonable investor of ordinary intelligence, or that, if it had, its contents were sufficient to place 3 such an investor on notice of the probability of fraudulent conduct by The Hartford. 4 After oral argument, Appellees provided us, pursuant to Federal Rule of Appellate 5 Procedure 28(j), with the Third Circuit’s decision in DeBenedictis v. Merrill Lynch & Co., Inc., 6 492 F.3d 209 (3d Cir. 2007). Appellees assert that this decision “bears directly upon” 7 Appellants’ arguments regarding the extent to which “storm warnings” must be company- 8 specific, and that it is consistent with Lentell in holding that articles and press releases that do not 9 identify the defendant by name can be sufficient to trigger a duty of inquiry. 10 DeBenedictis concerned allegations that Merrill Lynch’s brokers had a conflict of interest 11 with regard to the sales of Class B mutual fund shares – shares for which they received larger 12 commissions – by selling these shares to investors even though the mutual funds might not be in 13 the investors’ best interest. 492 F.3d at 210. The fund’s registration statements contained tables 14 and charts setting forth the fees and expenses for different classes of shares, and a description of 15 how the fee structure for each class affects its desirability. Id. at 210-13 16 The district court dismissed the complaint as time-barred, having taken judicial notice of 17 articles from USA Today and Time Magazine – published more than two years before the lawsuit 18 was initiated – which reported the conflict of interest that existed in the mutual fund industry 19 with respect to Class B shares. Id. at 214-15. The court also took notice of NASD news releases 20 reporting that brokers were disciplined for making unsuitable recommendations in selling Class 21 B shares to customers. Id. The district court found that the “registration statements, coupled 22 with the news articles and NASD press releases, sufficed to trigger inquiry notice.” 492 F.3d at 23 215 (emphasis added) (internal quotation marks omitted). The Third Circuit affirmed, 43 1 concluding that the materials relied on by the district court were sufficient to put an investor on 2 inquiry notice. Id. at 218-19. 3 Appellees argue that, in the Third Circuit’s view, the non-specific news articles in 4 DeBenedictis were sufficient to place plaintiffs on inquiry notice. But we do not read 5 DeBenedictis as holding that the non-specific articles and press releases were sufficient, without 6 more, to trigger inquiry notice. The Third Circuit observed, “News reports are not given weight 7 by courts in a vacuum, but rather have significance in cases where investors are presumed to 8 have read prospectuses . . . and other information related to their investments.” Id. at 217 9 (internal quotation marks omitted). This conclusion is consistent with our holding in LC Capital 10 Partners, where the reserve charges themselves were the “principal warnings,” and an industry 11 newsletter simply “[a]lso contribut[ed] to a duty of inquiry.” 318 F.3d at 155. 12 In Debenedictis, the widespread news coverage of mutual funds’ sales practices regarding 13 Class B shares should have alerted a reasonable investor to consult the prospectuses he received 14 to determine whether he was a victim of the fraud described in the news reports. Similarly, in 15 Shah, the specific articles about Morgan Stanley’s specific business practices would have alerted 16 a Morgan Stanley investor to the probability of fraud. In this case, however, the non-specific and 17 barely publicized news reports in this record did not reasonably provide Appellants with a 18 similar level of knowledge or awareness of the fraud alleged. Given the sparse and non-specific 19 coverage in mainstream media reports, The Hartford’s shareholders would not have been on the 20 lookout for unpublicized state court pleadings or obscure state regulatory filings, and thus they 21 would not have been reasonably aware of those public materials. 22 Appellees contend that actual notice of certain facts may be presumed. See Appellees’ 23 Br. at 37. However, the only Second Circuit case cited by Appellees to support this broad 44 1 proposition is an inapposite thirty-year-old decision that found omissions in a proxy statement to 2 be immaterial because a reasonable investor would have had certain information readily 3 available to him. See id. (citing Seibert v. Sperry Rand Corp., 586 F.2d 949, 953 (2d Cir. 1978).) 4 Appellees also argue that the difficulty of obtaining information should not be a criterion for 5 triggering the duty of inquiry, because if it were, investors would be mere passive actors with 6 respect to their investments. See Appellees’ Br. at 58. This contention, too, ignores the 7 objective character of the inquiry notice standard. Even if a particular plaintiff-investor acted 8 passively with regard to his or her investments, that plaintiff would be charged with knowledge 9 of information reasonably accessible to an investor of ordinary intelligence. 10 Given the objective standard for inquiry notice, there is an inherent sliding scale in 11 assessing whether inquiry notice was triggered by information in the public domain: the more 12 widespread and prominent the public information disclosing the facts underlying the fraud, the 13 more accessible this information is to plaintiffs, and the less company-specific the information 14 must be. In finding that the “storm warnings” were sufficient to trigger inquiry notice in this 15 case, the District Court necessarily concluded that an ordinary investor would come across the 16 judicially noticed materials submitted by Appellees, i.e., that these materials were reasonably 17 accessible. Based on the record before us, we simply cannot agree. 18 With the exception of the New York Times article, the two Financial Times articles, and 19 the Chicago Tribune blurb, the articles judicially noticed by the District Court were from 20 industry newsletters. (J.A. 212-15, 229-36, 265-66.) These articles were not sufficient to put 21 investors on inquiry notice. We have never affirmed the dismissal of a complaint as time-barred 22 based on a story that appeared only in a specialty publication, as opposed to mainstream press 23 reports that are more likely to come to the attention of an investor of ordinary intelligence. We 45 1 are reluctant to affirm such a dismissal where, as here, the specialty publication coverage is 2 relatively minimal and we do not know the publications’ circulation or the extent to which a 3 reasonable investor would be aware of the limited coverage. 4 Appellees place too much importance on LC Capital Partners, given the significant 5 differences between that case and this one. First, in LC Capital Partners, we made clear that 6 other publicly available documents were important to creating the duty to inquire. 318 F.3d at 7 155. Second, the National Underwriter article in LC Capital Partners included company- 8 specific information with regard to its discussion of the defendant’s problems that gave rise to 9 the “principal warnings.” Id. Third, the National Underwriter article in LC Capital Partners 10 repeated information that had already been the subject of previous lawsuits and disclosures. 11 Even assuming, arguendo, that an investor of ordinary intelligence would regularly 12 search various newsletters and information databases for news about the companies in which he 13 or she invests, this endeavor would not have been fruitful for an investor searching for 14 information about The Hartford, given the dearth of specific information about The Hartford or 15 insurers generally in the press articles judicially noticed by the District Court. It is for this 16 reason that the District Court could only take notice of the press articles for what they said about 17 the insurance industry as a whole. This is also why the District Court set July 25, 2001, as the 18 notice triggering date (when the Turner II lawsuit named subsidiaries of The Hartford as 19 defendants). 20 21 2. Regulatory Filings 22 23 As part of their argument that the totality of public information put Appellants on inquiry 24 notice by July 25, 2001, Appellees rely on the state regulatory filings that they submitted. They 46 1 also point to the two-page excerpt they submitted from The Hartford’s Form 10-K for the year 2 2000 (filed with the SEC on March 23, 2001). 3 The District Court briefly mentioned the regulatory filings, but it does not appear to have 4 given these filings any weight in holding that Appellants’ claims were time-barred. Staehr, 460 5 F. Supp. 2d at 338 (“There is no doubt that the lawsuits and press accounts clearly spelled out 6 the existence of contingent commission kickback schemes between insurance companies and 7 brokers.” (emphasis added)). 8 Furthermore, the state filings do not provide specific information that would suggest to a 9 reasonable investor the probability of fraud by The Hartford. In Newman, we held that, even 10 where the defendant’s Form 10-K contained specific information about write downs and 11 accounting charges that later formed the basis for fraud allegations, the plaintiffs were not placed 12 on inquiry notice by the 10-K, because the statements contained therein were not presented with 13 “clarity sufficient to place Plaintiffs on inquiry notice that there was a probability of fraud.” 335 14 F.3d at 194. The 10-K did not indicate that certain costs stemmed from fraud, so it was 15 reasonable for plaintiffs not to inquire further, particularly in light of the defendant’s “seemingly 16 benign explanation” for the costs. Id. 17 Similarly, The Hartford’s filings did not indicate that the contingent commission 18 expenses stemmed from fraudulent schemes. On the insurance exhibits, the category 19 “Commission and Brokerage” and the column heading “Acquisition, Field Supervision and 20 Collection Expenses” – under which the “contingent commissions” fall – are seemingly benign 21 categories of expenses in the absence of further explanation. The same can be said for the 22 “Revenue Recognition” subsection of The Hartford’s Form 10-K, within which the term 23 “contingent commissions” is used. 47 1 Appellees maintain that these filings “specifically refer to payments of contingent 2 commissions,” which is the very evil decried in the complaint. See Appellees’ Br. at 44 3 (emphasis added). The Form 10-K does use the phrase “contingent commission” in one 4 subsection. This is quite significant, because we normally will charge investors with notice of 5 the information in the 10-K. However, the 10-K at issue here did not define the term at all or 6 give any indication of its significance. The term was defined in some other newspaper articles, 7 and we might be inclined to charge investors with notice of information in the newspaper articles 8 when it is sufficiently widespread. However, the record The Hartford has compiled in this case 9 shows only a very small number of articles explaining the significance of the term “contingent 10 commissions,” and for the reasons discussed, we cannot conclude that a reasonable investor of 11 ordinary intelligence would have encountered these articles. Hence, we cannot assume that an 12 investor reading the 10-K would have understood the reference to “contingent commissions.” 13 We emphasize that our decision in this case is not meant to cast doubt on the usefulness 14 of 10-Ks and other disclosure documents generally as a means of putting investors on 15 constructive notice of facts about the company. We find significant in this case the fact that the 16 Defendant was able to put forward only a vanishingly small set of articles. More articles in 17 mainstream publications discussing the “contingent commissions” issue, combined with a 18 reference in the 10-K, would present a different case. It suffices to say that the record adduced 19 here was not enough to convince us that a reasonable investor would have notice, as a matter of 20 law, of the probability of fraud. 21 3. Lawsuits 22 23 Finally, with regard to the four lawsuits filed in different state courts, the Daniel and 24 Village of Orland Hills lawsuits do not mention The Hartford at all. The Turner I suit briefly 48 1 mentions the company’s name but does not specifically accuse it of wrongdoing, let alone the 2 wrongdoing that is the subject of this action. Thus, the Daniel, Village of Orland Hills, and 3 Turner I lawsuits would not have alerted a reasonable investor of ordinary intelligence to inquire 4 about inflated financial performance and bid rigging at The Hartford. 5 However, the Turner II lawsuit does name as defendants nine insurance subsidiaries of 6 The Hartford (although not the defendant in this case, The Hartford Financial Services Group, 7 Inc.), alleging that they violated California law by paying kickbacks to certain insurance brokers. 8 The allegations in the Turner II complaint are similar to the claims of kickbacks and 9 misrepresentations alleged by Appellants in this case. For this reason – and because the District 10 Court found that inquiry notice was triggered as of the Turner II filing date – we pay close 11 attention to the Turner II complaint. 12 Although Turner II does not allege that The Hartford’s subsidiaries misled investors by 13 touting The Hartford’s business success without disclosing that this success was premised on 14 bid-rigging and revenue inflation schemes, “storm warnings” need not detail the entire fraud. 15 See Dodds, 12 F.3d at 352. If a complaint alleges various frauds, an investor need not know the 16 details of each fraud in order to be placed on inquiry notice. For example, in LC Capital 17 Partners, the plaintiffs alleged several types of misconduct, including under-reserving for 18 insurance claims and predatory pricing. 318 F.3d at 150. We affirmed the dismissal of the 19 complaint, stating that the storm warnings related to under-reserving should have “alert[ed] any 20 reasonable investor that something is seriously wrong.” Id. at 155. Thus, despite Appellants’ 21 characterization of our precedents, these decisions would not prevent us from finding that Turner 22 II triggered inquiry notice, even though that complaint did not discuss the bid-rigging scheme 23 that is one of the bases of their complaint. 49 1 Appellants argue that, at most, the state court complaints submitted by Appellees might 2 have suggested that there was a possible conflict of interest within the insurance industry, but not 3 a widespread fraud being perpetrated by The Hartford. Appellants rely on Lentell to support this 4 argument. See 396 F.3d at 170. 5 We find Appellants’ argument unpersuasive as it pertains to the contents of the Turner II 6 lawsuit. The Turner II complaint did not merely signal the existence of a possible conflict of 7 interest in the insurance industry, but rather was grounded on an alleged failure by The Hartford 8 to disclose the payment of contingent commissions. For example, Turner II alleged that The 9 Hartford’s subsidiaries “fail[ed] . . . to disclose the . . . Undisclosed Fees and Commissions to 10 policyholders,” and that they “falsely represent[ed] to policyholders that the Included Brokers 11 would only be paid a disclosed commission . . . .” (J.A. 144.) This allegation does not just hint 12 at a temptation by The Hartford to commit fraud, but rather is the same basis of fraud alleged by 13 Appellants in the Complaint. 14 Although we do not underestimate the importance of the bid-rigging schemes alleged in 15 this case, we agree with the District Court that the contingent commission kickback schemes are 16 central to the Complaint, and that the alleged deception and fraud by The Hartford concerns the 17 failure to disclose these schemes. See Staehr, 460 F. Supp. 2d at 339. Turner II described the 18 failure of The Hartford’s subsidiaries to disclose the kickback schemes; thus, there was some 19 public information out there that suggested a substantial amount of the fraud alleged by 20 Appellants. See id. at 338. Specific allegations that The Hartford’s subsidiaries failed to 21 disclose kickback schemes, which are made in Turner II, should alert a reasonable investor that 22 something is wrong, LC Capital Partners, 318 F.3d at 155, and thus would trigger a duty to 50 1 inquire further. This conclusion is bolstered by our statement in Dodds that an investor need not 2 have “notice of the entire fraud being perpetrated to be on inquiry notice.” 12 F.3d at 352. 3 Because the Turner II complaint contains allegations that are similar to some of the key 4 allegations in the instant complaint, we agree with the District Court that the filing of the Turner 5 II lawsuit would trigger inquiry notice— but only if an investor of ordinary intelligence would 6 have been reasonably aware of the complaint. The dispositive issue, therefore, is whether the 7 Turner II complaint was reasonably accessible to an ordinary investor. We conclude that it was 8 not. 9 As we have already explained in detail, the Turner II lawsuit was not referenced in any of 10 the news articles or regulatory filings offered by Appellees. It would be unreasonable to expect 11 an ordinary investor to be aware of this lawsuit under these circumstances. We cannot say that 12 the reasonable investor of ordinary intelligence would or should have known about a lawsuit 13 filed in an unlikely venue (a California state court), that received no publicity whatever (not even 14 in niche publications) and that did not, at least based on the record before us here, result in 15 published or broadly disseminated opinions within the relevant time period. It has not been 16 demonstrated that this obscure and unpublicized lawsuit was reasonably accessible to such an 17 investor. 18 Appellees argue that, if the plaintiff in Turner II could detect “storm warnings” that 19 would cause him to file a lawsuit containing allegations of fraud against The Hartford’s 20 subsidiaries as of July 2001, then so, too, could the Appellants in this case. There is a 21 fundamental flaw to this argument, however: Scott C. Turner, the plaintiff in Turner II, was not 22 an investor of ordinary intelligence, but rather a lawyer affiliated with the firm Anderson, Kill & 23 Olick P.C.— a law firm that touts its insurance law practice and is known for battling 51 1 commercial insurers. 2 An insurance lawyer might very well have a sophisticated understanding 2 of the insurer-broker business, but he does not set the standard. The fact that an investor of 3 more-than-ordinary intelligence filed a lawsuit against subsidiaries of The Hartford in 2001 4 cannot be used as a bellwether for the adequacy of the “storm warnings.” 3 5 Following oral argument, Appellees submitted two recent unpublished opinions from this 6 Court concerning inquiry notice: Masters v. GlaxoSmithKline, 271 F.App’x 46 (2d Cir. Mar. 26, 7 2008), and Domenikos v. Roth, 2008 WL 2329315 (2d Cir. June 5, 2008). Of course, these 8 summary orders have no precedential effect on our analysis. See Local R. 32.1(b). However, the 9 facts of the two cases – which involved highly publicized class actions that could not have 10 escaped the notice of any reasonable investor - are so different from the facts before us that 11 nothing in the cited orders causes us to question our conclusion in this case. 12 13 14 III. CONCLUSION 15 16 In light of the foregoing, we VACATE the District Court’s decision granting Appellees’ 17 motion to dismiss the complaint as time-barred, and we REMAND this case to the District Court 18 for further proceedings not inconsistent with this opinion. We express no views about the merits 19 of the other grounds for dismissal raised by Appellees in their motion to dismiss, since the 20 District Court never reached them. 2 See J.A. 213, Roberto Ceniceros, Suit Asks Brokers to Return Contingent Commissions, Business Insurance, Feb. 14, 2000, at 1; see also Anderson Kill & Olick, http://www.andersonkill.com/about_Anderson_Kill.asp (last visited November 14, 2008). 3 Perhaps for this same reason, the February 14, 2000 Business Insurance article reported the senior litigation counsel for Aon Corporation as saying, “Let everybody draw their own conclusions” about the fact that an attorney was the only named plaintiff in the Turner I lawsuit. (See J.A. 213-14.) 52