Florida State Board of Administration v. Green Tree Financial Corp.

270 F.3d 645

• Court: Court of Appeals for the Eighth Circuit
• Decided: October 25, 2001
• Docket: 99-3536, 99-3586 and 99-3587
• Status: Published

Opinion

JOHN R. GIBSON, Circuit Judge.

In this case we must assess the sufficiency of three complaints under the new pleading standards made applicable to securities fraud cases by the Private Securities Litigation Reform Act of 1995, P.L. No. 104-67, 109 Stat. 737. The Florida State Board of Administration, which invests state employees’ pension funds, and two putative classes of investors, one for stock purchasers and one for option traders, have each brought Rule 10b-5 2 actions against Green Tree Financial Corporation, its CEO, Lawrence Coss, and other executives, 3 on the theory that they fraudulently overstated Green Tree’s financial value. The investors allege that they bought securities in a market affected by this fraud and that when the true facts came to light, Green Tree’s stock price tumbled and they lost their money. The district court held that none of the three complaints pleaded facts giving rise to a strong inference of knowing or reckless misconduct, and so the court dismissed the complaints with prejudice. Florida State Bd. of Admin, v. Green Tree Fin. Corp. (In re Green Tree Fin. Corp. Stock Litig), 61 F.Supp.2d 860, 878 (D.Minn.1999). The Florida Board and both investor classes appeal, contending that their complaints were indeed sufficient to plead securities fraud. We reverse.

I.

For the purpose of a motion to dismiss, we take the facts from the complaints. Green Tree is a financial services corporation that originally specialized in lending money on house trailers, referred to as “manufactured housing,” although it later diversified into other kinds of lending. Because manufactured housing loans are classified as “sub-prime,” the interest rates are very high — as much as 200 to 400 basis points (two to four percentage points) above residential mortgage rates. Green Tree rose to prominence by pioneering the securitization of manufactured housing loans, which means that it pooled large numbers of these loans and put them into a trust, which sold securities for which the loans served as collateral. The securities entitled the purchaser to fixed interest and principal payments under the loans. Importantly, Green Tree did not relinquish all rights under the securitized loans; instead, it retained the right to keep a portion of the loan payments and it retained the obligation to service the loans. Green Tree’s profit was therefore the spread between the interest it charged the borrowers and the interest it promised the instrument-holders, minus Green Tree’s costs of servicing the loans. However, Green Tree retained the risk of losses from loan defaults, as well as the risk that borrowers would prepay their loans before incurring interest charges, so that the expected interest would never materialize. By absorbing these risks, Green Tree could take high-interest-rate manufactured-housing loans and turn them into low-interest-rate securities, thereby creating the profitable spreads that fueled Green Tree’s growth.

As it securitized each loan pool, Green Tree booked a current gain on the transaction, even though the expected profits would not actually come into its coffers until much later and even though the amount of those profits would fluctuate with Green Tree’s success in collecting the loan payments. This “gain-on-sale revenue” was the force that drove Green Tree’s reported earnings during the periods covered by the complaints. Gain-on-sale receivables made up 63.8% of the net revenue that Green Tree originally reported for 1996. In addition to the income, Green Tree also recorded the present value of the securitizations as balance sheet assets called “Excess Servicing Rights Receivable” or “Interest Only Securities.” In 1996 “Excess Servicing Rights Receivable” was the greatest single asset on Green Tree’s balance sheet.

In order to estimate the present value of the securitizations, Green Tree had to assume three things: (1) the discount rate (reflecting the lesser value of a dollar to be paid in the future than a dollar paid today); (2) the loan default rate; and (3) the loan prepayment rate. The numbers chosen for these assumptions played a crucial role in deciding the amount Green Tree would report as earnings on the securitiza-tions and as the value of the Excess Servicing Rights Receivable. Choosing a figure for prepayment rate was especially difficult because if interest rates fell substantially or if other lenders competed aggressively, borrowers would be likely to refinance in large numbers. If that happened, the loans and consequently, the expected profit on them, would simply disappear. A discrepancy between assumed prepayment rates and actual prepayment experience could have a serious effect on Green Tree’s earnings and balance sheet.

The crux of this suit is the investors’ allegation that Green Tree used “unrealistic and unreasonable” assumptions in its gain-on-sale accounting, thus overvaluing its assets and overstating its earnings. In particular, the investors allege that the actual prepayment experience from 1995 to 1997 varied so much from the prepayment rate Green Tree assumed in its gain-on-sale accounting for 1994 and 1995 securiti-zations that Green Tree’s financials and other publicly filed reports during the 1995-1997 period were materially false. The investors concede that Green Tree publicly disclosed its actual prepayment experience. However, they allege the defendants refused to disclose what prepayment assumptions Green Tree had used in its gain-on-sale accounting and that this constituted omission of a material fact necessary to prevent its financial reports from being misleading. Green Tree stated in its public filings with the SEC that its management regularly reevaluated Green Tree’s prepayment assumptions in light of actual experience. According to the complaints, the defendants either knew that the discrepancy between the actual prepayment experience and the assumed rates rendered the financials materially false, or alternatively, recklessly disregarded this fact.

In any event, Green Tree’s assumptions about prepayment rates turned out to be seriously inadequate. Green Tree announced on November 13, 1997, that it planned to increase the prepayment reserve in its 1997 financials by an estimated $125-150 million because of higher than expected prepayment rates on manufactured housing loans. This announcement caused a precipitous decline in Green Tree’s stock price, from $42 to $30.75. The day after the announcement, in a conference call with industry analysts, CEO Coss disclosed the prepayment assumptions Green Tree had been using for the 1994 and 1995 pools. 4 He assured analysts, “I don’t see any set of circumstances that would cause us to [write down the 1994-1995 loan pools] again. And if we thought there was any set of circumstances that would, we would have provided more.” He also stated that a revision of prior earnings was not appropriate.

On January 27, 1998, Green Tree took an addition to its reserves of $190 million, rather than the $125-150 million addition to reserves anticipated in the November announcement. At the same time, it took a $200 million reduction of its previously reported 1996 earnings. In the press release accompanying the financials, Green Tree explained the decision to make larger adjustments than it had announced before; in reviewing its financial modeling, it “determined it had not fully considered the effect upon its interest only securities of partial prepayments (principal curtailments) and the impact that higher prepayments have on projected future interest due to investors on a weighted average basis.” Green Tree’s stock fell as low as $18.

On April 7, 1998, about two months after the close of the class periods, Green Tree announced that it would be acquired by Conseco, Inc., in a merger. On July 6, 1998, Conseco took charges totaling $350 million to write down the carrying value of Green Tree’s interest-only securities, reflecting adjustment to assumptions for prepayment rates, discount rates, and default rates.

The investors allege that several motives prompted the defendants to spread misleading financial information through the market. First, they allege that the compensation of defendant Coss and other Green Tree executives was tied to Green Tree’s financial results, which made the defendants want to maximize Green Tree’s reported earnings. In particular, CEO Coss had a remarkable contract awarding him 2.5% of Green Tree’s pre-tax income (minus other executives’ bonuses). Coss’s contract was to expire December 31, 1996. Accordingly, it was in Coss’s personal interest to report as much income as possible before the expiration of his lucrative contract. Other motives alleged are: (1) the defendants’ desire to maintain a high corporate credit rating to maximize the selling price of the loan pools; and (2) Green Tree’s need to demonstrate exemplary financial results in order to fend off a pending derivative suit alleging that the company wasted corporate assets by paying excessive executive compensation.

The stock purchasers allege that class members bought Green Tree stock from July 15, 1995 through January 27, 1998, and the options traders claim to have traded options from July 15,1995 though January 26, 1998. The Florida State Board alleges that it bought Green Tree stock from January 30, 1996 through January 27, 1998. The investors contend that they bought stock or traded options in a market in which the stock price was artificially inflated because of the defendants’ misrepresentations about Green Tree’s financial condition and that they lost money when the truth came out and the stock price fell.

II.

The stock purchasers’ class, the option traders’ class and the Florida State Board of Administration each filed a complaint alleging that Green Tree and the defendant officers of Green Tree violated section 10(b) of the Securities Exchange Act of 1934, 15 U.S.C. § 78j(b) (1994), and Rule 10b-5, 17 C.F.R. § 240.10b-5 (2001). They also alleged that the individual defendants were control persons of Green Tree and therefore were liable under section 20(a) of the Exchange Act, 15 U.S.C. § 78t(a) (1994), for Green Tree’s 10b-5 violations.

The defendants moved to dismiss the complaints for failure to state a claim un~ der Fed.R.Civ.P. 12(b)(6) and 9(b) and the pleading provisions of the Private Securities Litigation Reform Act of 1995, 15 U.S.C. §§ 78u-4 and 78u-5 (Supp. V 1999).

Applying section 78u-4(b)(l), 5 which requires that a complaint specify with particularity each statement alleged to have been misleading, the district court held that most of the investors’ factual allegations regarding the misleading statements and omissions were sufficiently particular to survive the dismissal motion. 61 F.Supp.2d at 871. The allegation that the court held not to be “particular” was the stock purchaser class’s allegation that Coss told an unnamed money manager in 1998 that Green Tree had been aware of the prepayment problem for some time, but had hoped it would go away. The district court held that this allegation did not state with particularity the “facts on which [the stock purchasers’] belief is formed,” 61 F.Supp.2d at 871 (quoting 15 U.S.C. § 78u-4(b)(l)), and therefore the court held that allegation did not comply with the statutory pleading requirements. Id. at 872.

The district court then turned to the central issue in this case-whether the complaints met the Reform Act’s requirements for pleading scienter, 15 U.S.C. § 78u-4(b)(2). 6 The district court held that the complaints failed to plead facts giving rise to a strong inference of scienter, as required by statute. 61 F.Supp.2d at 872.

Considering the allegations that the defendants were motivated to exaggerate Green Tree’s financial performance in order to increase the performance-based compensation of Green Tree’s executives, the court held that performance-based compensation alone does not give rise to a strong inference of scienter. Id. at 873 (citing Acito v. IMCERA Group, Inc., 47 F.3d 47, 54 (2d Cir.1995), and Shields v. Citytrust Bancorp. Inc., 25 F.3d 1124, 1130 (2d Cir.1994)). The unusual size of Coss’s performance incentives, while “clearly shocking,” id., did not provide the extra ingredient necessary to establish scienter. Id. Moreover, the court held that the complaints did not even show that the misrepresentations increased Coss’s compensation, since he had to give back a proportional amount of his 1996 compensation after Green Tree restated its 1996 financial results in 1998. Id. The 1995 results were not restated, so the district court held that there was no showing that they were misstated. Id. Coss and the other executives’ 1997 compensation was based on year-end results, which were figured after the revelation of the prepayment problems. Id. The investors argued that even though the defendants may not have actually profited from the misstatements, they could have hoped to do so, and this would supply the motive for fraud. The court responded that any such hope would have been so irrational that it could not be regarded as a motive. Id. at 874. Finally, any possible showing of motive to misrepresent would be rebutted by the fact that defendants Potts, Finn, and Gottesman bought Green Tree stock on the open market during the time period of the alleged fraud. Id. The court rejected the investors’ other theories of motive as well. Id.

The district court then considered whether the complaints contained facts that, in the absence of a showing of motive, nevertheless amounted to strong circumstantial evidence of scienter. The complaints allege that defendants monitored actual prepayment rates as they occurred throughout the periods covered in the complaints. Id. at 875. The investors’ theory was that once the defendants knew the actual prepayments for the 1994 and 1995 pools exceeded their assumed rates, their continued publication of financial statements based on assumptions they knew to be inadequate demonstrates intentional or reckless misrepresentation. Id. The district court held that the complexity of the accounting issue precluded this inference. Id. Amending the assumptions for pools that had already been sold was only one possible response to the problem; other responses would be to increase the assumed prepayment rate for future pools, which the defendants did beginning in 1996, and to increase prepayment reserves, which Green Tree did throughout the relevant period. Id. at 876. Therefore, the district court held that knowledge of the disparity between the prepayment assumptions and actual experience for the 1994 and 1995 pools would not satisfy the statutory pleading requirements without some additional evidence of wrongdoing. Id. at 877.

The investors sought to supply that additional evidence in three ways: (1) by pointing to the restatement of 1996 financial results as an admission that Green Tree knew those results were misleading when published; (2) by alleging that Green Tree’s accounting practices violated generally accepted accounting practices (GAAP) in various respects; and (3) by juxtaposing Green Tree’s publicized estimate of $125-150 million in charge-offs in November 1997 with its actual adjustments of $390 million in January 1998. The court held that the restatement of 1996 results did not amount to an admission of wrongdoing. Id. at 876 and 877 n. 21. The court further held that alleged GAAP violations had to do with making reasonable prepayment assumptions, which was a sufficiently complex task that defendants’ failure to divine the future accurately could not be equated with fraudulent intent. Id. at 877. Nor would the district court infer scienter from the disparity between Green Tree’s November 1997 estimate of charge-offs and the much greater actual charge-offs taken three months later. “[T]he November 13 announcement was expressly preliminary: it informed the market that the negative results expressed therein were not final, and that a final statement would be published after further review.” Id.

The district court concluded that the complaints did not adequately allege scien-ter and therefore dismissed them with prejudice. Id. at 878.

III.

Rule 10b-5, promulgated by the Securities and Exchange Commission under section 10(b) of the Securities Exchange Act, 15 U.S.C. § 78j(b), makes it unlawful

for any person directly or indirectly, by use of any means or instrumentality or interstate commerce, or of the mails or any facility of any national securities exchange,

(a) To employ any device, scheme, or artifice to defraud,

(b) To make any untrue statement of a material fact or to omit to state a material fact necessary in order to make the statements made, in light of the circumstances under which they were made, not misleading, or

(c) To engage in any act, practice, or course of business which operates or would operate as a fraud or deceit upon any person,

in connection with the purchase or sale of any security.

17 C.F.R. § 240.10b-5 (2001). Although not explicitly mentioned in the text of the rule, scienter is nevertheless an essential element of a Rule 10b-5 claim. Alpern v. UtiliCorp United, Inc., 84 F.3d 1525, 1534 (8th Cir.1996). Scienter means “the intent to deceive, manipulate, or defraud.” See Carlon v. Thaman (In re NationsMart Corp. Sec. Litig.), 130 F.3d 309, 320 (8th Cir.1997) (quoting Ernst & ilmst v. Hochfelder, 425 U.S. 185, 193, 96 S.Ct. 1375, 47 L.Ed.2d 668 (1976)). Our Circuit ruled before enactment of the Reform Act that recklessness satisfied the Rule 10b-5 scienter requirement, see Alpern, 84 F.3d at 1534, and the defendants do not contend that the Reform Act changed the substantive nature of the scienter requirement. 7 Our Circuit has adopted a definition of recklessness

limited to those highly unreasonable omissions or misrepresentations that involve not merely simple or even inexcusable negligence, but an extreme departure from the standards of ordinary care, and that present a danger of misleading buyers or sellers which is either known to the defendant or is so obvious that the defendant must have been aware of it.

Camp v. Dema, 948 F.2d 455, 461 (8th Cir.1991) (stating requirements for establishing aider and abetter liability under Rule 10b-5); see also Sundstrand Corp. v. Sun Chemical Corp., 553 F.2d 1033, 1045 (7th Cir.1977).

The question before us is whether the facts pleaded in the complaints adequately plead scienter. At the outset we must determine what pleading standard the investors have to satisfy. Complaints brought under Rule 10b-5 and section 10(b) are governed by special pleading rules unique "<to securities cases, which Congress adopted in the Private Securities Litigation Reform Act of 1995, P.L. No. 104-67, 109 Stat. 737, as part of its attempt to curb abuses of securities fraud litigation. In the Reform Act, Congress enacted two heightened pleading requirements for securities fraud cases. The first requires that the complaint specify each false statement or misleading omission and explain why the omission was misleading. 15 U.S.C. § 78u-4(b)(l) (Supp. V 1999). The second requires that the complaint state “with particularity” facts giving rise to a “strong inference” that the defendant acted with the scienter required for the cause of action. 15 U.S.C. § 78u-4(b)(2) (Supp. V 1999).

There is disagreement in the various circuits about the meaning of the strong-inference-of-scienter pleading standard. Our circuit has not yet decided the question, and the district court declined to decide it because the court concluded the investors had not framed an adequate complaint under any of the possible standards. 61 F.Supp.2d at 870. The investors contend that to establish a strong inference of scienter, they have only to plead either (1) facts showing the defendants had the motive and opportunity to commit fraud or (2) facts constituting strong circumstantial evidence of scienter. The defendants contend that a complaint cannot survive under the Reform Act based on allegations showing motive and opportunity alone. Additionally, the defendants contend that the investors have not pleaded facts showing motive and opportunity, or any other theory adequate to raise a strong inference of scienter.

Where the statutory language is susceptible of competing interpretations, we resort to legislative history to help determine congressional intent. Arkansas AFL-CIO v. F.C.C., 11 . F.3d 1430, 1440 (8th Cir.1993) (en banc); Dubois v. Thomas, 820 F.2d 943, 949 (8th Cir.1987).

The strong inference standard has its genesis in case law antedating the Reform Act. Since its adoption in 1937, Federal Rule of Civil Procedure 9(b) has imposed a special pleading standard for fraud: “In all averments of fraud or mistake, the circumstances constituting fraud or mistake shall be stated with particularity. Malice, intent, knowledge, and other condition of mind of a person may be averred generally.” See generally 2 James Wm. Moore, Moore’s Federal Practice Chapter 9 Historical App. (3d ed.2001). Even though Rule 9(b) said that state of mind could be averred generally (rather than with particularity), some circuits began to apply a special standard in securities cases, requiring pleading of facts that would indicate scienter, rather than the mere conclusion that the defendant acted with scienter. See, e.g., Romani v. Shearson, Lehman, Hutton, 929 F.2d 875, 878 (1st Cir.1991); In re Time Warner Inc. Securities Litig., 9 F.3d 259, 268 (2d Cir.1993).

In particular, the Second Circuit began to require plaintiffs in fraud cases to plead facts that “give rise to a ‘strong inference’ that the defendants possessed the requisite fraudulent intent.” Beck v. Manufacturers Hanover Trust Co., 820 F.2d 46, 50 (2d Cir.1987). In Beck, Judge Newman elaborated in a discussion that became the black-letter law of the Second Circuit:

A common method for establishing a strong inference of scienter is to allege facts showing a motive for committing fraud and a clear opportunity for doing so. See e.g., Goldman v. Belden, 754 F.2d 1059, 1070 (2d Cir.1985). Where motive is not apparent, it is still possible to plead scienter by identifying circumstances indicating conscious behavior by the defendant, though the strength of the circumstantial allegations must be correspondingly greater.

Id. (citations omitted and emphasis added).

From this discussion in Beck, it is not apparent that Judge Newman meant to say that pleading facts that show a motive and opportunity to commit fraud would, alone, satisfy Rule 9(b). The case Judge Newman cited, Goldman v. Belden, 754 F.2d 1059, 1070 (2d Cir.1985), would not have supported such a rule; in Goldman the complaint alleged that the defendants had held themselves out as knowledgeable about the kind of facts they were alleged to have withheld from the public. 754 F.2d at 1070. Thus, there were facts in addition to motive and opportunity that suggested knowing misconduct. So, Beck might have meant not that motive and opportunity alone are enough, but that without them, a plaintiffs burden is high indeed. However, in a 1993 case, Judge Newman revisited the point and ruled that a complaint was sufficient to plead Rule 10b-5 fraud when the only facts supporting scienter were the opportunity and the defendants’ motives to avoid jeopardizing corporate negotiations with prospective partners and to enhance their corporation’s stock price. In re Time Warner Inc. Sec. Litig., 9 F.3d 259, 269-71 (2d Cir.1993).

The rule that pleading motive and opportunity alone would give rise to a strong inference of scienter was remarkable for two reasons. First, having the motive and opportunity to do wrong are certainly not the same as having the intent to do it. See In re Comshare, Inc. Sec. Litig., 183 F.3d 542, 551 (6th Cir.1999) (quoted with approval in Greebel v. FTP Software, Inc., 194 F.3d 185, 197 (1st Cir.1999)). The rule, taken literally, presumes that anyone who has the chance to profit by wrongdoing is likely to do so. This is a large leap.

The second remarkable thing about the assertion that pleading motive and opportunity alone gives rise to a strong inference of fraud is the size of the breach it would leave in a pleading standard that was said to be stringent and tough. Nearly every highly ranked executive of a company could be said to have the motive and the opportunity to profit by misstatements. “Greed is a ubiquitous motive, and corporate insiders and upper management always have the opportunity to lie and manipulate.” Bryant v. Avado Brands, Inc., 187 F.3d 1271, 1286 (11th Cir.1999) (quoting Carley Capital Group v. Deloitte & Touche, L.L.P., 27 F.Supp.2d 1324, 1339 (N.D.Ga.1998)).

Apparently because of the potentially wide applicability of “motive and opportunity,” the Second Circuit soon retreated from a literal application of the rule. “Motive and opportunity” evolved into a term of art, meaning something far narrower than what it appears to mean. Within a year after Time Warner, the Second Circuit held that allegations of commonly held motives, even powerful ones such as the desire to stay CEO of a company, would not supply scienter under Rule 9(b):

If motive could be pleaded by alleging the defendant’s desire for continued employment, and opportunity by alleging the defendant’s authority to speak for the company, the required showing of motive and opportunity would be no realistic check on aspersions of fraud, and mere misguided optimism would become actionable under the securities laws.

Shields v. Citytrust Bancorp, Inc., 25 F.3d 1124, 1130 (2d Cir.1994). Nor would it constitute a pleading of “motive” to allege that the defendant wanted to maintain his or her company’s profitability or credit rating, either for its own sake or to increase thereby the defendant’s own compensation, see Chill v. Gen. Elec. Co., 101 F.3d 263, 268 (2d Cir.1996); Acito v. IMC-ERA Group, Inc., 47 F.3d 47, 54 (2d Cir.1995), even though common sense would suggest that these may be the very motives that prompt many cases of deceptive misstatements. Instead, the Second Circuit came to interpret “motive and opportunity” as requiring allegations that the “defendants benefitted in some concrete and personal way from the purported fraud,” Novak v. Kasaks, 216 F.3d 300, 307 (2d Cir.), cert. denied, 531 U.S. 1012, 121 S.Ct. 567, 148 L.Ed.2d 486 (2000), with the prototypic case being insider trading. Id. (“[I]n the ordinary case, adequate motive arose from the desire to profit from extensive insider sales.”). But see Press v. Chem. Inv. Serv. Corp., 166 F.3d 529, 538 (2d Cir.1999) (in case against three corporations, sufficient allegation of motive to aver that defendants would benefit from four-day “float” on a $100,000 T-bill). Yet, the difference between a case in which the defendant allegedly lied to increase his or her performance-driven compensation and one in which he or she allegedly lied to profit from insider trading is not so much a difference in the type of motive as a difference in the evidence that the defendant intended to capitalize on the opportunity. In either case, the defendant stands to make money (how much money depends entirely on the facts of the cases). But in the insider trading case, trading at a particular time is circumstantial evidence that the insider knew the best time to trade because he or she had inside information not shared by the public. This in turn is circumstantial evidence that he or she kept the information from the public in order to trade on the unfair advantage. See e.g., Stevelman v. Alias Research Inc., 174 F.8d 79, 85 (2d Cir.1999) (holding that allegation that officers sold off large portions of stockholdings during period of alleged misrepresentations “supports the inference that [CEO] withheld disclosures that would depress his stock until he had profitably sold his shares”); Greebel, 194 F.3d at 197 & n. 10. In contrast, in the executive compensation case, there is no telltale action (other than the false statement) to suggest guilty knowledge or purpose.

Whatever the difficulties in applying the Second Circuit’s strong-inference-of-scien-ter standard, when Congress decided to reform securities litigation to eliminate strike suits, it regarded the Second Circuit standard as the “most stringent pleading standard” in the country. S.Rep. No. 104-98, at 15 (1995), reprinted in 1995 U.S.C.C.A.N. 679, 694. Cf. Maldonado v. Dominguez, 137 F.3d 1, 10 n. 6 (1st Cir.1998) (questioning whether Second Circuit standard was as rigorous as First Circuit standard). The Senate Committee intended to impose the Second Circuit standard in securities fraud cases in all circuits. The Committee chose to accomplish this by requiring, “In any private action to recover money damages, the plaintiff must, for each misstatement or omission, specifically allege facts giving rise to a strong inference that the defendant acted with the required state of mind.” S. Rep. 104-98, at 26, reprinted in 1995 U.S.C.C.A.N. at 705. However, the Committee expressly chose not “to codify the Second Circuit’s caselaw interpreting this pleading standard, although courts may find this body of law instructive.” Id. at 15, reprinted in 1995 U.S.C.C.A.N. at 694.

The subsequent legislative history shows a great deal of confusion about whether the strong-inference-of-fraud standard was the same standard with the motive-and-opportunity test as without it. Senator Specter clearly thought it was not. He introduced an amendment to add the motive-and-opportunity language into the text of the bill. 141 Cong. Rec. 17,286 (1995). 8 He reasoned that without this caveat, the strong-inferenee-of-scienter standard posed an impossible task for plaintiffs. “How do you get into somebody else’s head?” he asked. 141 Cong. Rec. 17,424 (1995). It is not clear, however, if Senator Specter’s objection was that the standard was simply so vague that it might be interpreted unevenly or whether he believed the motive-and-opportunity language was a substantive relaxation of the strong-inference standard. 9

The Specter amendment passed on the Senate floor, 141 Cong. Rec. 17,425 (1995), only to be deleted in the Conference Committee. The Conference Report stated: “Because the Conference Committee intends to strengthen existing pleading requirements, it does not intend to codify the Second Circuit’s case law interpreting this pleading standard.” H. Conf. Rep. No. 104-369, at 41 (1995), reprinted in 1995 U.S.C.C.A.N. 730, 740. In a footnote, the report adverted to the deletion of the Specter amendment: “For this reason, the Conference Report chose not to include in the pleading standard certain language relating to motive, opportunity, or recklessness.” Id. at 48 n. 23, reprinted in 1995 U.S.C.C.A.N. at 748.

Although the import of the Conference Report seems to be that any reference to the motive-and-opportunity test applied in the Second Circuit was disapproved, the legislative history does not support that reading. Admittedly, the opponents of the bill, including President Clinton, characterized the deletion of the Specter amendment as a decision to raise the pleading standard above the Second Circuit standard. 10 E.g., 141 Cong. Rec. at 38,194 (Clinton veto message); 141 Cong. Rec. at 88,205-07 (Senator Specter); 141 Cong. Rec. at 35,299 (Senator Biden). However, Senator Dodd, a manager of the bill, contended that the Specter amendment was not deleted because the Conference Committee wanted to disavow the Second Circuit’s case law, but because the Specter amendment did not codify the case law accurately. 141 Cong. Rec. at 35,266. Moreover, the part of the formulation that was not codified to Senator Dodd’s satisfaction was not the motive-and-opportunity part, but the second part of the test, which required strong circumstantial evidence of conscious or reckless misbehavior. Senator Dodd objected that the part of Judge Newman’s formulation calling for a stronger circumstantial showing where motive is not apparent had been omitted from the Specter amendment. Id.; 141 Cong. Rec. at 38,228. Taken at face value, Senator Dodd’s defense of the bill suggests that he had no objection to the motive-and-opportunity part of the Specter amendment. Senator Dodd continued to argue that courts should look to the existing Second Circuit caselaw for guidance in applying the strong-inference-of-scienter standard. 141 Cong. Rec. at 38,228. More broadly, Senators Dodd and Bradley both argued in support of the bill that the Second Circuit caselaw was sufficiently ambiguous that any attempt to further codify it would bog the bill down. 141 Cong. Rec. at 38,228 and 38,322. The bill as submitted by the Conference Committee was passed over the President’s veto. 141 Cong. Rec. at 38,354 (Senate); 141 Cong. Rec. at 37,807 (House).

In sum, the legislative history provides nothing but uncertainty about whether the motive-and-opportunity test is an inherent part of the strong inference standard or whether it modifies and relaxes that standard. Ultimately, the Act as passed does not resolve this question. 11

The Circuits that have interpreted the Reform Act have fallen into (at least) three camps. The Ninth Circuit inferred from the deletion of the Specter amendment by the Conference Committee that Congress expressly rejected the Second Circuit’s “two-pronged test” and raised the pleading standard to a level beyond that in the Second Circuit. In re Silicon Graphics Inc. Sec. Litig., 183 F.3d 970, 978-79 (9th Cir.1999).

At the other end of the spectrum, the Second and Third Circuits have held that in using the “strong inference” language, the Reform Act adopted a pleading standard “approximately equal in stringency to that of the Second Circuit,” In re Advanta Corp. Sec. Litig., 180 F.3d 525, 534 (3d Cir.1999); Novak v. Kasaks, 216 F.3d 300, 310 (2d Cir.2000), including the motive- and-opportunity formulation, Advanta, 180 F.3d at 534-35; Ganino v. Citizens Util. Co., 228 F.3d 154, 168-70 (2d Cir.2000). The Second Circuit qualified its conclusion by stating that “Congress’s failure to include language about motive and opportunity suggests that we need not be wedded to these concepts in articulating the prevailing standard,” Novak, 216 F.3d at 310, but that courts should nevertheless be guided by the motive-and-opportunity cases. 12 Id. at 311; cf. id. at 307 (referring to “complexity and uncertainty” involved in applying Second Circuit motive-and-opportunity test). The Second and Third Circuits also agree that the Reform Act’s additional requirement that facts pertaining to scienter be pleaded “with particularity” represents a “heightening of the [pleading] standard.” Novak, 216 F.3d at 810-11; Advanta, 180 F.3d at 534.

Occupying the middle ground, the First, Fifth, Sixth, Tenth, and Eleventh Circuits have held that the primary effect of the Reform Act is to require a pleading to state facts giving rise to a “strong inference of scienter.” Allegations of motive and opportunity may or may not rise to that level in a particular- case. See Nathenson v. Zonagen Inc., 267 F.3d 400, 409-11 (5th Cir.2001); City of Philadelphia v. Fleming Cos., 264 F.3d 1245, 1261-63 (10th Cir.2001); Helwig v. Vencor, Inc., 251 F.3d 540, 551 (6th Cir.2001) (en banc); Greebel v. FTP Software, Inc., 194 F.3d 185, 197 (1st Cir.1999); Bryant v. Avado Brands, Inc., 187 F.3d 1271, 1282-83 (11th Cir.1999). “While it is true that motive and opportunity are not substitutes for a showing of recklessness, they can be catalysts to fraud and so serve as external markers to the required state of mind.... ‘[MJotive’ and ‘opportunity’ are simply recurring patterns of evidence.” Helwig, 251 F.3d at 550.

Putting aside the Ninth Circuit standard, which gives the deletion of the Specter amendment a more pointed reading than it will bear, the split in the other Circuits is more apparent than real. As we discussed above, supra at 656, the Second Circuit has dramatically constricted the types of “motive and opportunity” that it will recognize as sufficient to plead scienter. It will not allow plaintiffs to proceed based on widely held motives such as “(1) the desire to maintain a high corporate credit rating or otherwise sustain ‘the appearance of corporate profitability, or of the success of an investment, [or] (2) the desire to maintain a high stock price in order to increase executive compensation or prolong the benefits of holding corporate office.” Novak, 216 F.3d at 307 (internal citations omitted). Even complaints based on insider trading must allege more than that the defendant benefitted from trading because of a false statement or misleading omission; the insider trades have to be “unusual,” either in the amount of profit made, the amount of stock traded, the portion of stockholdings sold, or the number of insiders involved, before they will give rise to the required inference of scienter. See Rothman v. Gregor, 220 F.3d 81, 94 (2d Cir.2000); In re Advanta, 180 F.3d at 540-541; Oran v. Stafford, 226 F.3d 275, 290 (3d Cir.2000). This is the same kind of inquiry undertaken by courts that do not adhere to the motive-and-opportunity formulation. See Nathenson, 267 F.3d at 420; Helwig, 251 F.3d at 552; Greebel, 194 F.3d at 198. The search in the Second Circuit line of cases, as well as in the other circuits, is for facts that give a strong reason to believe that there was reckless or intentional wrongdoing. See Greebel, 194 F.3d at 197 (facts must support strong inference of scienter, rather than merely reasonable inference). Taken as a whole, the cases simply do not substantiate the fear that courts applying the motive-and-opportunity formulation will permit pleadings to go forward without facts strongly suggesting wrongdoing.

The Reform Act itself adopted only the strong-inference-of-scienter standard, without codifying the particular methods of satisfying the standard. The Second Circuit cases interpreting that language, both leading up to the Reform Act and after its enactment, give us valuable guidance about what factors help to establish such an inference. At the same time, we also look to case law from other circuits developing their own criteria for the badges of fraud. E.g., Greebel, 194 F.3d at 196-98; Helwig, 251 F.3d at 552. Ultimately, however, we must take care to use subsidiary formulae as an aid to interpreting the strong-inference standard and not as a substitute for it.

Therefore, we can say three things about motive and opportunity allegations. First, motive and opportunity are generally relevant to a fraud case, and a showing of unusual or heightened motive will often form an important part of a complaint that meets the Reform Act standard. Second, in some cases the same circumstantial allegations that establish motive and opportunity also give additional reason to believe the defendant’s misrepresentation was knowing or reckless. For instance, in insider trading cases, the timing of trades shows motive and opportunity, but it may also provide additional circumstantial evidence that the defendant knew of an advantage. Such allegations may meet the Reform Act standard, but if so it is because they give rise to a strong inference of scienter, not merely because they establish motive and opportunity. Third, when the complaint does not show motive and opportunity of any sort-either the unusual, heightened motive highlighted in the Second Circuit cases, or even an everyday motive such as keeping one’s job-then other allegations tending to show scienter would have to be particularly strong in order to meet the Reform Act standard. Cf Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986) (in Sherman Act context, where there appeared to be no rational motive to break the law, a plaintiff would have to adduce a particularly strong case in order to survive summary judgment).

IV.

This case was decided under both the Federal Rules of Civil Procedure and the Reform Act. The Reform Act modifies the ordinary Rule 12(b)(6) dismissal mechanism in two limited ways. First, whereas under Rule 12(b)(6), we must assume all factual allegations in the complaint are true, see Goss v. City of Little Rock, 90 F.3d 306, 308 (8th Cir.1996), under the Reform Act, we -disregard “catch-all” or “blanket” assertions that do not live up to the particularity requirements of the statute. See Advanta, 180 F.3d at 535. Cf. Parnes v. Gateway 2000, Inc., 122 F.3d 539, 550 (8th Cir.1997) (under Rule 9(b) district court properly struck conclusory allegations).

Second, under Fed.R.Civ.P. 12(b)(6), the plaintiff is entitled to all reasonable inferences that may be drawn from the allegations of the complaint. Young v. City of St. Charles, 244 F.3d 623, 627 (8th Cir.2001). However, under the Reform Act, a securities fraud case cannot survive unless its allegations collectively add up to a strong inference of the required state of mind.

Congress has effectively mandated a special standard for measuring whether allegations of scienter survive a motion to dismiss. While under Rule 12(b)(6) all inferences must be drawn in plaintiffs’ favor, inferences of scienter do not survive if they are merely reasonable .... Rather, inferences of scienter survive a motion to dismiss only if they are both reasonable and “strong” inferences.

Greebel, 194 F.3d at 195-96. See Helwig, 251 F.3d at 553 (“[T]he strong inference requirement means that plaintiffs are entitled only to the most plausible of competing inferences.”).

We review de novo the district court’s dismissal of complaints pursuant to Rule 12(b)(6), Breedlove v. Earthgrains Baking Cos., 140 F.3d 797, 799 (8th Cir. 1998), and by analogy we will apply the same standard of review to dismissal pursuant to the Reform Act, 15 U.S.C. § 78u-4(b)(3)(A).

V.

We must evaluate the cogency of the investors’ allegations against the strong inference standard. The investors argue that Coss and the other executives’ performance-based compensation gave the defendants a motive not to disclose the prepayment problems. 13 The stock purchasers class, for example, alleged:

In 1995 and 1996, Green Tree had the highest paid business executive in the entire United States. In 1995 and 1996, defendant Coss was paid, in cash and deeply discounted stock, $65 million and $102 million, respectively. Under the terms of his 1991 employment agreement, which was in effect until the end of 1996, Coss was paid 2.5% of the Company’s pre-tax income, in addition to a base salary of over $430,000. After Green Tree restated its 1996 earnings, Coss was required to repay $25.9 million of his 1996 bonus.

As a result of the approaching expiration of the 1991 employment agreement, Coss knew that 1996 would be the last year for which he would receive bonus compensation valued in the tens of millions of dollars_Coss’s 1997 compensation was valued at approximately $4.8 million, about $97 million less than the value of his original 1996 compensation.

Even in the Second Circuit, pleading that a defendant’s compensation depends on corporate value or earnings does not, by itself, establish motive to fraudulently misrepresent corporate value or earnings. Novak v. Kasaks, 216 F.3d 300, 307 (2d Cir.2000); Acito v. IMCERA Group, Inc., 47 F.3d 47, 54 (2d Cir.1995). Here, the investors allege that Coss’s arrangement established a legally significant motive because the amount of his compensation was based on a percentage of Green Tree’s pre-tax earnings, and his contract was set to expire at the end of 1996, making it urgent for Coss to maximize Green Tree’s earnings for that year. Green Tree originally booked earnings for 1996 that allowed Coss to earn $102 million. Roughly a year later, Green Tree found it necessary to revise its 1996 earnings retrospectively.

We conclude that the magnitude of Coss’s compensation package, together with the timing coincidence of an overstatement of earnings at just the right time to benefit Coss, provides an unusual, heightened showing of motive to commit fraud. The motive and opportunity allegations regarding Coss are therefore an important part of a circumstantial case against him and consequently, against Green Tree, which was subject to Coss’s control.

The defendants contend that Coss’s compensation is irrelevant because Coss wound up giving back a proportional amount of his 1996 compensation after the 1996 earnings were restated in January 1998. But Coss did not necessarily know at the time of the alleged misrepresenta tions and omissions that it would turn out that way. He could have acted recklessly to pile up earnings before his contract ran out, gambling that he would get away with it. The ultimate profitability of a course of conduct is not conclusive of intent. Just as we cannot countenance pleading fraud by hindsight, neither can we infer innocence by hindsight because the alleged misdeeds did not pay off.

The district court held that, even without such hindsight, such a gamble was so unlikely to succeed that the court would not infer that Coss tried it:

[Tjhere is no allegation in the pleadings from which it can be inferred that defendants could have concealed (forever) a significant gap between projected and actual earnings, even if prepayment speeds eventually had slowed. Furthermore, this alleged motive is premised on mere hope: even assuming that defendants knew that the company’s earnings were inflated and that a later turnaround could have been used to conceal past poor performance, plaintiffs allege nq facts which suggest that defendants had reason to believe that economic forces beyond their control would change dramatically enough to correct the alleged earnings woes.

61 F.Supp.2d at 873-74. To the contrary, the allegations provide powerful support for the idea that defendants hoped that the prepayment problem would come out in the wash-why else would they have waited until 1998 to reveal and remedy the problem? The complaints plead that defendants knew their assumed prepayment rate and they knew their actual rates as those rates were unfolding. This amounts to knowledge of the prepayment discrepancy. If the valuation of the interest-only securities turned out to be significantly wrong, there can be no question that it would be material to the accuracy of Green Tree’s financials. Assuming the defendants knew of the prepayment discrepancy, if we deprive them of the expectation that this discrepancy would be offset by other factors or later experience, we deprive them of an innocent explanation for their failure to correct their financials. Indeed, the defendants’ theory of this case is that they failed to correct the financials because the accounting was sufficiently complex that they reasonably believed the financials were fairly stated despite the prepayment discrepancy. Somewhere in between the defendants’ theory that the assumptions were still reasonable and the district court’s theory that the defendants could not have hoped to conceal the inadequacy of the assumptions, lies the investors’ theory. The investors allege that the defendants knew facts that showed the assumptions were materially inadequate, but the defendants recklessly attempted to sweep the problem under the rug hoping a change in the economy would ameliorate the problem. We cannot conclude that this is a case in which the motive theory is too irrational to add to the weight of other circumstantial allegations, and we therefore must disagree with the district court’s determination to the contrary.

Notably, the complaints do not allege that the executives sold their stock during the class periods. 14 Indeed, the defen dants point to documents filed with the S.E.C. indicating that the executive defendants owned more Green Tree stock at the end of the class period than at the beginning. Defendants Finn, Gottesman, and Potts bought more than $870,000 in Green Tree stock during the stock and option class period. Apparently, only Richard Evans sold stock during the class period. The defendants argue that these purchases are so inconsistent with the idea that defendants intentionally inflated the value of Green Tree stock that they, “defeat” the investors’ pleading of motive.

The shareholders respond that the executives’ stock transactions are outside the four corners of the complaints and are therefore not to be considered on a motion to dismiss. The executives’ stock transactions are documented in public filings required to be filed with the SEC, which other circuits have considered on a motion to dismiss. See Rothman v. Gregor, 220 F.3d 81, 88 (2d Cir.2000); Bryant v. Avado Brands, Inc., 187 F.3d 1271, 1275-81 (11th Cir.1999) (taking judicial notice under Fed. R.Evid. 201);' Advanta, 180 F.3d at 540; Lovelace v. Software Spectrum Inc., 78 F.3d 1015, 1017-18 (5th Cir.1996). The investors do not dispute the accuracy of the defendants’ assertions that they bought stock, and in fact the shareholders actually argue that the defendants’ purchases are relevant as a cover-up.

But whether or not the documents detailing defendants’ trading history can be considered on motion to dismiss, they do not neutralize the motive allegations because they do not directly contradict them. In cases in which plaintiffs sought to use stock transactions as evidence of motive, the Second Circuit has held that certain defendants’ abstention from trading stock so undercut the investors’ allegations of motive that the complaints failed to state a claim. Acito, 47 F.3d at 54; San Leandro Emerg. Med. Plan v. Philip Morris Cos., 75 F.3d 801, 814 (2d Cir.1996); accord Advanta, 180 F.3d at 539-40; see also Nathenson, 267 F.3d at 420; Ronconi v. Larkin, 253 F.3d 423, 436 (9th Cir.2001); Helwig, 251 F.3d at 571-72 (Kennedy, J., dissenting). In this case, however, the investors do not rely on allegations of inside transactions, but on other ■ motives, such as the hope of huge bonuses, that are not directly undermined by their abstention from trading.

Coss argues that since his bonus was in the form of stock, not cash, he would have only benefitted from the fraud by selling his stock. This argument wounds but does not defeat the investors’ theory of Coss’s motive, since the investors allege that Coss’s bonus was in cash as well as stock. We cannot resolve a factual dispute on a motion to dismiss. Furthermore, getting the stock bonus would have been a first step in Coss’s plan to benefit from manipulating Green Tree’s earnings. We may not dismiss the investors’ complaint merely because the alleged plan did not come to fruition. The defendants may well make effective use at trial of the fact that they held their stock, but this fact does not neutralize the investors’ other motive allegations for purposes of the motion to dismiss.

The other executives besides Coss object that they did not share Coss’s lucrative employment contract and that it is not plausible that they would commit fraud to enrich someone else. The compensation allegations against the other executives are obviously weaker evidence of motive than those against Coss and Green Tree. The investors did not plead that the other executives had incentive contracts that were about to expire or that their compensation was comparable to Coss’s. Potts was- alleged to have received a bonus of $1.25 million in 1996 and Richard Evans $350,000. The option traders argue, “The primary motive was to enrich Coss, and along the way a trickle down effect would be felt in the bonuses of the other executives.” This falls short of alleging the “concrete and personal” benefit that would add significantly to the allegations against the other executives.

The investors also allege that the defendants were motivated by the desire to maintain a high credit rating for Green Tree so that its securitizations would continue to sell at good prices. The desire to maintain a high credit rating is universally held among corporations and their executives and consequently does not contribute significantly to an inference of scienter. Novak, 216 F.3d at 307; San Leandro, 75 F.3d at 814.

The investors claim that defendants were motivated during the class period to show exemplary performance by Green Tree to enhance their defense of a derivative suit for corporate waste that was not filed until January 23, 1997, a year and a half into the class period alleged by the stockholders’ and option traders’ classes. The investors seem to concede that defense of a non-existent suit cannot supply a motive for the first year and a half of the alleged misrepresentation. Moreover, the defendants made their November 1997 and January 1998 revelations while the state suit was pending. While the pendency of a derivative suit could supply an important motive in some circumstances, we cannot say that it automatically satisfies the scien-ter requirement. The inference to be drawn depends on the facts of the case, and the timing issues in this case undermine the notion that the derivative suit played an important role in the events at issue.

The options traders’ class and the shareholders’ class also argue in their briefs that the defendants were motivated to overstate Green Tree’s financial results in order to make it more attractive to a potential buyer. They did not plead any such facts. The district court said that even if they had pleaded this theory, it would have been too implausible to suggest that Green Tree hoped a buyer would purchase the company without performing its own examination of Green Tree’s financial status. 61 F.Supp.2d at 874. Another problem with the theory is that the only merger on the horizon was the merger with Conseco, Inc., a transaction announced April 7, 1988-months after the November 1997 and January 1998 revelations. This is too thin a reed on which to hang an inference of scienter existing during the class period. Cf. Rothman, 220 F.3d at 94 (whether motive to inflate stock sufficiently pleaded by alleging impending acquisition “might well depend on the particular circumstances of the case”).

While this is not the kind of case in which the facts showing motive and opportunity alone create a strong inference of scienter, nevertheless, Coss’s extraordinary compensation package, on the very eve of expiring, created a powerful incentive to see to it that Green Tree made plenty of money before his contract expired. For Coss and for Green Tree, this factor is an important part of the overall picture of scienter. It appears that the investors’ motive allegations against the other executives are basically derivative from their case against Coss. They argue that “Coss was the dominant force at Green Tree” and that the Green Tree board of directors lacked independence from Coss. Although these contentions do not amount to a showing of an unusual, heightened motive for the other executives, they at least paint a rational and plausible picture of why the executives would have gone along with the alleged fraud. Cf. Kalnit v. Eichler, 264 F.3d 131, 139-40 (2d Cir.2001) (motive allegation nonsensical). Thus, if the motive allegations do not add much to the investors’ claims against the other executives, neither is there an utter lack of rational motive, which would require the investors to make a stronger circumstantial case than would otherwise be necessary.

VI.

In addition to the motive-and-opportunity allegations, the complaints contain other facts that give rise to a strong inference of fraud. In broad outline, the investors pleaded with particularity that defendants knew the prepayment assumptions on which their gain-on-sale accounting for the 1994 and 1995 pools were based, they knew that actual experience in 1995, 1996 and the first three quarters of 1997 deviated from those premises greatly, and they issued financials that did not take account of the disparity between the assumptions and actual experience. Specifically, Green Tree’s gain-on-sale accounting assumed that the 1994 and 1995 pools would experience prepayment at a rate referred to as “100 MHP,” 15 meaning roughly that after twenty-four months the borrowers would be prepaying the loans at the rate of 6% per annum. In fact, the loans were being prepaid at a much faster rate, with the majority of the pools running well above the 6% per annum rate at ten months. “At 24 months, the loans were being prepaid at a rate of between 9.0% and 16.30%.” The shareholders have appended to their complaint a chart indicating that the problem was pronounced as early as July 1995. The complaints alleged that the defendants were informed of the actual prepayment rates “on at least a monthly basis.” .Green Tree’s SEC filings stated: “Actual default and prepayment experience are reviewed quarterly.” In the November 14, 1997 conference call with industry analysts, Coss stated that Green Tree had been taking additional reserves in connection with this prepayment problem since the fourth quarter of 1996. There is no dispute that the prepayment problem led to Green Tree’s $390 million write-down in January 1998.

One of the classic fact patterns giving rise to a strong inference of scienter is that defendants published statements when they knew facts or had access to information suggesting that their public statements were materially inaccurate. See Novak v. Kasaks, 216 F.3d 300, 311 (2d Cir.), cert, denied, 531 U.S. 1012, 121 S.Ct. 567, 148 L.Ed.2d 486 (2000); Hollín v. Scholastic Corp. (In re Scholastic Corp. Sec. Litig.), 252 F.3d 63, 76 (2d Cir.2001), petition for cert. filed, 70 U.S.L.W. 3164 (U.S. Aug. 30, 2001) (No. 01-397); Howard v. Everex Sys. Inc., 228 F.3d 1057, 1064 (9th Cir.2000); City of Philadelphia v. Fleming Cos., 264 F.3d 1245, 1260-61 (10th Cir.2001) (in case of omissions, scienter proved by knowledge of omitted fact plus knowledge that omission likely to mislead). Here, the investors alleged that defendants published statements with knowledge of facts indicating crucial information in the statements was based on discredited assumptions. These allegations give rise to a strong inference of scienter. See Rothman v. Gregor, 220 F.3d 81, 91 (2d Cir.2000) (where defendants regularly assessed likelihood of recouping investment in asset, failure to ex pense investment that obviously would not be recouped supported strong inference of scienter). Additionally, the sheer size of the $390 million write-down adds to the inference that the defendants must have been aware the problem was brewing. See Rothman, 220 F.3d at 92; In re Scholastic Corp., 252 F.3d at 77.

The defendants respond that this is too simplistic a view of a case involving sophisticated financial transactions in which the accounting decisions were difficult and even very large mistakes could fall within the realm of good faith. The defendants contend:

[T]he securities that Green Tree issues are complex, with multiple layers (tranches) of debt at varying rates, making all of Green Tree’s valuation calculations complex. For example, with respect to the “weighted average’ interest rate issue [which Green Tree cited as causing greater write-downs than anticipated in November 1997], prepayments cause high priority tranches of the securities (which bear lower interest rates) to pay down more quickly, so the weighted average interest rate to be paid out to investors in the future increases.

Undoubtedly, the accounting issues are complex; whether they were handled within the parameters of good faith decision-making or whether the decisions amounted to recklessness will surely be the focus of any trial of this case. We will not prejudge that issue. But neither the district court, nor we, can conduct a battle of experts on a motion to dismiss. Rather, we must assume the truth of the allegations pleaded with particularity in the complaint. The strong-inference pleading standard does not license us to resolve disputed facts at this stage of the case.

The defendants contend that they did respond to the discrepancy between actual and assumed prepayment rates by adding to their reserves. 16 This response does not rob the investors’ allegations of their effect. First, although adding to reserves could correct the discrepancy, in light of the $390 million write-down announced in January 1998, it is clear that whatever response defendants took was not adequate. It was, after all, necessary to amend the 1996 financials retrospectively in 1998 to reduce reported earnings by $200 million. Moreover, in the November 14, 1997 conference call, Coss said that Green Tree had begun increasing reserves in response to the prepayment problem in the third quarter of 1996. This supports the inference that Green Tree did not begin responding to the prepayment problem until the end of 1996, whereas the investors have alleged that the problem was apparent by July 1995. Coss also characterized those additions to reserves as “relatively modest.” Whether defendants could have believed during the class period that the reserves were an adequate response is a question of fact that cannot render the complaints inadequate, lest the heightened pleading requirements of the Reform Act replace the function of a trial.

The investors have pleaded that during the various class periods, the defendants had in their possession facts that rendered their financial results materially false when they published them. Additionally, the investors have pleaded facts giving Coss, the dominant force within Green Tree, a substantial and urgent reason to delay owning up to the prepayment problem. Although the defendants other than Green Tree and Coss had less of a personal interest in putting off the hour of reckoning, the investors’ theory that the others were subject to domination by Coss fits within a highly plausible theory of fraud. While the defendants have many good arguments explaining how and why they thought their statements proper, none of these arguments saps the investors’ allegations of their force at this procedural stage. The facts pleaded with particularity add up to a strong inference of scienter, thus meeting the Reform Act standard. 17

VII.

Finally, the shareholders take issue with the district court’s holding that paragraph 178 of their complaint lacks the particularity required by the Reform Act. 61 F.Supp.2d at 871-72 (citing 15 U.S.C. § 78u-4(b)(l)). Paragraph 178 alleges:

[T]hroughout the Class Period, defendants knew or recklessly disregarded that Green Tree’s revenue, income and earnings per share were materially overstated and that critical assumptions used in calculating gain-on-sale revenue were unreasonable and contradicted by actual experience. For example, in or around January, 1998, defendant Coss had a telephone conversation with a money manager, in which Coss admitted that Green Tree had been aware of the Company’s prepayment problem for some time but had delayed making any sort of disclosure until November, 1997 because he had hoped that the problem would go away. In the same telephone call, Coss also admitted that defendants only decided to disclose the prepayment problem to the investing public when it became apparent that the situation was getting worse.

Contrary to the district court’s and the shareholders’ assumption, it is clear that this allegation goes to scienter, not to the making of a misrepresentation, and therefore is governed by section 78u-4(b)(2), which requires each fact supporting the inference of scienter to be stated with particularity. Therefore, we need not interpret the requirement of section 78u-4(b)(1), which states, “[I]f an allegation regarding the statement or omission [alleged to be false or misleading] is made on information and belief, the complaint shall state with particularity all facts on which that belief is formed.” (Emphasis added.) The question before us is not whether a plaintiff is truly compelled to plead all facts on which it bases its information and belief, see Novak, 216 F.3d at 313-14 (holding section 78u-4(b)(l) does not literally require pleading of all facts, so long as facts pleaded provide adequate basis for believing statements were false); instead, we must decide whether allegations quoting an individual and describing him generally, but not identifying him by name, satisfy the particularity requirement of subsection (b)(2).

Whether pleading with particularity requires the identification of the speaker whose words are relied on is not apparent from the face of the statute. Some opponents of the Reform Act argued in the House of Representatives that the House version of the Reform Act would require pleading names of confidential informants. See, e.g., 141 Cong. Rec. H2849 (March 8, 1995) (statement of Rep. Dingell). The Second Circuit sensibly refused to give weight to these “hyperbolic statements of legislators attempting (unsuccessfully) to amend the proposed act to lighten plaintiffs’ pleading burden.” Novak v. Kasaks, 216 F.3d 300, 313 (2d Cir.), cert, denied, 531 U.S. 1012, 121 S.Ct. 567, 148 L.Ed.2d 486 (2000). Our pre-Reform Act case of Parnes v. Gateway 2000, Inc., 122 F.3d 539, 549 (8th Cir.1997), interpreted the particularity requirement of Rule 9(b) to entail pleading of “the who, what, when, where, and how: the first paragraph of any newspaper story.” Id. (quoting DiLeo v. Ernst & Young, 901 F.2d 624, 627 (7th Cir.1990)). There, we quoted First Circuit authority that an allegation based on information and belief should set forth the source of the information. Id. (quoting Romani v. Shearson Lehman Hutton, 929 F.2d 875, 878 (1st Cir.1991)). This rule was developed as a corollary to the rule that allegations of fraud regarding matters peculiarly within the opposing party’s knowledge could be based on information and belief, so long as accompanied by a statement of the facts on which the belief was founded. See Luce v. Edelstein, 802 F.2d 49, 54 n. 1 (2d Cir.1986); Harold S. Bloomenthal & Samuel Wolff, Securities and Federal Corporate Law § 16:12 n. 9, 16:25.6 n. 2 (1998). One treatise refers to the rule permitting information and belief pleading as “providing] a road map around Rule 9(b) by asserting allegations based on speculation as to the underlying facts.” Bloomenthal & Wolff § 16:25.6. The treatise continues that this loophole was aggravated by the provision of Fed. R.Civ.P. 11(b) “under which counsel can attempt to avoid the sanction bullet by certifying that the allegations ‘are likely to have evidentiary support after a reasonable opportunity for further investigation or discovery.” ’ Bloomenthal & Wolff § 16:25.6. Since a primary goal of the Reform Act was to prevent baseless strike suits, S. Rpt. No. 104-98 at 4, reprinted in 1995 U.S.C.C.A.N. at 683, our practice under Rule 9(b) of requiring disclosure of facts relied on for pleadings based on information and belief should obviously be continued under the Reform Act.

However, the part of paragraph 178 at issue here is not explicitly pleaded on information and belief about facts peculiarly within the defendants’ control, nor does it state that the allegation is contingent on further discovery. Cf. Lirette v. Shiva, 999 F.Supp. 164, 165 (D.Mass.1998) (distinguishing between allegations based on information and belief and those “supported by some document or statement on personal knowledge by a potential witness”), quoted in Bloomenthal & Wolf at § 16:25:10. Paragraph 178 of the shareholders’ complaint recites details of a particular conversation with a particular person. If the shareholders’ attorneys do not have a factual basis for this allegation, they will be subject to Rule 11 sanctions. Therefore, there is already a mechanism in place to deter and punish fabrication. Consequently, there is no need for the name of the money manager to be pleaded in the complaint, although it appears it will soon have to be disclosed under Fed. R.Civ.P. 26(a)(1)(A) if the shareholders intend to make use of it.

Though we conclude that the allegation about the unknown money manager is adequately particular, it is not an especially valuable addition to the shareholders’ showing of scienter. They have already alleged that Coss knew the prepayment assumptions and the actual prepayment experience. Coss said in the November 14, 1997 conference call that Green Tree had been monitoring the problem carefully and adding to its prepayment reserves since the last quarter of 1996. The money manager allegation is little more than cumulative.

:f: :ji 45 ❖ ❖ ❖

We reverse the district court’s dismissal of the investors’ complaints and remand for further proceedings consistent with this opinion.

2 . 17 C.F.R. § 240.10b-5 (2001).

3 . In addition to Green Tree and Coss, all the investors named as defendants Robert D. Potts, Green Tree’s former Chief Operating Officer, and Edward L. Finn, Green Tree’s Chief Financial Officer. The stock purchasers’ and option traders' classes also named Joel H. Gottesman, General Counsel to Green Tree. Only the stock purchasers named Richard G. Evans, Executive Vice President of Green Tree, and only the option traders named Robley D. Evans, Controller and Principal Accounting Officer.

4 . The stockholders’ complaint alleges:

The company finally admitted in a conference call with analysts on November 14, 1997 that, throughout the Class Period, it had been assuming that its 1994 and 1995 loans would prepay at a rale of 6% after 24 months when, in fact, several pools of loans originated in early 1995 carried more than 10% prepayment rates when they were only 14 months old.

5 . 15 U.S.C. § 78u-4(b)(l) provides:

In any private action arising under this chapter in which the plaintiff alleges that the defendant—

(A) made an untrue statement of a material fact; or

(B) omitted to state a material fact necessary in order to make the statements made, in light of the circumstances in which they were made, not misleading;

the complaint shall specify each statement alleged to have been misleading, the reason or reasons why the statement is misleading, and, if an allegation regarding the statement or omission is made on information and belief, the complaint shall state with particularity all facts on which that belief is formed.

6 . 15 U.S.C. § 78u-4(b)(2) provides:

In any private action arising under this chapter in which the plaintiff may recover money damages only on proof that the defendant acted with a particular state of mind, the complaint shall, with respect to each act or omission alleged to violate this chapter, state with particularity facts giving rise to a strong inference that the defendant acted with the required state of mind.

7 . There is now substantial agreement among the Circuits that have considered the cuestión that 15 U.S.C. § 78u-4(b)(2) was not i -tended to alter the substantive standard for si i enter. See Nathenson v. Zonagen Inc., 267 F.3d 400, 409 (5th Cir.2001); City of Philadelphia v. Fleming Cos., 264 F.3d 1245, 1258-60 (10th Cir.2001); Greebel v. FTP Software, Inc., 194 F.3d 185, 198-201 (1st Cir.1999); Phillips v. LCI Int’l, Inc., 190 F.3d 609, 620 (4th Cir.1999); Bryant v. Avado Brands, Inc., 187 F.3d 1271, 1283-84 (11th Cir.1999); Hoffman v. Comshare, Inc. (In re: Comshare, Inc. Sec. Litig.), 183 F.3d 542, 548-49 (6th Cir.1999); In re Advanta Corp. Sec. Litig., 180 F.3d 525, 534 (3d Cir.1999); Press v. Chemical Inv. Serv. Corp., 166 F.3d 529, 537-38 (2d Cir.1999). In In re Silicon Graphics Inc. Sec. Litig., 183 F.3d 970, 977 (9th Cir.1999), the Ninth Circuit stated that the Reform Act adopted a heightened substantive standard for scienter, that of “deliberate recklessness,” which "reflects some degree of intentional or conscious misconduct.” In a later case, however, the Ninth Circuit limited Silicon Graphics ' holding to a pleading requirement, stating that "the [Reform Act] did not alter the substantive requirements for scienter under § 10(b).” Howard v. Everex Sys., Inc., 228 F.3d 1057, 1064 (9th Cir.2000).

8 . The Specter amendment provided:

[A] strong inference that the defendant acted with the required state of mind may be established either—

(A) by alleging facts to show that the defendant had both motive and opportunity to commit fraud; or

(B) by alleging facts that constitute strong circumstantial evidence of conscious misbehavior or recklessness by the defendant.

141 Cong. Rec. 17,286 (1995).

9 . Ironically, Senator D’Amato objected to the Specter amendment because he thought it would increase the difficulty of pleading. 141 Cong. Rec. 17,424 ("To be quite candid with you, I think -it places too great a burden on the plaintiffs.”).

10 .The Conference Committee also revised the language of the bill that required facts relevant to scienter to be "specifically alleged”; after the Conference the bill stated that such facts had to be "state[d] with particularity.” Senator Specter contended that this, too, raised the bill’s pleading standard above the Second Circuit standard, 141 Cong. Rec. at 38,206, but Senators Dodd and Bradley contended that this was not a substantive change. 141 Cong. Rec. at 38,227 and 38,-322.

11 . The option traders contend that statements in a Conference Report pertaining to 1998 legislation give further support to the argument that Congress intended to endorse the Second Circuit rule. The Conference Report, H.R. Conf. Rep. 105-803, at 15 (Oct. 9, 1998), only refers in a general way to the Second Circuit pleading standard and does not address the specific question of whether Congress intended to adopt the motive-and-opportunity standard.

12 . The S.E.C., in an amicus brief, argues, "The Reform Act adopts or leaves it to the courts to adopt the ‘motive and opportunity and ‘strong circumstantial evidence’ tests.”

13 . The district court held that the complaints adequately alleged opportunity to defraud, and the defendants do not dispute that point here.

14 . Unusual insider trading is a prominent theme in many Second Circuit cases holding that scienter was pleaded by facts showing motive. E.g., Hollín v. Scholastic Corp. (In re Scholastic Corp. Sec. Litig.), 252 F.3d 63, 74— 75 (2d Cir.2001), petition for cert, filed, 70 U.S.L.W. 3164 (U.S. Aug. 30, 2001) (No. 01-397); Novak, 216 F.3d at 308. But see ZVI Trading Corp. Employees’ Money Purchase Pension Plan & Trust v. Ross (In re Time Warner Inc. Sec. Litig.), 9 F.3d 259, 269-71 (2d Cir.1993) (holding motive adequately pleaded in case with no insider stock transactions). The Fourth Circuit has said that a claim of motive based on a defendant’s benefit from the increase in value of a corporation’s stock must include allegations that the defendant sold his stock. Phillips v. LCI Internat’l, Inc., 190 F.3d 609, 622 (4th Cir.1999).

15 . "MHP” means manufactured housing prepayment.

16 . The investors contend that we should not consider the increase in reserves because the figures are "unsubstantiated,” presumably meaning they are not pleaded in the complaints. Actually, they are. The Florida Board pleaded the amount of prepayment reserves for each of the last three quarters of 1996, which show that Green Tree increased reserves. All three complaints alleged that Green Tree's financials were misleading because the reserves were inadequate. The investors cannot base their allegations on Green Tree's financials and then prevent the court from looking at the financials.

17 . Because we have concluded that the investors have satisfied the Reform Act standard, we need not discuss their other various arguments augmenting their showing of scienter, such as the arguments that they have established scienter by alleging that Green Tree violated GAAP and that it concealed the actual prepayment rate by certain credit practices, and the argument that scienter may be inferred from the short time between Coss's statement about the magnitude of losses in November, 1997 and the charge off of greater amounts in January 1998.