In Re. Ikon v. City of Philadelphia

277 F.3d 658, 2002 U.S. App. LEXIS 470, 2002 WL 29662

• Court: Court of Appeals for the Third Circuit
• Decided: January 11, 2002
• Docket: 01-1553
• Status: Unknown

Opinion

OPINION OF THE COURT

GREENBERG, Circuit Judge.

This shareholders’ class action case comes on before this court on plaintiffs- appellants’ appeal from a February 6, 2001 order of the district court granting summary judgment to the defendant-appellee Ernst & Young LLP (“Ernst”). See In re IKON Office Solutions Sec. Litig., 131 F.Supp.2d 680 (E.D.Pa.2001). Appellants are representatives of a certified class consisting of all persons who purchased common stock, convertible preferred stock, and/or call options of IKON Office Solutions, Inc. (“IKON”) from October 15, 1997, through August 13, 1998. In their complaint they alleged that Ernst, IKON’s accounting firm, violated section 10(b) of the Securities Exchange Act, 15 U.S.C. § 78j(b), and Rule 10b-5, 17 C.F.R. § 240.10b-5, promulgated thereunder, by issuing an unqualified audit report approving IKON’s financial statements for fiscal year 1997 knowing that they overstated IKON’s pre-tax income or, even if Ernst did not have actual knowledge of the overstatement, by recklessly performing its audit.

The district court granted Ernst’s motion for summary judgment on the grounds that the appellants failed to raise a genuine issue of material fact with respect to two elements of a prima facie section 10(b) claim: scienter (that Ernst harbored an intent to deceive or acted with reckless disregard for the truth and accuracy of IKON’s financial disclosures) and causation (that the inflated value of IKON’s stock price dropped when the market reevaluated the security after a corrective disclosure). In addition, and in the alternative, the court granted Ernst’s motion for partial summary judgment, ruling that it could not be liable to certain members of the class under section 10(b) by reason of an October 15, 1997 press release IKON issued because Ernst itself did not communicate misrepresentations to investors in the press release — the only activity proscribed by the statute.

Because the record fails to establish a triable issue with respect to scienter, we will affirm the judgment of the district court without addressing loss causation or whether Ernst can be held liable under section 10(b) for IKON’s October 15, 1997 press release.

I. BACKGROUND

A. Factual History

IKON, which is headquartered in Mal-vern, Pennsylvania, supplies copiers, printing systems, and related services throughout the United States, Canada, and Europe. Its shares are traded publicly on the New York Stock Exchange. Between 1995 and 1998, IKON embarked on a “transformation” business plan in which it acquired and consolidated close to 200 independent copier, technology-services, and outsourcing and imaging companies. See J.A. 1331 (Jarrell Report). IKON intended to become an international provider of “office technology solutions,” serving as a single source for networking services, office technology, and software needs, rather than simply distributing and servicing office products in domestic markets. See J.A. 1330-31 (Jar-rell Report).

Ernst, a “Big Five” accounting firm, 1 served as the independent, outside auditor of IKON’s' September 30 fiscal year-end financial statements for a number of years, including the time encompassing the audit of IKON’s 1997 consolidated financial statements. See J.A. 113-14 (2d Am. Cplt. at ¶ 101). Ernst designed its year-end audits to evaluate whether IKON’s financial statements accurately and fairly reflected its financial position in accordance with Generally Accepted Accounting Principles (“GAAP”). 2 In addition, Ernst performed certain internal audit functions for IKON, such as monitoring and evaluating its compliance with its own internal accounting policies and procedures. See J.A. 115 (2d Am. Cplt. at ¶ 105).

This dispute focuses on the soundness of Ernst’s audit for fiscal year 1997. On October 15, 1997, after Ernst had completed the bulk of its audit work, 3 IKON issued a press release discussing fourth-quarter and year-end results. See J.A. 5150-55. The release reported income from continuing operations totaling $204.9 million for fiscal 1997, a 15 percent increase from fiscal 1996. See J.A. 5151. Ernst reviewed the press release before it was issued without proposing any modifications. See J.A. 3523 (Dillon Dep. 2627).

On December 24, 1997, Ernst publicly issued its unqualified, or “clean,” audit opinion, 4 stating that it had conducted its audit in accordance with Generally Accepted Auditing Principles (“GAAS”), 5 and con-eluding that IKON’s 1997 financial statements fairly reflected its financial position. See J.A. 4701 (Graham Report). Relevant portions of that unqualified audit opinion, which appeared in IKON’s 1997 Annual Report to the Securities and Exchange Commission on Form 10-k, follow:

We have audited the accompanying consolidated balance sheets of IKON Office Solutions, Inc ... and the related consolidated statements of income, changes in stockholders’ equity and cash flows for each of the three years in the period ended September 30, 1997.... We conducted our audits in accordance with generally accepted auditing standards. Those standards require that we plan and perform the audit to obtain reasonable assurance about whether the financial statements are free of material misstatement. An audit includes examining, on a test basis, evidence supporting the amounts and disclosures in the financial statements. An audit also includes assessing the accounting principles used and significant estimates made by management as well as evaluating the overall financial statement presentation. We believe that our audits provide a reasonable basis for our opinion. In our opinion, the consolidated financial state- mente referred to above present fairly, in all material respects, the financial position of IKON Office Solutions, Inc., and subsidiaries at September 30, 1997 and 1996, and the consolidated results of their operations and their cash flows for each of the three years in the period ended September 30, 1997, in conformity with generally accepted accounting principles.

See J.A. 117 (2d Am. Cplt. at ¶ 109).

Following the release of the audit opinion and IKON’s contemporaneous December 24, 1997 10-k filing with the SEC, share values of IKON common stock experienced a net gain. See J.A. 1371 (Chart Common 1). Within a matter of months, however, IKON’s prospects soured. On April 22, 1998, before the stock market opened, IKON announced that its second quarter earnings for fiscal year 1998 would be $0.35 per share, instead of $0.38 as expected by analysts. IKON also warned that third and fourth quarter earnings would fall below expectations. See J.A. 1624-25 (PR Newswire). IKON cited several reasons for the earnings shortfall, including issues related to its transformation process, competitive pressures, and costs associated with product rationalization. IKON, however, did not mention accounting charges to rectify discrepancies in its 1997 financial statements.

Later that morning the investment banking firm of Goldman, Sachs & Co. downgraded IKON to “market perform” from “recommend list,” while Prudential Securities downgraded IKON from “buy” to “hold.” See J.A. 1340 (Jarrell Report). As a result, the price of IKON common stock dropped precipitously from $34.625 a share on April 21, 1998, to close at $25.25 a share on April 22, 1998 — a one-day decline of 27.08 percent. See J.A. 1371 (Chart Common 1).

IKON’s stock continued to decline over the remainder of the spring and summer of 1998. On June 26, 1998, after the market closed, IKON announced that it would miss its earnings estimate for the third quarter of 1998. See J.A. 1341 (Jarrell Report). The following trading day, June 29, 1998, shares of its common stock declined $6.75 to close at $15.31. See J.A. 1371 (Chart Common 1). On July 9, 1998, IKON CEO John Stuart resigned and was replaced by IKON Vice-President James Forese. See J.A. 1342 (Jarrell Report).

With its market performance and economic prospects deteriorating, IKON engaged Ernst to review the books of each of its North American and United Kingdom business services, a project known as the “Special Procedures.” See J.A. 1131-32 (Letter to Dinkelacker). Specifically, IKON instituted the Special Procedures when internal audit work in IKON’s Florida district revealed signs of operational deficiencies and accounting errors, including problems with reconciling intercompa-ny transactions. See J.A. 1087-1130 (June 2, 1998 IKON Report on Florida district).

To assist in this process, the IKON Board of Directors hired the accounting firm of Arthur Andersen (“Andersen”) to review Ernst’s work. See J.A. 3184-85 (McAleer Dep. 138-41). Andersen was permitted to review all work papers prepared in connection with Ernst’s 1997 audit, IKON’s first and second limited 1998 quarterly reviews, and the Special Procedures.

On August 4, 1998, IKON issued a press release indicating that it was conducting a “full review of operations previously announced.” See J.A.1915 (IKON Press Release). On August 14, 1998, after the conclusion of an exhaustive review process, IKON announced the results of the Special Procedures: it would take a $110 million charge against earnings — $94 million in pre-tax charges applied to its 1998 third quarter earnings, and would restate its previously reported, unaudited 1998 second quarter earnings to reflect $16 million in pre-tax charges. See J.A. 2459 (IKON Press Release). The $110 million in charges included $28 million to cover defaults on leases, $20 million for unpaid accounts receivable, $85 million for adjustments due to the breakdown in internal controls at four operating units, $20 million due to asset impairment, and $7 million in miscellaneous adjustments. The press release did not disclose whether the $110 minion charge also corrected errors made in connection with IKON’s 1997 year-end consolidated financial statements. The price of IKON common stock rose by 62 cents, from $9.31 to $9.94 a share on August 14 following the announcement of the results of the Special Procedures.

B. Procedural History

Shortly after IKON announced these disclosures, appellants commenced 16 actions against IKON and certain individual defendants related to it, alleging violations of sections 10(b) and 20(a) of the Securities Exchange Act of 1934, 15 U.S.C. §§ 78j(b) and 78t(a), and of Rule 10b-5, 17 C.F.R. § 240.10b-5. In June 1999, prior to the close of discovery, appellants filed an amended complaint, adding Ernst as a defendant on the claim under section 10(b) and Rule 10b-5 alleging, in essence, that Ernst knew or should have been aware that IKON’s 1997 financial statements substantially overstated pretax income.

Thereafter, the district court consolidated the actions and appointed the lead plaintiffs, the City of Philadelphia through its Board of Pensions and Retirement, Oliver Scofield, and Lawrence Porter, to represent the interests of a class certified to incorporate all those who acquired IKON securities between October 15, 1997 (the day of IKON’s press release regarding its 1997 year-end results), and August 13, 1998 (the day before IKON announced the $110 million in charges against earnings).

In November 1999, the IKON defendants agreed to a settlement with the class for $111,000,000. The district court approved the settlement on May 9, 2000, leaving Ernst as the sole remaining defendant. See In re IKON Office Solutions, Inc. Sec. Litig., 194 F.R.D. 166 (E.D.Pa.2000).

On February 6, 2001, one month prior to trial, the district court on Ernst’s motion entered summary judgment in its favor completely extinguishing the case on two independent grounds — appellants’ failure to adduce sufficient evidence of both scien-ter and loss causation. In addition, the court, as we have indicated, granted Ernst a partial summary judgment. Appellants filed a timely notice of appeal on March 5, 2001.

C. Jurisdiction

We have jurisdiction over this appeal of a final judgment of the district court pursuant to 28 U.S.C. § 1291. The district court exercised federal question jurisdiction under 28 U.S.C. §§ 1331 and 1337, because the case arose under the Securities Exchange Act of 1934, 15 U.S.C. § 78aa, and Rule 10b-5 promulgated thereunder.

II. DISCUSSION

A. Standard of Review

We review an order granting summary judgment de novo, applying the same test the district court employed. See Lucent Info. Mgmt., Inc. v. Lucent Tech. Inc., 186 F.3d 311, 315 (3d Cir.1999). Fed. R.Civ.P. 56(c) provides, in pertinent part, that a court may grant summary judgment only if “the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to judgment as a matter of law.”

In determining whether summary judgment is warranted, we view the record and draw inferences in a light most favorable to the non-moving party. See Arnold M. Diamond, Inc. v. Gulf Coast Trailing Co., 180 F.3d 518, 521 (3d Cir.1999). If a non-moving party fails to make a showing sufficient to establish the existence of an element essential to that party’s case on which it bears the burden of proof at trial, there is no issue as to a genuine issue of a material fact and thus the moving party is entitled to judgment as a matter of law. See Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S.Ct. 2505, 2510, 91 L.Ed.2d 202 (1986).

Moreover, a party will not be able to withstand a motion for summary judgment merely by making allegations; rather, the party opposing the motion must go beyond its pleading and designate specific facts by use of affidavits, depositions, admissions, or answers to interrogatories showing there is a genuine issue for trial. See Celotex Corp. v. Catrett, 477 U.S. 317, 324, 106 S.Ct. 2548, 2553, 91 L.Ed.2d 265 (1986); see also GFL Advantage Fund, Ltd. v. Colkitt, 272 F.3d 189, 199 (3d Cir.2001). Only evidence “sufficient to convince a reasonable factfinder to find all of the elements of[the] prima facie case” merits consideration beyond the Rule 56 stage. Keller v. Orix Credit Alliance, Inc., 130 F.3d 1101, 1108 (3d Cir.1997) (en banc).

B. Section 10(b)

Section 10(b) prohibits the “use or employ, in connection with the purchase or sale of any security, ... [of] any manipulative or deceptive device or contrivance in contravention of such rules and regulations as the Commission may prescribe.... ” 15 U.S.C. § 78j(b). Rule 10b-5, promulgated under section 10(b), makes it unlawful for any person “[t]o make any untrue statement of a material fact or to omit to state a material fact necessary to make the statements made in the light of the circumstances under which they were made, not misleading ... in connection with the purchase or sale of any security.” 17 C.F.R. § 240.10b-5(b).

To state a valid claim under section 10(b) and Rule 10b-5, a plaintiff must show that the defendant (1) made a misstatement or an omission of a material fact (2) with scienter (3) in connection with the purchase or the sale of a security (4) upon which the plaintiff reasonably relied and (5) that the plaintiffs reliance was the proximate cause of his or her injury. See GFL Advantage Fund, 272 F.3d at 212; Weiner v. Quaker Oats Co., 129 F.3d 310, 315 (3d Cir.1997).

1. Scienter

Liability under section 10(b) may extend to secondary actors in the securities markets, as for example where an outside accounting firm prepares a fraudulent audit report that it knows will reach and likely influence the investing public. See Central Bank of Denver, N.A. v. First Interstate Bank of Denver, N. A., 511 U.S. 164, 191, 114 S.Ct. 1439, 1455, 128 L.Ed.2d 119 (1994) (“[a]ny person or entity, including a lawyer, accountant, or bank ... may be hable as a primary violator under lobs’’); Semerenko v. Cendant Corp., 223 F.3d 165, 181 (3d Cir.2000) (the “in connection with” prong is satisfied where the misrepresentations are material and disseminated to the public in a medium upon which a reasonable investor would rely). Of course, imposition of liability on this basis is consistent with the primary pur pose of the Securities Exchange Act of 1934 which is to protect against manipulated stock prices by imposing strict and extensive disclosure requirements, irrespective of the type of actor that disseminates information to the investing public. See S.Rep. No. 792, 73d Cong., 2d Sess., 1-5 (1934), reprinted 1934 WL 1289.

However, by its terms, section 10(b) does not prohibit aiding and abetting. See Central Bank, 511 U.S. at 191, 114 5.Ct. at 1455. 6 To establish securities fraud, plaintiffs must establish a more exacting threshold of scienter — “a mental state embracing intent to deceive, manipulate or defraud,” Ernst & Ernst v. Hochfelder, 425 U.S. 185, 193 n. 12, 96 S.Ct. 1375, 1381 n. 12, 47 L.Ed.2d 668 (1976), or, at a minimum, “highly unreasonable (conduct), involving not merely simple, or even inexcusable negligence, but an extreme departure from the standards of ordinary care, ... which presents a danger of misleading buyers or sellers that is either known to the defendant or is so obvious that the actor must have been aware of it.” SEC v. Infinity Group Co., 212 F.3d 180, 192 (3d Cir.2000), cert. denied, 532 U.S. 905, 121 S.Ct. 1228, 149 L.Ed.2d 138 (2001) (citing McLean v. Alexander, 599 F.2d 1190, 1197 (3d Cir.1979)); see Healey v. Catalyst Recovery of Pa., 616 F.2d 641, 649 (3d Cir.1980) (section 10b recklessness closer to intentional conduct than indifference); see also In re Advanta Sec. Litig., 180 F.3d 525, 535 (3d Cir.1999). Simple computation errors or slight accounting mistakes will not suffice to establish scien-ter. 7

Appellants advance the following claims to demonstrate that Ernst knew of or was reckless in failing to discover deficiencies in IKON’s financial statements in connection with IKON’s October 15, 1997 press release and Ernst’s December 24, 1997 audit opinion: 8 (1) that Ernst faked to investigate sufficiently evidence of fraud by IKON or take into account other conspicuous risk factors or “red flags” that would have alerted Ernst to the fallacious computations; and (2) that Ernst imper-missibly relied on IKON’s internal controls in preparing its audit calculations. As a result of these alleged deficiencies and other violations of GAAS, appellants attribute $20.8 million in known errors and $30.1 million in reckless errors to Ernst’s audit opinion.

We have concluded, however, after an intensive study of the formidable record in this case, that even when viewed with hindsight it does not supply a basis from which to draw a reasonable inference that Ernst recklessly or knowingly issued a materially false and misleading audit opinion after reviewing IKON’s 1997 year-end financial statements. 9 While a determination of whether a party acted with scienter, intertwined as it may be with an assessment of witness credibility, often cannot be undertaken appropriately on summary judgment proceedings, 10 in this case, even accepting arguendo that, as appellants assert, $54.9 million of the $110 million charge taken at the conclusion of the Special Procedures should have been recorded as of September 30, 1997, we are satisfied that the record establishes that Ernst’s failure to do so was not without a good faith belief or reasonable basis.

At the outset, the magnitude of Ernst’s audit for fiscal year 1997 bears mentioning: six full scope audits, including one performed at IKON’s leasing arm, IKON Capital, Inc.; eight localized audits at IKON business units that generated 50 percent of its revenues; a host of procedures performed at IKON’s corporate headquarters; full reviews of previous years’ audits; extensive external testing of IKON’s account balances as the primary support for the audit opinion; work-output totaling 70 percent more than the hours budgeted to the Northern California dis trict (1,014 as opposed to 600), where the implementation of a new computer system impeded review; all told, over 10,000 hours of labor expended on an account to which 8,250 hours originally were budgeted (a 20 percent overall increase). See J.A. 5447-51 (Mulherin Decl. ¶¶ 11, 12, 18, 21,. 23); J.A. 4778-82 (workpaper); J.A. 3149, 4803-06 (workpapers); J.A. 3147 (workpa-per).

It is equally notable that Arthur Andersen, after conducting its independent review during the Special Procedures, found nothing that “would be a significant issue” regarding the quality of Ernst’s opinion, and that in coming to this conclusion Andersen was assessing the work of a major rival in the national accounting market. See J.A. 3185 (McAleer Dep. 142). Though not performing an outright audit of the 1997 opinion or financial statements for compliance with GAAS or GAAP, Andersen allocated hundreds of hours of labor to its examination of the work papers produced by Ernst in 1997. See J.A. 3184-85 (McAleer Dep. 138^14). After this examination Andersen concluded that the “balance sheet was solid” as of September 30, 1997, and that the computation of reserves was neither aggressive, nor conservative but rather “essentially in the middle, solid,” and, in fact, concurred in IKON’s decision not to restate the 1997 financial statements. See J.A. 3185, 3188 (McAleer Dep. 144, 155-56). One Andersen partner involved in reviewing the 1997 working papers, Thomas Costello, stated that Andersen became aware of “nothing ... that was viewed to be a significant deficiency” with respect to GAAS. See J.A. 3256 (Costello Dep. 208). Likewise, the Chairman of the Audit Committee for the IKON Board, James R. Birle, indicated that Andersen did not “find anything fundamentally wrong with what [Ernst] ha[d] done, so they gave them a passing grade.” See J.A. 3301 (Birle Dep. 113). While appellants are correct that Andersen’s conclusions do not provide cover categorically to insulate Ernst from liability, 11 the fact that Andersen endorsed IKON’s decision not to restate the 1997 financial statements nevertheless is highly probative of the competence of Ernst’s 1997 audit opinion and undermines any suggestion that Ernst could not reasonably have opined that IKON’s financial statements fairly presented its financial condition in accordance with GAAP. See In re Worlds of Wonder Sec. Litig., 35 F.3d 1407, 1426 (9th Cir.1994) (in establishing scienter, “[t]he plaintiff must prove that ... the accounting judgments which were made were such that no reasonable accountant would have made the same decisions if confronted with the same facts”).

We conclude that none of the specific errors appellants assert raises a material dispute of fact on which to predicate a finding of scienter. The record makes clear with respect to appellants’ claim that Ernst ignored allegations that IKON CFO Kurt Dinkelacker (“Dinkelacker”) had been “cooking the books” and supplying fraudulent numbers during the 1997 audit that Ernst took appropriate steps to determine whether the allegations had any merit. To begin with, the alleged accuser, IKON official Peter Shoemaker, denied under oath that he ever made the remark. See J.A. 2988-91 (Shoemaker Dep. 7-21). Furthermore, George Berry (“Berry”), the Ernst partner responsible for the IKON audit, testified that he understood the remark to mean that Dinkelacker had been accused of characterizing operational expenses as “transformation expenses.” See J.A. 3048 (Berry Dep. 57-58). However, the allegation was deemed groundless as Ernst just recently had reviewed IKON’s “transformation expenses” and found no improprieties with regard to the classification of expenses. See J.A. 351 (Nepa Dep. 10-11). Finally, IKON’s top management engaged outside counsel, two senior partners in the law firm of Ballard Spahr Andrews & Ingersoll, LLP, to investigate the allegation; they subsequently found “nothing to the ‘cooking the books’ allegation.” See J.A. 259-60 (PL Response to Ernst’s First Set of Requests for Admissions).

Appellants next argue that a trier of fact may make a finding of scienter from evidence that Ernst deliberately disregarded a warning from an IKON officer that various districts were being forced to perform “unnatural acts” and boost income by manipulating IKON’s reserves. Specifically, in July 1998, Berry received written notice of these allegations from IKON official Michael Dudek (“Dudek”), who relayed information that several IKON employees indicated that they were uncomfortable with certain accounting directions which they had been receiving from IKON Corporate. See J.A. 1136 (Dudek Memoran-da).

A jury could not plausibly infer an extreme departure from the standards of ordinary care on the basis of this evidence. First and most significantly, nothing in the record links the concerns raised late in 1998 regarding the manipulation of reserves to Ernst’s preparation of its audit opinion in 1997, the only time-period where Ernst’s state of mind is relevant to the section 10(b) claim involved here. Furthermore, the record establishes that Ernst responded appropriately to the allegations of misconduct raised in the Dudek memoranda: Berry ensured that top IKON management was aware of the allegations, encouraged IKON to hire an independent counsel to investigate (which it did — the Philadelphia law firm of Dechert Price & Rhoads), delivered the Dudek memoranda to the independent counsel, and incorporated additional steps into the Special Procedures to review the newly raised concerns. See J.A. 354-55 (Nepa Dep. 23-28). The mere fact that Ernst did not conduct its own fraud investigation or alert its field auditors to the allegations of fraud is not probative, as the relevant inquiry is bad faith, not judgment. 12 See In re Worlds of Wonder, 35 F.3d at 1426-27 (“self-righteous” assertion that defendant did not conduct an audit precisely as plaintiffs might have insufficient to establish fraud); see also Hochfelder, 425 U.S. at 206, 96 S.Ct. at 1387 (Congress did not intend for anyone to be made liable except for practices “acted other than in good faith”).

In an effort to obtain a reversal of the summary judgment and secure a jury trial, appellants lay out the following scenario: Ernst prepared a checklist entitled “Internal Control and Fraud Considerations” during the planning stage of the 1997 audit (J.A. 1144-63); the pre-printed, 20-page checklist details dozens of possible risk factors; Ernst checked “yes” for the presence of a handful of risk factors, including, inter alia, “unduly aggressive earnings targets,” “excessive interest in maintaining or increasing [IKON’s] stock price or earnings trend,” and a commitment to achieving “what appear to be unduly aggressive or unrealistic forecasts;” these responses, appellants maintain, demonstrate conclusively that a “significant risk of financial error existed” and raise a genuine issue of material fact as to whether Ernst knew that IKON personnel were under pressure to fake revenues and artificially inflate operating results. 13 See Br. of Appellants at 17,19.

Appellants fail, however, to explicate that the checklist contained 98 “no” answers to potential risks or that it was prepared in the pre-audit planning stages, before Ernst actually undertook its review of IKON’s 1997 financial statements to appraise the validity of these concerns. Indeed, the questionnaire itself makes clear that “the relative importance of risk factors varies among engagements” such that the form can provide “only a portion of the understanding about an entity’s internal control” that is required to plan an audit. See J.A. 1144r-45. The simple fact that Ernst identified IKON management’s strong preference for favorable earnings, standing alone, does not raise an inference of scienter sufficient to survive a summary judgment motion predicated on the absence of scienter. See Acito v. IMCERA Group, Inc. 47 F.3d 47, 54 (2d Cir.1995) (allegation that defendants were motivated to defraud the public because an inflated stock price would increase their compensation insufficient because “[i]f scienter could be pleaded on that basis alone, virtually every company in the United States that experiences a downturn in stock price could be forced to defend securities fraud actions”). In fact, rather than probative of recklessness, the document tends to corroborate Ernst’s diligence in conducting the 1997 audit, identifying potential risks at an early stage in accordance with professional standards. See AU § 316 (1998) (an auditor has a duty to assess the “risk of material misstatement of financial statements due to fraud”) (quoted in P. Schoen-feld Asset Mgmt. LLC v. Cendant Corp., 142 F.Supp.2d 589, 607 (D.N.J.2001)).

We also reject appellants’ argument that scienter credibly may be inferred from Ernst’s reliance on IKON’s defective internal controls. 14 While IKON’s internal accounting controls well may have been unreliable during fiscal year 1997, the record, except with respect to the fourth quarter, 15 does not connect these internal deficiencies to Ernst’s independent, external audit. To the contrary, Ernst adopted an “effective/not rely” approach during the 1997 year-end audit, 16 wherein, after scrutinizing IKON’s internal controls to confirm their effectiveness (and making recommendations to IKON officers for improvement where deemed necessary), Ernst nevertheless opted not to defer to IKON’s internal audit findings and processes. See J.A. 627 (Generalized Audit Program Steps, Year Ending Sept. 30, 1997) (“substantive testing without placing reliance on the related systems is the most efficient way to audit”); see also J.A. 5450 (Mulherin Deck ¶¶ 17-18) (“[W]e determined that IKON’s overall internal control environment was effective.... However, we would not rely on the testing of controls as the primary support for our audit opinion. Rather, we would perform substantive testing — such as analytical procedures and tests of details — to serve as our principal source of audit evidence”). Absent further evidence affirmatively linking IKON’s internal controls to Ernst’s audit opinion, the record does not raise a genuine issue of material fact from which a jury could conclude that Ernst knowingly or blindly adhered to faulty internal controls or accounting practices. See Monroe v. Hughes, 31 F.3d 772, 775 (9th Cir.1994) (deficiencies in internal controls of a company are immaterial to the audit report itself because in essence they are matters of which only management should be aware); Danis v. USN Communications Inc., 121 F.Supp.2d 1183, 1195 (N.D.Ill.2000) (without more, auditor’s knowledge about problems in a client’s operational systems could support an inference only of negligence, not recklessness).

Having failed to present concrete evidence from which a jury plausibly could conclude that Ernst acted in a reckless manner, appellants purport to raise a material dispute with respect to scienter by parsing the 1997 audit for specific defects as evidence that Ernst acted recklessly with respect to the 1997 audit as a whole.

However, the discovery of discrete errors after subjecting an audit to piercing scrutiny post-hoc does not, standing alone, support a finding of intentional deceit or of recklessness. See In re Software Toolworks Inc., 50 F.3d 615, 627 (9th Cir.1994) (“the mere publication of inaccurate accounting figures, or a failure to follow GAAP, without more, does not establish scienter”) (citations and internal quotation marks omitted); SEC v. Price Waterhouse, 797 F.Supp. 1217, 1240 (S.D.N.Y.1992) (misapplication of accounting principles by an independent auditor does not establish scienter). An audit does not guarantee that a client’s accounts and financial statements are correct any more than a sanguine medical diagnosis guarantees well-being; indeed, even an audit conducted in strict accordance with professional standards countenances some degree of calibration for tolerable error which, on occasion, may result in a failure to detect a material omission or misstatement. See AICPA General Standard No. 3 (audit requires only due professional care) (cited in Vladimir v. Deloitte & Touche LLP, 1997 WL 151330, at *5 (S.D.N.Y. Mar.31, 1997)).

Rather, the “objective of the ordinary examination of financial statements by the independent auditor is the expression of an opinion on the fairness with which they present financial position, results of operations, and changes in financial position in conformity with generally accepted accounting principles.” AU § 110.01 (quoted in United States v. Weiner, 578 F.2d 757, 786 n. 27 (9th Cir.1978)). In other words, in issuing an opinion, the auditor certifies only that it exercised appropriate, not flawless, levels of professional care and judgment. See La Rossa v. Scientific Design Co., 402 F.2d 937, 943 (3d Cir.1968) (“Those who hire (experts) are not justified in expecting infallibility, but can expect only reasonable care and competence. They purchase service, not insurance.”) (citing Gagne v. Bertran, 43 Cal.2d 481, 275 P.2d 15, 20 (1954)).

Thus, to give rise to section 10(b) liability for fraud, the mere second-guessing of calculations will not suffice; appellants must show that Ernst’s judgment— at the moment exercised — was sufficiently egregious such that a reasonable accountant reviewing the facts and figures should have concluded that IKON’s financial statements were misstated and that as a result the public was likely to be misled. Cf. Denny v. Barber, 576 F.2d 465, 470 (2d Cir.1978) (rejecting “fraud by hindsight” because the law does not expect clairvoyance).

Nothing in the record meets this demanding threshold. For example, appellants assert that Ernst had actual knowledge of $20.8 million in errors when it rendered its audit opinion. But they derive this figure from their singular and fastidious vision as to how the 1997 audit sensibly should have been conducted. The allegation does not, however, raise an inference that Ernst harbored an intent to deceive or exhibited a reckless disregard for the likelihood of fraud by exercising divergent, but nevertheless principled, methodologies in auditing IKON’s financial statements.

The tabulation breaks down as follows. First, appellants contend that the lease default reserve (the amount maintained to offset unpaid leases that are charged back to individual marketplaces) in IKON Southern California was understated by $4.2 million. Appellants derive this figure by subtracting the recorded reserve rate of 1.8 percent from the historical lease default rate of 6 percent and applying the difference (4.2 percent) to the Southern California portfolio of $99,786,309 million (roughly $4.2 million). See Br. of Appellants at 22. They point to an internal report, in 22 which Ernst auditors noted the 4.2 percent discrepancy between default rate and reserve rate, to demonstrate actual knowledge on the part of Ernst of the shortfall. See J.A. 4820-25 (Report).

Ernst challenges this calculation for its failure to adjust for “recoveries,” the value of equipment retrieved from any defaulting lessees or the recovery obtained from defaulting lessees through settlement, litigation, or otherwise. Notwithstanding this observation, even were we to accept the accuracy of appellants’ $4.2 million figure, 17 the “error” falls short of raising a material dispute with respect to scienter. To begin with, the record reveals that the internal audit report from which appellants purport to discern Ernst’s culpable state of mind was a draft (marked for “discussion purposes only”) and may not have been completed by the time Ernst concluded its 1997 audit. 18 More importantly, there is no evidence to rebut a conclusion that Ernst took reasonable steps to ensure the adequacy of IKON’s default reserves over all, 19 Appellants cannot establish scienter merely by pointing to the $85 million reserve adjustment registered during the Special Procedures in 1998, without affirmatively linking that amount to problems known as of September 80, 1997. See Reply Br. of Appellants at 14.

Appellants emphasize a $2.8 million overstatement in Los Angeles inventory. However, as Ernst correctly notes, that figure is predicated on a draft summary of findings for the Southern California district, 20 which detailed a potential risk of misstatement in the event that inventory discrepancies remained unresolved. It does not, however, suggest that Ernst failed to address these concerns before publication of the audit opinion. 21 Quite simply, appellants fail to explain how Ernst’s awareness of discrete inventory problems in one IKON district is tantamount to scienter, to a mental state suggesting an intent to conceal IKON’s true financial position from investors or a realistic expectation of likely public confusion regarding IKON’s overall financial health.

Appellants point next to a “known” $3.6 million shortfall in questionable receivables that were transferred from IKON Management Services unit (“IMS”) to IKON Document Services (“IDS”). Appellants allege that Ernst knew of the “IDS/IMS Bad Debt Reserve” when its field auditors alerted Ernst personnel in Philadelphia to the fact that IDS did not maintain a separate reserve for dubious accounts receivable. See J.A. 315 (Summary Review Memo).

However, quite apart from this warning, nothing in the record supports the claim that IDS was in fact under-reserved for 1997 or that Ernst Philadelphia did not take appropriate steps to evaluate accounts receivable. Furthermore, as Ernst points out, auditing estimated reserves for doubtful accounts is a highly imperfect undertaking that requires an assessment of the risk that accounts may be defaulted on. As there is no evidence to suggest that Ernst’s method of predicting collectibilty was unreasonable or grossly inconsistent with acceptable accounting practices, there is no basis to conclude that Ernst fraudulently certified that the reserve for doubtful accounts in IKON’s consolidated financial statements comported with GAAP. 22

Appellants assert that Ernst missed approximately $4.0 million in known IDS in-tercompany balances during the 1997 audit, but the evidence on this point suggests nothing more than simple error, at best approaching negligence, not scienter. Approximately $7.3 million in intercompany errors came to light during the Special Procedures, a figure that includes the $4.0 million related to IDS and IMS. See J.A. 786-89 (Special Procedures Summary of Adjustments). Appellants maintain that Ernst “knew” of the imbalance both from a broad familiarity with IKON’s historic problems with expunging redundant balances and from explicit written warnings sent to Ernst Philadelphia from Ernst Houston emphasizing the need to eliminate accounts at IDS/IMS.

Yet those written warnings contain only a very general admonition “to ensure in-tercompany accounts eliminate on a consolidated basis,” and appellants offer no additional evidence to show that the $4.0 million figure discovered in 1998 corresponds with the imbalance identified by Ernst Houston in 1997. Moreover, to the extent that Ernst may have erred in failing to eliminate intercompany assets from pre-tax earnings, an inference of recklessness or intent to deceive that otherwise might be drawn cannot survive the fact that Ernst thoroughly reviewed and tested IKON’s intercompany balances. See J.A. 3489 (Nepa Dep. 593); J.A. 5262-5289 (workpaper); J.A. 928-29 (October 10, 1997 letter).

Finally, appellants claim that there was a known misrepresentation of $6.2 million involving the 1997 Summary of Audit Differences (“SAD”), the list documenting discrepancies between IKON’s numbers and Ernst’s conclusions. However, as appellants have acknowledged, Ernst’s failure to require IKON to adjust its financial statements by $6.2 million is only material when considered in the context of the $14.6 million in claimed errors with respect to lease default reserve; inventory, accounts receivable, and intercompany balances that we already have addressed. See Br. of Appellants at 27. Considering that these errors have been discounted, without some additional evidence insinuating scienter, the $6.2 million pertaining to the SAD does not supply a basis to establish scienter.

The $31 million in allegedly “reckless” misstatements appellants charge also cannot be the basis to establish scienter. First, the allegation that Ernst’s failure to reconcile accounts resulted in a $10.4 million overstatement does not support a finding of the requisite scienter. 23 The claim that appellants’ experts, in retrospect, would have compared assets against balance sheets on a monthly basis or that IKON’s internal policy called for monthly-reconciliation of accounts payable, accounts receivable, cash, and inventory, does not illustrate that Ernst’s incongruous auditing tactics were unjustified, let alone reflected an intent to defraud or rash disregard for the likelihood of deception.

The balance of allegedly “reckless” errors relate to $20.6 million of shortfalls in accounts receivable reserves. 24 Appellants’ experts opined that IKON’s transformation initiative seriously impaired account collectibility and, therefore, Ernst should have enforced IKON’s more conser vative policy to estimate appropriate reserve mínimums.

However, as discussed previously, 25 highlighting different accounting methodologies that Ernst might have employed— particularly in the nebulous context of establishing reserves for vulnerable accounts — does not suggest that the approach actually chosen was an extreme departure from ordinary care.

For example, appellants point to the fact that IKON Northern California’s deficient computer system could not properly assess the accounts receivable reserves. Yet the record confirms that Ernst took reasonable, additional steps to audit those reserves. In its September 30, 1997 summary review memorandum, Ernst details the host of problems identified and procedures instituted during the audit of IKON Northern California. See J.A. 887-893. After IKON conducted an internal audit of approximately 100 accounts receivable and recommended a reserve of at least $7 million, Ernst auditors reviewed the work and actually proposed increased reserves for doubtful accounts receivable. See J.A. 4802 (showing $5.3 million adjustment on the Summary of Audit Differences). Indeed, rather than “stripping” reserves as appellants have maintained, IKON substantially increased its overall allowance for doubtful accounts by some $20 million on Ernst’s recommendation in 1997. See J.A. 5292, 5298 (workpapers). Plainly, without some additional support, the GAAP methodologies advanced by appellants’ experts fall short of transforming Ernst’s estimates of accounts receivable reserves into actionable fraud.

In short, appellants’ citations to the record, like misshapen jigsaw pieces that collectively fail to reveal the picture embedded within the puzzle, simply raise no inference that Ernst opined on IKON’s consolidated financial statements with knowing or reckless disregard for the truth. 26 There is no reasonable basis in the record to doubt that Ernst harbored an honest and well-founded belief in the accuracy of its audit, and, without further substantiating evidence, a jury may not premise' a finding of willful or knowing conduct to defraud or recklessness merely by judging between competing but nevertheless sound accounting methodologies.

2. Loss Causation

As there is ho triable issue with respect to scienter, appellants’ prima facie section 10(b) claim fails, and we need not reach the issue of causation.

III. CONCLUSION

For the foregoing reasons, we will affirm the judgment of the district court entered February 6, 2001.

1 . The "Big Five" accounting firms are Arthur Andersen LLP, Deloitte & Touche LLP, Ernst & Young LLP, KPMG LLP and Pricewater-houseCoopers LLP. For many years, the accounting industry was dominated by eight national firms. In 1989, the "Big Eight” was reduced to six members with the mergers of Ernst & Whinney and Arthur Young into Ernst & Young and Touche Ross and Deloitte Haskins & Sells into Deloitte & Touche. In 1998, the "Big Six” became the "Big Five” as Price Waterhouse merged with Coopers Lyb-rand to become PricewaterhouseCoopers.

2 . GAAP is "a technical accounting term that encompasses the conventions, rules, and procedures necessary to define accepted accounting practices at a particular time.” See American Institute of Certified Public Accounts ("AICPA”), Statement of Auditing Standards No. 69, ¶ 69.02 (1992), quoted in Sanders v. Jackson, 209 F.3d 998, 1001 n. 3 (7th Cir.2000). GAAP principles and standards provide a common framework by which financial statements from divers companies may be compared and adjudged.

3 . The final audit opinion was dated October 15, 1997, with the exception of note 8, dated October 27, 1997. See J.A. 4701 (Graham Report).

4 . An ''unqualified” or "clean” audit opinion is the highest level of assurance that an auditor can give on an organization's financial statements. Accountants will "qualify” their opinion where discrepancies are identified in a client’s financial statements.

5 . GAAS are the standards prescribed by the Auditing Standards Board of the American Institute of Certified Public Accountants for the conduct of auditors in the performance of an examination. See SEC v. Arthur Young & Co., 590 F.2d 785, 788 n. 2 (9th Cir.1979).

6 . In Central Bank, the Supreme Court considered whether section 10(b) liability could extend to actors who do not commit a manipulative or deceptive act within the meaning of section 10(b) but who instead aid and abet a violation. In that case, the plaintiffs sought to hold a bank that served as indenture trustee for two separate bond issues liable under section 10(b) and Rule 10b-5 for agreeing to delay an independent review of appraisal of land which secured bonds until approximately six months after the bonds were to be sold. The bonds were defaulted on before completion of the independent appraisal. See 511 U.S. at 168, 114 S.Ct. at 1443. The Supreme Court, finding no basis within the language or legislative history of the statute and reluctant to allow plaintiffs to circumvent the reliance requirement of section 10(b) simply by demonstrating that a defendant gave some modicum of aid to those engaged in proscribed activities, refused to find an implied private right of action to impose aiding and abetting liability under section 10(b). See id. at 176-80, 114 S.Ct. at 1447-50.

7 . To prevail on a section 10(b) claim, however, appellants need not demonstrate that the audit for fiscal year 1997 amounted to a “pretended audit” as Ernst maintains throughout its brief. In McLean, we suggested the type of circumstantial evidence that could support an inference of bad faith, for instance, "[a] showing of shoddy accounting practices amounting at best to a pretended audit, or of grounds supporting a representation so flimsy as to lead to the conclusion that there was no genuine belief back of it.” McLean, 599 F.2d at 1198 (citations and internal quotation marks omitted) (emphasis added). We did not intend in McLean, however, to restrict the scienter threshold to the precise contours of this. list. Rather, if a plaintiff can show — by whatever means — that defendants "did not have an honest belief in the truth of their statements, then they are liable, so far as (scienter) is concerned.” Id.

8 .The district court entered judgment for Ernst as to the period from October 15, 1997, until December 24, 1997, in which appellants predicated their theory of liability on IKON's October 15 press release. The district court held that IKON's press release, which neither mentioned Ernst by name, nor attributed any representations to Ernst, could not "form a basis for a Section 10(b) claim,” because, in light of Central Bank, Ernst at best facilitated the principal actor's disclosures and, therefore, did not "make" a material misstatement (or omission) upon which the investing public relied. See 131 F.Supp.2d at 685 n. 5 (citing Wright v. Ernst & Young LLP, 152 F.3d 169, 175 (2d Cir.1998)). Noting further that some courts have applied a "substantial participation” test with respect to primary liability for secondary actors, the district court found no basis for liability to attach to Ernst under this standard because the company neither drafted nor directed any meaningful aspect of the press release such that IKON's representations could be attributed to Ernst. See id. (citing Cashman v. Coopers & Lybrand, 877 F.Supp. 425, 433 (N.D.ILL.1995)). While a number of courts have rejected the "substantial participation” test as inconsistent with Central Bank's repudiation of aiding-and-abetting liability for section 10(b), see Anixter v. Home-Stake Production Co., 77 F.3d 1215, 1226 (10th Cir.1996), we need not address this issue because we affirm the district court on scienter grounds only. As appellants conceded at oral argument, if the record cannot plausibly support an inference of intent to deceive or recklessness on Ernst's part in preparing its 1997 audit, then appellants’ section 10(b) claim must fail for the entirety of the class period, including from October 15, 1997, to December 24, 1997.

9 . As this litigation is well beyond the pleading stage, appellants overstate the significance of caselaw suggesting that allegations of violations of GAAP or GAAS, coupled with allegations of ignoring "red flags,” can be sufficient to withstand a motion to dismiss in a securities fraud action. See Reply Br. of Appellants at 10-11 (citing, inter alia, In re Baan Co. Sec. Litig., 103 F.Supp.2d 1, 17 (D.D.C.2000); In re Leslie Fay Cos., Inc. Sec. Litig., 835 F.Supp. 167, 175 (S.D.N.Y.1993)).

10 . See Hunt v. Cromartie, 526 U.S. 541, 552, 119 S.Ct. 1545, 1552, 143 L.Ed.2d 731 (1999) ("Credibility determinations, the weighing of the evidence, and the drawing of legitimate inferences from the facts are jury functions”) (citations and internal quotation marks omitted).

11 . This is especially so when it is considered that as a condition to being awarded the contract, Andersen agreed with Ernst, in writing, not to report to IKON’s board if it determined that the 1997 audit did not comport with accepted auditing standards. See J.A. 1239-40 (Letter to Arthur Andersen); 3256-67 (Costello Dep. 208-09).

12 . Under the AICPA's professional norms, an auditor need conduct its own investigation after discovering information which relates to previously evaluated financial statements only if the information existed at the time of the original report, is deemed reliable, and would have changed the report previously made. See J.A. 1137-39 (AU § 561, "Subsequent Discovery of Facts Existing at the Date of the Auditor's Report,” ¶¶ .04-05). Furthermore, AUS 561 advises the auditor to "discuss the matter with his client at whatever management levels he deems appropriate, including the board of directors, and request cooperation in whatever investigation may be necessary.” See J.A. 1137 (¶ .04). Here, Ernst ensured that board members were aware of the allegations, confirmed that IKON had retained independent investigators as well as outside accountants, and, in all respects, fully cooperated in the investigation.

13 . Appellants also rely on an August 22, 1997 memorandum from Dinkelacker to suggest Ernst's actual awareness of IKON’s fraudulent scheme to overstate income. The memo, addressed to various IKON district presidents and corporate officers and entitled "Followup on Cash Generation Initiative and 4Q Forecast,” states, in relevant terms, as follows:

I would like to have a telephone conversation ... on IKON's current cash flow position and the cash results for July. We need to make every effort to generate significant cash flows for the remainder of the year ... the updated forecast for the fourth quarter [ ] is not a pretty picture ... we must pull out all the stops to achieve 150 million in operating income from the field for the fourth quarter. IKON can not afford another miss. I would like to discuss certain opportunities relating to accounts receivable and inventory valuation reserves, and I want to ensure that we have an understanding around other general issues including the propriety of bonus accruals.

J.A. 1241 (emphasis in original).

However, even accepting appellants’ proffered interpretation that the memo served as an "injunction” or "command” from Dinke-lacker to strip IKON operating units of reserves for doubtful accounts and lease defaults, see Br. of Appellants at 14-15, insofar as we are aware nothing in the record suggests that anyone from Ernst was aware of this memo as of December 24, 1997, when the final audit opinion issued.

14 . "Internal accounting controls” refers to the mechanism by which companies monitor their accounting system (their individualized method of processing transactions) for errors and irregularities in order to safeguard company assets and ensure that records are sufficiently reliable. See SEC v. World-Wide Coin Invs., Ltd., 567 F.Supp. 724, 750 (N.D.Ga.1983) (citing SEC ruling ¶ No. 82,815, reprinted in 36 Bus. Law. 3 (April 1981)).

15 . In that quarter, Ernst employed a "roll forward method” that contemplated some degree of reliance on IKON's internal controls. Certain location teams performed audit procedures three months prior to fiscal year-end (June 30, 1997), and then "rolled forward” the results to fiscal year-end by doing limited testing of account balances and internal controls for the interim three-month period. See J.A. 5450 (Mulherin Decl. ¶ 18). Appellants have not specified how Ernst's limited reliance on IKON’s internal accounting processes undermined or otherwise affected the final audit opinion.

16 .Appellants place undue emphasis on the fact that Ernst’s 1998 "ineffective/not rely” audit strategy by comparison was far more effective, for example, confirming more than 600 accounts receivable in the Northern California district, as opposed to a sample size of 26 generated in 1997. See J.A. 1012 (IKON Northern California district, Accounts Receivable Confirmation, Sept. 30, 1997); 1061 (IKON Northern California, Trade Accounts Receivable, Sept. 30, 1998). Again, however, the relevant inquiry for purposes of section 10(b) is recklessness bordering on an intent to deceive, a finding to which these facts do not lend support.

17 . It should be emphasized that IKON-Southern California accounted for only 2.2 percent of IKON's revenues and 0.7 percent of IKON's assets for 1997. IKON revenues for 1997 totaled over $5.1 billion. See J.A. 5151 (Press Release).

18 . The draft report states that the compliance audit at IKON Southern California was completed as of October 24, 1997. See J.A. 4820 (Report). However, Ernst's workpapers suggest that the final audit report for IKON Southern California was not completed during fiscal 1997. See J.A. 4828-31 (workpa-pers). Furthermore, Carmen Nepa, Ernst's senior audit manager, testified that he did not become aware of the internal audit report for Los Angeles until the first quarter of fiscal 1998, after Ernst had issued its 1997 opinion. See J.A. 3476 (Nepa Dep. 542-43).

19 . See J.A. 3478 (Nepa Dep. 550-51) ("the allowance for lease default was analyzed on a consolidated basis ... a difference at one location would not — would not matter to me as long as I — as long as we concluded the consolidated reserve was adequate”); see also J.A. 4889 (Slack Dep. 52) (vice president of IOS Capital, the financial leasing subsidiary of IKON, testifying that estimated losses for the lease portfolio in 1997 reasonably matched the amount reserved by IKON to account for defaults).

20 . The report, generated at the close of fiscal year 1997 by an internal audit, stated that there was a "need for an adjustment” with respect to approximately $2.8 million of unreconciled inventory and warned of an "increased risk of material misstatement” if the inventory issues were not investigated and resolved. See J.A. 4821 (Report).

21 . On the other hand, we are troubled by the fact that Ernst has not come forward with evidence of affirmative steps taken to resolve the risk. Though the threshold for scienter is considerable, we are wary to raise the bar even higher and insulate auditors who craftily choose not to memorialize confirmed problems or to qualify their observations with highly equivocal terms like "risk,” "potential,” and "likelihood.”

22 . There has been substantial debate as to whether IKON accounting policy is equivalent to GAAP, such that Ernst's deviation from IKON’s internal benchmarks for accounts receivable reserves necessarily would constitute deviation from GAAP and raise an inference of scienter. We finally may put that dispute to rest. IKON organized accounts receivable according to the length of time that an account was outstanding and fixed the amount of reserves required to cover those accounts (33 ¡6 percent of the amount reserved for receivables outstanding 91-150 days, 662k percent for accounts aged 151-180 days, 100 percent for amounts aged greater than 180 days). See J.A. 4781 (workpaper). Ernst claims that it did not adhere strictly to IKON’s policy in calculating adequate reserve amounts because, though useful to IKON in ensuring uniformity and guaranteeing a minimum reserve level for all districts, it was inflexible and did not take into account the presence of factors unique to individual districts. This discrepancy, however, is not dis-positive because IKON’s internal accounting standards would not have yielded the only reserve level acceptable under GAAP. GAAP is a term of art that encompasses a wide range of acceptable procedures. See Thor Power Tool Co. v. Comm’r, 439 U.S. 522, 544, 99 S.Ct. 773, 787, 58 L.Ed.2d 785 (1979) (GAAP "are far from being a canonical set of rules that will ensure identical accounting treatment of identical transactions.... [R]ather, [they] tolerate a range of 'reasonable' treatments, leaving the choice among alternatives to management”); In re GlenFed, Inc. Sec. Litig., 42 F.3d 1541, 1549 (9th Cir.1994) (en banc) (as accounting concepts are flexible, circumstances will give rise to fraud only where differences in calculations are the result of a falsehood, "not merely the difference between two permissible judgments”); Godchanx v. Conveying Techniques, Inc., 846 F.2d 306, 315 (5th Cir.1988) (a reasonable accountant may choose to apply any of a variety of acceptable procedures when preparing a financial statement). As such, the fact that appellants can demonstrate that Ernst's allowance for doubtful accounts deviated from IKON's accounting policies does not raise a plausible inference that Ernst was reckless or intended to publish materially false information, let alone deviated from GAAP.

23 . This figure regarding reconciliations was calculated during the Special Procedures, one year after Ernst issued its 1997 clean audit opinion. See J.A. 788-89 (Special Procedures Summary of Adjustments).

24 . There has been considerable ambiguity surrounding the $20.6 million figure cited by appellants. Initially, appellants included the allegedly "known” $3.6 million shortfall in IDS/IMS reserves to arrive at this figure. See Br. of Appellants at 28; see also Reply Br. of Appellants at 1; Transcript of Oral Argument at 74 ("[tjhat item would probably also be included in the 20 plus million dollar account receivable ... There is a duplication.”). Subsequent to oral argument, appellants submitted a memorandum on October 16, 2001, in which they now maintain that the $20.6 million does not double count the $3.6 million shortfall in IDS/IMS reserves and instead consists of $17,089,000 in inadequate accounts receivable allowance and $3,234,000 in intercompany accounts that did not eliminate. Appellants presented similar figures to the district court. See 131 F.Supp.2d at 693 n. 3. Nevertheless, in the interest of fairness, we address the $20.6 million only with respect to shortfalls in accounts receivable reserves, as Ernst was not presented with an opportunity to address the $3,234,000 in in-tercompany accounts that allegedly were not eliminated. See Dillinger v. Caterpillar, Inc., 959 F.2d 430, 446-47 (3d Cir.1992). In any event, appellants have not directed us to portions of the record that suggest recklessness on the part of Ernst for failing to eliminate the $3.3 million in intercompany balances. To the contrary, the record suggests that Ernst reconciled intercompany accounts at the consolidated level. See J.A. 5262 (work-paper).

25 . See note 22 supra and accompanying text.

26 . We do not suggest, however, that individual defects in an audit could not, in the aggregate, create an inference of scienter, particularly at the summary judgment stage. To the contrary, in many cases the most plausible means to prevail on a section 10(b) claim against an auditor — without that ever-elusive "smoking gun” document or admission — will be to show how specific and not insignificant accounting violations collectively raise an inference of scienter. We conclude only that the record before us in this case does not rise to such a level.