Opinion
CLASS ACTION NO. C99-01729 WHA
September 5, 2000
ORDER GRANTING MOTION TO DISMISS IN PART AND DENYING IN PART AND SETTING PRETRIAL DATES
INTRODUCTION
In this securities class action, this order DENIES defendants' motion to dismiss the consolidated complaint insofar as the complaint arises out of defendants' allegedly misleading allocation of "In Process Research and Development" ("IPRD") in mergers and acquisitions. This order GRANTS the motion to dismiss the remainder of the complaint for failing to satisfy the pleading requirements of the Private Securities Litigation Reform Act (PSLRA) and In re Silicon Graphics, Inc., Securities Litigation, 183 F.3d 970 (9th Cir. 1999). In addition, this order GRANTS the motion to dismiss in its entirety as to defendants Zachary A. Nelson and Peter R. Watkins.
STATEMENT
This is another of the recent securities case arising out of allegedly excessive write-offs for IPRD in mergers and acquisitions. For financial reporting purposes, such combinations may be reported pursuant to either the "purchase" or the "pooling of interests" methods of accounting. When the "purchase" method is used (as in six of the acquisitions at issue in the instant case), the total purchase cost of the acquisition must be "fairly" allocated, based on the comparative values, across all of the constituent assets acquired. One such asset — at the heart of the present suit — is IPRD. Legitimate IPRD may be (indeed, must be) written off immediately as an acquisition cost. This write-off is attractive to acquiring companies because it eliminates amortization of assets that would otherwise serve as a drag on future earnings. Excessive IPRD allocations, however, will lead to artificially inflated earnings. That is the nub of this case.
See Alabaster v. Bastians, Civ. Action No. 99-10237 NG (D.Mass. July 27, 2000); In re Engineering Animation Securities Litigation, 2000 U.S. Dist. LEXIS 5118 (D. Iowa March 24, 2000)
In 1975, the Financial Accounting Standards Board issued FASB Interpretation No. 4 regarding treatment of RD expenses in an acquisition accounted for by the purchase method. The Board drew a distinction between acquired assets resulting from RD (such as blueprints, formulas and patents) and acquired assets to be used in research and development activities of the combined enterprise (such as "materials and supplies, equipment and facilities and perhaps even a specific research project in process"). Most critical to this order, the Board imposed a requirement of fair allocation. The fair allocation of the purchase price across these (and all other) categories of assets was, according to the Board, to be made in accordance with Accounting Principles Board Opinion No. 16. Once the fair allocation was determined, the accounting by the combined enterprise for IPRD, the Board stated, was to be governed by FASB 2. Under FASB 2, certain IPRD (but not all IPRD) must be written off at the time of acquisition whereas all other RD must be capitalized and amortized as an expense over time. FIN 4 stated:
The subsequent accounting by the combined enterprise for the costs allocated to assets to be used in research and development activities shall be determined by reference to Statement No. 2. Paragraph 12 of Statement No. 2 requires that costs identified with research and development activities shall be charged to expense when incurred unless the test of alternative future use in paragraph 11(a) or 11(c) is met. That requirement also applies in a business combination accounted for by the purchase method. Accordingly, costs assigned to assets to be used in a particular research and development project and that have no alternative future use shall be charged to expense at the date of consummation of the combination. Therefore, the accounting for the cost of an item to be used in research and development activities is the same under paragraphs 11 and 12 of Statement No. 2, whether the item is purchased singly, or as part of a group of assets, or as part of an entire enterprise in a business combination accounted for by the purchase method.
(¶ 5) (emphasis in original) (footnote omitted).
In both scenarios, all RD costs will be expensed but the scenarios differ as to when. In the case of IPRD to be used in a particular RD project without alternative future uses, the write-off must occur as of the date of the acquisition. All other RD costs must be amortized over future periods as expenses. It is a question of timing. All of the foregoing has been in the accounting literature since well before the events in question.
* * *
On September 9, 1998, the Chief Accountant for the Securities Exchange Commission wrote a letter to the Chair of the SEC Regulation Committee of the American Institute of Certified Public Accountants ("AICPA"). The letter observed that in the 1990's, there had been a disturbing trend of larger and larger write-offs of IPRD in acquisitions involving public companies:
Prior to the 1990's, large write-offs of IPRD were rarely reported, even though acquisitions of high technology companies occurred frequently. Although there was no change in the relevant accounting literature, IPRD write-offs increased significantly in the 1990's. More intense merger activity in the technology sector may explain some of the increase, but abuses in the valuation of IPRD also are suspected. This trend of larger write-offs could undermine public confidence in financial statements and presents significant challenges for the accounting profession.
(Def. Supp'l App. Exh. 1).
The letter gave the following examples of abuses, among others:
A company acquired a product which it intended to continue to market. An updated version of the product had been released shortly before the acquisition and only a small amount of work towards the next release was completed. However, the vast majority of the purchase price was allocated to IPRD.
An appraisal of a company's IPRD was based on forecasted revenues for the next seven years. The forecast included estimated revenues not only for the existing product and the succeeding version under development at acquisition date, but also estimated revenues for later versions not then under development. The cost and time necessary to develop any future version would be much greater without the "core technology" included in the currently marketed version. Yet the appraisal attributed substantially all revenues from all the future product versions envisioned by the acquiring company to the value of IPRD.
* * *
A company allocated nearly all of the purchase price to IPRD, but the target company had not incurred or disclosed expenditures prior to the acquisition that would indicate any significant research and development effort.
The letter encouraged the AICPA SEC Regulations Committee to provide "additional guidance" to the profession concerning IPRD. In this regard, the SEC letter noted:
APB 16 requires that the allocation of purchase cost be based on "fair value." In some circumstances, some appraisers have defined fair value not as "fair market value" (e.g., the exchange price between a willing buyer and seller), but as "investment value to a particular buyer" (e.g., the value of the assets to the acquiring company). That approach is not defined in the accounting literature and is not appropriate to the extent it varies from "fair value."
The value of IPRD should be determined separately from all other acquired assets. The fair value of IPRD relating to an enhancement of an existing technology should not be different, depending on whether or not the acquirer already owns the technology itself. Value flowing from acquired brand names, customer relationships, and engineering and marketing resources should not be attributed to IPRD. Estimation of the fair value of IPRD requires consideration of factors specific to that asset, such as its stage of completion at the acquisition date, the complexity of the work completed to date, the difficulty of completing development within a reasonable period of time, technological uncertainties, the costs already incurred, and the projected costs to complete.
* * *
The "income approach" typically involves allocation of forecasted cash flows between IPRD and other assets. Allocations of cash flows among acquired assets should be challenged if they do not give full recognition to the contribution of existing products, core technologies and other acquired assets that must be capitalized. Cash flows attributable to development efforts, including the effort to be completed on the development effort underway, and development of future versions of the product that have not yet been undertaken, should be excluded in the valuation of IPRD.
* * *
If an existing product is acquired as part of the acquisition, and continues to be marketed by the purchaser, the fair value of the product should be capitalized. It should not be considered IPRD, even if the buyer intends to significantly modify or alter the product in the future. (There still could be an IPRD allocation if in fact there was an ongoing enhancement to the existing product in process at the acquisition date.) The staff has noted that FASB Statement No. 86, Accounting for the Costs of Computer Software to be Sold, Leased or Otherwise Marketed, defines a product enhancement to include "Improvements to an existing product that are intended to extend the life or improve significantly the marketability of the original product. Enhancements normally require a product design and may require a redesign of all or part of the existing product."
(Ibid.) (emphasis added).
The SEC letter concluded by stating that restatement of financial statements "will be necessary if a registrant's valuation IPRD is materially misleading" (ibid.).
On September 29, 1998, SEC Chairman Arthur Levitt made remarks entitled The Numbers Game to the NYU Center for Law and Business concerning "the more common accounting gimmicks we've been seeing." He decried the practice of "earnings management," the pressure on management to meet or beat Wall Street earnings projections. Calling the practice "merger magic," Chairman Levitt addressed abuses in large write-offs ascribed to IPRD as follows:
I am not talking tonight about the pooling versus purchase problem.
Some companies have no choice but to use purchase accounting — which can result in lower future earnings. But that's a result some companies are unwilling to tolerate.
So what do they do? They classify an ever-growing portion of the acquisition price as "in-process" Research and Development, so — you guessed it — the amount can be written off in a "one-time" charge — removing any future earnings drag.
(Joint Response of Plaintiff and Defendants in Response to Court's Order Dated July 26, 2000, at Tab B, page 4).
In essence, Chairman Levitt noted that acquiring companies prefer to take a one-time "extraordinary" charge against earnings associated with the "acquisition" so as to free the future earnings streams from expenses and thus make the future earnings from operations appear more profitable than they really are. The complaint herein so alleges as to Network Associates. It is, of course, a fundamental purpose of financial reporting to portray the profitability of operations accurately.
Despite the SEC's expressed position on proper allocation of IPRD under the existing standards, some in the software industry have complained that the SEC view of those standards amounts to new rules (Def. Br. at 12; Judith Burns, SEC May Face Lawsuit Over In-Process RD Charges, Dow Jones News Service, February 3, 1999, included in Defendants Supplemental Appendix in Support of Joint Response to Court's Order Dated July 26, 2000). Defendant, however, points to no accounting literature supporting this proposition or supporting its IPRD methodology. Possibly as a result of such reactions, the Financial Accounting Standards Board voted on February 24, 1999 to propose an amendment that expressly requires purchased IPRD to be recognized as an asset and expensed over its estimated useful life (In-Process RD: Capitalize or Expense? — An Accounting Roller Coaster Ride, Electronic News, March 15, 1999 included in Defendants Supplemental Appendix in Support of Joint Response to Court's Order Dated July 26, 2000). This amendment is apparently still at the draft stage and does not yet impose any new official requirement.
* * *
Network Associates, Inc., a Delaware Corporation with its principal place of business and corporate offices in Santa Clara, California, is a leading developer and provider of networking security and management of software products, such as anti-virus and network fault software. In 1995, Network acquired AIM Technology, Inc., for $7.2 million and allocated the entire cost of the acquisition to IPRD. In 1996, it acquired Vycor Corporation for nine million dollars and allocated $7.8 million to IPRD. Altogether, from 1995 to 1998, Network acquired six companies for a total of $394.6 million and allocated $289.5 million — 73% — to IPRD, broken down as follows (all in millions):
Acquiree Price IPRD Percentage AIM Technology, Inc. $ 7.2M $ 7.2M 100% Vycor Corporation $ 9.0M $ 7.8M 87% Cinco Networks, Inc. $ 26.0M $ 23.7M 91% Pretty Good Privacy, Inc. $ 37.7M $ 30.9M 82% Magic Solutions, Inc. $ 140.3M $ 97.0M 70% Cybermedia, Inc. $ 174.3M $ 122.2M 70% The foregoing figures are admitted by defendants. Plaintiffs originally alleged that the IPRD allocations were actually higher but by letter to the Court agreed to the foregoing. The prices shown include all consideration given, including assumption of debt.The acquired companies had established products, many of which were trumpeted in press releases issued by Network itself at the time of the acquisitions (Lead Plaintiff Vatuone's Response to Court's August 2, 2000, Request for Information). For example, AIM sold "Sharp Shooter," "a logical extension of Network General's core business." Vycor sold the Enterprise Family of help-desk products and was intended to become the consolidated repository for their network management line. Cinco Networks sold an "award winning" NetXRay monitoring-and-analysis tool and Network expected to begin distributing Cinco's products immediately after the acquisition. Pretty Good Privacy was a pioneer in email-encryption technology and developed PGP, a worldwide de facto standard for Internet email and file encryption. Magic Solutions had an installed base of more then 5,000 organizations and sold Support Magic 4.0, a widely-recognized support package. Finally, Cybermedia sold "self-healing" technology for PCs including leading brand names that were fixtures in the top ten of PC Data's monthly top selling software list (ibid.).
As stated, even though the acquired companies had established products in the market — and in some cases, top-selling brands in the market — and even though the acquired companies had established intellectual property, Network allocated the vast majority of the purchase prices to IPRD. This alone raises a question concerning "fair allocation." Network made its allocation, it appears, by projecting that the existing products and technology would be promptly phased out and superseded by substantially different products. So characterized, the latter's future income stream(s) dwarfed the former's, which was, as stated, assumed to be very short-lived. The income streams were then discounted to present value. The larger category was deemed assets "to be used" for future products, products still "in process," and thus classified as IPRD. It was expensed immediately as an acquisition cost. Little value was assigned to the actual existing and successful products and technology of the acquired entity.
On September 10, 1998, in the wake of the SEC letter to the AICPA, Defendant William L. Larson, Network's CEO and Chairman, convened a conference call of securities analysts and addressed Network's own history of IPRD write-offs. He stated, according to the consolidated complaint (§ 68(a)-(c)):
Of course, as a result of that we tried to take as large of an in-process technology write-off as third party auditors would allow under all the laws governing the SEC and the accounting profession and consistent with GAAP. We are 100% consistent with GAAP, we stand by these write-offs. . . . This was absolutely in the best interest of shareholders. We stand by that. . . . and so that significantly reduced the amortization we would have going forward of goodwill which is very large since there are essentially no assets in software companies. . . .
* * *
So, if we look at the Q2 charges for instance, which some short-sighted shorts on CNBC decided after never talking to management to make very erroneous and scandalous, practically libelous comments about, we had about, you know, we had $97 million in in-process technology write-offs, we had to pay for four bankers, $15 million in bankers' and legal fees, $20 million in severance, and only about $15 million in impaired assets, principally leasehold consolidations. There's nothing in there that we do not stand by 100%. There's nothing in there that could not withstand significant scrutiny by any and each and every one of you. There's nothing in those charges that any and each and every one of you would not have done if you had been me or our CFO or our auditors. So, I think it is pure poppycock that somebody could get on a public forum and make libelous statements about a company, but that is called freedom of the press, it's the environment we live in. What is really sad is for those of you who know us, and seen us perform consistently, to believe such statements and that I have come on this call now to reassure you all which I'm more than happy to do, but it's not the most effective use of my time, given that you are investors in this company I should be out talking to customers.
* * *
So, in closing, we are bullish on the quarter. The numbers in our write-offs are squeaky clean; we stand by them. The inventories are lean in the channel. We have more than adequate reserves. . . .
Chairman Larson further stated that the company had been building "little honey pots that you can go to, because when you make these acquisitions in our business planning, you always project the sequential decline in the revenue the acquired company and we have to go back to our core businesses to get faster growth to cover the potential 10%, to 15% decline in revenue. And that is where you see the releasing of backlog . . . and you can only do that if you have created, you know, those acorns out there" (Compl. ¶ 68). Although Chairman Larson further stated that the accounting treatment was not only "100% consistent with GAAP," but also "some of the most conservative accounting in the world," the company was, not long thereafter, to reverse field and to reverse $230 million in IPRD write-offs.
On December 23, 1998, the SEC sent a letter to Prabhat K. Goyal, the CFO of Network. This letter followed the Levitt speech and SEC letter to the AICPA by a few weeks. The letter concerned various Forms 10-K, 10-Q and 8-K filed by Network for 1997 and 1998. In part, the SEC letter to Network questioned the IPRD allocations:
We note that you allocated a substantial portion of the purchase price to acquired in-process technology (IPRD) in connection with the Cybermedia ($122 million) and Magic Solutions ($97 million) acquisitions. Supplementally tell us and revise to disclose the nature of the IPRD acquired from Cybermedia and Magic Solutions, respectively, how you determined the value of the IPRD costs, and how these costs meet the definition of research and development costs under paragraphs 8 and 9 of SFAS 2. Tell us how you determined the IPRD costs had no alternative future uses (paragraph 11(c) of SFAS 2). Give us (a) your analysis to support the purchase price allocation including a copy of any independent appraisals, (b) copies of any press releases and/or information presented to your board of directors regarding these acquisitions. Confirm that the projections are consistent with or the same as any other projections furnished to outside parties, such as lenders, or developed internally for other purposes. If you cannot confirm this, please thoroughly explain any differences and quantify the impact of the differences on the valuation.
(Appendix A to Lead Plaintiff Vatuone's Opp. to Defendant's Motion to Dismiss at page 5).
The record does not contain Network's response, but in April 1999, Network substantially reversed its prior write-offs of the IPRD. This action evidently avoided an SEC enforcement action. This reversal resulted in capitalization of most of the acquisition purchase prices, as follows (in millions):
Acquiree Original Restated Restated IPRD IPRD Percentage Write-Off Write-Off of Total Price AIM Technology, Inc. $ 7.2M 0 0 Vycor Corporation $ 7.8M 0 0 Cinco Networks, Inc. $ 23.7M $ 5.2M 20% Pretty Good Privacy, Inc. $ 30.9M $ 3.9M 10% Magic Solutions, Inc. $ 97.0M $ 27.0M 19% Cybermedia, Inc. $ 122.2M $ 22.8M 13% The alleged misrepresentations in the company's financial statements — rooted in the IPRD practice — were made throughout the period 1995 through 1998, including during Chairman Larson's conference with analysts in September 1998. On January 6, 1999, the company issued a press release reporting it had received a comment letter for the SEC concerning the IPRD and stating that the SEC had recently issued "new guidelines" concerning IPRD (Compl. ¶ 69). The company further stated that the impact of the IPRD issue would be no more than $.02 to $.03 per quarter (Compl. ¶ 72). In reaction to this announcement, the stock price dropped from $59.938 to $58.188 (ibid.).On April 6, 1999, however, the company announced that it was reversing $169 million in prior IPRD write-offs for fiscal year 1998 and $45 million for fiscal year 1997 and that IPRD amortization expense for fiscal year 1998 was increased to $43 million and for fiscal year 1997 $13 million. Overall, the company reversed IPRD write-offs of $230.4 million for the period 1995-99 (Compl. ¶¶ 73-74). The company also revised downward its first quarter 1999 revenue and earnings projections. In response to both factors (IPRD and downward projection), the stock price dropped 27.1% (Compl. § 78).
On April 19, 1999, the company released its first quarter 1999 results. The results showed that the impact of the IPRD change was $0.12 per share, not the $0.02 to $0.03 previously represented (Compl. ¶ 80). On the same day, the company cut its second quarter 1999 estimate of sales drastically, from $230 million to $20 million. Defendants explained that the lower sales figure was due to the need to clear inventory from the company's distribution channels (Compl. ¶ 82). In reaction to these announcements, the stock price dropped another 27.5% (to $4.187 per share). These consolidated suits immediately followed.
ANALYSIS In Process Research and Development
Overstating IPRD is misleading to investors because it makes ongoing sales operations look more profitable than they really are. Because companies are often valued based on the present value of their future net earnings streams, eliminating costs from the equation can inflate the present value of the enterprise and thus the current stock price. The large amounts involved here were plainly material or so it surely must be held at the pleading stage. The main question presented by the motion is whether the heightened pleading requirement for scienter imposed by the PSLRA has been met.
The PSLRA raised the pleading bar in securities cases by requiring plaintiffs to allege "with particularity facts giving rise to a strong inference" that each defendant "acted with the required state of mind." 15 U.S.C. § 78u-4(b)(2). The required state of mind in the Ninth Circuit for non-forward-looking statements such as the IPRD write-offs is deliberate recklessness. In re Silicon Graphics, Inc. Sec. Litig., 183 F.3d 970, 974, 985 (9th Cir. 1999). The required state of mind, however, can be proven and pled through circumstantial evidence. It will be rare that a wrongdoer will admit to the required state of mind. The PSLRA calls for a "strong inference," not an outright confession or an airtight case at the pleading stage. See In re Peoplesoft, Inc. Sec. Litig., No. C 99-00472, slip op. at 5 (N.D.Cal. May 26, 2000).
Two precedents have addressed IPRD in securities cases. Judge Longstaff rejected a PSLRA motion to dismiss in In re Engineering Animation Securities Litigation, 2000 U.S. Dist. LEXIS 5118 (D. Iowa March 24, 2000). His order stated:
The Court notes [footnote omitted] that the allocation of IPRD by defendants appears to have been done at a time when many companies were engaging in similar practices.
* * *
The fact that many companies were doing the same thing, however, does not excuse a fraudulent practice. Further, the amount by which EAI re-allocated its IPRD calculations, by 70% and 81%, supports a contention that this was not just a marginal judgment call by EAI's accountants. It appears EAI consciously choose to show a very significant portion of the value of their acquisitions as IPRD. The magnitude of their reporting decisions was so large that defendant's IPRD allocation accounting, and related statements, could be shown to be fraudulent. See also in re Leslie Fay Companies, Inc. Securities Litig., 835 F. Supp. 167, 174-75 (S.D.N.Y. 1993) (impliedly finding that the magnitude of the alleged fraud is appropriate to consider when determining whether to dismiss a securities fraud class action). Further the Court cannot find as a matter of law that these statements are immaterial [footnote omitted].
* * *
Even if the Eighth Circuit were to find motive and opportunity is not an available method to evidence scienter under the PSLRA, the recklessness standard has also been met by the plaintiffs in the instance. The original IPRD allocations had to be reduced by 70%, a nearly four million dollar reduction, and by 81%, a nearly eight million dollar reduction. Further, the original IPRD figures were very significant parts of the total acquisition transaction — 97% in the Sense-8 acquisition. EAI was telling investors they were buying a company's future growth, only later to change their minds and tell the investing public that what they had really purchased were assets. The GAAP violations alleged here, along with other circumstances, create a strong inference of scienter. See In re Digi Intern, Inc. Securities Litig., 6 F. Supp.2d 1089, 1098 (D.Minn. 1998) ("A violation of GAAP, combined with other circumstances suggesting fraud, may create a strong inference of scienter") (citing Rehm v. Eagle Finance Corp., 954 F. Supp. 1246, 1255-56 (N.D.Ill. 1997) [footnote omitted] and Marksman Partners, 927 F. Supp. 1297 (C.D.Cal. 1996)).
In Alabaster v. Bastians, Civ. Action No. 99-10237 NG, (D.Mass. July 27, 2000), on the other hand, Judge Gertner rejected an IPRD theory. She found that "plaintiffs' allegation fail to established that LH was aware that its accounting methods were improper at the time they were employed." She noted that plaintiff had "offered no other allegations of warning signs which should have put LH on notice that its conduct was improper." Id., slip op. at 19. Neither precedent is on all fours. The instant case must turn on its own facts and circumstances, although we shall return to these precedents for comparison below.
Here, it seems clear that there was a GAAP violation; otherwise, the company would not have restated hundreds of millions in IPRD. A GAAP violation alone, however, does not prove scienter. Defendants contend that nothing more than an honest difference of interpretation with the SEC has been shown. The following circumstances taken together, however, raise a strong inference as to the scienter of the company itself and defendants Larson and Goyal:
First, although a GAAP violation alone does not translate into a per se securities violation, the magnitude of the GAAP violation is a circumstance that may be taken into account in assessing scienter. See In re Engineering Animation, 2000 U.S. Dist. LEXIS at *33 (and cases cited therein). Here, the magnitude was large. The reversed IPRD was $230 million.
Second, the repetitive character of the violation is relevant. Here, in every one of six acquisitions using the purchase method, Network allocated the vast majority of the purchase price (and in one case all of the purchase price) to IPRD, and precious little to the proven products and core technology of the acquired company. On its face, this does not seem to be a "fair" allocation. Third, it is hard to reconcile the company's lop-sided allocations with its press releases at the time of the acquisitions, most of which extolled the target company's extant products and technology. To restate only one example, Magic Solutions was hailed as a "pioneer" with "an installed base of more than 5000 organizations." Its SupportMagic 4.0 was "widely recognized." The release said that the product would continue to be sold and supported (Lead Plaintiff Vatuone's Response to Court's August 2, 2000, Request for Information). In contrast to these public statements, however, Network actually allocated little value to the acquired core products and core technology. Network gave short shrift to the value of the core technologies of the acquired companies for public reporting purposes even while trumpeting that technology at the time of the acquisition. Such disproportionate allocations, the SEC said, failed under GAAP to "give full recognition to the contribution of existing products, core technologies and other acquired assets that must be capitalized" (Sept. 9, 1998, Letter at 3). The fundamental GAAP principle called for (and still calls for) a "fair allocation." These allocations were so imbalanced when viewed in the context of the company's own releases as to support a strong inference of scienter. Perhaps an innocent explanation will emerge at trial, but at this stage the blatant inconsistency supports a strong inference of accounting fraud.
Fourth, the practice of heavy IPRD allocation was intentional. This seems clear from defendant Larson's comments. In September 1998, he stated that the company had "tried to take as large of an in-process technology write-off" as the auditors would allow under GAAP and the SEC rules (Compl. ¶ 68(a)). This was done, he said, precisely to avoid future amortization. Although Larson publicly stated that this was "some of the most conservative accounting in the world," it seems reasonably clear that the company actually and consciously chose the opposite — "aggressive accounting." It did so specifically to reduce a drag on ongoing net income. In turn, that boosted, one would reasonably suppose, the stock price by relieving the future net earnings stream of the burden of material expenses, at least insofar as the market determined price based on the present value of the future net earnings stream.
Defendants contend, as stated, that the IPRD allocations were sincere judgment calls based on appraisals and audits and that others in the industry did the same thing. Perhaps. But any good-faith reliance on professional opinions would be a defensive matter for trial. The defendants signed financial reports and filed them with the SEC, making direct public representations as to the financial condition of the company. Defendants pointed to no accounting literature justifying their methodology or their disproportionate allocations. The PSLRA does not require the complaint to disprove every conceivable innocent explanation for the GAAP violation. Plaintiff must plead facts and circumstances that give rise to a strong inference of deliberate recklessness. See Silicon Graphics, 183 F.3d at 974. This they have done for the reasons stated. They have raised a strong inference of deliberately reckless disregard of the GAAP requirement that the acquisition price be "fairly allocated" across all assets acquired, including proven products and core technologies. In contrast to Alabaster, the instant case reveals that the defendants, through their own press releases, knew that the acquired companies had significant core technologies and products quite apart from any in process research and development. What is more, the IPRD taken in the present case was many times greater and more disproportionate than in the Alabaster case. The circumstances pled in the present case raise a strong inference that the company recklessly or deliberately allocated excessive amounts to IPRD.
Revenue Recognition
The complaint also claims that defendants violated GAAP through "old-fashioned revenue recognition schemes" (§§ 2, 102, 103, 105). Plaintiffs allege that defendants delivered to distributors products that they knew would ultimately be returned — obsolete products and more product than could reasonably be sold — and thus "stuff[ed] the channels of distribution" (§ 29; Plaintiffs' Attachment A to Joint Response to Court's Order Dated July 26, 2000 ¶¶ 5). This, in turn, enabled defendants to meet quarterly estimates by misleadingly reporting earnings that did not really exist and that would eventually have to be reversed (ibid.). By April 1999, however, defendants were forced to acknowledge that inventory needed to be cleared from Network's distribution channels (§§ 81-83). Defendants explained that they had been forced to revise their future expectations and realign their inventory due to so-called Y2K issues and sales cycles (ibid.). Near the end of the complaint, one lengthy multi-part paragraph enumerates GAAP principles that plaintiffs believe defendants have violated. Although it is not completely clear because the principles listed are not explicitly connected to any of the practices described earlier in the complaint, several of these might be applicable to the revenue recognition allegations (§ 90(q), (w), (z)). In particular, the requirement of FAS No. 48 seems relevant: that for software vendors revenue cannot be recognized before delivery is complete or while substantial contingencies exist as to full and final payment or potential cancellation (§ 90(q)).
These allegations fail to meet the strict pleading standards of the PSLRA. The PSLRA requires that where, as here, plaintiffs are pleading "on information and belief, the complaint shall state with particularity all facts on which that belief is formed." 15 U.S.C. § 78u-4 (b)(1)(B). The complaint presents no such facts. Plaintiffs, in response to the Court's request, submitted identifications, by job and sometimes office location, of the individuals who were the sources of those facts (Plaintiffs' Attachment A to Joint Response to Court's Order Dated July 26, 2000 ¶ 5; Plaintiff's Letter of August 16, 2000). They did not, however, present the names of those sources, wanting to respect the sources' desire to remain anonymous. More importantly, they also did not describe how they gained the relevant information from the sources, and what form that information took — whether it was written or oral; whether it consisted of recollections or contemporaneous descriptions, vague hints or detailed explanations. Cf. Silicon Graphics, 183 F.3d at 985. Without deciding whether the names should have been disclosed, the totality of these problems provides little meat to the bare bones allegations.
Even were the Court to find that the allegations in regard to revenue recognition were stated with enough particularity to satisfy 15 U.S.C. § 78u-4(b)(1)(B), the Court cannot find that those allegations satisfy the requirements of 15 U.S.C. § 78u-4(b)(2). As discussed above, the PSLRA requires that the plaintiff state with particularity "facts giving rise to a strong inference of deliberate recklessness or intent." Silicon Graphics, 183 F.3d at 985. Bunched together at the end of the complaint are a series of paragraphs alleging that defendants had the requisite scienter in regard to their various misleading statements. Several of these may be relevant to the revenue recognition allegations (§§ 93, 99-101, 104). Plaintiffs' supplemental submissions also contain facts intended to demonstrate scienter (Plaintiffs' Attachment A to Joint Response to Court's Order Dated July 26, 2000 ¶ 5).
Plaintiffs suggest that defendants' sales of stock while stock value was inflated due to improper revenue recognition suggest scienter (Compl. ¶ 104), and that defendant Larson's statements in his September 10, 1998, conference call indicate scienter (§ 93). This is insufficient. Although such factors do, perhaps, give rise to an inference of scienter, it can hardly be described as a strong inference. Larson's statements in the conference call were too vague to be clearly applied to these particular allegations, and seem more relevant to defendants' IPRD violations. As for the stock sales, although "[i]nsider trading in suspicious amounts or at suspicious times is probative of bad faith and scienter," In re Apple Computer Sec. Litig., 886 F.2d 1109, 1117 (9th Cir. 1989), it is suspicious "only when it is `dramatically out of line with prior trading practices at times calculated to maximize the personal benefit from undisclosed inside information,'" id., quoted in Silicon Graphics, 183 F.3d at 986, 987. Here, plaintiffs fail to demonstrate why the stock sales were suspicious enough to demonstrate scienter. Defendants' alleged insider trading took place before November 30, 1998, several months before the April 1999 drop in Network stock value (Compl. ¶ 11-14). Finally, plaintiffs' supplemental submissions, while helpful in providing factual specificity regarding the alleged improper revenue recognition itself, do not demonstrate deliberate recklessness or intent (Plaintiffs' Attachment A to Joint Response to Court's Order Dated July 26, 2000 ¶ 5). In fact, the supplemental submissions indicate the opposite: that when Network realized that its use of "letters of intent" to book sales created problems because of customers later canceling orders, they discontinued the practice. Plaintiffs thus fail to state a claim in regard to their allegations of improper revenue recognition.
Forecasts for First and Second Quarter 1999
Plaintiffs also allege that defendants made misleading forward-looking statements, specifically their revenue and earnings predictions concerning the first and second quarters of 1999 (Compl ¶¶ 75-84). On March 15, 1999, defendants reaffirmed their previous revenue and earnings estimates for these quarters, predicting revenues of $280 to $285 million for the first quarter and $230 to $260 for the second quarter. Only three weeks later, however, defendants had to concede that those estimates were significantly inaccurate (Compl ¶ 99). On April 6, 1999, defendants announced a downward adjustment of at least $30 million in their first quarter estimates to $245 to $250 million (Compl ¶¶ 77-78). During a conference call on April 19, 1999, the actual final revenue for the quarter, $245 million, was announced (Compl ¶ 80). Once the additional amortization charges necessitated by the SEC investigation were taken into account, this adjustment impacted earnings by approximately $0.12 per share. Second quarter forecasts were adjusted downward even more steeply on April 19, 1999. At that time defendants revealed that they would ship only $20 million of product in that quarter, a drop in predicted revenue of at least $210 million (Compl ¶¶ 82-83, 99).
There is no denying that such dramatic downward adjustments, announced so shortly after reiteration of the previous high estimates, legitimately give rise to suspicion. The PSLRA, however, steps in the way of this logical suspicion giving rise to liability. Under the PSLRA, as plaintiffs are pleading on information and belief, they must state with particularity all facts on which that belief is formed. See 15 U.S.C. § 78u-4(b); Silicon Graphics, 183 F.3d at 984. Here, the complaint is generally vague about the details of the conversations with analysts in which the financial projections in question were allegedly made (Compl. ¶¶ 75-77). Cf. Wilson v. CKS Group, No. C 98-4229 MMC, slip op. at 10 (N.D.Cal.
Mar. 21, 2000) (dismissing claims based upon alleged statements to analysts where complaint did not identify "any specific facts regarding those conversations").
Even where plaintiffs can specify which defendants participated in the conversations and some of the content of their remarks (Compl. 76(g); Plaintiffs' Attachment A to Joint Response to Court's Order Dated July 26, 2000 ¶¶ 2-3), the complaint fails to overcome another hurdle of the PSLRA in this area as well. The PSLRA establishes a high standard of scienter for forward-looking statements. Whereas defendants' IPRD allocations constituted allegedly misleading statements about present conditions, any financial projections defendants made about the first and second quarters of 1999 were "forward-looking statements." 15 U.S.C. § 78u-5(I)(1) (defining forward-looking statements). As such, defendants may not be found liable for those projections unless plaintiffs allege facts creating a strong inference that defendants made the challenged statements with actual knowledge that the statements were false or misleading. See 15 U.S.C. § 78u-5(c)(1)(B). This is true regardless of whether the forward-looking statements were accompanied by "meaningful cautionary statements." Id. This standard of actual knowledge of falsity at the time the statements are made is even more difficult to satisfy than the Silicon Graphics deliberate recklessness standard.
The complaint lacks sufficient facts to indicate that defendants had actual knowledge that the statements were false or misleading at the time they were made. Bare allegations of such knowledge, without supporting facts, are certainly insufficient (Compl. ¶ 98). Plaintiffs note that the close proximity of the projections and revisions suggests that the (dramatically lower) revised estimates were already known at the time of the projections (Compl. ¶¶ 97, 99). This is a logical inference, but standing alone it is not "strong." Plaintiffs also suggest that the transparency of defendants' explanations of the SEC inquiry into Network's treatment of IPRD (claiming that the SEC had issued "new guidelines" when in fact it had not (Compl. ¶ 96)) and of its inaccurate earnings and revenues estimates in 1999 (blaming its lower revenues on Y2K problems even though these allegedly did not affect competitors) indicates knowledge of the truth. These explanations, however, were not so patently false as to create a strong inference of scienter. Finally, plaintiffs allege that defendants' "insider trading" demonstrates scienter (Compl. ¶¶ 104-5). As discussed above, plaintiffs fail to demonstrate why defendants' stock sales were so dramatically out of line with their prior trading practices as to be suspicious. See Apple Computer, 886 F.2d 1109 at 117.
Although especially the precipitous $210 million drop in second quarter revenue predictions does raise an inference of fraud, on this record the inference is not strong enough to state a claim. The complaint "lacks sufficient detail and foundation necessary to meet either the particularity or strong inference requirements of the PSLRA." Silicon Graphics, 183 F.3d. at 984.
Pooling of Interest Method of Accounting
Plaintiffs also allege that defendants made materially false and misleading statements in presenting their acquisition of several companies pursuant to the pooling of interest method of accounting (Compl. §§ 39, 47, 57, 67). These allegations fail to meet the pleading standard of the PSLRA. See 15 U.S.C. § 78u-4(b)(2). The complaint merely lists the names of the acquired companies and the year of their acquisition, then recites a rote claim that defendants did not make the proper disclosures in reporting those acquisitions. Plaintiffs do not even address the issue of scienter in regard to these claims. They do not plead any facts that could lead to an inference — much less a strong one — that defendants acted with the required state of mind, deliberate recklessness. See Silicon Graphics, 183 F.3d at 984 Such boilerplate assertions of wrongdoing fail to state a claim under the PSLRA.
Defendants Nelson and Watkins
Plaintiffs have failed to allege the necessary detail as to defendants Nelson and Watkins. Nelson's and Watkins' positions of responsibility at Network and their stock sales during the class period are detailed in the complaint (§§ 12, 13). No further specifics regarding their statements, actions, or scienter are given in the complaint, however. Plaintiffs' opposition to defendants' motion to dismiss details Nelson's and Watkins' responsible roles in product development in more depth, but (even if these assertions were considered as part of the complaint for purposes of satisfying the pleading requirements of the PSLRA) does not allege that these defendants participated in any allocations of IPRD or made any specific forward-looking statements (Opp. 18-19). The complaint does not give enough information about these defendants' position at Network to determine whether their day-to-day work would even involve knowledge of the alleged misleading statements, and indeed neither the Complaint nor the Opposition directly alleges that they did know of any such statements. Plaintiffs allege that both Nelson and Watkins were involved in product development, but do not present facts indicating that they had any knowledge of or involvement with the accounting or marketing side of the business. This is insufficient to state a claim against them, even under the "group published information" doctrine. As to defendants Nelson and Watkins, therefore, the motion to dismiss is GRANTED in its entirety.
CONCLUSION
1. Subject to a motion to certify and a ruling thereon, the class period shall be January 20, 1998 to April 6, 1999. By September 22, 2000, plaintiffs shall file an amended consolidated complaint that deletes all alleged misrepresentations not directed to claims of IPRD mis-allocation, deletes defendants Nelson and Watkins, and adds the supplemental material supplied at and after oral argument concerning the IPRD issue only, including the press releases. Leave to amend is otherwise DENIED, for the Court has already considered the further material the lead plaintiff has offered to add.
2. No further motions to dismiss shall be made. The Court has already considered the supplemental material. Defendants shall answer by October 6, 2000.
3. Discovery may commence immediately. Reasonable discovery may be taken regarding: All of the acquisition accounting concerning fair allocation of all assets.
Evidence concerning the process for making the allocations.
All statements made to the board of directors, internally, or by or to the investment bankers concerning the acquisitions.
All press releases concerning the acquisitions.
All internal memoranda that dealt with the purpose of the acquisitions.
Evidence in possession of the acquired companies that would reasonably relate to the value of the assets acquired.
Evidence concerning plans for use of the existing core technology and existing products of the acquired company at or about the time of the acquisition or accounting treatments in question
Evidence going to the factors set forth in the SEC letter of September 9, 1998 concerning proper allocation to IPRD such as, without limitation, the relative amount of attention given to the various acquired assets and assumed liability during due diligence and presentation to board of directors
Evidence going to factors specific to the assets acquired, such as their stage of completion at the acquisition date, the complexity of the work completed to date, the difficulty of completing development within a reasonable period of time, technological uncertainties, the costs already incurred, and the projected costs to completion.
Evidence concerning communication with appraisers.
Evidence concerning communication with auditors.
Audit work papers.
Evidence that reasonably relates to the causes for the decline in Network stock value on January 6, 1999, and April 6, 1999. Subject only to the foregoing and to the following proviso, no discovery shall be allowed concerning revenue recognition issues.
Defendants shall allow reasonable discovery, regardless of time period, relating to evidence they expect to offer in their defense at trial.
Discovery is not limited strictly to the specific subjects referenced above. The Court has used the foregoing as illustrative. There will likely be other permissible discovery areas.
4. By September 15, counsel are to meet and to confer and to devise a process for Network to produce its documents (or substantially to do so). They shall submit the plan to the Court by September 15, 2000. The undersigned will supervise discovery disputes.
5. Trial shall commence on June 4, 2001 at a time to be determined at the final pretrial conference. The final pretrial conference shall be at 2:30 p.m. on May 21, 2001. The last day to file dispositive motions shall be on April 5, 2001. Fact discovery shall be completed by March 2, 2001 and detailed expert reports exchanged by March 2, 2001. Rebuttal reports shall be due 14 days later. All expert discovery shall close on March 30, 2001.
IT IS SO ORDERED.
.