Two Recent First Circuit Court of Appeals Decisions Clarify Corporate Duty of Disclosure In Public Offerings and In Post-Offering Communications
The Computervision Decision
In 1992, Computervision, formerly known as Prime Computer, undertook a public offering of $600 million in stock and notes. During the three years prior to the public offering, Prime/Computervision's business had suffered as the world increasingly turned to PCs, rendering "mini-computers," such as those of Prime/Computervision, increasingly less marketable. Between 1989 and 1992, Prime/Computervision had lost nearly $1 billion.
Within Prime/Computervision, however, the software business had performed well. Prime/Computervision manufactured state-of-the-art CAD/CAM software used by an array of Fortune 100 manufacturing companies to design complex products. In response to the increasing losses in the computer business, and the financial burden remaining from a 1989 leveraged buyout, Prime/Computervision's management altered its business strategy and, as part of that effort, undertook the public offering in August of 1992.
The business plan behind the public offering was straightforward; Prime/Computervision would stop making computers, change the company's name to Computervision, and focus its efforts primarily on the sale of CAD/CAM software and running its substantial computer servicing business. Needless to say, substantial risks were associated with this ambitious business plan. Accordingly, the prospectus was highly cautionary and outlined an array of relevant risk factors. In addition, during the course of the due diligence process, the underwriters reduced the anticipated selling price of Computervision securities from $19.00 a share to $12.00 a share.
In mid-August of 1992, the public offering took place at a reduced price of $12.00 per share. However, six weeks after the offering took place, Computervision announced that revenues for the third quarter then ending would be lower than internal plans and lower than the preceding year's third quarter. Although the servicing business performed well, the higher-margin software sales had suffered in the face of the September 1992 European monetary crisis. Since 66% of Computervision's sales came from overseas markets, the unpredictable collapse of the "European exchange mechanism" had a devastating impact on Computervision's software sales.
The reaction of the stock market to these events was swift and the trading price of Computervision shares dropped to a low of approximately $5.50 per share. The well-organized plaintiffs' securities bar reacted just as swiftly; within one day, the first of eighteen lawsuits was on file. Eventually all of the lawsuits were consolidated in the federal court in Boston, Massachusetts, discovery was conducted over a six-month period, including extensive production of documents and depositions of all participants in the offering, and, by the summer of 1995, the trial court concluded that the plaintiffs could not make out a case and dismissed the action.
The plaintiffs then appealed to the First Circuit, the federal appeals court responsible for cases arising in eastern New England. The central issue before the Court of Appeals was whether the Complaint established a claim under the Securities Act of 1933. Reduced to its simplest element, the 1933 Act requires that when companies issue new securities they are responsible for insuring that all material information is accurately disclosed. This requirement, along with the strict liability aspects of the 1933 Act, drive the comprehensive and extensive "due diligence" process that is the backbone of underwriting.
On appeal, the plaintiffs focused on intra-quarter financial data at the time of the IPO. "Intra-quarter" data is simply the various techniques companies use to try _ uncertainly _ to measure performance during each week of a quarter, often against internal forecasts which, for many good reasons, are seldom disclosed publicly. The plaintiffs argued that the "intra-quarter" data allegedly showed variances between Computervision's internal forecast for the quarter and the performance to date and predicted the disappointing results that occurred six weeks later. However, like intelligence intercepts during the months prior to Pearl Harbor, sometime the "predictive" pattern of data can only be seen with the benefit of hindsight.
The defendants argued that requiring disclosure of "intra-quarter" data would amount to requiring companies to disclose projections and would obscure the bright line between historical data (which must be disclosed) and projections (which may, but need not, be disclosed). The securities laws have long been clear that, while encouraged, projections or forecasts need not be included in a prospectus.
In deciding the case, the Court accepted the plaintiffs' theory as a viable basis for a claim under the 1933 Securities Act. In short, the Court stated that "[t]he price set for an offering of securities is essentially a forecast." From this premise, the Court concluded "to the extent current information up to the date of the offering was not incorporated into the prices, the statements in the Prospectus presented a misleading half-truth because they suggested that the underwriters and Computervision took into consideration current estimates of business potential and earnings prospects." The Court, however, rejected the plaintiffs' claim because, after years of litigation and discovery, they had presented no evidence to support their theory.
The challenge Computervision poses is to determine what intra-quarter data must be disclosed for the quarter in which an offering takes place. Naturally, accurate financial and other historical data through the quarter completed prior to any offering must be included. However, Computervision raises the question of when must the prospectus disclose information about the company's financial performance to date during the uncompleted quarter in which the offering takes place. This same question had been answered by the Court of Appeals a few weeks earlier in the Shaw v. Digital Equipment case. There, the court concluded that the plaintiffs had sufficiently alleged that the company had been in possession of material intra-quarter financial performance data so late in the quarter (i.e., two weeks before the quarter ended) that, if the allegations were true, Digital had an obligation to include such information. Accordingly, the Court sent the case back to the district court to allow discovery and the development of the record.
The different result in Computervision was driven, at least in part, by two factors. First, in Computervision, the plaintiffs had full discovery and yet still had failed to produce specific and meaningful allegations of fact that would support their theory with any detail. Second, the Computervision offering had taken place almost precisely in the middle of the quarter. Accordingly, there was less basis to infer that the company knew, or could have known, of the likely results for the quarter.
Computervision answered some of the questions that had been raised by the Digital decision. Clearly, the fact that an offering takes place in the middle of the quarter diminishes the basis to assert that intra-quarter data could predict the quarterly results. Especially in technology companies with "hockey stick" revenue patterns (and assuming appropriate disclosure of such patterns in the prospectus) such an inference, without more, is insufficient. Computervision also reaffirms the fact that companies issuing securities are not obligated to make predictions about future events, including the likely performance for the quarter in which the offering takes place.
However, Computervision and Digital are reminders to underwriters and issuers that they must be particularly sensitive to any signs during a quarter that anticipated revenue and earnings levels might not be met. Things like declining bookings, variances between budgeted and actual revenues, decreased "order visibility," and lengthening DSO are all potential markers for a poor quarter_especially with the benefit of hindsight. Since class action plaintiffs always have the benefit of hindsight, when such factors are known, issuers and underwriters must give serious consideration to disclosing the specific risk that such "intra-quarter" factors might cause a quarterly variance in financial performance as soon as the quarter in which the offering is taking place.
On August 12, 1996 the Court of Appeals again spoke to the issue of disclosure under the securities laws, this time in the context of the 1934 Securities Act. Here, the Court reaffirmed important principles that have been central to how the First Circuit treats open market security fraud cases.
Summa Four, Inc., based in Manchester, New Hampshire, is a leading producer of high technology telecommunications "switches" used to process telephone communications and to deliver an array of specialized services. Summa Four's business characteristically entails a relatively small number of orders in a given quarter, each of which is, on average, very substantial. Typical customers are companies such as AT&T, Sprint and MCI, who purchase one or more Summa Four switches, often configuring them with custom software and other applications. As a result, the delay of one or two orders at the end of a quarter can have a substantial impact on the results for the quarter.
In the quarter ending June 30, 1994, the Company did not close a number of anticipated orders and, as a result, on July 5, 1994, the Company announced revenues for the quarter that were below the range forecasted by analysts who followed the company. Summa Four's stock dropped sharply.
Almost immediately, a class action lawsuit was filed in the United States District Court in Concord, New Hampshire. The allegations focused on a number of accurate historical statements and attempted to advocate that the statements created a duty on the part of the company to provide intra-quarter data. For example, in January of 1994, in commenting on the quarter that had just ended, the company accurately stated that, in some international markets, demand had been increasing for Summa Four's switch. The plaintiffs contended that this type of accurate statement created a duty on the part of the company to disclose intra-quarter budget variations for the first two months of the quarter.
In affirming the district court's dismissal of the action, the Court of Appeals rejected the plaintiff's argument. The Court reiterated that accurate reports of historical data do not create an obligation to comment upon, project or speculate about future quarters. This distinction draws a bright line between accurate reports of historical events and informed speculation about future prospects. Despite concerted efforts by the plaintiffs' bar to obscure this line, the Court has affirmed that the line is clear and meaningful. Whether in press releases, analyst conference calls or SEC reporting, when companies limit their commentary to accurate reports of historical events, they are unlikely to be burdened with a duty to comment on current events or to speculate about future events, at least in the First Circuit.
The Lessons of Computervision and Summa Four
Both Computervision and Summa Four dealt with intra-quarter data. In both cases, the plaintiffs were attempting to advocate a standard that will obscure the bright line between accurate reports of historical data and informed speculation about future prospects.
In Computervision, the Court has stated that, at least as a matter of theory, the line might not be as clear as it once was. Issuers and underwriters need to give careful consideration when securities are priced not only to the company's performance through the quarter just ended, but right up until the time of pricing based on weekly performance to that point. More importantly, disclosure must adequately address the risk that partial quarterly results may be at variance with internal projections and explain the ramifications of such risks.
In Summa Four, on the other hand, outside the IPO context, the Court declined to expand the limited circumstances when intra-quarter data must be provided. Rather, when companies limit their commentary to accurate reports of past results, they will seldom be compelled to disclose later developments (other than at the regular reporting period) even if those results are less rosy. Nevertheless, in reporting on quarters that have just ended, companies should make clear that they are limiting their commentary to the closed quarter, not the current quarter. When such practices are followed, the chance for dismissing shareholder suits at an early stage (or even discouraging the filing of such suits in the face of a disappointing quarter) will remain realistic in the First Circuit.
In the end, the Court was sensitive to the special nature of IPOs. In cases involving IPOs (such as in Computervision), companies will continue to be held to an extremely high standard. Winning dismissal will be challenging unless it is clear that, even with the benefit of discovery, the plaintiffs simply cannot articulate a claim. Where the line is drawn between uncertain, intra-quarter speculation, on the one hand (which need not be disclosed), and "hard," predictive intra-quarter data on the other hand (which does have to be disclosed), remains somewhat unclear and will have to be developed in future cases. Once companies go public, however, the rules are much clearer. When the allegations amount to no more than accurate historical reporting and measured commentary on the company's current operations (such as in Summa Four), the courts will remain willing to dismiss the case early.
