Bruce Ericson
Stacie Kinser

One of the most important ways a company can try to avoid potential liability under the federal securities laws is to incorporate precautionary disclosure in its public statements and regulatory filings. However, in a June 23, 2015 decision in In re Harman International Industries Securities Litigation (here), the D.C. Circuit provided a reminder to companies on the importance of keeping their precautionary disclosures up-to-date.


In the following guest post, Bruce A. Ericson and Stacie Kinser of the Pillsbury Winthrop Shaw Pittman LLP law firm take a detailed look at the D.C. Circuit’s recent opinion and consider the decision’s practical implications for companies’ precautionary disclosures. Ericson is a partner and Kinser is an associate at the Pillsbury law firm. Ericson is also Managing Partner of Pillsbury’s San Francisco Office, and Co-Head of Pillsbury’s Securities Litigation and Enforcement Team. A version of this article previously was published as a Pillsbury client alert and on Law 360.


I would like to thank Bruce and Stacie for their willingness to publish their article as a guest post on my site. I welcome guest post submissions from responsible authors on topics of interest to this site’s readers. Please contact me directly if you would like to submit a guest post. Here is Bruce and Stacie’s guest post.




SEC Rule 10b-5 makes it unlawful to misstate a material fact (or omit to say something if the omission would render misleading what you do say) in connection with the purchase or sale of a security. The Private Securities Litigation Reform Act (PSLRA) created a safe harbor for statements that are forward-looking and accompanied by meaningful cautionary language. In a recent decision, the D.C. Circuit revisited the standard for forward-looking statements, and placed special emphasis on the accompanying cautionary language, holding that statements which fail to account for historical facts cannot be meaningful. The opinion should serve as a timely reminder for companies to review and update their cautionary language.

The Harman Case

In re Harman Int’l Indus., Inc. Sec. Litig.[1] is a 10b-5 action against the successor to Harman Kardon, the iconic maker of home and automobile audio equipment. The case focuses on Harman’s statements about inventories, margins and sales of Harman’s personal navigation devices (PNDs) made at a time in 2007 when Harman was considering a venture capital-backed going-private transaction.

Harman announced the possible going-private deal in April 2007. The same day, Harman’s CEO predicted a growth in PND sales in Europe. He noted that Harman’s inventory of PNDs in Europe had grown substantially, but said Harman planned to reduce the inventory significantly by its year’s end (June 30). Harman’s stock price rose markedly in response to these announcements and held steady until September 2007, when Harman announced that the going-private transaction had been abandoned. The stock price then fell 24%. It fell again—this time by 40%—in January 2008, when Harman lowered its EPS forecast, blaming poor PND sales. And it fell again—this time by 15%—in February 2008, when Harman announced its results for Q2 of FY2008, noting PND sales had fallen by $29 million compared to the same period in FY 2007, in part due to the sale out of inventory of older products at substantial discounts.

The Challenged Statements—and the District Court’s Ruling

Shareholders sued Harman and three of its officers, alleging that they had violated Rule 10b-5 on three occasions between April and September 2007:

  • In April 2007, when the CEO predicted sales of 618,000 PNDs in Europe by the end of the fiscal year (6/30/07) while noting growing inventories and anticipated margin challenges, plus actual sales of only 300,000 units through the first three quarters of the fiscal year.
  • In August 2007, when Harman filed its 10-K for FY 2007, which said “Sales of aftermarket products, particularly PNDs, were very strong during fiscal 2007.”
  • In September 2007, when the CFO predicted overall Q1 sales would exceed Q1 FY 2007 by 15% and predicted “the growth and expansion of [the PND] business primarily in Europe.”[2]

The complaint alleged that these statements were materially false and misleading because Harman had recently updated its PND design and was stuck with a large inventory of older generation products that it could sell, if at all, only at substantial discounts—a situation exacerbated by increasing competition and shrinking margins in Europe. The district court dismissed the complaint, holding that the April and September forecasts fell within the statutory safe harbor and that the August statement (about “very strong” sales) was “puffery.”

The D.C. Circuit Reverses as to All Three Statements

After reviewing the case law in other circuits,[3] the appeals court turned to the forward-looking statements. The court noted that the cautionary statements made during the April and September conferences were cursory (as is typical), but referred the listener to Harman’s 10-K. Accordingly, the court considered the risk factors set forth there. These mentioned the growth in inventory (stating the dollar amount) and said that sales could suffer “if the Company failed to “develop, introduce and achieve market acceptance of new and enhanced products,” that it had to “maintain and improve existing products, while successfully developing and introducing new products,” and could “experience difficulties that delay or prevent the development, introduction or market acceptance of new or enhanced products,” as well as that competitors could “introduce superior designs or business strategies, impairing [the Company’s] distinctive image and [its] products’ desirability.”[4] These statements, defendants argued, pointed out the risks of obsolescence and increased competition.

The court disagreed, holding that this language was misleading in light of historical facts and the complaint’s allegations that the inventory not merely could become obsolete but already was obsolete, that sales in 2006 had been weak, and that there was no reason to think that obsolete 2006 models would sell any better in 2007. References to amassed inventory did not convey that the inventory already was obsolete. As the court put it, “[t]he cautionary language … is too general and fails to account for the materialization, rather than abstract possibility, of the important risk posed by PND obsolescence.”[5] The court noted with disapproval that Harman continued to use the same cautionary language in the 2007 annual report despite continued deterioration in its PND sales after April. Responding to defendants’ argument that they need only identify risks, the court emphatically disagreed: “To conclude … that even where a risk has materialized a company need only warn that it is a ‘risk,’ would render misleading cautionary language sufficient, a result neither the statutory text, nor legislative history, nor precedent supports.”[6] Citing earlier opinions, the court reasoned that the safe harbor does not protect “a person who warned his hiking companion to walk slowly because there might be a ditch ahead when he knows with near certainty that the Grand Canyon lies one foot away.”[7]

Making Your Cautionary Language Meaningful

The PSLRA safe harbor was designed to facilitate dismissal of challenges to forward-looking statements at the pleadings stage, before any discovery. But as Harman shows, even on a motion to dismiss, courts will take a hard look at the cautionary language and dismiss the complaint only if the cautionary language truly is meaningful. That means eschew boilerplate, be specific, don’t misstate historic facts and, above all, update your language as things change.

  • Don’t rely on boilerplate—but do incorporate specific risk factors by reference.

The April 2007 conference call in Harman began with the moderator’s standard warning that forward-looking statements made during the call “are subject to risks and uncertainties.” The court rejected this as boilerplate on which Harman could not rely.[8] But the court did not demand that pages of risk factors be intoned at the start of a conference call; it had no problem with the moderator incorporating by reference Harman’s previous 10-K. The court’s problem, rather, was with what the 10-K said, and didn’t say.

  • Warn of specific, important risks.

To work, cautionary language must be specific and meaty. This does not mean your crystal ball must be perfect, but you must warn of important factors that could cause actual results to differ from predictions.[9] And tailor your warnings to the predictions you are making. An earnings release with guidance on future quarters probably requires a different set of warnings than does a specific prediction made along with the announcement of a specific event or transaction.[10]

An even newer opinion—Julianello v. K-V Pharm. Co.—illustrates the point.[11] K-V, a pharmaceutical company, obtained “orphan drug” status for a drug (by statute, seven years of exclusive sales rights), then predicted that the FDA would enforce the orphan drug status by barring compounders from offering knock-offs, while cautioning about “the possibility that any period of exclusivity may not be realized.”[12] After the company hiked the price of the drug by 14,900%, the resulting hue and cry caused the FDA to announce that it would not take action against compounders offering equivalent drugs. Investors sued, the district court dismissed the complaint, and the investors appealed. But this time they lost. The Eighth Circuit held that the company’s cautionary language “explicitly identified the risks associated with the FDA’s presumed enforcement of exclusivity”; therefore, the safe harbor applied.[13]

K-V did not caution against precisely what happened—a public outcry, angry senators, and a one-off relaxation of exclusivity by the FDA. But the cautionary language need not identify “the factor that ultimately belies a forward-looking statement.”[14] It is sufficient that the cautionary language warns of risks “of a significance similar to that actually realized” such that investors are on notice of the dangers of investing. Id. As the court in Harman recognized, “perfect clairvoyance may be impossible.”[15]

  • Don’t misstate historical facts.

Harman reminds us that companies must state historical facts accurately. A lie about the past cannot be saved by burying it in a prediction about the future. And if something is an actuality, calling it a “risk” does not suffice: don’t talk of future risks of limited magnitude if the harm has already begun to materialize.[16]

  • Update your cautionary language.

Above all, review and update your cautionary language regularly. Regular attention to cautionary language will help prevent it from ossifying into boilerplate. It also will provide an opportunity to warn of new, specific and important risks, and to note any change in facts and circumstances that may have occurred since your last SEC filing. This is no place to put one’s mind on autopilot: it is dangerous simply to incorporate by reference or cut-and-paste cautionary language that may have gone stale.

[1] ___ F.3d ___, 2015 WL 3852089 (D.C. Cir. June 23, 2015).

[2] Harman, 2015 WL 3852089, at *3-*6.

[3] Harman, 2015 WL 3852089, at *8-*10.

[4] Harman, 2015 WL 3852089, at *10.

[5] Harman, 2015 WL 3852089, at *11-*13.

[6] Harman, 2015 WL 3852089, at *14.

[7] Harman, 2015 WL 3852089, at *9. The court also rejected defendants’ argument that the August statement in the 10-K about “very strong sales” was just puffery: “Unlike the statements in cases on which the Company relies, the statement was tied to a product and a time period and it was not too vague to be material.” Id. at *16.

[8] It has long been the rule that “boilerplate warnings won’t do” as meaningful cautionary language. Asher v. Baxter Int’l Inc., 377 F.3d 727, 732 (7th Cir. 2004); Southland Sec. Corp. v. INSpire Ins. Solutions, Inc., 365 F.3d 353, 372 (5th Cir. 2004) (“merely a boilerplate litany of generally applicable risk factors” does not meet the requirement for meaningful cautions).

[9] Harman, 2015 WL 3852089, at *10.

[10] That said, you probably also will want to incorporate by reference the risk factors in your most recent 10-K and 10-Q—at least if they are still valid.

[11] ___ F.3d ___, 2015 WL 4032102 (8th Cir. July 2, 2015).

[12] Julianello, 2015 WL 4032102, at *2.

[13] Julianello, 2015 WL 4032102, at *6.

[14] Harris v. Ivax Corp., 182 F.3d 799, 807 (11th Cir. 1999) (emphasis in original).

[15] Harman, 2015 WL at 3852089, at *10.

[16] Harman, 2015 WL at 3852089, at *9, *12.