Securities class actions are interesting for many reasons. They involve large stakes, and so they also attract outsized personalities. They are also more strictly regulated than other class actions. And, as a result, they often lead to unusual tactics for class action litigation. For example, as Professor Mark Gideon points out in his article Recanting Confidential Witnesses in Securities Litigation, 45 Loy. U. Chi. L.J. 575 (2014), because of the heightened pleading requirements they face, securities plaintiffs rely increasingly on confidential witnesses to prop up their complaints, particularly when pleading scienter.

Professor Gideon expresses concern about those courts that have either (1) discounted the testimony of confidential witnesses, or (2) allowed defendants discovery into the information alleged confidentially.

Recanting may not occur as often as some observers suggest, but it has occurred on a number of occasions. And some percentage, perhaps a substantial percentage, of this recanting is false. Various solutions to the problem of flawed CWs have been proposed, but some of these proposals are as flawed as the witnesses. Permitting the depositions of CWs prior to the resolution of motions to dismiss is both unwise and contrary to the express provisions of the PSLRA. Likewise, courts should decline to consider declarations by recanting CWs in the context of motions to dismiss.

From my perch having defended securities class actions, the “problem” with confidential witnesses is how they change the tactics involved in class litigation. Any time a defendant is faced with a complaint, it will conduct an internal review to determine how many of the allegations are valid, and how many are false. If those allegations are buttressed by a confidential witness’s testimony, then the defendant will do its best to identify (and, if appropriate, speak to) the witness so it can determine the truth of the allegations it must defend. That will occur regardless of whether there is a discovery stay. And if the witness is willing to recant on the record, the defendant can bring that to the court’s attention through a Rule 11 motion.

Similarly, my intuition is that courts tend not to trust CWs in complaints because an individual with damning information about corporate misconduct has other avenues than just talking to a plaintiff’s investigator; they can, for example, file a qui tam action or go through the SEC’s whistleblower program.

In fact, the presence of a confidential witness may act as a red flag for many defense counsel, alerting them to allegations that cannot be traced to an identifiable source, and therefore may actually be less reliable than other sources.

Which is not to say that plaintiffs have no tools to allege scienter; among other things, the Supreme Court has allowed them to use inferences from investor-relation campaigns to do so.

So what’s the takeaway here? For plaintiffs (and I know there are some who read this blog), be careful how much you rely on a confidential witness. And for defendants, if you see a confidential witness in a complaint, press further. Those are often the areas that don’t withstand scrutiny.