In recent years, Stanford Law School Professor Michael Klausner has led research on several critical issues involved with class action securiteis litigation and SEC enforcement actions.In the guest post below, Professor Klausner and his colleague Jason Hegland describe the two databases they have built in support of their research efforts and detail some additional findings their research has produced. The authors also invite comments and inquirites regarding their research tools and their analysis. I would like to thank Professor Klausner and Jason Hegland for their willingness to publish their article on this site. Here is their guest blog post:
In two recent articles in the PLUS Journal, we presented some simple statistics from the database we have been building over the past few years. One article is on the extent to which D&O insurance proceeds are paid into settlements. The other is on the timing of settlements and dismissals. The two articles can be found on the PLUS website through their search bar or here and here. Our plan is to follow up with a series of more detailed and sophisticated statistical analysis of both securities class actions and SEC enforcement actions. In this post, we summarize some additional findings and invite comments from those of you who are involved in litigating, underwriting, brokering, and paying claims. We begin by describing two databases that we have built and which we intend to maintain going forward. One covers securities class actions and the other covers SEC enforcement actions.
The Securities Class Action Database
This database covers all securities class actions since 2000 and includes over 50 categories of data. The basic data items include, for example:
· Names and positions of all individual defendants
· Names of third party defendants
· Names of lead plaintiffs and lead counsel
· Comprehensive case history, including details of all complaints filed, motions to dismiss and for motions summary judgment.
· Settlements, including individual payments, third party payments, and attorneys’ fees
· Courts and judges
In addition, we have some unique data points:
· The amount that insurers paid into settlements
· The stage of the proceedings at which a case settled (e.g. before the first motion to dismiss was ruled on, after the first motion was ruled on but before a later motion was finally denied)
· The nature of the misstatement (e.g. financial or nonfinancial misstatement, restatement involved)
· Evidence of scienter alleged (e.g. insider sales, confidential witnesses, parallel regulatory actions)
The SEC Enforcement Action Database
Our database on SEC enforcement actions covers civil and administrative actions involving public company misstatements filed from 2000 forward. The following are some examples of the data we have collected:
· Names of all individual, corporate and third party defendants
· Specific violations alleged (e.g. Sections 10(b), 17, 13(b)(5)).
· Allegations regarding evidence of scienter
· Whether cases were tried, settled or dismissed
· Case history
· Timing of settlement (e.g. upon filing, following motion for summary judgment)
· Penalties imposed on each defendant in settlement or following trial
· Courts and judges
Some Findings Regarding Dismissals
Dismissal rates range from 31% to 59% in any given year. Among cases filed in the years 2000 through 2005, 39% were dismissed, and among cases filed in the years 2006 through 2010, 45% of cases were dismissed. This difference, however, is not statistically significant.
Most cases (68%) were resolved on the basis of a single consolidated complaint. Thirty percent of cases were either dismissed with prejudice on the basis of the first consolidated complaint, or dismissed without prejudice and not refiled. Nine percent of cases were voluntarily dropped before the motion to dismiss was made. So a total of 37% of cases were dismissed or dropped without a second consolidated complaint being filed. Another 31% of cases were settled either before the ruling on the first motion to dismiss or after the ruling but before a second complaint was filed.
Courts vary somewhat with respect to dismissals, but since the nature of cases varies geographically, it is difficult to separate the court from the nature of the case. Figure 1 below shows the distribution of final dismissals (with prejudice) across first, second and third consolidated complaints for the four high-volume courts and for the remainder of courts as a group. Of the courts with a relatively high volume of securities class actions, the Southern District of New York dismisses cases disproportionately on the basis of the first consolidated complaint, and it rarely allows a plaintiff to file a third consolidated complaint. The other three high-volume courts more frequently give plaintiffs more opportunities to file second or third complaints. In the remainder of the country, courts are similar to the SDNY.
Some Findings Regarding the Timing of Class Action Settlement
Because the timing of settlements has a substantial impact on the cost of litigation, we collected detailed data on when settlements occur—at what stage of the litigation process and after how much time has passed. To simplify the presentation here, we divide the litigation process into three phases:
· Early Pleading: Anytime prior to a ruling on the first motion to dismiss.
· Late Pleading: After the first motion to dismiss has been granted without prejudice but before a later motion has been denied. This is the period during which a plaintiff is filing a second or later consolidated complaint.
· Discovery: Anytime after a motion to dismiss has been denied and a case heads into discovery. This phase includes cases settled soon after the motion has been denied and cases that settle on the eve of trial or even pending appeal.
Roughly half of all settlements occur in one of the two pleading stages—that is, before a court has finally denied a motion to dismiss and allowed a case to proceed to discovery. The other half of settlements occur after the motion has been denied and a case heads toward discovery. Among those cases, on average, settlement occurs 16 months after the motion was denied.
As shown in Figure 2, there is a slight difference in settlement timing across courts, with cases in the Central District of California settling in the Early Pleading Phase more often than cases in other court, and cases in the Northern District of California settling in the Discovery Phase more often than in other courts. These differences do not seem to be related to the patterns we see above with respect to differences in dismissals across courts.
D&O insurer payments into settlements also vary with settlement timing. This is shown in Figure 3, below. The lowest percentage paid is for settlements at the Early Pleading Stage, which is probably attributable to retentions. Earlier settlements tend to be lower than later settlements (though not across the board), and of course litigation expenses are lower for cases that settle early. An interesting finding, however, is that across settlements in all phases, insurer contributions to settlements were higher among cases filed in the second half of the past decade than in the first half. Figure 3 shows insurer contributions across the three phases for cases filed between 2000 and 2005 and cases filed from 2006 to 2010. (Many cases filed after 2010 are still ongoing, so we omit those).
Individual Liability in Class Actions and SEC Enforcement Actions
Individual liability is obviously a concern of directors and officers. In earlier research, one of use showed that outside directors’ risk of liability is extremely low, in class actions or in any other type of suit. Officers make made out-of-pocket contributions to class action settlements more often. In cases filed from 2006 to 2010, officers made payments into settlements in 2% of settlements (less than 1% of all cases filed). It is difficult to say whether these were cases where evidence of misconduct was strong, but to get some perspective on whether individual liability is a danger in the absence of strong evidence, we look at officers’ payments in class actions with parallel SEC actions—cases presumably with relatively strong evidence of misconduct. As shown in Figure 4, among the 60 pairs of resolved cases filed between 2006 and 2010 in which the SEC imposed a serious penalty, there were only 5 class actions in which the officers made an out-of-pocket payment. This suggests that even when the merits are relatively strong in class actions, the likelihood of a personal payment is low.
In class actions, the mean and median individual payments are $11.7 million and $600,000, respectively. In SEC actions, penalties against individuals involved in the 60 cases reflected in Figure 4 are shown in Figure 5, below.
Note: Frequencies refer to cases, which can involve multiple defendants. Means and medians refer to penalties per person.
Finally, we thought data on the duration of SEC actions, once filed, would be of interest. The mean and median duration of a case against an individual officer (as opposed to cases against the company, which settle more quickly) are 14 months and 8 months, respectively. This, however, includes a large number of cases that are settled simultaneously with filing or within a month of filing. Figure 6 provides a more detailed breakdown. Note, however, that these figures do not include the length of time that the SEC spends investigating a case.
We have provided here a relatively small sampling of the sorts of basic information we can extract from our database. We would welcome comments, questions, and explanations on what we have described here as well as questions that D&O Diary readers think would be interesting to address with the data we are collecting. We would be happy to follow up with additional blog posts.