Home About RSS

Data Visualization Samples

Hello again! While we’ve been quiet on the blog lately, we’ve been anything but in the offices of Stanford Securities Litigation Analytics. We are just a couple weeks away from a beta launch of our website. Even cooler, we’ve been working closely with Professor David Engstrom on his class certification case appearing before the United States Supreme Court.

However, today I’m here to present some cool (we think) data visualization ideas and solicit feedback from our readers. For fellow geeks in the audience, our developers will be working within the D3.js and NVD3.org libraries of data visualizations.  Please note, these graphics DO NOT feature any real data of any real kind, but are simple drawings with completely made up data and case references. PLEASE don’t reference these charts for any actual insights into securities litigation or details of any specific case.  So, with no further ado –

Settlement Distribution Chart

Distribution of Settlements Chart

The idea behind this graphic is to allow users to search for a sample of cases (based on any number of criteria), then graph the distribution of settlements within that sample.  The background of the chart help to delineate settlement sizes by quartile.  In addition to the quartile ranges, a line is drawn to show where the mean settlement value falls within the distribution of cases.  Each bar represents a case within the sample, and rolling over any bar provides a snapshot of information about that one specific case.  Clicking on the bar will bring a user to that particular case’s page where more detailed information about the case can be discovered.


Relationship “Force Direction” Chart

Relationship network

This chart is my personal favorite.  Not only does it look really cool, it quickly conveys an enormous amount of information.  In this example here, we compare the relationships between 4 groups – judges, settlement mediators, lead counsel and defense counsel.  By hovering a cursor over a single “node”, say the mediator Leo Papas, all of his connections are immediately highlighted.  A user can see that Mr. Papas has mediated cases before 3 different judges and worked with 3 different defense and 3 different plaintiffs firms in the past.  By actually clicking on his node, a list of cases will pop up on the right side of the screen that shows which other entities connect with Mr. Papas in a given case.  Clicking on one of the cases listed will direct the user to that case page.


Litigation Proceedings Chart

Litigation Proceedings Chart

This chart shows the number of cases that reach into later and later phases of litigation, with all cases starting at an initial complaint being filed, and the numbers growing smaller and smaller as a case progresses.  Each bar represents an important moment in the litigation process, such as a motion to dismiss being filed or a ruling on a motion for summary judgment.  Hovering the cursor over a given bar will pop up breakdown of case status to that point in the litigation proceeding, as well as an average settlement value for cases settling in that range.


Settlement Distribution Box Plots

Box Plots Chart

Box plots are another great way to visualize the distribution of settlement sizes.  Unlike the first graph, the box plot can show those distributions across an additional axis of data.  In this example, distributions of settlements are compared across phases of litigation.  Other comparisons may include allegation type, parallel SEC vs. no parallel SEC, IPO / Secondary Offering / Non-IPO cases, etc.  To address issues with scale, outlier settlements would be omitted, but a link would be provided at the bottom of each plot to display those omitted outliers.  Each portion of the box and whisker can be highlighted to show aggregate data for the cases falling within that quartile, as well as a list of those cases appearing on the right side of the screen.

So those are our current ideas.  Since each chart requires some amount of web developer time to code and deploy within our web framework, we welcome an feedback so we can target the most useful or most desired charts first.  Additionally, if there are any tweaks you can suggest, those comments would be most appreciated!

Stanford Securities Litigation Analytics data to feature in an upcoming US Supreme Court Case

We are extremely proud to announce that Stanford Securities Litigation Analytics provided data used in a brief written by David Engstrom (faculty, Stanford Law School) for the US Supreme Court on class certification, which has now been selected to be heard by the USSC. Below is an excerpt from the Argument Summary of the brief. If you would like to read the entire brief, please send us an email and we will provide a copy.

At issue in this case is a procedural tool of vital importance to class action practice under Federal Rule of Civil Procedure 23: American Pipe’s rule that the filing of a class action complaint protects potential class members from the running of limitations periods. In American Pipe, this Court wisely rationalized class action law and policy under Rule 23, promoting judicial economy in the adjudication of class action claims while protecting the rights of putative class members. Because it would serve both of these crucial ends, American Pipe’s protective rule should be construed to apply to the three-year limitations period in § 13 of the Securities Act.

American Pipe variously described its protective rule as “commencing” the action on behalf of all potential class members; as “suspending” the limitations period on their behalf; and as “tolling” that limitations period. Regardless of which of these labels attaches, it is plain what this Court did in American Pipe and why it did it. For Rule 23 to operate as designed, potential class members must be protected in their ability to make meaningful decisions about how to pursue their rights, including whether to intervene or file an independent action if class certification is denied. If absent class members did not enjoy protection under American Pipe, they would be compelled to take protective action, either intervening or filing independent lawsuits, in order to avoid being subsequently time-barred. The result would be wasteful and burdensome protective filings, a significant drain on federal court resources, and, most important of all, curtailment of rights.

As petitioner has ably explained, the Second Circuit’s decision cannot be reconciled with either the Court’s decision in American Pipe or the text and structure of § 13. There is no relevant semantic difference between § 13’s one- and three-year limitations provisions. See Pet. at 28-30. And given that the American Pipe rule is thoroughly entwined with Rule 23, as evidenced by the Court’s careful analysis of Rule 23’s structure and purpose in American Pipe itself, its application to § 13’s three-year limitations period cannot possibly be a judicial exercise of equitable discretion. See Pet. at 25-27. Amici endorse each of these plain textual and doctrinal grounds for reversing the Second Circuit’s decision below.

Yet the Second Circuit also ignored on-the-ground litigation realities and, in particular, the efficiency costs of its refusal to apply American Pipe to § 13’s three-year limitations period. Empirical evidence suggests these costs will be large – with estimates reported below showing that the Second Circuit’s position could yield protective filings in well more than half of securities class actions that reach a court order on class certification and at least one-quarter of all filed securities class actions. The Second Circuit’s decision, if allowed to stand, will thus undermine “a principal purpose” of the American Pipe rule: to promote the “efficiency and economy of litigation.” Chardon v. Fumero Soto, 462 U.S. 650, 659 (1983) (quoting American Pipe, 414 U.S. at 553).

The Second Circuit likewise ignored the disruptive and distortive effect of its decision limiting American Pipe’s reach on class action practice under the securities laws, particularly the suppression of Rule 23’s vitally important opt-out mechanism. That threat is even deeper because the Second Circuit’s decision will create perverse incentives for litigants to delay pre-trial proceedings for as long as possible in order to extinguish the rights of potential class members who might seek to go it alone. In short, the Second Circuit’s refusal to apply American Pipe’s protective rule to § 13’s three-year limitations period achieves the worst of all worlds: a flurry of wasteful filings in district courts across the country by sophisticated investors with the capacity to know about and monitor the litigation, and lost rights for everyone else.

Finally, the Second Circuit did not just ignore the policy implications of its decision. It also ignored the Court’s instructions in American Pipe itself when it held, in the alternative and with no analysis beyond its prior conclusion that § 13’s three-year limitations provision creates a “substantive” right, that applying American Pipe’s rule would violate the Rules Enabling Act. This is surprising, for the American Pipe Court rejected just such an argument and then specifically instructed that the validity of applying American Pipe cannot be assessed via a narrow- gauge focus on the limitations provision at issue or whether that provision embodies a “substantive” or “procedural” policy choice. Instead, the validity of applying American Pipe, the Court directed, turns on whether its application would be “consonant with the legislative scheme.” 414 U.S. at 558. The Second Circuit’s failure to heed this instruction – or, indeed, to engage in any recognizable inquiry at all in finding an Enabling Act violation – underscores the pressing need for this Court’s guidance. In fact, had the court asked whether applying American Pipe’s protective rule would be “consonant” with Congress’s securities litigation scheme, it could not have held the way it did.

For all of these reasons, the Court should act, and act now, to resolve the urgent question of American Pipe’s reach.

Bankrutpcies in Class Actions

Happy New Year everyone!  Apologies for the long silence on our blog.  It’s been a busy holiday season perfecting our new database, writing papers, beginning work on creating a public-facing website with user-available analytic tools among a myriad of smaller projects.  We (somewhat) recently wrote a guest piece on Kevin LaCroix’s fantastic D&O Diary site looking at class actions when issuers have filed for bankruptcy.  An excerpt of that post is available below, and the full article can be read here:

“One out of every seven securities class actions filed since 2000 involves a company in bankruptcy. This important subset of class actions has some important features that warrant empirical examination. In this blog, we use our database of securities class actions filed from 2000 to the present to shed light on how cases involving bankrupt companies differ from cases against solvent corporations.[i] Specifically, we address the following questions.

  • Are cases involving bankrupt corporations more successful than cases against solvent corporations?
  • Is the timing of settlement affected?
  • How protective is D&O insurance for officers and directors when a company is bankrupt? Specifically, how much does D&O insurance pay out and how frequently do individual officers or directors make personal, out-of-pocket payments into a settlement?
  • Is there a basis for inferring that additional Side A coverage would have provided more protection for those individuals who paid into settlements?”

Major strides in our technology development

It’s been a while since the SSLA team has written on our blog, but much has been happening behind the scenes here.  We’re very excited to have moved into a production environment for coding private shareholder cases in our new, Drupal-based database.  Our previous database became too limiting in function and flexibility as our project’s scope continued to grow, and we spent quite a while experimenting with alternative platforms before finally settling on Drupal.  The biggest change is our new event-driven system, which enables us to track and update cases in real-time.  Each event in a case triggers a stream of data collection with backward and forward looking dependencies on other data.  Sound confusing?  It is, but suffice to say that this will give us the clearest, and most concise picture of what’s going on in a given case, as well as within the general securities litigation environment.

While development continues in our new SEC and DOJ database, we’ve also begun the work of UI / UX design for the public facing website.  Check out some of our mockups!

Our goal for the public facing site is to create a tool that will allow users to query our database using a near-limitless combination of criteria, with the results generating aggregate analytics and sophisticated, interactive data visualizations.  For example, if a user wanted to see a “baseball card” of statistics for cases involving biotech companies with market caps over $500m facing a restatement-related class action, they might see a results page like this:

Of course, while all this IT development is going on, our team of research assistants is busy collecting more and more data on class actions and SEC actions.  We’ve demonstrated early success with experiments in machine learning and natural language processing for data capture and are exploring ways to utilize this technology to expand the scope of our research into other areas, such as FCPA-related SEC enforcements.

Speaking of SEC actions, in light of recent stories of judicial activism in holding up approval of SEC settlements, Michael Klausner and Jason Hegland are working on a guest blog post for the Harvard Law School Forum on Corporate Governance and Financial Regulation (a copy of the blog will be posted here, as well).  In it, we will explore SEC targeting of senior level executives and their subordinates, as well as provide a more detailed look at SEC penalties in these cases.

Stay tuned for more updates!

The Cost of Discovery and Summary Judgment Motions

In our recent articles published in the PLUS Journal, available through the PLUS website or here and here, we tackle the broad question of settlement timing in class actions and its relationship to settlement amounts.  As can be expected, settlement amounts increase as litigation progresses into later stages, which we’ve defined as:

  • 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.

Here we take a deeper look into settlements occurring within the last  phase, Discovery:

As we can see from the graph, cases that settle at some point during the discovery process but before any Motions for Summary Judgment are filed will settle for approximately $42 million, on average.  That amount rises to $63 million once the Motion for Summary Judgment is filed and defendants decide it’s better to settle rather than risk an adverse ruling on the motion.  For a case to proceed this deep into litigation, the judge has already weighed in on its merits.  As such, it’s not surprising that very few Motions for Summary Judgment are ruled on in the defendants’ favor.  Most motions for summary judgment are denied. Out of 80 motions ruled on in cases filed between 2000-2012, 56 were granted and 84 were denied.Once a summary judgment motion is denied, the price of settlement nearly doubles to $120 million, on average.  That amount is more than four times the average settlement size of a case settling prior to the judge’s ruling on the first Motion to Dismiss ($23.3 million).

Please share any comments or questions you may have below, and if you have a suggestion of blog topic, let us know!

Audit Committee Targeting in Class Actions.

In my last post I addressed the decline in securities class action filings over the decade from 2001-10.  Utilizing SLA’s database I highlighted the concurrent decrease in restatement-related cases and postulated that Sarbanes-Oxley regulation played a role in modulating corporate restatement filings.  As the nature of cases changed during this period so did plaintiff targeting of Audit Committee defendants.  Targeting of other defendant types such as the CEO or Board Chairman remained relatively constant, though CFOs were named about 10% less often in the 2009-10 period.

Restatements rose following SOX and then declined.  Class actions related to restatements followed this trend.  Audit committee members tend to be targeted in cases involving restatements more than in other cases, including cases involving alleged financial misstatements where there is no restatement.  Consequently, Audit Committee members’ involvement in securities class actions generally followed the same pattern.  In fact, Audit Committee members were named disproportionately more often in restatement-related cases following SOX, so the rise and decline in cases naming them was even more pronounced than the rise and decline of restatement-related cases.

The Stanford SLA team encourages you to become a registered user.  We welcome your questions and comments!

New Securities Class Action Filings are Declining.

In his April 30th D&O Diary post, Kevin LaCroix highlights the recent decline in Securities Class Action filings.  He cites two recent publications, a PwC study and an Advisen report, which reference that decline most notably during the second half of 2012 through the first quarter of 2013.  After crunching some numbers, we view the recent reports as part of a longer running trend in securities litigation.

Utilizing our comprehensive database of Class Actions filed against publicly traded companies for disclosure related misstatements, we see that new filings drop from about 180 per year to about 130 per year in the decade spanning 2001 to 2010.  Roughly paralleling that trend is a concurrent reduction in restatement related Class Action cases.  Restatement related cases accounted for 30-40% of filings in the first half of the decade before falling precipitously to about 12% later in the period.  Non-restatement accounting-related cases have remained constant as a percentage of filings while non-accounting and merger-related cases trend upward in both absolute numbers and as a percentage of filings.

One possible explanation for the decline in restatement related class actions lies in the landscape sculpted by the Sarbanes-Oxley Act of 2002 (SOX).  Restatements rose following SOX and then declined.  Class actions related to restatements followed this trend while the filing of new Securities Class Actions decreased overall.  Since suits filed against companies that did not file a restatement are more likely to be dismissed it follows that fewer Securities Class Actions ultimately reach a settlement.

The Stanford SLA team encourages you to become a registered user.  We welcome your questions and comments!

When Class Actions Settle, Who Pays?

In its May 2013 issue, the PLUS Journal will publish SLA’s second of two articles exploring the resolution of securities class action litigation.  The article, How Protective Is D&O Insurance in Securities Class Actions?-An Update, explores the extent to which companies, directors, officers, and D&O insurers contribute to class action settlements.  Utilizing its extensive database, the SLA team analyzed class actions filed from 2006-2010, and settled by the end of 2012.  We believe our analysis may benefit parties that litigate these cases, negotiate coverage with D&O carriers, and others.

Our analysis reveals that, on the whole, D&O Liability Policies are highly protective of both companies and their directors and officers.  In fact, D&O insurers contributed funds in 85% of settlements for which we found data on such payments. Officers paid out of pocket in 2% of cases despite receiving harsh penalties in parallel cases brought by the SEC 92% of the time.

Does settlement size or the severity of a misstatement correlate with the protection afforded by a D&O Policy? How does the nature of an alleged misstatement correlate with the targeting of particular officers or directors, for example, the CFO or audit committee member?  For the answers to these questions and further analysis we encourage you to read the article upon publication in the PLUS Journal.  Also check our site regularly as we highlight subjects trending in the world of securities litigation.

The Stanford SLA team encourages you to become a registered user.  We welcome your questions and comments!

When are Class Actions Dismissed and When do they Settle?

In the April 2013 issue of the PLUS Journal, we at Stanford Securities Litigation Analytics (SLA) published part one of two articles, When Are Securities Class Actions Dismissed, When Do They Settle, and For How Much?-An Update, exploring the timing and correlates of dismissals and settlements in recently filed securities class action cases.  In his blog, The D&O Diary, Kevin LaCroix posted his synopsis of the article.  Utilizing its extensive database, the SLA team analyzed class actions filed from 2006-2010, and resolved by the end of 2012.  We believe our analysis may benefit parties that litigate these cases, negotiate coverage with D&O carriers, and others.

Within the 652 case sample, 40% settled, 32% were dismissed with prejudice, 11% were voluntarily dropped, and 18% are still ongoing.  Despite the paradigm that frames class actions as lengthy and litigious, most cases reached a resolution relatively quickly. In fact, 58% of class actions in the study were dismissed, dropped, abandoned, or settled before the filing of a Second Consolidated Complaint.  Among cases that settled, 43% did so before a Motion to Dismiss was denied and the case moved toward discovery.

You might ask yourself: what other factors correlate with the timing of dismissals, or the size of settlements?  Does the nature of an alleged misstatement affect dismissal rates? Does the size of the settlement correlate with settlement timing or class recovery of losses?  We recommend you follow the links to our article and Mr. LaCroix’s blog post for additional analysis.  Also be on the lookout for part 2 of the PLUS article series, which explores the extent to which companies, directors, officers, and D&O insurers contribute to class action settlements.

The Stanford SLA team encourages you to become a registered user.  We welcome your questions and comments!

In defense of economics

Three years ago, Harvard economists Ken Rogoff and Carmen Reinhart published a paper that became  one of the most talked about studies in the media, politics and field of economics.  However, just this past week a University of Maryland economics grad student discovered errors in the mathematical formulas used, as well as underlying problems within the data itself.  This prompted a recent podcast by NPR’s Planet Money titled “How Much Should We Trust Economists?” published on April 19, 2013.

Justin Wolfers, an economist at the University of Michigan, states that in his analysis of economics papers regarding the death penalty, he found errors of various kinds in 90% of published studies anywhere between one month and 10 years after their publication.  In some cases, errors were in methodologies, and others the errors occurred within the underlying math.

It’s understandable why such common errors could create a of mistrust of economics.  Macro economists, as Planet Money points out, suffer a data problem.  We’re unable to run controlled experiments where specific variables are easily isolated and explained.  The study of securities litigation is especially susceptible to this problem.  The data itself is extremely difficult to reliably find, and even more difficult to accurately collect.  These issues are compounded by the relatively small universe of cases relative to other fields of study – small errors can have a big impact on the accuracy of any analysis.  We’re acutely aware of this problem at SLA, and take painstaking steps to minimize the risks and impacts of errors when counseling practitioners in our industry; we’re confident that our dataset is the most carefully collected one of its kind.

However, quality data collection is only half the battle of for quality analytics.  The other major hurdle that economists face is the challenge of teasing out, and measuring the influence of single variables against the entire dataset.  For instance, we find that class actions with a parallel SEC action are highly correlated to cases in which a restatement occurs.  Likewise, our data shows that cases with the largest damages will settle for the most amount of money.  All of this is obvious on the surface, but presents a challenge in understanding the true dynamic among various factors in a class action and its ultimate outcome.  We are continuing to refine our models every day to provide the most useful analytics to practitioners.  If you are interested in helping in our development, please send us a note.

As Mr. Wolfers points out, “The best defense of economics is not that we’re good, it’s that everything else is terrible.”