Opinion
Civil Action No. 02–12146–NMG.
2012-05-17
BRICKLAYERS AND TROWEL TRADES INTERNATIONAL PENSION FUND, Goodman Family Trust, James Uphoff, Malka Birnbaum, Neil McCarty, and Rodney Narbesky, Consolidated Plaintiffs, v. CREDIT SUISSE FIRST BOSTON, Jamie Kiggen, Frank Quattrone, Laura Martin, and Elliot Rogers, Defendants.
Daniel S. Sommers, Steven J. Toll, Cohen, Milstein, Hausfeld & Toll, Washington, DC, Donald R. Hall, Frederic S. Fox, Joel B. Strauss, Melinda D. Campbell, Gregory Michael Egleston, Egleston Law Firm, Menachem E. Lifshitz, New York, NY, Thomas G. Shapiro, Todd S. Heyman, Adam M. Stewart, Edward F. Haber, Boston, MA, for Plaintiffs. Avi S. Gesser, Dharma B. Frederick, Jennifer Newstead, Lawrence Portnoy, Daniel J. Schwartz, Jonathan K. Chang, Lawrence Portnoy, Davis, Polk & Wardwell, Henry Putzell, III, Law Office of Henry Putzell, III, Michael W. Mitchell, Jeff E. Butler, Warren L. Feldman, Alejandra De Urioste, New York, NY, Robert A. Buhlman, Siobhan E. Mee, Jason D. Frank, Sarah G. Kim, Jeffrey B. Rudman, Stephen A. Jonas, Adam Hornstine, Timothy J. Perla, Wilmer, Cutler, Pickering, Hale and Dorr, LLP, Jonathan A. Shapiro, Eben P. Colby, James R. Carroll, Skadden, Arps, Slate, Meagher & Flom, Boston, MA, Barbara A. Winters, Kenneth G. Hausman, Mark A. Sheft, Howard, Rice, Nemerovski, Canady, Falk Rabkin, San Francisco, CA, for Defendants.
Daniel S. Sommers, Steven J. Toll, Cohen, Milstein, Hausfeld & Toll, Washington, DC, Donald R. Hall, Frederic S. Fox, Joel B. Strauss, Melinda D. Campbell, Gregory Michael Egleston, Egleston Law Firm, Menachem E. Lifshitz, New York, NY, Thomas G. Shapiro, Todd S. Heyman, Adam M. Stewart, Edward F. Haber, Boston, MA, for Plaintiffs. Avi S. Gesser, Dharma B. Frederick, Jennifer Newstead, Lawrence Portnoy, Daniel J. Schwartz, Jonathan K. Chang, Lawrence Portnoy, Davis, Polk & Wardwell, Henry Putzell, III, Law Office of Henry Putzell, III, Michael W. Mitchell, Jeff E. Butler, Warren L. Feldman, Alejandra De Urioste, New York, NY, Robert A. Buhlman, Siobhan E. Mee, Jason D. Frank, Sarah G. Kim, Jeffrey B. Rudman, Stephen A. Jonas, Adam Hornstine, Timothy J. Perla, Wilmer, Cutler, Pickering, Hale and Dorr, LLP, Jonathan A. Shapiro, Eben P. Colby, James R. Carroll, Skadden, Arps, Slate, Meagher & Flom, Boston, MA, Barbara A. Winters, Kenneth G. Hausman, Mark A. Sheft, Howard, Rice, Nemerovski, Canady, Falk Rabkin, San Francisco, CA, for Defendants.
MEMORANDUM & ORDER
GORTON, District Judge.
This case is a consolidated securities class action in which the court-appointed lead plaintiff, the Bricklayers and Trowel Trades International Pension Fund, asserts claims on behalf of the class of individuals (“plaintiffs”) who purchased common stock of AOL–Time Warner, Inc. (“AOL”) from January 12, 2001, through July 24, 2002 (“the Class Period”). The defendants include Credit Suisse First Boston (USA), Inc. (“CSFB–USA”), Credit Suisse First Boston, LLC (“CSFB”), its wholly-owned subsidiary, and four individuals who were employed by CSFB during all or part of the Class Period (collectively, “defendants”). The individual defendants include James Kiggen and Laura Martin, former CSFB research analysts responsible for investment research coverage of AOL during the Class Period. Kiggen and Martin reported to defendants Frank Quattrone, the former Senior Managing Director and Global Head of CSFB's Technology Group, and Elliot Rogers, the former Managing Director and Global Director of Technology Research.
I. The Complaint
The Second Amended Consolidated Class Action Complaint (“the complaint”) asserts two counts: 1) CSFB, Kiggen and Martin made material misstatements and omissions in violation of section 10(b) of the Securities Exchange Act (“Exchange Act”), 15 U.S.C. § 78j(b), and Rule 10b–5(b) promulgated thereunder, 17 C.F.R. § 240.10b–5 (“Count I”), and 2) CSFB–USA, CSFB, Quattrone and Rogers acted as “control persons” in violation of section 20(a) of the Exchange Act, 15 U.S.C. § 78t(a) (“Count II”).
To prove allegations of securities fraud under Rule 10b–5, a plaintiff must establish:
1) that the defendants made a material misrepresentation in connection with the purchase or sale of a security;
2) that the misrepresentation was made with scienter;
3) reliance, i.e., but for the misrepresentation, an investor would not have purchased or sold the security;
4) economic loss, i.e., the investor lost money as the result of said purchase or sale; and
5) loss causation, i.e., a causal connection between the misrepresentation and the economic loss.
In re PolyMedica Corp. Sec. Litig., 432 F.3d 1, 6–7 (1st Cir.2005).
Plaintiffs' theory of recovery in this case is commonly referred to as the fraud-on-the-market scenario. Plaintiffs allege, under that theory, that 1) CSFB's overly optimistic and intentionally misleading reports, upon which class investors relied in purchasing the AOL stock, artificially inflated the stock price and 2) after the market learned of the deception, the price of the stock declined.
II. Procedural history
In 2003, various cases were consolidated to comprise this action. In September, 2006, former United States District Judge Nancy Gertner denied defendants' motions to dismiss and, two years later, certified the class. Most recently, on August 26, 2011, 2011 WL 3813204, Judge Gertner provisionally denied defendants' motions for summary judgment without ruling on defendants' objection to the testimony of plaintiffs' expert Dr. Scott Hakala. On that occasion she stated:
Summary judgment—without a full Daubert hearing—is an inappropriate way to decide [whether Dr. Hakala's event study and testimony will be admitted at trial.]. Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). The expert issues cannot be determined by precedent, pointing out that Plaintiffs' expert was accepted or rejected in this or that case so long as the Defendants are making, in effect, an as-applied challenge on these facts, in this context. It requires more than warring affidavits and strident briefs. It requires an evidentiary hearing. And in the hearing the threshold question is not just the reliability of the expert testimony under the Federal Rules of Evidence 702. It is also whether a jury would fully understand the attacks and counterattacks as they play out in the instant case.
Judge Gertner pointed out that reliance and loss causation, two central elements of plaintiffs' securities fraud case, “necessarily rely on [the] expert testimony” of Dr. Hakala, implying that if the Court were to exclude the testimony of Dr. Hakala after a Daubert hearing, it would necessarily have to revisit its summary judgment decision.
Upon Judge Gertner's retirement from the bench, the case was transferred to this Session. Pending before the Court are defendants' motion to preclude the expert opinions of Scott Hakala, M. Laurentius Marais, Bernard Black and Reinier Kraakman, which plaintiffs oppose, and plaintiffs' motion to preclude the expert opinions of Rene Stulz and John Deighton, which defendants oppose. On December 20, 2011, the Court convened a Daubert hearing at which Dr. Hakala and Dr. Marais testified and counsel were afforded an opportunity to expound at length on their respective motions. The Court took the motions under advisement and, upon further reflection and analysis, renders the following decision.
III. Defendants' motion to preclude the expert opinion of Dr. Scott Hakala
Dr. Hakala prepared an event study to measure the impact, if any, of defendants' allegedly fraudulent statements and omissions on the value of AOL stock during the Class Period.
A. Event studies
A conventional securities fraud event study is conducted as follows: an economist performs a regression to estimate the relationship between a stock's “actual return” (the difference between closing prices on two consecutive days) and the movement of one or more indices representingan average of the stock prices for several companies which make up the market and/or industries in which the firm operates. This first step allows the economist to predict how the stock price should move on any given day based on the movement of the indices (“the expected return”) and thereby provides a benchmark for all companies within a particular market. The estimated expected return is then used as the baseline against which the stock's actual return on pre-selected event days is measured. The expected return is thus a measured expectation of what the normal stock price movement would have been if the event had not occurred. If the difference between the expected return and the actual return on an event day is statistically significant, it may be attributed to the event occurring on that day, provided that the study controls for confounding factors. A. Craig MacKinlay, Event Studies in Economics and Finance. 35 J. Econ. Literature 13, 13–35 (1997); Jay W. Eisenhofer, Geoffrey C. Jarvis & James R. Banko, Securities Fraud, Stock Price Valuation, and Loss Causation: Toward a Corporate Finance–Based Theory of Loss Causation. 59 Bus. Law. 1419, 1425–26 (2004).
An event study is an accepted method of measuring the impact of alleged securities fraud on a stock price and often plays a “pivotal” role in proving loss causation and damages in a securities fraud case. In re Williams Sec. Litig., 496 F.Supp.2d 1195, 1272 (N.D.Okla.2007). Given the difficulty inherent in proving the effect, if any, of a single news item on the price of a stock, many courts require them in such cases. See Fener v. Operating Eng'rs Const. Indus. & Misc. Pension Fund (LOCAL 66), 579 F.3d 401, 409 (5th Cir.2009) (“[T]he testimony of an expert—along with some kind of analytical research or event study—is required to show loss causation.”); In re Imperial Credit Indus., Inc. Sec. Litig., 252 F.Supp.2d 1005, 1015 (C.D.Cal.2003) (“[A] number of courts have rejected or refused to admit into evidence damages reports or testimony by damages experts in securities cases which fail to include event studies or something similar.”) (internal citations and question marks omitted). While event study techniques have become more sophisticated over the years, their basic format and methodology have not materially changed. S.P. Kothari and Jerold B. Warner, Econometrics of event studies. Handbook of Corporate Finance: Empirical Corporate Finance (2004). In other words, there is no great dissension in the financial econometric community about how to conduct a proper event study.
B. Standard
The admission of expert evidence is governed by Federal Rule of Evidence 702, which codified the Supreme Court's holding in Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), and its progeny. United States v. Diaz, 300 F.3d 66, 73 (1st Cir.2002). Rule 702 charges a district court with determining whether: 1) “scientific, technical, or other specialized knowledge will assist the trier of fact,” 2) the expert is qualified “by knowledge, skill, experience, training, or education” to testify on that subject, 3) the expert's proposed testimony is based upon “sufficient facts or data,” 4) that testimony is the product of “reliable principles and methods” and 5) the expert “applies the principles and methods reliably to the facts of the case.” Furthermore, a critical inquiry is whether the expert “employs in the courtroom the same level of intellectual rigor that characterizes the practice of an expert” on securities fraud event studies. Kumho Tire Co. v. Carmichael, 526 U.S. 137, 152, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999).
The Court must be vigilant in exercising its gatekeeper role because of the latitude given to expert witnesses to express their opinions on matters about which they have no firsthand knowledge and because an expert's testimony may be given substantial weight by the jury due to the expert's status. See Daubert, 509 U.S. at 595, 113 S.Ct. 2786;Kumho Tire, 526 U.S. at 148, 119 S.Ct. 1167. This consideration looms large in securities fraud cases, where complex event studies are often the only evidence used to establish reliance and loss causation.
The Court must, nonetheless, keep in mind that
vigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence.
Daubert, 509 U.S. at 596, 113 S.Ct. 2786. If an expert's testimony is within “the range where experts might reasonably differ,” the jury, not the trial court, should be the one to decide among the conflicting views of different experts. Kumho Tire, 526 U.S. at 153, 119 S.Ct. 1167.
C. Analysis
Defendants contend that Dr. Hakala's event study is unreliable because it flouts established event study methodology and draws unreasonable conclusions from the data presented. Specifically, defendants argue that Dr. Hakala 1) studies the wrong days, 2) overuses so-called “dummy variables,” 3) disregards prior disclosures and 4) fails to control for confounding factors. The Court addresses defendants' arguments seriatim.
1. Event day selection
Accepted event study methodology teaches that an economist should exclude from the regression the days the alleged fraudulent misrepresentations were made (“inflationary days”), the days the market learned the truth (“corrective days”) (together, “event days”) and any other days on which major company-specific news, such as a quarterly earnings statement, was reported (“material news days”). By removing those days from consideration, an economist can establish a baseline of normal stock price volatility against which the change in price on event days can be compared. If stock price volatility is abnormally high on event days, and the study properly controls for other industry- and company-specific information released to the marketplace unrelated to defendants' alleged fraud (“confounding factors”), the inflation or deflation may fairly be attributed to the defendants. MacKinlay, supra, at 13–35; Eisenhofer, supra, at 1425–26.
To summarize: an economist should select as the event days to be studied the days the allegedly fraudulent misrepresentations were made and the days the market learned the truth. Dr. Hakala admittedly did not proceed in this fashion. Instead of selecting as event days the 34 days on which, according to the complaint, CSFB issued misleading reports, Dr. Hakala selected only 12 of those days, deeming the remaining 22 irrelevant to his study. His approach, while unconventional, does not necessarily subvert his study. It is, after all, certainly possible that some misleading analyst statements impact the price of a stock while others do not. It does, however, compel this Court to scrutinize his selection of event days.
Of the 57 identified event days, Dr. Hakala labels 21 as inflationary (i.e., days defendants' alleged misrepresentations purportedly inflated the value of AOL stock) and 36 as corrective (i.e., days the value of AOL stock correspondingly deflated when the market purportedly learned the truth). A review of the information released to the market on those days exposes the fallibility of his classifications.
Take, for example, February 1, March 7 and April 2, 2001, days on which CSFB issued reports allegedly misrepresenting AOL's earning potential and omitting material information that, if revealed, likely would have tempered investor expectations. Instead of treating those event days as inflationary, Dr. Hakala inexplicably deems them corrective disclosures. In so doing, his study not only contravenes established event study methodology, it belies the very allegations plaintiffs made in the complaint, i.e., that CSFB reports inflated AOL's stock price.
Nor were these isolated mistakes. Dr. Hakala's event study is replete with event days that appear to have been selected more for their volatility than for their actual relationship to defendants' alleged fraud. June 7, 2001 and February 5 and 26, 2002 are examples of days on which what little AOL news released to the market was positive and, yet, AOL's stock price declined. Instead of concluding that the news did not cause the price deflation and attributing it to any of the myriad other variables that affect stock prices, Dr. Hakala blamed the decline on defendants and labeled the days as corrective even though the news released was positive.
Rather than study the market's reaction to the misrepresentations alleged in the complaint, Dr. Hakala cherry-picked unusually volatile days and made them the focus of his study. If the stock price increased sharply, he attributed it to the defendants (even if no CSFB reports were released on that day). If the stock price decreased sharply, he called it a corrective disclosure (even if the news released was positive). The Court concludes, as did United States District Judge Rya Zobel in In re Xcelera.com Sec. Litig., No. 00–11649–RWZ, 2008 WL 7084626, at *2 (D.Mass. Apr. 25, 2008), that, “[q]uite simply, [Dr. Hakala's] theory does not match the facts.”
2. Dummy variables
A dummy variable is a variable that takes the value of one or zero to indicate the presence or absence of some effect in an event study. Dummy variables allow financial economists to control for the effect of event days and material news days on the price of a stock. They are a common and accepted component of securities fraud event studies, although courts have cautioned against their overuse. See In re Northfield Labs., Inc. Sec. Litig., 267 F.R.D. 536, 548 (N.D.Ill.2010); Xcelera, 2008 WL 7084626, at *1.
Defendants assert that Dr. Hakala's event study uses a much higher percentage of dummy variables than is considered acceptable in the financial econometric community and that such overuse renders his study unreliable. Instead of dummying out the 34 event days on or after which defendants issued analyst reports and any other days significant AOL-related news appeared, as defendants suggest would have been proper, Dr. Hakala dummied out virtually every day any AOL-specific news was reported (211 of the 388 days in the study period, a breathtaking 54%). In so doing, Dr. Hakala assumed for the purpose of his study that AOL stock trades “normally” less than half the time.
Apparently, this is not the first time that Dr. Hakala has been criticized for overusing dummy variables. In the event studies he prepared for the Northfield and Xcelera cases, he dummied out 117 of 1,383 trading days (8%) and 130 of 343 trading days (38%), respectively. The Xcelera court explained that Dr. Hakala's use of dummy variables was not a reliable method supported by the academic literature. 2008 WL 7084626, at *1 (“Although the academicliterature supports the use of dummy variables for events in which significant company-specific news is released, no peer-reviewed journal supports the view that dummy variables may be used on all dates on which any company news appears.”). The Northfield court explained why:
Dr. Hakala's excessive use of dummy variables understates the usual volatility of the stock [to 3.66% from 4.09%], which has the effect of making it appear that news had a greater affect [sic] on price than it actually had.
267 F.R.D. at 548. Both courts excluded Dr. Hakala's event studies after concluding that his overuse of dummy variables rendered them unreliable.
Exclusion is likewise warranted here. In this case, Dr. Hakala used dummy variables for a greater number of days and a higher percentage of the study period (211 days and 54%, respectively) than he did in his event studies in Northfield and Xcelera. If those courts were correct in excluding his event studies for the reasons articulated therein, as this Court believes they were, it follows a fortiori that his event study should be excluded here.
Looking beyond the numbers, Dr. Hakala's use of dummy variables in the event study he prepared for this case artificially deflated the baseline volatility of AOL's stock price during the Class Period, thereby making its volatility on days CSFB issued reports appear greater than it really was. Because the stock price of any company is likely to be more volatile on material news days, removing those days from the study creates a downward bias in the study's measure of price volatility. As a result, Dr. Hakala's study cannot establish that it was the reports of the CSFB analysts, rather than the tangle of factors that affect the price of a stock, that inflated AOL's share price during the Class Period.
3. Prior disclosures
At the class certification stage, plaintiffs established that AOL stock was traded on an efficient market (i.e., that the market price of AOL stock at any given time reflected all information publically available at that time). As a result, the Court ruled that they were entitled to the fraud-on-the-market presumption of reliance, which allows them to presume that investors relied on defendant's alleged misstatements, in order to prove the second element of Count I.
The fraud-on-the-market presumption is little more than legal recognition of an established economic principle. In an efficient market, a company's stock reacts immediately to new material information. PolyMedica, 432 F.3d at 14. If an executive or analyst makes a material misrepresentation, we can presume that investors relied upon it in purchasing the stock. As a corollary, a company's stock in an efficient market should generally not react to reiterations of previously released information. See Teachers' Ret. Sys. of La. v. Hunter, 477 F.3d 162, 187–88 (4th Cir.2007). A possible exception to this principle might be when the information is later released by a more reputable source or to a larger market. Reframed, the principle might read: a company's stock in an efficient market should not, without good reason, react to reiterations of previously released information.
Dr. Hakala's event study is also unreliable because it repeatedly ignores the efficient market principle. The study attributes volatility in AOL's stock price to the reports of defendant analysts when, at the time of the inflation or deflation, an efficient market would have already priced in the reports. For example, Dr. Hakala labels April 18, 2002 as a corrective date, and attributes stock price deflation to the defendants, even though the information released on that day, Deutsche Bank's lowered estimate and price target, was released nine days earlier without any corresponding impact.
Plaintiffs may not at the same time presume an efficient market to prove reliance and an inefficient market to prove loss causation. They may not have their cake and eat it too.
4. Confounding factors
To establish a causal link between stock price movement and misrepresentations or corrective disclosures, an economist must control for confounding factors, i.e., other industry- or company-specific information released to the market unrelated to the alleged fraud. In re Omnicom Grp., Inc. Sec. Litig., 541 F.Supp.2d 546, 554 (S.D.N.Y.2008). Financial economists have at their disposal a number of sophisticated methods to isolate the causal effect of different news items released on the same day.
One such method, intra-day trading analysis, charts the minute-by-minute stock price and the exact time each news item was released to the market and then compares the two to discern whether any causal relationship can be drawn. See Laura Starks, Discussion of Market Microstructure: An Examination of the Effects on Intraday Event Studies, 10 Contemp. Acct. Res. 355, 383–86 (1994). For example, if the price of a company's stock rose a staggering $10 per share on the same day that two positive reports about that company were released, one would examine the market's reaction to each report. If the stock price remained fairly level after the first report but rose dramatically after the second, the increase could fairly be attributed to the second report and the first report could be discounted as a causal factor. The converse could be inferred if the $10 spike occurred immediately after the first report. An intra-day trading analysis is not proof of a causal relationship but it provides a factfinder with a basis for determining whether and to what extent a single news release impacted the price of a stock on an otherwise confounded event day.
An event study that fails to disaggregate the effects of confounding factors must be excluded because it misleadingly suggests to the jury that a sophisticated statistical analysis proves the impact of defendants' alleged fraud on a stock's price when, in fact, the movement could very well have been caused by other information released to the market on the same date. In re Williams Sec. Litig., 558 F.3d 1130, 1143 (10th Cir.2009), aff'g, 496 F.Supp.2d 1195 (N.D.Okla.2007); Xcelera, 2008 WL 7084626, at *4–5.
Given the extraordinary volume of AOL-related news in the marketplace during the Class Period, Dr. Hakala had a herculean task. His principal approach was to read all the AOL-related news released on a given day and to make subjective judgments about which news impacted the stock price. It would be just as scientific to submit to the jurors evidence of defendants' alleged fraud and AOL's stock fluctuations and let them speculate whether the former caused the latter. See Williams, 558 F.3d at 1143 (finding that expert's subjective intuitions about which news moved a stock price “would be no less speculative and unreliable if reached by jurors”). After all, Dr. Hakala is an expert in financial econometrics, not in securities trading; that is why plaintiffs hired Professors Black and Kraakman. Courts require rigorous event studies to avoid such fallacious post-hoc inferences.
As a result, confounding factors pervade Dr. Hakala's event study. On many event days, information was released from a variety of different sources (e.g., stock analysts, news media and AOL itself) about different aspects of AOL's business (from acquisitions to advertising to subscriptions). Some of the information was positive, some negative. Instead of conducting an intra-day trading analysis for each event day with confounding information (which is, to say, nearly all of them) in order to provide the jury with some basis for discerning the cause of the stock price fluctuation, Dr. Hakala either attributed a rough proportion of the movement to each report or blamed it all on the defendants.
The problem is not just that many event days are hopelessly confounded and not readily attributable to the defendants' alleged misconduct; it is, more importantly, that Dr. Hakala made unreasonable judgments about which factors likely caused stock price movement on event days. For instance, Dr. Hakala attributes AOL's stock price increase on April 18, 2001 to a CSFB report even though AOL itself reported rising first-quarter earnings that morning and the CSFB report merely passed along that news.
There are at least three pitfalls associated with Dr. Hakala's theory. First, it makes little sense to conclude that investors relied upon the CSFB report and ignored the earnings numbers upon which it was based. Second, the academic literature on securities fraud event studies recommends excluding from an event study days with such firm-specific confounding announcements because of a) the difficulty of disentangling the response to the earnings announcement from the response to the analyst forecast revision and b) the typically minor impact of an analyst report issued immediately after an earnings announcement. See Zoran Ivkovic and Narasimhan Jegadeesh, The Timing and Value of Forecast and Recommendation Revisions, 73 J. Fin. Econ. 433, 448 (2004). Third and finally, an intra-day trading analysis conducted by Dr. Stulz suggests that the CSFB report did not cause the stock increase: AOL stock reached its end-of-the-day share price of $49 by 11 a.m., nearly five hours before CSFB issued its report.
In sum, Dr. Hakala's failure to isolate the effect of defendants' alleged fraud from other industry- and company-specific news reported on event days confounds his event study and renders it unreliable.
5. Conclusion
This judicial officer is inclined to let experts testify. The crucible of cross-examination is the usually the best way to assess the reliability of expert testimony. Juries are the hallmark of our legal system and they deserve far more credit than they are given to discern the truth through the fog of competing expert testimony.
Until now, no expert has been precluded on Daubert grounds from testifying in this Session of the United States District Court for the District of Massachusetts. Had Dr. Hakala's event study suffered from only one of the four methodological defects identified by this Court, or suffered from those flaws jointly but to a lesser degree, today's ruling might have been different. Given, however, the pervasiveness of Dr. Hakala's methodological errors and the lack of congruity between his theory and the data, the Court is compelled to exercise its role as gatekeeper and to exclude his event study as unreliable.
IV. Summary judgment
The same deficiencies that render Dr. Hakala's event study inadmissible also prevent the study, even if it were admitted, from raising a triable issue of fact on loss causation.
A. Standard
The role of summary judgment is “to pierce the pleadings and to assess the proof in order to see whether there is a genuine need for trial.” Garside v. Osco Drug, Inc., 895 F.2d 46, 50 (1st Cir.1990) (quoting Fed.R.Civ.P. 56 advisory committee's note). The burden is on the moving party to show, through the pleadings, discovery and affidavits, “that there is no genuine issue as to any material fact and that the moving party is entitled to judgment as a matter of law.” Fed.R.Civ.P. 56(c).
A fact is material if it “might affect the outcome of the suit under the governing law.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). “Factual disputes that are irrelevant or unnecessary will not be counted.” Id. A genuine issue of material fact exists where the evidence with respect to the material fact in dispute “is such that a reasonable jury could return a verdict for the nonmoving party.” Id.
Once the moving party has satisfied its burden, the burden shifts to the non-moving party to set forth specific facts showing that there is a genuine, triable issue. Celotex Corp. v. Catrett, 477 U.S. 317, 324, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986). The Court must view the entire record in the light most favorable to the non-moving party and indulge all reasonable inferences in that party's favor. O'Connor v. Steeves, 994 F.2d 905, 907 (1st Cir.1993). Summary judgment is appropriate if, after viewing the record in the non-moving party's favor, the Court determines that no genuine issue of material fact exists and that the moving party is entitled to judgment as a matter of law.
To survive summary judgment on loss causation in a fraud-on-the-market case, plaintiffs must show that their losses “are attributable to some form of revelation to the market of the wrongfully concealed information.” In re Worldcom, Inc. Sec. Litig., No. 02 Civ. 3288(DLC), 2005 WL 2319118, at *23 (S.D.N.Y. Sept. 21, 2005). The revelation can occur through one or more discrete corrective disclosures or it can “leak out” more gradually. Williams, 558 F.3d at 1137–38. Either way, plaintiffs must establish the mechanism by which the truth was revealed. Id. at 1138. It is not enough simply to note that
the market had learned the truth by a certain date and, because the learning was a gradual process, attribute all prior losses to the revelation of the fraud. The inability to point to a single corrective disclosure does not relieve the plaintiff of showing how the truth was revealed; he cannot say, “Well, the market must have known.”
Id.
Whether plaintiffs proceed under a corrective disclosure theory or a leakage theory, they must establish that the market reacted to the revelation of the alleged misrepresentation or the facts underlying it, not to other negative information about the company unrelated to the alleged fraud. In re Oracle Corp. Sec. Litig., 627 F.3d 376, 392 (9th Cir.2010). For that reason, an expert “must be careful not to connect each and every bit of negative information about a company to an initial misrepresentation,” Williams, 558 F.3d at 1140, such that “every announcement of negative news [is labeled a] corrective disclosure,” In re Motorola Sec. Litig., 505 F.Supp.2d 501, 546 (N.D.Ill.2007).
The market need not know at the time that the practices in question constitute a “fraud,” nor label them “fraudulent”, but in order to establish loss causation, the market must learn of and react to those particular practices themselves.
Oracle, 627 F.3d at 392.
The corrective disclosure must also “present facts to the market that are new, that is, publicly revealed for the first time.” Katyle v. Penn Nat'l Gaming, Inc., 637 F.3d 462, 473 (4th Cir.2011). Consistent with the efficient market principle, if investors already knew the truth, the drop in stock price could not be attributed to the disclosure. Id.;see also FindWhat Investor Grp. v. FindWhat.com, 658 F.3d 1282, 1312 n. 28 (11th Cir.2011) ( “[B]ecause a corrective disclosure must reveal a previously concealed truth, it obviously must disclose new information, and cannot be merely confirmatory.”).
Finally, to survive summary judgment, plaintiffs must isolate the extent to which the decrease in stock price was caused by the disclosure and not, as the Supreme Court has warned, “the tangle of [other] factors affecting [stock] price,” such as
changed economic circumstances, changed investor expectations, new industry-specific or firm-specific facts, conditions, or other events, which taken separately or together account for some or all of that lower price.
Dura Pharm., Inc. v. Broudo, 544 U.S. 336, 343, 125 S.Ct. 1627, 161 L.Ed.2d 577 (2005); see also In re Scientific Atlanta, Inc. Sec. Litig., 754 F.Supp.2d 1339, 1371–76 (N.D.Ga.2010) (“[I]n order to defeat summary judgment, plaintiffs in a securities fraud case must present evidence disaggregating the fraud and non-fraud-related causes of the plaintiff's loss.”); Omnicom, 541 F.Supp.2d at 554 (“Because the law requires the disaggregation of confounding factors, disaggregating only some of them cannot suffice to establish that the alleged misrepresentations actually caused Plaintiffs' loss.”); Williams, 558 F.3d at 1140 (affirming district court's grant of summary judgment on loss causation because plaintiffs' expert failed to disaggregate confounding factors).
B. Application
This brings us to the case at bar, where the “tangle of factors” affecting AOL's stock price during the Class Period was akin to the Gordian knot. AOL was covered extensively by over 40 investment firms, many if not most of which projected aggressive earnings estimates for AOL during the Class Period. As one of the first corporate entities to bring under one roof “old” and “new” media, and one of the major players in the burgeoning internet marketplace, AOL was abnormally susceptible to stock price volatility. With the merger also came exposure to different industries, yet another factor obscuring the true cause(s) of its stock price movement.
Dr. Hakala's event study purports to disentangle those factors and pinpoint with precision the effect of defendants' alleged fraud on AOL's stock price. To do so, Dr. Hakala identifies 36 corrective disclosure dates on which, he claims, defendants' alleged fraud was revealed to the market and caused a decline in AOL stock price. Careful review of the information disclosed to the market on those dates reveals three infirmities in Dr. Hakala's methodology that prevent plaintiffs from raising a triable issue of fact on loss causation.
First, Dr. Hakala fails to link many of the events he labels as corrective disclosures to the defendants' alleged fraud. Plaintiffs allege in the complaint that CSFB knew, yet failed to disclose, that
1) a substantial weakening in the advertising markets would likely prevent AOL from reaching its earning projections;
2) AOL engaged in inappropriate accounting activities which artificially inflated its revenue statements; and
3) layoffs had occurred which, if made public, would negatively impact AOL's stock price.
In a fraud-on-the-market case such as this one, in which defendant analysts are alleged to have fraudulently misstated a company's earning potential, plaintiffs must do more than show that “the market reacted to the purported ‘impact’ of the alleged fraud—the earnings miss.” Oracle, 627 F.3d at 392. To qualify as a corrective disclosure in this case, information released to the market must pertain to at least one of the foregoing topics about which defendants allegedly concealed information. See Lentell v. Merrill Lynch & Co., Inc., 396 F.3d 161, 172–73 (2d Cir.2005).
Yet many of the corrective disclosures identified by Dr. Hakala do not contain any mention of weakening advertising markets, inappropriate accounting practices or stealth layoffs. Information released on many of the identified corrective disclosure dates pertained to subscription rates (January 31, 2001, and January 2 and 4, March 12 and April 11, 2002), unrelated acquisitions (January 31 and December 3, 2001), management changes (April 11, 2002), projection downgrades by other firms (December 7, 2001, and January 3 and 4, 2002) and AOL's quarterly earnings miss (October 17, 2001).
Second, as discussed in greater depth in Section II.C.3 supra, many of the so-called corrective disclosures did not include new information. Assuming an efficient market, as plaintiffs have, the information disclosed to the market on many of the identified corrective disclosure days was already incorporated into the AOL stock price. Dr. Hakala should not have attributed it to defendants' alleged fraud.
Third, as discussed in greater depth in Section II.C.4 supra, Dr. Hakala failed properly to isolate the extent to which the stock price deflation was caused by the disclosures and not by other confounding factors.
This is not the first time a court has premised, at least in part, its entry of summary judgment on Dr. Hakala's failure to disaggregate confounding factors. In Omnicom, the court held that Dr. Hakala's event study could not establish loss causation, and granted summary judgment for defendants as a result, because
to the extent that any corrective disclosures exist, the event study does not isolate their effect on Omnicom's stock price from that of [other] negative reporting [and] the effect of post-Enron changed investor expectations.
541 F.Supp.2d at 554. Affirming the district court's ruling on appeal, the Second Circuit reiterated that his event study “does not suffice to draw the requisite causal connection” between the corrective disclosure and the fraud alleged in the complaint. In re Omnicom Grp., Inc. Sec. Litig., 597 F.3d 501, 512 (2d Cir.2010).
The court reached the same result in Scientific Atlanta:
Dr. Hakala's analysis fails to disentangle the effect of the new information regarding customer inventory levels from SA's new, negative characterization of how industry-wide trends were affecting it specifically.
754 F.Supp.2d at 1379. It would be a mistake to submit the evidence to the jury, the court reasoned, because the jury would have no basis for determining how much, if any, of the plaintiffs' loss should be attributed to defendants' fraudulent conduct. Id.
This Court reaches the same conclusion. Nearly all of the 54 event days are confoundedin some way; many of the days, hopelessly so. It is not enough to divide each day's economic loss by the number of analyst reports on that day and attribute a corresponding proportion to the defendants, without demonstrating that the defendants' report was a substantial cause of the loss. See id. at 1376. Nor is it sufficient simply to speculate that one report caused the loss instead of another,
without providing the factfinder with a basis for evaluating the relative effects of [potential] competing causes, thereby determining which factors were substantial and which were relatively minor or inconsequential.
Id. As the Supreme Court has emphasized, “[t]o ‘touch upon’ a loss is not to cause a loss, and it is the latter that the law requires.” Dura, 544 U.S. at 343, 125 S.Ct. 1627. This Court finds, as did the Omnicom and Scientific Atlanta courts, that Dr. Hakala's “partial disaggregation of confounding factors is insufficient to establish that the alleged misrepresentations actually caused plaintiffs' loss.” Scientific Atlanta, 754 F.Supp.2d at 1379.
C. Conclusion
Thus the Court concludes, as did the Fifth Circuit in Fener, that “once we disregard Hakala's flawed event study, the rest of his testimony is insufficient to prove loss causation.” 579 F.3d at 410;see also Xcelera, No. 00–11649–RWZ (allowing defendants' renewed motion for summary judgment in light of Court's exclusion of Dr. Hakala's event study); Williams, 496 F.Supp.2d at 1290–95 (granting summary judgment for defendants on loss causation after excluding expert witness's testimony as unreliable). Because plaintiffs have failed to raise a triable issue of fact on the element of loss causation, defendants are entitled to summary judgment on Counts I and II.
V. Other experts
Finding that the exclusion of Dr. Hakala's event study and testimony entitles the defendants to summary judgment on Counts I and II, the Court declines to assess whether the remaining experts should be permitted to testify.
ORDER
In accordance with the foregoing, defendants' motion to preclude the expert opinions of Scott Hakala, M. Laurentius Marais, Bernard Black and Reinier Kraakman (Docket No. 303) is ALLOWED, in part, and DENIED, in part. The motion is allowed as it pertains to Dr. Scott Hakala and denied as moot as it pertains to Dr. Marais and Professors Black and Kraakman. Plaintiffs' motion to preclude the expert opinions of Rene Stulz and John Deighton (Docket No. 310) is DENIED as moot.
Having ruled that Dr. Hakala's event study is inadmissible and having concluded that, without it, plaintiffs cannot raise a triable issue of fact as to loss causation, the Court grants summary judgment sua sponte on Counts I and II in favor of the defendants.
So ordered.
MEMORANDUM & ORDER
In a Memorandum and Order entered on January 13, 2012 (“the January 2012 M & O”), this Court excluded the event study of Dr. Scott Hakala and granted summary judgment in favor of the defendants. Pending before the Court is plaintiffs' motion to reconsider that ruling.
I. Background
The background and procedural history of this case was recounted in detail in the January 2012 M & O and will be rehearsed here only to the extent necessary. In 2003, various cases were consolidated to comprise this securities fraud class action over which United States District Judge Nancy Gertner was assigned to preside. In September 2006, Judge Gertner denied defendants' motions to dismiss and, two years later, certified the class.
In a Memorandum and Order entered on August 26, 2011, 2011 WL 3813204, Judge Gertner provisionally denied defendants' motions for summary judgment without ruling on the admissibility of expert testimony. She explained that “summary judgment—without a full Daubert hearing—is an inappropriate way” to decide the complicated expert issues presented. Because reliance and loss causation, two central elements of plaintiffs' case, “necessarily rely on expert testimony,” Judge Gertner intimated that if Dr. Hakala's testimony were excluded, her conditional summary judgment decision would have to be revisited.
Upon Judge Gertner's retirement, the case was transferred to this Session. In December 2011, a Daubert hearing was held at which counsel were given an opportunity to be heard with respect to their expert preclusion motions and Drs. Hakala and Marais testified. On January 13, 2012, the Court rendered its opinion. After finding four critical methodological flaws in Dr. Hakala's event study, the Court excluded that study as unreliable and granted summary judgment sua sponte in favor of the defendants on the grounds that, without the event study, plaintiffs could not raise a triable issue of fact as to loss causation. Shortly thereafter, plaintiffs filed the instant motion for reconsideration.
II. Motion for Reconsideration
A. Standard
Ruling on a motion for reconsideration requires a court to “balance the need for finality against the duty to render just decisions.” Davis v. Lehane, 89 F.Supp.2d 142, 147 (D.Mass.2000). Reconsideration is warranted only in a limited number of circumstances:
if the moving party presents newly discovered evidence, if there has been an intervening change in the law, or if the movant can demonstrate that the original decision was based on a manifest error of law or was clearly unjust.
United States v. Allen, 573 F.3d 42, 53 (1st Cir.2009). Such motions are not vehicles for pressing arguments which could have been asserted earlier or for “re-arguing theories previously advanced and rejected.” Palmer v. Champion Mortg., 465 F.3d 24, 30 (1st Cir.2006). The granting of such a motion is “an extraordinary remedy which should be used sparingly.” Id.
B. Application
Plaintiffs present procedural and substantive grounds for reconsideration. First, they contend that the Court misapplied the rules governing the admissibility of expert testimony. Second, they assert that the Court erred by granting summary judgment sua sponte without affording them notice and an opportunity to present curative evidence.
1. Expert Preclusion
The Court's ruling to preclude Dr. Hakala's testimony was rendered after a thorough examination of his expert report, multiple rounds of briefing and a full Daubert hearing which included live testimony from Dr. Hakala himself. Because plaintiffs do little more than advance arguments already considered and rejected by this Court, their motion to reconsider the Court's expert preclusion ruling will be denied.
2. Notice
A district court may enter summary judgment sua sponte as long as “discovery is sufficiently advanced that the parties have enjoyed a reasonable opportunity to glean the material facts,” and the party against whom judgment is to be entered has been given “notice and a chance to present its evidence on the essential elements of the claim.” Berkovitz v. Home Box Office, Inc., 89 F.3d 24, 29 (1st Cir.1996). Formal notice is not required. As then-Judge Breyer explained,
[B]eing “on notice” does not mean that [plaintiff] had to receive a formal document called “notice” or that the district court had to say the words “you are on notice” or even that the court had to explicitly tell [plaintiff], “I am thinking of ordering summary judgment for [defendants] sua sponte.”
Nat'l Expositions, Inc. v. Crowley Maritime Corp., 824 F.2d 131, 133 (1st Cir.1987). Rather, the issue is whether the party against whom judgment is to be entered has been given “a meaningful opportunity” to present the best evidence and legal arguments in support of its position. Id. at 31 (affirming sua sponte grant of summary judgment where the plaintiff had repeatedly addressed the relevant facts and legal issues); see also Young v. City of Providence, 404 F.3d 4, 13 n. 4 (1st Cir.2005) (finding that plaintiff had sufficient notice because the basis for the Court's ruling was raised in defendants' initial summary judgment motion, to which plaintiff had responded).
According to plaintiffs, defendants' motion for summary judgment was denied in full by Judge Gertner and, by entering summary judgment sua sponte, this Session effectively second-guessed her decision. Plaintiffs, however, gloss over the fact that Judge Gertner expressly declined to rule on whether triable issues of fact remained with respect to reliance and loss causation and indicated that if the Court were to preclude the testimony of Dr. Hakala after a Daubert hearing, her summary judgment decision would need to be reconsidered. Far from second-guessing Judge Gertner, this Session has followed her guidance.
Indeed, Judge Gertner specifically suggested that the best course of action would be “follow Judge Zobel's lead in In re Xcelera.com Securities Litigation.” In that securities fraud case, summary judgment was provisionally denied until the Court had the chance to rule upon the admissibility of the expert testimony after a Daubert hearing, whereupon the Court precluded the expert (Dr. Hakala, coincidentally) from testifying and granted summary judgment in favor of defendants. This Session has done likewise.
By unambiguously deferring a ruling on loss causation until after the Daubert hearing, Judge Gertner's opinion put plaintiffs on notice that such issues remained pending before the Court. Plaintiffs were given ample opportunity to convince the Court that Dr. Hakala's report was sufficient to establish loss causation. Their inability to do so does not furnish a basis for reconsideration.
Plaintiffs also complain that the Court unfairly changed course after leading them to believe that the action would proceed to trial. As evidence, they point to statements made by the Court at the November 21, 2011 status conference indicating that the case would likely proceed to trial and the Daubert hearing would not be as comprehensive as counsel envisioned. Plaintiffs misconstrue the Court's comments. First, by stating that the chances were high that the case would go to trial, the Court was merely notifying the parties that its general inclination is to let expert witnesses testify. Dr. Hakala is, in fact, the first expert witness precluded from testifying by this Session in many years. Second, the reference to the comprehensiveness of the Daubert hearing was, as the context makes clear, intended to inform the parties that the hearing would be limited in duration. The Court was not degrading its importance.
Finally, plaintiffs maintain that they were prejudiced by the Court's sua sponte grant of summary judgment because it prevented them from submitting curative evidence in the form of the Mulcahey Report, a 1,000–page expert report purporting to demonstrate that the methodological flaws pointed out by the Court are correctable. Plaintiffs fail to explain their untimely submission of this report more than three years after the close of fact and expert discovery. The same flaws of which the Court took notice were identified by the defendants as early as April, 2007. Plaintiffs had ample opportunity to respond to them and may not reopen expert discovery at this late stage of the case.
ORDER
In accordance with the foregoing, plaintiffs' motion for reconsideration (Docket No. 397) is DENIED.
So ordered.