From Casetext: Smarter Legal Research

In re BP P.L.C. Sec. Litig.

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF TEXAS HOUSTON DIVISION
Dec 6, 2013
MDL NO.: 10-md-2185 (S.D. Tex. Dec. 6, 2013)

Summary

noting "statistical regression analys[e]s that examine the effect of an event on a dependent variable, such as a corporation's stock price," are "commonly used in securities fraud class actions"

Summary of this case from Sec. & Exch. Comm'n v. World Tree Fin.

Opinion

MDL NO.: 10-md-2185 Civil Action No. 4:10-md-2185

12-06-2013

IN RE: BP p.l.c. SECURITIES LITIGATION


HON. KEITH P. ELLISON


MEMORANDUM AND ORDER

Pending before the Court is Plaintiffs' Motion for Class Certification. (Doc. No. 652.) Having reviewed the motion, Defendants' response (Doc. No. 664), Plaintiffs' reply brief in support of their motion (Doc. No. 677), Defendants' sur-reply (Doc. No. 688), Plaintiffs' response to Defendants' sur-reply (Doc. No. 691-1), all papers in support thereof, and having heard oral argument, the Court finds that Plaintiffs' Motion for Class Certification (Doc. No. 652) must be DENIED.

All docket references are to Multi-District Litigation No. 10-md-2185.

I. BACKGROUND AND PROCEDURAL HISTORY

A. Actionable Misstatements

This case is a putative class action against three BP corporations and two BP executives ("Defendants"). Plaintiffs assert securities fraud claims on behalf of purchasers of BP American Depositary Shares ("ADSs") on the New York Stock Exchange ("NYSE") who were allegedly injured by a series of misrepresentations and omissions made by Defendants between 2007 and 2010. According to Plaintiffs, the misleading nature of these statements and omissions was made clear by the April 20, 2010 explosion of the Deepwater Horizon drilling rig; the ensuing oil spill into the Gulf of Mexico; and various events and disclosures in the months following the explosion. Three times, the Court has engaged at length with Plaintiffs' theories of Defendants' liability. A more complete explication of the relevant facts and allegations can be found in the lengthy opinions issued on those occasions. See In re BP p.l.c. Securities Litig. ("BP I"), 843 F. Supp. 2d 712 (S.D. Tex. 2012); In re BP p.l.c. Securities Litig. ("BP II"), 852 F. Supp. 2d 767 (S.D. Tex. 2012); In re BP p.l.c. Securities Litig. ("BP III"), 922 F. Supp. 2d 600 (S.D. Tex. 2013).

Following the Court's decisions on Defendants' three motions to dismiss, Plaintiffs' claims have been narrowed such that only four broad categories of alleged misstatements and omissions remain in the case. These are:

• Statements touting BP's progress in implementing the recommendations of the independent commission known as the "Baker Panel" following the 2005 explosion at the Company's Texas City refinery. The Baker Panel was convened to review and suggest improvements to BP's safety practices, the efficacy of which was seriously in doubt following a series of high-profile safety mishaps. The Baker Panel released a report in January 2007 (the "Baker Report"), which included a series of specific recommendations intended to improve BP's safety culture and processes. Plaintiffs claim that, following the release of the Baker Report, Defendants repeatedly publicized their progress on the Report's recommendations as a way to assuage the public that BP had turned a corner on safety. In reality, according to Plaintiffs, nothing about BP's safety programs had changed, and BP remained an accident waiting to happen. Alleged misstatements in this category were made in November 2007, February 2008, April 2008, December 2008, and March 2010.
• Statements describing BP's Operating Management System ("OMS") as a system being applied across all of BP's lines of business, worldwide, in an attempt to standardize safety processes. Statements in this category were allegedly misleading because they omitted that OMS would not govern safety practices at contractor-owned sites, such as the Deepwater Horizon drilling rig. Statements in this category were also allegedly misleading because they represented that OMS had been implemented in the Gulf of Mexico by the time of the Deepwater Horizon explosion, when Plaintiffs claim it had not. Alleged misstatements in this category were made in February 2009, March 2009, April 2009, February 2010, March 2010, and April 2010.
• Statements from two agency filings—the Initial Exploration Plan ("IEP") and the Gulf of Mexico Regional Oil Spill Response Plan ("OSRP")—describing BP's ability to respond to a catastrophic deepwater oil spill. According to Plaintiffs, these statements were grossly inaccurate, and BP had no contingency plans and
no adequate response equipment for a disaster. The documents were filed with the relevant federal agency, the U.S. Department of the Interior's Minerals Management Service ("MMS"), in March 2009 and June 2009 respectively. Plaintiffs claim that they were publicly available documents upon filing. They were also scrutinized in the media following the Deepwater Horizon explosion.
• Statements made after the April 20, 2010 Deepwater Horizon explosion regarding the magnitude of the resulting oil spill. According to Plaintiffs, Defendants perpetuated the fiction that the spill was only approximately 5,000 barrels per day, even as internal BP estimates showed that the true number was much higher. Alleged misstatements in this category were made in late April 2010 and May 2010.

Although all the surviving misrepresentations in this case can be sorted into the above four categories, the first two categories are closely related. Specifically, OMS was a response to one of the Baker Report recommendations. Therefore, the alleged misrepresentations regarding OMS might be considered an extension, or a subset, of the alleged misrepresentations regarding BP's progress on the Baker Panel recommendations.

B. Lead Plaintiffs

Pursuant to procedures required by the Private Securities Litigation Reform Act ("PSLRA"), the Court previously had occasion to consider and appoint lead plaintiffs and lead counsel in this action. Two groups of prospective lead plaintiffs—"NY/OH" and the "Ludlow Plaintiffs"—vied for appointment.

The NY/OH group consisted of (1) Thomas P. DiNapoli, Comptroller of the State of New York, as Administrative Head of the New York State and Local Retirement Systems and sole Trustee of the New York State Common Retirement Fund ("New York"), and (2) Ohio Attorney General Mike DeWine, statutory litigation counsel for the Ohio Public Employees Retirement System ("Ohio"). They proposed a fairly long class period, stretching from June 2005 until June 2010. As the proposed class period indicates, they alleged that Defendants engaged in misrepresentations both before and after the Deepwater Horizon explosion itself.

During the early stages of this case, the Ohio Attorney General represented four Ohio public pension funds. (Doc. No. 113, at ¶ 32.) In the most recent pleading, however, he represents only the Ohio Public Employees Retirement System. (Doc. No. 339 ("SAC"), at ¶ 36.)

In addition to their securities fraud claims based on their purchases of BP ADSs on the NYSE, NY/OH brought securities fraud claims based on their purchases of BP Ordinary Shares on the London Stock Exchange. The Ludlow Plaintiffs did not bring these types of claims. The Ordinary Shares claims were dismissed in February 2012 and are no longer germane to this case. In re BP p.l.c. Securities Litig. ("BP I"), 843 F. Supp. 2d 712, 796 (S.D. Tex. 2012).

The Ludlow Plaintiffs consisted of four individual investors: Robert H. Ludlow, Peter D. Lichtman, Leslie J. Nakagiri, and Paul Huyck. They proposed a much shorter class period, from March 2009 until April 2010. They did not allege that Defendants committed securities fraud after the Deepwater Horizon explosion. They accused NY/OH of proposing an improperly long class period in order to maximize the extent of their BP holdings—a factor for consideration in the lead plaintiff analysis. They also asserted that NY/OH's emphasis on post-explosion fraud would dilute the class's pre-explosion claims.

Following extensive briefing, the Court elected to appoint both sets of plaintiffs: NY/OH as lead plaintiffs for the proposed class, and the Ludlow Plaintiffs as lead plaintiffs for a wholly subsumed subclass. The Court was motivated by concerns that the two groups had articulated significantly different theories of the alleged fraud:

First, whereas the Ludlow Plaintiffs' claims center on BP's statements about the safety of its drilling operations in the Gulf of Mexico in the thirteen months leading up to the Deepwater Horizon explosion, New York & Ohio argue more generally that BP made fraudulent statements between 2005 and 2010 about its safety precautions both in the Gulf of Mexico and elsewhere. Second, in extending the class period to June 1, 2010, New York & Ohio focus a good deal of attention on fraud allegedly committed after April 20, 2010, such as statements about the rate at which oil was escaping from the rig and about BP's attempts to stop the flow. A substantial portion of New York & Ohio's losses derive from purchases of BP [ADSs] by one of the Ohio funds on May 3 and May 25, 2010, several weeks after the Deepwater Horizon explosion.
In re BP Securities Litig., 758 F. Supp. 2d 428, 438 (S.D. Tex. 2010). The Court attributed the competing theories to the fact that NY/OH were "net sellers" of ADSs during the proposed class period proposed by the Ludlow Plaintiffs. Id. In other words, because NY/OH's losses were "concentrated outside of the Ludlow Period," they were incentivized to "present different legal theories than other plaintiffs." Id. The Court concluded that NY/OH had not shown themselves to be typical and adequate lead plaintiffs for the so-called Ludlow Period. Id. At the same time, the Ludlow Plaintiffs were similarly inadequate to represent plaintiffs who purchased before and after the Ludlow Period. Id. at 439. To accommodate the interests of all members of the proposed class, the Court opted to appoint NY/OH as lead plaintiffs of the class, and the Ludlow Plaintiffs as lead plaintiffs of a subclass. Id. at 442.

Following the Court's designation of lead plaintiffs, NY/OH and the Ludlow Plaintiffs filed separate complaints. Defendants moved to dismiss both complaints; the Ludlow Complaint was dismissed in its entirety, and the NY/OH Complaint was partially dismissed. See BP II, 852 F. Supp. 2d at 820; BP I, 843 F. Supp. 2d at 799. NY/OH and the Ludlow Plaintiffs then jointly filed a single consolidated, amended complaint: the Second Consolidated Amended Class Action Complaint, or "SAC." (Doc. No. 339.) Defendants moved for partial dismissal of the SAC, which was partially granted and partially denied. See BP III, 922 F. Supp. 2d at 640-41. Discovery has commenced on the surviving claims in the SAC, and the Court has entered a docket control order with trial scheduled for August and September 2014. (Doc. No. 582.)

II. PLAINTIFFS' PROPOSED CERTIFICATION

New York, Ohio, and the Ludlow Plaintiffs have filed a motion for class certification under Rule 23. They seek the following certification:

All persons and entities who purchased or otherwise acquired BP's ADSs between November 8, 2007 and May 28, 2010 [and were injured thereby] ("Class"), as well as all persons and entities who purchased or otherwise acquired BP's ADSs between March 4, 2009 and April 20, 2010 [and were injured thereby] ("[Ludlow Subclass]"). Excluded from the Class and [Subclass] are Defendants, directors and officers of BP, their families and affiliates, as well as the retirement accounts of Defendants and BP's directors and officers.
(Doc. No. 652-1 ("Mot."), at 2.) As made clear above, the proposed "Class Period" is November 8, 2007 to May 28, 2010. The proposed "Ludlow Subclass Period" is subsumed within the Class Period and is March 4, 2009 to April 20, 2010 (the date of the Deepwater Horizon explosion). New York and Ohio are proposed representatives for the Class; the Ludlow Plaintiffs are proposed representatives for the Ludlow Subclass.

The bracketed phrase has been proposed by Defendants in the event that Plaintiffs' motion is granted. (Doc. No. 664 ("Opp."), at 15 n.6.) Plaintiffs do not oppose the addition. (Doc. No. 677 ("Reply"), at 12 n.14.)

III. LEGAL STANDARDS

Federal class actions are governed by the requirements of Federal Rule of Civil Procedure 23. Unger v. Amedisys Inc., 401 F.3d 316, 320 (5th Cir. 2005). The party seeking class certification bears the burden of showing that Rule 23 has been satisfied. Id. (citing Berger v. Compaq Computer Corp., 257 F.3d 475, 479-80 (5th Cir. 2001)); see also Wal-Mart Stores, Inc. v. Dukes, — U.S. —, 131 S. Ct. 2541, 2551 (2011) ("Rule 23 does not set forth a mere pleading standard. A party seeking class certification must affirmatively demonstrate his compliance with the Rule—that is, he must be prepared to prove that there are in fact sufficiently numerous parties, common questions of law or fact, etc."). The requirements for class certification under Rule 23(a) are:

(1) The class is so numerous that joinder of all members is impracticable;
(2) there are questions of law or fact common to the class;
(3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and
(4) the representative parties will fairly and adequately protect the interests of the class.
Fed. R. Civ. P. 23(a)(1)-(4).

"To obtain class certification, parties must satisfy Rule 23(a)'s four threshold requirements, as well as the requirements of Rule 23(b)(1), (2), or (3)." M.D. ex rel. Stukenberg v. Perry, 675 F.3d 832, 837 (5th Cir. 2012) (citing Maldonado v. Ochsner Clinic Found., 493 F.3d 521, 523 (5th Cir. 2007)). Plaintiffs in this case seek certification under Rule 23(b)(3), which requires the Court to find two additional requirements satisfied: predominance and superiority. See Unger, 401 F.3d at 320. "The predominance element requires a finding that common issues of law or fact 'predominate over any questions affecting only individual members.'" Id. (quoting Fed. R. Civ. P. 23(b)(3)). This requirement is more demanding than the commonality prong of Rule 23(a) "because it 'tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation.'" Id. (quoting Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 623 (1997)). Finally, a party seeking class certification must show that a class action "is superior to other available methods for fairly and efficiently adjudicating the controversy." Fed. R. Civ. P. 23(b)(3).

IV. DEFENDANTS' OPPOSITION

Defendants oppose certification. They organize their objections to class certification under the following four principal arguments:

(1) Defendants claim that Rule 23(a) has not been satisfied because New York and Ohio are atypical and inadequate representatives of the proposed Class. Defendants' argument is grounded in the specifics of New York and Ohio's trading activity during the Class Period, as described below. (Doc. No. 664 ("Opp."), at 10-20.)
(2) Defendants contend that the Ludlow Plaintiffs have not established that the Rule 23 prerequisites are satisfied for the proposed Ludlow Subclass. (Id. at 20-22.)
(3) Defendants argue that the predominance and superiority requirements of Rule 23(b)(3) are not met with regards to Plaintiffs' claims based on the IEP and OSRP, because the fraud-on-the-market presumption is not available for those claims. (Id. at 23-36.)
(4) Defendants contend that the predominance requirement of Rule 23(b)(3) is not met because Plaintiffs have not established that damages can be calculated on a class-wide basis that is consistent with their theories of liability. (Id. at 37-40.)

V. ANALYSIS

To certify a class in this action, the Court must find that Plaintiffs have satisfied each of the four requirements of Rule 23(a), as well as both requirements of Rule 23(b)(3). Each will be addressed in turn.

A. Rule 23(a)

1. Numerosity

"To satisfy the numerosity prong, 'a plaintiff must ordinarily demonstrate some evidence or reasonable estimate of the number of purported class members.'" Pederson v. Louisiana State Univ., 213 F.3d 858, 868 (5th Cir. 2000) (quoting Zeidman v. J. Ray McDermott & Co., Inc., 651 F.2d 1030, 1038 (5th Cir. 1981)). In support of numerosity, Plaintiffs cite the following facts:

(1) BP's ADSs are listed on the NYSE. (Doc. No. 652-4, at 137.)
(2) According to BP's 2010 Annual Report, approximately 115,000 registered holders held approximately 815 million ADSs as of February 18, 2011. (Id.)
(3) On average, during the Class Period, more than 7.4 million BP ADSs traded each day. (Doc. No. 652-5 ("Coffman Report"), at Exh. 1.)

Defendants do not address numerosity in their opposition, apparently conceding that it exists in this case. By identifying the volume of trading in BP ADSs on an open and efficient market, Plaintiffs have satisfied the requirement that the class is so "numerous" as to make joinder "impracticable." Fed. R. Civ. P. 23(a); Harman v. LyphoMed, Inc., 122 F.R.D. 522, 524 (N.D. Ill. 1988) (finding numerosity element met when average weekly trading volume during the class period exceeded one million shares).

2. Commonality

Under the commonality prong, a plaintiff must demonstrate that "there are questions of law or fact common to the class." Fed. R. Civ. P. 23(a)(2). To meet this requirement, Plaintiffs' claims "must depend upon a common contention" which is "capable of classwide resolution." Wal-Mart, 131 S. Ct. at 2551. In other words, the determination of the truth or falsity of the so-called "common contention" must "resolve an issue that is central to the validity of each one of the claims in one stroke." Id.

Plaintiffs have identified several foundational questions common to the class, primarily concerning Defendants' course of conduct and whether Defendants violated federal securities law. These include whether the alleged misstatements were materially misleading and whether they were made with the requisite scienter. (Mot. at 5.) Defendants do not contest that commonality is met, and the Court agrees that common questions regarding Defendants' liability satisfy this requirement of Rule 23(a). See In re IndyMac Mortgage-Backed Securities Litig., 286 F.R.D. 226, 233 (S.D.N.Y. 2012) ("[Commonality] is 'plainly satisfied' in a securities case where 'the alleged misrepresentations . . . relate to all the investors, [because the] existence and materiality of such misrepresentations obviously present important common issues.'") (quoting Korn v. Franchard Corp., 456 F.2d 1206, 1210 (2d Cir. 1972)).

3. Typicality

To assess typicality, a court must determine that "the claims or defenses of the representative parties are typical of the claims or defenses of the class." Fed. R. Civ. P. 23(a)(2). "Typicality focuses on the similarity between the named plaintiffs' legal and remedial theories and the legal and remedial theories of those whom they purport to represent." Lightbourn v. County of El Paso, Tex., 118 F.3d 421, 426 (5th Cir. 1997). As the Supreme Court has explained, both the commonality and the typicality requirements "serve as guideposts for determining whether under the particular circumstances maintenance of a class action is economical and whether the named plaintiff's claim and the class claims are so interrelated that the interests of the class members will be fairly and adequately protected in their absence." Gen. Tel. Co. of Southwest v. Falcon, 457 U.S. 147, 157 n.13 (1982).

Plaintiffs contend that their claims share the essential characteristics of claims common to all members, in that (1) they purchased ADSs during the Class Period; (2) at prices artificially inflated by Defendants' misrepresentations; and (3) they suffered losses when the truth was revealed to the market and the price corrected. According to Plaintiffs, they are subject to no unique defenses not applicable to class members generally. (Mot. at 6.)

a. New York and Ohio

Defendants dispute that New York and Ohio's claims are typical of the class. They emphasize that New York and Ohio were both net sellers of ADSs during the Class Period, primarily because of their trading activity before the April 20, 2010 explosion. (Opp. at 11-13.) The pertinent details of New York and Ohio's trading activity are as follows:

• At the beginning of the Class Period, Ohio held 528,650 ADSs. By October 24, 2008, Ohio had sold all of its ADSs. It did not make any more purchases of ADSs until after the Deepwater Horizon explosion. After the explosion, Ohio purchased ADSs on May 3, 2010 and May 25, 2010. At the end of the Class Period, Ohio held 169,000 ADSs. (Doc. No. 677-3, at 2-3.)
• At the beginning of the Class Period, New York held 644,000 ADSs. It both bought and sold ADSs between November 8, 2007 and the date of the explosion,
but its sales outnumbered its purchases by approximately 2 to 1 (949,368 ADSs sold versus 498,094 ADSs purchased). After the explosion, New York bought ADSs on April 29, 2010 and May 26, 2010, and sold ADSs on May 27, 2010. Post-spill, its purchases outnumbered its sales by approximately 2 to 1 (20,080 ADSs purchased versus 10,510 ADSs sold). At the end of the Class Period, New York held 202,296 ADSs. (Doc. No. 677-2, at 2-3; Doc. No. 23-1, at 87-88.)

According to Defendants, New York and Ohio are atypical of the class because they profited from the alleged fraud. (Opp. at 15-18.) Defendants' expert, Dr. René Stulz, utilizes a methodology known as "constant dollar" to calculate that New York profited between $12.7-16.4 million, and Ohio profited approximately $12.6 million during the Class Period. (Doc. No. 664-18 ("Stulz Report"), at ¶¶ 12-13 & Ex. 1.) Because they profited from the alleged fraud, Defendants continue, New York and Ohio are subject to the unique defenses that: (1) they did not rely on Defendants' alleged misstatements and (2) they have no standing because they suffered no injury. (Opp. at 11-13, 15-18.)

New York and Ohio counter that the appropriate inquiry, for purposes of Rule 23(a), is not whether they were net sellers over the Class Period, but whether they suffered losses. (Doc. No. 677 ("Reply"), at 3-4.) New York and Ohio maintain that they did suffer losses during the Class Period, under two recognized accounting methodologies: Last-In-First-Out ("LIFO") and First-In-First-Out ("FIFO"). (Reply at 3.) They also claim to have suffered legally cognizable damages from their investments in BP ADSs, citing a "rudimentary" damages inquiry known as "retained share." (Id. 5-7.) Before addressing the results of these various calculations, as applied to New York and Ohio's trade histories, it is appropriate to define the terms cited by the parties.

Constant Dollar: As described at various points, the "constant dollar" approach appears to equate the amount of a stock price-drop on a given corrective disclosure day (after adjusting for confounding events, such as general market movement) with the "inflation" created by a preceding misrepresentation. This
"constant dollar" amount of inflation is then carried back to the date of the preceding misrepresentation, such that any share purchased in the interim is deemed "inflated" by that amount. (Doc. No. 677-4 ("Coffman Reply"), at ¶ 60.)
Retained Share: This approach involves identifying the number of shares purchased during the class period that are retained at the end of the class period, and "calculating damages by subtracting the purchase price for these retained shares from either (1) the average of the daily closing price of the stock during the 90 day period beginning at the end of the class period (if the share was not sold during the 90 day period) or (2) the higher of the actual sale price or an average of the daily closing price from the end of the class period to the date of sale (if a share was sold within the 90 day period)." Perlmutter v. Intuitive Surgical, Inc., No. 10-CV-03451-LHK, 2011 WL 566814, at *6-7 (N.D. Cal. Feb. 15, 2011).

In this opinion, the Court will use the term "corrective disclosure" to refer to any event alleged by Plaintiffs to have revealed that a prior representation made by one of the Defendants was untrue or misleading. The Court notes the potential confusion that this term can engender when some "corrective" events are materializations of concealed or understated risks. See In re Vivendi Univ., S.A. Securities Litig., 634 F. Supp. 2d 352, 363 n.9 (S.D.N.Y. 2009) (noting that "many courts in [this] district consider corrective disclosures to be a separate conceptual category from 'materalization of the risk'"). However, because the parties employ the "corrective disclosure" term in their briefing and arguments, the Court will do so as well.

In the Court's understanding, the above terms describe competing methods of calculating securities fraud damages. LIFO and FIFO, by contrast, are methods of accounting for sold shares in order to perform the above "constant dollar" or "retained share" calculations.

LIFO: Under the LIFO method, shares purchased most recently are assumed to have been sold first. (Stulz Report at ¶ 13, n.6.)
FIFO: Under the FIFO method, shares purchased first are assumed to have been sold first. (Id.)

Plaintiffs claim that New York suffered $2.24 million in losses over the Class Period under LIFO or $8.79 million in losses under FIFO. (Reply at 3; Doc. No. 677-2, at 2-3.) They claim that Ohio suffered $1.8M in losses under LIFO or $4.88 million in losses under FIFO. (Reply at 3; Doc. No. 677-3, at 2-3.) In addition to these overall trading losses, Plaintiffs claim that New York and Ohio suffered damages from Defendants' alleged fraud because each "at the end of the Class Period . . . held a significant number of shares that were purchased during the Class Period[.]" (Reply at 5.)

Plaintiffs criticize Defendants' expert for employing the "constant dollar" methodology to calculate Plaintiffs' alleged "gains" over the Class Period. They contend that Dr. Stulz's calculations are unreliable under Federal Rule 702 and should be struck, and that Defendants' attendant arguments regarding New York and Ohio's atypicality and lack of standing then necessarily fail. (Reply at 7-8.)

To be clear, Plaintiffs' and Defendants' experts agree that the "constant dollar" methodology, in the context of a case alleging misstated or concealed risk, does not yield a "but for" price that the stock would have traded at absent the alleged fraud. (Doc. No. 664-18 ("Stulz Report"), at ¶¶ 58-62; Doc. No. 677-4 ("Coffman Reply"), at ¶¶ 48-52.) Nonetheless, Plaintiffs' expert contends that the "constant dollar" methodology is the appropriate measure of Plaintiffs' damages from the alleged fraud. (Coffman Reply at ¶¶ 53-55.) Neither he nor Plaintiffs have satisfactorily explained the economic or legal justification for this position.

Plaintiffs also argue—regardless of whether the constant dollar approach can be used to approximate New York and Ohio's "profits" from the fraud—that it is inappropriate to "net" a plaintiff's alleged profits from any Class Period sale of shares purchased prior to the Class Period against his later losses. Plaintiffs cite a number of cases in which courts conclude that "netting"—or "aggregating"—a plaintiff's unwitting profits from pre-Class Period holdings would not serve the purposes of the securities laws, which is to deter fraud. (Reply at 6-7.) Defendants counter that Plaintiffs have identified the few cases in which "netting" is rejected—none of which is binding—but that the majority view, endorsed by the Fifth Circuit, is that damages from securities fraud should be calculated by netting the gains and losses from the plaintiff's ongoing trading strategy. (Doc. No. 688 ("Sur-Reply"), at 11.)

Even assuming that Defendants are correct that a plaintiff's losses from the alleged fraudulent scheme must be netted against that plaintiff's "profits," the Court finds it unnecessary and inadvisable to attempt such a calculation at this stage. The calculation depends upon variables which are not and cannot be fixed without substantive merits determinations, including: (1) when the "inflationary" period begins (i.e., when the first material and knowingly or recklessly false misrepresentation was made) and (2) what the amount of inflation over the Class Period is.

The two damages approaches proffered by the parties—"netting" versus "transaction-based"—have been endorsed by various courts at various times and are the source of much confusion in this area of the law. See Samuel Francis, Note, Meet Two-Face: The Dualistic Rule 10b-5 and the Quandary of Offsetting Losses by Gains, 77 Fordham L. Rev. 3045, 3061-73 (2009). Defendants indicate that the Fifth Circuit has definitively chosen the "netting" approach, relying upon Wolf v. Frank, 477 F.2d 467 (5th Cir. 1973). Plaintiffs dispute Defendants' interpretation of Wolf. (Reply at 6 n.6.) The Court posits that the law is unsettled in this area precisely because so few securities fraud cases proceed to trial and the determination of damages. It recognizes that, if this case proceeds to summary judgment and trial, it will likely be called upon again to decide whether the "netting" or "transaction-based" approach is appropriate for Plaintiffs' claims. But, as described in the text, it need not wade into these muddy waters to decide the typicality and adequacy of New York and Ohio.

Moreover, the unique defenses indicated by Defendants—absence of injury and lack of standing—can be assessed with a more foundational inquiry. The Supreme Court has indicated that a securities fraud plaintiff has experienced injury if he, at the time of any alleged corrective disclosure, held shares which he purchased during the Class Period at allegedly inflated prices. See Dura Pharms., Inc. v. Broudo, 544 U.S. 336, 342-43 (2005). The Court finds that New York and Ohio meet this standard for injury during the Class Period and thus qualify as typical representatives for the Class regardless of the timing of their purchases. See Kalodner v. Michaels Stores, Inc., 172 F.R.D. 200, 208 (N.D. Tex. 1997) (concluding that a "hypothetical conflict between early and late purchasers does not destroy typicality"). "[W]ere the rule otherwise, there could never be a class action in securities fraud cases because a representative plaintiff would potentially be needed for each day of the class period, since on each day the mix of information available to the public would vary." Feldman v. Motorola, Inc., Civ. A. No. 90 C 5887, 1993 WL 497228, at *6 (N.D. Ill. Oct. 14, 1993).

b. The Ludlow Plaintiffs

Defendants raise no specific typicality arguments regarding the Ludlow Plaintiffs, although they fault the Ludlow Plaintiffs generally for not attempting to establish that the proposed Ludlow Subclass meets each requirement of Rule 23. (Opp. at 20-22.) The Court agrees that the requirements of Rule 23 must be established for each proposed class and subclass. See Fed. R. Civ. P. 23(c)(5) ("[S]ubclasses . . . are each treated as a class under this rule."); see also Simms v. Jones, --- F.R.D. ---, Civil Action Nos. 3:11-CV-0248-M, 3:11-CV-345-M, 2013 WL 3449538, at *9 (N.D. Tex. July 9, 2013) ("When class certification is sought for multiple subclasses, each individual subclass is treated as its own class and must meet the requirements of Rule 23."). It also agrees that Plaintiffs provided scant information in their original motion regarding the need for the Ludlow Subclass or how the Ludlow Subclass independently satisfies Rule 23. Defendants have overstated the latter deficiency, however. Because the Ludlow Subclass Period is wholly subsumed within the Class Period, most of the Rule 23 requirements for the Ludlow Subclass hinge upon the same evidence and the same arguments advanced in support of the Class as a whole. Specifically, in the matter of typicality, the Ludlow Plaintiffs allege that they purchased BP ADSs during the Ludlow Subclass Period which were artificially inflated due to Defendants' misleading statements and omissions, and that they retained those shares until after an alleged corrective disclosure. (Mot. at 6.)

The one exception is Rule 23(a)'s adequacy requirement, which obligates the Court to assess the abilities and knowledge of the proposed class representative. Plaintiffs' failure to substantiate the Ludlow Plaintiffs' adequacy in their opening submission is addressed below in Part V.A.4.b of this opinion.

More troubling than the Ludlow Plaintiffs' failure independently to establish their compliance with the requirements of Rule 23 is their silence as to the need for the Ludlow Subclass at all. As described above, the Ludlow Subclass was instituted at the lead plaintiff stage of this case due to concerns that NY/OH and the Ludlow Plaintiffs intended to pursue fundamentally different theories of securities fraud against Defendants. Since that decision was made, Plaintiffs' claims and the Class Period have been significantly narrowed by the Court's rulings on Defendants' motions to dismiss. Indeed, the live pleading is a joint effort by both NY/OH and the Ludlow Plaintiffs. Counsel for both sets of lead plaintiffs report a productive and amicable working relationship—one to which the Court can attest, and one that has proved beneficial to the prosecution of this case. (Nov. 18, 2013 Hrg. Tr. at 6-7.) In summary, the Court can discern no conflict or competing theory of liability which necessitates the Ludlow Subclass as presently defined. Plaintiffs—including the Ludlow Plaintiffs—do not dispute this analysis and, in fact, have revised their motion to request certification of a single class. (Id. at 7.)

Anticipating this revised request, Defendants argue that the Ludlow Plaintiffs "have never sought to represent" the longer Class Period and would be unable to do so. (Opp. at 20 n.11.) To the extent that Defendants base their argument on the claim that the Ludlow Plaintiffs are "atypical" representatives for the Class, this would seem to contradict their position regarding New York and Ohio. It is undisputed that the Ludlow Plaintiffs were net purchasers of ADSs during the Class Period; therefore, they suffered a loss regardless of whether damages are calculated pursuant to a "netting" theory or a "transaction-based" theory. To the extent that Defendants base their argument on the Ludlow Plaintiffs' previous position—from the lead plaintiff stage of the case—that NY/OH's proposed class period was too long and diluted the class's most meritorious claims, the Court declines to judge the Ludlow Plaintiffs' typicality by reference to a position taken in opposition to a proposed class period and a set of claims which are now obsolete. As noted above, a securities fraud plaintiff need not have purchased his shares in any particular time frame of the alleged fraud to be typical of the class. See Feldman, 1993 WL 497228, at *6. By any relevant measure, the Ludlow Plaintiffs satisfy Rule 23(a)'s typicality requirement for the proposed Class as a whole.

4. Adequacy

Section 4 of Rule 23(a) requires that plaintiffs demonstrate that "the representative parties will fairly and adequately protect the interests of the class." Thus, to assess representativeness, courts must examine "class representatives, their counsel, and the relationship between the two." Berger, 257 F.3d at 479. Class representatives must have sufficient knowledge and understanding so as to be able to control or prosecute the litigation. Feder v. Elec. Data Systems Corp., 429 F.3d 125, 129-30 (5th Cir. 2005). In other words, they must be willing and able "to take an active role in and control the litigation and to protect the interests of absentees." Berger, 257 F.3d at 479 (internal quotation marks omitted). Their counsel should likewise be zealous and competent. Id. (citing Horton v. Goose Creek Indep. Sch. Dist., 690 F.2d 470, 484 (5th Cir. 1982)). Finally, courts should endeavor to uncover conflicts of interest between named plaintiffs and proposed class members. Feder, 429 F.3d at 130. It is commonly acknowledged, however, that "[d]ifferences between named plaintiffs and class members render the named plaintiffs inadequate representatives only where those differences create conflicts between the named plaintiffs' and the class members' interests." Berger, 257 F.3d at 480 (emphasis added).

a. Conflicts of interest

Although typicality and adequacy are separate components of the analysis, they overlap to the extent that a proposed representative subject to unique defenses may prove an inadequate representative for the class. See Beck v. Maximus, Inc., 457 F.3d 291, 301 (3d Cir. 2006) ("A proposed class representative is neither typical nor adequate if the representative is subject to a unique defense that is likely to become a major focus of the litigation."). Defendants argue that this problem exists for Ohio and New York, due to their atypical trading activity and alleged profits from the purported fraudulent scheme. (Opp. at 14-15.) According to Defendants, Ohio and New York have no incentive to establish that the price of ADSs was inflated in the pre-spill period, when Ohio and New York were both net sellers. Put another way, Defendants contend that Ohio and New York have a conflict of interest with class members, due to the timing of their trades, which makes them inadequate class representatives.

To fully appreciate Defendants' argument, it is necessary to conceptualize the types of misrepresentations which have been alleged in this case and how they operate over the span of the Class Period. Defendants are alleged to have misrepresented BP's efforts to improve its process safety processes, representations which—if true—would have had the effect of lowering BP's exposure to catastrophic risk. These misrepresentations took the form of statements regarding the Baker Panel and its recommendations, as well as statements regarding a specific Baker Panel recommendation: OMS. All of these statements took place prior to the April 20, 2010 Deepwater Horizon explosion.

Defendants are also alleged to have misrepresented the extent of the oil spill following the April 20, 2010 explosion. All of these statements took place following the explosion.

It is appropriate, therefore, to view the April 20, 2010 Deepwater Horizon explosion as an inflection point in the Class Period. This is because purchases of ADSs prior to the explosion could not have been motivated by Defendants' misrepresentations regarding the oil flow rate. Conversely, purchases of ADSs subsequent to the explosion were not likely motivated by Defendants' misrepresentations regarding its process safety improvements. Viewing the Class Period as relatively self-contained periods both before and after the Deepwater Horizon explosion, it becomes clear that Defendants have identified a conflict among class members that rises above mere hypothesis or speculation. See Ward v. Dixie Nat. Life Ins. Co., 595 F.3d 164, 180 (4th Cir. 2010) ("[A] conflict will not defeat the adequacy requirement if it is 'merely speculative or hypothetical[.]'") (quoting Gunnells v. Healthplan Services, Inc., 348 F.3d 417, 430 (4th Cir. 2003)); Bradburn Parent/Teacher Store, Inc. v. 3M (Minn. Mining and Manuf. Co.), No. Civ. A. 02-7676, 2004 WL 414047, at *5 (E.D. Pa. Mar. 1, 2004) ("'[M]ost courts hold that [a] conflict [between class members] must be more than merely or hypothetical' before a named representative can be deemed inadequate.") (quoting 5 James Wm. Moore, et al., Moore's Federal Practice 23.25[4][b] (3d ed. 2003)). Specifically, in the creation of a damages model, Plaintiffs will be tasked with addressing how to account for the multiple frauds which they have alleged. Class members who purchased prior to the Deepwater Horizon explosion may very well have interests different from, and possibly competing with, class members who purchased after that event.

Plaintiffs do not concede that the misleading nature of the Baker Panel and OMS statements was fully revealed by the Deepwater Horizon explosion. They contend that the issue of whether any given corrective disclosure is partial or full should be left to a finder of fact. (Reply at 9-11.)

Fortunately, this conflict can be ameliorated by the creation of a "Pre-Explosion Subclass" and a "Post-Explosion Subclass." See In re PaineWebber Ltd. P'ship Litig., 174 F.R.D. 35, 36-37 (S.D.N.Y. 1996) ("[S]ubclasses may be created when there are conflicts of interest among class members."); see also In re Oil Spill by the Oil Rig "Deepwater Horizon" in the Gulf of Mex. on April 20, 2010, 910 F. Supp. 2d 891, 918 (E.D. La. 2012) (recognizing that "[s]ubclassing is . . . one of many available options for limiting the possibility of intraclass conflicts"). Because Ohio held no ADSs at the time of the Deepwater Horizon explosion, it is an inadequate representative for the Pre-Explosion Subclass. Ohio purchased ADSs following the explosion, however, which it held until the end of the Class Period. Therefore, Ohio is a potentially adequate representative for the "Post-Explosion Subclass."

New York also purchased ADSs following the explosion, which it held until the end of the Class Period. But New York, unlike Ohio, held ADSs at the time of the explosion which it had purchased within the pre-explosion portion of the Class Period. Therefore, the Court discerns no conflict or impediment that prevents New York from serving as a representative of the Class throughout the proposed Class Period.

Depending upon whether the LIFO or FIFO accounting method is used, at the time of the explosion, New York held either 21,913 or 192,726 BP ADSs which had been purchased within the Class Period. (Doc. No. 677-2, at 2-3.)

Finally, the Ludlow Plaintiffs purchased ADSs prior to the explosion which they held through the explosion, but they did not purchase any additional ADSs after the explosion. Therefore, the Ludlow Plaintiffs are potentially adequate representatives for the "Pre-Explosion Subclass" only.

b. Other measures of adequacy

The adequacy prong of Rule 23(a) is not limited to the absence of a conflict of interest, however. It is also directed to the ability of the class representative to discharge its role of monitoring and directing the litigation. New York and Ohio are both large, sophisticated institutional investors, with significant assets and in-house resources directed specifically at this litigation and experience in other major securities fraud cases. (Mot. at 7-9.) It is clear that they satisfy this measure of adequacy, as Defendants do not dispute.

In their opposition to class certification, Defendants raise concerns about the Ludlow Plaintiffs' adequacy as class representatives. Defendants highlight that several of the Ludlow Plaintiffs, in their depositions, expressed come confusion about what it means to sue in a representative capacity; what misrepresentations Defendants are alleged to have made; and why the Subclass Period differs from the Class Period. (Opp. at 21 n.12.) The Court has reviewed the relevant portions of the Ludlow Plaintiffs' depositions. It finds that their testimony reflects the very basic knowledge that an individual investor should be expected to have, given the complicated nature of the claims alleged here. See Makor Issues & Rights, Ltd. v. Tellabs, Inc., 256 F.R.D. 586, 601 (N.D. Ill. 2009) ("'[I]t is well established that in complex actions such as securities actions, a plaintiff need not have expert knowledge of all aspects of the case to qualify as a class representative[.]'") (quoting Wagner v. Barrick Gold Corp., 251 F.R.D. 112, 118 (S.D.N.Y. 2008)). Moreover, the personal information regarding the Ludlow Plaintiffs submitted by Plaintiffs in their reply brief satisfies the Court as to their sophistication and ability to participate in this complex securities fraud case. (Reply at 16-17.)

The Court does not condone Plaintiffs' failure to include this information in their opening submissions. Because Defendants have had the opportunity to respond by way of sur-reply, however, the Court will consider the information submitted in the first instance on reply.

More troubling is the fact that the Ludlow Plaintiffs have apparently convened with counsel only once during the course of the litigation, in August 2013. (Opp. at 21 n.12.) This is in contrast to the Ludlow Plaintiffs' collective promise, while vying for Lead Plaintiff position, to meet on a regular basis. (Doc. No. 48-3, at ¶ 4.) The PSLRA was enacted, in part, to combat a model of securities class action in which plaintiffs serve as mere figureheads, ceding the responsibility to monitor and control the litigation to counsel. See Berger, 257 F.3d at 484 ("[I]n complex class action securities cases governed by the PSLRA, the adequacy standard must reflect the governing principles of the Act and, particularly, Congress's emphatic command that competent plaintiffs, rather than lawyers, direct such cases."). The Ludlow Plaintiffs' apparent disengagement from this case is of significant concern to the Court.

Ultimately, however, the Court sees benefit in the inclusion of individual investors in the prosecution of these claims. It is persuaded that the Ludlow Plaintiffs are sufficiently knowledgeable to serve as representatives. And it is heartened that one of the Ludlow Plaintiffs—Les Nakagiri—personally attended the hearing on class certification. Therefore, while not without reservations, the Court concludes that the Ludlow Plaintiffs are adequate representatives for the Pre-Explosion Subclass.

Finally, adequacy must be assessed for the proposed class representative's counsel. New York is represented by the firm Cohen Milstein Sellers & Toll PLLC; Ohio is represented by the firm Berman deValerio; and the Ludlow Plaintiffs are represented by Cotchett, Pitre & McCarthy, LLP and the Mithoff Law Firm. All of these firms have considerable experience prosecuting class action securities fraud cases, as their respective firm resumes attest. (Doc. No. 652-6; Doc. No. 652-7; Doc. No. 652-8; Doc. No. 652-9.) Additionally, the Court has had the benefit of observing the work performed by Plaintiffs' counsel during the course of this case to date. The work has been more than adequate, as Defendants concede by their silence.

B. Rule 23(b)(3)

1. Predominance

"To gain class certification under Rule 23(b)(3), a proposed class must satisfy Rule 23(a) and '[c]ommon questions must predominate over any questions affecting only individual members, and class resolution must be superior to other available methods for the fair and efficient adjudication of the controversy.'" Maldonado, 493 F.3d at 525 (quoting Amchem, 521 U.S. at 615) (internal quotation marks omitted). The predominance inquiry is "more demanding than the commonality requirement of Rule 23(a), and as such, mandates caution, particularly where 'individual stakes are high and disparities among class members great.'" Bell Atl. Corp. v. AT&T Corp., 339 F.3d 294, 301-02 (5th Cir. 2003) (quoting Amchem, 521 U.S. at 625). Defendants argue that the predominance requirement is not met with regards to one category of alleged misstatements—those found within Defendant BP Exploration & Production's regulatory filings with the MMS—because individualized questions of reliance will overwhelm the case. They also argue that the predominance requirement is not met for any of Plaintiffs' claims because they have not established that damages are capable of class-wide measurement.

a. Reliance

Predominance is a test readily met in cases alleging securities fraud. See Amchem, 521 U.S. at 625. This is largely due to the fraud-on-the-market presumption, which excuses securities fraud plaintiffs from establishing individual reliance on an alleged misrepresentation in certain circumstances. See Smilovits v. First Solar, Inc., --- F.R.D. ---, No. CV12-00555-PHX-DGC, 2013 WL 5551096, at *3 (D. Ariz. Oct. 8, 2013).

The fraud-on-the-market presumption theorizes that, when securities are sold on an open and efficient market, public and material information about the company is quickly incorporated into the price of the security, and each investor who trades in the security thereafter is "defrauded" to the extent that the information was false and artificially inflated the price of the security beyond its true value. See Basic Inc. v. Levinson, 485 U.S. 224, 244, 247 (1988). Although there had been some doubt in the past regarding what must be shown at the certification stage to invoke the fraud-on-the-market presumption, the Supreme Court has clarified that a proposed class representative need not establish materiality or loss causation to invoke the presumption and thereby satisfy Rule 23(b)(3) predominance. See Amgen Inc. v. Conn. Retirement Plans and Trust Funds, 133 S. Ct. 1184, 1195-99 (2013) (explaining that materiality need not be shown at class certification because it is an "objective" question with a "class-wide" answer, therefore necessarily satisfying Rule 23(b)(3) predominance); Erica P. John Fund, Inc. v. Halliburton Co., 131 S. Ct. 2179 , 2186 (2011) (explaining that loss causation need not be shown at class certification because it is "has no logical connection to the facts necessary to establish the efficient market predicate to the fraud-on-the-market theory"). Rather, the inquiry at this stage is focused on "trade timing, market efficiency, and publicity[.]" Erica P. John Fund, Inc. v. Halliburton, Inc., 718 F.3d 423, 432 (5th Cir. 2013), cert. granted by Halliburton Co. v. Erica P. John Fund, Inc., No. 13-317, 2013 WL 4858670, at *1 (U.S. Nov. 15, 2013).

For clarity, the applicability of the fraud-on-the-market presumption is vital to class certification because otherwise individualized questions of class members' reliance would overwhelm common questions regarding Defendants' liability, and Rule 23(b)(3) predominance would not be met. See Basic Inc. v. Levinson, 485 U.S. 224, 242 (1988); Regents of the Univ. of Cal. v. Credit Suisse First Boston (USA), Inc., 482 F.3d 372, 383, 394 (5th Cir. 2007).

Defendants dispute the applicability of the fraud-on-the-market presumption with regards to alleged misrepresentations found in two MMS regulatory filings: the IEP and OSRP. They claim that Plaintiffs have failed to establish two prerequisites necessary to invoke the presumption: publicity and market efficiency. Because the Court finds Defendants' publicity argument persuasive, it will not address market efficiency.

Defendants contend that it is appropriate to treat each alleged misstatement as a separate claim and analyze predominance accordingly. (Opp. at 24-25.) Plaintiffs have not argued that this approach is improper.

To invoke fraud-on-the-market, Plaintiffs "must demonstrate that the alleged misrepresentations were publicly known (else how would the market take them into account?)[.]" Halliburton, 131 S. Ct. at 2185. Defendants contend that the OSRP was not accessed by any member of the public until after the oil spill, and that the IEP was accessed by only a small handful of people (approximately 22) prior to the spill. (Opp. at 27-32.) They submit affidavits from representatives of the successor agency to MMS and a forensic expert in order to establish these facts. (Doc. No. 664-12 (Miller Decl.); Doc. No. 664-13 (FOIA Chart); Doc. No. 664-15 (Schoennagel Decl.); Doc. No. 664-16 (McGurk Aff.).) According to Defendants, because the preponderance of the evidence does not support that the IEP and OSRP were publicized or otherwise publicly known before the oil spill, the alleged misstatements contained within those documents could not have been incorporated into the ADS price, and each plaintiff will be required to establish his or her own reliance on the documents. (Opp. at 23-32.)

Plaintiffs have very weak retorts to Defendants' analysis. First, Plaintiffs rely on their complaint allegations that the IEP was "available to the public" and that the OSRP was "publicly filed." (SAC ¶¶ 337, 349.) They note that the Court rejected Defendants' argument at the motion to dismiss stage. (Mot. at 18.) But a motion to dismiss must be decided on the allegations in the pleadings, while at class certification the Court must look beyond the pleadings to determine whether a preponderance of the evidence indicates that common questions of fact or law will predominate over individualized questions during the course of the litigation. See Comcast Corp. v. Behrend, 133 S. Ct. 1426, 1432 (2013) (noting that district court faced with motion to certify a class under Rule 23(b)(3) must "take a 'close look' at whether common questions predominate over individual ones") (quoting Amchem, 521 U.S. at 615); see also Wal-Mart, 131 S. Ct. at 2551 ("Rule 23 does not set forth a mere pleading standard."). This more searching standard indicates that Defendants have raised a valid objection to Plaintiffs' invocation of the fraud-on-the-market presumption.

Plaintiffs also claim that Defendants publicized the OSRP following the oil spill by releasing a statement regarding the resources BP had marshaled to combat the spill. (Mot. at 19.) The press release cited, dated April 22, 2010, contained none of the specific statements of fact from the OSRP which the Court previously found actionable as fraud. As such, it does not assist with the publicity element for fraud-on-the-market.

Defendants contend—and Plaintiffs have not shown otherwise—that the alleged misrepresentations from the IEP and OSRP were first publicized in news articles from late April and early May 2010, respectively. (Opp. at 28-29, 31.) Defendants further argue that, because the new articles were critical and cast doubt on the accuracy of the purported misrepresentations, they could not have inflated the price of BP's ADSs. (Id. at 29, 31.)

Interestingly, Plaintiffs do not respond to the publicity problem in their reply briefing. They criticize Defendants' market efficiency argument. They claim that Defendants have not rebutted the fraud-on-the-market presumption because their arguments are about materiality. (Reply at 19.) But they do not suggest Defendants have incorrectly stated the law regarding publicity; they do not dispute Defendants' conclusions about publicity; and they do not offer any rebuttal evidence regarding publicity. In short, Plaintiffs have all but conceded that they have no evidence that the IEP and OSRP misstatements were known by the market and incorporated in the ADS price prior to the Deepwater Horizon explosion. As such, the fraud-on-the-market presumption cannot be invoked, and Plaintiffs' claims based on these alleged misstatements are not appropriate for class action treatment.

b. Damages

Plaintiffs indicate that class members' damages will be calculated by use of an event study. (Mot. at 20.) "An event study is a statistical regression analysis that examines the effect of an event on a dependent variable, such as a corporation's stock price." In re Imperial Credit Industries, Inc. Securities Litig., 252 F. Supp. 2d 1005, 1014 (C.D. Cal. 2003). Event studies are commonly used in securities fraud class actions. See In re Diamond Foods, Inc. Securities Litig., --- F.R.D. ---, 2013 WL 1891382, at *12 (N.D. Cal. May 6, 2013).

Defendants dispute that Plaintiffs' assertion as to what they will do satisfies their burden on class certification. They argue that Plaintiffs have not established by a preponderance of the evidence that damages can be calculated on a class-wide basis using a model consistent with their theory of liability. Defendants base their argument on Comcast Corp. v. Behrend, 133 S. Ct. 1426 (2013).

The facts in Comcast involved Comcast's practice of "clustering" by swapping infrastructure in some geographical areas with the infrastructure of competitors in other geographical areas. This practice permitted Comcast to achieve dominance in certain geographical markets. The market at issue in the case was the Philadelphia "Designated Market Area" ("DMA"). As a result of clustering, Comcast's market share in the Philadelphia DMA went from 23.9% to 69.5% over a 10-year period. See 133 S. Ct. at 1430.

Plaintiffs brought antitrust claims against Comcast, theorizing that clustering had "harmed subscribers in the Philadelphia cluster by eliminating competition and holding prices for cable services above competitive levels." Comcast, 133 S. Ct. at 1430. Plaintiffs posited that Comcast's anti-competitive behavior resulted in higher prices through four different mechanisms of action, including "reduc[ing] the level of competition from 'overbuilders,' companies that build competing cable networks in areas where an incumbent cable company already operates." Id. at 1430-31. The trial court determined that this "overbuilder-deterrence" theory was the only theory capable of class-wide proof and limited certification to this theory. Id. at 1431.

In support of their motion for certification, plaintiffs submitted the expert report of Dr. James McClave. Dr. McClave had run a regression model comparing actual cable prices in the Philadelphia DMA with "but for" prices derived from benchmark counties outside Philadelphia. See Comcast, 133 S. Ct. at 1431, 1438-39. "Dr. McClave acknowledged . . . [that] the model did not isolate damages resulting from any one theory of antitrust impact." Id. at 1431. As noted above, three of the Plaintiffs' four posited theories had been rejected for class treatment. As a result, the Supreme Court concluded that Dr. McClave's model did not "measure only [the] damages attributable" to the "overbuilder-deterrence" theory and therefore could not "possibly establish that damages are susceptible of measurement across the entire class for purposes of Rule 23(b)(3)." Id. at 1433. The Court further emphasized that it is not sufficient for class action plaintiffs to provide a classwide damages model; they must also show that the damages model is "'consistent with [their] liability case[.]'" Id. (citation omitted). Without this second step, Rule 23(b)(3) predominance would be "reduce[d] . . . to a nullity." Comcast, 133 S. Ct. at 1433.

In this case, Defendants argue that Plaintiffs' proposed event study methodology will violate the Comcast requirement that class-wide damages must hew to Plaintiffs' theories of liability. To illustrate the potential disconnects between an event study and the multiple frauds alleged in this case, Defendants' expert, Dr. Stulz, created an event study similar to the type used by Plaintiffs' expert, Mr. Chad Coffman, in other securities fraud cases. Dr. Stulz's event study assumes a "constant dollar" methodology of damages, and does not disaggregate inflation from multiple alleged disclosures. Thus, assuming an initial misrepresentation on the first day of the Class Period and eight corrective disclosure dates as identified by Mr. Coffman in one of his reports, Dr. Stulz's event study returns $27.26 of inflation, beginning on November 8, 2007 and continuing throughout most of the Class Period. (Doc. No. 689-1, at 2-3.) Defendants assert that Dr. Stulz's event study is substantially similar to one that Plaintiffs will attempt to use at trial. Defendants have identified multiple ways in which this damages methodology is inconsistent with Plaintiffs' theories of liability, including:

As corrective disclosures occur, this inflation slowly dissipates from the stock price, until the amount of inflation at the time of the last corrective disclosure is $0.

(1) Although the pre-spill misrepresentations understated a known risk, the "constant dollar" approach compensates investors for the full value of the stock price drop from the materialization of that risk. This overcompensates investors for their harm. (Stulz Report at ¶¶ 58-62.)
(2) The model does not disaggregate "inflation" according to the type of misrepresentation corrected or risk disclosed. This means that investors may receive a measure of damages attributed to misrepresentations which could not have influenced their purchase. (Sur-Reply at 8-9; Doc. No. 6881 ("Stulz Sur-Reply"), at ¶ 12.)
(3) The amount of inflation remains constant over the Class Period. This means that the repetition of certain statements—particularly, the Baker Report and OMS statements—is not shown to have had any cumulative inflationary effect, and that the amount of damages to be awarded will be similarly unresponsive to the jury's specific findings of liability. (Sur-Reply at 4-6.)

To explain this phenomenon, Dr. Stulz offers the following simplified example:

Imagine that a company announced that it was going to draw a marble from an urn of 100 marbles, of which 99 were black and one was red. If the company drew a red marble, it would have to pay $1 million. Prior to finding out the outcome, the company's market value would reflect the expected loss from this lottery of 1% of $1 million, or $10,000. If the company subsequently drew a red marble, the market value would have fallen $990,000 to reflect the new information—the certainty of a $1 million loss. If, however, contrary to the company's statement, there were two red marbles (increasing the probability of drawing a red marble), the share price would still have fallen when the company drew a red marble. In order to understand the value implication of the company's misstatement that there was only one red marble, the relevant issue is what the market value would have been, prior to the drawing, had the company told the truth. In this case, the market value would have reflected an expected loss of $20,000, only $10,000 lower than the actual market value, not the $990,000 less that would be implied by looking at the reaction to the drawing of a red marble.

Plaintiffs respond to the above criticisms by noting that the event study which prompted them was performed by Defendants, not Plaintiffs. (Nov. 18, 2013 Hrg. Tr. at 99, 103-04.) They also claim that their event study—which, again, has not yet been created—will provide options to the jury to disaggregate inflation by the type of misrepresentation, should the jury agree with Defendants that the misrepresentations were not part of a single, cohesive fraudulent scheme. (Id. at 11-14.) Of course, this feature would only address the second of the criticisms noted above; Plaintiffs tacitly concede that they have no plans to address the first or third.

These retorts only crystallize Defendants' point that Plaintiffs have failed to meet their burden of showing that damages can be measured on a class-wide basis consistent with their theories of liability. The Court agrees with Defendants that Comcast signals a significant shift in the scrutiny required for class certification. (Nov. 18, 2013 Hrg. Tr. at 122-23 (analogizing Comcast to the more demanding judicial scrutiny of notice pleading announced by Bell Atl. Corp. v. Twombly, 550 U.S. 544 (2007)).) Prior to Comcast, the Court may have been satisfied that Plaintiffs' invocation of the event study methodology alone showed the predominance of common issues. This is because the need to perform individual calculations of damages will not overwhelm common issues of liability when those calculations are "mathematical and formulaic." See Bell Atl., 339 F.3d at 307. Simply invoking the event study methodology may have been sufficient to indicate that this case would fall into the category of cases in which individual damages calculations are "mathematical or formulaic." In other words, it likely satisfies the first half of Comcast's test: that damages be measurable on a class-wide basis. But it does not satisfy the second half of that test. It does not assuage the Court that the class-wide damages methodology proposed will track Plaintiffs' theories of liability, as the Supreme Court expressly required in Comcast before a class may be certified. See 133 S. Ct. at 1433.

Following Comcast, circuit and district courts have rigorously examined proposed damages methodologies in putative class action cases for disconnects between damages and liability. Some of these examinations have resulted in denials of class certification. See, e.g., Turnbow v. Life Partners, Inc., Civil Action No. 3:11-cv-1030-M, 2013 WL 3479884, at *15-18 (N.D. Tex. July 9, 2013); see also In re Rail Freight Fuel Surcharge Antitrust Litig.—MDL No. 1869, 725 F.3d 244, 253-55 (D.C. Cir. 2013) (vacating district court's decision to grant class certification and remanding for consideration of whether plaintiffs' proposed damages model satisfies Comcast). Plaintiffs cannot avoid this hard look by refusing to provide the specifics of their proposed methodology. See In re Montano, 493 B.R. 852, 860 (Bankr. D.N.M. 2013) ("If in fact Class 2 damages could be measured class-wide, Plaintiffs had an obligation to come forward with evidence thereof. They did not, and Comcast does not allow them the luxury of waiting until trial."). Without a more complete explication of how Plaintiffs propose to use an event study to calculate class members' damages, and how that event study will incorporate—and, if necessary, respond to—the various theories of liability, the Court cannot certify this litigation for class action treatment.

Specifically, the Court notes the possibility that some of Defendants' alleged misrepresentations may not survive summary judgment, or may not be found meritorious at trial. Comcast suggests that a damages model which cannot accommodate variations in the liability findings may be grounds for decertification at a subsequent stage of the case.
--------

2. Superiority

According to Plaintiffs, resolving the question of Defendants' liability and the right of approximately 900,000 ADS purchasers to receive damages in this consolidated class action is vastly superior to the pursuit of hundreds of thousands of individual actions. (Mot. at 21.) This fact is often acknowledged in securities fraud cases. See, e.g., In re Deutsche Telekom Ag Securities Litig., 229 F. Supp. 2d 277, 282 (S.D.N.Y. 2002) ("Class actions are generally well-suited to securities fraud cases such as this one because they avoid the time and expense of requiring all class members to litigate individually. Given the large number of potential plaintiffs and the commonality of their claims, certifying the class will permit a fairer and more efficient adjudication of the controversy than requiring individual adjudications would.").

If Plaintiffs had established that damages can be calculated on a class-wide basis consistent with their theories of liability, the Court would be inclined to agree that the superiority requirement of Rule 23(b)(3) has been met. But such a conclusion is impossible without assurances that Plaintiffs' class-wide damages methodology will accurately reflect any liability findings in this case.

VI. CONCLUSION

It is ordered that Plaintiffs' Motion for Class Certification (Doc. No. 652) is DENIED. Plaintiffs have failed to discharge their burden to establish that damages in this case can be measured on a class-wide basis consistent with their theories of liability. Because the Court's ruling is based in large part on a recent Supreme Court decision that—in this Court's opinion—has appreciably changed the landscape for class certification, the Court finds it would be inequitable to disallow Plaintiffs a second attempt to establish the elements necessary for class action treatment. Therefore, should Plaintiffs wish to supplement and re-urge their Motion, they are instructed to do so in conformity with the rulings expressed in this Order within 30 days.

IT IS SO ORDERED.

SIGNED at Houston, Texas, on this the sixth day of December, 2013.

_____________

KEITH P. ELLISON

UNITED STATES DISTRICT JUDGE

(Stulz Report at ¶ 59.)


Summaries of

In re BP P.L.C. Sec. Litig.

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF TEXAS HOUSTON DIVISION
Dec 6, 2013
MDL NO.: 10-md-2185 (S.D. Tex. Dec. 6, 2013)

noting "statistical regression analys[e]s that examine the effect of an event on a dependent variable, such as a corporation's stock price," are "commonly used in securities fraud class actions"

Summary of this case from Sec. & Exch. Comm'n v. World Tree Fin.
Case details for

In re BP P.L.C. Sec. Litig.

Case Details

Full title:IN RE: BP p.l.c. SECURITIES LITIGATION

Court:UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF TEXAS HOUSTON DIVISION

Date published: Dec 6, 2013

Citations

MDL NO.: 10-md-2185 (S.D. Tex. Dec. 6, 2013)

Citing Cases

City of Cape Coral Mun. Firefighters' Ret. Plan v. Emergent Biosolutions, Inc.

Comcast , 569 U.S. at 35, 133 S.Ct. 1426. In In re BP p.l.c. Securities Litigation (BP I) , No. 10-MD-2185,…

Sec. & Exch. Comm'n v. World Tree Fin.

Statistical evidence can be useful in securities cases, and its admission is no novelty.See, e.g., United…