Searching over 5,500,000 cases.

Buy This Entire Record For $7.95

Download the entire decision to receive the complete text, official citation,
docket number, dissents and concurrences, and footnotes for this case.

Learn more about what you receive with purchase of this case.

Klein v. Td Ameritrade Holding Corp.

United States District Court, D. Nebraska

September 14, 2018

GERALD J. KLEIN, on behalf of himself and all similarly situated; Plaintiff,


          Joseph F. Bataillon, Senior United States District Judge.

         This matter is before the court on an objection filed by lead plaintiff Roderick Ford, [1] Filing No. 234, to the Findings and Recommendation (“F&R”) of the United States Magistrate Judge, Filing No. 233, on the lead plaintiff's motion for class certification, appointment of class representative, and appointment of class counsel, Filing No. 187. This is a putative class action filed by retail equity traders harmed by uniform order routing practices implemented by defendants TD Ameritrade Holding Corporation, TD Ameritrade, Inc., and Frederic Tomczyk (collectively, “Defendant” or “TD Ameritrade”). The plaintiff contends that defendant's order routing practices involve use of computer algorithms to send its customers' equity orders to venues that pay the defendant the most money, without regard to whether the venues would provide the best possible execution of those orders. The plaintiff further alleges defendant failed to disclose the practice to its customers. The plaintiff contends the practices are inconsistent with the defendant's duty of best execution and seeks remedies for violations of Sections 10(b) and 20(a) of the Securities Exchange Act of 1934, 15 U.S.C. § 78a et seq. The plaintiff contends that the purported class suffers economic loss due to orders for securities trades being unfilled, underfilled, filled at a suboptimal price, and/or filled in a manner that adversely affects the order's performance post-execution. He seeks, on behalf of the class, damages and injunctive relief.

         Plaintiff seeks certification of a class defined as follows:

[A]ll clients of TD Ameritrade between September 15, 2011 and September 15, 2014 who placed orders that did not receive best execution, in connection with which TD Ameritrade received either liquidity rebates or payment for order flow, and who were thereby damaged (the “Class”).

(Filing No. 187.). He also seeks certification of a class for injunctive relief under Federal Rule of Civil Procedure 23(b)(2):

[A]ll clients of TD Ameritrade between September 15, 2011 and September 15, 2014 who placed orders in connection with which TD Ameritrade received either liquidity rebates or payment for order flow and who continue to be clients of TD Ameritrade (the “Injunctive Class”).

         The only relief sought by the purported Rule 23(b)(2) class is an injunction requiring defendant to change its common order routing practices. Alternatively, the plaintiff seeks class certification of a limited issue class on liability under Rule 23(c)(4).

         In opposition to the plaintiff's motion for class certification, the defendant argues that proof of economic loss and reliance will require extensive individualized inquiries into evidence specific to each class member and each of their orders. Solely for purposes of the class certification motion, TD Ameritrade does not dispute that Ford's allegations regarding its order routing policies-and TD Ameritrade's substantial defenses to those allegations-could be proven or disproven with class-wide evidence. See Filing No. 194, Defendant's Brief at 4. It contends however, that common evidence cannot prove the required elements of economic loss and reliance and the individualized inquiries relating to those elements predominate.

         After a hearing on March 27, 2018, the Magistrate Judge recommended that the plaintiff's motion be denied. Filing No. 233, F&R at 1. The Magistrate Judge stated that “[p]laintiff has not shown that he can prove economic harm on a class-wide basis through use of a tested, complete algorithm.” Id. at 8. She also found “there are certain aspects necessary to evaluate economic loss in a best execution case that simply cannot be captured through algorithmic analysis[, ]” namely, “an individual's state of mind or investment strategy.” Id. at 9. She found that “[g]iven the individual, order-by-order inquiries necessary to determine whether each individual customer actually sustained an economic loss, Rule 23(b)(3)'s predominance requirement has not been satisfied.” Id. She also found the proposed class fails to satisfy Rule 23(b)(3)'s superiority requirement, finding resolution of the claims through a class action would create case management issues and would be an inefficient allocation of judicial resources. Id. at 10. Further, she stated, in light of those individual inquiries, a limited certification under Federal Rule of Civil Procedure 23(c)(4) would not promote efficiency or judicial economy. Id. at 13.

         The lead plaintiff objects to the recommendation. Filing No. 234. He reasserts the arguments made to the Magistrate Judge and maintains that class certification is appropriate based on the common question of whether the defendant sought to maximize order flow revenue in such a manner that caused a failure to satisfy the duty to provide best execution of their clients' trades.

         I. BACKGROUND

         A. Procedural History

         The procedural history of the action is relevant to the court's determination. Earlier in this action, the court denied defendant's motion to dismiss the plaintiff's federal securities fraud claim, finding that the presumption of reliance established in Affiliated Ute Citizens of Utah v. United States, 406 U.S. 128, 153-54 (1972), applies to the plaintiff's claims. Filing No. 104, Memorandum and Order at 17; see Zola v. TD Ameritrade, Inc., 172 F.Supp.3d 1055, 1061 (D. Neb. 2016), appeal dismissed (May 18, 2016), aff'd, 889 F.3d 920 (8th Cir. 2018) (affirming dismissal of state law claims under the Securities Litigation Uniform Standards Act of 1998 (“SLUSA”), 15 U.S.C. § 78bb(f)(1), and noting that the gravamen of the plaintiff's claims involves a misrepresentation or omission of a material fact in connection with the purchase or sale of a covered security); see also Newton v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 259 F.3d 154, 174-75 (3d Cir. 2001), as amended (Oct. 16, 2001) (applying Affiliated Ute presumption of reliance to class claims alleging fraudulent acceptance by brokers of orders under duty of best execution).

         Also, the court sustained the defendant's motion for a protective order and denied the plaintiff's request for discovery of extensive class-wide trading records. Filing No. 163, Order. In moving to limit discovery to representative order data, the defendant made the following argument:

Economic loss is an essential element of liability under Section 10(b), but Lead Plaintiff does not need to prove or even “approximate” it for all class members prior to class certification. That is a merits inquiry. Rather, at this stage of the proceedings, Lead Plaintiff must show that he has a valid methodology for proving economic loss at the merits stage by common proof that does not vary by class member. If he cannot make that showing by reference to more than 10, 000 equity orders, he will not be helped by more trade data. Further, if Lead Plaintiff required sufficient discovery to prove economic loss at trial before the Court could rule on class certification, it would entirely subsume the merits inquiry into the class certification inquiry and render meaningless the Court's limitation on discovery (as well as Fed.R.Civ.P. 23(c)(1)(A) and 26(b)(1)).

Filing No. 155, Defendant's Brief at 4-5 (emphasis in original, footnote omitted).

         Based on that line of reasoning, and on the defendant's concession, “for purposes of the class certification process, [of] many of the plaintiff's elements of proof for class certification, ” Magistrate Judge Thalken granted the defendants request for a protective order. Filing No. 163, Order at 3-4 (noting that “[TD Ameritrade] contends the chief issues for class certification are economic loss and reliance”). The Magistrate Judge limited discovery, at that point in the litigation, to trading records of representative accounts. Id. In response to the plaintiff's objections to the Magistrate Judge's order, the defendant stated “[t]he question of whether economic loss can be assessed on a common basis for the entire class can be resolved by consideration of Lead Plaintiff's equity order data, and nothing else.” Filing No. 171, Defendant's Brief at 4-5. This court affirmed the Magistrate Judge's order, finding that the plaintiff had not refuted TD Ameritrade's contention that “the question of whether economic loss can be assessed on a common basis for the entire class can be resolved by consideration of the representative plaintiffs' equity order data” and agreeing that the plaintiff had not demonstrated that he could not “make a showing of economic loss from a representative sample.” Filing No. 177, Order at 4.

         The court later overruled the defendant's motion to exclude the testimony of the plaintiff's experts Haim Bodek and Dr. Shane Corwin under Daubert v. Merrell Dow Pharms., 509 U.S. 579, 589 (1993). Filing No. 208, Memorandum and Order at 7-8. The court found that “[b]oth experts appear to be qualified to testify” and that the experts opinions were “based on methodology that appears reliable relevant-similar methods have been used in other securities cases.” Id. at 7.

         B. Facts

         For purposes of this motion, defendant concedes that the securities fraud elements of misrepresentation or omission and scienter are capable of class-wide proof. Proof of the elements of reliance and economic loss are the subject of the parties' dispute on class-action status.

         The record shows that Dr. Shane Corwin and Haim Bodek are experts in the field of financial markets. Bodek and Dr. Corwin both describe how an out-of-pocket loss can be measured for each TD Ameritrade customer. See Filing No. 189-2, Ex. 2, Declaration of Haim Bodek (“Bodek Decl.”) at 8-14; Filing No. 189-3, Ex. 3, Expert Report of Shane A. Corwin (“Corwin Report”) at 4-5. Bodek performed an analysis of the orders placed by plaintiff Roderick Ford and former plaintiff Kwok L. Shum. Filing No. 189-2, Bodek Decl. at 6. He also utilized historical market data obtained from Thesys Technologies LLC (“Thesys”) and International Continental Exchange, Inc. (“ICE”), and various derived and enhanced data, including, for example, a “fast” version of the National Best Bid or Offer (“NBBO”), [2] as part of his analysis. Id. Derived data includes, for example, data that is based on the Kwok and Shum order data and market data, such as data indicating whether an order was (i) afforded price improvement, meaning that it received a price better than the NBBO, or (ii) subject to a trade-through, meaning that it was executed at a price inferior to the NBBO. Id. The evidence presented by the plaintiff shows the experts used computer analysis and metrics that are widely used and accepted in the industry, by regulators such as the Financial Investments Regulatory Agency (“FINRA”) and the Securities Exchange Commission (“SEC”) and in academic studies to examine the execution quality of securities orders.

         Bodek states that the duty of best execution requires that brokers seek the best price, including the chance to obtain a better price than the NBBO. Id. at 4. At the hearing, he testified he had enough order data to have confidence in the statistical soundness of his conclusions. Filing No. 229, Transcript (“Tr.”) at 75. He testified that his analysis of Shum's and Ford's trading data showed that both had been harmed by TD Ameritrade's order-routing practices. Id. at 78-80.

         Dr. Corwin described Bodek's methodology as follows:

The methodology involves three stages. The first stage involves the initial application of three metrics to calculate the harm associated with failure to provide best execution, the second stage uses an order book analysis to analyze whether the orders-whether orders could have been executed if routed differently, and the third stage uses a series of exclusions to eliminate cases where the harm may not be attributable to the actions of TD Ameritrade.[3]

Id. at 32. He further explained that “[t]he order book analysis is performed by an algorithm which takes into account order data from all of the different venues as well as the trade data and order data from TD Ameritrade.” Id. at 45. Bodek's findings were consistent with the plaintiff's theory that defendant TD Ameritrade routed orders to venues that offered the greatest order routing payments rather than venues that were likely to offer the best execution quality. Bodek concluded that:

[the] failure of the duty of best execution was endemic in TD Ameritrade's order routing practices, resulting in systematic economic harm in the form of exposure to impermissible trade-throughs, increased adverse selection, and lost opportunity costs, and that the magnitude of these harms could be calculated for each affected TD Ameritrade customer through algorithmic analysis.

Filing No. 189-2, Bodek Decl. at 14.

         At the hearing, Bodek stated that he would need all customer orders to properly run an order book analysis without over counting the harm. Filing No. 229, Tr. At 114. He also stated that his algorithm and modern computing could handle all the data involved in this litigation. Id. at 114. The order book analysis Bodek proposes would involve hundreds of millions of data points over a multi-year period. Id. at 107. He agreed with Dr. Kleidon that any reliable analysis of economic loss would need to take into account certain exclusions, exceptions or exemptions (including whether the security was a non-U.S. security, whether the order was subject to one or more of the several trade through exemptions in Rule 611 of Regulation NMS, whether the order was placed outside the market's regular hours, and whether an order was routed during a trading halt or during a fast market, or was part of an oversized order exclusion). Id. at 108, 110. Bodek testified that all of the exclusions could be excluded by his code in the same manner that he filtered other exclusions. Id. at 99; see also Filing No. 189-2, Ex. 2., Bodek Decl. at 16-18.

         Bodek's initial findings concerning overall execution quality and harm were supported by an additional analysis by Dr. Corwin showing that the sample orders received substantially inferior price improvement and by Dr. Corwin's conclusion that TD Ameritrade's order routing practices did not comply with its duty of best execution. Id. at 55-56. He testified that he determined that Bodek's methodology could be used to identify best execution failures and the associated harm with customer orders, and the methodology could be used as a basis to form a damages model. Id. at 18. He also stated that although he had not seen the computer code associated with Bodek's algorithm, he reviewed and verified the output with respect to Shum's trading data. Id. at 56.

         Corwin testified that “that best execution as a responsibility applies to each order individually” and that a “customer's individual trading strategy or investment decisions either before or after each order has no impact on the assessment of harm related to providing best execution on the original order.” Id. at 42-43. Dr. Corwin explained that a full list of exclusions necessary to yield precise damage figures would require discovery that was not provided to the plaintiff. Id. at 46. (“The best place to determine an appropriate list of exclusions would be to understand TD Ameritrade's routing systems themselves, because those systems and the evaluation of execution quality related to that would have to take those exclusions into consideration. . . . We've not been given that information.”).

         Defendant's expert Dr. Allan Kleidon criticized Bodek's methodology. He testified that “the algorithm that has been proposed by plaintiffs and Mr. Bodek here in this matter does not determine whether there's economic loss for everybody in the class, in the putative class, or by how much in this particular matter.” Id. at 145. He stated that “the methodology proposed by Mr. Bodek does not demonstrate that it is possible to identify economic loss for each individual member of the putative class absent individualized inquiry, and, in fact, in my opinion, such individualized inquiry would be necessary to establish ...

Buy This Entire Record For $7.95

Download the entire decision to receive the complete text, official citation,
docket number, dissents and concurrences, and footnotes for this case.

Learn more about what you receive with purchase of this case.