In Re: Checking Account Overdraft Litigation

RESPONSE in Support re MOTION to Strike Testimony From Plaintiffs' Expert Witness Arthur Olsen

S.D. Fla.June 11, 2012

sf-3152700 1 UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF FLORIDA CASE NO. 1:09-MD-02036-JLK IN RE: CHECKING ACCOUNT OVERDRAFT LITIGATION MDL No. 2036 THIS DOCUMENT RELATES TO: Steen v. Capital One, N.A. E.D. La. Case No. 2:10-cv-01505-JCZ-KWR S.D. Fla. Case No. 1:10-cv-22058-JLK DEFENDANT CAPITAL ONE, N.A.’S REPLY IN SUPPORT OF MOTION TO EXCLUDE TESTIMONY FROM PLAINTIFFS’ EXPERT WITNESS ARTHUR OLSEN I. INTRODUCTION In their motion for class certification, Plaintiffs rely upon Arthur Olsen to try to show that it is possible to develop and apply an automated process to Capital One’s customer transaction data that will identify those customers who have allegedly been harmed by Capital One’s posting order policies and, consequently, will identify those customers who are members of Plaintiffs’ proposed Class. As Capital One showed in its opening brief, Olsen’s testimony fails in several significant respects to meet the standard required of expert witnesses by Federal Rule of Evidence 702 and Daubert v. Merrell Down Pharmaceuticals. Rather than address Capital One’s concern that Olsen has not articulated any realistic means of addressing the undisputed gaps in Capital One’s customer transaction data, Plaintiffs’ opposition simply repeats the empty promise that Olsen can analyze the data here because he analyzed the data in Gutierrez. Rather than try to cure the many deficiencies Capital One identified in Olsen’s proposed methodology for determining “harm” (such as the failure to account for NSF fees and fee reversals), Plaintiffs simply urge the Court to ignore them. Because Plaintiffs have not met their burden to show that Olsen’s testimony is admissible under Daubert, the testimony must be excluded. Perhaps recognizing that Olsen’s testimony is not capable of doing what Plaintiffs proffered it to do, Plaintiffs’ opposition attempts to change course and claim that Olsen has been Case 1:09-md-02036-JLK Document 2751 Entered on FLSD Docket 06/11/2012 Page 1 of 11 2 sf-3152700 tasked only with applying predetermined formulas to preexisting data, and not with determining which customers have been harmed by Capital One’s posting policies. This new position contradicts Plaintiffs’ statements about the purpose of Olsen’s testimony set forth in their class certification motion. Even if this were the case, however, Olsen’s testimony must still be rejected. Daubert requires that expert testimony assist the trier of fact in determining a fact at issue. If Olsen’s testimony cannot assist the Court to determine the issue at hand – whether an automated process can be developed to identify those customers who were harmed by Capital One’s posting policies – the testimony will not assist the Court in deciding Plaintiffs’ class certification motion. Regardless of the purpose for which it is offered, Olsen’s testimony is inadmissible under Daubert and must be excluded. II. ARGUMENT A. Olsen’s Testimony Must Be Excluded Because His Conclusion That He Can Analyze Capital One’s Consumer Transaction Data To Determine Class Membership Is Not Supported By Sufficient Facts Or Data. It is Plaintiffs’ burden on class certification to demonstrate that “questions of law or fact common to class members predominate” over questions affecting only individual members. Fed. R. Civ. P. 23(b)(3). In attempting to meet this burden, Plaintiffs assert that the “predominance of common questions” is shown by “the nature of the proof required to ascertain members of the Class” and calculate their damages; namely, “[b]oth exercises will be accomplished by Plaintiffs’ highly-qualified expert, Arthur Olsen, through an automated process using Capital One’s detailed transaction history records.” (Plaintiffs’ Motion for Class Certification (“Mot. for Class Cert.”), filed under seal Dec. 20, 2011, at 4.) But as Capital One showed in its opening brief, Olsen’s report summarily concludes that he can develop an automated process to identify class members using Capital One’s consumer account records, without describing the facts or assumptions that permit him to reach that conclusion and without accounting for undisputed facts about the availability and accessibility of Capital One’s archived transaction data. (ECF No. 2454 (“Mot.”) at 5-7.) This is not a “battle of the experts” and has nothing to do with the way Plaintiffs propose to calculate damages. Rather, this is a straightforward question of whether Olsen has shown – in a way that is sufficiently reliable under Daubert – that he can do what Plaintiffs rely upon him to do. Capital One showed that he has not. Plaintiffs’ opposition brief offers nothing more than unsupported conclusions to the contrary. (See e.g., Plaintiffs’ Opposition to Capital One’s Motion to Exclude Testimony (“Opp’n”), filed under seal May 29, 2012, at 5-7.) Case 1:09-md-02036-JLK Document 2751 Entered on FLSD Docket 06/11/2012 Page 2 of 11 3 sf-3152700 The most significant problem Capital One identified is Olsen’s complete failure to address the unique issues presented by Capital One’s customer transaction data for all dates prior to December 11, 2008. (See Mot. at 6-7.) As Chip Gremillion explained in his 30(b)(6) deposition, Capital One’s historical transaction data “exists in a reporting archive . . . in a proprietary . . . encrypted file format” in which data cannot be searched or manipulated. (Pls.’ Ex. 6, Gremillion Depo 31:10-32:18.) Although small amounts of data can be exported piecemeal out of the report archive and into a text file format, it is “impractical to export very large reports out of the system.” (Id.) The data cannot be exported in the large quantities that Plaintiffs are seeking because “trying to save more than [200] or 300 pages of reports creates bottlenecks on the network . . . . [One] particular report averages [70,000] to 100,000 pages,” so it would take “a considerable amount of time to download a single day’s worth of reports.” (Id. (emphasis added); see also id. 78:8-79:18 (“[O]nce you attempt to export more than 100 pages at a time, the system is not equipped to deliver that . . . amount of mass of data.”); id. 81:14-82:2 (same limitations as to different data); id. 83:7-83:16 (same); Pls.’ Ex. 4, Donnelly Depo 47:13- 49:3 (“[O]ne day’s worth [of consumer data] would be roughly 100,000 pages.”).) Ken Donnelly, another 30(b)(6) deponent, further explained that historical data reports can only be exported to text files individually, with one report representing one day’s worth of data, and therefore it would take “unlimited funds” to export all the consumer transaction data Plaintiffs would require. (Pls.’ Ex. 4, Donnelly Depo 47:13-48:3.) Asked whether it would be possible to export all the necessary data to a searchable format through a single automated process, Donnelly responded that “you need a good expert because I have not figured out a way to do more than one day at a time.” (Id.) Although Olsen was fully aware of these issues, his declaration does not even mention the virtual impossibility of translating Capital One’s historical account data into a readable format on which he can perform the rest of his proposed analysis, let alone suggest any method for handling this complication. First, Plaintiffs argue that this oversight is “[not] relevant to the reliability of Olsen’s analysis.” (Opp’n at 13-14.) Of course it is relevant: it is Plaintiffs who put forth Olsen’s testimony to support their argument that Class members can be identified and damages can be calculated “through an automated process using Capital One’s detailed transaction history records.” (Mot. for Class Cert. at 4, 18, 36.) As Olsen’s testimony does not demonstrate with any reliability how he can accomplish this, his testimony is not helpful to the Court in deciding Plaintiffs’ motion for class certification; therefore his testimony must be Case 1:09-md-02036-JLK Document 2751 Entered on FLSD Docket 06/11/2012 Page 3 of 11 4 sf-3152700 excluded under Daubert. See Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 591-92 (1993) (The requirement that expert testimony “assist the trier of fact . . . goes primarily to relevance,” and therefore requires a valid connection between the testimony and the pertinent inquiry “as a precondition to admissibility.”). Second, Plaintiffs argue that Olsen has satisfied any obligation he does have to explain his methodology because, from his review of interrogatory responses and deposition testimony, “Olsen determined that [Capital One] maintained and is able to provide him with access to all of the information and data needed to perform ‘the same type of comprehensive analysis that [he] performed in the Gutierrez case.’” (Opp’n at 7.) But this is not a methodology; this is a conclusion. Plaintiffs cannot simply substitute Gutierrez – a different case involving different parties, different data systems, and different retention policies – in place of the “sufficient facts or data” required by Daubert and the Federal Rules of Evidence.1 Moreover, Plaintiffs do not even suggest that the Gutierrez case involved the same kinds of data translation problems at issue here.2 Plaintiffs’ assertion that “Olsen demonstrated that he is able to perform the same calculations he made in Gutierrez” by performing exemplar calculations on the accounts of the named Plaintiffs is without merit because, as Olsen acknowledged in his declaration, his exemplar calculations did not take into account any transaction data pre-dating December 11, 2008. (Appendix IV to Plaintiffs’ Mot. for Class Cert. (“Olsen Decl.”) ¶ 51.) Third, Plaintiffs argue that, even if Olsen’s proposed analysis depends on the astronomically expensive and laborious piecemeal approach described by the 30(b)(6) deponents – which of course it must – considerations of time and expense cannot be considered in determining whether Olsen’s testimony is admissible to support their motion for class certification. (Opp’n at 14-15.) Plaintiffs are wrong. As Capital One explained, “litigating the plaintiffs’ claims as [a class action] no matter what the cost . . . runs counter to the policies underlying Rule 23(b)(3),” which is meant to encompass “those cases in which a class action would achieve economies of time, effort and expense,” (Mot. at 6 n.2, citing Andrews v. AT&T 1 The analysis in Gutierrez v. Wells Fargo on which Plaintiffs rely is currently pending appellate review before the Ninth Circuit Court of Appeals. (Case Nos. 10-17468, 10-16959, 17689.) Oral argument was held on May 15, 2012. 2 Nor do Plaintiffs suggest that similar problems were at issue in PNC Bank’s or TD Bank’s motions before this Court to strike Olsen’s declaration. Indeed, Plaintiffs do not identify any way in which this Court’s PNC Bank or TD Bank orders are similar to the issues presented here; this is because the facts and issues raised in those Orders were very different from the facts and issues raised here. Therefore, those orders have no application to this motion. Case 1:09-md-02036-JLK Document 2751 Entered on FLSD Docket 06/11/2012 Page 4 of 11 5 sf-3152700 Co., 95 F.3d 1014, 1025 (11th Cir. 1996)), and class certification is not appropriate where determining class membership is “too administratively difficult,” (Mot. at 7 n.2, citing Perez v. Metabolife Int’l, Inc., 218 F.R.D. 262, 269 (S.D. Fla. 2003).) Plaintiffs ignore these cases entirely. Because Olsen does not propose a methodology for identifying class membership or calculating damage which is helpful to the Court at the class certification stage, his testimony is not admissible under Daubert. 509 U.S. at 590. In addition to Olsen’s failure to address the issues affecting customer transaction data dating before December 11, 2008, Capital One showed in its motion that Olsen has also failed to set forth “sufficient facts or data” to support his conclusion that he is capable of analyzing customer transaction data after that date. (Mot. at 5-6.) Olsen’s declaration states that his proposed analysis would be “similar” to the one he performed in Gutierrez, but with “small adjustments” to account for the differences in Capital One’s data (Olsen Decl. ¶¶ 46, 50); but nowhere does Olsen identify what adjustments would be required or how his analysis in this case would differ. This is another issue that Plaintiffs ignore completely in their opposition brief. In short, Olsen’s conclusion that he can apply an automated process using Capital One’s customer transaction data to determine class membership and calculate damages is not supported by sufficient facts or data, and therefore falls short of the Daubert standard for admissibility. Olsen’s testimony should be excluded on this basis alone. B. Olsen’s Testimony Must Be Excluded Because His Conclusion That He Can Determine Which Customers Were Harmed By Capital One’s Posting Order Policy Is Based On Several Incorrect And Unsupported Assumptions. Olsen’s testimony should be excluded for a second, independent reason. As Capital One demonstrated in its opening brief, Olsen’s conclusion that his methodology can be used to determine which customers were harmed as a result of Capital One’s posting policy, and can therefore be used to identify which customers are members of the proposed Class, falls far short of the Daubert reliability standard because it is premised on several incorrect and unsupported assumptions. Plaintiffs’ opposition brief does nothing to cure these defects. Instead, Plaintiffs attempt to belatedly reframe the supposed purpose of Olsen’s testimony to avoid addressing these defects altogether. This Court should reject this attempt and exclude Olsen’s testimony. 1. Plaintiffs’ Opposition and Plaintiffs’ class certification briefing make contradictory assertions regarding the purpose of Olsen’s testimony. In their motion for class certification, Plaintiffs assert that Olsen’s testimony shows that it is possible to identify by common proof which customers were harmed as a result of Capital Case 1:09-md-02036-JLK Document 2751 Entered on FLSD Docket 06/11/2012 Page 5 of 11 6 sf-3152700 One’s posting policies and, consequently, which customers are members of the proposed Class. (Mot. for Class. Cert. at 4, 18, 36.) Assuming Olsen’s methodology is truly intended to accomplish this goal, it falls far short of the Daubert standard for the reasons described in Capital One’s opening brief. Perhaps realizing that Olsen’s methodology cannot reliably determine which customers were harmed, Plaintiffs now shift their strategy and claim that the only thing Olsen was asked to do was to “locate the relevant transactional data contained in [Capital One’s] databases and to write a program to perform computations using that data.” (Opp’n at 5.) Whether those computations result in a reliable picture of which customers were harmed by Capital One’s posting policies, Plaintiffs now say, is a merits question that can only be answered by the fact finder at trial. (Id. at 1, 11.) First, this is an incorrect statement of the law; it is Plaintiffs’ burden on class certification to show that the members of the class they seek to certify suffered harm and “consequently, have Article III standing” to assert claims against Capital One. See Conigliaro v. Norwegian Cruise Line, No. 05-21584-CIV-ALTONAGA/Turnoff, 2006 U.S. Dist. LEXIS 95576, at *24-25 (S.D. Fla. Aug. 31, 2006); see also U.S. Parole Comm’n v. Geraghty, 445 U.S. 388, 415 n.8 (1980) (“Certification is no mere formality. It represents a judicial finding that injured parties other than the named plaintiffs exist [and] provides a definition by which they can be identified.”); In re Hydrogen Peroxide Litig., 552 F.3d 305, 317 n.17 (3d Cir. 2008) (concern for merits avoidance “should not be talismanically invoked to artificially limit a trial court’s examination of the factors necessary to a reasoned determination of whether a plaintiff has met her burden of establishing each of the Rule 23 class action requirements.”) (citations omitted). Second, if it really is the case that Plaintiffs do not intend for Olsen’s testimony to show which class members were harmed, then Olsen’s testimony is not relevant to the arguments for which Plaintiffs cite it, and therefore it would not be helpful to the Court in deciding class certification. The requirement that expert testimony “assist the trier of fact . . . goes primarily to relevance,” and therefore requires a valid connection between the testimony and the pertinent inquiry “as a precondition to admissibility.” Daubert, 509 U.S. at 591. 2. Olsen’s methodology is based on the incorrect assumption that re-ordering transactions can reliably show whether customers were harmed by Capital One’s posting policies. As Capital One has shown, Olsen’s testimony that he can identify which customers were “harmed” as a result of Capital One’s posting policy by re-ordering transactions according to an alternative posting order is based on the incorrect assumption that, had an alternative posting Case 1:09-md-02036-JLK Document 2751 Entered on FLSD Docket 06/11/2012 Page 6 of 11 7 sf-3152700 order actually been in place, all factors would remain exactly the same except the order in which transactions were posted and the number of overdraft fees that Capital One would charge within the span of a single day. (Mot. at 7-9 (citing Olsen Decl. ¶¶ 17, 46-47, 53).) This assumption is incorrect because, as Capital One showed, many additional factors would change under an alternative posting order scenario, including an increase in NSF fees charged by Capital One, an increase in late fees charged by third parties, and an increase in overdraft fees charged by Capital One due to the resubmission of returned checks. (Mot. at 7-9.) Plaintiffs do not dispute that Olsen’s assumption is incorrect or that, because of this, Olsen cannot accurately determine which customers were “harmed” by Capital One’s posting policies. Instead, the sum of Plaintiffs’ response is that the Court should simply ignore this basic defect in Olsen’s methodology. First, Plaintiffs suggest that the Court may ignore any economic harm that would flow to customers under an alternative posting order scenario because it is too “speculative.” (Opp’n at 9.) To the contrary: as Capital One showed, the increased NSF and overdraft fees that customers would incur under an alternative scenario derive predictably from Capital One’s consumer checking account policies, and the fact that third parties (such as landlords or student loan companies) predictably charge fees for late payments is beyond dispute. (Mot. at. 7-9.) It is Olsen who provides no analysis showing that a static re-ordering of transactions accurately reflects harm, and therefore it is Olsen whose methodology is “so speculative and unrooted in reality as to be inherently unreliable as well as unhelpful.” (Opp’n at 9, citing Anderson v. Hale, No. CIV-02-0113-F, 2002 U.S. Dist. LEXIS 28281, at *20-22 (W.D. Okla. Nov. 4, 2002).) Plaintiffs bear the burden on this issue, and Plaintiffs have failed to meet their burden. See Hendrix ex rel. G.P. v. Evenflo Co., 609 F.3d 1183, 1194 (11th Cir. 2010) (“The proponent of the expert testimony bears the burden of showing” that it is admissible under each prong of the Daubert standard.). Second, Plaintiffs suggest that the Court can ignore the undisputedly false assumption at the heart of Olsen’s methodology because it concerns only the “weight” of Olsen’s testimony and not its “admissibility.” (Opp’n at 10-12.) Plaintiffs are wrong. Before admitting the testimony of any expert witness, courts in the Eleventh Circuit must undertake a “rigorous” inquiry to determine whether “the methodology by which the expert reaches his conclusions is sufficiently reliable.” Hendrix, 609 F.3d at 1194. Where an expert’s methodology “makes an assumption based on a false inference,” as Olsen’s methodology does here, it is not sufficiently reliable under Daubert, and therefore it is not admissible. (Mot. at 7 (citing Kilpatrick v. Berg, Case 1:09-md-02036-JLK Document 2751 Entered on FLSD Docket 06/11/2012 Page 7 of 11 8 sf-3152700 Inc., 613 F.3d 1329, 1343 (11th Cir. 2010).)3 Plaintiffs’ cases do not say otherwise. This is not a question of whether Olsen took every minor variable into account for purposes of calculating damages at trial under circumstances where reasonable minds could disagree. The issue here is whether Olsen presents a sufficiently reliable method for determining, on the basis of common evidence, who has been harmed and consequently who is a member of the proposed class. At most, Olsen has shown that his methodology can demonstrate the change in overdraft fees arising from a re-ordering of transactions within the confines of a single day. Olsen assumes that this method properly identifies those customers who have been harmed by Capital One’s posting policies, but does not provide any facts or data showing this assumption to be true. Capital One has shown that it is not true. Olsen’s testimony therefore is not helpful to the Court in deciding Plaintiffs’ motion for class certification, and must be excluded from consideration.4 3. Olsen’s methodology cannot reliably determine who was harmed because it fails to identify those customers whose overdraft fees were reversed by Capital One. Even assuming that Olsen’s proposed methodology of re-ordering transactions and counting the resulting overdraft fees was an appropriate way to identify which customers have been “harmed” by Capital One’s posting order policies, which it is not, Olsen does not show that he can reliably apply even this method. As Capital One demonstrated in its opening brief, Olsen’s methodology is not capable of identifying those customers whose overdraft fees were reversed by Capital One, and who therefore were not “harmed” by Capital One’s posting policy. (Mot. at 9-10.) Plaintiffs acknowledge that Olsen’s methodology is not capable of accurately accounting for fee reversals, and Plaintiffs do not dispute that customers who were initially 3 Plaintiffs’ attempt to distinguish Kilpatrick fails. (See Opp’n at 11.) Plaintiffs state that the expert in Kilpatrick “failed to present a reliable study establishing a direct causal link between the plaintiff’s . . . injury and the purported cause,” whereas Olsen “directly links the breach of contract . . . with the resulting harm.” (Id.) As shown above, Olsen’s methodology does not directly link customers’ injury to Capital One’s posting practices; to the contrary, neither Olsen nor Plaintiffs offer any methodology for assessing whether a given customer suffered actual harm. 4 Plaintiffs’ assertion that, if they proceed under a breach of contract theory, they need not show that Capital One’s conduct was the cause of customers’ harm – and need not take into account whether Plaintiffs would have suffered greater harm under Plaintiffs’ proposed alternative posting scenarios – is baffling. (See Opp’n at 8, 10.) “Causation is an essential element of damages in a breach of contract action, and . . . a plaintiff must prove that a defendant’s breach directly and proximately caused his or her damages.” Diesel Props. S.R.L. v. Greystone Bus. Credit II LLC, 631 F.3d 42, 52-53 (2d Cir. 2011) (citations omitted); Favrot v. Favrot, 68 So. 3d 1099, 1109 (La. Ct. App. 2011) (party liable for damages “caused by” his breach). Capital One’s posting policy cannot be said to have “caused” harm to a customer if that customer would have suffered a greater economic injury under Plaintiffs’ proposed alternative posting order. See 24 Williston on Contracts § 64:1 (4th ed.) (relevant damages measure is “an amount reasonably calculated to make [the plaintiff] whole and neither more nor less”). Case 1:09-md-02036-JLK Document 2751 Entered on FLSD Docket 06/11/2012 Page 8 of 11 9 sf-3152700 charged allegedly wrongful overdraft fees, but then had those fees reversed by Capital One, would not be considered “harmed.” But again, Plaintiffs ask the Court to overlook the defect. Plaintiffs’ argument is without merit. First, Plaintiffs argue that because Capital One does not maintain its customer records in a way that automatically links fee reversals to specific fee assessments, the need for actual data showing who has been harmed should be replaced with an inference that works to the disadvantage of Capital One (here, the LIFO or 30-Day inferences). (Opp’n at 15.) Plaintiffs’ only support for the use of such an inference is Gutierrez. (Opp’n at 15-16, citing Gutierrez v. Wells Fargo, 730 F. Supp. 2d 1080, 1139 (N.D. Cal. 2010).) But Gutierrez incorrectly applied the law on this point. The only case cited by the Gutierrez court is Bigelow v. RKO Radio Pictures, 327 U.S. 251, 264 (1946), where the Supreme Court held that a defendant may not object to a plaintiff’s reasonable estimate of damages where the defendant’s “wrongful action” prevents plaintiffs from obtaining a more precise measure. 730 F. Supp. 2d at 1139 n.28. Recent Eleventh Circuit authority confirms that “an adverse inference is drawn from a party’s failure to preserve evidence only when absence of that evidence is predicated on bad faith.” Bashir v. Amtrak, 119 F.3d 929, 931 (11th Cir. 1997). Such an inference is not permissible where – as is the case here – the absence of evidence stems from a party’s storage or destruction of documents in the ordinary course of business pursuant to a routine document retention policy. Managed Care Solutions, Inc. v. Essent Healthcare, Inc., 736 F. Supp. 2d 1317, 1333 (S.D. Fla. 2010); Corporate Fin. Inc. v. Principal Life Ins. Co., No. 05-20595-CIV-UNGARO-BENAGES, 2006 U.S. Dist. LEXIS 84540, at *7-10 (S.D. Fla. Nov. 20, 2006) (same). Second, Plaintiffs urge the Court to permit use of the LIFO or 30-Day inference on a class-wide basis on the ground that “individual methods of proof” will be no more precise. (Opp’n at 15.) This is not true; fee reversals are exactly the type of situation for which an individualized, account-level review is superior to the anonymous, automated review that Plaintiffs seek. At an individual level, the fact finder can review notes placed in a customer’s file about the reason for the reversal, the testimony of the customer who sought the reversal and the employee who authorized it, and the customers’ contemporaneous account activity. All of these things would shed light on the connection between the reversal and the initial fee, and they would all be missed by Olsen’s proposed approach. Plaintiffs’ only support for their conclusion about individual review is Gutierrez, but Gutierrez did not contemplate these factors, and the only case it cited is inapposite. See Gutierrez v. Well Fargo, No. C 07-05923 WHA, 2010 U.S. Case 1:09-md-02036-JLK Document 2751 Entered on FLSD Docket 06/11/2012 Page 9 of 11 10 sf-3152700 Dist. Ct. LEXIS 29117, at *34 (N.D. Cal. March 26, 2010) (citing Bigelow, 327 U.S. at 251). Because Olsen’s testimony does not set forth a sufficiently reliable methodology for identifying consumers whose allegedly wrongful overdraft fees have been waived, and who therefore were not harmed by Capital One’s posting policies and are not members of the proposed Class, Olsen’s testimony must be excluded under Daubert. III. CONCLUSION For the reasons stated above and in Capital One’s opening brief, Capital One respectfully requests that the Court grant its Motion to Exclude Testimony from Plaintiffs’ Expert Witness Arthur Olsen. Dated: June 11, 2012 Respectfully submitted by, MORRISON & FOERSTER LLP By: /s/ James R. McGuire JAMES R. McGUIRE JAMES R. McGUIRE (pro hac vice) RITA F. LIN (pro hac vice) MORRISON & FOERSTER LLP 425 Market Street San Francisco, CA 94105 Telephone: 415.268.7000 Facsimile: 415.268.7522 Attorneys for Defendant CAPITAL ONE, N.A. Case 1:09-md-02036-JLK Document 2751 Entered on FLSD Docket 06/11/2012 Page 10 of 11 1 sf-3152700 CERTIFICATE OF SERVICE CASE NO.: 1:09-MD-02036-JLK I HEREBY CERTIFY that on June 11, 2012, I filed the following document with the Clerk of the Court using CM/ECF. I also certify that the foregoing document is being served this day either by Notice of Electronic Filing generated by CM/ECF or by U.S. mail on all counsel of record entitled to receive service. DEFENDANT CAPITAL ONE, N.A.’S REPLY IN SUPPORT OF MOTION TO EXCLUDE TESTIMONY FROM PLAINTIFFS’ EXPERT WITNESS ARTHUR OLSEN /s/ Janie Fogel Janie Fogel MORRISON & FOERSTER LLP 425 Market Street San Francisco, CA 94105 Telephone: 415.268.7000 Facsimile: 415.268.7522 Case 1:09-md-02036-JLK Document 2751 Entered on FLSD Docket 06/11/2012 Page 11 of 11