State of Cal. ex rel. Edelweiss Fund v. JPMorgan Chase & Co.

• Court: California Court of Appeal
• Decided: April 27, 2023
• Docket: A163264
• Status: Published

Opinion

Filed 4/27/23 CERTIFIED FOR PUBLICATION

IN THE COURT OF APPEAL OF THE STATE OF CALIFORNIA

FIRST APPELLATE DISTRICT

DIVISION FOUR

STATE OF CALIFORNIA ex rel. EDELWEISS FUND, LLC, et al., A163264 Plaintiffs and Appellants, v. (City & County of San Francisco JPMORGAN CHASE & COMPANY Super. Ct. No. CGC-14-540777) et al., Defendants and Respondents.

Plaintiff-Relator Edelweiss Fund, LLC (Edelweiss) brought a qui tam action against several financial institutions and subsidiaries under the California False Claims Act (Gov. Code, § 12650 et seq.)1 (CFCA). In its operative seventh amended complaint, Edelweiss alleges that defendants contracted to serve as remarketing agents (RMAs) to manage California variable rate demand obligations (VRDOs): tax-exempt municipal bonds with interest rates reset by RMAs on a periodic basis, typically weekly. It alleges that defendants violated the CFCA by submitting false claims for payment for these remarketing services, knowing they had failed their obligation to reset the interest rate for the California VRDOs at the lowest possible rate that would enable them to sell the series at par (face value). Instead, defendants “engaged in a coordinated ‘Robo-Resetting’ scheme where they mechanically set the rates en masse without any consideration of the

1 Undesignated statutory references are to the Government Code. individual characteristics of the bonds or the associated market conditions or investor demand” and “impose[d] artificially high interest rates on California VRDOs, the exact opposite of what California hires them to accomplish.” Edelweiss also alleges that defendants had conspired to violate the CFCA by colluding to inflate interest rates on these VRDOs. Edelweiss alleges that it performed a forensic analysis of rate resetting during a four-year period, which revealed that defendants regularly grouped VRDOs with “vastly different” characteristics into “buckets,” and applied the same absolute rate change (a given number of basis points) to each bucket. Edelweiss identifies eight factors that made these bonds allegedly dissimilar: credit quality of the issuer, credit quality of the letter of credit provider, type of liquidity support facility, source of revenue, economic sector, the size of the issuance, the state of the issuance, and fraternity (applicable to small issues in which the buyers may have some special affiliation or fraternity with the issuer, like a university). Edelweiss alleges that it also studied credit rating upgrades for California VRDO issuers and identified “dozens of specific instances” in which the upgrade did not result in a relative decrease in the interest rate. It alleges that in these instances, defendants “set the interest rate of a VRDO at a level higher than it should have been, taking the relevant circumstances into consideration, including the characteristics of the VRDO at issue and the market preferences for it.” Edelweiss further alleges that various former employees of defendants “stated and corroborated” that defendants engaged in this robo-resetting scheme. A former employee of Wells Fargo stated that it initially determined by how many basis points the VRDO’s interest rate should differ relative to

2 the SIFMA index,2 and then “almost never” made adjustments to the spread, resetting rates of VRDOs in “large groups” by the same number of basis points. A former Citi employee described the VRDO market as “the ‘biggest joke of a market of all time’ ” and said that it should have operated on the basis of prevailing market conditions, but did not. The trial court sustained defendants’ demurrer to the seventh amended complaint without leave to amend, concluding that Edelweiss had not pleaded sufficiently particularized factual allegations. It reasoned that the allegations “may be consistent with fraud,” but the pleading lacked particularized allegations about how the defendants set their VRDO rates and did not support a reasonable inference that “the observed conditions were caused by fraud, as opposed to any other factors that may have influenced the relevant financial markets during the relevant time period.” On appeal, Edelweiss argues that the trial court applied an overly burdensome particularity requirement and erroneously concluded that Edelweiss had failed to adequately plead its claims. While allegations of a CFCA claim must be pleaded with particularity, we conclude that the trial court required too much to satisfy this standard. We also reject defendants’ alternative argument that Edelweiss’s claims are foreclosed by the CFCA’s public disclosure bar (§ 12652, subd. (d)(3)(A)). Accordingly, we reverse.

2SIFMA refers to the Securities Industry Financial Markets Association swap index, which “tracks the average interest rate for highly- rated VRDOs reset on a weekly basis.”

3 BACKGROUND3 1. CFCA The CFCA was enacted by the Legislature in 1987, patterned on the federal False Claims Act (31 U.S.C. § 3729 et seq.). (Rothschild v. Tyco Internat. (US), Inc. (2000) 83 Cal.App.4th 488, 494.) It imposes liability for civil penalties and treble damages on any person who (1) “[k]nowingly presents or causes to be presented a false or fraudulent claim for payment or approval”; (2) “[k]nowingly makes, uses, or causes to be made or used a false record or statement material to a false or fraudulent claim”; or (3) “[c]onspires to commit a [CFCA] violation.” (§ 12651, subd. (a)(1)–(3).) The CFCA was designed “ ‘to prevent fraud on the public treasury,’ ” and thus it “ ‘must be construed broadly so as to give the widest possible coverage and effect to the prohibitions and remedies it provides.’ ” (City of Pomona v. Superior Court (2001) 89 Cal.App.4th 793, 801–802 (Pomona).) The CFCA was also intended “to supplement governmental efforts to identify and prosecute fraudulent claims made against state and local governmental entities.” (Rothschild v. Tyco Internat. (US), Inc., supra, 83 Cal.App.4th at p. 494.) Accordingly, the CFCA contains qui tam provisions authorizing private relators to bring actions on behalf of California to seek redress for a violation of the act. (§ 12652, subd. (c)(1).) A qui tam plaintiff must file a CFCA complaint under seal and serve it on the Attorney General, and include a written disclosure of the plaintiff’s material evidence and information. (Id., subd. (c)(2), (3).) After the Attorney General notifies the court that both it and the appropriate prosecuting authority decline to

3 The following is a brief summary of some of the factual and procedural background in this case, which we set out to provide context to the issues raised on appeal. Additional facts are included in our legal discussion.

4 proceed with the action, the qui tam plaintiff has the right to prosecute the action. (Id., subd. (c)(8)(D)(iii).) Qui tam claims based on certain categories of information are foreclosed by what is known as the CFCA’s “public disclosure bar.” (State ex rel. Bartlett v. Miller (2016) 243 Cal.App.4th 1398, 1407 (Bartlett).) Section 12652, subdivision (d)(3)(A) (section 12652(d)(3)(A)) states: “The court shall dismiss an action or claim under this section, unless opposed by the Attorney General or prosecuting authority of a political subdivision, if substantially the same allegations or transactions as alleged in the action or claim were publicly disclosed in any of the following: [¶] (i) A criminal, civil, or administrative hearing in which the state or prosecuting authority of a political subdivision or their agents are a party. [¶] (ii) A report, hearing, audit, or investigation of the Legislature, the state, or governing body of a political subdivision. [¶] (iii) The news media.” The public disclosure bar, however, does not apply if the person bringing the action is the “original source of the information.” (Id., subd. (d)(3)(B).) “Original source” includes an individual who has voluntarily provided the information to the state before filing an action or “[h]as knowledge that is independent of, and materially adds to, the publicly disclosed allegations or transactions.” (Id., subd. (d)(3)(C)(ii).) If the qui tam plaintiff proceeds and is successful, the plaintiff is typically entitled to receive between 25 and 50 percent of the recovery achieved in the case. (§ 12652, subd. (g)(3)–(5) [subject to exceptions for government employees and individuals who planned or initiated the CFCA violation].) The CFCA thus “seeks to induce private ‘whistleblowers,’ uniquely armed with information about false claims, to risk the failure of their qui tam suits in hopes of sharing in a handsome recovery if they

5 succeed. Indeed, this prospect of reward may be the only means of inducing such private parties to come forward with their information.” (State ex rel. Harris v. PricewaterhouseCoopers, LLP (2006) 39 Cal.4th 1220, 1231.) “ ‘The driving force behind the false claims concept is the providing of incentives for individual citizens to come forward with information uniquely in their possession and to thus aid the Government in [ferreting] out fraud.’ ” (Ibid.) 2. This Action In 2014 Edelweiss filed a qui tam action on behalf of the state of California against several financial institutions and subsidiaries alleging a single cause of action for violation of the CFCA. Edelweiss is a limited liability company and its sole principal, B. Johan Rosenberg, is a “registered municipal advisor with more than 20 years of experience advising municipalities and other clients on issuing securities, particularly VRDOs and other municipal bonds.” Rosenberg is not a current or former employee of any of the defendants. In July 2019, after Edelweiss had twice amended its complaint, the trial court dismissed the defendants named in the original complaint for failure to serve them with the complaint and summons within three years after the action was commenced. (Code Civ. Proc., § 583.210, subd. (a).) The trial court, however, declined to dismiss the additional defendants who had been added in the second amended complaint. We affirmed this order in State ex rel. Edelweiss Fund, LLC v. JP Morgan Chase & Co. (2020) 58 Cal.App.5th 1113, 1126. In August 2019 the trial court sustained the remaining defendants’ demurrer to the second amended complaint with leave to amend. Edelweiss filed a third amended complaint and, by stipulated order, a fourth amended complaint. The trial court sustained defendants’ demurrer to the fourth

6 amended complaint with leave to amend. It found that the allegations remained insufficient due to lack of particularity. It also addressed, and rejected, defendants’ alternative argument that the action was prohibited by the public disclosure bar because it relied on interest rates published on the Electronic Municipal Market Access (EMMA) website.4 Relying on Bartlett, the trial court concluded that EMMA is not a “report” of the state or “news media” triggering the public disclosure bar. (Bartlett, supra, 243 Cal.App.4th at pp. 1413–1414 [no public disclosure bar based on Securities and Exchange Commission (SEC) filing accessible on its online database].) Edelweiss filed a fifth amended complaint, and the trial court again sustained defendants’ demurrer with leave to amend for lack of particularity. Edelweiss then filed a sixth amended complaint with a motion for leave to file a seventh amended complaint. The trial court granted the motion. In March 2021 Edelweiss filed its seventh amended complaint. 3. Seventh Amended Complaint The seventh amended complaint asserts a single cause of action for violation of the CFCA against JPMorgan, Citigroup, Wells Fargo, Bank of America, Merrill Lynch, Morgan Stanley, Barclays, Royal Bank of Canada (RBC), and Piper Jaffray. It alleges that, since 2009, these defendants each served as the RMA for between 54 and 408 VRDOs issued by California (either the state or some local governmental entity). The seventh amended complaint included the following allegations regarding VRDOs, remarketing agreements, defendants’ “robo-resetting” scheme, and the resulting harm to California.

4EMMA is a publicly accessible website operated by the Municipal Securities Rulemaking Board (MSRB). Effective April 1, 2009, MSRB rules require RMAs to report VRDO interest rate resets and this information is available through EMMA.

7 a. VRDOs VRDOs are variable rate, tax-exempt bonds primarily issued by state and local government entities to raise money to fund various long-term projects or infrastructure. The interest rate on VRDOs is reset regularly, typically weekly or even daily. VRDOs are an attractive financing option for these government entities because they can borrow money for long periods of time while paying lower, short-term interest rates that are reset regularly. VRDOs are attractive to investors because they are a low-risk, high-liquidity, and tax-free investment. b. Remarketing Agreements VRDO issuers contract with RMAs to manage these bonds. The seventh amended complaint references “representative examples” of remarketing agreements with each of the eight defendants, and quotes a provision in these agreements entitled “Rate Resetting Obligation.” The quoted provision in the example agreement with Bank of America reads: “The Weekly Rate shall be the rate determined by the Remarketing Agent to be the lowest rate which would enable the Remarketing Agent to sell the [Series 2005B2 and B3 Bonds] for delivery on the effective date of such rate at a price (without regard to accrued interest) equal to 100% of the principal amount thereof.” (Italics and boldface omitted.) The quoted provisions in the example agreements with Barclays, Citigroup, and JPMorgan contain similar language. The quoted provision in the example agreement with Morgan Stanley reads: “The Weekly Rate shall be the rate of interest per annum determined by the Remarketing Agent to be the minimum interest rate which, if borne by such [St. Joseph’s Series 2011B, C & D Bonds] under Prevailing Market Conditions, would enable the Remarketing Agent to sell such [St. Joseph’s

8 Series 2011B, C & D Bonds] on the effective date of such rate at a price (without regarding accrued interest) equal to the principal amount thereof.” (Italics and boldface omitted, and italics added.) The quoted provision in the example agreement with Piper Jaffray also contains similar “prevailing market conditions” language. The quoted provision in the example agreement with RBC reads: “The Variable Rate determined by the Remarketing Agent on each Variable Interest Computation Date shall be that rate of interest which, if borne by the [Turnleaf Series 2003A Bonds], would, in its reasonable professional judgment, on the basis of prevailing financial market conditions, be the interest rate necessary, but which would not exceed the interest rate necessary, to be borne by the [Turnleaf Series 2003A Bonds] in order for the market value of the [Turnleaf Series 2003A Bonds] on such Variable Interest Computation Date to be 100% of the principal amount thereof (disregarding accrued interest) if the [Turnleaf Series 2003A Bonds] were sold on such Variable Interest Computation Date; provided, however, that in no event shall the Variable Rate at any time exceed the Maximum Rate.” (Italics and boldface omitted, and italics added.) According to the seventh amended complaint, to determine the lowest interest rate, an RMA must consider the “unique characteristics” of the VRDO, including “the identity of the issuer, the source of repayment, the issuer’s credit rating, the sector of the finance project, the location of the finance project, the existence of a liquidity provider, the type of liquidity provided, and the credit rating of the liquidity provider.” c. “Robo-Resetting” Scheme The seventh amended complaint alleges that defendants violated their “Rate Resetting Obligation” to reset the interest rate for California VRDOs at

9 the lowest possible rate that would enable them to sell the series at par.5 Instead, defendants “engaged in a coordinated ‘Robo-Resetting’ scheme where they mechanically set the rates en masse without any consideration of the individual characteristics of the bonds or the associated market conditions or investor demand” and “impose[d] artificially high interest rates on California VRDOs, the exact opposite of what California hires them to accomplish.” Edelweiss offers four bases to support this theory. First, it alleges that Edelweiss performed “an extensive forensic analysis of the interest rates and other market data” from April 1, 2009, to November 14, 2013, “for the California VRDOs (and VRDOs issued by other states) for which Defendants have served as the RMA.” Edelweiss concluded that the interest rates on “buckets” of unrelated VRDOs were set collectively, moving up or down in “lock-step fashion.” Second, it alleges that Edelweiss performed a “study of credit-rating upgrades for California VRDO issuers” and identified “dozens of examples” where the upgrade “did not result in a decrease in the interest rate each Defendant set for the VRDO at issue, relative to the interest rates the Defendant set on the VRDOs of other issuers whose credit ratings were not upgraded.” Third, it alleges that seven former employees of defendants “stated and corroborated” that defendants engaged in this robo-resetting scheme.

5 It also alleges that defendants had a “Remarketing Obligation” to “use their best efforts to ‘remarket’ at the lowest possible interest rate the VRDOs to other investors when the existing investor ‘puts’ or ‘tenders’ the bond back to the RMA for a return of its investment (at face value plus interest).” Edelweiss does not make any argument that it adequately pleaded a CFCA claim based on this obligation, and accordingly we deem any such argument waived. (Benach v. County of Los Angeles (2007) 149 Cal.App.4th 836, 852.)

10 Fourth, it alleges that defendants’ inflated VRDO pricing “is evident from a comparison of their average pricing to the average pricing of 7-day AA non-financial commercial paper.” According to Edelweiss, “[t]his is the type of security most closely analogous to VRDOs because both are of a short-term nature, have a low rate of default, are issued by non-financial entities, and are approved investments for money market funds.” It alleges that in a “properly functioning market,” the interest rates on California VRDOs should not exceed 75 percent of 7-day AA non-financial commercial paper because unlike VRDOs, commercial paper is not tax-exempt and therefore is a less attractive investment. It alleges that, from 2008 onwards, the interest rate for defendants’ California VRDOs was higher than for commercial paper. The seventh amended complaint also alleges that defendants conspired to violate the CFCA by colluding to inflate interest rates on these VRDOs. It alleges that this collusion was evident from (1) “cross-bank bucketing” of VRDOs showing the same kind of “lock-step” price movements across defendants; (2) defendants’ “coordinated response” to certain market events; (3) defendants’ use of the J.J. Kenny Index;6 and (4) the relationships between defendants. d. Alleged Harm The seventh amended complaint alleges that California has been harmed because it (1) paid for remarketing services it did not receive; (2) paid higher interest on its VRDOs; and (3) paid for “letter of credit” services— where the provider must purchase the VRDO from a redeeming investor if the RMA is unable to find another investor—that were “rarely called upon”

6The J.J. Kenny Index is a third-party index of VRDO rates that is “regularly published in the form of an e-mail to over 200 subscribers, including the Defendants in this action.”

11 because of these inflated interest rates. It estimated damages of “over $641 million,” including $155 million in remarketing fees, $350 million in inflated rates, and $136 million in letter of credit fees during the 2009 to 2013 period. 4. Demurrer Sustained Without Leave to Amend Defendants demurred to the seventh amended complaint. The trial court sustained the demurrer without leave to amend. It found that Edelweiss had pleaded express contractual provisions from the RMA agreements that required defendants to set the interest rate at the lowest rate that would enable them to sell the VRDO series at par. It concluded, however, that Edelweiss had not pleaded particularized allegations of falsity, i.e., that defendants’ claims for payment were rendered false because they submitted the claims knowing they had materially breached those contractual obligations. Given that Edelweiss had not pleaded a viable theory of predicate liability, it concluded that the conspiracy claim also failed. This appeal followed. We received an amicus curiae brief submitted jointly by the Chamber of Commerce of the United States of America, the California Chamber of Commerce, and the American Bankers Association, as well as Edelweiss’s response to the amicus brief. 5. Edelweiss Actions in Other States In addition to its CFCA complaint, Edelweiss filed similar qui tam actions in four other states. Defendants’ motions to dismiss were denied in Illinois, New York, New Jersey, and Massachusetts. In Illinois, the trial court concluded that the allegations were sufficiently particular to state a claim under the Illinois False Claims Act and rejected dismissal based on the public disclosure bar.

12 In New York, the appellate court affirmed the denial of defendants’ motion to dismiss upon concluding that Edelweiss had adequately alleged a claim at this stage and that the proposed deficiencies fell into the realm of specifics that Edelweiss was not required to plead. (State ex rel. Edelweiss Fund, LLC v. JP Morgan Chase & Co. (2020) 140 N.Y.S.3d 1, 2.) Defendants were alleged to have bucketed VRDOs and treated them identically, sometimes for years, and without apparent economic justification, given the disparate characteristics of the VRDOs that Edelweiss itemized. (Id. at p. 3.) In New Jersey, the trial court denied defendants’ motion to dismiss after noting that it seemed like more of a summary judgment motion and that the question of whether Edelweiss “can prove the allegations contained in the fourth amended complaint is not the issue before me[.]”7 It found that Edelweiss had alleged with proper specificity that defendants reset rates for dissimilar VRDOs en masse and through a process that could not result in the lowest rates that defendants were required to obtain, and that “a contrary interpretation or an alternative explanation simply presents a factual issue for discovery.”

7 Edelweiss has made two requests for judicial notice on appeal. We granted the May 9, 2022 request, but deferred ruling on the November 19, 2021 request until the merits of the appeal. (See People v. Preslie (1977) 70 Cal.App.3d 486, 493–494.) We now grant Edelweiss’s request to take judicial notice of Exhibit 1, a transcript of decision in State of N.J. ex rel. Edelweiss Fund, LLC v. JP Morgan Chase & Co. (N.J.Super.Ct., Sept. 13, 2021, No. 885–15). Edelweiss also asks us to take judicial notice of Exhibit 2, an article from The Bond Buyer; defendants oppose the request. We deny this request. (Vons Companies, Inc. v. Seabest Foods, Inc. (1996) 14 Cal.4th 434, 444, fn. 3 [reviewing courts generally do not take judicial notice of evidence not presented to the trial court absent exceptional circumstances]; In re Noreen G. (2010) 181 Cal.App.4th 1359, 1389, fn. 13 [denying request for judicial notice where existence of newspaper article is irrelevant and truth of its contents is not judicially noticeable].)

13 In Massachusetts, the action was dismissed after the court found that EMMA was “news media” under the public disclosure bar of the Massachusetts False Claims Act (MFCA).8 That ruling was affirmed in Rosenberg v. JPMorgan Chase & Co. (2021) 487 Mass. 403 (Rosenberg).) DISCUSSION Edelweiss argues that the trial court erred in sustaining defendants’ demurrer to the seventh amended complaint for two reasons. First, it argues that the trial court applied an overly burdensome particularity requirement, effectively adopting the federal plausibility standard that California law does not support. Second, it argues that the trial court erred in concluding that the seventh amended complaint did not adequately plead falsity. Before turning to the merits of these arguments, we begin with the applicable standard of review. 1. Standard of Review “In reviewing the sufficiency of a complaint against a general demurrer, we are guided by long-settled rules.” (Blank v. Kirwan (1985) 39 Cal.3d 311, 318.) We review the operative complaint “de novo to determine whether the complaint alleges facts sufficient to state a cause of action under any legal theory or to determine whether the trial court erroneously sustained the demurrer as a matter of law.”9 (Aguilera v. Heiman (2009) 174 Cal.App.4th 590, 595.) We construe the complaint in a reasonable manner and assume the truth of properly pleaded factual allegations that are not inconsistent with other allegations, exhibits, or

8 The language of the MFCA public disclosure bar mirrors the language of the CFCA public disclosure bar. (Mass. Gen. Laws ch. 12, § 5G, subd. (c).) 9 While we would apply an abuse of discretion standard to review the trial court’s denial of leave to amend, no such argument has been raised in this appeal. (Blank v. Kirwan, supra, 39 Cal.3d at p. 318.)

14 judicially noticed facts. (Genis v. Schainbaum (2021) 66 Cal.App.5th 1007, 1015.) We need not accept as true, however, contentions, deductions, or conclusions of fact or law. (Blank v. Kirwan, supra, 39 Cal.3d at p. 318.) Appellant bears the burden of demonstrating that the trial court erred. (Aguilera v. Heiman, at p. 595.) With this standard in mind, we turn to some predicate determinations made by the trial court that affect our review. 2. Implied Certification Claim As part of its overarching arguments regarding the pleading standard and the sufficiency of its allegations, Edelweiss challenges two underlying determinations made by the trial court regarding what Edelweiss had alleged and what it was required to plead. First, the trial court determined that Edelweiss had alleged an “implied certification” claim. California courts distinguish between express and implied certification claims under the CFCA. (San Francisco Unified School Dist. ex rel. Contreras v. Laidlaw Transit, Inc. (2010) 182 Cal.App.4th 438 (Contreras).) An express certification claim may allege, for example, that a defendant submitted an invoice for payment under a contract and that the invoice expressly asserted compliance with the requirements of the contract. (Id. at p. 447.) The falsity of the claim thus flows from the expressly false statement on the invoice. (Id. at p. 448.) Contreras, however, explained that the CFCA also supports an implied certification claim. (Id. at p. 449.) If the contractor has knowledge of its non-compliance with the contract but nonetheless chooses to seek payment without informing the government— even if the claim for payment does not contain an expressly false statement— then “it is a fraud appropriately within the scope of the CFCA.” (Id. at p. 453.)

15 Edelweiss argues that it has alleged not only an implied certification claim, but also an “ ‘archetypal qui tam False Claims action’ ” based on a “ ‘literal false or fraudulent’ ” claim for payment. We are not persuaded. “The archetypal qui tam FCA action is filed by an insider at a private company who discovers his employer has overcharged under a government contract.” (United States ex rel. Hopper v. Anton (9th Cir. 1996) 91 F.3d 1261, 1265–1266 [explaining that federal FCA was originally enacted during Civil War to combat widespread fraud by government contractors submitting inflated invoices and shipping faulty goods].)10 Here, Edelweiss does not allege that defendants should have claimed some lower amount for payment than they actually did. Nor does it allege any other express false statement in defendants’ claims for payment. Instead, it alleges that defendants impliedly certified compliance with their contractual obligations by submitting payment for those services, and that this implied certification was false because defendants knew those services had not been performed as promised. Second, the trial court determined that because Edelweiss had alleged an implied certification claim, it was required to adequately plead falsity based on a failure of defendants’ express contractual “Rate Resetting Obligation” as set forth in their RMA agreements: to reset the interest rate for California VRDOs at the lowest possible rate that would enable them to sell the series at par. In other words, Edelweiss could not plead falsity by alleging only the failure of an implied obligation based on its interpretation of those agreements; namely, that defendants were required to reset the rates of each California VRDO “individually.”

10Given the “very close similarity” of the CFCA to the federal False Claims Act, “it is appropriate to turn to federal cases for guidance in interpreting the [CFCA].” (Pomona, supra, 89 Cal.App.4th at p. 802.)

16 We begin by observing that the implied certification claim in Contreras involved an alleged failure to comply with express contractual terms, so the court had no occasion to consider the viability of a claim premised on an implied obligation. (Contreras, supra, 182 Cal.App.4th at p. 443.) But Contreras explained that liability under the CFCA is distinguished from liability for any run-of-the-mill breach of contract insofar as not every breach of a “material” contract term will be material to the government’s decision to pay. (Id. at p. 456.) We thus agree with Contreras that “the materiality requirement is a meaningful limit to the scope of liability under the CFCA” (id. at p. 457), but we find no basis to conclude as a matter of law that compliance with an implied contractual term can never be material to the decision to pay. That said, what we view as material to the government’s payment decision in this case is defendants’ compliance with the express obligation to reset each VRDO’s interest rate at the lowest possible level to enable them to sell the series at par; Edelweiss itself acknowledges that the remarketing agreements do not mandate a specific process that defendants must use to arrive at it. Nonetheless, it follows from defendants’ rate-resetting obligation that they must employ some methodology that is in principle capable of allowing them to comply with it, and at this stage of the litigation we accept as true Edelweiss’s allegation that the amount an interest rate for a particular VRDO should rise or fall is a function of changes in that VRDO’s “unique characteristics” (or investor preference for those characteristics) across the categories Edelweiss identifies. The seventh amended complaint alleges that defendants’ approach to rate setting neither did nor could take

17 into account the relevant factors for each VRDO.11 But we need not decide whether such an allegation would suffice to state a claim in the absence of additional particularized allegations to support an inference that defendants failed to reset rates at the lowest possible level because, as explained below, we conclude that the seventh amended complaint adequately supplies them. 3. Pleading with Particularity CFCA actions are subject to “ ‘heightened fraud-like pleading requirements.’ ” (State of California ex rel. McCann v. Bank of America, N.A. (2011) 191 Cal.App.4th 897, 906 (McCann).) “ ‘Every element of the cause of action for fraud must be alleged in the proper manner (i.e., factually and specifically).’ ” (Committee on Children’s Television, Inc. v. General Foods Corp. (1983) 35 Cal.3d 197, 216.) This particularity requirement “necessitates pleading facts which ‘show how, when, where, to whom, and by what means’ ” the false representations were made. (Lazar v. Superior Court (1996) 12 Cal.4th 631, 645.) “ ‘As in any action sounding in fraud, the allegations of a [CFCA] complaint must be pleaded with particularity. The complaint must plead “ ‘the time, place, and contents of the false representations, as well as the identity of the person making the misrepresentation and what he obtained thereby.’ ” ’ ” (McCann, at p. 906, quoting Pomona, supra, 89 Cal.App.4th at p. 803.) Allegations of the

11To the extent this allegation is merely a conclusion drawn from Edelweiss’s analysis of the interest rate data, it adds nothing to that analysis. But, as explained further below, the allegation has independent support in statements by former employees that, for example, the remarketing desk was inadequately staffed or could have been “replaced by three monkeys,” and that the VRDO market was the “ ‘biggest joke of a market of all time’ ” and did not operate on the basis of prevailing market conditions.

18 defendant’s knowledge, however, may use “conclusive language.” (Pomona, at p. 803.) In McCann, this court explained that qui tam actions “are meant to encourage private whistleblowers, uniquely armed with information about false claims, to come forward,” and thus these insiders “should have adequate knowledge of the fraudulent acts to comply with the pleading requirement.” (McCann, supra, 191 Cal.App.4th at p. 907.) The heightened pleading requirement “ ‘serves not only to give notice to defendants of the specific fraudulent conduct against which they must defend, but also “to deter the filing of complaints as a pretext for the discovery of unknown wrongs, to protect [defendants] from the harm that comes from being subject to fraud charges, and to prohibit plaintiffs from unilaterally imposing upon the court, the parties and society enormous social and economic costs absent some factual basis.” ’ ” (Id. at p. 909.) Edelweiss argues that the trial court erred by assessing whether it had satisfied this particularity requirement under the federal “plausibility” standard, which requires a complaint to allege sufficient facts to state a claim to relief that is plausible on its face. (Bell Atlantic Corp. v. Twombly (2007) 550 U.S. 544, 570.) Instead, Edelweiss contends, the analysis should have followed the California standard: that a demurrer does not “test the truth of the plaintiff’s allegations” and we must assume the truth of “facts that reasonably can be inferred from those expressly pleaded.” (Franceschi v. Franchise Tax Bd. (2016) 1 Cal.App.5th 247, 256.) In its order, the trial court explained it was “not evaluating whether the facts Plaintiff pled with specificity are plausible, but whether the facts Plaintiff pled with specificity, which must be taken as true, provide reasonable support for the conclusions Plaintiff draws from them. . . . At a basic level, the particularity requirement

19 would serve no purpose if there was no connection between the particularized factual allegations and the ultimate conclusions.” While we see no error in this articulation of the standard, we conclude that the trial court required too much to satisfy it. As described above, Pomona defines the particularity requirement to include “ ‘ “the time, place, and contents of the false representations.” ’ ” (Pomona, supra, 89 Cal.App.4th at p. 803.) Edelweiss did that here. It alleged that, between 2009 and 2013, defendants submitted false claims for payment, and that those claims were false because defendants’ robo-resetting practices caused them to violate their express contractual obligations to reset the interest rate for California VRDOs at the lowest possible rate that would enable them to sell the series at par. These allegations satisfy the dual purposes of the particularity requirement: to put defendants on notice of the alleged fraudulent conduct and deter fishing expeditions for unknown wrongs. (McCann, supra, 191 Cal.App.4th at p. 909.) There can be no question that Edelweiss’s allegations gave defendants sufficient notice of the conduct at issue. Moreover, Edelweiss identified the particular alleged wrongs (violation of defendants’ contractual rate resetting obligations and conspiracy to commit said violations) and, as detailed below, provided at least “ ‘ “some factual basis” ’ ” to support those alleged wrongs. (Ibid.) 4. “Robo-Resetting” Allegations The seventh amended complaint contains four categories of allegations regarding the “robo-resetting” scheme: (1) Edelweiss’s bucketing analysis; (2) Edelweiss’s study of credit rating upgrades; (3) statements from former employees; and (4) Edelweiss’s commercial paper comparison. As a preliminary matter, we do not find the allegations regarding the commercial paper comparison to be sufficient. Edelweiss attempts to show

20 that defendants artificially inflated VRDO rates by alleging that the average VRDO rates from 1998 to 2007 were 75 percent of the average commercial paper rate, but the average VRDO rates in years 2008 to 2013 were higher than the average commercial paper rate. The seventh amended complaint, however, also includes alleged statements from three employees (Employees 1, 2, and 3) who worked for defendants during the 1998 to 2007 period indicating that defendants engaged in the same robo-resetting practices and inflation of VRDO rates during this time. If defendants’ alleged rate-resetting practices were the same both when the average VRDO rate was lower than the average commercial paper rate and when it was higher, it is not reasonable to infer that those practices explain the claimed inversion, at least in the absence of additional information that the seventh amended complaint does not supply. Moreover, Edelweiss’s additional allegations that the VRDO rate remained higher than commercial paper after 2013 was contradicted by judicially noticed data showing that the average VRDO rate was lower than the average commercial paper rate in 2014, 2015, 2017, 2018, and 2019. We thus reject Edelweiss’s conclusions from the commercial paper rate comparison because they are contradicted by its own factual allegations and judicially noticed facts. (Blank v. Kirwan, supra, 39 Cal.3d at p. 318; Genis v. Schainbaum, supra, 66 Cal.App.5th at p. 1015.) We address the remaining three categories in turn. a. Bucketing Analysis Edelweiss alleges that it performed a forensic analysis of defendants’ rate resetting from April 1, 2009 to November 14, 2013, which revealed that defendants regularly applied the same absolute rate change (a given number of basis points) to unrelated California VRDOs. When the pattern continued

21 often enough, Edelweiss considered those VRDOs to have been “bucketed” together. These “buckets” are not necessarily the groupings of VRDOs that defendants themselves made. Rather, they are patterns in the rate-resetting data that meet a definition furnished by Edelweiss. Specifically, Edelweiss considered a particular VRDO to be part of the same “bucket” if “its week- over-week rate change in a given week matched the most common week-over- week rate change among the bonds priced by a Defendant RMA that week, for at least twenty-six consecutive weeks . . . and at least 80 percent of the time.”12 So defined, Edelweiss’s analysis found “bucketing” for between 58 percent (126 of 217 for Bank of America) and 100 percent (all 59 for Piper Jaffray) of the California VRDOs in the portfolios studied. Edelweiss further alleges that California VRDOs in the same “bucket” had “widely divergent characteristics,” including issuer credit quality, letter of credit quality, type of liquidity support facility, revenue source, economic sector, and size. Edelweiss alleges that the “significant differences” in these factors “should have led to differences in their interest rate changes over time,” and that defendants could not have complied with their rate resetting obligation given

12Edelweiss states that it selected these threshold qualifiers “to account for the periodic sale of certain bonds or their departure from the market,” and that applying them “most accurately captures Defendants’ anomalous interest rate resetting practices and the extent of the intentional interest rate inflation in which Defendants have engaged.” If Edelweiss had employed different qualifying thresholds—if the required period were something other than 26 consecutive weeks, or if the rate change were required to match the most common rate change something other than 80 percent of the time—then the percentage of VRDOs that were “bucketed,” the number of buckets, and the number of VRDOs in each bucket, would likely have been different as well.

22 that they applied matching rate resets for a months-long period across 841 bucketed California VRDOs. Defendants contend that this analysis is inadequate for two principal reasons. First, they argue that because Edelweiss is unable to allege how defendants actually reset rates, there is no basis to infer that defendants failed to exercise their judgment when determining the lowest rate, or that the rate of any given California VRDO was too high. Second, they contend that the analysis does not directly identify any California VRDO for which the rate was reset higher than it should have been on any particular date. We are not persuaded by defendants’ first argument because we must credit Edelweiss’s allegations that (1) the “buckets” consisted of VRDOs that differed across the multiple characteristics Edelweiss identified; (2) these characteristics (or investor demand for them) change in different ways over time; (3) the lowest interest rate that would enable the series to be sold at par in a given week is a function of these changes from week to week (i.e., prevailing market conditions for that particular VRDO); and (4) given the dynamics of the market, one could not reasonably expect VRDOs with vastly different characteristics to require an identical interest rate adjustment in absolute terms week after week if the objective were to select the lowest interest rate for each VRDO that would enable the series to be sold at par. Defendants suggest that “Edelweiss’s point is a little like observing that the prices for regular and premium gasoline moved together in a given time period, and that those products have different characteristics.” Since “the prices are already different in absolute terms,” defendants write, “identical price changes do not mean that the price for regular gasoline was inflated.” But if a plaintiff alleged that the ingredients for regular and premium were very different and the cost of each ingredient changed in different ways from

23 week to week, yet the price charged for regular or premium rose or fell by the same amount, it would be reasonable to infer that the price of at least one of the two did not accurately reflect changes in the cost of its ingredients. Moreover, Edelweiss alleges that the interest rate change applied by a defendant to every VRDO in a given bucket was “calibrated to ensure the lowest quality, riskiest VRDO in the bucket” would trade at par, increasing the likelihood that lockstep adjustments would result in an interest rate that was higher than necessary for some of the VRDOs in the bucket. It is true, as defendants point out, that Edelweiss’s definition of a “bucket” allows non-matching rate changes up to 20 percent of the time—a threshold that is not particularly well explained—and nothing in the seventh amended complaint explains why or under what circumstances a VRDO would not receive a matching rate change. But Edelweiss need not allege defendants’ exact rate resetting procedure if the allegations it does offer support the inference that defendants were knowingly failing to reset rates on some VRDOs at the lowest possible level, as we believe they do. The same is true for defendants’ contention that Edelweiss “cherry-picked” the period it studied. Such an argument may have force in the context of a summary judgment motion, but at the pleading stage we cannot say that a period of twenty-six weeks is so short as to be meaningless. Ultimately defendants may be able to offer an “innocent” explanation for the patterns Edelweiss discerns in the interest rate data, but that explanation, if it exists, likely relies on information that we cannot consider when ruling on a demurrer, or depends on rejecting allegations that for now we are required to credit. (See Perdue v. Crocker National Bank (1985) 38 Cal.3d 913, 922.) As for defendants’ second argument, we agree that the bucketing analysis does not identify specific VRDOs for which Edelweiss alleges the

24 interest rate was too high. But we do not view the bucketing analysis in a vacuum; Edelweiss supplied these allegations in its analysis of credit rating upgrades, to which we now turn. b. Study of Credit-Rating Upgrades Edelweiss alleges that it performed a study of credit-rating upgrades for California VRDO issuers to determine whether an upgrade in the credit rating of a VRDO issuer resulted in a decrease in the interest rate on that VRDO, which it alleged should occur “[a]ll else being equal.” Edelweiss identifies approximately three dozen examples—Exhibit K to the seventh amended complaint, which has 39 entries—in which an upgrade by Moody’s (a credit rating agency) did not result in a relative interest rate decrease for that VRDO, determined by the first rate reset after the upgrade. In other words, while the interest rate on the VRDO may have fallen, the decrease was not greater (in absolute terms) than the changes defendant applied to the interest rates “on the VRDOs of other issuers whose credit ratings were not upgraded.” Defendants argue that the credit study allegations are flawed because Edelweiss failed to include allegations assessing “other characteristics” of the bonds and “market forces.” But a demand that Edelweiss exhaustively detail its analysis in the complaint—effectively leaving nothing to explore in discovery—makes too much of the particularity requirement. Edelweiss alleges that the study provides “dozens of specific instances in which a Defendant set the interest rate of a VRDO at a level higher than it should have been, taking the relevant circumstances into consideration, including the characteristics of the VRDO at issue and market preferences for it.” Each such alleged instance is set forth in Exhibit K. These allegations are sufficient, as they “directly identify” 39 particular rate resets for particular

25 VRDOs where defendants purportedly violated their rate resetting obligation. (McCann, supra, 191 Cal.App.4th at p. 910.)13 c. Former Employee Statements Edelweiss alleges that “[s]even former employees of Defendants have stated and corroborated that Defendants shirked their contractual and regulatory obligations and instead engaged in the rate-setting misconduct that Relator’s forensic analyses revealed.” Employee 4 worked on the remarketing desk at Wells Fargo from approximately 2008 through 2015. This employee described the rate resetting process as follows: when Wells Fargo initially became the remarketing agent for a California VRDO, it determined by how many basis points the VRDO’s interest rate should differ relative to the SIFMA index. Thereafter, Wells Fargo “almost never” made adjustments to the spread. The weekly rate resetting involved adjusting “large groups of VRDOs, including California VRDOs, by the same absolute number of basis points. As a result, large groups of VRDOs were adjusted by the same number of basis points regardless of the percentage increase, or decrease, that the number of basis points represented.” Employee 4 stated that this process “did not result in the lowest rate for all the VRDOs for which Wells Fargo was the remarketing agent, which was required to be obtained by the remarketing agreements.” Instead, it “was designed to ensure that VRDOs, including California VRDOs,

13 We note that only some of the bonds in Exhibit K are elsewhere identified as having been bucketed during the 2009 to 2013 period Edelweiss analyzed, and some of the dates listed in Exhibit K are outside that period. However, even the entries that do not correspond to the bucketing analysis are relevant because the seventh amended complaint alleges that “the misconduct in which Defendants engaged is ongoing” and that Edelweiss’s “claims extend to all VRDOs for which each Defendant served as an RMA during the statutory period.”

26 were not ‘put’ back to Wells Fargo as the remarketing agent, which would require Wells Fargo to undertake inventory risk.” Employee 4 “closely monitored” the rate resetting practices of the other defendants and was aware of their methodologies; he believes they engaged in a rate resetting process that was “substantially the same.” We conclude that these statements support Edelweiss’s allegations that defendants used robo- resetting on California VRDO rates and, taken together with the bucketing and credit study allegations, support the reasonableness of the inference that defendants violated their rate-resetting obligation. We are not persuaded by defendants’ arguments to the contrary. They contend that the statements are insufficient because Employee 4 did not allege exactly how Wells Fargo determined the initial spread to the SIFMA index, or exactly how Wells Fargo determined its subsequent rate resets. Again, we deem that level of specificity to exceed what is required at the pleading stage. Defendants also argue that the example of lockstep rate resets described by Employee 4 was contradicted by other judicially noticed data. Employee 4 explained that on January 20, 2010, 32 of the 51 California VRDOs were adjusted by 10 basis points “without regard” to the relative percentage change this represented for each VRDO. Of those 32 California VRDOs, 21 had a 100 percent rate increase, 5 had a 77 percent rate increase, 5 had a 43 percent rate increase, and 1 had a 2 percent rate increase. The trial court granted defendants’ request for judicial notice of rate resets the week of November 11, 2009 for these VRDOs. Defendants argue that the rate resets in this week for the 32 California VRDOs referenced above “contradicted” Employee 4’s statement that Wells Fargo “almost never” made adjustments to the SIFMA index spread. But the same cherry-picking complaints made by defendants can be applied here: a single week of rate

27 reset data does not render the allegation by Employee 4—that Wells Fargo rarely adjusted the SIFMA index spread—defective on its face. With this judicially noticed data, defendants provide their counter analysis of a different time period using different methodology and reaching a different result. Nor does this data necessarily contradict the general allegation that defendants used bucketing to reset rates: a majority of the 32 California VRDOs had either a 1 or 7 basis point adjustment the week of November 11, 2009. Employee 5 is a former senior employee who worked in the remarketing function for Barclays from approximately 2008 through 2012. Employee 5 alleges that Barclays “did not expend sufficient resources to staff the remarketing function in a manner that would have resulted in Barclays’ setting rates on the VRDOs for which it was the remarking agent at the lowest rate,” and that Employee 5 was “regularly ‘yelled at’ ” by management that VRDO rates were being reset too low and should be increased. Similarly, Employee 7 (employed by Citi for 20 years, and then an unnamed financial institution from 2006 to 2016) believed that the VRDO market was operated by remarketing agents as the “ ‘biggest joke of a market of all time’ ” and did not operate on the basis of prevailing market conditions. Again, while not sufficient to state a cause of action in and of themselves, the statements by Employees 5 and 7 add allegations regarding the feasibility of defendants’ compliance with their rate resetting obligations, and if all Edelweiss’s allegations were proven true, support the reasonableness of the inference that those obligations were violated.14

The statements by Employee 6 relate mostly to the commercial paper 14

comparison that we deem insufficient, and we do not find that they add anything meaningful.

28 Employees 1 and 2 worked at JPMorgan before the period at issue in the seventh amended complaint (2009 onwards). The same is true for Employee 3, who worked at Citi from the early 1990s through 2008. Thus, while these allegations may add little to what has already been said, the employees described practices consistent with those in the period at issue in the complaint. According to Employee 1, the VRDOs for which the employee was responsible “were divided into a small number of groups (two to four)” called “buckets,” each of which was reset based on a spread to the SIFMA index that remained static for long periods of time.15 Employee 2 described a similar process, and stated that JPMorgan “did not engage in any individualized consideration of any particular VRDOs when setting rates.” According to this employee, the joke within the business unit was that “the whole remarketing desk could be replaced with three monkeys.” Employee 3 stated that Citi and other remarketing agents “generally did not consider VRDOs on an individualized basis, but rather reset rates en masse,” and that issuers would have a “coronary” if they were aware that Citi “was not watching VRDO rates carefully in order to get the lowest possible rates.” Taken together, these statements add support for the inference from the rate-resetting data that defendants did not evaluate, for each VRDO, the factors that Edelweiss alleges it is necessary to consider in order to reset

15 The trial court wrote that it saw no allegation that JPMorgan “adjusted the rates of large groups of unrelated California bonds,” as bonds “were divided into a small number of groups.” But we read Employee 1’s statement to mean that JPMorgan divided all of its VRDOs into only two to four buckets, and then applied the same rate reset for each bucket. This allegation is consistent with the bucketing allegations regarding JPMorgan’s practices during the time period at issue in the seventh amended complaint, as well as the practices of other defendants.

29 rates at the lowest possible level that would allow the series to be sold at par, and that their failure to do so resulted in rates that were too high. 5. Conspiracy Claim In addition to its claim that defendants violated the CFCA by submitting false claims, Edelweiss also alleges that defendants conspired to commit a violation of the CFCA by colluding to inflate interest rates on California VRDOs. (§ 12651, subd. (a)(3).) “To support a conspiracy claim, a plaintiff must allege the following elements: ‘(1) the formation and operation of the conspiracy, (2) wrongful conduct in furtherance of the conspiracy, and (3) damages arising from the wrongful conduct.’ ” (AREI II Cases (2013) 216 Cal.App.4th 1004, 1022.) A conspiracy claim under the CFCA is subject to the same heightened standard of pleading with particularity. (McCann, supra, 191 Cal.App.4th at p. 906; see also Prakashpalan v. Engstrom, Lipscomb & Lack (2014) 223 Cal.App.4th 1105, 1136 [“Where fraud is alleged to be the object of the conspiracy, the claim must be pleaded with particularity”].) Defendants argue that Edelweiss’s conspiracy claim suffers from a “threshold failure” because its underlying allegations for CFCA violation are not sufficient. We reject this argument for all the reasons described above. Defendants then go on to challenge Edelweiss’s specific allegations regarding conspiracy. Edelweiss alleges that defendants’ conspiracy is evident by the “cross- bank bucketing” of VRDO interest rate resets showing “the same kind of lock- step price movements even across banks.” Edelweiss provides two examples of cross-bank buckets: (1) 440 VRDOs from JPMorgan, Citi, Morgan Stanley, Wells Fargo, and Bank of America with an average maximum 26-week matching rate for the 2009 to 2013 period of 95 percent; and (2) 231 VRDOs

30 from Citi, Morgan Stanley, Wells Fargo, and Bank of America with a matching rate of 91.2 percent. Edelweiss also reviewed the rates for 10 VRDOs with dissimilar characteristics selected at “random” for the year 2012. Edelweiss alleges that the interest rate changes were “almost identical” across the board. In response to these allegations, defendants repeat their argument that Edelweiss was required to allege a rate for particular VRDOs that should have moved differently in a particular week. But Edelweiss has alleged a coordinated robo-resetting scheme for at least 710 particular VRDOs (these two examples and the credit-rating upgrade study) in a specified time period, and that this scheme resulted in rates that were higher than they should have been. These allegations are sufficient to draw a reasonable inference of wrongful conduct in furtherance of the conspiracy. Edelweiss offers three other allegations regarding defendants’ purported conspiracy.16 First, it alleges that Moody’s downgraded the short- term credit rating of Bank of America in June 2012. Despite the downgrade, VRDOs with Bank of America credit “continued to move in lock step with the other VRDOs in this bucket.” Edelweiss alleges that this was caused by defendants’ agreement to “ignore the downgrade” and continue its coordinated pricing. Second, Edelweiss alleges that defendants used indexing services like J.J. Kenny to exchange information about future VRDO rate- setting and adjust planned rates. Third, Edelweiss alleges that defendants have opportunity and incentive to inflate VRDO rates because they serve as a

16 Edelweiss also alleges that defendants coordinated their rate resets after the Federal Reserve’s decision in December 2015 to raise the target for the federal funds rate by 25 basis points. Because this allegation depends on a comparison to commercial paper rates at the time, we do not consider it sufficient.

31 letter of credit provider on VRDOs where another defendant was the RMA, and defendant RMAs own money market funds that invest in the same VRDOs. Defendants argue that these allegations lack particularity. Evidence of a conspiracy “ ‘ “ ‘ “may be inferred from the nature of the acts done, the relation of the parties, the interests of the alleged conspirators, and other circumstances.” ’ ” ’ ” (Novartis Vaccines & Diagnostics, Inc. v. Stop Huntingdon Animal Cruelty USA, Inc. (2006) 143 Cal.App.4th 1284, 1296.) While defendants’ alleged interrelationships, opportunities to collude, and incentives to do so may not be alone sufficient to plead conspiracy, they again support the reasonableness of the inference drawn from all of Edelweiss’s allegations that defendants conspired in robo-resetting California VRDO rates. (See Balistreri v. Turner (1964) 227 Cal.App.2d 236, 242 [defendant had motive because plaintiff supported opponent in reelection to union]; AREI II Cases, supra, 216 Cal.App.4th at pp. 1008, 1023 [reasonable inference that defendant investment bankers knew officer was a convicted felon because they were extensively involved in background checks and drafting related memorandum].) In sum, we conclude that Edelweiss has adequately pleaded its CFCA claims based on the alleged violation of defendants’ contractual obligations and alleged conspiracy to commit such violation. We thus conclude that the trial court erred in sustaining defendants’ demurrer to the seventh amended complaint. 6. CFCA Public Disclosure Bar Defendants argue that even if Edelweiss has pled its claims with sufficient particularity, those claims are independently foreclosed by the CFCA’s public disclosure bar. Defendants concede that this argument was

32 not repeated in their demurrer to the seventh amended complaint, but state that it was expressly reserved for appeal after the trial court rejected the argument on defendants’ previous (successful) demurrers. Upon appeal of a final judgment, we may review “any intermediate ruling, proceeding, order or decision which involves the merits or necessarily affects the judgment or order appealed from or which substantially affects the rights of a party.” (Code Civ. Proc., § 906.) Under these circumstances, and given that Edelweiss has not argued or demonstrated any prejudice, we address the merits of the argument. After the parties submitted their briefing on appeal, the Attorney General filed a notice opposing dismissal of the action pursuant to the public disclosure bar. Section 12652(d)(3)(A) provides for the dismissal of an action under the public disclosure bar “unless opposed by the Attorney General or prosecuting authority of a political subdivision.” Accordingly, if the Attorney General files an opposition to a pending motion to dismiss in the trial court, the notice forecloses application of the public disclosure bar. (Ibid.) But because the Attorney General filed its notice here for the first time on appeal, it raises the question of whether the opposition has been forfeited. (See Holmes v. California Nat. Guard (2001) 90 Cal.App.4th 297, 319, fn. 13 [government defendants waived argument never raised or asserted below].) We need not decide this question, however, as we conclude that the public disclosure bar does not apply.17 We begin by addressing two threshold issues. Defendants first contend, without citation to any California authority, that the original complaint is the operative pleading for purposes of the

17Given this conclusion, we need not consider defendants’ opposition to the Attorney General’s notice, and we deny the application by Edelweiss and joinder by the Attorney General for leave to file a reply.

33 public disclosure bar. We reject this contention in light of California cases analyzing application of the public disclosure bar to the allegations in amended complaints. (E.g., Bartlett, supra, 243 Cal.App.4th at p. 1402 [SEC filings cited in first amended complaint did not trigger application of public disclosure bar]; City of Hawthorne ex rel. Wohlner v. H&C Disposal Co. (2003) 109 Cal.App.4th 1668, 1686 [public disclosure bar not applicable to claims in fourth amended complaint].) Even considering the allegations of the seventh amended complaint, defendants maintain that the action is barred because it relies on publicly disclosed data. Specifically, defendants contend that the VRDO interest rate resets underlying Edelweiss’s bucketing analysis were made available in real time and for later review on EMMA. They also contend that the commercial paper comparison relied on information available on Bloomberg and the Federal Reserve Economic Data (FRED) website. In order for the public disclosure bar to apply, however, defendants concede that the information must be “material” to the claim. (State of California v. Pacific Bell Telephone Co. (2006) 142 Cal.App.4th 741, 745 [qui tam action cannot proceed if the material elements of a false claim are already in the public domain].) Given our conclusion that the commercial paper allegations are insufficient to support Edelweiss’s claim, information from Bloomberg and FRED cannot support application of the public disclosure bar here. That leaves us to decide whether the interest rate reset information on EMMA triggers application of the public disclosure bar: that substantially the same allegations or transactions as alleged in the action were publicly disclosed in the certain enumerated categories under section 12652, subdivision (d)(3)(B). We turn to those categories, as they “limit our review.” (State of California v. Pacific Bell Telephone Co., supra, 142 Cal.App.4th at

34 p. 749.) “ ‘If those fora are not implicated the inquiry is at an end.’ ” (Bartlett, supra, 243 Cal.App.4th at p. 1407.) Defendants argue that EMMA falls under either of two section 12652, subdivision (d)(3)(B) categories: as a “report” of the state, or as “news media.” We are not persuaded. The interest rate reset information on EMMA is clearly not a “report” of the state; it is provided by RMAs and made available by MSRB, a non-governmental self-regulatory organization. (BOKF, NA v. Estes (9th Cir. 2019) 923 F.3d 558, 561.) Defendants cite no authority to the contrary. (Cf. Rosenberg, supra, 487 Mass. at p. 458 [concluding that official statements containing remarketing agreements constituted “reports”]; Schindler Elevator Corp. v. United States ex rel. Kirk (2011) 563 U.S. 401, 414 [written responses by Department of Labor to Freedom of Information Act requests are federal “reports” under federal FCA].) Whether the interest rate reset information on EMMA constitutes a public disclosure in “news media” presents a closer question of statutory interpretation. In answering such a question, “ ‘our primary task is to determine the lawmakers’ intent’ ” and the process “to ascertain that intent may involve up to three steps.” (MacIsaac v. Waste Management Collection & Recycling, Inc. (2005) 134 Cal.App.4th 1076, 1082.) First, we look to the words of the statute itself as “chosen language is the most reliable indicator of its intent.” (Id. at p. 1082.) “If the statutory language is clear and unambiguous, our task is at an end, for there is no need for judicial construction.” (Id. at p. 1083.) When the plain meaning of the text does not resolve the question, we proceed to the second step and turn to maxims of construction and extrinsic aids, including legislative history materials. (Ibid.) If ambiguity remains, we “must cautiously take the third and final

35 step” and “apply ‘reason, practicality, and common sense to the language at hand.’ ” (Id. at p. 1084.) While the plain language of section 12652(d)(3)(A) is not dispositive here, it supports the conclusion that interest rate reset information on EMMA is not a disclosure in “news media.” The term “news media” is not defined anywhere in the CFCA. When a term goes undefined in a statute, we give the term its “broad ordinary meaning.” (Schindler Elevator Corp. v. United States ex rel. Kirk, supra, 563 U.S. at p. 408.) We agree with defendants that “news media” must be interpreted “to encompass the many ways in which people in the modern world obtain financial news, including from publicly available websites on the Internet.” (Rosenberg, supra, 487 Mass. at p. 460.) The term clearly includes online financial news sources “providing summaries or analysis of trends in market transactions.” (Ibid.) But we depart from Rosenberg in its comparison of such sources to the online repository of interest rate reset data on EMMA; it is not the same thing. Bartlett supports our view. In that case, the defendants argued that information in SEC filings qualified as a public disclosure by the “news media” because those filings were accessible on the SEC’s online public database. (Bartlett, supra, 243 Cal.App.4th at p. 1414.) Bartlett explained: “To be sure, the advent of online news sites, blogs and social media has blurred the line between what has traditionally been considered news media and other forms of public discussion. . . . Still, wherever that fuzzy line now is between news media and some other form of publicly accessible information, we have little difficulty concluding that disclosures in forms available only on the SEC’s online public database are not disclosures by the news media no matter how broadly that term is interpreted.” (Ibid.) The SEC database at issue in Bartlett is an apt comparison: like EMMA, it is an

36 automated collection of information that entities are required to submit, and its purpose is to accelerate the receipt, dissemination, and analysis of this time-sensitive information for the benefit of investors, corporations, and the market. (Id. at p. 1414, fn. 9.) In ascertaining the meaning of language in a statute, courts also turn to general dictionaries but “must exercise ‘great caution’ when relying on a dictionary definition of a common term to determine statutory meaning because a dictionary ‘ “is a museum of words, an historical catalog rather than a means to decode the work of legislatures.” ’ ” (A.S. v. Miller (2019) 34 Cal.App.5th 284, 293, fn. 4; see also De Vries v. Regents of University of California (2016) 6 Cal.App.5th 574, 591.) While the term “news media” is not commonly defined in dictionaries, the term “news” has been defined as a “ ‘report of a recent event,’ ” or “ ‘the presentation of a report on recent or new events in a newspaper or other periodical or radio or television.’ ” (Rosenberg, supra, 487 Mass. at p. 459, quoting Webster’s New Universal Unabridged Dictionary 1295 (2003).) “Media” has been defined as “ ‘[t]he means of communication, as radio and television, newspapers, and magazines, that reach or influence people widely.’ ” (Ibid.) Read together, the term “news media” includes “ ‘methods of communication that are used to convey information about recent events or other information that would commonly be found in newspapers, news broadcast, or other news sources.’ ” (Silbersher v. Allergan Inc. (2020) 506 F.Supp.3d 772, 806, revd. on other grounds (9th Cir. 2022) 46 F.4th 991.) “ ‘Accordingly, the extent to which the information typically conveyed by a source would be considered newsworthy is relevant to whether it is a news media source.’ ” (Ibid.) Again, we see no basis to conclude that an online repository containing defendants’ daily or

37 weekly submission of interest rate reset data would be considered generally newsworthy. Our interpretation is also consistent with the structure of section 12652(d)(3)(A), which explicitly limits its application to public disclosures in particular fora. “It is a maxim of statutory interpretation that courts should give meaning to every word of a statute and should avoid constructions that would render any word or provision surplusage.” (Tuolumne Jobs & Small Business Alliance v. Superior Court (2014) 59 Cal.4th 1029, 1038.) “ ‘An interpretation that renders statutory language a nullity is obviously to be avoided.’ ” (Id. at p. 1039.) If the interest rate data here were considered a disclosure by “news media” simply because EMMA is a publicly available website, it would effectively swallow the fora limitations of section 12652(d)(3)(A). As Bartlett explained, “the Legislature may well conclude it should amend CFCA to include information available through the Internet as one of the categories within the public disclosure bar,” but “we have no power to do what the Legislature has to date chosen not to do.” (Bartlett, supra, 243 Cal.App.4th at p. 1414.) Finally, the legislative history of section 12652(d)(3)(A) further supports the conclusion that interest rate information on EMMA is not a disclosure in “news media.” As originally enacted in 1987, the provision did not yet allow for opposition by the Attorney General or prosecutor, but contained the same prohibition on qui tam actions “based upon the public disclosure of allegations or transactions in a criminal, civil, or administrative hearing, in an investigation, report, hearing, or audit conducted by or at the request of the Senate, Assembly, auditor, or governing body of a political subdivision, or from the news media.” (Stats. 1987, ch. 1420, § 1.) This provision was “intended to reward the first whistleblower, and to prevent

38 other ‘bandwagon’ whistleblowers from reaping the benefits of the disclosure.” (Sen. Com. on Judiciary, Analysis of Assem. Bill No. 1441 (1987– 1988 Reg. Sess.) as amended July 9, 1987, p. 8.) In other words, the Legislature understood that qui tam actions would help protect the public treasury, and thus implemented the public disclosure bar to protect the incentives for a qui tam plaintiff to “aid the Government in ferreting out fraud.” (Assem. Com. on the Judiciary, Hearing on A.B. 1441, the California False Claims Act (May 6, 1987), testimony of David Huebner, p. 3.) An online repository of interest rate reset information does not, in our view, constitute the type of disclosure in “news media” and subsequent “bandwagon” problem the Legislature sought to preclude. The CFCA was amended in 2012 to conform its provisions to the federal FCA: section 12652(d)(3)(A) now allows the Attorney General or prosecutor to permit qui tam lawsuits based on publicly disclosed information. (Stats. 2012, ch. 647, § 3.) The three fora categories of section 12652(d)(3)(A) remained unchanged, as they already mirrored the federal FCA. (31 U.S.C. § 3730, subd. (e)(4)(A).) We agree with defendants that we can turn to published federal authority for guidance in interpreting the CFCA’s public disclosure bar (Pomona, supra, 89 Cal.App.4th at p. 802), but the cases they cite do not alter our conclusion. United States ex rel. Osheroff v. Humana, Inc. (11th Cir. 2015) 776 F.3d 805, for example, applied the federal FCA public disclosure bar to a qui tam action based on advertisements contained in newspapers and publicly available websites of health clinics that were intended to disseminate information about the services and programs provided by the clinic. (Id. at p. 813.) United States ex rel. Green v. Service Contract Education & Training Fund (D.D.C. 2012) 843 F.Supp.2d 20 similarly concluded that a promotional page of an organization’s website

39 containing information about its operation and training program constituted “news media” for the purposes of the federal FCA. (Id. at p. 32.) Unlike the websites in Osheroff and Green, the interest rate reset data on EMMA is not promotional information. And none of the cases cited by defendants support their position that all publicly available websites are “news media.” (Cf. United States ex rel. Hong v. Newport Sensors, Inc. (2018) 728 Fed.Appx. 660, 662–663 [declining to address argument as relator did not independently challenge holding in unpublished district court decision that most public webpages are “news media”].) Indeed, federal courts have explicitly rejected that view. (E.g., Silbersher v. Allergan, Inc., supra, 506 F.Supp.3d 772, 807 [rejecting argument that information disclosed on patent website falls under “news media” category “simply because it can be found on the Internet”].) We agree with Bartlett that, while the Internet has certainly expanded the meaning of “news media” to include certain information publicly available online, it does not include the information at issue here. (Bartlett, supra, 243 Cal.App.4th at p. 1414.) In sum, we conclude that the interest rate reset information from EMMA is not a public disclosure from “news media” as contemplated by section 12652(d)(3)(A).18 The public disclosure bar is thus inapplicable and does not require dismissal of Edelweiss’s CFCA claim.

18 Given this conclusion, we need not address Edelweiss’s alternative arguments that the public disclosure bar does not apply here because (1) the interest rate data from EMMA did not disclose “ ‘substantially the same allegations or transactions’ ” as alleged in its action; and (2) Edelweiss falls within the “ ‘original source’ ” exception.

40 DISPOSITION The June 25, 2021 judgment is reversed and the case is remanded for further proceedings. Edelweiss is entitled to its costs on appeal. GOLDMAN, J.

WE CONCUR:

STREETER, Acting P. J. WHITMAN, J.*

*Judge of the Superior Court of California, County of Alameda, assigned by the Chief Justice pursuant to article VI, section 6 of the California Constitution.

41 Trial Court: City and County of San Francisco Superior Court

Trial Judge: Honorable Anne-Christine Massullo

Counsel for Plaintiffs and STEYER LOWENTHAL Appellants: BOODROOKAS ALVAREZ & SMITH, Allan Steyer, Jill M. Manning, Jill K. Cohoe THE LAWRENCE LAW FIRM, Jeffrey W. Lawrence SCHNEIDER WALLACE COTTRELL KONECKY, Todd M. Schneider, Jason H. Kim, Matthew S. Weiler, James A. Bloom CONSTANTINE CANNON, Ari Yampolsky Counsel for Defendant and JONES DAY, Michael P. Conway, Respondent Wells Fargo Bank, Matthew J. Silveira, Margaret Adema N.A.: Maloy Counsel for Defendant and GREENBERG TRAURIG, William J. Respondent JPMorgan Chase Goines Bank, N.A. and J.P. Morgan Securities LLC Counsel for Defendant and WILLIAMS CUTLER PICKERING Respondent Bank of America HALE AND DORR, Matthew Benedetto Corporation, Bank of America N.A., and Merrill Lynch, Pierce, Fenner & Smith Inc. Counsel for Defendants and KEESAL, YOUNG & LOGAN, Peter R. Respondents Citigroup Global Boutin, Christopher A. Stecher Markets Inc., Citibank N.A., Citigroup Financial Products Holdings Inc., and RBC Capital Markets, LLC Counsel for Defendants and SIDLEY AUSTIN, Matthew J. Dolan Respondents Morgan Stanley & Co. LLC, Morgan Stanley Bank, N.A., Morgan Stanley Capital Services Inc., and Morgan Stanley Capital Group Inc.

42 Counsel for Defendant and KEESAL, YOUNG & LOGAN, Ben Respondent Piper Jaffray & Co., Suter and Piper Jaffray Financial Products Inc. Counsel for Defendants and SKADDEN, ARPS, SLATE, MEAGHER Respondents Barclays Capital Inc. & FLOM, Jack P. DiCanio, Kasonni M. and Barclays Bank PLC Scales

Counsel for Chamber of Commerce KING & SPALDING, Ethan P. Davis, of the United States of America, Matthew V.H. Noller California Chamber of Commerce, and American Bankers Association as Amici Curiae on behalf of Respondents