The City of Philadelphia v. Bank of America Corporation

Background

These consolidated proposed class actions were brought by the City of Philadelphia, the Mayor and City Council of Baltimore, and the Board of Directors of the San Diego Association of Governments, acting as the San Diego Regional Transportation Commission. They sued eight banks—Bank of America, Barclays, Citigroup, Goldman Sachs, JPMorgan Chase, Morgan Stanley, the Royal Bank of Canada, and Wells Fargo—along with various related entities.

The plaintiffs alleged that the banks violated federal antitrust law and state contract and fiduciary-duty laws by conspiring from February 1, 2008, through November 30, 2015, to keep interest rates artificially high on Variable Rate Demand Obligations, or VRDOs. VRDOs are long-term bonds issued by municipalities and other public or charitable entities whose interest rates are generally reset weekly or daily. The plaintiffs alleged that the banks, which served as remarketing agents, coordinated through communications and third-party pricing services rather than competing to set the lowest possible rates.

The plaintiffs moved to certify a nationwide class and a related contract subclass under Rule 23 of the Federal Rules of Civil Procedure. They relied on reports from Dr. William Schwert and Dr. Rosa Abrantes-Metz. The banks moved under Rule 702 of the Federal Rules of Evidence and the Supreme Court’s decision in Daubert to exclude some or all of those experts’ testimony. The banks also opposed class certification, principally arguing that individual issues would prevent common questions from predominating.

Expert-evidence motions

The court denied the banks’ motion concerning Dr. Schwert. He offered two regression models intended to measure class-wide impact and damages: a multiple-dummy-variable model and a backcasting model. The court found that the models used reasoned and supported methodological choices, including variables addressing systemic economic conditions and features of individual VRDOs.

The banks argued that Dr. Schwert failed to account for the Financial Crisis, the European Sovereign Debt Crisis, supply and demand, inventory levels, and other factors. The court concluded that these objections primarily concerned the accuracy and weight of the models rather than their admissibility. The court also rejected the argument that the models were shown to produce a systemic number of false positives. It acknowledged that the model effectively produced false positives during March 2020, but found that those four weeks represented less than one percent of the relevant periods and occurred outside the alleged conspiracy period.

The court also rejected the banks’ argument that the models masked too many uninjured class members. Dr. Schwert’s analysis estimated that less than two percent of VRDOs never had an inflated rate, which the court found to be a small enough percentage at this stage. The court emphasized that the banks relied in part on modified versions of Dr. Schwert’s models created by their rebuttal expert rather than showing that Dr. Schwert’s models themselves were unworkable.

The court likewise rejected the challenge to Dr. Abrantes-Metz’s testimony. Her analysis examined the VRDO market’s structure and concluded that a conspiracy could have been effective and that the banks had a mechanism for coordinating through base rates. Her rate study also used qualitative and quantitative analysis to examine the relationship between the banks’ base rates and VRDO rates. The court found her responses to the banks’ criticisms thorough and reasoned and concluded that her models were capable of showing common class-wide impact.

Class certification

The court held that the plaintiffs satisfied the predominance requirement of Rule 23(b)(3). The banks did not dispute that the other Rule 23 requirements—numerosity, commonality, typicality, adequacy of representation, ascertainability, and superiority—were met. The central common questions included whether the banks conspired and whether that alleged conspiracy caused VRDO rates to be inflated.

The banks argued that rate-setting involved individualized consideration of credit quality, inventory levels, historical performance, bond size, tax status, and industry sector. The court held that those arguments did not defeat certification because the plaintiffs’ expert testimony provided common evidence of class-wide injury. Whether the alleged conspiracy, rather than other factors, caused any overcharge remained a question for the merits stage.

The court also rejected the banks’ arguments concerning synthetic fixed-rate transactions, which can combine VRDOs with interest-rate swaps. The court acknowledged that individualized questions might arise concerning which bonds involved swaps, when the bonds and swaps were entered, and whether an issuer was fully protected from VRDO rate increases. But the banks did not identify how many class members were actually unharmed. The court treated those questions as closely connected to damages and held that they did not defeat predominance.

The court rejected the banks’ argument that conduit bond issuances made class membership and direct-payor identification unmanageable. It characterized that argument as an administrative-feasibility objection and held that Rule 23 did not require the proposed method of identifying class members to be easy or administratively perfect. The court also rejected the timeliness argument, concluding that common questions concerning concealment could predominate over individual questions about class members’ knowledge or diligence. It noted that subclasses could be created later if individual issues required them.

Order and effect

The court granted the plaintiffs’ motion to certify a nationwide class consisting of persons and entities that directly paid interest expenses on qualifying VRDOs under remarketing agreements with the defendants from February 1, 2008, through November 30, 2015. It excluded the defendants and their employees, affiliates, parents, subsidiaries, co-conspirators, and the United States government.

The court also granted certification of the contract subclass consisting of persons and entities that were parties to qualifying remarketing agreements with a counterparty defendant during the same period, subject to the same exclusions. It granted the request to appoint Quinn Emanuel Urquhart & Sullivan LLP, Wollmuth Maher & Deutsch LLP, and Susman Godfrey LLP as class counsel.

The court’s conclusion states that the class-certification motion was granted and the defendants’ motion to preclude Dr. Schwert and Dr. Abrantes-Metz was denied. The court directed the plaintiffs to submit proposed notice and opt-out procedures and a letter addressing sealing and redaction requests. It also directed the clerk to terminate ECF Nos. 362 and 386.