No. 25-4458

D.C. No. 2:24-cv-05028MWF-JC

Appeal from the United States District Court for the Central District of California Michael W. Fitzgerald, District Judge, Presiding

Argued and Submitted April 24, 2026 Pasadena, California

Filed August 19, 2026Page 2 Before: Stephen A. Higginson, Jacqueline H. Nguyen, and Daniel A. Bress, Circuit Judges. *

Opinion by Judge Bress

SUMMARY **

The panel reversed the district court's order denying PennyMac Mortgage Investment Trust’s motion to dismiss shareholder Roberto Verthelyi’s putative class action alleging a violation of California’s Unfair Competition Law (“UCL”), Cal. Bus. & Prof. Code § 17200, which prohibits “any unlawful, unfair or fraudulent business act or practice.”

In 2023, the London Inter-Bank Offered Rate (“LIBOR”), a widely used rate index governing trillions of dollars’ worth of financial contracts, was discontinued. To replace LIBOR, the Federal Reserve Bank selected the Secured Overnight Financing Rate (“SOFR”). To address any problems arising from the transition from LIBOR to SOFR, Congress passed the Adjustable Interest Rate (LIBOR) Act, which provides that existing contracts will be governed by SOFR unless the contract contains an acceptable fallback 'benchmark replacement'-that is, a Page 3permissible rate or index of rates to apply if LIBOR is unavailable.

PennyMac, a real estate investment trust, offered two sets of shares to investors governed by corporate charter documents, known as Articles Supplementary. If the LIBOR rate was not available, the Articles provided for three cascading fallbacks. PennyMac concluded that the third fallback provision—providing that the dividend rate shall be calculated at the dividend rate in effect for the immediately preceding Dividend Period—should apply. Verthelyi alleged that PennyMac’s failure to convert its rates to SOFRbased rates was both “unlawful” and “unfair” under the UCL because a fixed rate cannot serve as a benchmark replacement under the LIBOR Act.

The panel held that PennyMac’s third fallback rate, as set forth in the Articles, was “a benchmark, or an interest rate or dividend rate” under the LIBOR Act, 12 U.S.C. § 5802(3). Therefore, Verthelyi’s argument failed under the plain text of the statute. There is no requirement that a 'benchmark replacement' be a floating rate. Although Congress replaced LIBOR with SOFR when the contracting parties failed to provide adequate contingencies if LIBOR became unavailable, the LIBOR Act did not purport to override private contractual agreements that contained a valid benchmark replacement. The panel concluded that PennyMac’s application of the third fallback provision was not 'unlawful' under the UCL. The panel also held that Verthelyi could not obtain relief under the “unfair” prong of the UCL.

Accordingly, the panel held that PennyMac's third fallback provision was not an inadequate benchmark replacement under the LIBOR Act simply because it Page 4resolves to a fixed rate, and the district court erred in denying PennyMac’s motion to dismiss on that ground. The panel remanded to the district court so that the parties may advance any further arguments they might have regarding whether PennyMac's application of the third fallback provision violates the LIBOR Act.

Catherine Pratsinakis (argued), Dilworth Paxson LLP, Philadelphia, Pennsylvania; Nicole Lavallee and Daniel E. Barenbaum, Berman Tabacco, San Francisco, California; for Plaintiff-Appellee.

Steven M. Farina (argued) and Melissa B. Collins, Williams & Connolly LLP, Washington, D.C.; Matthew D. Umhofer and Jonas P. Mann, Umhofer Mitchell & King LLP, Los Angeles, California; for Defendants-Appellants.

Page 5BRESS, Circuit Judge:

In 2023, the London Inter-Bank Offered Rate (“LIBOR”), a widely used rate index governing trillions of dollars’ worth of financial contracts, was discontinued. To facilitate the transition from LIBOR and to prevent “disruptive litigation related to existing contracts,' 12 U.S.C. § 5801(a)(3), Congress passed the Adjustable Interest Rate (LIBOR) Act. This statute replaces LIBOR in LIBORbased legacy contracts with a Federal Reserve-selected replacement rate—but only if the contract fails to identify a suitable “benchmark replacement” for LIBOR as a fallback. Id. § 5803(a)(2)(A). This case concerns whether a fallback dividend rate provision in a shareholder agreement that resolves to a fixed rate qualifies as a “benchmark replacement' under the LIBOR Act. Id. § 5802(3). We hold that it does.

For decades, banks and other financial institutions used LIBOR as a benchmark to set interest rates. LIBOR was calculated by averaging the self-reported interest rates that certain major banks with significant operations in London anticipated being charged by their peer banks for an interbank loan. LIBOR was reported for five global currencies (the U.S. dollar, the euro, the British pound, the Japanese yen, and the Swiss franc), and seven different maturities, or tenors (overnight and one week, as well as one, two, three, six, and twelve month periods), with the threemonth U.S. dollar-denominated rate being the mostPage 6 commonly used. At its height, LIBOR was 'used as a benchmark rate in more than [$200 trillion] worth of contracts worldwide.” 12 U.S.C. § 5801(a)(1).

However, the self-reported nature of LIBOR made it easy to manipulate. After a series of regulatory investigations revealed how several large banks rigged LIBOR quotes by pressuring other institutions into reporting artificially inflated or deflated anticipated lending rates, British financial regulators announced that LIBOR would be discontinued in 2021. This deadline was later extended to 2023 for U.S. dollar-denominated LIBOR.

To replace LIBOR, the Federal Reserve Board selected what was viewed as a more reliable alternative, known as the Secured Overnight Financing Rate (“SOFR”), which began publication in 2018. SOFR is a floating interest rate keyed to the actual costs of overnight repurchasing contracts that use U.S. government bonds as collateral. In response to LIBOR's phasing out, many financial institutions began replacing LIBOR with SOFR in their existing floating-rate instruments. By January 2022, banks and financial institutions had stopped using U.S. dollar-denominated LIBOR for new loans in the United States.

But Congress foresaw that the transition from LIBOR to SOFR would create a problem, because 'a significant number of existing contracts that reference LIBOR [did] not provide for the use of a clearly defined or practicable replacement benchmark rate when LIBOR is discontinued.” 12 U.S.C. § 5801(a)(2). Congress also recognized that the cessation of LIBOR 'could result in disruptive litigation related to existing contracts that do not provide for the use of a clearly defined or practicable replacement benchmark rate.” Id. § 5801(a)(3).

Page 7To address this situation, Congress in 2022 passed the LIBOR Act. The LIBOR Act states that its purpose is “to establish a clear and uniform process, on a nationwide basis, for replacing LIBOR in existing contracts the terms of which do not provide for the use of a clearly defined or practicable replacement benchmark rate, without affecting the ability of parties to use any appropriate benchmark rate in new contracts.” Id. § 5801(b)(1). But the statute also 'allow[s] existing contracts that reference LIBOR but provide for the use of a clearly defined and practicable replacement rate, to operate according to their terms.” Id. § 5801(b)(3).

To this end, the LIBOR Act provides that existing contracts will be governed by SOFR unless the contract contains an acceptable fallback “benchmark replacement”— that is, a permissible rate or index of rates to apply if LIBOR is unavailable. Id. § 5802(1), (3), (6). The statute specifically provides that '[o]n the LIBOR replacement date,” SOFR “shall be the benchmark replacement for any LIBOR contract that . . . (1) contains no fallback provisions; or (2) contains fallback provisions that identify neither(A) a specific benchmark replacement; nor (B) a determining person.' Id. § 5803(a). Importantly for this case, the term 'benchmark replacement' is defined as 'a benchmark, or an interest rate or dividend rate (which may or may not be based in whole or in part on a prior setting of LIBOR), to replace LIBOR or any interest rate or dividend rate based on LIBOR, whether on a temporary, permanent, or indefinite basis, under or with respect to a LIBOR contract.' Id. § 5802(3) (emphasis added). The LIBOR Act emphasizes that it is not intended to “alter or impair . . . any LIBOR contract that contains fallback provisions that identify a benchmark replacement that is not based in any way on any LIBOR value.” Id. § 5803(f)(2).

Page 8In short, Congress directed that existing contracts that reference LIBOR would default to SOFR unless they contain a fallback provision that, among other things, see id. § 5803(b)(1), meets the requirements of a 'benchmark replacement.”

Defendant PennyMac Mortgage Investment Trust is a real estate investment trust managed by defendant PNMAC Capital Management, LLC (collectively, “PennyMac”). In 2017, PennyMac issued two sets of shares to investors: the Series A and Series B Fixed-to-Floating Preferred Shares. The terms for the Series A and B shares are identical in material respects and are governed by PennyMac’s corporate charter documents, known as Articles Supplementary (“Articles”). Under the Articles, shareholders were initially entitled to receive quarterly dividends at a fixed rate (8.125% per year for Series A, and 8% for Series B) for the first several years after the shares' issue date. Then, in 2024, those dividend rates would convert to floating rates equal to the value of three-month dollar-denominated LIBOR plus a fixed rate of either 5.831% or 5.99% per year (for Series A and B shares, respectively). 'Three-Month LIBOR' is defined in the Articles as LIBOR “for deposits in U.S. dollars having an index maturity of three months . . . as such rate appears on' Reuters's LIBOR publication service.

If that LIBOR rate is not available, the Articles provide for three cascading fallbacks. First, PennyMac will attempt to request U.S. dollar-denominated interbank lending rates for a three-month maturity period from national banks in the London or New York interbank lending market. Second, if there are not enough quotes available, 'the Three-Month LIBOR for the applicable period will be the same as for the Page 9immediately preceding Dividend Period.” Third, and most relevant here, 'if there was no such Dividend Period, the dividend shall be calculated at the dividend rate in effect for the immediately preceding Dividend Period.”

When LIBOR was discontinued in 2023-one year before the shares were set to transition to a floating rate— PennyMac concluded that the third fallback provision should apply, as there were no available interbank lending quotes and no LIBOR value for the 'immediately preceding Dividend Period.” Accordingly, in August 2023, PennyMac announced that the shares’ interest rates would be calculated in perpetuity based on their initial fixed rates rather than transitioning to a floating rate. This allegedly caused the value of the shares to drop from $25 to $22.77 for Series A shares and to $22.40 for Series B shares, resulting in a $40 million loss in market capitalization.

Displeased with this development, shareholder Roberto Verthelyi sued PennyMac on behalf of himself and a class of similarly situated investors under California’s Unfair Competition Law (“UCL”), which prohibits “any unlawful, unfair or fraudulent business act or practice.' Cal. Bus. & Prof. Code § 17200. Verthelyi alleged that PennyMac's failure to convert its rates to SOFR-based rates was both “unlawful” and “unfair” under the UCL because, in his view, a fixed rate cannot serve as a 'benchmark replacement' under the LIBOR Act. Verthelyi asserted that if PennyMac had instead converted the fixed rate dividends to SOFRbased floating rate dividends, Series A shareholders would have received a dividend rate of approximately 11.19% on March 24, 2024, and Series B shareholders would have received a dividend rate of approximately 11.35%substantially higher than the now-permanent 8.125% and 8% fixed rates.

Page 10PennyMac moved to dismiss the complaint under Rule 12(b)(6). PennyMac first argued that it did not violate the LIBOR Act because the Articles' third fallback provision qualifies as an adequate 'benchmark replacement.' 12 U.S.C. § 5802(3). In PennyMac’s view, a fixed dividend rate is a 'dividend rate . . . to replace LIBOR,' and therefore satisfies the statutory requirements for a “benchmark replacement.' Id. And if PennyMac's conduct complied with the LIBOR Act, Verthelyi could not state a claim under the UCL. Second, PennyMac argued, in the alternative, that because its corporate charter document contained a choiceof-law provision selecting Maryland law, the UCL should not apply.

The district court denied PennyMac’s motion to dismiss. In response to PennyMac's statutory argument, the court concluded that '[t]he parties' divergent interpretationsboth of which appear plausible from the plain meaning of the text—indicate some level [of] ambiguity within the statute.” It then concluded that Verthelyi 'has proffered sufficient evidence from the legislative history indicating that a principal concern of the Act may have been to prevent floating-rate instruments from unfairly converting into fixed-rate instruments.' Verthelyi therefore sufficiently alleged that PennyMac violated the LIBOR Act. As for the choice-of-law issue, the district court concluded that California law applied.

PennyMac sought certification to file an interlocutory appeal under 28 U.S.C. § 1292(b), which the district court granted as to both issues. A motions panel of this court granted PennyMac permission to appeal. We review de novo the district court's ruling on a Rule 12(b)(6) motion. See Milliken v. Bank of Am., N.A. , 162 F.4th 1030, 1032 (9th Cir. 2025). Because we resolve the LIBOR Act question in Page 11PennyMac’s favor, we do not reach the choice-of-law issue, and assume (as Verthelyi alleges) that California law applies.

The UCL prohibits “any unlawful, unfair or fraudulent business act or practice.” Cal. Bus. & Prof. Code § 17200. 'Unlawful,' 'unfair,' and 'fraudulent' are three separate varieties of wrongful activity under the UCL. Levitt v. Yelp! Inc. , 765 F.3d 1123, 1130 (9th Cir. 2014). We first address Verthelyi’s claim under the “unlawful” prong of the UCL, which “borrows violations of other laws and treats them as unlawful practices that the unfair competition law makes independently actionable.” Id. (quoting Cel-Tech Commc’ns, Inc. v. L.A. Cellular Tel. Co. , 973 P.2d 527, 539– 40 (Cal. 1999)). In other words, to state a claim, Verthelyi must show that PennyMac’s actions violated a law—here, the LIBOR Act. And that requires us to interpret the meaning of “benchmark replacement.” 12 U.S.C. §§ 5802(3), 5803(a)(2)(A). To our knowledge, no other court has interpreted this (or any) provision of the LIBOR Act. We therefore interpret the statute from first principles, starting, as always, with the text. See, e.g. , Amazon.com Servs., LLC v. Teamsters Amazon Nat’l Negotiating Comm. , 163 F.4th 624, 629 (9th Cir. 2025).

At the outset, the LIBOR Act sets out a multi-step framework for evaluating whether a LIBOR-based rate should be replaced with a SOFR-based rate. This process begins with § 5803(a) of the statute, which states that:

On the LIBOR replacement date, the Boardselected benchmark replacement [ i.e. ,

Page 12SOFR] 1 shall be the benchmark replacement for any LIBOR contract that, after giving any effect to [§ 5803(b)]—

(1) contains no fallback provisions; or (2) contains fallback provisions that identify neither— (A) a specific benchmark replacement; nor (B) a determining person.

12 U.S.C. § 5803(a).

In this case, PennyMac's Articles contain fallback provisions, so § 5803(a)(1) presumptively does not apply (although fallback provisions can be knocked out by § 5803(b), as we discuss below, we do not definitively resolve the question of whether § 5803(b) knocks out PennyMac’s fallbacks). In addition, because no “determining person” is at issue here, § 5803(a)(2)(B) does not apply.² See id. § 5802(10). So the only relevant provision in § 5803(a) for our purposes is § 5803(a)(2)(A), Page 13which requires us to determine whether the Articles identify a “specific benchmark replacement.”

The next subsection, § 5803(b)—which § 5803(a) references—informs our answer to that question. Section 5803(b) provides that certain types of fallback provisions are invalid because they use inadequate replacement measures. It states:

On the LIBOR replacement date, any reference in the fallback provisions of a LIBOR contract to—

1. (1) a benchmark replacement that is based in any way on any LIBOR value, except to account for the difference between LIBOR and the benchmark replacement; or
2. (2) a requirement that a person (other than a benchmark administrator) conduct a poll, survey, or inquiries for quotes or information concerning interbank lending or deposit rates;

shall be disregarded as if not included in the fallback provisions of such LIBOR contract and shall be deemed null and void and without any force or effect.

Id. § 5803(b).

PennyMac's Articles contain three different fallback provisions, two of which are plainly disqualified by § 5803(b). The first fallback, which requires PennyMac to request quotes for interbank lending rates from various national banks in the London or New York markets, is Page 14knocked out by § 5803(b)(2), as it involves 'inquiries for quotes or information concerning interbank lending or deposit rates.' The second fallback, which is linked to 'Three-Month LIBOR . . . for the immediately preceding Dividend Period,” is knocked out by § 5803(b)(1). That is because it references a presently undetermined value of LIBOR due to the discontinuation of LIBOR before the Articles transitioned to a floating rate. Id. § 5803(b)(1).

That leaves only the third fallback, which provides that if the second fallback does not apply, “the dividend shall be calculated at the dividend rate in effect for the immediately preceding Dividend Period.” As applied by PennyMac, the third fallback resolves to a fixed rate. The specific issue presented to us in this interlocutory appeal is whether a fixed rate qualifies as a 'benchmark replacement' under the LIBOR Act, and that is what we decide here. See 28 U.S.C. § 1292(b).

The LIBOR Act expressly defines the term “benchmark replacement” as:

[A] benchmark, or an interest rate or dividend rate (which may or may not be based in whole or in part on a prior setting of LIBOR), to replace LIBOR or any interest rate or dividend rate based on LIBOR, whether on a temporary, permanent, or indefinite basis, under or with respect to a LIBOR contract.

Id. § 5802(3) (emphasis added). 'Benchmark' is itself defined two subsections earlier as “ an index of interest rates or dividend rates that is used, in whole or in part, as the basis of or as a reference for calculating or determining any Page 15valuation, payment, or other measurement.” Id. § 5802(1) (emphasis added).

Putting these two definitions together allows us to resolve the question of whether PennyMac’s third fallback counts as a “benchmark replacement.” Recall that the final fallback provides that “the dividend shall be calculated at the dividend rate in effect for the immediately preceding Dividend Period.” And because LIBOR was discontinued before the floating rate could take effect, the “dividend rate in effect' before then was the initial fixed rate-namely, 8.125% and 8% for Series A and B shares, respectively.

While a fixed rate is not a “benchmark” because it is not an index of rates, the statute defines “benchmark replacement” to mean “a benchmark, or an interest rate or dividend rate.” 12 U.S.C. § 5802(3) (emphasis added). The word “or” in “ordinary use is almost always disjunctive, that is, the words it connects are to ‘be given separate meanings.’” United States v. Woods , 571 U.S. 31, 45-46 (2013) (quoting Reiter v. Sonotone Corp. , 442 U.S. 330, 339 (1979)). Here, the statute says that a “benchmark replacement” can be either “a benchmark” or “an interest rate or dividend rate.” 12 U.S.C. § 5802(3). If Congress had wanted a “benchmark replacement” to only be a “benchmark,” it would have simply used the term 'benchmark,' rather than defining an entirely new term. See Lockhart v. United States , 577 U.S. 347, 356 (2016) ('We assume that Congress used two terms because it intended each term to have a particular, nonsuperfluous meaning.' (quoting Bailey v. United States , 516 U.S. 137, 146 (1995))).

Given all of this, we now ask: is a fallback rate “calculated at the dividend rate in effect for the immediately preceding Dividend Period,” as set forth in the Articles, “aPage 16 benchmark, or an interest rate or dividend rate” under the LIBOR Act? 12 U.S.C. § 5802(3). The answer is yes, as the 'dividend rate in effect for the immediately preceding Dividend Period' is-naturally-a 'dividend rate.' Id. Therefore, Verthelyi’s argument fails under the plain text of the statute. This interpretation is squarely in line with the statute’s stated purpose: to allow contracts with valid benchmark replacements 'to operate according to their terms.” Id. § 5801(b)(3).

Verthelyi raises several points in rebuttal. Most prominently, he argues that the phrase “an interest rate or dividend rate” in the definition of “benchmark replacement” does not stand independently. 12 U.S.C. § 5802(3). In his view, 'an interest rate or dividend rate' is an appositive phrase 3 that explains the term 'benchmark,' which is defined as a floating rate. Id. § 5802(1). As Verthelyi argues in his briefing, § 5802(3) 'does not employ grammar associated with a series of elements,' as it 'includes a comma after the last element, uses the conjunction 'or' twice, and lacks a serial (or Oxford) comma after ‘interest rate.’” By this logic, if Congress had intended to provide a choice between 'a benchmark,' 'an interest rate,' or a 'dividend rate,' it would have written the phrase as 'a benchmark, an interest rate, or dividend rate” instead of “a benchmark, or an interest rate or dividend rate.”

Page 17We find this grammatical argument unpersuasive. Most importantly, in the statute's definitions section, Congress already defined “benchmark” as “an index of interest rates or dividend rates that is used, in whole or in part, as the basis of or as a reference for calculating or determining any valuation, payment, or other measurement.' 12 U.S.C. § 5802(1). It would not make sense for Congress to then redefine the term in a more cursory way as “an interest rate or dividend rate” two subsections later, in a way that would do “no independent work.” Pulsifer v. United States , 601 U.S. 124, 142 (2024). Verthelyi “offer[s] no account of what function' the clause 'or an interest rate or dividend rate' 'would serve on [his] proposed interpretation.' Advoc. Health Care Network v. Stapleton , 581 U.S. 468, 477 (2017).

Indeed, Verthelyi’s purported additional definition is not just redundant, but runs into the problem that an “index of interest rates or dividend rates” is not the same as “an interest rate or dividend rate.' In the financial context, an index refers to “a system of numbers used for comparing values of things that change according to each other or a fixed standard.” Index , Cambridge Dictionary, https://dictionary.cambridge.org/us/dictionary/english/index [https://perma.cc/68K4-UL49]; see also Milliken , 162 F.4th at 1031-32 (examining a statute governing 'changes in [interest rates] according to operation of an index'). For instance, the U.S. Prime Rate is an index calculated based on “the base rate on corporate loans posted by at least 70% of the 10 largest U.S. banks.' Milliken , 162 F.4th at 1031.

In contrast, the terms “interest rate” and “dividend rate” merely describe a “percentage amount” paid or received over a period of time, whether fixed or floating. See Interest Rate , Cambridge Dictionary, https://dictionary.cambridge.org/ dictionary/english/interest-rate [https://perma.cc/454Q- Page 18RX5V]. Indexes (and the rates they produce) necessarily change when their input values change. But “interest rates” and “dividend rates” are not so limited—both terms could refer to either a fixed or a floating rate. So it is not correct to say, under the LIBOR Act, that the terms 'benchmark' and 'an interest rate or dividend rate' are merely two different ways of saying the same thing.

Thus, the more natural reading of § 5802(3) is that “a benchmark' is meant to refer to an index of rates under § 5802(1), and the phrase “an interest rate or dividend rate” is meant to refer to a single interest or dividend rate, whether fixed or floating ( e.g. , a fixed 8% interest rate or a floating dividend rate calculated at 3% plus the value of the U.S. Prime Rate). This distinction is underscored by the use of the word “an,” as well as the use of the singular “rate,” rather than the plural “rates.” 12 U.S.C. § 5802(3). It is further underscored by the fact that commas are used to set off the entire phrase, not just 'interest rate' or 'dividend rate' individually. And the LIBOR Act consistently uses the phrase 'interest rate or dividend rate' in the definitions section, which indicates that Congress understood these two terms as a single set, as distinct from an index of rates. See 12 U.S.C. §§ 5802(1), (3), (13).

The statute on its face also supplies reasons for why Congress may have written § 5802(3) in the way that it did. For one, it makes sense to put “interest rate” and “dividend rate' together against the term 'benchmark,' as both are rates, rather than indexes of rates. Id. § 5802(1). Additionally, recall that the definition of benchmark replacement includes a parenthetical: “a benchmark, or an interest rate or dividend rate ( which may or may not be based in whole or in part on a prior setting of LIBOR ), to replace LIBOR.” Id. § 5802(3) (emphasis added). Although that Page 19parenthetical is not at issue here, the sentence structure shows that by separating with a comma “a benchmark” from “an interest rate or dividend rate,” Congress intended for the parenthetical to modify the phrase 'an interest rate or dividend rate,” but not “a benchmark.” See, e.g. , Barnhart v. Thomas , 540 U.S. 20, 26 (2003) (noting that a qualifying phrase 'should ordinarily be read as modifying only the noun or phrase that it immediately follows”). This makes sense because a benchmark ( i.e. , an index of rates) cannot be calculated based on a single 'setting.' In contrast, an individual 'interest rate or dividend rate' can be set “based in whole or in part on a prior setting of LIBOR,' i.e. , a previously determined historical value of LIBOR. Id. § 5802(3).

Verthelyi further argues that PennyMac’s fallbacks were only meant to operate when LIBOR was temporarily unavailable, and the LIBOR Act did not authorize conversions of temporary fallbacks to permanent ones. But that is contrary to the statutory text, which provides that a benchmark replacement is one designed to “replace LIBOR . . . whether on a temporary, permanent, or indefinite basis .” Id. § 5802(3) (emphasis added). Even if Verthelyi is correct that the fallbacks were originally intended to operate temporarily, that fact is irrelevant in the face of the statute’s plain text. The LIBOR Act does not inquire into the original purpose of a fallback provision. Nor is it concerned with whether the parties specifically anticipated the cessation of LIBOR when drafting their agreement. Instead, the LIBOR Act is concerned with function: whether the fallback provision identifies a “specific benchmark replacement,” id. § 5803(a), and whether it can operate self-sufficiently in a post-LIBOR world. Id. § 5803(b).

Page 20Finally, Verthelyi turns to legislative history. Any effort to use legislative history to override the statute’s plain text of course cannot succeed, because Congress’s “authoritative statement is the statutory text, not the legislative history.' Exxon Mobil Corp. v. Allapattah Servs., Inc. , 545 U.S. 546, 568 (2005). Regardless, the legislative history does not change matters.

Verthelyi first points to a floor statement by the sponsor of the LIBOR Act, Congressman Brad Sherman, stating that 'trillions of outstanding loans . . . have adjustable interest rates' that are 'tied to . . . LIBOR,' and Congress was “dealing . . . with adjustable rate mortgages, business loans and securities.” 167 Cong. Rec. H7482 (daily ed. Dec. 8, 2021) (statement of Congressman Bradley J. Sherman). Second, Verthelyi points to testimony by Mark Van Der Weide, the general counsel of the Federal Reserve Board, who testified that “many floating-rate notes and securitizations have problematic fallback language' that “convert [them] to fixed-rate instruments at the last published value of LIBOR.” The End of LIBOR: Transitioning to an Alternative Interest Rate Calculation for Mortgages, Student Loans, Business Borrowing, and Other Financial Products: Hearing Before the Subcomm. on Inv. Prot., Entrepreneurship, and Cap. Mkts. of the H. Comm. on Fin. Servs. , 117th Cong. 54 (2021) (statement of Mark Van Der Weide, General Counsel, Bd. of Governors of the Fed. Reserve Sys.). Van Der Weide further testified that “[f]ederal legislation should be targeted narrowly to address legacy contracts that have no fallback language, that have fallback language referring to LIBOR or to a poll of banks, or that convert to fixed-rate instruments.” Id. at 55.

Verthelyi puts more weight on this legislative history than it can bear. For one, 'scattered floor statements by Page 21individual lawmakers' are 'among the least illuminating forms of legislative history.' Stapleton , 581 U.S. at 481 (quoting NLRB v. SW Gen., Inc. , 580 U.S. 288, 307 (2017)); see also United States v. Tan , 16 F.4th 1346, 1352 (9th Cir. 2021) (same). And witness testimony before a committee is even less persuasive. See Kelly v. Robinson , 479 U.S. 36, 51 n.13 (1986) ('[N]one of those statements was made by a Member of Congress, nor were they included in the official Senate and House Reports. We decline to accord any significance to these statements.”).

But even if we give these statements some weight, they do not prove Verthelyi's theory. Congressman Sherman's statement merely noted that LIBOR was used as the index for many adjustable-rate financial instruments. Congressman Sherman did not say what should happen under the LIBOR Act when a floating-rate instrument is converted to a fixed-rate instrument under an adequate fallback provision. Van Der Weide's testimony fares no better, as he was merely expressing a concern over conversions to “fixed-rate instruments at the last published value of LIBOR.” Van Der Weide was not testifying about the meaning of any statutory language, and regardless, the plain text of the LIBOR Act must control over whatever policy preferences any one witness may have had. See Am. Rivers v. FERC , 201 F.3d 1186, 1204 (9th Cir. 1999) (emphasizing that “legislative history-no matter how clear—can’t override statutory text”).

In sum, the third fallback provision in the PennyMac agreements identifies a “benchmark replacement” under the LIBOR Act. 12 U.S.C. §§ 5802(3), 5803(a)(2). There is no requirement that such a 'benchmark replacement' be a floating rate. Although Congress replaced LIBOR with SOFR when the contracting parties failed to provide Page 22adequate contingencies if LIBOR became unavailable, the LIBOR Act does not purport to override private contractual agreements that contain a valid “benchmark replacement.” See id. § 5801(b)(3) (providing that the LIBOR Act “allow[s] existing contracts that reference LIBOR but provide for the use of a clearly defined and practicable replacement rate, to operate according to their terms”); id. § 5803(f) (providing that “[n]othing in this chapter may be construed to alter or impair . . . except as provided in subsection (b), any LIBOR contract that contains fallback provisions that identify a benchmark replacement that is not based in any way on any LIBOR value”). The LIBOR Act closes gaps in contracts; it does not rewrite otherwise valid contracts. Because a fixed rate can be a benchmark replacement under § 5802(3)'s definition of 'benchmark replacement,” PennyMac’s application of the third fallback provision is not, for that reason, “unlawful” under the UCL. See Levitt , 765 F.3d at 1130.

Finally, Verthelyi argues that even if he cannot show an underlying violation of the LIBOR Act based on his 'nofixed rate” theory, he can still obtain relief under the “unfair” prong of the UCL. When deciding whether a practice is “unfair” under the UCL, California courts consider, among other things, 'whether the practice is 'immoral, unethical, oppressive, unscrupulous or substantially injurious to consumers,’” or “whether the practice’s impact on the victim outweighs 'the reasons, justifications and motives of the alleged wrongdoer.’” Doe v. CVS Pharmacy, Inc. , 982 F.3d 1204, 1214-15 (9th Cir. 2020) (quoting Morgan v. AT&T Wireless Servs., Inc. , 99 Cal. Rptr. 3d 768, 784 (Cal. Ct. App. 2009)). Verthelyi’s argument fails for two reasons.

Page 23For one, the LIBOR Act expressly preempts 'any provision of any State or local law, statute, rule, regulation, or standard . . . relating to the selection or use of a benchmark replacement or related conforming changes.” 12 U.S.C. § 5806(1). This preemption provision reflects Congress's stated intent 'to establish a clear and uniform process, on a nationwide basis, for replacing LIBOR in existing contracts' that 'do not provide for the use of a clearly defined or practicable replacement benchmark rate.” Id. § 5801(b)(1). Here, the core of Verthelyi’s “unfairness” claim under the UCL is that even if the dividend rates for the shares were set in compliance with § 5802(3) of the LIBOR Act, PennyMac should have used SOFR instead of the predetermined contractual fallback. But that is a claim “relating to the selection or use of a benchmark replacement,” and so is preempted. Id. § 5806(1).

Relatedly, California courts have created a safe harbor for lawful conduct under the UCL. Under this doctrine, if the government “has permitted certain conduct or considered a situation and concluded no action should lie, courts may not override that determination.” Alvarez v. Chevron Corp. , 656 F.3d 925, 933 (9th Cir. 2011) (emphasis omitted) (quoting Cel-Tech , 973 P.2d at 541); see also Webb v. Smart Document Sols., LLC , 499 F.3d 1078, 1082 (9th Cir. 2007). Here, the LIBOR Act squarely “allow[s] existing contracts that reference LIBOR but provide for the use of a clearly defined and practicable replacement rate[ ] to operate according to their terms.” 12 U.S.C. § 5801(b)(3). The safe harbor doctrine applies because Congress did not disallow a fixed rate from serving as a “benchmark replacement” under § 5802(3). Indeed, as the district court correctly anticipated, “[s]hould the Ninth Circuit agree with Defendants’ interpretation of the Act,” “there is a high likelihood that aPage 24 resolution in Defendants' favor would result in Plaintiff's claim failing under both federal preemption and the UCL’s safe-harbor doctrine.”

* * *

For the reasons we have explained, PennyMac's third fallback provision is not an inadequate benchmark replacement under the LIBOR Act simply because it resolves to a fixed rate. The district court erred in denying PennyMac's motion to dismiss on that ground. But it is unclear from the record below if the parties dispute whether PennyMac's application of the third fallback provision violated the LIBOR Act in some other respect. Because this case came to us in an interlocutory posture, we remand to the district court so that the parties may advance any further arguments they might have regarding whether PennyMac’s application of the third fallback provision violates the LIBOR Act.⁴