National Bank Act Preempts State Minimum Interest Mandates on Mortgage-Escrow Accounts Under Barnett Bank’s “Nature and Degree” Test

1. Introduction

These consolidated interlocutory appeals arise from two putative class actions brought by mortgage borrowers (Alex Cantero; Saul R. Hymes and Ilana Harwayne-Gidansky) against Bank of America, N.A. (“BOA”), a federally chartered national bank. The plaintiffs alleged BOA unlawfully refused to pay at least 2% interest on their mortgage-escrow balances as required by New York’s GOL § 5-601, and pleaded contract and related theories.

BOA moved to dismiss on the ground of federal preemption: federal banking law authorizes national banks to make real-estate loans and, as an incident of that power, to establish mortgage-escrow accounts without state-imposed interest mandates. The district court denied dismissal (and certified the preemption question). The Second Circuit previously held the statute preempted, but the Supreme Court vacated and remanded, directing a “nuanced comparative analysis” under the Barnett Bank standard rather than a categorical “control” test. On remand, the Second Circuit again holds GOL § 5-601 preempted.

Key issue: Whether a state “interest-on-escrow” requirement “prevents or significantly interferes” with national bank powers under 12 U.S.C. § 25b(b)(1)(B), applying the preemption standard “in accordance with” Barnett Bank of Marion County v. Nelson.

2. Summary of the Opinion

The majority (Judge Park, joined by Chief Judge Livingston) reverses the district courts’ orders denying BOA’s motions to dismiss and remands. Applying the Supreme Court’s remand instructions from Cantero v. Bank of Am., N.A., 602 U.S. 205 (2024), the court conducts a “nuanced comparative analysis” focusing on:

• Whether the state law affects a national bank power: GOL § 5-601 affects the exercise of national banks’ mortgage-lending power (12 U.S.C. § 371(a)) and incidental powers (12 U.S.C. § 24 (Seventh)) by dictating a minimum interest term of escrow accounts.
• Nature of interference: The law targets banks and limits banks’ power to set escrow terms; the court finds it comparable to laws preempted in Fidelity Federal Savings and Loan Ass'n v. de la Cuesta and Barnett Bank of Marion County v. Nelson, and unlike generally applicable laws upheld in McClellan v. Chipman and common-law-consistent regimes upheld in Anderson National Bank v. Luckett.
• Degree of interference: A mandated interest floor increases costs and burdens efficient mortgage lending; the court finds the interference at least as severe as the advertising restriction preempted in Franklin National Bank of Franklin Square v. New York, and potentially more severe because pricing mandates intrude more directly than advertising limits.

Judge Pérez dissents, arguing the majority effectively revives an overbroad “flexibility/control” approach, insufficiently grounds “significant interference” in practical consequences, and misreads the Supreme Court’s preemption precedents.

3. Analysis

3.1 Precedents Cited (and How They Shape the Decision)

A. The governing standard: Barnett Bank of Marion County v. Nelson

The opinion treats Barnett Bank of Marion County v. Nelson, 517 U.S. 25 (1996), as the lodestar because Dodd-Frank codifies its test: state consumer financial laws are preempted only if, “in accordance with” Barnett Bank, they “prevent[] or significantly interfere[]” with national bank powers (12 U.S.C. § 25b(b)(1)(B)). The Supreme Court’s remand in Cantero v. Bank of Am., N.A., 602 U.S. 205 (2024), required the Second Circuit to compare the “nature and degree” of interference to the Supreme Court’s banking-preemption cases rather than relying on an abstract “control” formulation.

B. “Not preempted” comparators: National Bank v. Commonwealth; McClellan v. Chipman; Anderson National Bank v. Luckett

• National Bank v. Commonwealth, 76 U.S. 353 (1869): The court uses this case to illustrate that a law can “target” banks (Kentucky’s tax on bank stock) yet still not be preempted if it does not meaningfully hinder the bank’s functions. It supports the majority’s threshold distinction between laws that merely affect banks and those that affect the exercise of banking powers.
• McClellan v. Chipman, 164 U.S. 347 (1896): Cited for the principle that generally applicable state laws (fraudulent transfer restrictions) ordinarily apply to national banks as part of “their daily course of business” and are unlikely to be preempted even if they incidentally burden banking activities. The majority distinguishes GOL § 5-601 as not generally applicable: it is a bank-targeted pricing mandate.
• Anderson National Bank v. Luckett, 321 U.S. 233 (1944): Used to explain that even bank-targeting laws may survive if they are consistent with generally applicable common-law schemes (Kentucky’s rebuttable-presumption escheat regime). The majority contrasts this with GOL § 5-601, which is not framed as a general background rule of property/contract but as a specific consumer financial requirement imposed on mortgage-lenders’ escrow pricing.

C. “Preempted” comparators: First National Bank of San Jose v. California; Franklin National Bank of Franklin Square v. New York; Fidelity Federal Savings and Loan Ass'n v. de la Cuesta; Barnett Bank of Marion County v. Nelson

• First National Bank of San Jose v. California, 262 U.S. 366 (1923): Presented as a preempted example where state law was unusually harsh (a non-rebuttable presumption leading to escheat for mere dormancy) and therefore significantly interfered with banking by deterring depositors. The majority finds consumer-attraction/deterrence effects inconclusive for escrow-interest mandates and does not rely heavily on this prong.
• Franklin National Bank of Franklin Square v. New York, 347 U.S. 373 (1954): The pivotal “degree-of-interference” anchor. New York’s restriction on using “savings” in ads was preempted because it impaired national banks’ ability to exercise their power efficiently—advertising was a practical necessity. The majority analogizes: escrow accounts are a crucial risk-mitigation feature of mortgage lending, and dictating a minimum interest rate impedes the efficient, competitive structuring and pricing of that feature.
  The majority’s key move is treating GOL § 5-601 as a constraint on an “essential” incident of mortgage lending, much like the ad restriction constrained an essential incident of receiving savings deposits.
• Fidelity Federal Savings and Loan Ass'n v. de la Cuesta, 458 U.S. 141 (1982): Emphasized for the idea that state law may be preempted even where dual compliance is not “physically impossible,” because state restrictions can still undermine broad federal permissions—particularly where federal law grants flexibility/optionality (there, due-on-sale clauses “at its option”) and the state curtails that flexibility.
• Barnett Bank of Marion County v. Nelson, 517 U.S. 25 (1996): Used both as standard and comparator: a Florida law preventing banks from selling insurance was preempted because it significantly interfered with a broad, unqualified federal permission. The Second Circuit reads federal statutory structure (RESPA’s extensive escrow regulation without an interest mandate; TILA’s selective incorporation of state interest rules for a different subset of loans) as signaling broad federal permission for banks to decide whether and how much interest to pay on most escrow accounts.

D. Additional authorities shaping the majority’s analysis

• McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316 (1819): Appears mainly as historical backdrop (the earlier Second Circuit decision had leaned on “control” and the “power to destroy” logic). On remand, the majority deemphasizes this and instead operationalizes the Supreme Court’s “nature and degree” comparison.
• Cuomo v. Clearing House Ass'n, LLC, 557 U.S. 519 (2009): Cited to illustrate that generally applicable regimes (e.g., fair-lending/false-advertising enforcement) typically survive; the majority uses it to reinforce that targeted, bank-specific pricing mandates stand on a different footing.
• Conti v. Citizens Bank, N.A., 157 F.4th 10 (1st Cir. 2025): The Second Circuit explicitly rejects the First Circuit’s contrary holding (Rhode Island’s similar escrow-interest law not preempted), criticizing (i) insufficient attention to RESPA and TILA and (ii) understatement of how pricing mandates can materially affect operations.
• Kivett v. Flagstar Bank, FSB, 154 F.4th 640 (9th Cir. 2025): Cited for the proposition that escrow-interest mandates raise costs and can affect bank behavior. The majority also cites a dissenting view in Kivett to underscore that pricing mandates may interfere even more than advertising restrictions.
• In re Cap. One 360 Sav. Acct. Int. Rate Litig., 779 F. Supp. 3d 666 (E.D. Va. 2024) and Ill. Bankers Ass'n v. Raoul, 760 F. Supp. 3d 636 (N.D. Ill. 2024): Used as persuasive support for the intuition that state-imposed pricing terms can be particularly intrusive under the national bank preemption framework.

3.2 Legal Reasoning

A. The court’s post-remand framework

Following the Supreme Court’s remand instruction, the Second Circuit formalizes a two-step inquiry:

1. Does the state law affect the exercise of a national bank power? If not, no preemption under § 25b(b)(1)(B).
2. If yes, does it “prevent or significantly interfere”? This requires a “practical assessment” of (i) the nature of the interference and (ii) the degree of the interference, using textual/structural analysis, comparison to precedents, and “common sense.”

B. Banking power identified: mortgage lending and escrow-term setting as an incident

The majority characterizes mortgage-escrow accounts as a “clear logical outgrowth” of the express power to make real-estate loans (12 U.S.C. § 371(a)) and the incidental powers clause (12 U.S.C. § 24 (Seventh)). Because escrow accounts are framed as a “crucial risk mitigation tool,” dictating a minimum interest rate is treated not as peripheral consumer protection but as an operational constraint on a core feature of national-bank mortgage lending.

C. “Nature” of interference: targeted pricing mandate vs generally applicable background law

The “nature” analysis turns on two distinctions drawn from the precedent set:

• General applicability cuts against preemption: Like the fraudulent transfer rule in McClellan v. Chipman, laws of general application that incidentally affect banks are less likely to be preempted.
• Bank-targeted restrictions on broad federal permission cut toward preemption: Like the prohibitions at issue in Fidelity Federal Savings and Loan Ass'n v. de la Cuesta and Barnett Bank of Marion County v. Nelson, a law that targets banks and narrows their discretion over the terms of a federally authorized product is more likely to “significantly interfere.”

GOL § 5-601 is placed in the latter category: it expressly applies to mortgage institutions and sets a minimum interest floor for escrow balances, and therefore is not treated as a neutral background rule of property or contract.

D. Federal statutory structure (RESPA and TILA) as evidence of broad federal discretion

A central (and contested) move is the majority’s inference from federal statutory “architecture”:

• RESPA regulates escrow accounts extensively but does not mandate interest—suggesting Congress left interest as a bank-by-bank business decision for most escrow accounts.
• TILA requires interest on escrow for certain mortgages “in the manner as prescribed by an applicable State or Federal law,” but that express incorporation is limited to a subset of loans, implying (in the majority’s view) that similar state mandates outside that subset are preempted.

This is where the dissent parts ways: it argues congressional silence should not be treated as an affirmative “broad permission” that displaces state consumer financial regulation, and it warns that reframing federal powers as “flexibility” risks recreating the disapproved “control” test.

E. “Degree” of interference: efficiency and cost, evaluated without an evidentiary record

The majority takes seriously the Supreme Court’s direction that courts may use “common sense” and do not need record evidence of real-world effects. It then reasons:

• Escrow administration entails costs; a mandatory 2% interest floor “raises the cost to national banks to use escrow accounts” and can force recoupment through other pricing or reduced offerings.
• This practical impact is “similarly severe” to the impairment found preemptive in Franklin National Bank of Franklin Square v. New York, because it constrains an incidental power necessary to efficient exercise of an express power.
• Pricing mandates are arguably more intrusive than advertising restrictions (an argument supported by the Kivett v. Flagstar Bank, FSB dissent and district-court analogies to deposit-interest constraints).

The majority declines to rest on consumer-attraction/deterrence effects (the First National Bank of San Jose/Anderson National Bank v. Luckett axis), finding the net effect uncertain: consumers might like interest, but banks might offset with fees or reduced mortgage availability.

3.3 Impact

A. Doctrinal impact within the Second Circuit

• Concrete rule for escrow-interest mandates: State laws that target banks and impose minimum interest terms on mortgage-escrow accounts are likely preempted when they function as pricing mandates that substantially affect efficient mortgage lending—especially where federal law is read to confer broad discretion over escrow compensation.
• Methodological clarification: The opinion operationalizes Cantero v. Bank of Am., N.A., 602 U.S. 205 (2024), by making the “nature”/“degree” rubric explicit and by rejecting any requirement of empirical record proof for interference.

B. Practical impact for lenders, borrowers, and state regulators

• National banks in New York: At least as to the claims at issue, they are not required to pay the 2% GOL § 5-601 interest on mortgage-escrow balances.
• State consumer protection design: The opinion signals that bank-targeted pricing floors are especially vulnerable. States seeking to influence escrow outcomes may face tighter constraints unless they operate within federally incorporated spaces (e.g., the subset of TILA-covered escrow accounts) or through more generally applicable legal regimes.
• Litigation landscape: The decision deepens an inter-circuit tension with Conti v. Citizens Bank, N.A., setting up a stronger possibility of future Supreme Court review to harmonize treatment of state interest-on-escrow laws.

C. Administrative developments (OCC proposals)

While not the formal basis of decision, the majority’s reasoning aligns with the OCC’s December 2025 proposals: Real Estate Lending Escrow Accounts, 90 Fed. Reg. 61099, and Preemption Determination: State Interest-on-Escrow Laws, 90 Fed. Reg. 61093. If finalized and later found persuasive, such agency action could further influence how courts frame the “nature” of interference for escrow-term regulation—though the dissent persuasively emphasizes that conclusory assertions of “flexibility” should not substitute for Barnett’s “significant interference” analysis.

4. Complex Concepts Simplified

• Preemption: A state law is unenforceable if it conflicts with federal law. Here, the question is not “any conflict,” but whether the state rule “prevents or significantly interferes” with national bank powers.
• National bank “powers” and “incidental powers”: Congress expressly authorizes some activities (like making real-estate loans). “Incidental powers” are additional actions banks may take that are necessary or useful to carry out the banking business (like structuring escrow arrangements to protect collateral).
• Mortgage-escrow account: Money the borrower pays monthly that the lender/servicer holds to pay property taxes and insurance. It reduces default risk and protects the collateral (the home).
• Barnett standard (“nature and degree”): Courts compare the kind of interference (e.g., targeted prohibition vs generally applicable rule) and how severe it is (e.g., efficiency/cost impacts) against the Supreme Court’s bank-preemption cases, rather than applying a simple label like “controls banking.”
• Rebuttable vs non-rebuttable presumptions (escheat cases): A rebuttable presumption can be overcome with proof; a non-rebuttable presumption cannot. In the banking cases, harsher non-rebuttable rules were treated as more likely to deter customers and thus more likely preempted.

5. Conclusion

On remand from the Supreme Court, the Second Circuit reaffirms that New York’s GOL § 5-601 is preempted as applied to national banks’ mortgage-escrow accounts. The court’s central contribution is methodological: it concretely applies the “nuanced comparative analysis” demanded by Cantero v. Bank of Am., N.A., 602 U.S. 205 (2024), by assessing both the nature (a bank-targeted pricing mandate narrowing broad federal discretion) and the degree (a substantial efficiency- and cost-related interference comparable to Franklin National Bank of Franklin Square v. New York) of the state law’s interference.

The dissent underscores the fault line that will likely animate future litigation: whether federal silence (and selective incorporation of state law in statutes like TILA) should be read to create a broad federal “permission” that preempts state consumer financial protections, and whether “common sense” suffices to prove significant interference without a factual record. With an explicit split from Conti v. Citizens Bank, N.A., the decision positions state interest-on-escrow mandates as a recurring and increasingly ripe question in National Bank Act preemption.