TL;DR

Background: U.S. patients pay more than patients in other wealthy countries for the same branded prescription drugs. Supporters of most-favored-nation (MFN) pricing treat that gap as proof Americans overpay, and propose tying U.S. reimbursement to foreign prices. The idea has moved from a blocked 2020 Medicare rule into today’s trade debate. In April 2025, the U.S. Commerce Department opened a Section 232 investigation into pharmaceutical imports, with the Bureau of Industry and Security later advancing an onshoring framework that would condition tariff relief on company-specific MFN-pricing agreements.

But… The price gap does not prove overpayment in any meaningful economic sense. It reflects how other countries suppress branded-drug prices through monopsony purchasing by single-payer systems, health-technology assessments that set reimbursement below patients’ willingness to pay, and external reference pricing that turns one country’s suppressed price into another’s ceiling. Foreign governments use these tools to underpay for innovative medicines while shifting R&D costs onto U.S. patients. Pegging American reimbursement to those prices would import the distortion U.S. policy should confront.

Moreover… Pharmaceutical innovation depends on margins earned during a limited patent window. Compressing those margins lowers the expected returns that determine whether tomorrow’s medicines get developed, especially high-risk therapies for patients with few alternatives.

The better response would treat foreign price suppression as an anticompetitive trade distortion for targeted enforcement, not copy it into the U.S. market through the back door of a national-security tariff.

KEY TAKEAWAYS

Sticker Shock, Distorted Mirror

Per-drug comparisons make the United States look like a serial overpayer. U.S. prescription-drug prices average roughly 2.56 times those in comparable OECD countries, and about 3.44 times as much for branded drugs. But that narrow, highly visible slice of the market is not the whole thing. 

Generics account for roughly 90% of U.S. prescriptions, and U.S. generic prices are the lowest among peer countries. Weighted by real-world prescribing volume, rather than the highest-priced branded drugs, net Medicare and Medicaid prescription costs run about 18% below those in Germany, France, the United Kingdom, Canada, and Japan. 

The aggregate mismatch is stark. In 2022, the United States accounted for roughly half of worldwide prescription-drug revenue, but only about 13% of volume across countries. Among OECD countries, it generated about 60% of revenue on 24% of volume. American consumers generate more than 70% of OECD pharmaceutical profits, even though the United States accounts for about 34% of OECD gross domestic product.

The Cure Gets Pricier to Invent

Drug development has high fixed costs, but low marginal costs. Bringing a new molecule to market costs billions, and more than 90% of candidates fail. Once a drug wins approval, each additional dose costs relatively little to produce. 

So the patent-protected window matters. Prices during that period finance the entire research portfolio, including the many candidates that never reach patients. A price control does not merely shift surplus from producers to consumers. It weakens the signal that tells firms whether high-risk, long-horizon research is worth undertaking. 

Studies using demographic shifts and regional variation find that innovation responds strongly to expected revenue, with U.S. innovation tracking expected revenue at an elasticity of roughly 0.43. Given a projected $0.5 trillion to $1 trillion revenue reduction under negotiation-style proposals, those elasticities imply meaningfully fewer new drugs. 

Medicare price cuts for durable medical equipment offer the cleanest natural experiment because they share drugs’ relevant cost structure. Where reimbursement fell by an average of 61%, more-exposed manufacturers cut R&D spending by 53%. U.S. patents fell 75%, device submissions fell 25%, and revenue declined 44%. New entry dropped by 49%, driven by a 90% collapse in entry by U.S. manufacturers. Outsourcing to foreign producers rose by 28%, and adverse-event rates climbed as production moved offshore. 

Because price caps hit the most successful products, they truncate the right tail of returns that finances the rest of the portfolio. Frontier therapies get squeezed first. The same mechanism cuts against Section 232’s premise: Compressing margins can shrink innovation, push manufacturing abroad, and weaken the supply-chain resilience the investigation is supposed to protect.

The Ratchet Wrench

Even for those who think U.S. branded-drug prices should fall, an MFN policy is a lousy tool. MFN pricing and external reference pricing spread the lowest administratively set price across markets. If the price a manufacturer accepts in a small country becomes the ceiling in larger ones, the manufacturer may delay launches, skip markets, or hide the real price in confidential rebates. Those responses already cluster in heavily referenced countries. A U.S. MFN policy would magnify them, with foreign patients often paying the price. 

In its 2020 Medicare rule, later enjoined and withdrawn, CMS acknowledged that some patients could lose access to existing providers and face longer travel, lower-efficacy alternatives, or delayed and forgone treatment. Evidence from the Inflation Reduction Act’s “maximum fair price” program, which relies on similar administered-price mechanics, shows post-enactment declines in small-molecule oncology trials, consistent with the law’s price-setting timeline.

Don’t Import the Disease

The Section 232 inquiry identifies a real problem: Foreign pricing institutions suppress U.S. returns, shift R&D costs onto U.S. purchasers, and discourage domestic production. One answer would be to treat qualifying foreign systems as anticompetitive market distortions: government interventions that weaken competition and distort price formation without an overriding public-policy justification, while giving favored interests an artificial advantage. 

The right remedy is targeted, distortion-calibrated tariffication, not broad sectoral tariffs or domestic price imitation. A tariff should target the specific institutions producing the harm and should be bounded by the measured distortion rather than set for blunt deterrence. The Office of the U.S. Trade Representative could pursue targeted enforcement and negotiations aimed at transparency, nondiscrimination, and a more proportionate foreign contribution to global pharmaceutical research and development.

The United States should confront foreign price suppression, not import it. A policy that copies foreign price controls may look like hard bargaining. In practice, it would weaken the innovation base, disadvantage future patients, and mistake the symptom for the disease.

For more on this topic, see the ICLE issue brief “Don’t Import the Distortion: Why MFN Drug Pricing Would Weaken U.S. Innovation” by Kristian Stout.

TL;DR

Background: For more than a century, the United States has tried to keep banking and commerce separate, barring commercial firms from owning banks and banks from owning commercial enterprises. The idea reflected a distinctly American distrust of concentrated financial power, concern that banks could dominate local economies, and fear that commercial activities could threaten financial stability and the deposit-insurance fund. Laws from the National Bank Act through Glass-Steagall to the Bank Holding Company Act of 1956 turned those concerns into a structural rule for an era of geographically isolated community banks. 

But… That world no longer exists. Interstate banking, embedded finance, platform economics, banking-as-a-service (BaaS), and emerging tools like agentic AI have blurred the line between banking and commerce. Meanwhile, national and global competition has increased pressure for a more efficient financial system.

The separation has always been imperfect. Today, it is increasingly one-sided: technology firms offer bank-like services, while chartered banks remain bound by rules written for a different era. 

However… Modern regulation no longer depends on rigid structural barriers. Consolidated supervision, risk-based capital, stress testing, affiliate-transaction limits, and near real-time reporting can address the conduct that actually threatens stability and competition. Other advanced economies permit much greater integration without sacrificing financial stability. Rather than trying to preserve an obsolete divide, policymakers should regulate risk directly and judge firms by what they do, not by whether they fit outdated legal categories. 

KEY TAKEAWAYS

A Wall Made of Old Fears

The separation of banking and commerce did not arise from economic science. It reflected a political culture deeply suspicious of concentrated financial power.

From the fight over the Banks of the United States through Jacksonian populism and the Progressive Era campaign against the “Money Trust,” policymakers repeatedly favored dispersed credit over concentrated banking. They accepted the resulting inefficiencies as the price of limiting private financial influence.

The fear was practical, not theoretical. Todd Zywicki compares it to the company store, where employers paid workers in scrip redeemable only at employer-owned shops, giving them control over wages, credit, and retail. States responded with “anti-truck” laws requiring payment in lawful currency. Banking policy followed the same logic: lawmakers feared that institutions controlling credit could also control commerce.

Financial crises reinforced that instinct. The National Bank Act of 1864 responded to the instability of the “Free Banking Era,” when state-chartered banks issued their own notes and “wildcat banks” often issued currency that proved worthless. Rather than centralize banking, Congress preserved what Jamie Grischkan calls a “geographically segmented and peculiarly fragmented financial structure” that limited competition in service of democratic ideals.

The New Deal cemented the modern separation regime. The Senate Banking Committee’s “Pecora hearings” exposed conflicts of interest at depositor-backed institutions. The Banking Act of 1933 (better known as Glass-Steagall) separated commercial and investment banking and created federal deposit insurance. The Bank Holding Company Act of 1956 extended the separation to bank holding companies, creating one of the world’s most complex banking regulatory systems.

The Wall Has Holes

Structural separation began eroding decades ago. Interstate-banking liberalization culminated in the Riegle-Neal Act, which eliminated geographic limits. The Gramm-Leach-Bliley Act dismantled much of Glass-Steagall’s wall between commercial and investment banking.

Today, banking and commerce are deeply intertwined. Retailers extend credit. Technology platforms route payments and hold balances. BaaS lets fintechs own the customer relationship while supervised banks hold deposits and bear the regulatory burden. The Bank Holding Company Act’s categories of “bank” and “commercial” no longer fit economic reality. The wall has been bypassed, and the institutions most constrained by it are often the ones regulators already supervise most closely.

The original case for separation reflected the limits of an earlier supervisory state. When regulators could not effectively oversee complex institutions, structural prohibitions substituted for direct oversight. That approach also shielded incumbents, limited competition, and left many Americans outside the financial system.

Modern supervision makes that tradeoff less necessary. Consolidated supervision, risk-based capital, stress testing, liquidity requirements, affiliate-transaction limits, advanced analytics, and near real-time reporting allow regulators to target specific risks instead of banning integration outright.

The Wall Is Optional

The United States is unusual. Most advanced economies allow universal banking and closer bank-commerce ties without the systemic crises that separation’s defenders predict. That does not make integration risk-free, but it does show that structural separation is not the only path to financial stability.

The real risks—credit misallocation, self-dealing, and improper access to the federal safety net—are conduct problems. They should be addressed directly through tools such as Sections 23A and 23B of the Federal Reserve Act and Regulation W, rather than by banning particular ownership structures.

A modern framework should regulate risk, not legal labels. Congress could begin by amending Section 4(k) of the Bank Holding Company Act to permit de minimis commercial activities subject to quantitative limits, disclosure, and supervision. A modernized Regulation W could extend arm’s-length rules to equivalent platform arrangements, protecting insured banks without prohibiting integration.

Portability Beats Prohibition

Today’s biggest concern is not that a bank might own a hardware store. It is that digital platforms might use network effects and closed ecosystems to lock in customers.

The better response is openness, not separation. Interoperability, data portability, and low switching costs preserve consumer choice and competitive pressure without banning integration. Data governance should likewise regulate how data is used, not who owns it, while supervision should scale with the risks a firm actually poses.

Banking and commerce already interact. The real challenge is not keeping them apart, but ensuring they compete fairly, innovate responsibly, and protect consumers and financial stability.

For more on this topic, see the ICLE white paper “Tear Down This Wall: Rethinking the Separation of Banking and Commerce” by Todd J. Zywicki.

I.   Introduction

The International Center for Law & Economics (ICLE) submits these comments in response to the National Credit Union Administration’s (NCUA) interim final rule (Rule), which clarifies that federal credit unions (FCUs) may charge noninterest charges and fees, including interchange fees from credit- and debit-card transactions, even when those fees are set by or in consultation with third parties.[1] The Rule further concludes that federal law preempts the Illinois Interchange Fee Prohibition Act (IFPA), which prohibits interchange fees on the tax and gratuity portions of payment-card transactions and restricts the use of payment-card transaction data.[2]

ICLE is a nonprofit, nonpartisan research center that applies law & economics to public-policy questions. Its work promotes sound economic analysis and consumer welfare. ICLE scholars have written extensively on payment-card markets, interchange fees, and payment regulation. These comments draw substantially from ICLE’s white paper, “Regulating State Interchange Fees: Evaluating the Likely Effects of the IFPA,”[3] and related research,[4] largely adapting earlier comments submitted to the Office of the Comptroller of the Currency (OCC) regarding its interim final rule and order.[5]

The IFPA is a novel state law that, absent preemption or a stay, prohibits payment-card issuers, networks, and processors from collecting interchange fees on transaction amounts attributable to state and local sales taxes or gratuities. Although the law purports to reduce merchants’ payment-acceptance costs, it would create economic and legal problems that far outweigh its claimed benefits.

The IFPA would impose a cumbersome two-track compliance regime. Under one approach, merchants would transmit tax and gratuity data in real time during payment authorization. Under the other, merchants would seek rebates for interchange fees collected on exempt amounts after the transaction. Both approaches would impose substantial technological, operational, and compliance costs on payment networks, processors, merchants, and financial institutions while making the payments system less efficient.

Interchange fees are not merely transaction costs. They help payment-card networks balance the two-sided market connecting cardholders and merchants. Interchange revenue supports fraud prevention, payment security, rewards programs, insurance benefits, and continued investment in payment-system innovation. Eliminating interchange fees on taxes and gratuities would disrupt that balance and reduce issuer revenue. Financial institutions would likely respond by reducing rewards, increasing account or card fees, raising borrowing costs, or some combination of the three. Experience with interchange-fee regulation under the Durbin Amendment, as well as in the European Union and Australia, suggests such interventions often reduce consumer benefits without producing corresponding reductions in retail prices. The IFPA also would shift a substantial share of its costs to consumers outside Illinois, giving the law significant extraterritorial effects.

Those effects would multiply if other states enacted similar laws. The IFPA effectively requires national payment systems to accommodate Illinois-specific rules, forcing costly nationwide changes. Numerous states have considered or are considering comparable legislation.[6]  A patchwork of state requirements would fragment the national payments system, increase transaction costs, discourage innovation, and reduce efficiency. The risk is especially acute because states have proposed materially different exemptions and compliance mechanisms. As the NCUA recognized in the Rule, these laws also raise substantial constitutional and federal-preemption concerns.

FCUs have challenged the IFPA as preempted by the Federal Credit Union Act (FCU Act). In that litigation, the federal district court held that NCUA regulations issued under the FCU Act did not preempt the IFPA’s Interchange Fee Provision under conflict-preemption principles, but that the FCU Act’s grant of incidental powers did preempt the law’s Data Usage Provision.[7]

Ultimately, the IFPA’s limited benefits for some merchants would come at the expense of broader harms to consumers, financial institutions, and the payments ecosystem. If replicated nationwide, similar laws would undermine the efficiency and uniformity of the integrated national payments system on which modern commerce depends. The NCUA correctly concluded that FCUs may charge interchange fees and other noninterest charges necessary to participate in modern payment markets. The NCUA also correctly recognized that payment networks may establish those fees because they perform an essential coordinating function within the payment-card ecosystem. The Rule therefore properly preempts the IFPA’s Interchange Fee Provision as applied to FCUs. Sound law & economics principles likewise support extending that preemption to all participants involved in the interchange-fee process.

II.   Electronic Payments and Interchange Fees

Electronic payments are indispensable to modern commerce. During the COVID-19 pandemic, they helped sustain the U.S. and global economies as cash use and face-to-face transactions fell sharply.[8] Studies show that payment cards—whether physical cards or mobile wallets—offer substantial advantages over cash for most transactions.[9] Consumers benefit from greater convenience, stronger fraud protection, short-term liquidity through interest-free grace periods, and, in most cases, zero liability for fraud. Many cards also provide purchase protection, travel insurance, cashback rewards, airline miles, hotel points, and other benefits. These advantages help explain why most consumers prefer cards to cash or checks (Figure 1).

FIGURE 1: Payment-Instrument Share as Proportion of Number of Payments

SOURCE: Federal Reserve Board [10]

Merchants likewise benefit from electronic payments. Payment cards generally speed checkout, increase sales, and reduce the theft and fraud risks associated with cash.[11] When Chicago-based quick-service chain Epic Burger went cashless in 2017, founder David Friedman cited faster transactions, improved safety, and fewer counting errors.[12] Mercedes-Benz Stadium in Atlanta similarly reported shorter transaction times, higher per-capita spending, and lower operating costs after going cashless in 2018.[13] Restaurants and sports venues in Illinois and nationwide have since followed suit.[14]

Payment systems succeed only when both consumers and merchants participate. That participation depends on trust. Payment networks, issuing banks, acquiring banks, and payment processors invest billions of dollars in fraud prevention, cybersecurity, infrastructure, and system maintenance to sustain that trust.[15] Consumers, in turn, use payment cards because they value security, convenience, insurance, rewards, and related benefits. Issuing banks recover much of the cost of providing those benefits through interchange fees.[16]

Interchange fees remain widely misunderstood. They are often described simply as “transaction costs.”[17] That description is incomplete. Interchange fees do cover operating costs, including fraud prevention and network maintenance. They also transfer value from merchants to consumers by financing cardholder benefits, such as rewards, purchase protection, and short-term credit.

All payment systems impose costs.[18] Cash is expensive to print, transport, secure, and process, although many of those costs remain hidden because governments subsidize currency production and distribution.[19] Checks also entail costs and risks, including fraud, processing delays, and nonpayment. Merchants further benefit from legal rules requiring checks to clear at par, which shifts some costs to consumers and financial institutions.[20] Regardless of how costs are allocated, no payment system survives unless both merchants and consumers perceive net benefits from using it.[21] Continued market adoption demonstrates that both sides generally benefit.

Electronic payments differ from cash and checks primarily because the costs of those older payment systems are often obscured by subsidies or legal rules that shift costs away from merchants. There is therefore no economic basis for exempting taxes or gratuities from interchange fees, just as there would be no basis for exempting portions of cash or check transactions from the costs of those payment systems. The visibility of interchange fees does not justify selectively shifting payment-system costs away from merchants.

Economists describe these transfers as “cross-side subsidies” because one side of a two-sided market subsidizes participation on the other.[22] Such arrangements are common. Advertisers subsidize newspapers, search engines, smartphone applications, and large language models. The economics literature has long recognized the importance of cross-side subsidies, and the U.S. Supreme Court recognized their central role in payment-card markets:

Sometimes indirect network effects require two-sided platforms to charge one side much more than the other. For two-sided platforms, “‘the [relative] price structure matters, and platforms must design it so as to bring both sides on board.’” The optimal price might require charging the side with more elastic demand a below-cost (or even negative) price. With credit cards, for example, networks often charge cardholders a lower fee than merchants because cardholders are more price sensitive. In fact, the network might well lose money on the cardholder side by offering rewards such as cash back, airline miles, or gift cards. The network can do this because increasing the number of cardholders increases the value of accepting the card to merchants and, thus, increases the number of merchants who accept it. Networks can then charge those merchants a fee for every transaction (typically a percentage of the purchase price). Striking the optimal balance of the prices charged on each side of the platform is essential for two-sided platforms to maximize the value of their services and to compete with their rivals.[23]

The Supreme Court’s discussion concerned American Express, a three-party payment-card network that combines network operations, card issuance, and acquiring functions within a single firm. The IFPA, by contrast, targets only four-party payment-card networks, in which network operations, issuing, and acquiring are performed by separate entities. Only four-party networks use interchange fees because issuing and acquiring institutions are distinct.

III.   The IFPA’s Economic Costs

Although the IFPA is unusual in exempting only portions of interchange fees, many jurisdictions have adopted interchange-fee caps and similar regulations. Those experiences offer useful lessons about the law’s likely effects. The Durbin Amendment to the Dodd-Frank Wall Street Reform and Consumer Protection Act is especially instructive.

Before the Durbin Amendment, debit-card interchange fees helped issuing banks fund consumer benefits, including debit-card rewards and free checking accounts with low minimum-balance requirements. After the Federal Reserve implemented Regulation II, many banks reduced or eliminated those benefits.

Empirical studies by Federal Reserve economists found that Regulation II’s interchange-fee caps led covered banks to raise account fees, increase minimum-balance requirements, reduce access to free checking, and eliminate debit-card rewards.[24] Those changes disproportionately harmed lower-income households and increased the number of unbanked and underbanked consumers.[25] At the same time, there is little evidence that merchants passed their interchange-fee savings through to consumers as lower retail prices.[26]

The same economic logic applies to the IFPA. The law restricts interchange fees on both sales taxes and gratuities. It applies to credit cards, debit cards, and general-use prepaid cards, and reaches every major participant in the payments ecosystem, including issuers, payment-card networks, acquiring banks, and processors. In effect, it prohibits every participant in the payment chain from “charging or receiving” interchange fees on sales taxes and gratuities.

Compliance would require substantial technological and operational changes. Issuers and payment networks would need to upgrade information-technology systems and transaction-processing software to identify taxes and gratuities, modify authorization systems, and implement tracking or rebate mechanisms for post-transaction adjustments.[27] Yet the IFPA appears not to apply to three-party networks, such as American Express and Discover, because those systems do not rely on interchange fees. The law therefore disadvantages four-party networks while favoring functionally similar competitors without any apparent economic justification.

Implementing the IFPA would require far more than simply “turning off” interchange fees on taxes and gratuities. It would require systemwide changes to the nation’s electronic-payments infrastructure. Payment-card networks and processors would need to revise transaction-message formats and modify the algorithms used to calculate interchange fees for Illinois transactions. At a minimum, compliance would require:

• Reliably identifying Illinois sales taxes and gratuities, including precise amounts that vary by product and locality;
• Modifying authorization and settlement-message formats to transmit that information;
• Altering clearing and settlement systems so interchange fees apply only to permitted transaction amounts; and
• Maintaining recordkeeping and dispute-resolution systems to process merchant claims for reimbursement of interchange fees collected on taxes and gratuities.

Even while holding at summary judgment that the IFPA’s Interchange Fee Provision was not preempted, the district court acknowledged the law’s extraordinary operational complexity. The court observed that compliance would likely require technical capabilities that do “not currently exist, but could possibly exist through additional investment.”[28] It likewise described the resulting compliance costs as “undeniable” and even “staggering.”[29] The NCUA appropriately recognized the same concern, describing the IFPA as a “complex and potentially unworkable” regulatory scheme with enormous potential liability.[30]

Those compliance costs would be substantial on their own, particularly for smaller merchants that rely on the rebate mechanism, which would likely require labor-intensive manual processing. They would fall especially hard on FCUs. Nearly 90% of FCUs hold less than $1.1 billion in assets.[31] Their cooperative structure and limited statutory powers also constrain their revenue sources. Unlike commercial banks, for example, FCUs generally cannot engage in business lending. As a result, they have fewer opportunities to absorb new compliance costs, making the IFPA disproportionately burdensome for smaller credit unions.

The reduction in interchange-fee revenue would impose an additional—and likely larger—cost. Interchange fees finance rewards programs, fraud prevention, and other cardholder benefits. Eliminating interchange fees on taxes and gratuities could reduce issuer revenue by roughly 0.1% of the transaction value for purchases involving Illinois merchants.[32] For large issuers, that could amount to tens of millions of dollars annually.[33] Issuers, both inside and outside Illinois, would likely respond in one or more of several ways.

A. Reduced Cardholder Rewards and Benefits

FCUs would likely respond to the IFPA first by reducing cardholder rewards and benefits. Credit-card issuers finance rewards programs primarily through interchange fees. If the IFPA reduces interchange-fee revenue from Illinois transactions by 10% or more, issuers will have less revenue to support existing rewards programs.

Experience elsewhere strongly suggests this outcome. Following interchange-fee regulation under the Durbin Amendment in the United States, Reserve Bank of Australia regulations, and the European Union’s Interchange Fee Regulation, issuers reduced the generosity of rewards programs.[34] In many cases, those reductions were widespread.

Issuers could also scale back other benefits financed through interchange fees, including travel insurance, purchase protections, and related cardholder services. Australian and European issuers adopted similar measures after interchange-fee regulation.[35] If Illinois remains the only state to impose these restrictions, nationwide FCUs may be able to limit some reductions to Illinois cardholders. Illinois-based credit unions, by contrast, would have fewer options because their membership is concentrated in the state. Even so, any nationwide reduction in benefits would force cardholders outside Illinois to subsidize the Illinois operations of large merchants.

FCUs could instead create Illinois-specific rewards programs to offset lost revenue. That approach would require amendments to nearly every cardholder agreement, along with revisions to agreements with rewards partners such as airlines and hotels. State-specific programs would also complicate customer communications and increase administrative costs.

A more likely response would be modest nationwide reductions in rewards. For example, an FCU that currently offers 2% cashback might reduce rewards to 1.95%, or slightly devalue points across its card portfolio. Economically, reducing rewards functions as a price increase for consumers. Research consistently shows that consumers bear most of the costs of interchange-fee restrictions, while merchants pass through little, if any, of their savings as lower retail prices. Once compliance costs are taken into account, many merchants may realize no net savings at all and therefore have nothing to pass on.[36]

Although nationwide changes would still require amendments to cardholder and partner agreements, they would be simpler to administer than maintaining separate Illinois-specific programs. They would also spread the costs across all cardholders. In effect, Illinois policy would require consumers nationwide to subsidize the Illinois operations of large merchants.

The likelihood of reduced rewards and benefits would increase substantially if additional states adopted similar laws. Even if Illinois remains the only state to impose these restrictions, FCUs may still reduce benefits for Illinois-based members or for transaction categories with especially large tax components, such as gasoline purchases that include substantial fuel taxes.

B. Higher Card and Account Fees

Issuers have often responded to interchange-fee price controls by introducing or increasing cardholder fees. In Australia, for example, banks increased average annual credit-card fees by roughly 50% following interchange-fee regulation.[37] Illinois consumers could face similar outcomes. Cards that currently carry no annual fee could begin charging one, while cards with more generous rewards could become more expensive.

Debit cards and associated checking accounts could also become more costly. That is precisely what occurred after the Durbin Amendment. Some covered banks initially proposed monthly debit-card usage fees to offset lost interchange revenue. Following public backlash, many instead increased monthly account fees and raised minimum-balance requirements for free checking.[38]

Those changes disproportionately burden lower-income consumers, who are less likely to maintain large account balances and more likely to rely on low-cost banking products. Experience with interchange-fee regulation demonstrates that reduced interchange revenue can diminish access to affordable banking services and increase the number of unbanked and underbanked households.

Because FCUs issue a substantial share of the debit and credit cards used in Illinois, the IFPA’s effects would likely extend beyond the state’s borders. If FCUs increase fees across broader customer portfolios rather than creating Illinois-specific pricing, consumers nationwide would bear part of the cost of Illinois’ interchange-fee restrictions. The IFPA would therefore produce significant extraterritorial effects.

Acquiring banks and payment processors could likewise increase their fees to offset both reduced revenue and higher compliance costs, particularly for smaller merchants that rely on the rebate process. As a result, many merchants may realize little, if any, net savings. Some small retailers that do not collect gratuities could even see their total payment-processing costs increase. As ICLE has previously documented, one reason merchants realized smaller-than-expected savings after the Durbin Amendment was that acquirers increased their own fees.[39]

C. Higher Borrowing Costs

Issuers could also respond to the IFPA by increasing borrowing costs, including annual percentage rates (APRs). After the European Union adopted the Interchange Fee Regulation—which capped interchange fees at 0.2% for debit transactions and 0.3% for credit transactions—the spread between the European Central Bank’s policy rate and credit-card APRs widened.[40]

The IFPA alone may not produce a measurable increase in APRs. Even so, reduced interchange-fee revenue would place upward pressure on borrowing costs, particularly for higher-risk borrowers. Consumers ultimately would bear those costs.

If FCUs and other issuers apply higher borrowing costs across their broader card portfolios rather than limiting them to Illinois accounts, the effects would extend well beyond Illinois. As with reduced rewards and higher account fees, the IFPA would shift part of its costs to consumers nationwide.

D. Fragmentated State Payment Rules

Payment-card networks currently operate under largely uniform nationwide interchange-fee schedules. The IFPA would disrupt that system. If other states adopt similar laws exempting sales taxes, gratuities, or other transaction components from interchange fees, payment networks would likely have to maintain multiple state-specific interchange regimes.

That fragmentation would increase operational complexity, compliance costs, and administrative burdens. Networks and processors would need to track differing state requirements, identify covered transactions in real time, and apply different interchange calculations depending on the jurisdiction and transaction type.

The risk is not hypothetical. States have already proposed materially different approaches. Colorado, for example, recently considered legislation that would have exempted smaller banks from interchange-fee restrictions, while many other proposals would exempt sales taxes but not gratuities.[41] If states continue to adopt divergent rules, payment-card networks would face an increasingly complex patchwork of compliance obligations. At sufficient scale, that patchwork would undermine the efficiency and uniformity of the national payments system and could render the current nationwide interchange framework effectively unworkable.

E. Nationwide Increases in Interchange Fees

Payment-card networks could also respond by increasing default multilateral interchange-fee schedules nationwide to offset revenue losses from Illinois transactions. In practice, that would likely mean modest across-the-board increases in interchange fees throughout the country.

Such an approach would shift costs from Illinois merchants to merchants and consumers in other states. Out-of-state merchants would pay higher interchange fees, while consumers outside Illinois would likely bear part of those costs through higher retail prices, reduced rewards, or other pricing adjustments. The IFPA would therefore create another form of extraterritorial cost shifting, spreading its costs nationwide to subsidize Illinois merchants.

IV.   The IFPA Conflicts with FCUs’ Federally Authorized Powers

At the preliminary-injunction stage, the district court applied ordinary conflict-preemption principles and held that the IFPA’s Interchange Fee Provision was not preempted because NCUA rules issued under the FCU Act appeared to address only “state laws regulating fees charged to credit union members in connection with an initial line of credit.”[42] At summary judgment, the court rejected application of the broader Barnett Bank standard under the National Bank Act to the FCU Act and adopted its earlier conflict-preemption analysis.[43]

By limiting which portions of a transaction may carry interchange fees, the IFPA imposes a price control on a fee structure that payment networks continuously calibrate to balance merchants and consumers. Disrupting that balance reduces the value of payment cards to consumers, which reduces usage and, in turn, diminishes value to merchants. The result harms both sides of the platform and materially interferes with FCUs’ exercise of their federally authorized powers.

The NCUA is therefore correct that the IFPA’s Interchange Fee Provision should be preempted as applied to FCUs. The Rule clarifies this preemptive effect by “stat[ing] explicitly that FCUs have authority to charge non-interest charges and fees related to permissible activities.”[44] That includes authority to charge “interchange fees from credit and debit operations.”[45]

The same logic extends further. Other participants whose activities are inseparable from FCUs’ federally authorized payment-card operations—including payment networks and processors—should receive similar preemption from the IFPA and comparable state laws.

As the district court explained in addressing the Data Use Provision:

Both parties admit that under the current system, if the Issuer associated with a transaction is exempt from the IFPA with respect to a transaction, then—to give effect to the Issuer’s exemption—other participants in the payment card ecosystem would need relief from the requirements of the IFPA for purposes of that transaction, though the Attorney General notes that said result may not be necessary were the global payment card ecosystem structured a different way.[46]

Applying “longstanding principles of equity,” the court concluded that the “Data Usage Limitation is so tied up in the federal entities’ powers that the preemptive effect must run to the Payment Card Networks and others involved in the payment process.”[47]

That reasoning accords with multisided-platform economics and applies equally to the IFPA’s Interchange Fee Provision. Payment-card networks operate integrated systems that set interchange fees, participation requirements, authorization protocols, clearing and settlement rules, and liability standards. Networks connect merchants and cardholders through issuing banks, acquiring banks, and processors, while transmitting the information needed to complete transactions. Issuing banks, in turn, use interchange-fee revenue to process payments and fund services for cardholders and deposit customers.

Payment-card networks are integrated platforms, not isolated contracts among merchants, banks, and processors. The NCUA’s Rule, like the OCC’s rule for nationally chartered banks and federal savings associations, recognizes that reality. Interchange fees are the pricing mechanism that holds the system together. Exempting taxes and gratuities from interchange fees would alter the balance between merchants and consumers, shift costs to consumers, and force operational changes across a nationwide—and indeed global—payments network.

That reality bears directly on preemption. A law that prevents FCUs from recovering costs and funding cardholder services conflicts with their federally authorized powers, even if the law formally targets payment networks or transaction components rather than FCUs directly. The NCUA is right to define “charge” to mean “directly or indirectly, through intermediaries, partners, payment networks, interchanges, or other third parties” because payment-card networks charge interchange fees on behalf of issuers, including FCUs.[48]

For the same reason, equitable relief cannot stop with issuing FCUs alone. Because the payment-card system operates as a coordinated platform, obligations imposed on one participant necessarily spill over to networks, processors, and other participants. Partial injunctions therefore cannot fully remedy the interference. The NCUA’s Rule and the OCC’s earlier rule and order are necessary first steps. But the IFPA’s Interchange Fee Provision should not apply to any participant in the interchange-fee process connected to federally regulated banks, savings associations, or FCUs, including payment networks and processors.

V.   The Case for Preemption

Absent preemption, the IFPA and similar state laws would create a fragmented and increasingly unworkable regulatory patchwork. Multiple states have already followed Illinois by proposing or adopting nonuniform interchange-fee restrictions that apply to different institutions, transaction components, and fee categories.

The IFPA’s inclusion of gratuities underscores the lack of any limiting principle. Taxes at least involve government levies that merchants collect and remit. Tips do not. Gratuities are voluntary payments from customers to service workers and function economically as wages. There is no meaningful economic distinction between interchange fees applied to wages embedded in the listed price of a meal or hotel room and interchange fees applied to tips paid through the same card transaction.

Illinois law still allows merchants to retain a limited sales-tax collection allowance for remitting taxes. No comparable state allowance or subsidy exists for processing gratuities. Including tips therefore suggests that the IFPA’s objective is not merely to align interchange fees with government functions, but to impose broader price controls on payment-card processing fees. In practice, exempting gratuities transfers costs from merchants to issuing banks and consumers.

Once states begin carving selected transaction components out of interchange fees, there is little reason the process would stop with taxes and tips. Other politically influential industries could seek exemptions for food, gasoline, medical care, public transportation, child care, vehicle repairs, electric-vehicle charging, or any other favored category. Legislatures would face constant pressure to create narrow and inconsistent exemptions based on political influence and public sentiment.

That logic quickly turns inward. If merchants selling goods on consignment do not retain the full purchase price, should interchange fees apply only to their markup? If a merchant collects government-imposed licensing or registration fees, should those amounts also be exempt? As carveouts multiply, the operational complexity of calculating interchange fees would increase dramatically.

The rapid spread of nonuniform state proposals shows why federal preemption is necessary. The FCU Act empowers the NCUA to prevent states from imposing inconsistent regulatory obligations on FCUs. Without preemption, interchange-fee regulation would become increasingly fragmented, politically driven, and economically disruptive.

VI.   Conclusion

The NCUA correctly concluded that the IFPA is preempted as applied to entities subject to its supervision. It also correctly clarified that when payment-card networks set interchange fees on behalf of FCUs, state efforts to regulate those fees are preempted as applied to the networks as well.

The same logic supports broader equitable relief for other participants that operate the payment-card system. As the IFPA litigation returns to the district court, the reasoning that supports preemption for FCUs also supports relief for payment networks, processors, and other participants whose functions are inseparable from FCUs’ federally authorized powers. Payment-card networks should therefore receive relief from both the IFPA’s Interchange Fee Provision and its Data Use Provision.

If the court enjoins enforcement against those participants, it should also enjoin enforcement against other state-regulated financial institutions, including credit unions. Otherwise, the IFPA would create an uneven regulatory landscape, distort competition in favor of exempted institutions, and leave the payments system with another half-working legal contraption—never an ideal design principle.

More broadly, the IFPA shows why federal preemption remains necessary in nationally integrated financial and payments markets. Interchange fees are not stray transaction costs; they are the pricing mechanism that helps balance a two-sided market connecting consumers and merchants. Allowing states to impose inconsistent carveouts for taxes, gratuities, or other favored transaction components would fragment the payments system, increase operational complexity, shift costs to consumers, reduce cardholder benefits, and undermine the uniform national framework established by federal law.

 

[1] Preemption—Federal Credit Union Non-Interest Charges and Fees, 91 Fed. Reg. 34,725 (June 9, 2026), https://www.govinfo.gov/content/pkg/FR-2026-06-09/pdf/2026-11559.pdf [hereinafter Interim Final Rule].

[2] See id. at 34,726 (“NCUA is issuing this interim final rule (IFR) to consolidate and clarify NCUA’s preemption rules. The IFR clarifies that FCUs have authority under the FCU Act to charge non-interest charges and fees, including interchange fees, and NCUA has exclusive authority over FCUs’ ability to charge non-interest charges and fees.”).

[3] Julian Morris & Ben Sperry, Regulating State Interchange Fees: Evaluating the Likely Effects of the IFPA, Int’l Ctr. L. & Econ. (July 7, 2025), https://laweconcenter.org/wp-content/uploads/2025/07/IFPA-Paper-2025.pdf.

[4] See, e.g., Ben Sperry & Julian Morris, Half a Swipe, Whole Lot of Mess: Platform Economics and the Interchange Fee Cases, Truth on the Mkt. (Feb. 26, 2026), https://truthonthemarket.com/2026/02/26/half-a-swipe-whole-lot-of-mess-platform-economics-and-the-interchange-fee-cases; Julian Morris, State Regulation of Interchange Fees, Int’l Ctr. L. & Econ. (Nov. 15, 2024), https://laweconcenter.org/resources/state-regulation-of-interchange-fees.

[5] Int’l Ctr. L. & Econ., Comments to the OCC on Preempting the Illinois Interchange Fee Act (May 28, 2026), https://laweconcenter.org/resources/icle-comments-to-the-occ-on-preempting-the-illinois-interchange-fee-prohibition-act.

[6] See, e.g., Emma Kinery, States Advance Bills to Regulate Credit Card Fees on Taxes and Tips, State Affairs (Apr. 28, 2026), https://pro.stateaffairs.com/co/finance/interchange-fee-bills-2026.

[7] See Illinois Bankers Ass’n v. Raoul, 819 F. Supp. 3d 882, 907-08 (N.D. Ill. 2026), vacated and remanded, 2026 WL 1291987 (7th Cir. 2026).

[8] See, e.g., Julian Morris, Todd J. Zywicki & Geoffrey A. Manne, The Effects of Price Controls on Payment-Card Interchange Fees: A Review and Update, Int’l Ctr. L. & Econ. (Mar. 4, 2022), https://laweconcenter.org/wp-content/uploads/2022/03/Payments-2021-Lit-Review.pdf.

[9] Julian Morris & Ben Sperry, The Cost of Payments: A Review, Int’l Ctr. L. & Econ. (Aug. 28, 2024), https://laweconcenter.org/resources/the-cost-of-payments-a-review.

[10] Berhan Bayeh et al., 2026 Diary of Consumer Payment Choice, Fed. Rsrv. (2026), https://www.frbservices.org/news/research/2026-findings-diary-consumer-payment-choice.

[11] Morris & Sperry, supra note 9; Claire Wang, Cash Me If You Can: The Impacts of Cashless Businesses on Retailers, Consumers, and Cash Use, Cash Prod. Off., Fed. Rsrv. Sys. (2019), https://www.frbsf.org/wp-content/uploads/sites/7/Cash-Me-If-You-Can-August2019.pdf.

[12] Daniel Gerzina, Epic Burger Is Now Cashless, Tamale Spaceship Closes Wicker Park Restaurant, More Intel, Eater Chi. (June 21, 2017), https://chicago.eater.com/2017/6/21/15846364/epic-burger-cashless-tamale-spaceship-closed-wicker-park-restaurant-am-intel.

[13] Id.

[14] See Morris & Sperry, supra note 9.

[15] See Julian Morris, The Hidden Wealth of Payment Cards: How Innovations in Payments Transform Society, Int’l Ctr. L. & Econ. (Dec. 19, 2024), https://laweconcenter.org/resources/the-hidden-wealth-of-payment-cards-how-innovations-in-payments-transform-society.

[16] See Todd J. Zywicki, The Economics of Payment Card Interchange Fees and the Limits of Regulation, Int’l Ctr. L. & Econ. (June 2, 2010), https://laweconcenter.org/images/articles/zywicki_interchange.pdf.

[17] See Aaron Klein et al., How Better Payment Systems Can Improve Public Transportation, Brookings Ctr. Regul. Mkts. (2023), https://www.brookings.edu/wp-content/uploads/2023/01/20230109_CRM_Klein_TransitPayments_final1.pdf.

[18] See Consumer Fin. Prot. Bureau, Taskforce on Consumer Financial Law 586-88 (2021).

[19] One frequently cited estimate found that a transition to a cashless economy could increase annual GDP by roughly 1% in advanced economies and by as much as 3% in developing economies. See Marks Massi, Godfrey Sullivan, Michael Strauß & Mohammad Khan, How Cashless Payments Help Economies Grow, Bos. Consulting Grp. (May 28, 2019), https://www.bcg.com/publications/2019/cashless-payments-help-economies-grow.

[20] The shift from checks to electronic payments financed by interchange fees helped drive the expansion of free checking accounts and reduce monthly maintenance fees as debit cards became more widely used.

[21] A consumer may choose to pay in Bitcoin and bear the associated costs. By contrast, a merchant that declines to accept Bitcoin incurs no comparable costs.

[22] See Morris, supra note 15.

[23] Ohio v. Am. Express Co., 585 U.S. 529, 536-37 (2018) (internal citations omitted).

[24] See, e.g., Morris, Zywicki & Manne, supra note 8; Mark D. Manuszak & Krzysztof Wozniak, The Impact of Price Controls in Two-Sided Markets: Evidence from U.S. Debit Card Interchange Fee Regulation (Fin. & Econ. Discussion Series No. 2017-074, Fed. Rsrv., July 2017), https://www.federalreserve.gov/econres/feds/the-impact-of-price-controls-in-two-sided-markets-evidence-from-us-debit-card-interchange-fee-regulation.htm; Benjamin S. Kay, Mark D. Manuszak & Cindy M. Vojtech, Bank Profitability and Debit Card Interchange Regulation: Bank Responses to the Durbin Amendment (Fin. & Econ. Discussion Series No. 2014-77, Fed. Rsrv., Sept. 2014), https://www.federalreserve.gov/econres/feds/bank-profitability-and-debit-card-interchange-regulation-bank-responses-to-thedurbin-amendment.htm.

[25] See Geoffrey A. Manne, Julian Morris & Todd J. Zywicki, Unreasonable and Disproportionate: How the Durbin Amendment Harms Poorer Americans and Small Businesses, Int’l Ctr. L. & Econ. (Apr. 25, 2017), https://laweconcenter.org/wp-content/uploads/2017/08/icledurbin_update_2017_final-1.pdf; Vladimir Mukharlyamov & Natasha Sarin, Price Regulation in Two-Sided Markets: Empirical Evidence from Debit Cards, 172 J. Fin. Econ. 104090 (2025), https://www.sciencedirect.com/science/article/pii/S0304405X25001023.

[26] See, e.g., Zhu Wang, Scarlett Schwartz & Neil Mitchell, The Impact of the Durbin Amendment on Merchants: A Survey Study, 100 Econ. Q. 183 (2014); Mukharlyamov & Sarin, supra note 25, at 10 (“Durbin-induced interchange fee savings for gas merchants were too small for their pass-through—even if full—to be discerned with statistical significance.”).

[27] Morris, supra note 15, at 19.

[28] See Illinois Bankers, 819 F. Supp. 3d at 904.

[29] Id.; see also id. at 896 (noting evidence that complying with the IFPA would be “extraordinarily expensive and will drive institutions out of the market”).

[30] Interim Final Rule, supra note 1, at 34,729 (“Despite the complex and potentially unworkable nature of the interchange fee prohibition, the IFPA exposes FCUs to penalties of $1,000 per transaction for failing to comply with its provisions. Given the upwards of 6.5 billion payment card transactions that occur yearly in Illinois, participants in the payment card [market] could be subject to as much as $6.5 trillion in liability per year for non-compliance with IFPA.”).

[31] Nat’l Credit Union Admin., Quarterly Credit Union Data Summary: Q1 2026, at 2 (2026).

[32] See Tax Found., Taxes in Illinois, https://taxfoundation.org/location/illinois (last visited June 30, 2026) (estimating Illinois’s average combined state and local sales tax rate at 8.96%. Assuming gratuities average 7% of sales based on a 15% tip, taxes and tips account for roughly 10% of a typical transaction. At a 1% interchange fee, exempting taxes and tips would reduce interchange-fee revenue by about 0.1% of the transaction value).

[33] According to the U.S. Census Bureau, Illinois retail sales totaled roughly $244 billion in 2022. See U.S. Census Bur., QuickFacts: Illinois, https://www.census.gov/quickfacts/fact/table/IL/PST045223 (last visited May 22, 2026). Assuming modest growth, retail sales likely exceed $250 billion in 2026. Interchange-fee revenue attributable to sales taxes and gratuities could therefore approach $250 million annually. For issuers with market shares of 4% or more, that would imply annual revenue losses of at least $10 million.

[34] See Morris, Zywicki & Manne, supra note 8.

[35] Id.; Iris Chan et al., The Personal Credit Card Market in Australia: Pricing Over the Past Decade, Rsrv. Bank Austl. (2012), https://www.rba.gov.au/publications/bulletin/2012/mar/pdf/bu-0312-7.pdf.

[36] For example, after enactment of the Durbin Amendment, many smaller merchants saw no price reductions because payment networks eliminated special discounts for small merchants and low-dollar transactions while acquirer fees increased. See Manne, Morris & Zywicki, supra note 25; see also Robert Shapiro & Jerome Davis, The Unanticipated Costs and Consequences of Federal Reserve Regulation of Debit Card Interchange Fees 4, Progressive Pol’y Inst. (2025), https://www.progressivepolicy.org/wp-content/uploads/2025/12/PPI_The-Unanticipated-Costs-and-Consequences-of-Federal-Reserve-Regulation-of-Debit-Card-Interchange-Fees_V3.pdf.

[37] Todd J. Zywicki et al., Price Controls on Payment Card Interchange Fees: The U.S. Experience (Geo. Mason L. & Econ. Rsch. Paper No. 14-18, 2014), https://www.law.gmu.edu/pubs/papers/14_18.

[38] Id.; see also Manne, Morris & Zywicki, supra note 25.

[39] See Zywicki et al., supra note 37; Manne, Morris & Zywicki, supra note 25.

[40] See Julian Morris, The Credit Card Competition Act’s Potential Effects on Airline Co-Branded Cards, Airlines, and Consumers, Int’l Ctr. L. & Econ. (Nov. 17, 2023), https://laweconcenter.org/resources/the-credit-card-competition-acts-potential-effects-on-airline-co-branded-cards-airlines-and-consumers.

[41] Julian Morris, Colorado’s Swipe-Fee Fix Is a Big-Box Gift That Would Burden Small Businesses, Colo. Sun (May 28, 2026), https://coloradosun.com/2026/05/28/opinion-colorado-swipe-fee-legislation.

[42] Illinois Bankers Ass’n v. Raoul, 2025 WL 409060, at *3 (N.D. Ill. Feb. 6, 2025).

[43] See Illinois Bankers, 819 F. Supp. 3d at 906–07.

[44] Interim Final Rule, supra note 1, at 34,727.

[45] Id. at 34,728.

[46] Illinois Bankers, 819 F. Supp. 3d at 892.

[47] Id. at 913.

[48] Interim Final Rule, supra note 1, at 34,727.

Europe keeps asking where its technology champions are. In Google Android, the Court of Justice of the European Union (CJEU) offered part of the answer: build a successful platform, and Brussels may spend the next decade treating its architecture as evidence.

The CJEU’s final judgment in Google Android, handed down last week, will be celebrated in Brussels as a triumph of public enforcement over Big Tech. It deserves a less triumphant reading.

The judgment ends an eight-year legal fight by leaving Google and Alphabet with a fine of roughly €4.125 billion for contractual practices tied to Android, Google Search, Chrome, and the Play Store. The court accepted that Google abused its dominance by using Android distribution terms, preinstallation conditions, anti-fragmentation obligations, and related arrangements to favor its own search and browser products.

The fine is painful. The precedent is worse.

The CJEU approved important parts of the General Court’s analysis. It allowed courts to consider economic context without systematically constructing a counterfactual. It also confirmed that liability does not always depend on proof that the practices could foreclose an “as-efficient competitor” (AEC). That test asks whether a rival as efficient as the dominant firm could compete despite the challenged conduct.

That doctrinal signal matters more than the penalty. A €4 billion fine stings. But the larger cost comes from the precedent’s effects on platform design, investment incentives, and the legal expectations facing future European technology firms. If Article 102 of the Treaty on the Functioning of the European Union (TFEU)—which governs abuse of dominance—condemns ordinary platform governance whenever rivals dislike the outcome, Europe will not get more innovation. It will get more litigation, more regulatory redesign of products, and fewer firms willing to build integrated platforms in the first place.

Read the full piece here.

I.   Introduction and Overview

The International Center for Law & Economics (ICLE) submits these comments in response to the Federal Communications Commission’s (FCC) Public Notice seeking comment on AT&T’s Petition for Declaratory Ruling.^[1] ICLE is a nonprofit, nonpartisan research center that applies law & economics methodologies to public-policy questions. Its work promotes sound economic analysis and consumer welfare, especially in fast-changing, technology-driven markets such as telecommunications.

The market has already moved from copper to fiber and wireless. Consumers have left legacy voice service in overwhelming numbers. Yet carriers must still maintain costly, aging copper networks for a small and shrinking group of subscribers.[2] The FCC recognized this problem in the Network Modernization Order, where it acted unanimously and on a bipartisan basis to clear away rules that keep obsolete copper in service.[3]

This proceeding presents a narrower question with real consequences. Can one state’s legacy rules block a transition that federal law authorizes and federal policy favors?

These comments make three points. First, California’s carrier-of-last-resort (COLR) and related universal-service requirements may once have served a useful purpose, but they now function as a prohibition on copper retirement. They require AT&T to keep its legacy network running even where the FCC has authorized discontinuance. Second, capital is finite and AT&T allocates it nationally. The costs of California’s mandate therefore do not stop at the state line. They fall on AT&T’s customers across its footprint through modern infrastructure that goes unbuilt. Third, if the FCC relies on the impossibility exception to Section 2 of the Communications Act, it need not find that AT&T must physically separate the interstate and intrastate components of plain old telephone service (POTS) before preempting California law.

II.   California Rules Block Copper Retirement

Federal law requires only FCC authorization for AT&T to discontinue POTS and retire the copper facilities that support it. California’s rules independently prevent AT&T from doing so.

The obstacle begins with the network itself. AT&T California provides interstate exchange access and intrastate telephone-exchange service over the same local loops, Class 4 and Class 5 switches, trunks, and supporting systems.[4] AT&T therefore cannot retire the facilities that carry interstate service while preserving the facilities that carry intrastate service. The network either remains in operation, or it does not.

California’s COLR rules create the main barrier. As the sole COLR in its service territory, AT&T must provide “basic service” to every residential household and serve every business customer on request throughout that territory. The California Public Utilities Commission (CPUC) describes “basic service” as technologically neutral, but its required elements are not. Directory assistance, white-pages listings, free operator services, and the ability to place and receive voice-grade calls “over all distances” are POTS-specific features that wireless, cable, and Voice over Internet Protocol (VoIP) providers do not include in modern offerings.[5]

The CPUC also has refused to allow any COLR to satisfy its obligation with nonwireline technology unless the CPUC first adopts service-quality standards for the substitute service. It has not adopted those standards for mobile wireless.[6] As a practical matter, AT&T can satisfy its COLR obligation only by continuing to operate the same legacy copper network the FCC has authorized it to retire.

AT&T also cannot escape the obligation by relinquishing its COLR designation. CPUC rules allow relinquishment only if another carrier first agrees to assume the obligation. No carrier will.[7] When more than 200 carriers had the chance to take on AT&T’s COLR duties, every one declined to accept these outdated and unfunded requirements.[8]

AT&T’s 2023 relinquishment application confirms the futility of California’s process. After more than a year of contested proceedings, extensive discovery, and statewide public hearings, the CPUC dismissed the application on a threshold motion and barred AT&T from reapplying for at least a year.[9]

California also layers procedural barriers on top of the substantive COLR requirement. State law requires basic service to be tariffed and bars detariffing of basic-exchange service. Any change to, or withdrawal of, that tariff requires advance notice and a formal CPUC proceeding that the CPUC may reject.[10] To discontinue POTS, AT&T also would need CPUC approval of customer-notice plans and would have to file a 19-point “exit plan” under the Mass Migration Guidelines. AT&T must continue providing existing service until the CPUC approves each submission.[11]

Each step requires separate CPUC approval. None carries an enforceable deadline that the CPUC honors. Any one can be denied outright. Together, these rules give California a practical veto over copper retirement. Even after the FCC authorizes discontinuance, California requires AT&T to keep powering, maintaining, and selling POTS over a copper network that serves roughly 3% of households and costs about $1 billion a year to operate. Those resources cannot then fund the fiber and wireless networks consumers actually use.[12]

III.   California’s Copper Mandate Harms Customers Nationwide

The case for copper retirement is, at bottom, an argument about waste. California’s COLR regime requires AT&T to pour finite capital into infrastructure whose costs are large and rising, even as its benefits approach zero.[13]

The customer base for legacy copper has collapsed. Nationwide, copper last-mile subscribers fell roughly 81% between 2014 and 2024, from about 66 million to 12.5 million. By 2024, about 79% of U.S. adults lived in wireless-only households, while fewer than 1% lived in landline-only households.[14] In AT&T’s California territory, the numbers are starker. The company spends about $1 billion a year to operate a copper network that serves only about 3% of households, and that share continues to fall as customers move to wireless and Internet Protocol (IP)-based alternatives.[15] Nationally, AT&T spends roughly $6 billion a year—close to 5% of revenue—on a shrinking legacy footprint.[16]

The savings from copper retirement are large and well-documented. Verizon’s migration of 4.5 million circuits to fiber generated roughly $180 million a year in savings and reduced maintenance dispatches by about 60%. All-fiber networks cost roughly $91 less per home each year to operate than copper-based digital subscriber line (DSL) networks.[17]

Energy savings account for much of the difference. One carrier’s copper service consumed roughly 172 kilowatt-hours per subscriber each year, compared with just 6 kilowatt-hours for fiber, a 97% reduction.[18] AT&T’s copper-to-fiber conversions saved an estimated 340,000 megawatt-hours of electricity in 2024 alone.[19] Completing the transition for remaining copper subscribers could save roughly $398 million to $830 million in annual energy costs.[20] A forced-maintenance mandate turns those recurring costs into a permanent drag on communications providers.

Forced maintenance also imposes public-safety costs that California’s COLR rules largely ignore. As copper prices climbed from about $2.29 per pound in 2020 to roughly $6 by early 2026, copper theft surged.[21] AT&T alone reported about 8,700 theft incidents in 2025 at a cost of nearly $76 million. Across the industry, roughly 15,540 theft-and-sabotage incidents between mid-2024 and mid-2025 disrupted service for an estimated 9.5 million customers, with collateral effects on 911 systems, hospitals, and military installations.[22]

By requiring carriers to keep valuable, deteriorating copper in the ground long after customers have left it, California’s rules make copper theft more attractive. The obsolescence problem compounds the risk. The core Class 5 switches that anchor the network—the Lucent 5ESS, Nortel DMS-100, and Siemens EWSD—have not been manufactured in decades, forcing carriers to scavenge replacement parts on secondary markets.[23] In one widely cited example, Tinker Air Force Base sourced 5ESS components on eBay.[24] A mandate to maintain service over equipment that carriers can no longer reliably repair is a mandate to manage slow, expensive failure.

The decisive point for customers is opportunity cost. Capital is scarce, and providers allocate investment from national budgets.[25] Every dollar locked into maintaining a copper office in California is a dollar unavailable for fiber and 5G deployment elsewhere in AT&T’s footprint. The roughly $1 billion a year that California’s COLR rules freeze reduces the resources available for next-generation deployment nationwide.

The economic stakes are significant. Completing nationwide fiber deployment would generate roughly $3.24 trillion in net present value and about 380,000 jobs.[26] International experience points in the same direction. WIK-Consult’s 2020 study found that lengthy regulatory notice periods delayed copper switch-off in Europe even where fiber was already available. That experience counsels shorter timelines once adequate alternatives exist.[27]

California’s regime therefore imposes costs in two directions. Within California, customers bear the costs of a less reliable, less energy-efficient, and theft-prone network maintained for their nominal benefit, even though nearly all of them have chosen newer alternatives. Across AT&T’s broader footprint, customers bear the cost of modern infrastructure that goes unbuilt because capital remains trapped in legacy copper.

The FCC’s modernization proceedings are designed to prevent that result. A state rule that stalls the transition after the FCC has authorized discontinuance does not preserve a meaningful benefit for Californians. It imposes a diffuse, recurring loss on everyone AT&T serves.

IV.   The FCC May Preempt State Copper Mandates

Express preemption under Section 214(c) offers the cleanest path to the relief AT&T seeks, but that path depends on whether AT&T’s California POTS is interstate or jurisdictionally mixed service subject to the FCC’s Section 214 authority.[28] Section 214(c) provides that, once the FCC authorizes discontinuance, a carrier may proceed “without securing approval other than such certificate.” California’s COLR and related requirements demand exactly those additional approvals. Section 214(c) therefore displaces them if the service falls within the FCC’s Section 214 authority.

Section 214 does not reach wholly intrastate services. If the FCC determines that AT&T’s California network is not jurisdictionally mixed, preemption would likely rest on the impossibility exception to Section 2(b) of the Communications Act, which preserves state authority over intrastate lines. Courts have applied that exception where it is impossible or impracticable to separate the interstate and intrastate components for the regulation at issue, and where state regulation would frustrate a valid federal objective.

A. Impossibility Includes Practical Impossibility

Even if the FCC concludes that AT&T could physically separate some interstate and intrastate components,[29] the impossibility exception does not require such a rigid inquiry. The governing standard, as the 8th U.S. Circuit Court of Appeals framed it, asks whether “it is not possible to separate the interstate and intrastate aspects of the service.”[30] Courts and the FCC have long understood “not possible” to include arrangements that may be imaginable in theory but unworkable in practice because of economic or operational burdens.[31]

The relevant question is therefore not whether an engineer could draw a jurisdictional line on a network diagram. It is whether forcing a carrier to separate the facilities would impose costs and operational burdens that defeat the federal interest the FCC seeks to protect.

The North Carolina decisions provide the basic illustration. The state argued that its regulation of customer-premises equipment could coexist with the FCC’s contrary federal rule because customers could, in theory, maintain one set of equipment for intrastate calls and another for interstate calls.[32] The 4th U.S. Circuit Court of Appeals upheld preemption, finding reasonable the FCC’s determination that duplicate equipment was “a practical and economic impossibility.”[33] The point is straightforward. The impossibility exception turns on real-world feasibility, not theoretical divisibility. A separation that exists only on paper, that consumers would not adopt, and that the market would not support does not defeat preemption.

The FCC’s analysis therefore need not focus only on whether physical separation is technically possible. It should also consider whether a provider could practically separate existing infrastructure to retire interstate services while maintaining copper networks for wholly intrastate calls. The Fahmy Declaration describes AT&T’s network in detail and explains why separation may be both physically and practically impossible. As the declaration explains:

AT&T California cannot cease offering long distance service and decommission its facilities provisioning long distance service without also decommissioning its facilities that provide intrastate service—these facilities are one and the same. Thus, either AT&T California continues to spend around $1 billion a year keeping its POTS network in California running or AT&T California does not. There is no option for a partial retirement of just the interstate long-distance part of the POTS network.[34]

Even if the FCC disagrees that separation is technically impossible, the record shows that separation would impose significant costs. The Fahmy Declaration also explains that such separation would leave California POTS customers able to place only in-state calls.[35] The existing rules also force providers to maintain copper connections to interstate networks, even where providers already have supplied customers with IP-based connectivity.

B. California’s Rules Frustrate Federal Policy

The second prong of the impossibility exception asks whether federal regulation is necessary to advance a valid federal regulatory objective.[36] The first prong concerns the technical and economic relationship between interstate and intrastate service. The second prong confirms that preemption serves a genuine federal aim rather than displacing state authority for its own sake.

That requirement is met here. The FCC has identified a clear federal objective: accelerating the retirement of legacy copper networks and the transition to next-generation IP-based infrastructure. California’s COLR regime works directly against that objective.

In the Network Modernization Order, the FCC found that legacy mandates “have been unduly prolonging the use of legacy networks and actually preventing providers from building modern ones.”[37] It adopted reforms designed to cut “the red tape that has both required providers to keep aging copper lines in place and effectively prevented them from investing in the modern infrastructure that Americans want and deserve.”[38] The FCC expressly identified the transition to next-generation networks as a federal regulatory objective and used its Section 214 authority to advance that objective. That is the kind of valid federal objective the impossibility exception protects.

California’s COLR regime conflicts with that objective at its core. It requires AT&T to continue offering “basic service” that, in practice, only POTS can satisfy. That requirement forces AT&T to keep its copper network powered, maintained, and in service indefinitely, regardless of the FCC’s authorization to discontinue service.

The FCC has already explained the problem. Where “state and local requirements prevent a provider from discontinuing the interstate portion of a legacy voice service for which the Commission has already granted discontinuance authorization,” those requirements “negate a valid federal regulatory objective.”[39] California’s rules hold federal authorization hostage to a state obligation that can be satisfied only by maintaining the very network the FCC sought to retire. Resources the FCC intended to free for broadband deployment remain trapped in obsolete copper, defeating federal policy when implementation matters most.

That conflict remains even if the CPUC claims its rules are “technologically neutral” and could be satisfied over a modern network.[40] As explained above, the practical operation of the COLR regime forecloses the copper-free compliance the CPUC describes. The FCC’s policy is not that carriers may modernize after completing another round of state proceedings. It is that carriers authorized to discontinue legacy service may “proceed with the . . . discontinuance” without further approval.[41] A state regime that substitutes its own timeline and conditions for the FCC’s judgment directly frustrates that objective.

V.   Conclusion

California’s carrier-of-last-resort regime belongs to a monopoly era that no longer exists. It singles out one provider, imposes a largely unfunded obligation that only obsolete copper can satisfy, and delivers shrinking benefits to customers who have overwhelmingly chosen fiber, wireless, and IP-based alternatives. It also strands roughly $1 billion a year that could otherwise support the networks consumers actually use.

The costs do not stop in California. Because AT&T allocates capital nationally, every dollar locked into maintaining legacy copper reduces the resources available for fiber and 5G deployment across the company’s footprint. California’s rules therefore impose a recurring opportunity cost on customers nationwide, while preserving a less reliable, less energy-efficient, and theft-prone network for a small and declining group of subscribers.

That regime cannot be reconciled with federal law or federal policy. Once the FCC authorizes discontinuance under Section 214, federal law permits the carrier to proceed without securing additional approval. A state requirement that conditions, delays, or second-guesses that authorization conflicts with the FCC’s decision to accelerate the transition away from legacy copper and toward next-generation networks.

The FCC should declare that California’s carrier-of-last-resort rules, tariffing requirements, and related obligations are preempted to the extent they impede AT&T from discontinuing plain old telephone service once the FCC has authorized discontinuance.

[1] AT&T Servs., Inc., Petition for Preemption and Declaratory Ruling, WC Docket No. 26-125 (filed May 20, 2026), https://www.fcc.gov/ecfs/document/1052056507747/1 [hereinafter AT&T Petition]; Wireline Competition Bureau Seeks Comment on AT&T’s Petition for Preemption and Declaratory Ruling, Public Notice, DA 26-520, WC Docket No. 26-125 (rel. May 22, 2026), https://docs.fcc.gov/public/attachments/DA-26-520A1.pdf.

[2] Eric Fruits & Brian Albrecht, Paying to Stand Still: Legacy Copper Mandates in a Fiber World, Int’l Ctr. for L. & Econ. (Feb. 27, 2026), https://laweconcenter.org/resources/paying-to-stand-still-legacy-copper-mandates-in-a-fiber-world.

[3] Reducing Barriers to Network Improvements and Service Changes; Accelerating Network Modernization, Report and Order, WC Docket Nos. 25-209, 25-208, ¶ 4 (rel. Mar. 27, 2026), https://docs.fcc.gov/public/attachments/DOC-419252A1.pdf [hereinafter Network Modernization Order].

[4] AT&T Petition, supra note 1, at 29.

[5] Id. at 12.

[6] Id.

[7] Id. at 4.

[8] Id. at 35.

[9] Id. at 16–18.

[10] Id. at 13–15.

[11] Id. at 46.

[12] Id. at 3, 10

[13] See Comments of the Int’l Ctr. for L. & Econ., Reducing Barriers to Network Improvements and Service Changes; Accelerating Network Modernization, WC Docket Nos. 25-208 & 25-209 (filed Aug. 22, 2025), https://laweconcenter.org/resources/icle-comments-to-the-fcc-on-the-copper-retirement-nprm [hereinafter ICLE Copper Retirement Comments].

[14] Fruits & Albrecht, supra note 2.

[15] Fruits & Albrecht, supra note 2; AT&T Petition, supra note 1, at 3, 9–10.

[16] AT&T Petition, supra note 1, at 3, 9–10.

[17] Fruits & Albrecht, supra note 2, at 9.

[18] Id. at 2.

[19] Id. at 1.

[20] Id. at 2.

[21] Id. at 10–11.

[22] Id.

[23] Id. at 12–13.

[24] Id. at 13.

[25] ICLE Copper Retirement Comments, supra note 13, at 9.

[26] Brattle Grp., Fiber Deployment Has Significant Incremental Economic Benefits (2024), https://www.brattle.com/insightsevents/publications/fiber-deployment-has-significant-incremental-economic-benefits-according-to-a-recent-brattle-report.

[27] WIK-Consult, Copper Switch-Off: European Experience and Practical Considerations (White Paper, Q3 2020), https://www.wik.org/fileadmin/Studien/2020/Copper_switch-off_whitepaper.pdf.

[28] 47 U.S.C. § 214(c).

[29] Declaration of Dr. Hany Fahmy ¶ 9 at 6, Petition of AT&T for Preemption and Declaratory Ruling, WC Docket No. 26-125 (filed May 20, 2026) (Ex. 1) [hereinafter Fahmy Declaration].

[30] Minn. Pub. Utils. Comm’n v. FCC, 483 F.3d 570, 578 (8th Cir. 2007).

[31] See id. (“It was proper for the FCC to consider the economic burden of identifying the geographic endpoints of VoIP communications in determining whether it was impractical or impossible to separate the service into its interstate and intrastate components.”).

[32] Id. at 578–79.

[33] Id.

[34] Fahmy Declaration, supra note 29, at 6.

[35] Id.

[36] Minn. Pub. Utils. Comm’n, 483 F.3d at 578.

[37] Network Modernization Order, supra note 3, ¶ 4.

[38] Id. ¶ 1.

[39] Id. ¶ 106

[40] AT&T Petition, supra note 1, at 35.

[41] Network Modernization Order, supra note 3, ¶ 106.

The American A.I. Sovereign Wealth Fund Act rests on a sweeping claim about the ownership of value created by artificial intelligence. Because AI models are trained on data generated by the public, the bill treats the resulting gains as a public resource subject to state control and redistribution.

Sen. Bernie Sanders’ (I-Vt.) proposal would require covered AI developers to transfer up to 50% of their corporate value to a new federal sovereign wealth fund. That fund would distribute “dividends” to the public and use its ownership stake to steer AI development “in the public interest.”

The bill therefore raises questions that go well beyond artificial intelligence. It implicates basic principles of value creation, property rights, corporate governance, political choice, and capital formation. Its central premise is that public data gives rise to public ownership. That premise confuses the availability of information with the entrepreneurial and technical process required to transform information into a productive asset.

This piece argues that the Sanders bill rests on four related errors. First, it treats raw data as the source of economic value, while discounting the entrepreneurial discovery and technical judgment that make data useful. Second, it assumes a federal commission can identify and impose a coherent public interest on a technology marked by conflicting preferences and rapid change. Third, it would weaken the market for corporate control and disrupt integrated firm structures that often reduce transaction costs and improve coordination. Fourth, it would distort capital formation by creating confiscation risk and encouraging firms to organize around a political threshold rather than consumer demand.

Read the full piece here.

The Supreme Court just told police they cannot turn your phone into a witness against you merely because you walked through the wrong patch of pavement. That is the good news. The less comforting news is that the Court reached that result by dragging some badly aging Fourth Amendment doctrine along for the ride.

In its June 29 Chatrie v. United States decision, the Court held 6-3 that law enforcement conducts a Fourth Amendment “search” when it forces companies like Google to turn over users’ location-history data through a geofence warrant.

That is a major win for digital privacy. A geofence warrant lets police demand information about every device in a defined area during a set time—essentially asking first and sorting suspects later. The Court was right to rein in that digital dragnet.

But peer under the hood of Justice Elena Kagan’s majority opinion, and the legal engine sputters. The Court reached the right destination, but took the scenic route through a swamp.

To preserve the aging Katz “reasonable expectation of privacy” test—and its creaky cousin, the third-party doctrine—the majority tied itself into a logical pretzel. Justice Neil Gorsuch, concurring only in the judgment, offered the cleaner and more textually grounded path the Court should have taken.

Read the full piece here.

The Digital Markets Act (DMA) was built to police digital gatekeepers. The European Commission now wants to test how far that metaphor can stretch—past app stores, social networks, and marketplaces, and into the server racks.

The Commission has reached the preliminary view that Amazon Web Services (AWS) and Microsoft Azure should be designated as gatekeepers under the DMA. That would mark two firsts. It would be the first time cloud computing falls within the DMA’s reach, and only the second time the Commission uses Article 3(8)—the provision that allows it to designate firms that do not meet the law’s numerical thresholds after conducting a market investigation. While the Commission did invoke Article 3(8) in its iPad OS designation, quantitative thresholds still did most of the work in that case.

AWS and Azure are the two largest cloud providers operating in the European Union. But the DMA’s user-number thresholds were built for consumer-facing platforms, not cloud computing, which is overwhelmingly a business-to-business service. Both providers therefore fall outside those thresholds. To designate them anyway, the Commission must do more than invoke the DMA’s built-in presumptions. It must show, with actual evidence, that AWS and Azure serve as “important gateways” and enjoy “entrenched and durable” market positions.

That makes these designations far more interesting than another lap around the DMA enforcement track. Cloud services do not obviously operate as “gates” in the way two-sided platforms do. They do not sit between business users and end users in the familiar app-store or marketplace sense. One of the DMA’s core rationales—increasing contestability by prying open bottlenecks controlled by gatekeepers—appears, at least at first glance, to be missing.

So the question is not merely whether AWS and Azure are large. Plainly, they are. The question is whether “gatekeeper” remains a meaningful legal category that separates firms with durable bottleneck power from firms that are big in competitive markets. Or is it just Brussels-speak for size, with the statutory criteria serving as the ceremonial chant before the inevitable designation?

Until now, the Commission has relied on the quantitative presumptions in Article 3(2) to designate gatekeepers. The qualitative criteria have never had to carry the load on their own. With AWS and Azure, they finally do.

Read the full piece here.