Research Programs
More
What are you looking for?
Showing Latest Publications
ICLE Issue Brief Executive Summary California Assembly Bill 1776, the COMPETE Act, would expand the state’s Cartwright Act to reach single-firm conduct for the first time in the Act’s 119-year history. It would...
California Assembly Bill 1776, the COMPETE Act, would expand the state’s Cartwright Act to reach single-firm conduct for the first time in the Act’s 119-year history. It would create state-law liability for monopolization, attempted monopolization, maintenance of monopoly power, and monopsonization, moving California well beyond federal Section 2 doctrine.
The bill lowers the core thresholds federal courts use to separate anticompetitive conduct from competition on the merits. It makes optional market-power thresholds, recoupment in predatory-pricing cases, cross-market balancing for multi-sided platforms, as-efficient-competitor analysis, and quantitative evidence of harm. It then directs courts to interpret the law liberally and maximize deterrence.
That framework invites over-enforcement. It would expose aggressive price competition to predatory-pricing claims, require one-sided analysis of multi-sided platforms, extend monopsony liability into unsettled labor-market doctrine, and replace the consumer welfare standard with a broad mandate to protect “all trade participants” and broader social interests.
The likely result is more strategic litigation, more settlements driven by defense costs, and more pressure on firms to raise prices, reduce service quality, limit product integration, or alter employment practices to avoid liability. Because many firms serving California operate nationally, AB 1776 would also set de facto national antitrust rules and raise unresolved dormant Commerce Clause concerns.
The Legislature should proceed cautiously before enacting the COMPETE Act as written. Any single-firm conduct reform should restore recoupment, permit cross-market balancing, require market-power thresholds, retain the consumer welfare standard, address extraterritorial effects, and exclude single-firm conduct from criminal liability.
California Assembly Bill 1776, the COMPETE Act (Competition and Opportunity in Markets for a Prosperous, Equitable, and Transparent Economy),[1] would significantly expand the Cartwright Act[2] —California’s primary antitrust statute—by extending it to reach single-firm conduct for the first time in the Act’s 119-year history. To date, the Cartwright Act has targeted concerted action, leaving unilateral conduct largely to federal antitrust law—the Sherman Act, the Clayton Act, and the Federal Trade Commission Act.[3] AB 1776 would upend that allocation by creating state-law liability for monopolization, attempted monopolization, maintenance of monopoly power, and monopsonization by a single firm.
Proponents frame this change as closing a “gap” in California law. But the absence of single-firm liability under the Cartwright Act reflects a longstanding division of labor between state and federal regimes, under which Section 2 of the Sherman Act governs unilateral conduct.[4] While some states maintain single-firm conduct provisions, many track federal law or include harmonization clauses. Those clauses typically direct courts to interpret state statutes consistently with federal law, even if they do not require strict conformity.[5] AB 1776 breaks from that approach. It expressly provides that federal precedent is not binding, inviting state courts to adjudicate claims outside established federal doctrine.
At the same time, the bill discards much of the analytical framework that gives content to single-firm liability. Section 16732 lists 10 requirements commonly applied by federal courts and provides that none is necessary to establish liability under California law. Section 16733 instructs courts to “liberally interpret” the statute while “maximizing” deterrence. Section 16730 further provides that federal precedent is relevant only to the extent it is consistent with California law. The bill thus expands liability while rejecting the doctrinal tools that have traditionally structured and limited its application—without offering a coherent alternative.
The bill rests on two contestable premises. First, it assumes that exclusionary single-firm conduct is widespread and harmful enough to justify a new state cause of action, but the statutory text does not specify that theory of harm, and the economics literature does not support it. Second, it rejects the doctrinal standards federal courts use to distinguish anticompetitive conduct from vigorous competition, yet offers no workable substitute.
AB 1776’s remedies—criminal liability, treble damages, private rights of action with low pleading thresholds, and no market-power requirement—are disproportionate to the harms its proponents identify, even on their own terms. The bill also lowers core liability standards. It eliminates any recoupment requirement for predatory pricing, bars cross-market balancing for multi-sided platforms, and does not require proof of market power. As a result, plaintiffs—both public and private—could prevail on claims that federal courts would reject.
These features carry significant spillover effects. The firms most likely to face liability under AB 1776 operate nationally or globally. California courts would therefore set de facto national antitrust standards by applying a lower liability threshold.
The bill applies to any firm engaged in trade or commerce in California, regardless of where it is incorporated or headquartered. A manufacturer based in Ohio, a Delaware-incorporated retailer, or a Texas-based logistics firm all face Cartwright Act liability if their conduct affects California markets or participants. The statute imposes no industry or size limits. Any company that employs California workers, sells to California consumers, or competes in California markets falls within its scope.
That breadth has practical consequences. The firms most exposed to AB 1776 are not primarily California-based. They are national and multinational companies whose California operations cannot be separated from their broader business models. Firms will not maintain one set of practices for California and another elsewhere. Instead, they will adjust their conduct nationwide to comply. The bill’s extraterritorial effects are not incidental—they are the predictable result of applying a state-specific antitrust regime to firms that operate across integrated national and global markets.
The Cartwright Act, enacted in 1907, formed part of a broader wave of state competition laws aimed at curbing cartel behavior and supplementing common-law remedies for monopolies and price-fixing.[6] The statute targets “trusts,” defined as combinations of two or more persons to restrain trade or suppress competition. That focus on coordinated conduct has anchored enforcement against horizontal price-fixing, bid-rigging, and market-allocation agreements.
By contrast, Section 2 of the Sherman Act governs unilateral conduct. A single firm violates federal law when it acquires or maintains monopoly power through exclusionary practices. The Cartwright Act contains no comparable provision. The California Supreme Court confirmed in Cianci v. Superior Court that the Act is “broader in range and deeper in reach” than the Sherman Act within its domain,[7] but courts have consistently recognized that this domain does not extend to unilateral conduct.[8]
As a result, California law does not provide a direct cause of action against a single firm that refuses to deal with a rival on terms that would enable competition, acquires nascent competitors to eliminate emerging threats, or uses market power to impose exclusionary contracts. Federal law reaches these forms of conduct under Section 2, but federal litigation is costly, time-consuming, and imposes demanding evidentiary burdens.
The California Law Revision Commission (CLRC) argues that state antitrust law can complement federal enforcement by expanding enforcement capacity, enabling additional private plaintiffs, and tailoring remedies to California-specific conditions.[9]
Assembly Bill 1776 largely codifies—nearly verbatim—the recommendations of the California Law Revision Commission (CLRC), which spent several years evaluating whether to extend state antitrust law to single-firm conduct. The bill adopts the CLRC’s core design choices: an industry-neutral framework that applies across the economy, a directive to maximize deterrence, and a break from federal antitrust doctrine as a binding guide.
Understanding AB 1776 therefore requires examining both the CLRC’s process and its proposed statutory framework. The bill translates those recommendations into enforceable law by expanding liability to unilateral conduct, lowering or eliminating traditional evidentiary screens, and granting courts broad discretion to define the contours of liability without a clear alternative framework.
The CLRC spent several years reviewing the state’s antitrust laws, culminating in recommendations that underpin Assembly Bill 1776. The CLRC proposed a new single-firm conduct provision that applies across the entire California economy, rather than targeting specific sectors such as technology. AB 1776 adopts this industry-neutral approach.
At the same time, the CLRC’s recommendations—enacted largely without modification—depart sharply from established federal antitrust principles. The CLRC proposed instructing courts to maximize deterrence of antitrust violations, drawing on Clayworth v. Pfizer, Inc.[10] AB 1776 § 16733 codifies that directive verbatim. The CLRC also introduced “Judicial Guidance”—a list of 10 federal evidentiary screens that “may constitute evidence” of a violation but are not required to establish liability.[11] AB 1776 § 16732 adopts that list in full, making optional key elements of federal analysis, including market power (subdivision (i)), recoupment in predatory pricing (subdivision (g)), cross-market balancing for multi-sided platforms (subdivision (f)), and the as-efficient-competitor test (subdivision (h)). The CLRC further proposed barring courts from offsetting harms in one market with benefits in another; AB 1776 § 16731(b) adopts that rule. In each instance, the bill tracks the CLRC’s recommended language.
While AB 1776 closely follows the CLRC’s recommendations, ICLE scholars and others sharply criticized the Commission’s process and conclusions. In formal comments, they argue that the decision to untether California antitrust law from the federal error-cost framework and the consumer welfare standard is misguided.[12] By favoring the risk of over-enforcement (false positives) over under-enforcement (false negatives), the CLRC’s approach risks chilling procompetitive conduct, including price cutting and vertical integration.
These critics also contend that the CLRC’s framework effectively rejects key U.S. Supreme Court precedents, including Brooke Group, Trinko, and Amex, and instead aligns California law more closely with European Union competition policy.[13] They further argue that the CLRC assumes federal antitrust law has failed, yet offers little empirical evidence that California consumers or businesses suffer from reduced competition due to gaps in current enforcement.
In practice, the CLRC’s framework lowers the evidentiary burden for plaintiffs without defining a clear alternative standard. It directs courts to reduce reliance on established screens but leaves them to determine what suffices to establish liability. These changes benefit plaintiffs’ attorneys, who gain broader grounds for treble-damages litigation; state enforcers, who gain expanded authority; and competitors seeking new claims against dominant firms. While these groups have legitimate interests, a deterrence-maximizing approach departs from the prevailing law & economics consensus, which emphasizes protecting consumer welfare and avoiding over-deterrence that can harm the broader economy.
AB 1776 would add Sections 16730 through 16733 to the California Business and Professions Code. The bill’s enactment clause states: “An act to add Sections 16730, 16731, 16732, and 16733 to the Business and Professions Code, relating to business regulations.” It does not amend Section 16720, the Cartwright Act’s definition of “trust,” which still requires “two or more persons.” Instead, new Section 16731(a)(1) provides that “restraint of trade” includes conduct “cognizable under Section 16720, whether directed, caused, or performed by one or more persons.” This incorporation-by-reference approach extends Cartwright Act liability to single-firm conduct—covering unilateral pricing, exclusive dealing, refusals to deal, product design, and contracting practices across all industries—without revising the statute’s core definitions.
Section 16731 makes it unlawful for one or more persons to restrain trade or to “monopolize or monopsonize, to attempt to monopolize or monopsonize, to maintain a monopoly or monopsony, or to combine or conspire with another person to monopolize or monopsonize.” The inclusion of “monopsonize” marks a significant expansion. It introduces buyer-side liability into California antitrust law, extending coverage to labor markets and other input markets.
Section 16730(b) underscores that shift by identifying “workers’ freedom to choose employment” as a protected competitive interest. The bill thus explicitly incorporates labor-market competition into the statute’s core objectives.
Section 16731 requires courts to evaluate anticompetitive effects and procompetitive justifications “within the same relevant market,” prohibiting cross-market balancing.
Section 16732 lists factors that may inform liability but are not required. These include: termination of a prior course of dealing; differential treatment of rivals; below-cost pricing; conduct that “makes no economic sense” absent a harmful purpose; quantitative evidence of harm; harm on multiple sides of a platform or harm outweighing benefits on another side; a probability of recoupment in predatory pricing; comparison to an as-efficient competitor; market-share or market-power thresholds recognized under federal Section 2; and a defined relevant market where direct evidence of market power exists.
Section 16733 directs courts to interpret California antitrust law liberally and to remain “mindful that California favors ‘maximizing’ effective deterrence of antitrust violations.” Section 16730 further provides that federal precedent is not binding and may be considered only to the extent it is “consistent with California law.”
Enforcement follows existing Cartwright Act mechanisms. Private plaintiffs, the attorney general, district attorneys, and qualifying city attorneys may bring suit. Available remedies include injunctive relief, treble damages, and attorneys’ fees. The Act’s criminal provisions also apply to violations of AB 1776.
AB 1776 rests on two core assumptions: that exclusionary single-firm conduct is both widespread and harmful enough to justify a deterrence-maximizing regime, and that courts can identify such conduct without the analytical screens that structure federal antitrust law. Both assumptions warrant scrutiny.
Sections IV.A through IV.C examine the economic and legal implications of that shift. Together, they highlight three throughlines: the tradeoff between false positives and false negatives, the risk of protecting competitors rather than competition, and the move away from the consumer welfare standard toward a set of competing policy goals without a clear method of resolution. By lowering evidentiary thresholds and prioritizing deterrence, AB 1776 increases the likelihood of over-enforcement, weakens the distinction between harmful and beneficial conduct, and leaves courts to resolve competing interests without a principled framework.
Antitrust enforcement produces two types of errors. A false positive condemns procompetitive conduct—penalizing behavior that benefits consumers. A false negative allows anticompetitive conduct to persist without remedy. Frank H. Easterbrook, writing before his appointment to the 7th U.S. Circuit Court of Appeals, identified the asymmetry between these errors that makes careful calibration essential: markets can partially self-correct false negatives, because above-competitive profits attract entry that erodes monopoly power over time.[14] False positives do not self-correct. An erroneous condemnation deters the challenged conduct across future periods through legal precedent, chilling similar conduct by other firms.[15] ICLE scholars apply this framework to California’s reform proposals in comments to the CLRC and subsequent analyses.[16]
A maximally deterrent antitrust regime increases the risk of false positives by design. It captures more anticompetitive conduct, but only by also condemning more procompetitive conduct that less aggressive standards would permit.
The CLRC explicitly prioritizes maximizing deterrence over calibrating enforcement to specific harms. It recommends departing from federal standards—such as recoupment and below-cost pricing requirements—which it characterizes as “rigid rules” that can “unduly restrict” enforcement.[17] While the CLRC rejects an “abuse of dominance” standard due to vague thresholds, it adopts a broad “restraint of trade” framework to reach a wider range of conduct than federal law.[18] AB 1776’s directive to maximize deterrence reflects a policy choice: accept more false positives to reduce false negatives, and systematically favor plaintiffs across theories of unilateral harm, regardless of the strength of the underlying claims.
ICLE’s comments note that this approach mirrors the European Union’s precautionary approach to antitrust, which assumes markets may not self-correct effectively. That approach can yield benefits, but “comes, almost by definition, at the expense of short-term growth.[19] The comments warn that adopting such a framework “is a costly policy stance in those circumstances where it is not clearly warranted by underlying risk and uncertainty,” particularly for a state like California, whose economy depends on innovation and startup growth.
The Supreme Court has long held that antitrust law protects “competition, not competitors.”[20] That distinction determines whether antitrust law serves consumers—here, California citizens broadly—or instead protects individual rivals. A firm that wins customers through lower prices harms competitors but benefits consumers and should not face antitrust liability.[21] By contrast, a firm that uses exclusionary threats to cut off rivals harms the competitive process itself and may warrant intervention. The analytical tools that AB 1776 makes optional—market-power analysis, recoupment, and multi-sided platform balancing—exist to distinguish between these scenarios.
AB 1776 replaces that filtering function with a directive to maximize deterrence. Courts instructed to resolve ambiguity broadly and favor deterrence will impose liability in more cases, including those where the conduct harms a rival but not competition. The result shifts the statute’s focus from protecting competition to protecting competitors.
The empirical case for this expansion is weak. In his August 2024 presentation to the CLRC, ICLE President Geoffrey Manne showed that claims of rising market concentration remain contested.[22] Some studies based on publicly traded firms and broad industry codes suggest increasing concentration, but more granular analyses point in the opposite direction. Gerard Hoberg and Gordon Phillips, accounting for multi-industry firms, find declining average Herfindahl-Hirschman Index (HHI) scores.[23] C. Lanier Benkard and co-authors report that “decreases in concentration over time are broad-based” and that this result “contradicts the prevailing popular opinion.”[24] As Manne emphasized, concentration data alone “says nothing about the amount of competition,” and therefore has no direct normative implication for antitrust policy.[25]
Federal antitrust law has long organized liability around consumer welfare—harm to consumers through higher prices, reduced output, or diminished quality. The Supreme Court described the Sherman Act as “a consumer welfare prescription” in Reiter v. Sonotone Corp.[26] This standard limits liability to conduct that harms consumers, not merely rivals, and provides a principled boundary for antitrust enforcement.
AB 1776 replaces that framework with a broader and less defined set of objectives. The bill declares that California antitrust law protects (1) “free and fair competition” (2) for “all trade participants, including workers and consumers,” and (3) “an environment that is conducive to the preservation of our democratic, political, and social institutions.”[27] These goals are distinct and often in tension, yet the statute does not explain how courts should resolve those conflicts. ICLE scholars warned during the CLRC process that extending protection to “trading partners” as a class—rather than focusing on consumers and the competitive process—risks politicizing enforcement and shielding less-efficient firms at consumers’ expense.[28]
These tensions arise in routine cases. A firm that lowers prices benefits consumers but harms rivals. A firm that integrates complementary features improves quality for users but reduces demand for standalone products. A firm that grows through successful competition may increase concentration, raising political concerns even if its conduct benefits consumers. Under the consumer welfare standard, courts resolve these conflicts by asking whether the net effect on consumers is positive. Under AB 1776’s “all trade participants” standard—combined with a directive to maximize deterrence—no comparable principle applies. Courts must weigh competing interests case by case, producing outcomes that are difficult to predict and hard for firms to anticipate in structuring their conduct.
Federal predatory-pricing doctrine rests on a simple economic insight: below-cost pricing harms consumers only if the firm can later recoup its losses through supracompetitive prices. AB 1776 discards that requirement without replacing it, exposing firms to liability for the very conduct—aggressive price competition—that antitrust law seeks to protect.
Sections V.A and V.B explain the implications of that shift. Together, they show that removing the Brooke Group recoupment screen eliminates a core safeguard against false positives, leaves courts without a coherent test to distinguish harmful from beneficial pricing, and predictably deters procompetitive price cutting. The result is a regime that increases liability risk while weakening the economic foundation of predatory-pricing law, to the detriment of consumers.
In Brooke Group Ltd. v. Brown & Williamson Tobacco Corp.,[29] the Supreme Court held that a predatory-pricing claim requires proof of two elements: (1) pricing below an appropriate measure of cost, and (2) a dangerous probability of recouping the losses through later supracompetitive pricing. The Court explained why both elements matter: “Without a dangerous probability of recoupment, it is highly unlikely that a firm would engage in predatory pricing,” and absent recoupment, “predatory pricing produces lower aggregate prices in the market, and consumer welfare is enhanced.”[30]
The logic of the recoupment requirement is straightforward. A firm that prices below cost without any realistic prospect of later monopoly pricing incurs losses with no path to recovery. Those low prices transfer value to consumers on every unit sold. The firm must eventually raise prices to competitive levels (inviting entry), exit the market, or continue absorbing losses—none of which reflects a strategy that harms consumers. Predatory pricing becomes plausible only when below-cost pricing represents a temporary sacrifice backed by an expectation of future monopoly pricing. Recoupment tests whether that expectation is economically credible.
Then-Judge Stephen Breyer underscored the error-cost stakes in Barry Wright Corp. v. ITT Grinnell Corp.: “the consequence of a mistake here is not simply to force a firm to forego legitimate business activity it wishes to pursue; rather, it is to penalize a procompetitive price cut, perhaps the most desirable activity (from an antitrust perspective) that can take place in a concentrated industry.”[31] ICLE’s comments to the CLRC and subsequent scholarship argue that eliminating the Brooke Group recoupment requirement would move California law toward the European model, which assesses pricing without regard to recoupment—a standard ICLE describes as having “no basis in economic theory or evidence.”[32]
Section 16732 of AB 1776 provides that a plaintiff need not establish “whether a defendant is likely to recoup losses from below-cost pricing” to prove a predatory-pricing violation. The bill offers no substitute analytical test. Instead, courts must evaluate such claims under a general mandate to interpret antitrust law liberally and maximize deterrence.
Under this framework, a firm that prices aggressively to gain market share, match a competitor’s promotion, or expand demand for a new product could face liability without any showing of market power, exclusionary intent, or a realistic prospect of recouping losses through supracompetitive pricing. A plaintiff need only persuade a court that prices fell below some measure of cost and that the conduct harmed a trade participant enough to justify liability.
Firms will respond predictably. The expected cost of a predatory-pricing claim equals the probability of liability multiplied by expected damages—treble damages plus attorneys’ fees. When firms cannot predict how courts will characterize their pricing—because no recoupment screen anchors the analysis and courts must maximize deterrence—they will price more conservatively. The conduct most likely to be deterred is precisely the aggressive price competition that antitrust law has traditionally protected.
The Supreme Court adopted the recoupment requirement in Brooke Group to address this problem. Before that decision, courts often condemned aggressive pricing based on intent or market structure, which led to frequent false positives.[33] Brooke Group added recoupment to reduce the risk of penalizing procompetitive price cuts. AB 1776 effectively returns California to the pre-Brooke Group regime—one federal courts abandoned because it deterred beneficial price competition.
Consumers bear the cost of that shift. When firms avoid aggressive pricing to limit litigation risk, prices rise above competitive levels. In concentrated markets, this means consumers pay more—not because firms exercise monopoly power, but because legal risk discourages price competition. Even familiar examples of sustained low pricing—such as Costco’s $1.50 hot dog and soda combo or its $4.99 rotisserie chicken—become harder to sustain under a regime that treats aggressive pricing as presumptively suspect.
Many of California’s most significant firms—including Google, Meta, Apple, and Amazon—operate multi-sided platforms that fund services on one side of the market with revenue from another. The Supreme Court recognized in Amex that platform conduct must be evaluated across both sides to avoid systematically biased results. AB 1776 prohibits that approach, requiring courts to assess competitive harm one side at a time while disregarding offsetting benefits on the other.
Sections VI.A through VI.D show how this shift departs from the economics of platform competition. Multi-sided markets depend on cross-market interactions, and pricing decisions on one side cannot be understood in isolation. By rejecting cross-market balancing and making multi-sided harm optional, AB 1776 removes the tools needed to distinguish harmful conduct from business models that benefit consumers overall. The result is a framework that risks misidentifying harm, over-deterring efficient cross-subsidization, and exposing a wide range of industries—not just technology—to liability for standard competitive practices.
A multi-sided platform serves multiple groups of users simultaneously, creating value by facilitating interactions among them. The economics of these platforms differ fundamentally from single-sided firms. A credit card network becomes more valuable to merchants as more consumers carry the card, and more valuable to consumers as more merchants accept it. These indirect network effects mean that pricing decisions on one side of the platform shape the size, composition, and behavior of the other.
Pricing therefore operates as an integrated system. A platform that charges merchants higher fees may use that revenue to fund consumer rewards, fraud protection, or broader acceptance, attracting more users and increasing transaction volume for merchants. The causal relationship runs in both directions. Evaluating a price increase on one side of the platform without accounting for effects on the other side produces an incomplete—and often misleading—assessment. That does not mean harms on one side can never outweigh benefits on the other. It means cross-market effects are central to competition among multi-sided platforms and cannot be ignored without risking harm to consumers.
Many of California’s most significant firms operate multi-sided platforms. Google funds free search and maps through advertising. Meta offers free social networking supported by advertising revenue. Apple’s App Store connects developers and users through a commission-based model. Amazon’s marketplace links third-party sellers with consumers while competing as a retailer. In each case, pricing decisions on one side of the platform are analytically inseparable from their effects on the other.
The Supreme Court addressed the implications of platform economics for antitrust analysis in Ohio v. American Express Co.[34] The Court held that, because Amex operates a two-sided transaction platform, evidence of a price increase on the merchant side “cannot, by itself, demonstrate an anticompetitive exercise of market power.”[35] It emphasized that “the two-sided market for credit-card transactions should be analyzed as a whole,” warning that focusing on only one side “tends to distort the competition that actually exists” and risks “mistaken inferences” that could chill legitimate competition.”[36] The Court therefore requires plaintiffs in two-sided platform cases to show anticompetitive effects across both sides of the market before the burden shifts to defendants to offer procompetitive justifications.
This framework reflects the underlying economics of multi-sided platforms. Costs imposed on one side and benefits delivered on the other are linked through the platform’s business model. Evaluating only costs or only benefits produces a systematically biased assessment of welfare effects. Herbert Hovenkamp, a leading antitrust scholar at the University of Pennsylvania Law School, explains that market power on a multi-sided platform cannot be inferred from conditions on a single side, because apparent price increases or market power may be offset by services or subsidies on the other.[37]
ICLE applies this logic directly to the CLRC’s proposed rejection of Amex. Evidence of a price effect on one side of a two-sided platform may reflect neutral, procompetitive, or anticompetitive conduct. Distinguishing among those possibilities requires examining both sides of the platform, rather than isolating one dimension of the market.[38]
Two provisions of AB 1776 displace the Amex framework. Section 16731(b) requires courts to evaluate anticompetitive effects and procompetitive justifications “within the same relevant market,” prohibiting cross-market balancing. Section 16732(f) further provides that plaintiffs need not show harm on more than one side of a multi-sided platform, or that harm on one side outweighs benefits on another.
Read together, these provisions require courts to assess platform conduct one side at a time and to disregard offsetting benefits on the other side. A platform that raises advertising prices to fund free consumer services would be judged solely on advertiser-side costs, without credit for consumer benefits. A payment network that charges merchants higher fees while providing cardholders with superior fraud protection and rewards could face liability based only on merchant-side effects.
The bill offers no economic justification for this approach. The CLRC dismisses Amex, asserting:
[Amex] created a confusing precedent as to the type and amount of evidence needed to show harm in cases involving two sided platforms. This case also used assumptions about the interconnectedness of the two sides that may not translate to market realities in other circumstances, and could allow firms to escape antitrust liability for causing harm on one side of a platform and masking it with benefits on the other side.[39]
ICLE scholars responded:
As in the Amex case itself, such an approach would confer benefits on certain platform-business users (in Amex, retailers) at the direct expense of consumers (in Amex, literal consumers of retail goods purchased by credit card). Adopting such an approach in California—whose economy is significantly dependent on multisided digital-platform firms, including both incumbents and startups—would imperil the state’s economic prospects and exacerbate the incentives for such firms to take jobs, investments, and tax dollars elsewhere.[40]
As in the Amex case itself, such an approach would confer benefits on certain platform-business users (in Amex, retailers) at the direct expense of consumers (in Amex, literal consumers of retail goods purchased by credit card).
Adopting such an approach in California—whose economy is significantly dependent on multisided digital-platform firms, including both incumbents and startups—would imperil the state’s economic prospects and exacerbate the incentives for such firms to take jobs, investments, and tax dollars elsewhere.[40]
By prohibiting cross-market balancing, AB 1776 adopts an analytical framework that will misidentify harm in multi-sided markets. A rule that counts costs on one side without crediting benefits on the other will find liability even when overall consumer welfare improves. Courts applying this rule cannot distinguish between conduct that harms consumers and conduct that harms only competitors. Any adverse effect on one group of users becomes actionable, regardless of offsetting benefits to others.
The predictable result is liability for business models the Supreme Court declined to condemn in Amex—models that often benefit consumers overall despite imposing costs on certain commercial users. Rather than address that distinction, AB 1776 forbids courts from considering it.
The same error extends to the bill’s treatment of vertical restraints. ICLE’s comments to the CLRC synthesize a large empirical literature showing that vertical integration and related practices produce predominantly procompetitive or neutral effects.[41] As former FTC Bureau of Economics Director Francine Lafontaine explains, when firms adopt vertical restraints, they typically improve product quality and service, benefiting consumers as well as producers.[42] Even more skeptical reviews acknowledge that few studies identify vertical practices that likely harm competition.[43] AB 1776’s decision to subject these practices to heightened scrutiny under a maximize-deterrence mandate conflicts with that empirical record.
The prohibition on cross-market balancing exposes any firm that subsidizes one group of users with revenue from another to antitrust liability. While technology platforms provide the most visible examples, this model appears across many industries.
Google offers free search, maps, and email to consumers while charging advertisers for access to those users. Its advertising prices cannot be evaluated in isolation from the consumer services they fund. Under AB 1776, a plaintiff could challenge those prices based solely on advertiser costs, with no credit for the consumer benefits they support.
The same structure applies to Meta, Apple, and Amazon. Meta funds free social networking through advertising. Apple charges App Store commissions while providing developers and users with software review, device integration, and payment infrastructure. Amazon charges third-party sellers marketplace fees while operating fulfillment and consumer-trust systems. A one-sided liability framework counts costs imposed on one group while ignoring the benefits delivered to another, even when both arise from the same transaction.
Credit card networks operate the same two-sided model the Supreme Court analyzed in Amex. Visa and Mastercard charge merchants interchange fees while providing consumers with fraud protection, rewards, and payment convenience. Under AB 1776, a challenge to those fees would proceed without accounting for those consumer benefits. Merchants may gain; consumers lose.
This structure extends beyond technology and finance. Newspapers and broadcasters sell advertising to fund content that consumers receive without direct payment. Grocery retailers use loyalty programs and data revenues to support lower prices. Health care systems cross-subsidize services, using profitable lines to fund emergency or community care. These models depend on integrated pricing across different user groups.
AB 1776’s one-sided analysis will identify harm where a complete assessment shows net consumer benefit. Courts will evaluate pricing and revenue models without the tools to measure their full effects. Firms that compete by delivering value across multiple sides of a market face the greatest exposure.
AB 1776 extends antitrust liability to monopsonization—buyer-side market power—and identifies workers’ freedom to choose employment as a protected competition interest. Federal law offers limited precedent for applying single-firm antitrust doctrine to labor markets, and the economics literature does not provide a settled framework for defining relevant labor markets or measuring employer wage-setting power.
Sections VII.A through VII.C show how the bill codifies this uncertainty. AB 1776 applies the same expanded liability structure used for product markets—no market-power threshold, liberal interpretation, and maximum deterrence—to an area where both doctrine and measurement remain underdeveloped. The result is a regime that lacks clear standards for identifying harm, extends liability across a wide range of employers, and encourages firms to adjust hiring and compensation practices to manage legal risk rather than compete aggressively for talent.
As noted above, AB 1776 extends the Cartwright Act to monopsonization and identifies workers’ freedom to choose employment as a protected competition interest.[44] Federal law offers little guidance on applying Section 2 to unilateral labor-market conduct by a single employer. The analytical tools for identifying monopsony power remain far less developed than those used to assess monopoly power in product markets.
Federal antitrust enforcement in labor markets has focused on horizontal agreements among employers—no-poach agreements and wage-fixing arrangements—rather than single-employer monopsonization. AB 1776 expands liability into an area where both the doctrine and the underlying economics remain unsettled.
ICLE scholars Geoffrey Manne, Brian Albrecht, and Dirk Auer argue that the economics literature lacks a clear consensus on how to assess labor-market power.[45] They note that standard tools used in product markets—geographic market definition, substitution analysis, and price-cost measurement—do not translate cleanly to labor markets. ICLE reiterates these concerns in its CLRC comments and subsequent analysis of the California proposals.[46]
In product markets, defining a relevant market turns on substitution. The “hypothetical monopolist” test asks whether a firm controlling all supply could profitably raise prices by 5%–10%. That inquiry maps onto observable behavior: where consumers go when prices rise. Economists can answer it using price and quantity data to estimate demand elasticities and cross-elasticities across products.
In labor markets, the parallel question is whether a single employer could profitably reduce wages. But the substitution analysis is far less tractable. Workers do not switch among employers as easily as consumers switch among products. Non-wage attributes—working conditions, career opportunities, organizational culture, job security, and proximity to family or professional networks—differentiate jobs in ways that are difficult to measure.
Individual circumstances further complicate the analysis. A software engineer in San Francisco may not view a position in Austin as a substitute, regardless of higher pay, due to family ties or housing investments—or she may, if the job offers remote work.[47] Whether two positions fall within the same labor market depends on worker-specific preferences and mobility constraints that aggregate data often cannot capture.
Geographic market definition has become especially uncertain with the rise of remote work. Pre-pandemic studies relied on commuting patterns. Today, many California workers are employed by firms based in other states, and many firms recruit across multiple regions. Research finds that employees may accept 5%–25% lower compensation for remote or hybrid work.[48] At the same time, California’s employment regulations—including expense-reimbursement requirements, daily overtime rules, and numerous local minimum-wage ordinances—have led some out-of-state employers to exclude California residents from remote hiring.[49] These factors make labor-market boundaries harder to define and monopsony power more difficult to measure.
AB 1776 applies the same expanded liability structure to monopsonization that it applies to monopolization: no market-power threshold, no required methodology for measuring labor-market power, a mandate for liberal interpretation, and an instruction to maximize deterrence. Courts applying this framework will lack settled tools to define relevant labor markets, measure wage-setting power, or distinguish aggressive competition for workers from anticompetitive conduct.
The exposure extends beyond technology firms. Any employer that hires a large share of workers with specialized skills in a given area—a hospital system employing specialized nurses, a logistics firm employing warehouse workers, or a media company employing niche editorial staff—could face monopsonization claims. Without a market-power threshold, firms cannot determine in advance whether their size or hiring patterns create liability risk. The open-ended standard, combined with a deterrence mandate, invites courts to resolve uncertainty in plaintiffs’ favor.
Firms will respond by adjusting employment practices. To reduce litigation risk, employers may avoid compensation strategies or hiring practices that could be characterized as evidence of wage-setting power, even when those practices reflect competition for talent. They may limit long-term commitments or rely more heavily on outsourcing to reduce direct labor-market exposure. These responses may narrow opportunities and reduce flexibility for the workers the statute aims to protect.
The consumer welfare standard gives courts a single, measurable question: whether the challenged conduct leaves consumers better off or worse. That inquiry distinguishes conduct antitrust law should reach from conduct it should not and allows courts to screen out claims that allege harm to competitors but not to competition.
AB 1776 replaces that standard with a multi-objective declaration protecting “free and fair competition” for “all trade participants” and the preservation of “democratic, political, and social institutions,” without any mechanism for resolving the conflicts and tradeoffs those objectives create. Sections VIII.A and VIII.B show that these conflicts are inevitable and that removing a clear limiting principle leaves courts to balance competing interests case by case. An “all-and-sundry welfare” approach does not guide that analysis—it invites inconsistent outcomes and risks undermining the consumer benefits antitrust law is designed to protect.
The consumer welfare standard does not privilege consumers over workers or other market participants as a normative matter. It provides courts with a coherent, measurable limit on what conduct antitrust law should reach. As the Supreme Court stated in Reiter v. Sonotone Corp., the Sherman Act is “a consumer welfare prescription.”[50] The Court reinforced in Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc. that plaintiffs must show harm to competition, not merely harm to themselves as competitors.[51] Together, these principles establish that antitrust law does not shield market participants from losing to superior rivals.
The standard shapes how courts analyze liability. When a plaintiff challenges pricing, exclusive dealing, or product integration, a court asks whether the conduct raised prices, reduced output or quality, or limited consumer choice. If not—if the conduct harmed a competitor while benefiting consumers—the court can dismiss the claim without full merits analysis. This screening function reduces litigation costs and filters out claims that, even if factually accurate, do not describe antitrust violations.
AB 1776 replaces the consumer welfare standard with a declaration that California antitrust law protects “free and fair competition” for “all trade participants, including workers and consumers,” and “an environment that is conducive to the preservation of our democratic, political, and social institutions.”
This formulation does not function as a workable legal standard. It combines three distinct objectives—protecting competition, protecting trade participants as a class, and preserving political institutions—without providing a method to resolve conflicts among them. Those conflicts arise routinely. A firm that lowers prices benefits consumers but harms rivals; both are “trade participants.” A firm that vertically integrates to secure inputs may benefit workers and customers through stability while disadvantaging independent suppliers; all fall within the statute’s scope. The consumer welfare standard resolves these tensions by asking whether consumers are better or worse off. AB 1776 provides no comparable principle. Combined with a mandate to maximize deterrence, the statute leaves courts to prioritize among competing interests case by case, producing inconsistent and unpredictable outcomes.
Murat Mungan and John Yun identify a further consequence of overbroad liability standards. When antitrust violations no longer clearly signal consumer harm, the reputational penalty that reinforces legal sanctions weakens.[52] Firms found to have harmed consumers through anticompetitive conduct face social and market consequences that amplify deterrence. Firms found liable for conduct that the public views as competitive or beneficial do not. By extending liability to conduct whose anticompetitive character is not self-evident, AB 1776 risks diluting this reputational mechanism and, with it, the overall deterrent effect of antitrust law.
AB 1776 systematically lowers the thresholds for antitrust liability below those federal courts have developed over decades. It makes core screening tools optional—market power, recoupment, and cross-market balancing—thereby expanding liability across a wide range of conduct that federal law would not reach.
Sections IX.A through IX.C show how these lower standards extend beyond California. Because federal law does not preempt state antitrust law, and firms operating in national or digital markets cannot realistically maintain California-specific business practices, the bill’s standards would shape conduct nationwide. California enforcement actions and settlements would set de facto national rules, while imposing compliance costs on firms across the country. That extraterritorial effect raises unresolved dormant Commerce Clause concerns about whether the burdens on interstate commerce are proportionate to California’s asserted interests.
AB 1776 establishes lower thresholds for antitrust liability than federal law. Section 16732 renders optional 10 factors that federal courts treat as requirements—not out of confusion about antitrust goals, but because experience shows each screens out claims that would condemn procompetitive conduct.
The market-power threshold requires a defendant to possess monopoly power—often inferred from a market share above 50% in a properly defined relevant market—before unilateral conduct can qualify as monopolization.[53] This requirement reflects a basic constraint: a firm without market power cannot harm the competitive process through unilateral action. Without the ability to control prices or exclude rivals, practices such as aggressive pricing, exclusive dealing, and refusals to deal remain disciplined by competition. AB 1776 removes this threshold, exposing firms of any size or position to monopolization claims.
Other federal screens serve similar functions. Recoupment distinguishes temporary price competition from predation; eliminating it collapses that distinction. The Amex framework requires cross-market analysis for two-sided platforms; prohibiting it produces systematically incomplete welfare assessments. Each screen targets a known source of error in antitrust adjudication.
The as-efficient-competitor test asks whether the challenged conduct would exclude a rival that matches the defendant’s efficiency, or only less efficient firms. That comparison identifies conduct that distorts competition rather than reflects it. AB 1776 makes this test optional, allowing liability where conduct harms only rivals that cannot match the defendant’s performance. Under this approach, even a low-price offering—such as Costco’s $1.50 hot dog and soda combo—could be characterized as unlawful predatory pricing by less efficient competitors.
These lower liability thresholds have national consequences. The California attorney general and private plaintiffs may bring antitrust cases in state courts against firms that operate nationwide. When those cases succeed under AB 1776’s standards—without market-power analysis, recoupment, or multi-sided market balancing—they produce precedents and settlements that shape firms’ conduct beyond California.
A firm that settles a predatory-pricing claim under AB 1776 will not maintain one pricing strategy in California and another elsewhere. For many businesses, especially digital platforms, geographic differentiation is costly or infeasible. Settlement terms adopted to resolve a California claim will apply across the firm’s operations. California courts, applying lower standards than federal courts, would effectively set national policy for the challenged conduct.
Federal antitrust law does not preempt state antitrust law under Parker v. Brown.[54] But when firms cannot segment their practices by state, California’s rules will govern their behavior nationwide. By enacting AB 1776, the California Legislature would influence not only in-state commerce but commercial conduct across the United States.
AB 1776’s extraterritorial reach raises concerns under the dormant Commerce Clause, which bars states from imposing burdens on interstate commerce that are “clearly excessive in relation to the putative local benefits.”[55] Courts have not resolved whether a state antitrust law that drives nationwide changes in firm behavior—particularly through digital markets—satisfies this standard. Shira Liu concludes that the constitutional question remains unsettled.[56]
What is clear is that AB 1776’s lower liability standards would impose compliance costs on firms across the country that may exceed any in-state benefits. A firm that adjusts its national pricing strategy to avoid California liability under a no-recoupment standard changes behavior for consumers in all 50 states based solely on California law. Whether such nationwide effects are proportionate to California’s regulatory interests remains unanswered by courts and unaddressed by the Legislature.
AB 1776’s combination of broad liability, low analytical thresholds, and treble damages creates strong incentives for strategic litigation—suits aimed at extracting settlements through defense costs rather than remedying anticompetitive harm. By weakening screening mechanisms and prioritizing deterrence, the bill increases both the volume and leverage of weak claims, shifting competitive disputes from the market to the courtroom.
AB 1776’s combination of low pleading standards, broad liability categories, no market-power threshold, and treble damages creates strong incentives for strategic antitrust litigation—claims filed not to remedy anticompetitive harm, but to leverage the cost of defense into favorable settlements.
When a statute presumes liability across a wide range of conduct and removes the analytical screens that would allow early dismissal, defending even a weak claim becomes costly. Antitrust litigation routinely requires extensive discovery, expert analysis, and depositions, often costing millions of dollars. A defendant that expects to prevail on the merits may still choose to settle because the cost of defense exceeds the cost of settlement. The resulting terms—pricing concessions, supply commitments, data-sharing obligations, or limits on business practices—reflect litigation pressure, not a judicial finding of anticompetitive conduct.
AB 1776’s directive to interpret antitrust law liberally and maximize deterrence intensifies these incentives. Courts operating under that mandate will be less likely to dismiss claims at the pleading stage, more likely to treat ambiguous evidence as supporting liability, and more inclined to resolve legal uncertainty against defendants. Each of these effects increases the expected value of weak claims—and, in turn, the number of such claims filed.
In practice, AB 1776’s private right of action will primarily benefit the plaintiffs’ bar and rivals seeking advantages they cannot achieve through competition. A less-efficient firm competing with a larger rival may be unable to match price or quality. Under AB 1776, it can instead threaten litigation—alleging predatory pricing (no recoupment required), cross-market harm in a platform business (no balancing permitted), or monopsony in labor markets (no market-power threshold). Each theory forces the defendant to incur substantial defense costs and creates settlement pressure regardless of the claim’s merits.
This form of litigation does not protect consumers. It protects the litigating rival at consumers’ expense. Settlements may require defendants to raise prices to avoid further predation claims, reduce the quality or scope of platform services to limit cross-market effects, or constrain employment practices to mitigate monopsony allegations. Each outcome benefits the rival’s competitive position while harming consumers or workers.
AB 1776 does not operate in isolation. Its departures from established doctrine are best assessed against three benchmarks: federal Section 2 law, which the bill systematically weakens across key elements; New York’s Twenty-First Century Antitrust Act, which has repeatedly stalled under scrutiny; and the European Union’s Digital Markets Act, whose early enforcement record illustrates the practical effects of a similar regulatory approach.
Sections XI.A through XI.C show a consistent pattern. Where AB 1776 diverges most sharply from established frameworks—by lowering liability thresholds, rejecting analytical safeguards, and prioritizing deterrence—it aligns with approaches that have either failed to gain legislative traction or produced mixed results in practice.
Table 2 compares AB 1776 with federal Section 2 doctrine across the doctrinal elements most relevant to the bill’s effects.
AB 1776 does not simply fill the Cartwright Act’s historical gap on single-firm conduct. It adopts more plaintiff-favorable standards on each contested element, allowing the California attorney general and private plaintiffs to prevail on claims that federal courts would reject.
New York State has considered extending its antitrust law to single-firm conduct since at least the 2019–2020 legislative session, when state Sen. Michael Gianaris (D-Queens) introduced the Twenty-First Century Antitrust Act as S. 8700-A.[57] He has introduced similar versions in each subsequent session.[58] The most recent version, S. 335, was introduced in January 2025.
The proposal would amend the Donnelly Act—New York’s primary antitrust statute governing restraints of trade—in two key ways. First, it would prohibit monopolization, attempted monopolization, and conspiracy to monopolize, treating unilateral conduct similarly to agreements. Second, it would establish an “abuse of dominance” standard for firms with a dominant position in product or labor markets. That standard would allow courts to infer dominance from direct or indirect evidence, including the ability to impose supracompetitive prices or subcompetitive wages. The bill would thus lower the legal thresholds required to establish antitrust violations.
In October 2023, Sen. Gianaris presented the proposal to the CLRC’s Single-Firm Conduct Working Group, which rejected it. The group concluded that terms such as “dominant position” and “abuse of dominance” were too vague to provide workable standards.[59] Although the New York Senate has passed versions of the bill in each session since its introduction, the Assembly has not taken it up.[60] That repeated inaction reflects sustained resistance to an analytically weak framework.
AB 1776 avoids the “abuse of dominance” label but reaches a similar result through different means: lower thresholds, fewer evidentiary requirements, and a mandate to maximize deterrence. The New York experience suggests a broader pattern. Proposals that closely track federal doctrine tend to survive scrutiny; those that depart most sharply from it tend to stall.
Although AB 1776 is not modeled on the European Union’s Digital Markets Act (DMA),[61] the two share underlying assumptions that make the DMA’s experience relevant. The CLRC declined to adopt a formal European “abuse of dominance” standard,[62] but AB 1776 incorporates similar logic through a different route. It mirrors the European approach in three respects: it removes market-power thresholds as a prerequisite for liability, prohibits cross-market balancing, and departs from the effects-based, consumer-welfare framework that has long distinguished U.S. antitrust law. ICLE has cautioned against this “Europeanization” of California law, urging continued alignment with consumer welfare, effects-based analysis, and error-cost discipline.[63]
The DMA provides the most prominent example of this approach in practice. In force since March 2024, it authorizes the European Commission to designate firms as “gatekeepers” based on size, user base, and the provision of “core platform services.”[64] The Commission has designated firms such as Alphabet, Amazon, Apple, ByteDance, Meta, Microsoft, and Booking.com, and imposed obligations related to interoperability, data portability, and self-preferencing.[65]
These obligations apply without requiring case-specific proof of market power, consumer harm, or the absence of efficiencies.[66] The DMA’s self-preferencing rules rest on a contested premise—that integration by a platform is inherently suspect—rather than recognizing that integration often generates efficiencies.[67] AB 1776 adopts a similar shortcut by making market power, recoupment, cross-market balancing, and as-efficient-competitor analysis optional. Both regimes relax traditional screens and presume harm without requiring case-specific evidence.
Early evidence from the DMA’s implementation illustrates the risks of that approach. A 2025 survey by the European Centre for International Political Economy (ECIPE), a Brussels-based policy institute, found that while most consumers support intervention in digital markets, 39% report needing more steps to complete tasks that were previously simple, and roughly one-third report more fragmented and confusing digital experiences.[68] A separate analysis by Carmelo Cennamo and co-authors estimates that DMA provisions could reduce revenues across EU service sectors by up to €114 billion due to lost efficiencies.[69] Sector-specific losses range from €4.4 billion to €59 billion in retail and €14 billion to €21 billion in accommodation. Neither study finds evidence of a meaningful shift in market structure toward alternative providers. Interoperability and unbundling mandates have reduced product quality, while the competitive landscape remains largely unchanged.
These outcomes reflect a broader concern. As ICLE has documented, the stated goals of digital competition policy—greater “contestability” and “fairness”—often diverge from their practical effect: redistributing value from successful firms to less-efficient rivals.[70] The mechanisms differ—conduct mandates in the DMA, expanded tort liability in AB 1776—but the underlying approach is similar: skepticism of market-power thresholds, rejection of efficiency balancing, and a willingness to second-guess business decisions without proof of consumer harm. The DMA’s record offers the best available evidence of how that approach operates in practice. California legislators should weigh that experience before adopting a framework that has degraded digital services for consumers without achieving its structural goals.
ICLE’s comments also point to broader macroeconomic patterns.[71] As the comments note, ECIPE research shows that U.S. GDP per capita exceeded the EU’s by 47% in 2010 and by 82% in 2021, a widening gap under differing regulatory approaches. Even studies often cited to support concerns about concentration cut the other way on closer review. For example, David Autor and his co-authors observe:
An alternative perspective on the rise of [large firms and increased concentration] is that they reflect a diminution of competition, due to weaker U.S. antitrust enforcement. Our findings on the similarity of trends in the United States and Europe, where antitrust authorities have acted more aggressively on large firms, combined with the fact that the concentrating sectors appear to be growing more productive and innovative, suggests that this is unlikely to be the primary explanation, although it may be important in some industries.[72]
AB 1776 raises potential concerns for criminal enforcement under the Cartwright Act. A core principle of American law is the presumption of innocence: prosecutors bear the burden of proving guilt beyond a reasonable doubt. As William Blackstone observed, “the law holds that it is better that ten guilty persons escape than that one innocent suffer.”[73] This principle reflects a broader tradeoff. Efforts to deter harmful conduct must be balanced against the risk of deterring lawful conduct, especially in areas of uncertainty.
Several doctrines operationalize that balance. The void-for-vagueness doctrine requires that criminal laws give clear notice of prohibited conduct. The rule of lenity requires courts to interpret ambiguous penal statutes narrowly in favor of defendants. Together, these constraints limit the scope of criminal liability and protect against overreach.
AB 1776’s interpretive mandates complicate this framework. Section 16733 directs courts to construe antitrust law liberally and to remain “mindful” that California favors “maximizing” deterrence. That instruction does not distinguish between civil and criminal cases. The Cartwright Act, however, carries criminal penalties—up to three years in county jail for individuals and fines up to $6 million for corporations[74] —and it is unclear whether those penalties extend to the new single-firm conduct provisions. This issue is not theoretical. In 2024, the California attorney general announced plans to resume criminal antitrust prosecutions for the first time in 25 years.[75]
In civil cases, broad construction and deterrence mandates may be aggressive but permissible. In criminal cases, they conflict with foundational constraints: proof beyond a reasonable doubt, the rule of lenity, and the requirement of fair notice. Clear notice is easier to establish when conduct is categorically unlawful. A cartel agreement among gas stations to fix prices presents little ambiguity. By contrast, whether below-cost pricing—such as Costco’s $1.50 hot dog and soda combo—violates antitrust law depends on a complex, fact-specific inquiry. Single-firm conduct typically falls under the rule of reason, not per se illegality, making criminal enforcement more difficult to justify.
Federal practice reflects this distinction. Criminal enforcement under the Sherman Act has focused on Section 1 collusion—price-fixing and bid-rigging—where conduct is per se unlawful.[76] Proposals to extend criminal liability to Section 2 monopolization have faced resistance, based on concerns about fair notice and the difficulty of proving specific intent to harm competition.[77]
AB 1776 heightens these concerns. Its instruction to construe the statute broadly operates in tension with lenity, while its open-ended structure—making federal standards optional without specifying what suffices for liability—raises vagueness issues. A prosecutor could argue that “liberal interpretation” defines the elements of the offense, with the reasonable-doubt standard applying only to proof of those elements. That approach would expand criminal liability in practice while formally preserving the burden of proof.
The result is uncertainty about what conduct is criminal. If California courts depart from federal antitrust doctrine without clear replacement standards, firms and individuals may lack fair notice of the law’s boundaries. AB 1776 does not address this tension, and it remains unclear whether its deterrence mandate can be reconciled with the constitutional limits that govern criminal punishment.
Assembly Bill 1776 adopts the California Law Revision Commission’s recommendations with minimal change. Those recommendations abandon the error-cost framework and the consumer welfare standard that have disciplined antitrust enforcement for decades. The bill makes optional each major evidentiary screen federal courts use to distinguish procompetitive from anticompetitive conduct—market-power thresholds, recoupment in predatory pricing, cross-market balancing for multi-sided platforms, the as-efficient-competitor test, and quantitative evidence of harm. In their place, it directs courts to maximize deterrence and construe liability broadly, a combination that predictably increases costly false positives. Markets can partially self-correct false negatives as entry erodes supracompetitive profits. False positives do not self-correct, because erroneous condemnation deters beneficial conduct going forward.
The consequences follow directly from these choices. Eliminating recoupment returns California predatory-pricing law to the pre-Brooke Group regime that federal courts rejected for overdeterring price competition. Prohibiting cross-market balancing mandates one-sided analysis of multi-sided platforms—counting costs without crediting offsetting benefits—that will mismeasure welfare for firms such as Google, Meta, Apple, Amazon, and the payment networks at issue in Amex. Extending liability to labor-market monopsony codifies an unsettled theory under an open-ended, deterrence-maximizing standard. Replacing the consumer welfare standard with a multi-objective declaration protecting “all trade participants” and broader social interests deprives courts of a coherent limiting principle.
The bill also creates strong incentives for strategic litigation. Broad liability, low analytical thresholds, treble damages, and a private right of action invite suits aimed at extracting settlements through defense costs. The likely results—higher prices, reduced service quality, or constrained business practices—benefit litigating rivals while harming consumers.
AB 1776’s reach extends beyond California. Firms serving California through digital channels cannot feasibly maintain California-specific business practices. As a result, the bill’s lower standards would set de facto national rules. That extraterritorial effect raises unresolved dormant Commerce Clause questions the Legislature has not addressed.
The interaction of AB 1776’s interpretive mandates with the Cartwright Act’s criminal penalties presents additional due-process concerns. An instruction to construe liability broadly and maximize deterrence conflicts with the rule of lenity and heightens vagueness risks, particularly if applied to single-firm conduct. With the California attorney general signaling a return to criminal antitrust enforcement, these concerns are immediate.
Comparative evidence reinforces these risks. The European Union’s Digital Markets Act—built on similar premises—has, after roughly two years, reduced the seamlessness of digital services for many users without producing measurable increases in competition. New York’s repeated failure to enact a structurally similar bill suggests that proposals departing sharply from federal doctrine face sustained scrutiny.
If the Legislature proceeds with single-firm conduct reforms, it should adopt the following changes:
AB 1776 is not a marginal adjustment. It represents a comprehensive departure from established antitrust doctrine. Any reform should preserve the analytical tools that distinguish harmful conduct from competition on the merits and should align enforcement with both economic evidence and constitutional constraints.
[1] Cal. Assemb. B. 1776, 2025–2026 Reg. Sess. (Cal. 2025).
[2] Cal. Bus. & Prof. Code §§ 16700–16770.
[3] State enforcers may sue under all three statutes. Private plaintiffs may sue under the Sherman and Clayton Acts, see 15 U.S.C. §§ 15, 26, but not under the FTC Act, which only the Federal Trade Commission may enforce. State attorneys general may bring parens patriae actions for Sherman Act violations under § 4C of the Clayton Act, 15 U.S.C. § 15c. Although the Clayton Act often targets concerted conduct—most notably merger review under § 7—it also reaches certain unilateral conduct. The Robinson-Patman Act, codified as an amendment to the Clayton Act, 15 U.S.C. § 13, governs some forms of unilateral price discrimination. Whatever one thinks of Robinson-Patman as a policy matter, it remains relevant to assessing whether any gap exists in the law governing single-firm conduct that A.B. 1776 would fill.
[4] 15 U.S.C. § 2.
[5] See Note, Antitrust Federalism, Preemption, and Judge-Made Law, 133 Harv. L. Rev. 2557 (2020).
[6] Thomas Greene, Robert C. Fellmeth, Thomas A. Papageorge & Kathleen J. Tuttle, A Century of Government Antitrust Enforcement Under the Cartwright Act, 17 Antitrust & Unfair Comp. L. Sec. St. B. Cal. 173 (2008).
[7] Cianci v. Superior Court, 40 Cal. 3d 903, 920 (Cal. 1985).
[8] See Asahi Kasei Pharma Corp. v. Cotherix, Inc., 204 Cal. App. 4th 1, 8 (2012); Flagship Theaters of Palm Desert LLC v. Century Theaters, Inc., 198 Cal. App. 4th 1366, 1386 (2011); Freehand Corp. v. Adobe Sys. Inc., 852 F. Supp. 2d 1171, 1185 (N.D. Cal. 2012).
[9] See Cal. Law Revision Comm’n, Tentative Recommendation: Antitrust Law: Single Firm Conduct (Dec. 2025), https://clrc.ca.gov/pub/Misc-Report/TR-B750.pdf (hereinafter CLRC Recommendation), at 4 (“At a minimum, adopting state law would allow such matters to proceed under state law, even if litigated in federal court.”), 7 (“Vertical integration in major California industries, along with the scale of certain digital platforms, presents competitive challenges not anticipated by the original antitrust drafters.”), 15 n.122 (“Since enactment of the Class Action Fairness Act of 2005, most consumer Cartwright Act class actions have proceeded in federal court. This shift has produced several side effects, including a tendency among federal judges to conflate Cartwright claims with federal antitrust claims, treating them as effectively identical even when they are not.”).
[10] CLRC Recommendation, supra note 9, at 13.
[11] CLRC Recommendation, supra note 9, at 15.
[12] Geoffrey A. Manne, Dirk Auer, Brian Albrecht & Lazar Radic, Int’l Ctr. for L. & Econ., Comments on Memorandum 2025-21 on the Draft Language for Single-Firm Conduct Provision (May 23, 2025), https://laweconcenter.org/wp-content/uploads/2025/05/CLRC-Comments.pdf (hereinafter ICLE 2025 CLRC Comments).
[13] CLRC Recommendation, supra note 9.
[14] Frank H. Easterbrook, The Limits of Antitrust, 63 Tex. L. Rev. 1, 2–4 (1984).
[15] False negatives may occasionally produce precedent, but they more often arise without enforcement and thus without comparable precedential effect.
[16] ICLE 2025 CLRC Comments, supra note 12; Geoffrey A. Manne & Dirk Auer, Int’l Ctr. for L. & Econ., Against the “Europeanization” of California’s Antitrust Law: Comments of the International Center for Law & Economics on the Single-Firm Conduct Expert Report (2024), https://laweconcenter.org/wp-content/uploads/2024/05/Comments-of-the-International-Center-for-Law-California-Law-Revision-Commission-Single-Firm-Conduct.pdf; Geoffrey A. Manne, Dirk Auer & Brian Albrecht, California’s Ill-Advised Turn Toward Europeanized Theories of Harm for Single-Firm Conduct, CPI Antitrust Chron. (June 2025), https://www.pymnts.com/cpi-posts/californias-ill-advised-turn-toward-europeanized-theories-of-harm-for-single-firm-conduct.
[17] CLRC Recommendation, supra note 9, at 19 (“Indicators of anticompetitive intent vary by circumstance, and rigid rules that demand specific fact patterns can unduly restrict enforcement.”).
[18] CLRC Recommendation, supra note 9, at 9, 11.
[19] ICLE 2024 CLRC Comments, supra note 16.
[20] Brown Shoe Co. v. United States, 370 U.S. 294, 320 (1962) (emphasis in original).
[21] Price reductions—always valuable to consumers—often harm less efficient competitors, which may be unable to match the better deal.
[22] Geoffrey A. Manne, Int’l Ctr. for L. & Econ., Understanding Concentration and Competition: Implications for California’s Antitrust Policy, Presentation Before the Cal. Law Revision Comm’n (Aug. 15, 2024), https://laweconcenter.org/wp-content/uploads/2025/05/Manne-CLRC-presentation-2024-08-15.pdf.
[23] Gerard Hoberg & Gordon M. Phillips, Scope, Scale and Concentration: The 21st Century Firm (Nat’l Bureau of Econ. Rsch., Working Paper No. 30672, 2022), https://www.nber.org/system/files/working_papers/w30672/w30672.pdf.
[24] C. Lanier Benkard, Ali Yurukoglu & Anthony Lee Zhang, Concentration in Product Markets (Nat’l Bureau of Econ. Rsch., Working Paper No. 28745, 2021), https://www.nber.org/system/files/working_papers/w28745/w28745.pdf.
[25] Manne, supra note 22.
[26] Reiter v. Sonotone Corp., 442 U.S. 330, 343 (1979).
[27] Cal. Assemb. B. 1776, § 16730 (Cal. 2025).
[28] ICLE 2025 CLRC Comments, supra note 12; Manne, Auer & Albrecht, supra note 16.
[29] Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209 (1993).
[30] Id. at 224.
[31] Barry Wright Corp. v. ITT Grinnell Corp., 724 F.2d 227, 234 (1st Cir. 1983).
[32] ICLE 2025 CLRC Comments, supra note 12; Manne, Auer & Albrecht, supra note 16.
[33] U.S. Dep’t of Justice, Competition and Monopoly: Single-Firm Conduct Under Section 2 of the Sherman Act 51 (2008), https://www.justice.gov/sites/default/files/atr/legacy/2009/05/11/236681.pdf.
[34] Ohio v. Am. Express Co., 585 U.S. 529 (2018).
[35] Id. at 2278.
[36] Id. at 2278, 2287.
[37] Herbert Hovenkamp, Antitrust and Platform Monopoly, 130 Yale L.J. 1952 (2021).
[38] ICLE 2025 CLRC Comments, supra note 12; Manne, Auer & Albrecht, supra note 16.
[39] CLRC Recommendation, supra note 9, at 20.
[40] ICLE 2025 CLRC Comments, supra note 12.
[41] ICLE 2024 CLRC Comments, supra note 16.
[42] Francine Lafontaine & Margaret Slade, Exclusive Contracts and Vertical Restraints: Empirical Evidence and Public Policy, in Handbook of Antitrust Economics 391 (Paolo Buccirossi ed., 2008).
[43] ICLE 2024 CLRC Comments, supra note 16, citing James C. Cooper, Luke M. Froeb, Dan O’Brien & Michael G. Vita, Vertical Antitrust Policy as a Problem of Inference, 23 Int’l J. Indus. Org. 639 (2005) (surveying the empirical literature and concluding that, although “some studies find evidence consistent with both pro- and anticompetitive effects . . . virtually no studies can claim to have identified instances where vertical practices were likely to have harmed competition”); James Cooper, Luke Froeb, Daniel O’Brien & Michael Vita, Vertical Restrictions and Antitrust Policy: What About the Evidence?, Comp. Pol’y Int’l 45 (2005) (finding that “some studies find evidence consistent with both pro- and anticompetitive effects . . . virtually no studies can claim to have identified instances where vertical practices were likely to have harmed competition”).
[44] Cal. Assemb. B. 1776, §§ 16730(b), 16731(a)(2) (Cal. 2025).
[45] Geoffrey A. Manne, Brian C. Albrecht & Dirk Auer, Labor Monopsony and Antitrust Enforcement: A Distorting Mirror, 74 DePaul L. Rev. 1119 (2025).
[46] See ICLE 2025 CLRC Comments, supra note 12; Manne, Auer & Albrecht, supra note 16.
[47] See ICLE 2025 CLRC Comments, supra note 12; Manne, Auer & Albrecht, supra note 16.
[48] Zoë Cullen, Bobak Pakzad-Hurson & Ricardo Perez-Truglia, Home Sweet Home: How Much Do Employees Value Remote Work?, 115 AEA Papers & Proc. 276 (2025).
[49] See Michael J. Nader, Managing a California Remote Work Policy: Determining Which Laws Apply, Ogletree Deakins: Insights (Apr. 1, 2022), https://ogletree.com/insights-resources/blog-posts/managing-a-california-remote-work-policy-determining-which-laws-apply; Justworks, HR Compliance for Remote-First Companies: Managing HR Compliance Across Remote Teams (2026), https://assets.ctfassets.net/mnc2gcng0j8q/4cMqUT8G5XIzKjNVqXaDEb/6b07b29650b0dcda8f47e8e7f9967bd4/HR_Compliance_for_Remote-First_Companies_-_Guide.pdf (noting that “[s]ome states may create excessive compliance burdens or tax exposure,” leading “companies to exclude them from approved remote work locations”).
[50] Reiter, supra note 26.
[51] Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 (1977) (“Plaintiffs must prove antitrust injury—i.e., injury of the type the antitrust laws were intended to prevent and that flows from what makes the defendant’s conduct unlawful. The injury must reflect the anticompetitive effect of the violation or of conduct the violation made possible.”) (emphasis in original).
[52] Murat C. Mungan & John M. Yun, A Reputational View of Antitrust’s Consumer Welfare Standard, 61 Hous. L. Rev. 569 (2024), https://scholarship.law.tamu.edu/cgi/viewcontent.cgi?article=2962&context=facscholar.
[53] United States v. Grinnell Corp., 384 U.S. 563, 571 (1966) (“We defined monopoly power as ‘the power to control prices or exclude competition.’ Such power may ordinarily be inferred from a predominant market share.”).
[54] Parker v. Brown, 317 U.S. 341 (1943); see also California v. ARC Am. Corp., 490 U.S. 93 (1989).
[55] Pike v. Bruce Church, Inc., 397 U.S. 137, 142 (1970).
[56] Shira Liu, Dormant Commerce Clause: A Potential Brake on State Antitrust Legislation, 33 Competition: J. Antitrust, UCL & Privacy Section 1 (2023) (observing that constitutional uncertainty persists because courts must apply the “notoriously unclear” Pike balancing test to assess whether the nationwide burdens of expansive state antitrust laws—such as those governing mergers or diverging from federal standards—clearly exceed their local benefits).
[57] S. 8700-A, 2019–2020 Reg. Sess. (N.Y. 2019), https://www.nysenate.gov/legislation/bills/2019/S8700.
[58] See, e.g., S. 933-C, 2021–2022 Reg. Sess. (N.Y. 2021), https://www.nysenate.gov/legislation/bills/2021/S933; S. 6748, 2023–2024 Reg. Sess. (N.Y. 2023), https://www.nysenate.gov/legislation/bills/2023/S6748; S. 335, 2025–2026 Reg. Sess. (N.Y. 2025), https://www.nysenate.gov/legislation/bills/2025/S335.
[59] The Bus. Council, S. 6748-B (Gianaris) / A. 10323 (Peoples-Stokes) (May 20, 2024), https://www.bcnys.org/memo/s6748-b-gianaris-a10323-peoples-stokes (reporting that the CLRC’s Single-Firm Conduct Working Group “outright rejected the Twenty First Century Anti-Trust Act and its proposed adoption of vague and undefined legal standards”).
[60] Jared P. Nagley & Helen Cho Eckert, Amending New York’s Donnelly Act: If at First You Don’t Succeed, Try, Try, and Try Again, Nat’l L. Rev. (Jan. 31, 2025), https://natlawreview.com/article/amending-new-yorks-donnelly-act-if-first-you-dont-succeed-try-try-and-try-again.
[61] Regulation (EU) 2022/1925 of the European Parliament and of the Council of 14 September 2022 on Contestable and Fair Markets in the Digital Sector and Amending Directives (EU) 2019/1937 and (EU) 2020/1828, 2022 O.J. (L 265) 1, https://eur-lex.europa.eu/legal-content/en/TXT/?uri=CELEX%3A32022R1925 (hereinafter Digital Markets Act).
[62] CLRC Recommendation, supra note 9, at 9 (“The Commission considered creating a distinct single-firm conduct framework for firms with significant market power, drawing on EU law prohibiting ‘any abuse by one or more undertakings of a dominant position within the internal market or a substantial part of it.’ The Commission ultimately declined to adopt such an approach, citing concerns about the vagueness and arbitrariness of defining thresholds for substantial market power, the use of differing conduct standards, and the failure of similar efforts in the United States.”).
[63] ICLE 2025 CLRC Comments, supra note 12, at 2–4; Manne, Auer & Albrecht, supra note 16.
[64] Digital Markets Act, supra note 61, arts. 2(1), 3(1)–(2).
[65] Eur. Comm’n, Gatekeepers Under the Digital Markets Act, https://digital-markets-act.ec.europa.eu/gatekeepers_en (last visited Apr. 22, 2026).
[66] Digital Markets Act, supra note 61, arts. 5–7 (establishing ex ante obligations for designated gatekeepers without requiring case-by-case effects analysis).
[67] Giuseppe Colangelo, Antitrust Unchained: The EU’s Case Against Self-Preferencing, 72 GRUR Int’l 538, 542–46 (2023); Lazar Radic, Opening the Walled Garden: Global Regulation and the Unbundling of Apple’s Ecosystem, Truth on the Mkt. (Mar. 19, 2026), https://truthonthemarket.com/2026/03/19/opening-the-walled-garden-global-regulation-and-the-unbundling-of-apples-ecosystem (“The MSCA—like the EU’s Digital Markets Act (DMA)—takes a different approach. It restricts forms of integration precisely where platforms often integrate for efficiency . . . These interventions assume, rather than demonstrate, that platform control is more likely to suppress than enhance competition . . . That assumption sits uneasily with the empirical literature on digital platforms”).
[68] Eur. Ctr. for Int’l Pol. Econ., What About Us? Consumer Response to the Digital Markets Act, Occasional Paper No. 10/2025, at 3 (2025), https://ecipe.org/publications/consumer-response-to-the-digital-markets-act.
[69] Carmelo Cennamo et al., Economic Impact of the Digital Markets Act on European Businesses and the European Economy 5 (CCIA Eur., June 2025), https://www.dmcforum.net/publications/economic-impact-of-the-digital-markets-act-on-european-businesses-and-the-european-economy.
[70] Lazar Radic, Geoffrey A. Manne & Dirk Auer, Regulate for What? A Closer Look at the Rationale and Goals of Digital Competition Regulations, 22 Berkeley Bus. L.J. 201 (2025), https://lawcat.berkeley.edu/record/1312409?v=pdf.
[71] ICLE 2024 CLRC Comments, supra note 16.
[72] David Autor, David Dorn, Lawrence F. Katz, Christina Patterson & John Van Reenen, The Fall of the Labor Share and the Rise of Superstar Firms, 135 Q.J. Econ. 645, 651 (2020) (citations omitted; emphasis added).
[73] 2 William Blackstone, Commentaries on the Laws of England 358 (George Sharswood ed., 1893).
[74] See S. 763, 2025–2026 Reg. Sess. (Cal. 2025).
[75] See, e.g., Niall E. Lynch & Sydney Kirlan-Stout, Criminal Antitrust Enforcement by the California Attorney General: What Can We Expect?, CPI Columns Cartel (Aug. 2024), https://www.pymnts.com/wp-content/uploads/2024/08/Cartel-Column-August-2024-Full.pdf.
[76] See Joseph Matelis & Daniel Richardson, Criminal Enforcement of Section 2 of the Sherman Act, 36 Antitrust 61, 65 (2022) (“Significantly, no grand jury appears to have returned a criminal Section 2 indictment in more than four decades, and nearly 50 years have passed since a grand jury returned an indictment based solely on a Section 2 charge.”); ABA Section of Antitrust L., Comments in Response to the Antitrust Modernization Commission’s Request for Public Comment on Criminal Penalties 5 (Nov. 14, 2005), https://govinfo.library.unt.edu/amc/public_studies_fr28902/criminal_pdf/051114_ABA_Criminal_Remedies.pdf (“For generations, the Antitrust Division has limited criminal enforcement to hard-core cartel conduct. While not an absolute guarantee against prosecuting other conduct, this practice reflects a long-standing, near-universal consensus that only such conduct warrants criminal prosecution.”).
[77] See Matelis & Richardson, supra note 76, at 66–68.
TOTM In standard-essential patent (SEP) licensing, every procedural tweak is also a skirmish over bargaining power. That is what makes licensing negotiation groups (LNGs) more than an obscure acronym in the...
In standard-essential patent (SEP) licensing, every procedural tweak is also a skirmish over bargaining power. That is what makes licensing negotiation groups (LNGs) more than an obscure acronym in the European Commission’s 2026 Technology Transfer Block Exemption Regulation (TTBER) and accompanying Guidelines (TTGs). LNGs would allow technology implementers to bargain collectively with rights holders. Depending on whom you ask, that is either a sensible way to reduce transaction costs—or a buyer cartel with a compliance memo.
The draft TTGs introduced a dedicated section on LNGs and, more notably, offered a soft antitrust safe harbor. In practice, qualifying LNGs would have avoided a full case-by-case assessment if they satisfied a defined set of conditions.
That approach did not come out of nowhere. A few months earlier, the European Commission signaled its position in an informal guidance letter issued jointly with the German competition authority, addressing the creation of the Automotive Licensing Negotiation Group.
That episode sets the stage for this post. It begins by situating LNGs within the broader SEP debate. It then examines the competition-law risks they raise, the limits of analogizing them to patent pools, and their uneasy fit with the Huawei framework.
Finally, it turns to the final TTGs. While the Commission dropped the proposed safe harbor, it kept a dedicated section on LNGs—raising the obvious question: was the intervention worth it?
Read the full piece here.
Regulatory Comments 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 . . .
The International Center for Law & Economics (ICLE) submits these comments in response to the Federal Communications Commission’s (FCC) public notice on reforms to the Lifeline program.[1] ICLE is a nonprofit, nonpartisan research center that applies law & economics methodologies to public policy. Its work promotes sound economic analysis and consumer welfare, particularly in dynamic, technology-driven markets such as telecommunications.
The Universal Service Fund (USF) has expanded access to voice and data services, but it has long faced persistent waste, fraud, and abuse.[2] The program relies on mandatory contributions passed through to consumers as line-item charges on phone bills.[3] Improper payments therefore raise the cost of connectivity and undermine the program’s purpose. The Commission is right to prioritize program integrity and limit excess costs.
The FCC has taken several steps over the past two decades to address these challenges.[4] The 2012 Lifeline reforms created the National Lifeline Accountability Database (NLAD), limited benefits to one per household, and established non-usage de-enrollment rules.[5] The 2016 reforms introduced the National Verifier to centralize eligibility determinations.[6] In 2019, the Commission added further safeguards, including restrictions on enrollment-based compensation, enhanced registration requirements, and stricter documentation rules.[7] These reforms improved oversight but did not eliminate systemic vulnerabilities.
Recent evidence underscores those remaining gaps. The FCC Office of Inspector General (OIG) found that, in three opt-out states, providers received about $5 million to enroll more than 116,000 deceased individuals.[8] Nearly 40,000 of those enrollments may have occurred after death, indicating fraud beyond delays in death-record updates.[9] The report also identified duplicate claims across multiple states.[10] These findings highlight structural weaknesses that require targeted reform.
The Commission should build on prior efforts with a disciplined, cost-benefit approach. Effective reforms must reduce waste, fraud, and abuse while preserving access for eligible households and maintaining provider participation. Measures such as eliminating the opt-out framework, adopting secondary consent verification, and improving data transparency offer targeted ways to strengthen integrity. At the same time, overly burdensome requirements—whether through rigid eligibility rules, excessive compliance obligations, or misaligned service standards—risk deterring participation and undermining the program’s goals.
Lifeline policy should reflect modern network realities and user needs. Minimum service standards should focus on the level of connectivity required for meaningful participation, not parity with average consumption. Prioritizing data-enabled service, streamlining reporting requirements, and aligning incentives with an all-IP communications environment would improve both efficiency and outcomes. The Commission should adopt reforms that deliver measurable integrity gains without sacrificing access, competition, or affordability.
The NPRM correctly targets waste, fraud, and abuse in the Lifeline program. The USF relies on a mandatory contribution assessed on telecommunications providers, which they pass directly to consumers as a line-item surcharge. The contribution factor now stands at 37%, making the surcharge one of the largest hidden taxes on telecommunications services.[11]
Improper payments directly harm the consumers the program intends to help. Funds paid to ineligible subscribers, deceased individuals, fraudulent enrollees, or providers that do not actually deliver service reduce the resources available to eligible households. The FCC owes ratepayers a clear duty to ensure these funds serve their intended purpose.
That said, eliminating waste, fraud, and abuse is only part of the Commission’s mandate. The agency must evaluate proposed reforms through a rigorous cost-benefit framework that accounts not only for improper payments, but also for the real costs that compliance burdens impose on eligible households and participating providers.[12] Section 254 of the Communications Act requires more than fraud reduction. It requires ensuring that low-income Americans can access quality services at affordable rates.[13]
Reforms that reduce fraud but also deter eligible households from enrolling or remaining in the program risk undermining that mandate. The same holds for reforms that increase provider costs to the point that participation declines, leaving eligible consumers with fewer or no service options.[14]
Provider participation is critical to the program’s competitive structure. Lifeline depends on a sufficient number of Eligible Telecommunications Carriers (ETCs) to ensure meaningful consumer choice and coverage across geographic areas and demographic groups. If new requirements significantly raise compliance costs—especially for smaller, non-facilities-based ETCs that serve most Lifeline subscribers on thin margins—providers may scale back service, exit high-cost markets, or leave the program entirely.
The Commission should evaluate each proposed reform with a clear-eyed assessment of who bears the costs and whether those costs are proportionate to the expected gains in program integrity.
The NPRM identifies a central problem: waste, fraud, and abuse in states that opted out of NLAD and the National Verifier.[15] The Commission should eliminate the opt-out framework and require all remaining states to transition to the federal National Verifier for eligibility verification and duplicate checking on a clear, reasonable timeline.
The evidence supports this change. The OIG Report found that providers in opt-out states received nearly $5 million in reimbursements for more than 116,000 deceased individuals over five years.[16] About 40% of those payments went to individuals who had died, or may have died, before initial enrollment.[17] These failures reflect structural weaknesses in state-based systems.
The National Verifier addresses those weaknesses. It connects directly to federal and state databases for real-time or near-real-time eligibility checks. It conducts automated death checks using authoritative federal mortality data and applies uniform identity-verification standards nationwide.[18] These safeguards do not depend on state policy choices.
State deviations can undermine program integrity. California’s decision to eliminate the Social Security number requirement for Lifeline applicants weakened identity verification and made it harder to detect duplicate enrollments. The Commission lacked a proactive mechanism to prevent that change and had to rely on revocation after the fact.[19] No comparable vulnerability exists under the federal system, where the FCC controls verification standards and processes across all states.
The opt-out framework also creates inequities among ratepayers. A low-income household in Texas faces a different verification process, documentation burden, and duplicate-checking regime than an identical household in a neighboring state that uses the National Verifier. The integrity of a federally funded program should not vary based on geography. Yet evidence shows higher rates of improper payments in opt-out states than in states using the federal system.[20] The Commission’s experience with the Affordable Connectivity Program further demonstrates that a uniform federal verification framework is both feasible and more effective than a patchwork approach.
A unified system would also reduce administrative burdens. Providers operating across both NLAD and opt-out states must comply with multiple verification frameworks, each with different processes and requirements.[21] That fragmentation increases compliance costs and complexity. Requiring all states to use the National Verifier would replace that patchwork with a single, consistent system that providers can implement once and apply nationwide.
FCC OIG investigations show that many consumers were enrolled in Lifeline without their knowledge or consent and never received service.[22] These unauthorized enrollments benefit only providers and agents who claim reimbursement. A secondary consent-verification requirement—conditioning enrollment or transfer on an affirmative response to a confirmation text or email—targets this fraud at its source. It does not raise eligibility barriers, add documentation, or require consumers to navigate complex processes. It simply ensures that the person listed on the application confirms their intent to enroll.
This requirement directly disrupts how enrollment fraud occurs. Fraud typically involves submitting an application using a real person’s information without consent or fabricating contact information to create fictitious subscribers.[23] Both schemes depend on the absence of a verification step requiring a response from the actual individual. A confirmation request sent to the applicant’s contact information creates an immediate check. If the contact information is fabricated, no one can respond, and the enrollment fails. If real information is used without consent, the individual receives an unexpected notice and can reject it, triggering scrutiny instead of completing a fraudulent enrollment.
This approach reflects standard fraud-prevention practices. Financial institutions, health care providers, and government agencies routinely use out-of-band confirmation to verify that the person initiating a transaction controls the relevant account.[24] Applying this well-established method to Lifeline is not experimental. It adopts a proven safeguard that works at scale in analogous contexts.
The benefits are even greater for transfers. When an ETC initiates a benefit transfer in NLAD, the system moves the subscriber to a new provider based solely on the initiating provider’s certification of consent. There is no independent verification that the subscriber actually approved the transfer.[25] The OIG Report identifies unwanted transfers as a significant source of harm.[26] Consumers can be switched without notice, lose existing service relationships, and sometimes lose their phone numbers. These transfers benefit only the receiving provider and the agent.
A secondary verification requirement closes this gap. Subscribers who did not authorize a transfer would receive a notice before it occurs and could block it. That shifts detection from after-the-fact remediation to real-time prevention.
This reform would impose minimal burdens on legitimate users. Enrollment already requires applicants to complete certification forms, provide identifying information, and verify eligibility through the National Verifier. A confirmation step—tapping a link or replying to a message—adds only seconds. Most smartphone users already perform similar actions routinely. For willing, eligible applicants, the burden is de minimis. It is far less onerous than existing documentation, annual recertification, or certification requirements.
The NPRM asks whether the Commission should revise the Lifeline program’s minimum service standards.[27] As the Commission evaluates the record, it should focus on what recipients actually need to benefit from connectivity.[28] The goal of a low-income support program is to ensure access to a threshold level of service sufficient for meaningful participation in modern economic and social life. That goal differs from providing parity with average or median consumer usage. Conflating the two has contributed to the Commission’s difficulties with the mobile broadband data-allowance update mechanism.[29]
The relevant question is not how much data the average smartphone user consumes or what median speeds look like. It is what level of service enables Lifeline subscribers to access core applications: job searches, telehealth, education, government services, emergency communications, banking, and basic social connection. The law & economics literature on broadband adoption consistently finds that the largest gains come from moving households from no connectivity to basic connectivity, not from incremental increases beyond that baseline.[30]
The applications that generate these gains are not bandwidth-intensive by modern standards. A telehealth visit typically requires about 1–6 Mbps of sustained throughput.[31] Job portals, government websites, and educational resources require far less. These are the services that justify a low-income subsidy program, and they function reliably at modest bandwidth levels consistent with current minimum service standards.
Even limited connectivity can be transformative. A household with access to a basic data plan can apply for jobs, communicate with health care providers, access benefits, and support children’s education. Those benefits do not depend on 29 GB of monthly data or gigabit speeds. They depend on reliable access to a basic level of connectivity.
Lifeline rules should maximize benefits to recipients while minimizing costs on consumers’ phone bills. Voice-only plans may meet some users’ needs, but adding basic data service imposes minimal additional cost on providers—especially if the Commission avoids overly burdensome minimum data-cap and bandwidth requirements. The benefits of including data, by contrast, are substantial.
On modern all-IP wireless networks, the marginal cost of providing basic broadband data alongside voice service is effectively zero. Voice over Long-Term Evolution (VoLTE) calls use only about 6.6 kbps to 23.85 kbps of bandwidth[32] and run over the same IP infrastructure and spectrum as data traffic. These networks were designed for data, with voice layered on top as a managed IP service, not as a separate circuit-switched function. As a result, the incremental cost of offering data access reflects the data allowance itself, not additional infrastructure. The Commission should recognize that the cost difference between a voice-only plan and a basic bundled broadband plan is far smaller than the current $4 reimbursement differential suggests.
The consumer benefits of data access are far greater. Broadband enables participation in the full range of modern economic and social activity. In 2026, essential services—employment, health care, education, government services, banking, and communication—primarily rely on IP-based applications. A subscriber limited to voice-only service cannot fully access these opportunities. Subsidizing voice without enabling meaningful participation risks wasting program resources.
The Commission’s broader policy goals reinforce this conclusion. The FCC has long encouraged the transition from legacy copper networks to all-IP infrastructure.[33] ICLE has consistently shown that this transition benefits both providers and consumers, while maintaining legacy networks imposes rising costs and diverts investment from next-generation services.[34] Existing rules can slow this transition by making migration more costly.
Continuing to subsidize a distinct voice-only Lifeline tier conflicts with these objectives. The Commission should not promote all-IP migration on one hand while preserving a subsidy structure that assumes voice remains a separate, standalone service. Aligning Lifeline with modern network realities requires prioritizing data-enabled service as the baseline.
The NPRM proposes a one-per-residence rule to address multiple Lifeline subscribers at a single street address.[35] Although well intentioned, a broad rule of this kind would exclude many eligible households. Many Americans share a residential address while maintaining separate economic lives. Multigenerational households are a common example. Pew Research data shows that roughly 25% of Asian, Black, and Hispanic Americans live in such arrangements.[36] These households often include distinct family units that would otherwise qualify for Lifeline support.
Rising housing costs have also increased shared living arrangements. Many young adults now live with parents while remaining financially independent.[37] Other households share a single address while renting individual rooms. Transitional housing presents similar challenges. Families recovering from homelessness, domestic violence, substance abuse, or incarceration often share addresses while maintaining separate eligibility for benefits. A bright-line, one-per-residence rule would deny support to many households that need it most.
The Commission should still address potential abuse. Requiring NLAD to display the number of Lifeline discounts claimed at a given address across all ETCs is a more targeted solution. Greater visibility would allow ETCs to identify potential duplicate claims while preserving access for legitimately independent households.
The NPRM seeks ways to improve program efficiency while reducing reporting burdens on ETCs.[38] The Commission should consolidate FCC Form 481 and FCC Form 555 into a single annual filing with one deadline, one submission portal, and one distribution to all relevant regulators.[39]
The two forms cover overlapping aspects of the same program. They apply to largely the same ETCs, require similar officer certifications, and duplicate administrative work. ETCs must track separate deadlines, maintain parallel workflows, and submit overlapping organizational and identifying information multiple times. A unified filing would eliminate this redundancy.
A consolidated form could combine Form 481’s financial and operational data with Form 555’s recertification metrics into a single annual snapshot of each ETC’s Lifeline participation. ETCs would submit this filing once per year through a centralized portal accessible to the Commission, the Universal Service Administrative Company (USAC), state commissions, and relevant Tribal and territorial governments. The Commission should set a deadline that aligns with the most practical point in the program’s annual cycle.
Dual filings impose the greatest burden on small- and mid-sized, non-facilities-based ETCs, which serve most Lifeline subscribers. Maintaining separate compliance processes, preparing multiple officer certifications, and submitting duplicative filings diverts resources from serving customers. These costs are not trivial.
Reporting burdens can also affect market participation. Providers considering entry must weigh compliance costs against expected Lifeline revenues. In low-density markets with thin margins, duplicative reporting can deter participation. Reducing unnecessary compliance costs would improve efficiency and strengthen the competitive provider ecosystem that supports universal Lifeline access.
The NPRM raises concerns about non-facilities-based ETCs and their association with higher rates of fraud. Enforcement actions show that these providers have been disproportionately linked to serious program-integrity failures. The current compliance-plan framework has not reliably detected or deterred systemic fraud. Greater scrutiny is warranted, and the Commission is right to revisit the conditions attached to its forbearance grant in light of a decade of experience.
At the same time, the Commission must account for the role these providers play in ensuring universal access. Non-facilities-based ETCs are not just a regulatory artifact. They are often essential to serving low-income consumers, particularly in markets where facilities-based providers do not participate in Lifeline.
Any redesign of the forbearance framework should reflect these market realities. Proposed measures—such as enhanced compliance plans, letters of credit, expanded financial disclosures, mandatory annual audits, and more frequent resubmission requirements—impose real costs. The Commission should weigh those costs against expected fraud-reduction benefits using the same cost-benefit framework applied elsewhere in this NPRM.
The NPRM presents several proposals to reduce waste, fraud, and abuse in the Lifeline program. Many of these reforms move in the right direction. The Commission should evaluate each proposal through a rigorous cost-benefit framework that accounts not only for program-integrity gains, but also for effects on participation, provider incentives, and consumer access.
Effective reform requires targeting fraud at its source while preserving access for eligible households. Measures such as eliminating the opt-out framework, requiring secondary consent verification, and improving data visibility within NLAD offer high-impact, low-burden ways to strengthen program integrity. At the same time, overly rigid rules—such as a one-per-residence limit or excessive compliance burdens on providers—risk excluding eligible households and reducing provider participation.
The Commission should also align Lifeline with modern network and consumer realities. Minimum service standards should reflect the level of connectivity needed for meaningful participation, not parity with average usage. Prioritizing data-enabled service over legacy voice-only offerings would better serve recipients and reflect the all-IP networks that now deliver communications services. Streamlining reporting requirements and reducing duplicative compliance obligations would further support a competitive provider ecosystem.
Lifeline succeeds when it connects eligible households to essential services at a reasonable cost to consumers who fund the program. Reforms should focus on that objective: target support to those who need it most, deliver services that meet real-world needs, and ensure that program rules strengthen—rather than undermine—access, participation, and efficiency.
[1] Lifeline and Link Up Reform and Modernization et al., WC Docket No. 11-42 et al., Notice of Proposed Rulemaking, FCC 26-8 (rel. Feb. 23, 2026), https://docs.fcc.gov/public/attachments/FCC-26-8A1.pdf (hereinafter NPRM).
[2] Lifeline and Link Up Reform and Modernization et al., WC Docket No. 11-42 et al., Third Report and Order, Further Report and Order, and Order on Reconsideration, 31 FCC Rcd 3962, ¶ 46 (2016), https://docs.fcc.gov/public/attachments/FCC-16-38A1.pdf (hereinafter 2016 Order).
[3] Paroma Sanyal & Coleman Bazelon, The Economics of Universal Service Fund Reform, The Brattle Group 4–5 (2023), https://incompas.org/wp-content/uploads/2024/10/The-Economics-of-USF-Reform-Brattle_FINAL.pdf.
[4] TracFone Wireless, Inc., Petition for Designation as an Eligible Telecommunications Carrier in New York et al., Order, 23 FCC Rcd 6206 (2008), https://docs.fcc.gov/public/attachments/FCC-08-100A1.pdf.
[5] Lifeline and Link Up Reform and Modernization et al., Report and Order and Further Notice of Proposed Rulemaking, 27 FCC Rcd 6656 (2012), https://docs.fcc.gov/public/attachments/FCC-12-11A1.pdf.
[6] 2016 Order, supra note 2.
[7] Bridging the Digital Divide for Low-Income Consumers et al., Fifth Report and Order et al., 34 FCC Rcd 10886 (2019), https://docs.fcc.gov/public/attachments/FCC-19-111A1.pdf.
[8] See FCC OIG, Advisory Regarding Deceased and Duplicate Lifeline Subscribers 4 (2026), https://www.fcc.gov/sites/default/files/FCC%20OIG%20Advisory%20Regarding%20Deceased%20and%20Duplicate%20Lifeline%20Subscribers.pdf (hereinafter OIG Report).
[9] Id. at 5
[10] Id. at 6.
[11] Proposed Second Quarter 2026 Universal Service Contribution Factor, Public Notice, DA 26-218 (rel. Mar. 16, 2026), https://docs.fcc.gov/public/attachments/DA-26-218A1.pdf.
[12] See Jerry Ellig, Why and How Independent Agencies Should Conduct Regulatory Impact Analysis, 28 Cornell J.L. & Pub. Pol’y 1 (2018), https://scholarship.law.cornell.edu/cgi/viewcontent.cgi?article=1489&context=cjlpp; see also Thomas Hazlett, Benefit–Cost Analysis in the 5.9 GHz Band, J. Benefit-Cost Analysis 1 (2025), https://www.cambridge.org/core/journals/journal-of-benefit-cost-analysis/article/benefitcost-analysis-in-the-59-ghz-band/2788B60F3B5C9F87788F844742C7186A.
[13] 47 U.S.C. § 254(b).
[14] Comments of NCTA—The Internet & Television Association, WC Docket No. 11-42, at 4 (Apr. 19, 2021), https://www.fcc.gov/ecfs/document/1041937652451/1.
[15] NPRM, supra note 1, ¶ 8.
[16] OIG Report, supra note 8, at 4.
[17] Id. at 5.
[18] U.S. Gov’t Accountability Off., GAO-21-235, FCC Has Implemented the Lifeline National Verifier but Should Improve Consumer Awareness and Experience (2021), https://www.gao.gov/assets/gao-21-235.pdf.
[19] Lifeline and Link Up Reform and Modernization, Order, DA 25-965 (WCB Nov. 20, 2025), https://docs.fcc.gov/public/attachments/DA-25-965A1.pdf.
[20] See OIG Report, supra note 8.
[21] Univ. Serv. Admin. Co., 2026 National Verifier Annual Report and Data (2026), https://www.usac.org/wp-content/uploads/lifeline/documents/Data/2026-National-Verifier-Annual-Report-and-Data.pdf.
[22] NPRM, supra note 1, ¶ 33.
[23] 2016 Order, supra note 2, Dissenting Statement of Commissioner Ajit Pai (“Sales agents schemed with consumers—who no longer had skin in the game—to enroll them in Lifeline multiple times, even if the consumer never qualified in the first place.”).
[24] Nat’l Inst. of Standards & Tech., U.S. Dep’t of Com., NIST Special Publication 800-63B, Digital Identity Guidelines: Authentication and Lifecycle Management § 5 (2017), https://pages.nist.gov/800-63-3/sp800-63b.html#sec5.
[25] Univ. Serv. Admin. Co., Benefit Transfers, https://www.usac.org/lifeline/national-lifeline-accountability-database-nlad/benefit-transfers (last visited Apr. 29, 2026).
[26] OIG Report, supra note 8, at 5.
[27] NPRM, supra note 1, ¶¶ 43–56.
[28] Jeffrey Westling, Redefining Broadband Speeds to Reflect User Needs, Am. Action F. (June 15, 2023), https://www.americanactionforum.org/insight/redefining-broadband-speeds-to-reflect-user-needs.
[29] NPRM, supra note 1, ¶ 46.
[30] Wolfgang Briglauer, Jan Krämer & Nicole Palan, Socioeconomic Benefits of High-Speed Broadband Availability and Service Adoption: A Survey, 48 Telecomm. Pol’y 7 (2024), https://doi.org/10.1016/j.telpol.2024.102808.
[31] FCC Consumer & Gov’t Affs. Bureau, Broadband Speed Guide, https://www.fcc.gov/sites/default/files/broadband_speed_guide.pdf (last visited Apr. 29, 2026).
[32] Eur. Telecomm. Standards Inst. (ETSI), Speech Codec Speech Processing Functions; Adaptive Multi-Rate—Wideband (AMR-WB) Speech Codec; General Description, ETSI TS 126 171 V19.0.0 (3GPP TS 26.171 Ver. 19.0.0 Rel. 19), https://www.etsi.org/deliver/etsi_ts/126100_126199/126171/19.00.00_60/ts_126171v190000p.pdf.
[33] Reducing Barriers to Network Improvements and Service Changes; Accelerating Network Modernization, WC Docket Nos. 25-209 & 25-208, Report and Order, FCC 26-19 (rel. Mar. 27, 2026), https://docs.fcc.gov/public/attachments/FCC-26-19A1.pdf; Advancing IP Interconnection; Accelerating Network Modernization; Call Authentication Trust Anchor, Notice of Proposed Rulemaking, FCC 25-73 (rel. Oct. 29, 2025), https://docs.fcc.gov/public/attachments/FCC-25-73A1.pdf.
[34] Comments of ICLE, Reducing Barriers to Network Improvements and Service Changes, WC Docket Nos. 25-209 & 25-208 (Aug. 22, 2025), https://www.fcc.gov/ecfs/document/10822002748400/1; Comments of ICLE, Advancing IP Interconnection; Accelerating Network Modernization; Call Authentication Trust Anchor, WC Docket Nos. 25-304, 25-208, 17-97 (Jan. 20, 2026), https://laweconcenter.org/wp-content/uploads/2026/01/ICLE-Comments-on-Advancing-IP-Interconnection.pdf.
[35] NPRM, supra note 1, ¶ 63.
[36] D’Vera Cohn et al., Financial Issues Top the List of Reasons U.S. Adults Live in Multigenerational Homes, Pew Rsch. Ctr. (Mar. 24, 2022), https://www.pewresearch.org/social-trends/2022/03/24/the-demographics-of-multigenerational-households (reporting that 24% of Asian, 26% of Black, and 26% of Hispanic Americans lived in multigenerational households in 2021, compared with 13% of White Americans).
[37] Id.
[38] NPRM, supra note 1, ¶ 71.
[39] Comments of USTelecom, in re Delete, Delete, Delete, GN Docket No. 25-133 (Apr. 11, 2025), https://www.fcc.gov/ecfs/document/104110786700283/1; Comments of WTA—Advocates for Rural Broadband, in re Delete, Delete, Delete, GN Docket No. 25-133 (Apr. 14, 2025), https://www.fcc.gov/ecfs/document/10411148117639/1.
ICLE Issue Brief Executive Summary American freight railroads are safer today than at any point in their history. Accident rates, hazardous-material releases, and employee fatalities have all declined . . .
American freight railroads are safer today than at any point in their history. Accident rates, hazardous-material releases, and employee fatalities have all declined over the past two decades, driven largely by private investment and innovation.
The Railway Safety Act, introduced after the 2023 East Palestine derailment, would impose prescriptive mandates, including crew-size requirements, federal detector standards, expanded inspections, and accelerated tank-car replacement timelines. These measures assume additional regulation will improve safety.
This issue brief argues the Act is unlikely to do so. It mandates sweeping rulemakings without cost-benefit analysis, relies on rigid requirements that risk locking in current technology, and includes provisions—especially the two-person crew mandate—that lack a clear evidentiary basis. It also expands regulation in ways that dilute focus on the highest-risk operations and may exceed the industry’s capacity to comply.
The costs are substantial. Compliance could reach billions of dollars, while the scale of accident reduction needed to justify those costs is implausible given current trends. Law & economics research further shows that regulatory accumulation in freight transportation raises costs, reduces output, and weakens innovation-driven growth.
A better approach is performance-based regulation: set measurable safety targets and allow firms flexibility in how to achieve them. Policymakers should preserve voluntary industry initiatives and support smaller carriers through targeted funding.
Freight rail safety is improving. The priority should be to sustain that progress, not impose mandates that risk undermining it.
American freight railroads are safer today than at any point in their history. Over the past two decades, the train accident rate per million train-miles has fallen by more than 38%,[1] h hazardous-materials train accident rates have declined by at least 61%,[2] and employee on-duty fatalities reached an all-time low of seven in 2023.[3] Bureau of Labor Statistics injury-rate data place rail transportation squarely in the middle of the American industrial landscape—comparable to general freight trucking and retail trade, and well below air transportation, warehousing, and urban transit.[4] Freight rail remains one of the safest modes of surface freight transport in the United States, and the gap continues to widen.
Political demand for sweeping new railroad safety mandates has nevertheless surged. The February 2023 derailment of Norfolk Southern Train 32N in East Palestine, Ohio—caused by an overheated wheel bearing[5] —pushed rail safety to the center of public debate. Congress responded by introducing and reintroducing the Railway Safety Act across two successive sessions.[6]
This issue brief argues that the proposed Railway Safety Act, despite its rhetorical appeal, suffers from three fundamental deficiencies. First, it mandates sweeping federal rulemakings without requiring cost-benefit analysis, in tension with Executive Order 14192’s deregulatory framework.[7] Second, its prescriptive mandates risk locking in current technology and discouraging the innovation that has driven the industry’s safety gains. Third, several prominent provisions—particularly the two-person crew mandate—appear driven less by safety evidence than by political economy concerns. A better approach would recognize the industry’s substantial safety gains and avoid layering on regulations that do not address identifiable problems. Short of that, policymakers should adopt performance-based regulation that sets measurable safety targets while preserving flexibility to innovate.
The Railway Safety Act also offers a useful case study in how political-economy dynamics can produce net-harmful regulation. A high-profile incident generated intense public demand for action. Concentrated interest groups then shaped the legislative response, while policymakers largely overlooked a factual record of improving safety and ongoing innovation. The result is a bill that would likely impose billions in compliance costs without demonstrated safety benefits, even as the industry’s baseline trajectory remains positive.
This pattern is not unique to rail policy. It reflects a broader regulatory tendency: policymakers confronting complex, technical systems often focus on prescriptive solutions crafted under political pressure, rather than evaluating how those mandates may displace more effective, innovation-preserving alternatives. Error-cost analysis and regulatory forbearance provide a more reliable framework for achieving public-interest goals without foreclosing beneficial market adaptation.
Recent academic work underscores the stakes. Bentley Coffey, Patrick McLaughlin, and Pietro Peretto show in a multi-industry endogenous-growth model that per-unit transportation costs constrain innovation-driven growth; policies that raise those costs—including cumulative regulation—can dampen the translation of manufacturing productivity gains into aggregate output.[8] Their companion empirical paper finds that a 5% increase in federal regulatory restrictions on a freight transportation mode raises unit costs by 0.8–2.3% and reduces quantities shipped by 1.4–4.1%, with effects that persist and compound over time.[9] In this context, the Railway Safety Act is likely to do more harm than good.
The Railway Safety Act first appeared as S. 576 in March 2023, introduced by Sen. Sherrod Brown (D-Ohio) in the immediate aftermath of the East Palestine derailment. The Senate Commerce Committee approved the bill in May 2023, but it never reached the floor during the 118th Congress.[10] Lawmakers reintroduced it in the House in February 2025 (H.R. 928) and again in the Senate on Feb. 24, 2026 (S. 3903), led by Sen. Jon Husted (R-Ohio) and cosponsored by Sens. Maria Cantwell (D-Wash.), Roger Marshall (R-Kan.), Eric Schmitt (R-Mo.), Tammy Baldwin (D-Wis.), Amy Klobuchar (D-Minn.), Bernie Moreno (R-Ohio), and John Fetterman (D-Pa.).[11]
The bill’s core provisions include a mandatory two-person crew for all Class I freight trains; a federal framework governing the spacing, maintenance, sensitivity, and data protocols for wayside defect-detection systems; an expanded definition of “high-hazard trains” (HHT) using a five-car threshold for flammable gases; accelerated tank-car retrofit and phaseout timelines; expanded pre-departure and locomotive inspection requirements; state notification requirements for hazardous-material movements; and higher civil penalties.[12]
These mandates sit uneasily alongside the current administration’s deregulatory framework. Executive Order 14192, issued Jan. 31, 2025, directs agencies to identify at least 10 existing regulations for repeal for every new rule they propose.[13] For fiscal year 2025, the order further requires that the total incremental cost of all new regulations be “significantly less than zero.”[14]
An executive order does not bind Congress, but it highlights a clear policy divide. The administration has articulated a pro-growth, deregulatory agenda, while some congressional constituencies continue to favor prescriptive mandates. That tension raises a central question: whether additional command-and-control safety rules make sense when, as discussed infra, the industry’s safety record continues to improve under the current framework.
The Railway Safety Act would require the U.S. Department of Transportation (DOT) and the Federal Railroad Administration (FRA) to issue a series of new rules with significant compliance costs. This is precisely the type of regulatory expansion Executive Order 14192 seeks to prevent. The interaction between the two is not merely theoretical. If the FRA attempted to comply with both simultaneously, it would need to repeal a substantial number of existing rules and demonstrate a net reduction in regulatory costs—an implausible outcome given the scope of the Act’s mandates.
Much of the political momentum behind the Railway Safety Act rests on the perception that train derailments are a growing crisis. A commonly cited comparison notes more than 1,300 U.S. derailments in 2019, compared with just 73 in the European Union, even though EU rail traffic spans more than five times as many rail-kilometers.[15] That comparison obscures more than it reveals.
First, roughly 60% of U.S. derailments occur in rail yards at very low speeds—the functional equivalent of parking-lot fender benders.[16] Mainline derailments, which pose the greatest risk to public safety, remain at historic lows. Second, the EU uses different reporting methodologies, emphasizing rates per billion train-kilometers and often excluding comparable low-severity yard incidents. Third, the underlying systems differ fundamentally. The U.S. network is freight-dominant and moves far more tonnage per rail-kilometer, while European systems are overwhelmingly passenger-oriented. Direct comparisons between the two are not meaningful.
The long-term trend is clear. Federal Railroad Administration data show equipment-caused train accidents have declined 38% since 2005,[17] and axle- and bearing-related accidents have fallen roughly 81% since the early 1980s. The agency attributes much of this progress to the widespread adoption of wayside hot-bearing detectors and related monitoring technologies.[18] These gains have occurred under the existing regulatory framework. Any policy change should therefore account for how new mandates might disrupt the innovation that produced them.
Prescriptive safety regulation typically rests on a familiar market-failure story: firms underinvest in safety because they do not fully internalize the social costs of accidents, and government mandates must close the gap. The Railway Safety Act adopts that logic. The freight rail industry’s response after East Palestine, however, complicates the narrative.
Within months of the derailment, Class I railroads deployed roughly 1,000 additional wayside detectors, lowered bearing-temperature alert thresholds, adopted industry-wide predictive analytics, and expanded first-responder access to real-time railcar data—investments that outpaced any plausible federal rulemaking.[19] These actions reflect strong private incentives: liability exposure, reputational capital, network reliability, and the economics of asset preservation. In this environment, the case for additional prescriptive regulation weakens. The proposed mandates would largely codify existing practices, but at higher cost and with less flexibility.
The industry has also committed to developing a shared standard for bearing-trending analysis—an algorithmic approach that identifies degradation patterns before absolute temperature thresholds trigger alerts.[20] This trajectory suggests that the Act does not correct a market failure. It instead layers compliance costs onto an already-functioning incentive structure, with effects that extend beyond the industry’s balance sheet.
The Minnesota Wayside Detector System Study, a legislatively mandated assessment completed in March 2026, reinforces this point.[21] The study surveys roughly five wayside-detection technologies deployed on Minnesota’s rail system and 11 additional detector types in use across North America, including hot-bearing detectors, wheel-impact load detectors, acoustic bearing detectors, dragging-equipment detectors, and machine-vision inspection portals. It also examines emerging systems such as distributed fiber-optic sensing and light detection and ranging (LiDAR)-based infrastructure monitoring.[22]
The study’s cost-benefit analysis proves especially revealing. It models three detector-spacing scenarios—10, 15, and 20 miles—applied statewide to nearly 1,000 miles of Class II and Class III track. In each case, recurring operating costs exceed quantified safety benefits, producing a negative cumulative net impact over a 10-year horizon.[23] Because a state legislature commissioned the study, not the industry, its findings carry particular weight. They undermine the premise that federally mandated detector standards would generate net benefits, especially for smaller carriers.
Federal Railroad Administration inspections point in the same direction. In January 2024, the FRA published its High-Hazard Flammable Train Route Assessment, summarizing inspections of more than 2,600 wayside detectors across 28 railroads. Inspectors found that railroads generally monitored detector performance closely, though they identified issues at roughly 120 sites, including calibration drift and inverted transducers.[24] The broader lesson is straightforward: voluntary industry standards often precede federal regulation, not the reverse, because firms can adapt more quickly than regulators.
Even setting aside the political-economy dynamics discussed above, the Railway Safety Act suffers from a set of analytical and structural defects that warrant separate attention. Its core provisions impose prescriptive, uniform mandates rather than risk-weighted, performance-based approaches that an error-cost framework would favor. The bill also omits the cost-benefit discipline that typically constrains major rulemakings. Several key terms—from the five-car threshold in the expanded high-hazard-train definition to fixed detector-spacing and two-person-crew requirements—appear to reflect negotiated compromises rather than empirical evidence about where risk actually concentrates.
These design choices have concrete consequences. They direct safety capital toward specifications likely to become obsolete before regulators finalize implementing rules. They shield incumbent detection technologies from competition by next-generation alternatives. They also impose fixed compliance costs on smaller carriers that do not track marginal risk on their networks. The result is a regulatory regime that treats a dynamic engineering problem as a static checklist—one that predictably produces less safety, less innovation, and more concentrated harm among the least well-resourced participants than the performance-based alternatives discussed above.
The most fundamental objection to the Act is structural. It mandates sweeping DOT rulemakings without requiring the agency to show that each rule’s benefits justify its costs. This approach departs from the longstanding cost-benefit framework set out in OMB Circular A-4—reinstated in its 2003 form by Executive Order 14192—which requires rigorous analysis for economically significant regulations.[25] More importantly, the Act’s prescriptive mandates would limit the Federal Railroad Administration’s ability to make evidence-based regulatory decisions. Even if the agency’s own analysis shows that a requirement’s costs exceed its safety benefits, the statute would require the agency to proceed.
Available estimates suggest that the compliance burden would run into the billions over a decade. One analysis places the cost of the detector-spacing provision alone between $1.1 billion and $2.2 billion.[26] The 2008 Positive Train Control mandate—a single, comparable requirement—cost U.S. freight railroads roughly $10 billion to $15 billion over its implementation period.[27] The Act would layer multiple additional mandates on top of that baseline, with cumulative costs that could rival or exceed the PTC experience.
To justify expenditures of that magnitude, proponents would need to show that the Act prevents a comparable volume of accidents. Before adopting such costly measures, Congress should at least direct the Congressional Research Service or the FRA to estimate the likely compliance burden.
Empirical benchmarks highlight the gap. An analysis of FRA accident data commissioned by the North Carolina Department of Transportation finds that the average property-damage cost of a rail incident is about $122,000 in 2020 dollars.[28] Even high-severity derailments—those involving roughly 40 or more loaded freight cars—reach only about $2.6 million.[29]
A simple break-even calculation illustrates the mismatch. Assume total compliance costs comparable to Positive Train Control—about $10 billion over a decade. Using the $122,000 average property-damage figure, the Act would need to prevent roughly 80,000 incidents to cover its costs. Even valuing each prevented event at $2.6 million, the Act would still need to prevent nearly 4,000 severe derailments over the same period. Neither scenario is plausible. Mainline derailments on Class I railroads number in the low hundreds each year, and the long-term trend continues to decline. No realistic set of prescriptive mandates could reduce incidents at the scale required to satisfy this cost-benefit threshold.
The empirical literature reinforces these concerns. Recent work finds that regulatory accumulation in freight transportation suppresses investment in productivity-enhancing innovation: a 5% increase in regulatory restrictions raises unit shipping costs by 0.8–2.3% and reduces freight volumes by 1.4–4.1%, with effects that compound over time.[30] These are not one-time adjustment costs; they represent persistent constraints on economic growth. The broader literature reaches similar conclusions. One estimate finds that federal regulation added since 1949 reduced annual U.S. output growth by about one percentage point on average, resulting in a 28% reduction in output by 2005.[31]
Expanding the HHT definition to include flammable gas may respond to East Palestine. The Act’s five-car threshold, however—far below the existing standard of 20 or more continuous tank cars or 35 total—rests on no clear risk assessment and would sweep a far larger set of trains into heightened regulation.[32] Existing voluntary “Key Train” practices, codified in Circular OT-55-R, already impose 50-mph speed limits and siding requirements on the highest-risk movements, including those carrying toxic-inhalation-hazard materials, large volumes of hazmat, or spent nuclear fuel.[33] Lowering the threshold risks shifting attention away from genuinely high-risk operations toward routine freight traffic, imposing costs without corresponding safety gains.
The Act’s expanded locomotive and pre-departure inspection requirements reflect a prescriptive, process-oriented approach. They dictate how railroads must inspect equipment rather than what safety outcomes they must achieve. Nothing in the record suggests that more manual inspections would have prevented the East Palestine derailment, which resulted from a bearing failure that developed between automated detector readings, not from an inspection lapse.[34]
The regulatory-design literature has long identified the limits of this approach. Cary Coglianese and David Lazer distinguish among technology-based regulation, which specifies means; performance-based regulation, which specifies ends; and management-based regulation, which requires firms to design their own safety systems to a specified vision.[35] Performance-based and management-based approaches generally outperform prescriptive mandates because they preserve flexibility and encourage efficient solutions. Railroads have already deployed acoustic bearing detectors, machine-vision portals, and trending-analysis algorithms. But the Federal Railroad Administration’s existing prescriptive framework has not adapted to allow these technologies to satisfy regulatory requirements, effectively reducing the return on safety innovation and likely discouraging further investment.
The Act’s provisions establishing federal standards for detector thresholds, spacing, and data protocols pose a particular risk to innovation. The relevant technology remains in flux. The Minnesota study identifies five wayside-detector types currently deployed in the state, 11 additional types in use elsewhere in North America, and a set of emerging technologies—including distributed fiber-optic sensing for landslide and rockfall detection, light detection and ranging (LiDAR)-based infrastructure monitoring, and IoT-enabled remote-condition systems—many still in pilot-project phases.[36] A federal rule written today would codify 2026-era technology. Given the Federal Railroad Administration’s multi-year rulemaking timelines, the resulting standards could become obsolete before they take effect.
Positive Train Control offers a cautionary example. Congress mandated PTC in 2008, locking in a technological specification before the system had fully matured. Implementation ultimately cost the industry roughly $10 billion to $15 billion and required repeated statutory deadline extensions.[37] The mandate imposed substantial costs and delays while foreclosing potentially superior alternatives.[38] The Act’s detector and tank-car provisions risk repeating that pattern.
Coffey, McLaughlin, and Peretto’s growth model explains why this matters beyond railroads. In their framework, transportation is an integral component of production, and per-unit transportation costs set a lower bound on market prices and returns to innovation. When prescriptive mandates raise those costs, they do more than reallocate resources within the transportation sector; they weaken the innovation-growth feedback loop across the broader economy.[39] Technology-freezing regulation in freight rail thus carries macroeconomic consequences.
Section 110 illustrates similar design flaws. It accelerates the phaseout of older-generation tank cars in certain Class 3 flammable-liquid service, setting a primary deadline of Dec. 31, 2027, with a one-year fallback to Dec. 31, 2028 if the secretary determines that manufacturing or retrofit capacity is insufficient. Section 110(b) directs the secretary to remove or revise any conflicting deadlines immediately, and Section 110(d) requires a Government Accountability Office (GAO) review of manufacturing and retrofit capacity within 18 months of enactment.
This acceleration largely duplicates a transition already in progress. Bureau of Transportation Statistics data show that compliance with DOT-117 standards rose from 56% in 2021 to 59% in 2022, with about 6,914 additional cars built or retrofitted in 2023—a trajectory that would converge on the current statutory deadline of May 1, 2029 without further intervention.[40] At most, the Act advances compliance by roughly 17 months. Section 110(b)’s directive to revise conflicting deadlines also risks creating a regulatory gap by voiding existing schedules before new ones take effect, introducing uncertainty into fleet planning and capital investment.
Capacity constraints compound the problem. The Bureau of Transportation Statistics reports that certified facilities produced or retrofitted about 6,914 tank cars in 2022, reflecting the system’s practical annual throughput.[41] Compressing the deadline to December 2027 would require the industry to process the remaining noncompliant fleet at a pace well above demonstrated capacity. If that compression leads to equipment shortages or service disruptions, it would impose costs on shippers and consumers without clear incremental safety benefits, given that safer tank-car designs are already entering the fleet through market-driven replacement.[42]
The Act’s two-person crew mandate presents a similar evidentiary gap. The Federal Railroad Administration concluded in 2019 that it “cannot provide reliable or conclusive statistical data” on whether one-person crews are safer or less safe than multiple-person crews.[43] The agency reversed course in its 2024 final rule without introducing new statistical evidence, relying instead on qualitative judgments about redundancy.[44] The mandate would override collective-bargaining agreements and eliminate the operational flexibility railroads use to tailor staffing to route conditions, traffic density, and available technology. Its primary effect would be to preserve employment levels—a legitimate policy goal in some contexts, but one better addressed through collective bargaining than through safety regulation.
If prescriptive mandates are the wrong tool, what should replace them? The regulatory-design literature and the railroad industry’s recent experience point to a performance-based approach.
First, the Federal Railroad Administration should adopt performance-based regulation. Instead of prescribing detector spacing, crew size, and inspection procedures, the agency should set measurable safety targets—derailments per million train-miles, hazmat release rates, bearing-failure rates—and allow railroads to determine how best to achieve them. Marc Scribner finds the FRA has been “highly prescriptive” and “slow to adopt performance-based alternatives,” and calls for a systematic shift toward outcome-oriented regulation.[45]
First, the Federal Railroad Administration should adopt performance-based regulation. Instead of prescribing detector spacing, crew size, and inspection procedures, the agency should set measurable safety targets—derailments per million train-miles, hazmat release rates, bearing-failure rates—and allow railroads to determine how best to achieve them. Marc Scribner finds the FRA has been “highly prescriptive” and “slow to adopt performance-based alternatives,” and calls for a systematic shift toward outcome-oriented regulation.
Third, smaller carriers need targeted support, not unfunded mandates. The Minnesota study shows that detector mandates would impose net losses on rural short-line railroads. Existing programs—the Consolidated Rail Infrastructure and Safety Improvements (CRISI) Program, Infrastructure for Rebuilding America (INFRA) grants, and state initiatives such as Minnesota’s Short Line Infrastructure Maintenance Tax Credit—can help smaller carriers adopt safety technologies at a pace consistent with their financial capacity.[46]
Finally, the macroeconomic evidence supports regulatory restraint. Coffey, McLaughlin, and Peretto find that regulatory accumulation in freight transportation reduces labor, fuel, and capital productivity across all major freight modes. Their counterfactual simulations show that these costs do not remain within the regulated firms; they propagate through the supply chain, raising prices, reducing shipment volumes, and weakening the innovation-driven growth process. Performance-based regulation that limits prescriptive rule accumulation, preserves investment incentives, and allows railroads to allocate resources toward the highest-return safety technologies would advance both safety and economic growth.
The Railway Safety Act responds to a real tragedy. But sound regulatory design requires more than urgency and good intentions. The Act imposes sweeping prescriptive mandates without cost-benefit justification, in tension with Executive Order 14192’s deregulatory framework. It targets risks imprecisely, substitutes uniform rules for evidence-based prioritization, and would require the Federal Railroad Administration to proceed with rulemakings even where costs exceed benefits.
These design flaws carry predictable consequences. The Act risks locking in current-generation technology, discouraging the innovation that has driven decades of safety gains, and imposing fixed compliance costs that fall hardest on smaller carriers. It would layer billions of dollars in new mandates onto an industry where accident rates, hazmat releases, and equipment-related failures have all declined over time under the existing framework. Its most visible provision—the two-person crew mandate—lacks a clear evidentiary basis and reflects political compromise more than demonstrated safety need.
The broader lesson extends beyond rail policy. Prescriptive, command-and-control regulation often treats dynamic engineering and operational problems as static compliance exercises. In doing so, it can displace more effective, innovation-preserving approaches while introducing economy-wide costs. As the law & economics literature shows, regulatory accumulation in freight transportation raises per-unit costs, suppresses productivity, and weakens the innovation-growth process that underpins long-run economic performance.
A better approach is available. Performance-based regulation would set clear, measurable safety targets—such as derailments per million train-miles or hazmat release rates—while allowing railroads to determine how best to meet them. Policymakers should preserve the voluntary, industry-led initiatives that have already accelerated safety improvements, continue to rely on the FRA’s collaborative advisory process, and provide targeted support to smaller carriers through existing funding programs. This framework aligns incentives, encourages technological progress, and directs resources toward the highest-return safety investments.
Freight rail safety is improving. The central policy challenge is not to react reflexively to high-profile incidents, but to sustain and accelerate that progress. The Railway Safety Act, as currently structured, risks doing the opposite—imposing large, certain costs in pursuit of uncertain and unquantified benefits.
[1] Fed. R.R. Admin., U.S. Dep’t of Transp., Train Accident (Not at Highway-Rail Crossings) Summary, https://data.transportation.gov/stories/s/dsuf-xcni (last visited May 1, 2026) (showing incidents fell from 4.139 per million miles in 2005 to 2.554 in 2025).
[2] Fed. R.R. Admin., U.S. Dep’t of Transp., Rail Safety Overview Report (1.12), https://data.transportation.gov/stories/s/Rail-Safety-Overview-Report-1-12-/dsuf-xcni (last visited May 1, 2026) (reporting 39 hazmat releases in 2005 and 15 in 2025); see also Ass’n of Am. R.Rs., Press Release, FRA 2023 Data Affirms Rail’s Strong, Sustained Safety Record (Mar. 4, 2024), https://www.aar.org/news/fra-2023-data-affirms-rails-strong-sustained-safety-record (claiming a roughly 75% reduction based on FRA data not currently available).
[3] Fed. R.R. Admin., U.S. Dep’t of Transp., Employee on Duty Fatalities, Injuries, and Illnesses, https://data.transportation.gov/stories/s/Employee-on-Duty-Fatalities-Injuries-and-Illnesses/khzx-vxu4 (last visited Apr. 24, 2026) (showing a 68% decline since 2005).
[4] Bureau of Lab. Stats., U.S. Dep’t of Lab., Table 1: Incidence Rates of Nonfatal Occupational Injuries and Illnesses by Industry and Case Type, 2024, https://www.bls.gov/iif/nonfatal-injuries-and-illnesses-tables/table-1-injury-and-illness-rates-by-industry-2024-national.htm (last visited May 1, 2026) (reporting incidence rates per 100 full-time workers: rail transportation (3.4), general freight trucking (3.1), retail trade (3.0), construction (2.2), grocery stores (4.1), air transportation (6.5), urban transit systems (6.1), and warehousing and storage (4.8)).
[5] Nat’l Transp. Safety Bd., Norfolk Southern Railway Derailment with Subsequent Hazardous Material Release and Fires, East Palestine, Ohio (Feb. 3, 2023), NTSB/RIR-24/05, https://www.ntsb.gov/investigations/AccidentReports/Reports/RIR2405%20CORRECTED.pdf.
[6] Railway Safety Act of 2023, S. 576, 118th Cong. (2023); Railway Safety Act of 2025, H.R. 928, 119th Cong. (2025); Railway Safety Act of 2026, S. 3903, 119th Cong. (2026); see also Press Release, Sen. Maria Cantwell, Cantwell, Husted, Colleagues Reintroduce Bipartisan Railway Safety Act (Feb. 24, 2026), https://www.commerce.senate.gov/2026/2/cantwell-husted-colleagues-reintroduce-bipartisan-railway-safety-act.
[7] Unleashing Prosperity Through Deregulation, Exec. Order No. 14,192, 90 Fed. Reg. 9,065 (Feb. 6, 2025).
[8] Bentley Coffey, Patrick A. McLaughlin & Pietro F. Peretto, Transportation, Innovation and Growth (Working Paper, Apr. 9, 2026), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=6444398.
[9] Bentley Coffey, Patrick A. McLaughlin & Pietro F. Peretto, Regulation and the Cost of Moving Goods (Hoover Inst. Econ. Working Paper, Mar. 20, 2026), https://www.hoover.org/sites/default/files/research/docs/26108-McLaughlin-Coffey-Peretto.pdf.
[10] S. 576, 118th Cong. (2023); see also Cong. Budget Off., Cost Estimate for S. 576 (Sept. 2023), https://www.cbo.gov/publication/59947.
[11] H.R. 928, 119th Cong. (2025); S. 3903, 119th Cong. (2026).
[12] S. 3903, 119th Cong. §§ 102–109 (2026).
[13] Exec. Order No. 14,192, § 3, 90 Fed. Reg. at 9,065, supra note 7.
[14] Id. § 3(b).
[15] See Alex N. Press, As Rail Executives Grow Richer, Train Derailments Have Become Commonplace, Jacobin (Feb. 2023), https://jacobin.com/2023/02/train-derailments-east-palestine-norfolk-southern-profits; see also Gregory Labelle, Letter: Does the U.S. Even Care About Rail Safety? The Numbers Suggest It Doesn’t, LehighValleyLive.com (Apr. 11, 2023), https://www.lehighvalleylive.com/opinion/2023/04/does-the-us-even-care-about-rail-safety-the-numbers-suggest-it-doesnt-letter.html. U.S. figures derive from FRA Form 6180.54 accident/incident data; see Fed. R.R. Admin., Off. of Safety Analysis, Train Accidents and Rates, https://safetydata.fra.dot.gov/officeofsafety/publicsite/query/TrainAccidentsFYCYWithRates.aspx. EU figures derive from Eurostat; see Eurostat, Railway Safety Statistics in the EU, https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Railway_safety_statistics_in_the_EU.
[16] Fed. R.R. Admin., U.S. Dep’t of Transp., Accident/Incident Data (2019), https://data.transportation.gov/stories/s/2ju5-8zxb (showing 784 yard derailments out of 1,344 total derailments in 2019).
[17] Fed. R.R. Admin., supra note 2.
[18] Fed. R.R. Admin., U.S. Dep’t of Transp., Off. of Rsch., Dev. & Tech., An Implementation Guide for Wayside Detector Systems (2019), https://railroads.dot.gov/elibrary/implementation-guide-wayside-detector-systems.
[19] See Ass’n of Am. R.Rs., Freight Railroads Announce Key Safety Measures in Drive to Zero Accidents (Mar. 2023), https://www.aar.org/news/freight-railroads-announce-key-safety-measures-in-drive-to-zero-accidents (announcing approximately 1,000 additional hot-bearing detectors, a 170°F alert threshold, and expanded trending analysis); Ian Jefferies, Ass’n of Am. R.Rs., Statement for the Record Before the Subcomm. on R.Rs., Pipelines & Hazardous Materials, H. Comm. on Transp. & Infrastructure 3 (July 23, 2024), https://www.congress.gov/event/118th-congress/house-event/LC73449/text [hereinafter Jefferies Statement] (citing FRA and AAR data and describing implementation of voluntary commitments, including expansion of AskRail to more than 2.3 million first responders); Fed. R.R. Admin., U.S. Dep’t of Transp., Safety Advisory 2023-01: Evaluation of Policies and Procedures Related to the Use and Maintenance of Hot Bearing Wayside Detectors, 88 Fed. Reg. 13,376 (Mar. 3, 2023); see also Cong. Rsch. Serv., East Palestine, OH, Train Derailment and Hazardous Materials Shipment by Rail: Frequently Asked Questions, R47435 (2024).
[20] CPCS Transcom for Minn. Dep’t of Transp., 2025 Wayside Detector System Study (2026), https://www.lrl.mn.gov/docs/2026/mandated/260603.pdf [hereinafter MnDOT CPCS Report]; see also Ass’n of Am. R.Rs., Freight Railroads Announce Key Safety Measures in Drive to Zero Accidents (Mar. 2023), https://www.aar.org/news/freight-railroads-announce-key-safety-measures-in-drive-to-zero-accidents.
[21] MnDOT CPCS Report, supra note 20.
[22] Id. §§ 2.1, 3.1 (cataloguing 16 distinct wayside detector technologies); id. at 34 (describing distributed fiber-optic sensing); id. (describing light detection and ranging (LiDAR)).
[23] MnDOT CPCS Report, supra note 20, ch. 6, Economic and Industry Impacts, at 53–58 (modeling 10-year cost-benefit scenarios at 10-, 15-, and 20-mile detector spacing on Class II and III track and finding negative cumulative net impacts across all scenarios).
[24] Fed. R.R. Admin., U.S. Dep’t of Transp., High-Hazard Flammable Train Route Assessment & Legacy Tank Car Focused Inspection Program: Summary Report (Jan. 22, 2024), https://railroads.fra.dot.gov/sites/fra.dot.gov/files/2024-01/HRA%20Final%20Report_01.22.24.pdf.
[25] Unleashing Prosperity Through Deregulation, supra note 7; Off. of Mgmt. & Budget, Circular A-4 (2003).
[26] Michael F. Gorman, Rail Safety Policy After East Palestine, Regulation (Summer 2023), https://www.cato.org/regulation/summer-2023/rail-safety-policy-after-east-palestine.
[27] Cong. Rsch. Serv., Positive Train Control (PTC): Overview and Policy Issues, R42637 (2018), https://www.congress.gov/crs-product/R42637; see also Ass’n of Am. R.Rs., Positive Train Control, https://www.aar.org/issue/positive-train-control (estimating industry PTC investment at approximately $15 billion).
[28] Steven Bert et al., The Comprehensive Cost of Rail Incidents in North Carolina, Report No. FHWA/NC/2020-44, at 7 & app. fig. 45, at A-19 (Inst. for Transp. Rsch. & Educ., N.C. State Univ., Dec. 2020).
[29] Id. at 7 fig. 6 & app. figs. 45 & 47, at A-19, A-21.
[30] Coffey, McLaughlin & Peretto, Regulation and the Cost of Moving Goods, supra note 9.
[31] John W. Dawson & John J. Seater, Federal Regulation and Aggregate Economic Growth, 18 J. Econ. Growth 137 (2013), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2223315.
[32] Compare 49 C.F.R. § 171.8 (current HHFT definition), with S. 576 § 102 (proposed HHT definition).
[33] Ass’n of Am. R.Rs., Recommended Railroad Operating Practices for Transportation of Hazardous Materials, Circular OT-55-R (eff. July 1, 2022), https://www.aar.org/wp-content/uploads/2022/07/2022-07-01-OT-55-R-CPC-KBD.pdf.
[34] Nat’l Transp. Safety Bd., supra note 5.
[35] Cary Coglianese & David Lazer, Management-Based Regulation: Prescribing Private Management to Achieve Public Goals, 37 Law & Soc’y Rev. 691, 693–96 (2003), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=297162.
[36] MnDOT CPCS Report, supra note 20, figs. 2 & 10, at 5–6 & 16–17, §§ 3.2.4–3.2.5, at 32, 34.
[37] Jerry Ellig & Patrick A. McLaughlin, Preventing a Regulatory Train Wreck: Mandated Regulation and the Cautionary Tale of Positive Train Control (Mercatus Ctr., Working Paper, June 2016), https://www.mercatus.org/media/57556/download.
[38] Id.
[39] Coffey, McLaughlin & Peretto, Transportation, Innovation and Growth, supra note 8.
[40] Railway Safety Act of 2026, S. 3903, 119th Cong. § 110(a), (c), (d) (2026); Bureau of Transp. Stat., U.S. Dep’t of Transp., Fleet Composition of Rail Tank Cars Carrying Flammable Liquids: 2023 Report to Congress (Sept. 2023), https://www.bts.gov/sites/bts.dot.gov/files/2023-09/BTS_Tank_Car_Report_To_Congress_9_13_2023.pdf (reporting monthly production rates); Cong. Rsch. Serv., Freight Rail Safety Issues in the 119th Congress, R47911 (2025) (summarizing industry concerns about accelerated retrofit timelines in current Railway Safety Act proposals).
[41] Bureau of Transp. Stat., U.S. Dep’t of Transp., Fleet Composition of Rail Tank Cars Carrying Flammable Liquids: 2023 Report to Congress v–vi (Sept. 2023), https://www.bts.gov/sites/bts.dot.gov/files/2023-09/BTS_Tank_Car_Report_To_Congress_9_13_2023.pdf.
[42] Id.
[43] Train Crew Staffing, 84 Fed. Reg. 24,735, 24,741 (May 29, 2019), https://www.govinfo.gov/content/pkg/FR-2019-05-29/pdf/2019-11088.pdf.
[44] Train Crew Size Safety Requirements, 89 Fed. Reg. 25,502, 25,508 (Apr. 9, 2024), https://www.federalregister.gov/documents/2024/04/09/2024-06625/train-crew-size-safety-requirements.
[45] Marc Scribner, Toward Performance-Based Transportation Safety Regulation, Competitive Enter. Inst. (Mar. 2017), https://cei.org/studies/toward-performance-based-transportation-safety-regulation.
[46] MnDOT CPCS Report, supra note 20, ch. 6.
Scholarship (ICLE) Abstract The abuse of economic dependence is undergoing a revival, driven by the growing policy emphasis on restoring fairness in digital markets. Legal provisions originally intended to address bargaining power...
The abuse of economic dependence is undergoing a revival, driven by the growing policy emphasis on restoring fairness in digital markets. Legal provisions originally intended to address bargaining power imbalances in traditional, brick-and-mortar settings are now being repurposed as part of the digital enforcement toolkit. Yet the scope and antitrust character of these rules remain contested. Indeed, a central paradox persists. If antitrust law aims to protect competition rather than individual competitors, it is unclear why abuses of economic dependence belong under its purview rather than contract law. Conversely, if such abuses do affect market dynamics, the distinction from abuses of dominant position becomes blurred. Drawing on the insights of transaction cost economics, this paper offers a critical analysis of national provisions on the abuse of economic dependence and outlines a framework for defining the scope and criteria of such abuse as a standalone antitrust offence, with the aim of ensuring consistency between its application and the principles of transaction cost economics.
Popular Media (Affiliate) Stablecoins—digital dollars that move instantly across borders without banks—are quickly becoming a serious part of the financial system. Global movement of stablecoins (primarily among capital . . .
Stablecoins—digital dollars that move instantly across borders without banks—are quickly becoming a serious part of the financial system. Global movement of stablecoins (primarily among capital markets) amounted to $10.9 trillion worldwide in 2025. And consumers around the world are exhibiting growing interest in using stablecoins for purchases, cross-border transactions, and peer-to-peer payments. Stablecoins are here to stay and are poised to occupy an increasingly large role in the economy. The real question is whether the United States will lead this shift or let it move overseas.
TOTM The European Commission published its draft “guidelines on the assessment of mergers under Council Regulation (EC) No 139/2004 on the control of concentrations between undertakings” . . .
The European Commission published its draft “guidelines on the assessment of mergers under Council Regulation (EC) No 139/2004 on the control of concentrations between undertakings” yesterday. The title does what titles of merger guidelines usually do: it lowers expectations. That is useful misdirection. The document itself is anything but dull.
The draft guidelines span more than 100 pages and raise a host of issues. This post zeroes in on one that should give competition lawyers pause: the quiet politicization of competition law through soft-law instruments that sidestep—and ultimately erode—the coherence of established frameworks.
On the surface, the document updates the European Union’s merger-review framework. Read more closely, and a different project emerges: a systematic effort to advance industrial policy through competition law, dressed up as technical refinement.
What follows breaks down how the guidelines do this, the mechanisms they deploy, and why that trajectory should concern you.
TOTM Brazil’s digital markets do not need a regulatory savior so much as a careful doctor. Bill 4,675/2025 arrives with the bedside manner of a reform, but the . . .
Brazil’s digital markets do not need a regulatory savior so much as a careful doctor. Bill 4,675/2025 arrives with the bedside manner of a reform, but the instruments of major surgery: a new bureaucracy, decade-long designations, and open-ended obligations for firms deemed systemically important. Before Congress scrubs in, it should ask whether the patient is actually failing—or whether Brazil’s existing antitrust tools are already doing much of the work.
Late last year, the Brazilian government submitted the bill to the Chamber of Deputies as part of its “Digital Brazil Agenda.” The proposal borrows from Europe’s Digital Markets Act (DMA), but it is not a straight copy. Its structure more closely resembles the United Kingdom’s Digital Markets, Competition and Consumers Act (DMCC).
Unlike the DMA, the bill would not impose a fixed list of obligations as soon as a company is designated. Instead, it creates a second-stage process in which the Administrative Council for Economic Defense (CADE) would study the designated firm’s markets and then decide which firm-specific duties to impose.
That may sound more restrained. It is still a major shift in Brazilian competition policy.
The bill would amend the Brazilian Competition Law (BCL) to create a new Digital Markets Superintendency (SMD) within CADE. It would empower CADE to designate firms as having “systemic relevance in digital markets” for up to 10 years. It would then allow the agency to impose tailor-made “special obligations” drawn from an open-ended statutory menu.
I have previously written here at Truth on the Market about several problems with this proposal. The bill risks quietly pushing aside the consumer-welfare standard and replacing it with vague goals like “the protection of the competitive process” and “the promotion of freedom of choice.”
Lazar Radic and I have also examined the institutional risks of creating the new SMD. That office would duplicate much of the work of CADE’s existing General Superintendence (SG), rather than building digital-market expertise inside CADE’s current investigative body.
The scale of this proposed overhaul deserved a more comprehensive look. To that end, Geoffrey Manne, Dirk Auer, and I recently published an International Center for Law & Economics (ICLE) white paper, “Digital Overreach: A Premature Turn to Ex Ante Regulation in Brazil.” Policymakers, legal practitioners, and academics should consult the full paper for a detailed economic and institutional assessment of the proposed regime.
This post highlights several of the paper’s central claims. Bill 4,675/2025 raises serious institutional concerns, and it may be unnecessary, given Brazil’s existing antitrust toolkit.
Europe’s early experience also offers a warning. Importing a DMA-style model could bring meaningful tradeoffs, including higher compliance and operational costs, more user friction, and further strain on Brazil’s already notorious “Custo Brasil”—the regulatory and structural cost of doing business in the country.
With that in mind, here are several points the Brazilian Congress should consider before enacting an ex ante regime like Bill 4,675/2025.
Regulatory Comments I. Introduction The International Center for Law & Economics (ICLE) appreciates the opportunity to respond to the European Commission’s consultation on proposed measures to specify . . .
The International Center for Law & Economics (ICLE) appreciates the opportunity to respond to the European Commission’s consultation on proposed measures to specify Alphabet’s obligations under Article 6(11) of the Digital Markets Act (DMA). Article 6(11) requires Alphabet to provide access to certain Google Search data to third-party providers of online search engines (OSEs) on fair, reasonable, and non-discriminatory (FRAND) terms.[1] ICLE is a non-profit, non-partisan global research and policy centre that advances evidence-based policy.
These comments address the Commission’s implementing choices under Article 8(2) DMA. They do not challenge Article 6(11) or Alphabet’s designation.[2] The central issue is how the Commission defines ‘effective compliance’. That definition will shape the scope of access, the design of anonymisation, the pricing framework, and the supervisory regime.
The Preliminary Measures risk shifting Article 6(11) from a data-access obligation to a tool for delivering competitor success. The provision does not support that shift. It requires access to specified data, anonymised, on FRAND terms. It does not guarantee that recipients will match Google’s quality, gain market share, or remain in the market. The specification should reflect that legislative choice.
Four themes guide these comments.
Effectiveness. Article 6(11) is an access obligation, not a rescue regime. Effectiveness should turn on whether third parties receive lawful, workable access on FRAND terms. It should not depend on uptake, market-share shifts, or beneficiary survival. Recital 32 confirms that the DMA’s concept of contestability is procedural and opportunity-based.
Data transferability. The premise that shared, anonymised data will generate comparable gains for recipients is uncertain. The literature shows diminishing returns to data and highlights the role of complementary inputs—crawler infrastructure, indexing, ranking systems, engineering capability, and experimentation. Anonymisation further reduces the value of shared data, especially by suppressing rare and tail queries. The Commission should require evidence that specific data fields will produce material improvements, rather than assume parity with Alphabet’s internal use.
Privacy and anonymisation. Search queries are highly sensitive. The Preliminary Measures’ layered design—technical anonymisation combined with extensive contractual restrictions—reflects the limits of anonymisation alone under the General Data Protection Regulation (GDPR). Privacy should operate as a binding constraint. Data minimisation should guide scope and retention, and staged access should mitigate risk as the recipient pool expands.
FRAND pricing. Search data is an information good characterised by high fixed costs and low marginal costs. The FRAND experience with standard-essential patents (SEPs) shows that pricing such assets is inherently indeterminate and yields a range of reasonable outcomes. The draft’s weighted-average-cost-of-capital (WACC) ceiling, exclusion of core investments from the cost base, and open-ended recipient class risk shifting pricing toward recipient ability to pay. Non-discrimination should instead prevent competitive disadvantage among similarly situated recipients, consistent with Unwired Planet v. Huawei. The regime should also allow for periodic ex post review, rather than rely on fixed terms that may not remain appropriate.
These themes reflect a common concern. The Commission’s specification should preserve the balance the DMA strikes—ensuring access while respecting privacy, proportionality, and investment incentives.
The Commission asks whether the proposed measures will be effective in practice.[3] That question should track the legal obligation: effective access to anonymised data on FRAND terms. It should not turn on beneficiary uptake, traffic, market-share shifts, or the survival of particular recipients.
Article 6(11) grants third parties a right of access to specified data, anonymised, on FRAND terms. It does not grant a right to succeed. Conflating ‘effective’ with ‘sufficient to make beneficiaries viable’ would distort the provision. Under that reading, any disappointing competitive outcome—rival exit, lower-quality results, or limited market-share gains—would invite claims that the measures are ineffective and must be tightened. Compliance would become a moving target, defined by the least efficient beneficiaries, rather than by whether Alphabet has met its data-access obligations.
There are strong reasons to reject that approach.
First, beneficiary outcomes depend on many factors beyond data access. Product quality, ranking architecture, engineering capacity, marketing, distribution, brand, and business-model fit all matter. In United States v. Google LLC, Google expert Edward Fox’s analysis implied that user-side data explained only about 3 per cent of the measured Google–Bing search-quality gap. Fox testified that the remaining difference was ‘not from user interaction data’, pointing instead to factors such as ‘innovation’ and ‘better algorithms’.[4] If beneficiaries underperform despite receiving data on FRAND terms, that result may reflect ordinary competition, not non-compliance.
Second, the empirical record on remedies that aim to rebalance market shares through forced disclosure or default changes is mixed. Users often revert to preferred services after intervention. In a 2016 experiment, Mozilla switched Firefox’s default search engine to Bing; Bing retained about 42 per cent of search volume by day 12, with retention declining over time.[5] In Europe, where the Android choice screen has operated since 2020, Google’s share of search has barely shifted.[6] The UK Competition and Markets Authority likewise recognised that, although click-and-query data may improve ranking quality, ‘the empirical evidence finds rapidly diminishing returns to scale’, with effects concentrated in rare queries.[7] These findings do not show that data sharing lacks value. They show that competitive outcomes reflect a broader set of forces. Market shares are therefore a poor proxy for compliance with the DMA’s data-access obligation.
Third, even proponents of data-access remedies warn against using them to override competitive selection. The Crémer–de Montjoye–Schweitzer report acknowledges that ‘the sharing or pooling of data can discourage competitors from differentiating and improving their own data collection and analytics pipelines’.[8] The Furman Report similarly cautions that such interventions carry trade-offs:
Requiring the opening up of a part of a business’s legitimately obtained data holding would be a significant intervention. Platforms would reasonably be concerned about the impact upon their business model, the legitimacy of requiring access to a significant asset, and the impact on incentives for investment in future data collection and management.[9]
These concerns weigh against reading Article 6(11) to require parity with Google.
Some may argue that the DMA’s ‘contestability’ objective permits an outcome-based test of effectiveness. The text does not support that view. Recital 32 defines contestability in procedural terms: the ability of undertakings to overcome barriers to entry and expansion and to challenge the gatekeeper on the merits.[10] The relevant question is whether the measures materially reduce the data-access barriers identified in Recital 61, on FRAND terms and subject to anonymisation, so that eligible third-party search engines can improve and optimise their services. It is not whether those firms achieve a particular market share, user base, query volume, or advertising-revenue shift. Those outcomes depend on factors outside Article 6(11), including product quality, brand, distribution, user preferences, innovation, privacy choices, and the scale and network characteristics of search.
Tying effectiveness to recipient outcomes would also create a predictable ratchet. Beneficiaries that find anonymised data less useful than expected will press for additional data fields. Beneficiaries that find FRAND prices too high will press for lower prices, potentially calibrated to their own constraints. Each step will be framed as necessary to make the measures ‘effective’. Each step will also erode the statutory constraints—privacy, proportionality, and dynamic incentives—that the DMA preserves.
A second, related concern is the implicit causal premise of the Preliminary Measures: that the same data Google uses internally to optimise its search service will, once anonymised and shared, produce comparable improvements for recipients. The literature does not support that assumption, and the empirical record is mixed. The Commission should treat this as a working hypothesis, not as a basis for broad scope or parity defaults.
The economic literature on data as an input is more cautious than the specification suggests. Anja Lambrecht and Catherine Tucker find that big data rarely satisfies the conditions for a sustained competitive advantage, and that ‘the simple act of amassing big data does not confer a long-term competitive advantage’.[11] Tucker summarises the evidence as showing ‘concave returns to data’—initial gains, followed by rapidly diminishing marginal benefits.[12] Hal Varian likewise notes that data scale faces statistical limits: because measurement accuracy increases with the square root of sample size, ‘you have to have four times as big a sample to get twice as good an estimate’.[13] Empirical work on internet search reaches a similar conclusion. Additional data can improve results, but its marginal value depends on context, including user-history depth, algorithmic quality, and system design.[14]
These findings align with how firms actually create value from data. Data is not a stand-alone input. It complements crawler infrastructure, indexing, ranking systems, machine-learning expertise, experimentation capabilities, and product design. The same dataset will produce different outcomes in different hands.[15] ICLE scholars describe this as ‘data immobility’: the value of data depends on the system that generates and uses it, and disclosure cannot transfer that value in the absence of complementary capabilities.[16]
The search-specific evidence points in the same direction. The United States v. Google LLC record includes expert testimony attributing only about 3 per cent of the Google–Bing quality gap to user-side data.[17] Other evidence suggests that Bing can perform comparably to Google in some contexts, despite Google’s larger data scale. That result is difficult to square with a theory that treats data as the single binding constraint. The court itself observed that Bing’s ‘search quality on Desktop measures up to Google’s’.[18]
The Commission’s anonymisation regime further limits transferability. The Preliminary Measures recognise that anonymisation suppresses long, rare, and tail queries—the queries where marginal returns to additional data are highest.[19] The U.S. data-sharing remedy in United States v. Google LLC similarly acknowledges that DMA-style anonymisation can remove the vast majority of queries before sharing.[20] The dataset proposed for disclosure is therefore, by design, stripped of much of the information that would drive marginal improvements.
These points do not undermine Article 6(11). They do, however, support two changes to the specification. First, the Commission should require evidence—not assumption—that the data fields included in the search dataset will deliver material improvements after anonymisation and under contractual limits. Second, the Commission should resist scope expansions based on unsupported claims of necessity. The idea of ‘parity with Alphabet’s own use’[21] may reflect a statutory aspiration, but it does not establish proportionality for every data field, recipient, or use case.
The Preliminary Measures rest on a strong premise: that technical and contractual anonymisation can produce data that is both useful for search optimisation and sufficiently anonymised to satisfy the General Data Protection Regulation (GDPR). Recital 61 of the DMA recognises the tension by requiring anonymisation that does not ‘substantially degrade’ usefulness.[22] The literature on search-query anonymisation, and the Commission’s own framework, show that this trade-off is real. The Commission should treat privacy as a binding constraint on the regime’s design, not as a parameter to relax in pursuit of beneficiary success.
Search queries can reveal highly sensitive information—health, finances, sexuality, location, politics, and ideology—often in granular, longitudinal detail. The 2006 AOL data release illustrates the risk. AOL disclosed 20 million search queries from 658,000 users, replacing identities with numeric pseudonyms. Within days, a New York Times reporter identified user #4417749 as Thelma Arnold of Lilburn, Georgia, based on her queries.[23] Search histories are uniquely revealing. They can expose a user’s health concerns, political interests, religious beliefs, financial anxieties, sexual orientation, and family issues. As privacy advocates have observed, access to search queries can be ‘akin to reading someone’s most complete and intimate diary’.[24]
The technical literature confirms that this risk is not limited to crude anonymisation. Latanya Sweeney showed that 87 per cent of the U.S. population can be uniquely identified using date of birth, gender, and postal code alone.[25] Yves-Alexandre de Montjoye and co-authors found that four spatio-temporal data points can identify 95 per cent of individuals in a large mobility dataset, even when the data is coarse.[26] Cynthia Dwork’s differential-privacy framework formalises the underlying trade-off. Under the ‘Fundamental Law of Information Recovery’, sufficiently accurate answers to enough queries will erode privacy.[27] Differential privacy does not eliminate this constraint; it manages cumulative privacy loss.
The Preliminary Measures themselves recognise these limits. Section 3.1 specifies technical safeguards—removal of identifiers, suppression of long and rare queries, generalisation of metadata, and k=50 / r=50 thresholding—to reduce re-identification risk to a ‘residual level’.[28] Section 3.2 then adds extensive contractual restrictions: prohibitions on attempts to determine which records relate to the same users, prohibitions on linking with auxiliary datasets, limits on augmentation that could weaken safeguards, and audit rights.[29] The Commission describes these contractual measures as necessary to reduce risk to an ‘insignificant level’.[30]
This layered design reflects an important reality. Technical measures alone do not satisfy the GDPR’s anonymisation standard, which requires that re-identification be impossible ‘taking account of all the means reasonably likely to be used’.[31] ICLE’s prior work on the DMA–GDPR interface makes the same point: anonymisation leaves residual risk, and gatekeepers remain the first line of defence.[32] The Commission’s reliance on contractual controls confirms that anonymisation here operates as a governance regime, not a binary switch. Weakening those controls would reintroduce the very risks the regime seeks to manage.
Privacy risk also depends on who receives the data. Risk is not a property of the dataset alone, but of the dataset and the recipient set. Data that poses limited risk in the hands of a single, well-governed recipient may pose far greater risk when shared with a broad and heterogeneous group. The Preliminary Measures extend access to any third-party undertaking offering an online search engine (OSE) in the European Economic Area (EEA), ‘including AI chatbots with OSE functionalities… even if the OSE is provided as part of a broader service’.[33] Expanding eligibility expands the adversary class. It may include recipients with sophisticated inference capabilities not fully addressed in the anonymisation literature.[34]
Against this backdrop, three principles should guide the specification.
First, treat privacy as a binding constraint. If shared data proves less useful than expected, the solution is not to weaken anonymisation, expand permissible uses, or reduce safeguards. The solution is to recognise that the regime delivers what Article 6(11) requires—lawful access on FRAND terms—and that beneficiaries must compete within those limits.
Second, apply data minimisation as the controlling principle for fields, recipients, and retention. The combination of a broad search dataset and a five-year retention period[35] sits at the more permissive end of what proportionality under Recital 61 can support.
Third, consider staged access. Provide synthetic or filtered datasets first, with fuller access contingent on audit, security review, and demonstrated compliance. Staged access aligns with comparable disclosure regimes and reduces the cost of correcting recipient-side failures.
The Preliminary Measures’ FRAND specification risks converting a data-access obligation into a competitor-subsidy regime. Search data, like standard-essential patents (SEPs) and app-store services, is a classic information good. It involves high fixed and sunk costs—collection, indexing, ranking, and quality maintenance—and near-zero marginal costs of replication.
As Carl Shapiro and Hal Varian explain, ‘Information is costly to produce but cheap to reproduce… cost-based pricing does not work’.[36] The two-decade SEP/FRAND experience confirms the point. Accepted methodologies yield a range of defensible outcomes, not a single price.[37] In Microsoft v. Motorola, the court took years to determine a FRAND rate, with a range spanning roughly thirty-fold. Unwired Planet v. Huawei accepted that FRAND yields a range of acceptable royalties, and Optis v. Apple increased the rate sevenfold on appeal.[38] The DMA has already generated a similar dispute in the Apple App Store proceedings, where the Commission has yet to converge on a stable outcome despite years of effort and a €500 million fine.[39]
The Preliminary Measures amplify these structural difficulties through three design choices.
First, paragraph 71 caps Alphabet’s return on capital employed at its weighted average cost of capital, effectively eliminating economic profit. Paragraph 72(i) relaxes that cap only where Alphabet shows it cannot recover costs from its own use of the data.[40] This approach ties pricing to Alphabet’s profitability, not to economic cost.[41] It operates as a transfer rule, not a pricing rule, and risks the under-compensation problem identified by Daniel Spulber and Christopher Yoo in regulated information industries.[42]
Second, paragraph 78 excludes ‘overhead, sunk costs, or investments in data collection, processing and storage not attributable to making the data available’.[43] Those investments create the data’s value. Pricing based only on the marginal cost of disclosure misprices the asset.
Third, paragraph 79(b) requires Alphabet to forecast ‘the expected number of eligible access recipients’.[44] That denominator is inherently unstable. The regime covers all OSE providers in the EEA, including AI chatbots with OSE functionalities. As eligibility expands, cost allocation becomes less predictable. Paragraph 75 compounds the problem by locking in FRAND terms for five years, followed by renegotiation without a clear benchmark. This structure is likely to increase, not reduce, disputes.[45]
Taken together, these features calibrate prices to recipient size and ability to pay, rather than to economic cost. Paragraph 74 makes this explicit for small and medium-sized enterprises (SMEs). It caps their charges at incremental cost and requires that other beneficiaries’ allocations be calculated as if all recipients faced that constraint. The result is cross-subsidisation. The burden of below-cost SME access shifts to other recipients.[46]
Article 6(11) does not support that outcome. It requires access on ‘fair, reasonable and non-discriminatory’ terms—not parity, and not pricing based on what recipients can afford. Properly understood, non-discrimination prevents competitive disadvantage among similarly situated licensees. It does not require identical terms for differently situated firms.[47] An access price designed to ensure viability for all would invert that principle. It would equalise outcomes by shifting the shortfall of less efficient recipients onto more efficient ones.
The incentive effects run in the same direction. Below-economic pricing weakens the gatekeeper’s incentives to invest in data quality, abuse detection, and privacy-preserving analytics. It also reduces recipients’ incentives to invest in independent crawl, ranking, and behavioural data. As noted above, even proponents of ex ante data-access duties warn that such regimes can dampen investment incentives.[48]
A more stable design would follow three principles.
First, the cost base should reflect the full economic cost of supplying the data. That includes compliance, audit, security, and monitoring costs, plus a return that reflects investment risk, not just ex post weighted average cost of capital (WACC).
Second, interpret non-discrimination in line with the SEP/FRAND tradition. The goal is to prevent competitive disadvantage among similarly situated recipients, not to impose identical terms across heterogeneous firms.
Third, build in review. The five-year lock-in, followed by renegotiation without a benchmark, trades short-term certainty for long-term instability. Periodic ex post review would better reflect the Commission’s practice and the dynamics of the market.
Article 6(11) is in force, and ICLE recognises the Commission’s authority to specify what effective compliance requires. The question is how that authority should be exercised. Three points follow.
First, effectiveness should track the legal obligation, not recipient outcomes. Article 6(11) guarantees access to anonymised data on FRAND terms; it does not guarantee competitive success. The DMA’s definition of contestability in Recital 32 is procedural and opportunity-based. It asks whether firms can overcome barriers and compete on the merits, not whether they achieve particular outcomes. The evidence reinforces this distinction. Competitive performance depends on multiple inputs—product quality, engineering, distribution, and business-model fit—not data access alone. A specification that equates effectiveness with competitor success will create a predictable ratchet: broader scope, lower prices, and weaker safeguards, with no principled stopping point.
Second, the Commission should address, not assume away, the core constraints identified above. Data is not a freely transferable input. Its value depends on complementary capabilities, and the empirical literature shows diminishing returns to scale. Anonymisation further limits transferability, especially where it removes tail queries that drive marginal improvements. At the same time, privacy is a binding constraint. The GDPR requires that re-identification be effectively impossible, and the Preliminary Measures’ layered technical-and-contractual regime reflects that reality. These constraints cannot be relaxed without undermining the legal framework.
The same is true for FRAND. Search data is an information good with high fixed costs and low marginal costs. The SEP/FRAND experience shows that pricing such assets yields a range of reasonable outcomes, not a single point. The current specification departs from that logic. The WACC ceiling, the exclusion of core investments from the cost base, and the open-ended recipient class together shift pricing toward recipient ability to pay rather than economic cost. That approach risks under-compensation, cross-subsidisation, and distorted incentives. It weakens investment by both the gatekeeper and recipients.
Third, the Commission should approach specification with discipline and restraint. The DMA aims to provide clarity through ex ante rules, but early experience shows that even narrow specifications can generate complex disputes and significant enforcement costs. Article 6(11) is broader and more operationally demanding than the provisions the Commission has addressed to date. That increases the risk of unintended effects on users, innovation, and privacy.
These themes point in the same direction. The Commission should anchor effectiveness in access, not outcomes; treat privacy and proportionality as binding constraints; recognise the limits of data transferability; and align FRAND with economic cost, rather than recipient viability. It should also build in mechanisms—such as staged access and periodic ex post review—that allow the regime to adjust without constant expansion.
The Commission has the authority to define what Alphabet must do. It should exercise that authority with care. Article 6(11) does not authorise a competitor-subsidy regime, and the specification should not evolve into one in the name of effectiveness.
[1] Eur. Comm’n, For Public Consultation in Case DMA.100209 — SP — Alphabet — Article 6(11): Preliminary Measures (16 Apr. 2026), https://digital-markets-act.ec.europa.eu/document/download/b3aed7f6-c45c-4bfa-b032-b8975a48bb06_en [hereinafter Preliminary Measures]; Eur. Comm’n, Case DMA.100209 — SP — Alphabet — Article 6(11) Google Search Data Sharing: Case Summary (16 Apr. 2026), https://digital-markets-act.ec.europa.eu/dma100209-consultation-proposed-measures-google-search-data-sharing_en [hereinafter Case Summary].
[2] Regulation (EU) 2022/1925 of the European Parliament and of the Council of 14 Sept. 2022 on Contestable and Fair Markets in the Digital Sector, art. 6(11), 2022 O.J. (L 265) 1 [hereinafter DMA]; Eur. Comm’n, Commission Designates Six Gatekeepers Under the Digital Markets Act (5 Sept. 2023), https://ec.europa.eu/commission/presscorner/detail/en/ip_23_4328.
[3] Case Summary, supra note 1, at 1 (‘Interested third parties are now consulted on these measures, in particular their effectiveness, completeness, and implementation timelines’).
[4] See Rebuttal Testimony of Professor Douglas W. Oard, Trial Ex. UPXD105 at 7, 32, United States v. Google LLC, No. 1:20-cv-03010-APM (D.D.C. 15 Nov. 2023) (reporting a measured Bing–Google IS4@5 gap of 3.924 and a frozen-versus-retrained Google effect of 0.113, and quoting Edward Fox’s testimony that the remaining ‘97 percent’ was ‘not from user interaction data’).
[5] United States v. Google LLC, Memorandum Opinion, No. 20-cv-3010 (APM), at 117 (D.D.C. 5 Aug. 2024), https://www.tn.gov/content/dam/tn/attorneygeneral/documents/pr/2024/pr24-59-Google.pdf; see also Geoffrey A. Manne, A Critical Analysis of the Google Search Antitrust Decision 16–17, Int’l Ctr. for L. & Econ. (14 Aug. 2024), https://laweconcenter.org/wp-content/uploads/2024/08/Manne-Google-Search-Decision-Analysis-2024-08-14.pdf [hereinafter Manne, Critical Analysis].
[6] Id.
[7] Competition & Mkts. Auth., Online Platforms and Digital Advertising: Market Study, Appendix I: Search Quality and Economies of Scale ¶ 34 (2020), https://assets.publishing.service.gov.uk/media/5fe4957c8fa8f56aeff87c12/Appendix_I_-_search_quality_v.3_WEB_.pdf.
[8] Jacques Crémer, Yves-Alexandre de Montjoye & Heike Schweitzer, Competition Policy for the Digital Era 9, Eur. Comm’n (2019).
[9] Digital Competition Expert Panel, Unlocking Digital Competition ¶ 2.87 (2019) [hereinafter Furman Report].
[10] DMA, supra note 2, recital 32.
[11] Anja Lambrecht & Catherine E. Tucker, Can Big Data Protect a Firm from Competition?, Competition Pol’y Int’l Antitrust Chron. 8 (Jan. 2017).
[12] Catherine Tucker, Digital Data, Platforms and the Usual [Antitrust] Suspects: Network Effects, Switching Costs, Essential Facility, 54 Rev. Indus. Org. 683 (2019).
[13] Tom Krazit, Google’s Varian: Search Scale Is ‘Bogus’, CNET (14 Aug. 2009), https://www.cnet.com/culture/googles-varian-search-scale-is-bogus.
[14] Maximilian Schäfer, Geza Sapi & Szabolcs Lorincz, The Effect of Big Data on Recommendation Quality: The Example of Internet Search 1 (DIW Berlin Discussion Paper No. 1730; DICE Discussion Paper No. 284, 2018), https://www.diw.de/documents/publikationen/73/diw_01.c.581628.de/dp1730.pdf; Maximilian Schäfer & Geza Sapi, Learning from Data and Network Effects: The Example of Internet Search 1 (DIW Berlin Discussion Paper No. 1894, 2020), https://www.diw.de/documents/publikationen/73/diw_01.c.798442.de/dp1894.pdf.
[15] Geoffrey A. Manne & Dirk Auer, From Data Myths to Data Reality: What Generative AI Can Tell Us About Competition Policy, Competition Pol’y Int’l (Feb. 2024) (observing that AI rivals such as OpenAI, Anthropic, and Perplexity emerged without massive incumbent data).
[16] See, e.g., Geoffrey A. Manne & Dirk Auer, Antitrust Dystopia and Antitrust Nostalgia: Alarmist Theories of Harm in Digital Markets and Their Origins, 28 Geo. Mason L. Rev. 1280 (2021).
[17] Fox testimony, in Oard, supra note 4.
[18] United States v. Google LLC, supra note 5, at 46.
[19] Preliminary Measures, supra note 1, ¶¶ 13–22 (describing technical anonymisation, including suppression of long queries and queries containing rare words and word combinations); Case Summary, supra note 1, at 3.
[20] Mikolaj Barczentewicz, Comparing the EU DMA to the Search-Query Data-Sharing Remedy in US v Google, Truth on the Mkt. (3 Sept. 2025), https://truthonthemarket.com/2025/09/03/comparing-the-eu-dma-to-the-search-query-data-sharing-remedy-in-us-v-google.
[21] Preliminary Measures, supra note 1, ¶ 3.
[22] DMA, supra note 2, recital 61.
[23] Michael Barbaro & Tom Zeller, Jr., A Face Is Exposed for AOL Searcher No. 4417749, N.Y. Times (9 Aug. 2006), https://www.nytimes.com/2006/08/09/technology/09aol.html.
[24] Chad Marlow & Jennifer Stisa Granick, Celebrating an Important Victory in the Ongoing Fight Against Reverse Warrants, ACLU (29 Jan. 2024), https://www.aclu.org/news/privacy-technology/fight-against-reverse-warrants-victory.
[25] Latanya Sweeney, Simple Demographics Often Identify People Uniquely 16 (Carnegie Mellon Univ. Data Privacy Working Paper No. 3, 2000).
[26] Yves-Alexandre de Montjoye, César A. Hidalgo, Michel Verleysen & Vincent D. Blondel, Unique in the Crowd: The Privacy Bounds of Human Mobility, 3 Sci. Rep. 1376 (2013); see also Yves-Alexandre de Montjoye et al., Unique in the Shopping Mall: On the Reidentifiability of Credit Card Metadata, 347 Science 536 (2015).
[27] Cynthia Dwork & Aaron Roth, The Algorithmic Foundations of Differential Privacy, 9 Found. & Trends Theoretical Comput. Sci. 211, 214 (2014) (‘the Fundamental Law of Information Recovery states that overly accurate answers to too many questions will destroy privacy in a spectacular way’); see also Cynthia Dwork, Differential Privacy, in Automata, Languages and Programming 1, 1–12 (Michele Bugliesi et al. eds., 2006), https://doi.org/10.1007/11787006_1.
[28] Preliminary Measures, supra note 1, ¶¶ 13–22; Case Summary, supra note 1, at 3 (technical measures reduce re-identification risk ‘to a residual level without unnecessarily degrading the quality or usefulness of the search data’).
[29] Preliminary Measures, supra note 1, ¶¶ 38–39 (prohibiting re-identification, sessionisation, linking, and augmentation), ¶¶ 40–48 (imposing purpose limitation), ¶¶ 36–37 (requiring auditor verification).
[30] Case Summary, supra note 1, at 3.
[31] Regulation (EU) 2016/679 (GDPR), recital 26.
[32] Miko?aj Barczentewicz, Comments of the International Center for Law & Economics on the Joint Guidelines on the Interplay Between the Digital Markets Act and the GDPR, Int’l Ctr. for L. & Econ. (2025), https://laweconcenter.org/resources/icle-comments-on-the-interplay-between-dma-and-gdpr.
[33] Preliminary Measures, supra note 1, ¶ 2.
[34] See Barczentewicz, supra note 20 (observing that the privacy-utility trade-off depends on who receives the data and how many recipients there are).
[35] Preliminary Measures, supra note 1, ¶ 12 (requiring data to be made available for at least five years).
[36] Carl Shapiro & Hal R. Varian, Information Rules: A Strategic Guide to the Network Economy 3 (Harv. Bus. Sch. Press 1999); see also William J. Baumol & J. Gregory Sidak, The Pricing of Inputs Sold to Competitors, 11 Yale J. on Reg. 171 (1994).
[37] Anne Layne-Farrar, A. Jorge Padilla & Richard Schmalensee, Pricing Patents for Licensing in Standard-Setting Organizations: Making Sense of FRAND Commitments, 74 Antitrust L.J. 671 (2007); J. Gregory Sidak, The Meaning of FRAND, Part I: Royalties, 9 J. Competition L. & Econ. 931 (2013).
[38] Microsoft Corp. v. Motorola, Inc., 696 F.3d 872 (9th Cir. 2012); Microsoft Corp. v. Motorola, Inc., 795 F.3d 1024 (9th Cir. 2015); Unwired Planet Int’l Ltd. v. Huawei Techs. Co. [2020] UKSC 37; Optis Cellular Tech. LLC v. Apple Retail UK Ltd. [2025] EWCA Civ 552.
[39] Press Release, Eur. Comm’n, Commission Finds Apple’s App Store Rules Breach Digital Markets Act (23 Apr. 2025), https://ec.europa.eu/commission/presscorner/detail/en/ip_25_1085.
[40] Eur. Comm’n, For Public Consultation in Case DMA.100209 — SP — Alphabet — Article 6(11): Preliminary Measures ¶¶ 71, 72(i) (16 Apr. 2026), https://digital-markets-act.ec.europa.eu/document/download/b3aed7f6-c45c-4bfa-b032-b8975a48bb06_en [hereinafter Preliminary Measures].
[41] Sidak, supra note 37.
[42] Daniel F. Spulber & Christopher S. Yoo, On the Regulation of Networks as Complex Systems: A Graph Theory Approach, 99 Nw. U. L. Rev. 1687, 1711–12 (2005).
[43] Preliminary Measures, supra note 1, ¶ 78.
[44] Id. ¶ 79(b).
[45] Id. ¶ 75.
[46] Id. ¶ 74.
[47] Unwired Planet Int’l Ltd. v. Huawei Techs. Co. [2020] UKSC 37, ¶¶ 112–14; Valéria Silva, FRAND-Licensing Litigation Across the Atlantic, Int’l Ctr. for L. & Econ. (8 Apr. 2025), https://laweconcenter.org/resources/frand-licensing-litigation-across-the-atlantic-a-comparative-assessment-of-us-and-uk-jurisprudence-on-telecom-disputes.
[48] See supra notes 8-9 and accompanying text.
