The State of Competition in the US Rail Sector

I. Introduction

Building on earlier International Center for Law & Economics (ICLE) research[1] into railroads and common-carrier obligations—highlighting how the U.S. rail industry’s competitive structure has evolved beyond common assumptions, with intermodal rivals, contractual arrangements, and capital-market discipline now central to constraining service and pricing—this issue brief offers an update on the state of rail-industry competition in 2025. In particular, it examines how the federal courts’ statutory interpretation in the wake of the 2024 Loper Bright decision are likely to intersect with the economic realities of modern rail markets, and how that intersection might reshape the analysis of regulatory authority.

Three focal points structure the discussion. First, the law’s treatment of market dominance requires a return to the statutory and economic foundations of the Staggers Act—an approach that recognizes product and geographic substitution, consistent with Congress’ original intent, rather than freezing competition analysis in a 1980s procedural mold.[2] Second, and in the same vein, the concept of revenue adequacy under §10704 must be understood as a floor designed to attract capital and network investment, not a ceiling that caps profitability. Third, the broad preemption provisions of the Interstate Commerce Commission Termination Act of 1995 (ICCTA) continue to function as pro-competitive infrastructure by preventing state and local overlays that would otherwise balkanize operations, raise costs, and shrink the effective competitive frontier.

Taken together, these themes show how the economically oriented frameworks of the Staggers Act and ICCTA remain relevant but must be applied with renewed textual discipline and contemporary evidence of competition. Statutory fidelity—sharpened by post-Loper Bright jurisprudence—and economic recognition of intermodal and geographic rivalry both point toward the same conclusion: regulatory interventions should be tightly bounded by text and justified only where competitive forces fail. In a networked industry such as rail, uniformity and competition are not opposing values; indeed, uniformity is what makes competition possible.

II. Statutory Text as First Principle: Grand Trunk in the Post-Loper Bright Era

The U.S. Supreme Court’s 2024 decision in Loper Bright Enterprises v. Raimondo marked a watershed moment in the history of U.S. administrative law.[3] In overruling Chevron U.S.A. v. Natural Resources Defense Council,[4] the Court abandoned the doctrine of complete judicial deference to “reasonable” agency interpretations of ambiguous statutes. While it remains premature to make definitive predictions about the post-Loper regulatory landscape, what is clear is that federal agencies will have to work harder to justify their own interpretations of relevant statutes.

In Chevron’s place, the Court has reaffirmed the Administrative Procedure Act’s command that “the reviewing court shall decide all relevant questions of law,” which ostensibly leaves a gap that will be filled by the pre-Chevron Skidmore deference.[5] The practical effect is that agencies may no longer rely on interpretive leeway to advance broad policy objectives untethered from statutory text. For regulated industries such as rail, which has historically operated under a densely detailed statutory framework, this shift places renewed emphasis on statutory fidelity and the limits of delegated authority.

The 7th U.S. Circuit Court of Appeals’ decision earlier this year in Grand Trunk v. STB demonstrates how courts are approaching rail regulation in the wake of Loper Bright, even without explicit reliance on that case.[6] At issue was the Surface Transportation Board’s (STB) reciprocal-switching rule, which would have compelled carriers to interchange traffic without a threshold finding that the incumbent’s service was inadequate.[7] For decades, both the Interstate Commerce Commission (ICC) and its successor, the STB, had construed 49 U.S.C. § 11102(c) to require such a predicate—phrased in earlier decisions as a showing of “actual necessity” or a “compelling reason,” and operationalized as a finding of inadequate service.[8]

The STB’s rule dispensed with this longstanding requirement, substituting a broader “public interest” inquiry.[9] The 7th Circuit vacated the rule, holding that the statutory text and history compelled the conclusion that inadequate service remained an indispensable prerequisite.[10] By treating that condition as optional, the STB was found to have exceeded its statutory authority. What’s notable is that the 7th Circuit reached this conclusion without invoking Loper Bright.

The court’s reasoning, however, exemplifies the post-Loper mode of review: a refusal to defer to an agency’s policy preferences where they are untethered from statutory language. Loper Bright made clear that ambiguity in a statute does not authorize agencies to create new regulatory regimes,[11] and Grand Trunk shows the application of this principle in practice—although under the guise of traditional statutory interpretation. Together, the two decisions underscore that the relevant inquiry is no longer whether an agency’s policy is “reasonable,” but whether Congress has authorized it expressly.

This shift is particularly significant for the economics of rail competition. It raises a heightened risk of invalidation for regulatory structures designed to reengineer market dynamics by administrative fiat, rather than by reference to statutory commands. In the rail market, a disciplined return to statutory text would also align with sound economic reasoning: statutes such as the Staggers Act[12] and the ICCTA[13] embody deliberate tradeoffs between competition and stability, and between rate flexibility and minimum service guarantees. Treating those compromises as pliable invitations to agency policymaking not only distorts the legal framework but unsettles the economic equilibrium that Congress sought to maintain.

III. Improving Competition Analysis: Statutory Tradeoffs and Economic Realities

The statutory compromises struck in the Staggers Act of 1980 and the ICCTA reflect a deliberate balance between deregulation and the choice to retain limited safeguards. On the one hand, Congress sought to restore competitive market dynamics by freeing carriers from pervasive rate and service controls that had pushed the industry toward insolvency. On the other hand, Congress preserved some tools—such as the common-carrier obligation and maximum-rate jurisdiction—to guard against the possibility of abuse where genuine market power persisted. These statutory tradeoffs are central to the economics of rail competition; they aim to promote efficiency and investment while preventing the reemergence of monopoly rents in captive-shipper markets.

Yet the STB’s competition analysis has not kept pace with these foundational compromises. Rate cases currently treat markets as static and narrowly defined, in contravention of Congress’ original intent. This is in tension with the reality that shippers today operate within far broader competitive environments.[14] Intermodal competition from trucking, pipelines, and even barges frequently constrains rail pricing, as does sourcing flexibility; shippers can alter production or distribution points to take advantage of regional rate differentials. By failing to incorporate these product and geographic alternatives into its analysis, the STB risks mischaracterizing the extent of rail-market power and, in turn, applying remedies that suppress efficient pricing or distort network investments.

This failure of modern competition analysis has emerged in rate cases where the STB refused to take account of many contemporary dynamics. In U.S. freight-rail regulation, a railroad’s “market dominance” must be established before the STB will review a rate’s reasonableness. Since 1998, however, the STB has refused to consider evidence of product or geographic competition when determining market dominance in rate cases.[15]

In other words, the STB examines only direct transportation alternatives (such as other rail carriers or modes) and ignores whether shippers could switch to other products or sourcing locations to avoid the challenged rail rate. This policy, affirmed in the early 2000s by the U.S. Circuit Court of Appeals for the D.C. Circuit,[16] was intended to simplify and expedite rate-dispute proceedings. But this narrow approach to market definition has become increasingly problematic in light of evolving competitive conditions. The STB’s exclusion of broader competition evidence is increasingly problematic—particularly given evolving market conditions in the freight-rail industry.

A. Market Dominance in Rail Regulation: Statutory Framework and STB Policy

Under federal law, the STB cannot review the reasonableness of a challenged rail rate unless the shipper first shows the carrier has “market dominance,” which is defined by statute as “an absence of effective competition from other rail carriers or modes of transportation to which a rate applies.”[17] Congress introduced this market-dominance test as part of railroad deregulation in the late 1970s and early 1980s in order to ensure that regulation applies only where a railroad has monopoly power.[18]

Early on, the ICC limited market-dominance inquiries to direct transportation competition—i.e., other railroads on the route (intramodal competition) or alternative modes like trucks, barges, or pipelines (intermodal competition).[19] The commission explicitly declined to consider “indirect” competition, such as product substitutes or geographic alternatives, and the D.C. Circuit upheld that limited approach as a reasonable and efficient reading of the statute.[20]

Soon after, however, the regulatory stance shifted. In the early 1980s the ICC reconsidered its rules and decided to include evidence of product and geographic competition in market-dominance determinations. For example, a railroad could fend off a market-dominance finding by showing that a captive shipper could avoid using its service by purchasing a substitute product (product competition) or by obtaining the commodity from or shipping it to a different location reachable by another carrier (geographic competition).[21] In upholding this approach, the 5th U.S. Circuit Court of Appeals opined that a fuller consideration of competition aligned with the Staggers Act’s goal for market-based rate setting where that was possible.[22]

The STB was created in 1996 to replace the ICC, and that change brought another policy reversal. By 1998, the STB concluded that accounting for product and geographic competition was impractical and counterproductive. In its December 1998 order Ex Parte No. 627, the STB announced it would “no longer consider evidence of product and geographic competition” in market dominance determinations.[23] The new rule, finalized in 1999, categorically excluded these indirect competition factors from the market-dominance phase. In practical terms, this meant that after 1999, a railroad could not defeat a market-dominance claim by arguing that the shipper had outside alternatives, such as different inputs or sourcing regions.

This policy was promptly challenged by the rail industry. In Association of American Railroads v. STB, however, the court upheld the STB’s discretion.[24] The D.C. Circuit found that the statutory definition of market dominance “does not require the STB to consider indirect (product and geographic) competition.”[25]

B. Why Excluding Broader Competition Is a Mistake in Today’s Environment

The STB’s refusal to consider product and geographic competition rests on an economically impoverished view of how rail markets work. By drawing an artificial boundary around “indirect competition,”[26] the board effectively assumes away the competitive constraints that matter most to shippers in the real economy. The result is a regulatory framework that too easily finds “market dominance” where none exists, inviting unnecessary interventions that distort both pricing and investment.

The core problem is that ignoring product and geographic factors produces a caricature of market power. Antitrust law and modern competition economics insist that market definition must account for reasonable substitutes—whether those substitutes take the form of alternative products or alternative sources of supply.[27] If a buyer can avoid an alleged monopolist’s rate increase by switching fuels or sourcing from a different region, the monopolist’s leverage is limited.

Yet the STB has chosen to exclude these very considerations. It treats the relevant market as confined to a single origin–destination pair and asks only whether another carrier or mode is available for that specific modality and geography. This narrow approach is at odds with the analytical frameworks applied by the U.S. Justice Department (DOJ) and Federal Trade Commission (FTC),[28] and it systematically increases the likelihood of false positives: findings of dominance in settings where real-world competition disciplines rates.

For example, in the 2023 Merger Guidelines, the federal antitrust agencies examine potential market concentration as follows:

Absent price discrimination, a relevant market is described by a product or group of products and a geographic area. In determining whether a hypothetical monopolist would be in a position to exercise market power, it is necessary to evaluate the likely demand responses of consumers to a price increase. A price increase could be made unprofitable by consumers either switching to other products or switching to the same product produced by firms at other locations. The nature and magnitude of these two types of demand responses respectively determine the scope of the product market and the geographic market.[29]

In other words, to determine the presence of market power in an antitrust case, it is necessary to look at the reasonable substitutes that would discipline that power. A rate-case evaluation has a nearly identical disposition: the question is whether a rail provider is exercising market power with rate increases. But to exclude relevant product and geographic markets is essentially to assume the outcome: a rail carrier is a monopolist because the STB decides its market is, by definition, a monopoly.

Concrete examples abound of how these product and geographic vectors undermine the STB’s position. In energy markets, railroads haul coal to power plants, but the rise of cheap natural gas and renewable alternatives has reshaped utility-fuel choices.[30] Abundant natural gas has transformed the generation mix, making coal a far less secure demand source. In practice, this means that even a railroad with sole access to a coal plant cannot raise rates with impunity; if rail delivery becomes too expensive, the utility can and will switch to gas-fired generation. The threat of substitution is a powerful competitive check. Yet under the STB’s framework, such a railroad is still deemed “market dominant,” and its rates subject to review and possible caps, as if no competition existed at all. Refusal to recognize product substitution is tantamount to automatically siding with shippers.

Geographic flexibility exerts similar pressures. Shippers frequently have multiple sourcing or destination options.[31] A utility facing high rail rates from Mine A may shift to Mine B, served by a different carrier or accessible by barge. Grain and chemical shippers may reposition shipments through different terminals or transloading points to arbitrage rate disparities. The economic reality is that the railroad’s residual demand curve is more elastic than it appears if one were to look solely at a single corridor. But because the STB bars geographic evidence, its decisions miss this elasticity and treat the carrier as more powerful than it is. In so doing, it risks intervening in markets that are already self-corrected through shipper substitution.

This critique goes further: by generating false positives, the STB creates significant dynamic inefficiencies. If a carrier is wrongly found to have market dominance, its rates may be capped below what competitive conditions would warrant. That misalignment can deprive railroads of revenues needed to recover high fixed costs and to invest in infrastructure. Over time, this undermines the revenue adequacy that Congress deliberately sought to restore through deregulation. Worse still, regulatory error can distort resource allocation—e.g., prolonging coal shipments that would otherwise exit the market in favor of gas, simply because regulators suppress the rail rate below its market-clearing level. Such distortions are deeply inefficient; they keep uneconomic flows alive while discouraging efficient capital deployment.

The economic disconnect has only grown more pronounced with time. The freight market of 2025 bears little resemblance to that of 1998.[32] Trucking competition has intensified, global supply chains have become more flexible, and advances in data and analytics make it easier than ever to quantify substitution possibilities. Rail-dependent industries—from agriculture to manufacturing—have become nimbler in shifting routes and suppliers. In this environment, the STB’s categorical refusal to acknowledge indirect competition looks increasingly anachronistic. What may once have been justified as a way to simplify cases now functions as a “myopic” view of competition that blinds regulators to the dynamics of a global, intermodal marketplace.

The inconsistency is striking when compared with the STB’s own treatment of other sectors. In pipeline cases, for instance, the STB has allowed evidence of product and geographic substitution to determine market power.[33] This discrepancy underscores that the exclusion is more the product of legacy procedural habits than of principled economic reasoning. For a regulator charged with relying on competition “to the maximum extent possible,” that is a fundamental flaw.[34]

C. Vulnerability of the STB’s Policy in the Post-Loper Bright Era

Ultimately, the courts originally upheld the STB’s exclusion of product and geographic competition primarily by invoking Chevron deference. In Association of American Railroads v. STB, the D.C. Circuit rejected the AAR’s challenge to the 1998 rule, not because it found the statute compelled the board’s reading, but because—under Chevron—the agency retained discretion to interpret the ambiguous statutory term “transportation” in its definition of market dominance.[35] As the D.C. Circuit explained in a follow-up case, the court explained:

Under Chevron the Board maintained discretion to interpret the market dominance definition so as to exclude indirect competition. Therefore, we deferred to its latest interpretation and accepted the Board’s reasoning.[36]

In other words, the judicial imprimatur rested not on the strength of the statutory text, but on a doctrine of agency deference.

That doctrinal foundation has now collapsed. In Loper Bright, the Court held that the Administrative Procedure Act requires lower courts to exercise independent judgment in deciding all questions of law, and that no special deference is owed to agency interpretations of ambiguous statutes.[37] This means that if the STB’s rule is challenged anew, the reviewing court cannot defer to the STB’s administrative judgment about efficiency. It must instead ask whether the statutory text and structure authorize the categorical exclusion of relevant competition evidence.

On its face, the statute does not. Section 10707 defines market dominance as the absence of effective competition from other rail carriers or modes of transportation,[38] and Congress’ broader policy directives stress reliance on “competition and the demand for services” wherever possible.[39] Those provisions would be difficult to square with the board’s prohibition on parties presenting economically relevant evidence of competition. Thus, even setting aside the practical market realities that make product and geographic competition economically salient, the STB’s position would be, at best, on tenuous legal footing in the post-Loper landscape. Federal courts must now scrutinize the statutory text directly, without deference to the STB’s view that excluding such evidence reduces complexity.

Once Chevron is stripped away, what remains is a rule that sacrifices substantive accuracy for administrative convenience—an approach courts are unlikely to uphold when the statutory policy tilts so heavily toward incorporating competitive forces. Put simply, the STB’s blanket prohibition may have survived under the deferential regime of Chevron, but it faces a far tougher slog under Loper Bright’s command of independent judicial review. Indeed, the board has conceded that product and geographic competition can effectively constrain rail rates.[40]

To remedy this legal infirmity, the STB should revisit its 1998 rule and restore consideration of product and geographic competition in market-dominance determinations, aligning its procedures with both the statutory text and contemporary competitive realities.

IV. Revenue Adequacy as a Floor, not a Ceiling

In the rail context, the statutory concept of “revenue adequacy” was introduced by §10704 of the Staggers Rail Act as part of Congress’ broader attempt to restore the financial health of an industry that had been brought to the brink of collapse by decades of pervasive rate and service regulation.[41] The statutory text instructs the STB to:

maintain … standards and procedures for establishing revenue levels for rail carriers … that are adequate, under honest, economical, and efficient management, for the infrastructure and investment needed to meet the present and future demand for rail services and to cover total operating expenses, including depreciation and obsolescence, plus a reasonable and economic profit or return (or both) on capital employed in the business.[42]

The central purpose was clear: Congress wanted private railroads to have the ability to attract capital and to maintain, modernize, and expand their networks.

Yet in recent decades, some have attempted to repurpose revenue adequacy as a tool to constrain railroads’ earnings, treating it as an implicit rate cap.[43] This view is in tension with both the statutory text and the economics of network investment. As Bernard Sharfman has argued, the text of §10704 links adequacy to the industry’s health and ability to attract capital; nothing in the statute authorizes the STB to use adequacy as the basis to lower maximum rate levels.[44] Revenue adequacy was meant to signal whether the system as a whole was capable of earning a competitive return on invested capital, not to set a ceiling for ratemaking in individual disputes. To invert adequacy from a floor into a cap is to treat a measure of industry health as if it were evidence of monopoly rents.

The economic problems with this inversion are significant. If regulators treat adequacy as a ceiling, they effectively punish railroads for achieving the very financial stability that Congress intended for them to attain. It also undermines market incentives to invest in long-lived, capital-intensive assets. Railroads face enormous, fixed costs in maintaining track, equipment, and safety systems, and their ability to fund these investments depends on sustained profitability.

Moreover, using adequacy levels as a de facto rate cap discourages carriers from pursuing efficiency gains or productivity improvements, since any financial gains will be clawed back through regulatory rate reviews. The result is a chilling effect on innovation and network expansion. As Sharfman emphasizes, this risks locking the industry into a “just enough to survive” equilibrium, when the statutory design was to ensure a robust, capital-attracting rail system.[45]

Instead, revenue adequacy should be treated as a health diagnostic, not as a lever for rate suppression.[46] In economic terms, adequacy is about dynamic efficiency: ensuring that the industry has sufficient resources to maintain and improve its infrastructure over time. Misusing it as a ceiling prioritizes short-term distributional gains for certain shippers at the expense of long-term system investment, a tradeoff that is both economically inefficient and contrary to congressional intent. Moreover, such a ceiling approach creates perverse incentives: railroads may avoid pursuing new efficiencies or cost savings, fearing that any margin will trigger more aggressive regulatory intervention.

In short, the proper reading of §10704 is that revenue adequacy functions as a threshold assurance—a guarantee that the system can attract the capital it needs. It is not a metric for determining whether railroads are “too profitable” or whether rates should be capped below levels permitted by competition and statutory maximum-rate standards. Using adequacy in this way violates both the statutory scheme and the basic principles of regulatory economics. As Sharfman concludes, adequacy should be viewed as a floor that protects the industry’s viability, not as a ceiling that undermines the very incentives Congress sought to preserve.[47]

Further, read through a post-Loper Bright lens, §10704’s architecture reinforces the reading that it was intended as a floor. The provision instructs the STB to assure adequacy so that private railroads can attract capital; it does not supply a freestanding mandate to cap earnings once that condition is met. Under de novo review, courts will ask whether Congress authorized the use the board proposes.

Treating adequacy as a ceiling is not a textual implementation of § 10704; it is a policy preference that reassigns the statute’s investment assurance into an earnings constraint. That is precisely the kind of atextual gloss that Loper makes difficult to defend. The safer—and more faithful—reading of the statutory text is that adequacy sets the investment floor, while any limits on rates must be justified case-by-case under the separate maximum-rate provisions and on a record that demonstrates actual market power.[48]

V. Preemption and the Networked Industries

Rail is the paradigmatic network industry. It relies on sunk, interconnected assets and generates value from seamless interstate operations. Congress codified that reality in the ICCTA, which vests the STB with exclusive jurisdiction over “transportation by rail carriers” and expressly preempts state and local regulation “with respect to regulation of rail transportation.”[49]

For nearly three decades, courts have read §10501(b) to bar state and local measures that seek to manage or govern rail operations, while tolerating truly general, neutral laws (e.g., building codes, fire and electrical safety) that do not unreasonably burden rail service. Appellate decisions have repeatedly enshrined this division. This congressional choice to make federal regulation “completely exclusive” underscores that preemption was an essential component of the ICCTA’s deregulatory design.[50] State or local schemes that function as rail-specific permitting, siting, or veto devices are preempted; truly general exercises of police power that do not intrude on operations typically survive.[51]

The doctrinal payoff of this express preemption is a single national baseline—not because uniformity is a value in itself, but because uniformity is part of what makes competition possible for a networked industry. It keeps assets fungible, schedules reliable, and intermodal/geographic substitutions viable.

Congress itself confirmed this structure by carving out narrow, express exceptions—e.g., the Clean Railroads Act of 2008,[52] which subjects certain solid-waste rail-transfer facilities to specified state and local oversight and gives the STB a tailored “land-use-exemption permit” role.[53] Thus, when Congress wants state overlay local rules within the rail domain, it does so explicitly. Absent such instruction, the ICCTA’s default is federal exclusivity over rail transportation.

Two additional clarifications matter. First, preemption analysis under §10501(b) is statutory; it is not a free-floating balancing test. The question is whether the challenged measure regulates rail transportation or amounts to a rail-targeted impediment to operations, facilities, or services. Second, overlapping statutes (e.g., the Federal Railroad Safety Act (FRSA))[54] have their own preemption/savings architectures; they do not diminish ICCTA’s independent field of exclusivity over economic regulation of rail transportation.

States often invoke the FRSA to justify regulatory overlays where the Federal Railroad Administration has not acted, but that argument misreads the statutory framework. The FRSA and ICCTA address distinct spheres—safety versus economic regulation—and the presence of an FRSA savings clause does not dilute the ICCTA’s field preemption over rail transportation. As the STB has explained, “when both the Board and FRA have jurisdiction, the later-enacted ICCTA preemption may impose restraints on state action even where the FRSA preemption does not.”[55]  The DOJ has taken the same position, emphasizing that:

[T]he FRSA cannot be read to create a loophole in the ICCTA that would permit a patchwork of state and local regulation over rail transportation simply because the regulations also happen to touch on “safety-related” matters.[56]

While Loper Bright casts a long shadow here as well, in the case of ICCTA §10501(b), the statutory command is so clear, and the legislative intent so unmistakably deregulatory, that both the STB and the courts should interpret preemption broadly and consistently. Congress paired express preemption with exclusive federal jurisdiction precisely to prevent state and local interference with rail operations.

In that context, Loper Bright reinforces, rather than undermines, the STB’s authority to articulate a uniform preemption policy grounded in the statute’s text. Both the board and reviewing courts should therefore err on the side of preemption: neutral, generally applicable laws may apply only insofar as they do not manage or govern rail transportation or impose unreasonable burdens on operations, while policy-driven efforts to steer routing, scheduling, equipment, or facility use are preempted at the threshold. The same textual discipline illustrated by Grand Trunk and the earlier market-dominance discussion applies no less to ICCTA preemption.

For a networked industry like rail, preemption is itself an infrastructure for competition. As discussed above, intermodal rivals and product/geographic substitutions provide discipline for modern rail markets. Patchwork regulation erodes those constraints in three predictable ways.

• Latency and uncertainty—state-by-state permits, facility-specific approvals, or rail-targeted fees—would lengthen cycle times and inject variance that shippers cannot hedge, dulling the price discipline of alternative routes or modes.
• Loss of fungibility—locomotive or equipment restrictions, conflicting emissions or crew mandates—reduces the interchangeability of assets across state borders, shrinking the practical option set for shippers.
• Cost layering—compliance costs accumulated across jurisdictions—raises delivered prices and compresses the competitive frontier that the STB should recognize in market-power analyses.

By contrast, the ICCTA’s uniform baseline lowers search and compliance costs and preserves the reliability and asset mobility on which effective competition depends.

A. Renewed State Overlays, Familiar Fault Lines

Various familiar pressure points have resurfaced since our 2024 report because of recent state initiatives. Several jurisdictions have adopted environmental controls tailored to rail operations, most notably “waste-by-rail” regimes that single out putrescible or construction-and-demolition material moved by rail.[57] These laws typically require enclosure, impose throughput or dwell-time caps, mandate buffer zones, or condition movement on rail-specific permits; they function, in practice, as commodity-specific operational vetoes, rather than neutral health measures.

Some states have also imposed rail-only fees framed as safety or hazardous-materials assessments. Typically, these are keyed to train movements, railcars, or hazardous contents, and dedicated to rail-specific funds coupled with inspection or planning mandates. Such fees operate less like neutral taxation and more like targeted regulation of operations and throughput.

For example, in 2023, Minnesota revived and increased a railroad/pipeline safety assessment that funds hazmat readiness—$4 million annually, with 70% charged to railroads according to route-miles—and with additional charges after “significant response” incidents. The state also created dedicated accounts for the hazmat rail-safety program.[58]

The California Air Resources Board’s (CARB) 2023 In-Use Locomotive Rule would have required railroad operators to fund emissions-indexed spending accounts; comply with a 30-minute idling cap and detailed reporting; and meet age-based operational limits, with zero-emission milestones for switchers and line-haul locomotives.[59] In that case, litigation followed. After CARB represented it would not enforce the rule—pending Environmental Protection Agency (EPA) action—it withdrew its authorization request in January 2025 and noticed a repeal. This left the rule as a salient—but ultimately withdrawn—example of a state rail-specific equipment/emissions mandate.[60]

B. Preemption as Pro-Competitive Infrastructure

ICCTA preemption is not a peripheral doctrine; it is integral to the statutory and economic synthesis developed here. As Loper Bright underscores, the statutory text controls and §10501(b) provides a uniquely clear textual command: exclusive federal jurisdiction, paired with express preemption of state and local regulation of rail transportation.

Reaffirming that command serves two core functions. First, it keeps regulatory authority within the statute’s bounds, ensuring that neither state overlays nor agency improvisations displace the structure Congress enacted. This was reflected in the jurisprudential theme running through Grand Trunk and beyond. Second, it preserves the conditions under which competition operates in a network industry. By blocking state interventions that would otherwise degrade reliability, raise delivered costs, and constrict product and geographic alternatives, ICCTA preemption protects the very competitive frontier the STB must recognize in its analyses.

In network industries, uniformity is not antithetical to competition; it is essential. Rail demonstrates this most vividly: only with a stable and uniform federal baseline can intermodal and geographic substitution discipline carrier pricing and investment choices. The ICCTA is the legal mechanism that delivers this uniformity, and in doing so, it ties the jurisprudential discipline of post-Loper Bright review to the economic realities of modern rail competition.

V. Conclusion

Read in context of the post-Loper Bright frame, the controlling question across rail regulation is no longer whether an agency’s policy is “reasonable” but whether Congress authorized it. Grand Trunk supplies the template. Begin with the statute, its structure, and settled predicates; do not creatively interpret around them. That same textual discipline requires that market-dominance and rate decisions must rest on the record’s evidence of competition, rather than on assumptions that markets are captive or static.

The economic corollary is straightforward. Where product and geographic substitution exist, competition analysis must admit it. Moreover, revenue adequacy under §10704 must be treated as the floor that Congress designed to sustain investment, not as an ad hoc ceiling on earnings. Where rates are to be restrained, the work belongs under §10707 on a case-specific record of market power. This alignment—text first, evidence next—preserves incentives to maintain and expand a capital-intensive network, while targeting intervention to those narrow settings where true dominance persists.

Finally, ICCTA preemption is not peripheral; it is infrastructure for competition in a network industry. A uniform federal baseline lowers latency and compliance costs, preserves asset fungibility and reliability, and keeps intermodal and geographic rivalry viable. Reaffirming robust §10501(b) preemption, modernizing competition analysis, and maintaining the floor reading of revenue adequacy would together yield a coherent legal-economic program. Statutory fidelity is not at odds with rail competition; it is the precondition for it.

[1] Kristian Stout, Common Carrier Reforms to Promote a Healthy Market in the Rail Industry, Int’l Ctr. for Law & Econ. (2025), https://laweconcenter.org/resources/common-carrier-reforms-to-promote-a-healthy-market-in-the-rail-industry.

[2] See H. Rep. 96-1430 (1980), at 96-97, https://www.congress.gov/bill/96th-congress/senate-bill/1946.

[3] Loper Bright Enterprises v. Raimondo, No. 22-451, 603 U.S. 369 (2024), available at https://www.supremecourt.gov/opinions/23pdf/22-451_7m58.pdf.

[4] Chevron U.S.A., Inc. v. NRDC, 467 U.S. 837 (1984).

[5] Daniel Deacon, Loper Bright, Skidmore, and the Gravitational Pull of Past Agency Interpretations, Notice & Comment (Jun. 30, 2024), https://www.yalejreg.com/nc/loper-bright-skidmore-and-the-gravitational-pull-of-past-agency-interpretations.

[6] Grand Trunk Corp. v. Surface Transp. Bd., No. 24-1811 (7th Cir. 2025), available at https://www.govinfo.gov/content/pkg/USCOURTS-ca7-24-01811/pdf/USCOURTS-ca7-24-01811-0.pdf.

[7] Id. at 2.

[8] Id. at 12.

[9] Id.

[10] Id. at 30.

[11] Loper Bright, supra note 3.

[12] S. 1946, 96th Cong. (1979–80) (Staggers Rail Act of 1980), https://www.congress.gov/bill/96th-congress/senate-bill/1946.

[13] Id.

[14] See Stout, supra note 1.

[15] STB Stands by Its Decision to Exclude Product and Geographic Competition Considerations from Rate Complaints, Prog. Railroading (Apr. 3, 2001), https://www.progressiverailroading.com/rail_industry_trends/news/STB-stands-by-its-decision-to-exclude-product-and-geographic-competition-considerations-from-rate-complaints–3366.

[16] Ass’n Amer. R.R. v. Surface Transp. Bd., 237 F.3d 676 (D.C. Cir. 2001).

[17] 49 USC 10707(a).

[18] See Public Law 94-210, 94th Cong., 90 Stat. 31 (Feb. 5, 1976) (Railroad Revitalization and Regulatory Reform Act of 1976), available at https://www.congress.gov/94/statute/STATUTE-90/STATUTE-90-Pg31.pdf; Staggers Rail Act, supra note 12.

[19] Ass’n Amer. R.R. v. Surface Transp. Bd., No. 01-1213 (D.C. Cir. Oct. 8, 2002), https://law.justia.com/cases/federal/appellate-courts/cadc/01-1213/01-1213a-2011-03-24.html.

[20] Atchison, Topeka & Santa Fe Ry. v. ICC, 580 F.2d 623 (D.C. Cir. 1978).

[21] Western Coal Traffic League v. United States, 719 F.2d 772 (5th Cir. 1983) (en banc).

[22] Ass’n Amer. R.R. v. Surface Transp. Bd., 306 F.3d 1108 (D.C. Cir. 2002).

[23] Prog. Railroading, supra note 15.

[24] AAR, supra note 22.

[25] Id.

[26] AAR, supra note 22.

[27] See, e.g., United States v. E. I. du Pont de Nemours & Co., 351 U.S. 377 (1956).

[28] See 2023 Merger Guidelines, U.S. Dep’t of Justice & Fed. Trade Comm’n (2023), § 4, https://www.justice.gov/atr/2023-merger-guidelines (broadly describing the mission of the FTC and DOJ to police antitrust harms in markets, including through the tools of market definition to determine where actual competition takes place).

[29] Id. (emphasis added).

[30] See, e.g., Electricity Demand in the Eastern United States Surged from Heat Wave, U.S. Energy Info. Admin., (Aug. 5, 2020), https://www.eia.gov/todayinenergy/detail.php?id=65604.

[31] See, e.g., A Study of Competition in the U.S. Freight Railroad Industry and Analysis of Proposals That Might Enhance Competition: Executive Summary, Christensen Assocs. (Revised Final Report, Nov. 2009), available at https://www.stb.gov/wp-content/uploads/files/docs/competitionStudy/Executive%20Summary.pdf (concluding that access to transportation alternatives creates competitive pressure that lowers rail rates); Kevin McCoy et al., The Importance of Highways to U.S. Agriculture, U.S. Dep’t of Agric. Agric. Mktg. Serv’n (Dec. 2020), available at https://www.ams.usda.gov/sites/default/files/media/Main_Highway_Report.pdf (noting that shippers will switch to highway transport as a lower-cost substitute when rail transport rates increase).

[32] See Stout, supra note 1.

[33] See STB Docket No. 41685, CF Industries, Inc., v. Koch Pipeline Co., LP, (May 3, 2000).

[34] 49 USC 10101(1).

[35] AAR, supra note 16.

[36] AAR, supra note 22.

[37] Loper Bright, supra note 3.

[38] 49 U.S.C. § 10707(a).

[39] 49 U.S.C. § 10101.

[40] See Market Dominance Determinations—Product and Geographic Competition, 3 S.T.B. 937, 945 n.49 (1998), available at https://www.stb.gov/wp-content/uploads/BV-3-Jan98-Dec98-937.pdf (acknowledging that “in certain circumstances product and geographic competition effectively limit railroad pricing”).

[41] 49 U.S.C. §10704.

[42] Id. at § 10704(a)(2).

[43] See, e.g., Russell Pittman, Two Roads Diverged: Two Alternate Strategies for Protecting Captive Freight Shippers in the “Americas” Model of Freight Rail Restructuring, U.S. Dep’t of Justice Antitrust Div. (Expert Analysis Group Discussion Paper EAG 24-2, Oct. 2024), https://www.justice.gov/atr/media/1372596/dl?inline (framing “revenue adequacy” as a constraint (ceiling) in certain ratemaking contexts to prevent monopoly pricing by claiming it is used to establish “constrained market pricing” in rate cases with captive shippers).

[44] Bernard S. Sharfman, The Significance of ‘Adequate Revenues’ Under the Staggers Rail Act of 1980, Transportation L.J. (forthcoming), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5236244.

[45] Id.

[46] Id.

[47] Id.

[48] There are other instances in the statute that support this view. See 49 U.S.C. § 10101(3) (stating U.S. policy is “to promote a safe and efficient rail transportation system by allowing rail carriers to earn adequate revenues, as determined by the Board”); 49 U.S.C. § 10707(a)(2) (requiring the board to “make an adequate and continuing effort to assist those carriers in attaining revenue levels prescribed under this paragraph”); 49 U.S.C. § 10704(a)(3) (directing the board to “annually determine which carriers are earning adequate revenues”); 49 U.S.C. § 10701(d)(2) (instructing the board to consider “the policy of this part that rail carriers shall earn adequate revenues” when evaluating rate reasonableness).

[49] 49 U.S.C. § 10501(b).

[50] See H.R. Rep. No. 104-311, at 95–96 (1995) (explaining that the ICCTA eliminated a limited state police-power exception “as unnecessary, in view of the Federal policy of occupying the entire field,” and that “[a]lthough States retain the police powers reserved by the Constitution, the Federal scheme of economic regulation and deregulation is intended to address and encompass all such regulation and to be completely exclusive”).

[51] Green Mountain R.R. Corp. v. Vermont, 404 F.3d 638, 643–46 (2d Cir. 2005) (ICCTA § 10501(b) broadly preempts state land-use permitting that functions as a rail-specific veto; notes that generally applicable health/safety codes may apply if they don’t unreasonably burden operations); N.Y. Susquehanna & W. Ry. Corp. v. Jackson, 500 F.3d 238, 252–54 (3d Cir. 2007) (ICCTA preempts state efforts that regulate operations, while allowing neutral laws of general applicability that do not unreasonably interfere); Friberg v. Kan. City S. Ry. Co., 267 F.3d 439, 443–45 (5th Cir. 2001) (Texas anti-blocking statute preempted because it directly regulates train operations/scheduling).

[52] 49 U.S. Code § 10908.

[53] 49 U.S. Code § 10909

[54] 49 U.S.C. §20109.

[55] STB Amicus Br. at 6, 9, Iowa, Chi. & E. R.R. v. Wash. Cnty., 384 F.3d 557 (8th Cir. 2004) (No. 03-3136), 2004 WL 6395939.

[56] Brief for the United States as Amicus Curiae, Ohio v. CSX Transp., 144 S. Ct. 545 (2024) (No. 22-459), 2023 WL 8112708, at *13.

[57] See, e.g., N.Y. Environmental Conservation Law § 27-0712, https://law.justia.com/codes/new-york/env/article-27/title-7/27-0712 (requires railcars hauling putrescible waste to be sealed with hard lids and non-putrescible waste to be covered with hard tarps); N.J.A.C. 7:26-2D.1, https://www.law.cornell.edu/regulations/new-jersey/N-J-A-C-7-26-2D-1 (imposes operational standards on rail carriers that transfer solid waste to/from rail cars, including sealed containers at all times and time-at-facility limits (e.g., a limit for putrescible waste of less than or equal to 72 hours)); New York City DSNY siting rules for putrescible and C&D transfer stations, available at https://www.nyc.gov/assets/dsny/downloads/about/adopted-rules/about-adopted-rule-siting-requirements-transfer-stations-0815.pdf (requires minimum buffer distances from residential/sensitive uses (e.g., 400–700 feet) and throughput conditions, keyed in part to whether material is moved by rail or vessel).

[58] Minn. Stat. § 299A.55 (2023), available at https://www.revisor.mn.gov/statutes/2023/cite/299A.55/pdf.

[59] Final Regulation Order: In-Use Locomotive Regulation, Cal. Air Res. Bd. (2022), Title 13, Cal. Code Regs. §§ 2478–2478.17, available at https://ww2.arb.ca.gov/sites/default/files/barcu/regact/2022/locomotive22/fro.pdf.

[60] Public Hearing to Consider the Proposed Repeal of the In-Use Locomotive Regulation: Staff Report, Initial Statement of Reasons, Cal. Air Res. Bd. (Apr. 29, 2025), available at https://ww2.arb.ca.gov/sites/default/files/barcu/regact/2025/locorepeal/isorlocorepealada.pdf.

I. Introduction

The Promoting Responsible and Open Codes Act (PRO Codes Act), introduced in June by U.S. Reps. Darrell Issa (R-Calif.) and Deborah Ross (D-N.C.),[1] has exposed a long-simmering tension within the world of standards development. On its face, the bill offers a straightforward promise: if a private standard is incorporated by reference into law, that standard must be made freely available to the public. Beneath that promise, however, is a complex set of questions about the economics of standards setting, the role of copyright in sustaining that system, the risks of sweeping statutory reforms that fail to account for the breadth of business models under consideration, and the further attendant risks when Congress wades into a complicated area of litigation.

In contrast to the bold claims made both by the bill’s sponsors[2] and supporters [3] and its critics,[4] the real promise and danger raised by the PRO Codes Act is not that it entrenches copyright, but that it hollows it out—transforming a property right into a conditional license revocable at the government’s whim. Worse, it risks normalizing broad “public-spirited” exceptions to copyright that are untethered from the careful doctrines courts have developed over decades. Properly understood, the PRO Codes Act is not a reinforcement of copyright but a distortion that would weaken it for the very organizations that depend on it most.

Standards are indispensable to the modern economy. They form the hidden architecture of markets and regulation, governing everything from boilers and electrical wiring to aircraft design, medical devices, and nuclear facilities. Their creation requires deep technical expertise and countless hours of volunteer effort, and their adoption by governments has long been praised as a means to harness expert knowledge without the expense and politicization of agency-drafted codes. Copyright revenues can be essential to sustaining this enterprise, ensuring that standards organizations can recover the costs of developing works that—once produced—can be copied at almost no cost.

The current debate over the PRO Codes Act, however, highlights that not all standards-development organizations (SDO) operate in the same way. Two distinct business models have emerged. On one side are voluntary-adoption SDOs, which rely on volunteer engineers and technical experts to produce consensus standards and sustain their work through private, generally commercial sales. On the other side are lobbying-oriented SDOs, which supplement technical work with lobbying campaigns to secure wholesale statutory adoption; they then typically invoke copyright to meter access to the law text.

Both models depend on copyright, and both are legally entitled to enforce it. But the lobbying model raises sharper questions because of the way it couples lobbying with property rights, creating the appearance—if not the reality—of something like “unclean hands.”[5] It is this dynamic, not copyright itself, that has drawn legislative attention, and that fuels the push for legislation like the PRO Codes Act.

The danger is that, by targeting copyright across the board, rather than tailoring its response to one business model, Congress could destabilize the entire ecosystem of standards development. The PRO Codes Act risks collapsing the distinction between the voluntary-adoption and lobbying approaches, and introducing a de facto compulsory-licensing regime that would undercut the very property rights on which standards creation depends.

This issue brief examines these models, demonstrating how each interacts with copyright. It also considers the law & economics implications of the PRO Codes Act’s attempt to redraw the line between public access and private incentives.

II. The PRO Codes Act Doesn’t Mean What You Think It Means

Reps. Issa and Ross have cast their bill as a measure that will “protect public access to federal rules,” while also recognizing the role of private standards in the regulatory process.[6] The sponsors’ stated goal is to assure citizens enjoy free online access to incorporated standards, while preserving the copyright protection that funds those standards’ development. The ensuring debate, however, has split the interested parties into two sharply opposed camps, each of which has framed the proposal in ways that obscure its real long-term effects.

Opponents of the bill have described it as a giveaway to standards bodies. UpCodes, a startup that has litigated against SDOs to force publication of model building codes online, argues that the bill “locks up the law” and empowers private organizations to monetize access to rules that bind the public.[7] Others have warned that the bill would “attempt to copyright the law” and may itself be unconstitutional,[8] arguing that it constitutes a privatization of law that undermines public access.[9]

Supporters of the bill, by contrast, characterize the PRO Codes Act as essential copyright protection, arguing that it would preserve the incentive structure for standards development while ensuring transparency.[10] Not surprisingly, the measure has been championed by the International Code Council (ICC) and the National Fire Protection Association (NFPA), the two SDOs whose lobbying activity, as discussed below, may have been an important factor in forcing Congress to consider this “fix.”

That this cross-cutting lineup of interested parties argues the positions they do is not without irony. Among the bill’s opponents are copyright skeptics, who see the measure as a copyright giveaway, but they are joined in opposition by voluntary-adoption SDOs, who fear the bill will undermine their business model. Meanwhile, while the bill’s supporters are typically pro-copyright institutions, the measure they are defending could transform copyright into a contingent right that would evaporate if incorporated into law unless certain access conditions are met. Put differently, while popular rhetoric treats the PRO Codes Act as strengthening copyright, a careful examination of the economic incentives of copyright in the SDO context suggests that it may fact weaken copyright by subjecting the development and promulgation of codes to a form of compulsory licensing.

The real issue in this debate is not whether citizens should have access to the law—no one disputes that principle—but how to achieve that goal without destabilizing the intellectual-property foundation on which standards development rests. The next sections examine two frequently employed business models of standards development—the voluntary-adoption model and the lobbying model—before returning to the PRO Codes Act to assess how it would affect each and why it risks undermining, rather than reinforcing, copyright.

III. Two Business Models of Standards-Setting Organization

The PRO Codes Act debate, though focused on a relatively narrow segment of the standards ecosystem, brings to the surface broader issues of incentive design, the structure of the standards marketplace, and the stability of copyright policy. To understand the potential downstream effects of any legislative intervention, it is useful to examine two popular models for the commercialization of standards development implicated by the proposed legislation.

A. The Voluntary-Adoption Standards Model

Voluntary-adoption SDOs emerged largely in the late 19th and early 20th centuries, often in response to public-safety crises or industrial inconsistencies. The American Society of Mechanical Engineers (ASME) turned to standardization in the 1880s following a series of deadly boiler explosions and soon published its boiler code to prevent further accidents.[11] Underwriters Laboratories (UL) was founded in 1894 by insurance companies to address rampant fire hazards from unsafe consumer products.[12] The American Society for Testing and Materials (now known as ASTM International), founded in 1898, arose from concerns about the inconsistent quality of steel rails used in the railroad industry.[13]

A common characteristic of these organizations is their reliance on volunteer technical committees composed of engineers, academics, government officials, and industry participants. Decisions are reached through consensus ballots and public review, which—under ANSI’s accreditation regime—must reflect a balance of interests to prevent domination by any single stakeholder.[14]

Voluntary-adoption SDOs—such as ASME, ASTM International, or the Institute of Electrical and Electronics Engineers (IEEE)—do not typically engage in extensive lobbying campaigns to secure wholesale statutory enactment of their codes. This is not to say these organizations never participate in the political process, but their business models do not generally depend on the lobbying process to secure statutory enactment the way the ICC and NFPA models do (as discussed below). ASME, for example, emphasizes that its codes are developed by volunteer committees and are “frequently referenced in regulations worldwide,” attributing such adoption to their technical credibility, rather than political advocacy.[15] Similarly, ASTM International attributes the widespread citation of its standards in law to “technical excellence and global consensus,” rather than lobbying.[16]

This reliance on expertise makes voluntary standards something of a “default” status for standards adoption, which is further supported by federal policy. Under the National Technology Transfer and Advancement Act[17] and OMB Circular A-119, federal agencies are directed to use voluntary-consensus standards wherever possible. At the margins, this reduces SDOs’ need to rely on lobbying campaigns to develop a sustainable business model.[18]

Examples of this model abound. ASME’s volunteer committees, for example, maintain the Boiler and Pressure Vessel Code, a core set of safety standards that federal regulators—including the Occupational Safety and Health Administration (OSHA) and the Nuclear Regulatory Commission (NRC)—incorporate by reference routinely.[19] ASTM International now maintains roughly 12,000 voluntary-consensus standards, widely used in construction, petroleum, environmental regulation, and manufacturing. Federal agencies such as the Environmental Protection Agency (EPA) and the Consumer Product Safety Commission (CPSC) incorporate ASTM International standards by reference into regulations governing emissions testing and product safety.[20] IEEE produces hundreds of standards for electrical and computer engineering, including the well-known IEEE 802 Wi-Fi protocols.[21] Founded 1905, the Society of Automotive Engineers (now known as SAE International) continues to supply aviation and automotive standards referenced by the Federal Aviation Administration (FAA) and U.S. Transportation Department (DOT).[22] The American Concrete Institute (ACI) produces the ACI 318 structural code, which is widely referenced in building regulations.[23]

1. The economics of the voluntary-adoption model

Voluntary-adoption SDOs typically do not rely heavily on dues or public appropriations. Instead, their primary source of revenue is the sale of standards documents. ASTM International, for example, derives roughly 85% of its funding from publication sales, with only about 15% from membership fees.[24]

Thus, such SDOs have generally argued that their ability to continue producing technical standards depends on revenue from the sale of copyrighted materials. Because developing consensus standards requires expert committees, extensive consultation, and periodic revision, SDOs maintain that the ability to earn copyright revenues is essential to their work.[25] Courts have acknowledged this claim, even as other courts have questioned whether industry and regulatory incentives might suffice without copyright protection.[26]

But what must be acknowledged is the uneven economics of standards setting: only some standards are incorporated widely into law or command significant market demand, while many others serve narrower or specialized purposes. While there has been, to date, no broadly available quantification of sales distribution across standards, it is reasonable to infer—drawing on patterns in other creative markets—that revenues are likely concentrated in just a handful of broadly adopted or “blockbuster” standards, which in turn help to sustain the development of more specialized standards with more limited commercial appeal.[27]

Given this reality, the voluntary-adoption model is uniquely vulnerable to shifts in copyright doctrine. Incorporation by reference (IBR) allows regulators to require compliance with an external standard by citing it directly in law, rather than reproducing the text. Moreover, IBR allows updates to the standards to be reflected in the law without the need to promulgate new regulations or pass new legislation. Regulators have generally held that incorporation does not place the standard into the public domain.[28] OSHA, for instance, has repeatedly confirmed that “copyright laws protect national consensus standards,” even when incorporated into OSHA rules.[29]

From an economic perspective, the voluntary-adoption model represents a unique solution to the public-goods problem. Standards have high fixed development costs but, once created, can be reproduced (or incorporated by reference) at negligible marginal cost. Absent some exclusion mechanism, there would be chronic underinvestment. Copyright creates a limited exclusion that allows SDOs to recoup costs, while volunteer labor minimizes the need for large financial outlays.

The voluntary-adoption model’s efficiency stems from its ability to pool technical expertise that governments could not feasibly replicate at scale. Agencies would face significant opportunity costs, both financial and political, if forced to draft thousands of engineering standards internally. Volunteer-consensus processes also produce broader stakeholder buy-in, which enhances compliance and legitimacy.[30] At the same time, the backdrop of copyright ensures that SDOs will be able to fund ongoing work, even for standards that do not generate large market demand.

B.  The Lobbying Model of Standards Development

In contrast to the voluntary-adoption model of standards development is what we can call the “lobbying model.” While it is true that this SDO business model similarly relies on engineering expertise and consensus-building activities, it is distinct in that it seeks to leverage copyright to encourage promulgation of standards-reliant codes. For example, the ICC advertises that an important part of its government-relations team’s mission is to “advocate for the adoption, implementation and enforcement of current building safety codes, standards and policies.”[31] The NFPA notes that its staff “engage[] with policymakers at all levels of government to support adoption and use of NFPA codes and standards.”[32] And at least one account of the relationship between these two organizations detailed the intense lobbying battle between ICC and NFPA in the early 2000s, when each group deployed “lobbyists, political ads, and grassroots campaigns” to persuade states and cities to adopt their specific codes.[33]

The evidence suggests it is likely this business model that has generated many of the conflicts that have inspired legislation like the PRO Codes Act.

1. Mechanics of the lobbying model of standards development

In contrast to the voluntary-adoption approach, the commercialization portion of the lobbying model centers on two sequential moves. First, a private body like the ICC or the NFPA drafts a comprehensive code and lobbies legislatures and agencies to adopt this code wholesale (often with minimal amendments). Second, once legal force is attached, the organization asserts copyright in the text of the law and sells access to the code via print, PDFs, and subscriptions, while offering only constrained read-only portals as a free option.

This two-stage maneuver—leveraging the political process to create a legal obligation and then invoking copyright to control the text—structures a business model in which legal adoption helps to create a captive market for the publisher’s editions.[34] This can generate significant revenue for the relevant standards bodies:

To understand the battle between NFPA and ICC, it is useful to avoid thinking of the two groups as associations of code geeks. Rather, think of them as two publishing houses engaged in a war for book sales. Codebooks are a big business in the United States and abroad. When a city or a state adopts one model code or the other, it means that thousands of code officials, architects, engineers and others must purchase new copies to keep on their desks. In 2004, NFPA pulled in $58 million from publication sales, while ICC earned $25 million.[35]

The ICC, which emerged through consolidation of various regional model-code bodies, most notably publishes the International Codes (“I-Codes”), which serve as the baseline building regulations for most U.S. jurisdictions.[36] The NFPA, founded in 1896, develops hundreds of safety standards, including NFPA 70 (National Electrical Code) and NFPA 101 (Life Safety Code), which are widely adopted by reference.[37] Once enacted, the ICC’s I-Codes and the NFPA codes are sold at textbook-level price points and supplemented by paid digital subscriptions that offer broader functionality.[38]

The lobbying model can create tension because it sits at the fault line between private copyright enforcement and the public’s right to know the law. As we discuss in the next section, courts have been forced to confront the economic and legal consequences of allowing private SDOs to both press for their texts to become legally binding and then restrict access to those same texts once enacted.

C. Relevant Case Law

The overlap between voluntary-adoption and lobbying-oriented SDOs becomes most visible in the courts. Whatever differences exist in their business models, both types of SDO ultimately confront the same question: what becomes of copyright once private standards are incorporated into binding law? Judicial treatment of this issue has been uneven across circuits and contexts, but it reflects a common baseline principle—citizens must be able to access the laws that govern them, even when those laws originate as privately authored texts.

Early decisions reflected a strong inclination to preserve copyright protection, notwithstanding incorporation. In Practice Management v. AMA, the 9th U.S. Circuit Court of Appeals upheld copyright in the American Medical Association’s (AMA) Current Procedural Terminology (CPT) coding system, despite its incorporation into Medicare regulations, as the court reasoned that government reference did not divest the AMA of authorship or originality.[39] Similarly, in CCC Information Services v. Maclean Hunter, the 2nd U.S. Circuit Court of Appeals confirmed copyright protection for an automobile-valuation guide, even though state law required insurers to use it.[40] These cases underscored the view that mere legal reliance on a work did not erase the statutory protections conferred by the Copyright Act—a view congenial to SDOs that depend on copyright revenues to sustain costly consensus-driven technical work.

The 5th U.S. Circuit Court of Appeals, however, took a different approach in Veeck v. SBCCI, where a citizen posted online the building codes of two Texas towns that had adopted a model code verbatim.[41] Sitting en banc, the court held that, once a model code is enacted as municipal law, it loses copyright protection under both the government-edicts doctrine and the merger doctrine.[42] The court reasoned that “public ownership of the law means precisely that ‘the law’ is in the ‘public domain’ for whatever use the citizens choose to make of it,” emphasizing that citizens must have free access to the rules that bind them.[43] The circuit court’s decision was rooted in the idea that, once enactment renders a standard a non-substitutable public good, exclusionary rights function as a toll on compliance, rather than as an incentive for creation.

Importantly, the Veeck court was careful to limit its holding to the wholesale adoption of model codes as law, noting that the case did not involve extrinsic technical standards. This distinguishes that opinion from cases that uphold copyright protection for incorporated technical works.[44] Veeck thus struck most directly at lobbying-oriented SDOs, whose revenue model depends on wholesale enactment of their codes. When enactment itself triggers loss of copyright, it undercuts the very mechanism on which they rely to convert technical documents into guaranteed markets. By contrast, voluntary-adoption SDOs generally see their standards incorporated by reference, rather than enacted wholesale, and thus remain better positioned to preserve copyright protection under prevailing doctrine.

This difference is not merely procedural: when a standard is incorporated by reference, it remains a privately authored work that is used by law but not transformed into law, allowing copyright to persist under, for example, the “reasonable availability” regime established by OMB Circular A-119. Wholesale adoption, however, turns the text into the law itself, triggering the “law is public domain” problem at the heart of Veeck. As discussed below, the PRO Codes Act threatens to collapse this distinction, treating both models as functionally equivalent and thereby extending statutory protection to the lobbying-oriented model, while destabilizing the voluntary-adoption one.

Subsequent cases have, however, shifted away from the government-edicts doctrine and instead relied on fair use to reconcile public access with copyright. In American Society for Testing & Materials v. Public.Resource.Org, the U.S. Circuit Court for the D.C. Circuit upheld nonprofit republication of standards incorporated by reference into law, reasoning that such dissemination was transformative: while SDOs published standards for commercial sale, Public.Resource.Org republished them to educate the public about legal obligations.[45] The court also found the “reading rooms” that SDOs maintained to be inadequate, noting they often barred printing, downloading, or robust searching.[46] Doctrinally, this fair-use holding applied across the board—sweeping in both lobbying-oriented and voluntary-adoption SDOs alike.

The ICC’s litigation against UpCodes, a startup offering searchable online building codes, pushed these tensions into the commercial realm. The district court echoed Veeck in holding that private entities cannot wield copyright to bar access to enacted law, while leaving unresolved questions about non-enacted provisions and advertising claims. The case highlights the remaining gray area: nonprofit republication for educational purposes has been immunized, while commercial republication is still subject to closer scrutiny, with courts weighing the public benefits of wider access against potential market substitution.

These cases did not emerge in a vacuum. The most acute disputes, such as Veeck and UpCodes, arose from the distinctive incentives of lobbying-oriented SDOs, which campaign for wholesale enactment of their codes and then invoke copyright to meter access once those codes have become law. That strategy created the scarcity dynamic that brought these conflicts into the courts. Voluntary-adoption SDOs like ASTM International have been swept into the same doctrinal current—most notably in Public.Resource.Org, where fair use applied to their standards incorporated by reference—but their exposure reflects the breadth of the doctrine, rather than the business model they pursued. The underlying political pressure for legislation such as the PRO Codes Act seems to stem far more from the lobbying-oriented posture of ICC and NFPA.

Moreover, this jurisprudence is critical to understanding the stakes of the PRO Codes Act. Courts have already developed tools to balance access and incentives, allowing standards to remain protected as intellectual property, while ensuring that the public is not priced out of compliance. The PRO Codes Act, by contrast, threatens to displace these doctrines with a blunt statutory rule that collapses distinctions among SDOs and risks destabilizing copyright protections across the board. The next section turns to that legislation, analyzing its likely effects on both voluntary-adoption and lobbying-driven models of standards development.

IV. The Law & Economics of the PRO Codes Act

Into this complicated case law, Congress is considering injecting the PRO Codes Act. Framed as a transparency measure, its core provision declares that, once a private standard is incorporated by reference into law, it must be posted online for free.[47] If the standards body fails to provide that access, it will lose its copyright in the work.[48]

Supporters argue that this rule addresses a basic democratic concern: citizens should not have to pay, or navigate paywalled reading rooms, to learn the rules that govern them.[49] For proponents, the problem is straightforward—codes that function as law are often accessible only through expensive books or restricted digital portals. By forcing free posting, the bill purports to guarantee that binding legal obligations are accessible to all.

At first glance, the policy appeal is strong. Courts in cases like Veeck and Public.Resource.Org have reasoned that, once private standards become law, restricting access through paywalls or high-friction portals imposes deadweight losses by charging tolls for access to a non-substitutable public good.

Yet the PRO Codes Act addresses this problem in a way that risks undermining copyright more broadly. By conditioning copyright protection on whether a standard has been incorporated by reference, the bill effectively converts government adoption into a trigger for compulsory licensing at zero price. This approach does not simply guarantee access; it extinguishes property rights whenever the state adopts a private work. That is to say, a third party gets to exercise veto power over an SDO’s property rights.

Critically, entities like ICC and NFPA are just as entitled as any other standards body to enforce their copyrights. This approach to funding their operations is not per se objectionable, even as litigation has challenged whether incorporated statutes should or should not qualify for a fair-use exception. What is relevant here is how the lobbying-oriented model appears to foment many of the legal crises to which the PRO Codes Act seeks to respond. To the extent there is a problem the courts have not addressed, the fix is not to create a major exception to copyright. It should be to examine the activity that generates such problems and tailor any appropriate responses to that activity. Here, it is likely the combination of political adoption and subsequent reliance on exclusivity that makes the lobbying-oriented model uniquely open to criticism.

Further, the design of the PRO Codes Act could disproportionately harm voluntary-adoption standards bodies, which do not lobby for their codes to be adopted wholesale, but whose works are frequently referenced in federal and state regulations—access to which is very likely already covered under fair use. Unlike lobbying-oriented SDOs, who seek enactment as part of their business model, voluntary-adoption SDOs have little control over whether their codes are cited by regulatory bodies. Under the PRO Codes Act, such adoption would effectively eliminate the SDOs’ copyrights, erasing the very revenue stream that sustains the cross-subsidized development of technical safety standards.

The economic consequences of such a shift are significant. If SDOs cannot rely on publication revenues, many would likely have to scale back or abandon development of standards that lack commercial scale. Government bodies could, in theory, step in with funding, but doing so would require major new appropriations and would risk politicizing technical work now performed through professional consensus.

The U.S. advantage in global standards setting has long rested on a decentralized ecosystem of private SDOs, drawing on volunteer expertise and sustained by intellectual property. Weakening the property rights that support this ecosystem would risk slowing code updates, reducing technical quality, and further risk ceding international leadership to rivals. China’s “Standards 2035” plan, for example, explicitly aims to supplant U.S. and European standards as global benchmarks.[50] A policy that destabilizes U.S. SDOs could inadvertently accelerate that shift.

The PRO Codes Act also raises a more general property-rights concern. Once Congress sets the precedent that state action extinguishes copyright, the principle could spread beyond standards. If textbooks, software, or other works used by government (or, indeed, other “deserving” third parties) were deemed incorporated into a work of “public concern,” would they also lose protection?

The PRO Codes Act’s narrow focus on civil-engineering codes masks the breadth of its doctrinal innovation. It would collapse the distinction between using a work in governance and expropriating it into the public domain. This is a distortion ultimately caused by using one particular business model as the spark to restructure copyright itself.

The PRO Codes Act would fundamentally alter the current balance of law and directly undermine the business models of the voluntary-adoption SDOs. Under existing doctrine, when a voluntary-consensus standard is incorporated by reference, it does not automatically fall into the public domain. Courts have occasionally curtailed copyright (for example, by recognizing nonprofit republication as fair use), but these rulings have been case-by-case and have left intact the broader rights of standards bodies outside those narrow contexts. Agencies guided by OMB Circular A-119 must ensure “reasonable availability,” a requirement that can be satisfied through limited reading rooms or online portals. This approach permits SDOs to continue selling full copies—which are searchable, printable, and portable—while meeting their public-access obligations.

The PRO Codes Act would replace that flexible system with a rigid statutory rule and would, in essence, codify the lobbying-oriented SDOs’ business model. Once a standard is incorporated by reference, it would have to be posted online for free. If the SDO failed to do so, it would lose its copyright altogether. This turns copyright from a stable property right into a contingent interest revocable by government action.

Even compliance would bring risks. By mandating open online posting, the bill strips SDOs of control over how their works are made available and threatens to erode revenue streams that depend on users purchasing official editions. In short, the bill destabilizes SDOs’ rights, undermines their ability to sustain their business model, and exposes them to uncompensated forfeiture of the intellectual property on which their work depends.

A more balanced policy response would start from the premise that copyright is worth preserving as the foundation of creative and technical enterprise. The problem with the lobbying-oriented model is not copyright per se, but the way it is coupled with lobbying for wholesale statutory adoption. The PRO Codes Act mistakenly attacks copyright across the board, rather than targeting the lobbying-driven dynamics that generate actual problems in public access. In doing so, it threatens voluntary-adoption organizations that exemplify the public-interest rationale for copyright protection.

It is likely that the best approach to this issue is to do nothing, leaving courts to continue calibrating the balance through doctrines like fair use and the government-edicts rule. That approach already ensures nonprofit republication and public-domain status where texts are enacted verbatim, while leaving SDOs’ revenue streams intact. In short, Congress should reject statutory “fair use” expansion in the form proposed by the PRO Codes Act. Codifying a blanket rule that incorporation extinguishes copyright unless the rightsholder gives away his property would put at odds two distinct issues: ensuring access to the law and preserving the economic incentives that sustain the private development of standards.

V. Conclusion

Copyright undergirds the distinct but fragile business models that sustain standards development. For voluntary-adoption SDOs, copyright revenues provide the cross-subsidies needed to maintain a vast portfolio of technical rules, most of which would never be commercially viable on their own. For lobbying-oriented SDOs, copyright plays a different role, functioning as part of a two-stage maneuver, alongside lobbying to secure wholesale adoption. Both models illustrate how property rights serve as the backbone of the standards ecosystem, even as they invite scrutiny from courts and policymakers.

The PRO Codes Act threatens to upend this balance by introducing compulsory licensing by stealth. By tying copyright to government incorporation, the bill does not merely guarantee access but distorts the structure of property rights, creating a precedent that could undermine innovation, erode U.S. competitiveness, and unsettle the careful balance between public access and private incentive. Framed as a transparency measure, it risks entrenching one contested business model, while destabilizing the broader ecosystem of standards development.

The economic stakes are clear. SDOs provide technical infrastructure on a scale and at a level of expertise that government bodies could not replicate without extraordinary cost. Their continued viability depends on predictable intellectual property protections. Undermining those rights in the name of transparency would erode the capacity for innovation and distort the broader architecture of property law.

In short, the PRO Codes Act is a miscalibrated response. It overcorrects problems created by lobbying-driven SDOs and, in doing so, threatens the sustainability of voluntary-adoption SDOs. The better path is not to abandon copyright, but to preserve and refine the flexible tools already available in law—fair use, licensing, and targeted judicial scrutiny. Those mechanisms have long balanced access with incentives, and they remain far better suited to the complexities of standards development than the blunt instrument of statutory compulsory licensing.

[1] Pro Codes Act, H.R. 4009, 119th Cong. (2025), https://www.congress.gov/bill/119th-congress/house-bill/4009.

[2] Press Release, Issa, Ross Introduce Legislation to Protect Public Access to Federal Rules, Regulations, Off. Rep. Darrell Issa (Jun. 23, 2025), available at https://issa.house.gov/media/press-releases/issa-ross-introduce-legislation-protect-public-access-federal-rules.

[3] See Copyright Alliance Issues Statement Supporting HJC Markup of Pro Codes Act, Copyright All. (Apr. 17, 2024), https://copyrightalliance.org/press-releases/pro-codes-act.

[4] See Pro Codes Act, UpCodes https://up.codes/pro-codes-act (last visited Oct. 2025); The Real Truth About the “Pro Codes” Act, Ass’n Res. Libs. (Mar. 22, 2024), available at https://www.arl.org/wp-content/uploads/2024/03/The-Real-Truth-about-the-Pro-Codes-Act.pdf.

[5] The equitable doctrine of “unclean hands” prevents a party from obtaining judicial relief when its own conduct has been unethical or in bad faith in relation to the matter at issue. See, e.g., Restatement (Second) of Torts § 940 (1979). This is not to suggest that enforcing copyrights constitutes an improper manipulation of legal process, but rather that invoking property rights as part of a strategy that leverages public authority to create a captive market raises analogous concerns.

[6] Issa, supra note 2.

[7] UpCodes, supra note 4.

[8] Canyon Brimhall, The Pro Codes Act Attempts to Copyright the Law – But Is It Constitutional?, R Street Inst. (Mar. 2024), https://www.rstreet.org/commentary/the-pro-codes-act-attempts-to-copyright-the-law-but-is-it-constitutional.

[9] ARL, supra note 4.

[10] Copyright Alliance Applauds Introduction of the Pro Codes Act, Copyright All. (Mar. 2022), https://copyrightalliance.org/press-releases/pro-codes-act.

[11] FAA, Voluntary Industry Standards and their Relationship to Government, Fed. Av. Admin. (1996), at 27, available at http://www.faa.gov/about/office_org/headquarters_offices/ast/media/vol_std.pdf.

[12] Id.

[13] Id. at 28.

[14] See, e.g., Frequently Asked Questions, ASTM Int’l, https://www.astm.org/faq (last visited Oct. 2025).

[15] Codes & Standards Development, Amer. Soc. Mech. Eng’rs., https://www.asme.org/codes-standards (last visited Oct. 2025).

[16] About ASTM International, ASTM Int’l, https://www.astm.org/about (last visited Oct. 2025).

[17] National Technology Transfer and Advancement Act of 1995, Public Law 104-113 (1995), https://www.nist.gov/standardsgov/national-technology-transfer-and-advancement-act-1995.

[18] OMB Circular A-119, Off. Mgmt. Bdgt. (2016), available at https://share.ansi.org/Shared%20Documents/About%20ANSI/Why-Voluntary-Consensus-Standards-Incorporated-by-Reference-into-Federal%20Government%20-Regulations-Are-Copyright-Protected.pdf

[19] Voluntary Industry Standards and Their Relationship to Government, Fed. Av. Admin. Off. Com. Space Transp. (FAA/AST) (1996) at 27, 42, available at http://www.faa.gov/about/office_org/headquarters_offices/ast/media/vol_std.pdf.

[20] S. Michael Gentine & Kelsie Sicinski, D.C. Circuit: “Fair Use” Means Sharing Safety Standards Incorporated by CPSC Is Fair Game, Arnold & Porter Kaye Scholer LLP (Oct. 30, 2023, https://www.arnoldporter.com/en/perspectives/blogs/consumer-products-and-retail-navigator/2023/10/dc-circuit-fair-use-means-sharing-safety-standards.

[21] IEEE 802 LAN/MAN Standards Committee (LMSC), Inst. Elec. Electronics Eng’rs., https://www.ieee802.org (last visited Oct. 9, 2025); The Evolution of Wi-Fi Technology and Standards, IEEE Standards Ass’n., https://standards.ieee.org/beyond-standards/the-evolution-of-wi-fi-technology-and-standards (last visited Oct. 9, 2025).

[22] Order 8110.116, Fed. Av. Admin. (Sep. 23, 2011), available at https://www.faa.gov/documentLibrary/media/Order/8110.116.pdf.

[23] 2022 ACI 318-19 Building Code Requirements for Structural Concrete and Commentary on Building Code Requirements for Structural Concrete (ACI 318R-19), Int’l Code Council, https://codes.iccsafe.org/content/ACI31819R2022P2 (last visited Oct. 9, 2025); Jack Moehle, Introducing ACI 318-19: Building Code Requirements for Structural Concrete, B. Safety J. (Aug. 26, 2019), https://www.iccsafe.org/building-safety-journal/bsj-technical/introducing-aci-318-19-building-code-requirements-for-structural-concrete.

[24] FAA/AST, supra note 19, at 42.

[25] Copyright in Standards Incorporated by Reference into Law and the Pro Codes Act, CRS Product No. R47656, Cong. Res. Svcs. (Aug. 8, 2025); Am. Soc’y for Testing & Materials v. Public.Resource.Org, Inc.., 896 F.3d 437 (D.C. Cir. 2018).

[26] Veeck v. Southern Building Code Congress Int’l, Inc., 293 F.3d 791 (5th Cir. 2002).

[27] See generally John Burroughs, Incorporated Standards and Fair Use: Remaining Uncertainties Set the Stage for Further Litigation, Univ. Chi. Bus. L. Rev. Online Edition (2024), https://businesslawreview.uchicago.edu/online-archive/incorporated-standards-and-fair-use-remaining-uncertainties-set-stage-further.

[28] Why Voluntary Consensus Standards Incorporated by Reference into Federal Government Regulations Are Copyright Protected, Amer. Nat’l. Standards Inst. (n.d.) at 13, available at https://share.ansi.org/Shared%20Documents/About%20ANSI/Why-Voluntary-Consensus-Standards-Incorporated-by-Reference-into-Federal%20Government%20-Regulations-Are-Copyright-Protected.pdf.

[29] Updating OSHA Standards Based on National Consensus Standards; Personal Protective Equipment, 74 Fed. Reg. 46,350 (Sep. 9, 2009), https://www.osha.gov/laws-regs/federalregister/2009-09-09.

[30] See discussion supra notes 15-20 and accompanying text.

[31] Advocacy – Government Relations, Int’l Code Council, https://www.iccsafe.org/advocacy (last visited Oct. 2025).

[32] Public Policy and Advocacy, Nat’l. Fire Prot. Ass’n., https://www.nfpa.org/advocacy (last visited Oct. 2025).

[33] Sarah Harney, The Code War, Governing (Aug. 11, 2010), https://www.governing.com/archive/Code-War.html.

[34] Id.; Protect Safety Codes by Supporting Advancement of the Pro Codes Act, Int’l Code Council (2024), https://www.iccsafe.org/building-safety-journal/bsj-news/protect-safety-codes-by-supporting-advancement-of-the-pro-codes-act-2; Free Access to NFPA Codes and Standards, Int’l Ass’n Fire Chiefs, https://www.iafc.org/topics-and-tools/resources/resource/free-access-to-nfpa-codes-and-standards (last visited Oct. 2025).

[35] See Harney, supra note 25.

[36] Id.

[37] James Peterkin & Scott Mason, Navigating IBC and NFPA Differences, Health Facilities Mgmt. (Jul. 14, 2022), https://www.hfmmagazine.com/articles/4500-navigating-ibc-and-nfpa-differences.

[38] International Building Code®, Int’l Code Council https://shop.iccsafe.org/2021-international-building-coder.html (as of October 2025, ICC’s 2021 IBC listed at $203 in softcover, with a $112 license for digital content); NFPA LiNK, Nat’l. Fire Prot. Ass’n., https://www.nfpa.org/product/nfpa-70-national-electrical-code-nec/p0070code/nfpa-70-national-electrical-code-nec-2023/7023sb (as of October 2025, NFPA 70 (2023) was listed at about $161 in softcover, with an approximately a $168 annual subscription fee for a license).

[39] Practice Mgt. Info. Corp. v. American Med. Ass’n, 121 F.3d 516 (9th Cir. 1997), opinion amended by (9th Cir. 1998) 133 F3d 1140.

[40] CCC Info. Servs. v. Maclean Hunter Mkt. Reports, Inc., 44 F.3d 61, 74 (2nd Cir. 1994).

[41] Veeck v. SBCI, supra note 26.

[42] See id. at 801, 795.

[43] Id. at 799.

[44] Id. at 804.

[45] Am. Soc’y for Testing & Materials v. Public.Resource.Org, Inc., 82 F.4th 1262, 1265 (D.C. Cir. 2023).

[46] Id. at 1270.

[47] Pro Codes Act, supra note 1, at § 123(b).

[48] Id.

[49] See, e.g., Copyright Alliance, supra note 3.

[50] Yi Wu, China Standards 2035 Strategy: Recent Developments and Implications for Foreign Companies, China Br. (Jul. 26, 2022), https://www.china-briefing.com/news/china-standards-2035-strategy-recent-developments-and-their-implications-foreign-companies.

Executive Summary

This issue brief analyzes the antitrust implications of the proposed $34.5 billion merger between Charter Communications Inc. and Cox Communications Inc. from a law & economics perspective. We examine the proposed transaction under both a traditional consumer-welfare analysis, as well as the more recent “America First” flavor of analysis. The transaction would combine two major cable operators to create the largest broadband provider in the United States, with approximately 35.9 million residential and business broadband subscribers, and 69.5 million “passings” across 46 states.

The communications market is undergoing significant transformation, characterized by dynamic multi-platform competition from fixed-wireless access, satellite broadband, and streaming services, which exert considerable pressure on traditional wireline providers. The proposed merger is primarily a geographic expansion, rather than a horizontal consolidation within overlapping markets—a distinction critical for antitrust analysis.

From an antitrust standpoint, the parties project to achieve approximately $500 million in annual cost synergies within three years. These efficiencies are anticipated to translate into consumer benefits through potentially lower prices, enhanced service quality, and increased investment in advanced technologies and product innovation, particularly in mobile offerings.

Regulatory bodies, including the U.S. Justice Department (DOJ) and the Federal Communications Commission (FCC), will review the merger. The DOJ will apply the Clayton Act’s “substantial lessening of competition” standard, while the FCC will use its “public interest” framework. Given the minimal geographic overlap and robust multi-modal competition, the deal should not expect major regulatory obstacles on competition grounds. But concerns arising from broader policy agendas—such as diversity, equity, and inclusion (DEI), and the potential for regulatory overreach unrelated to demonstrable competitive harms—may arise.

This brief concludes that, whether evaluated under the traditional consumer-welfare standard or under the emerging “America First” framework, the Charter/Cox merger should be approved. Under either framework, the transaction reflects a strategic adaptation to market pressures, rather than an attempt to monopolize. Under conventional antitrust principles, the absence of significant geographic overlap and the presence of verifiable, merger-specific efficiencies weigh heavily in favor of clearance. Under the newer approach, the deal’s alignment with infrastructure-investment goals, enhanced competitiveness against dominant rivals, and commitments to expand broadband access all further strengthen the case for approval.

Policymakers should focus on demonstrable, transaction-specific competitive effects and adopt a technology-neutral approach to avoid hindering beneficial innovation and investment.

I. Introduction

Charter Communications Inc. and Cox Communications announced a proposed $34.5 billion agreement to combine in May 2025.[1] The combined company would surpass Comcast to become the largest broadband provider in the United States, with approximately 35.9 million residential and business broadband subscribers and 69.5 million passings across 46 states.[2] The company would also be the largest pay-TV provider in the United States, with approximately 14.6 million subscribers.[3] The merged company will be named Cox Communications, but Charter’s Spectrum brand will be retained for consumer services.[4]

This issue brief employs a law & economics approach to evaluate the antitrust and public-policy implications of the proposed merger.

• Section II examines the ongoing transformation of the U.S. communications marketplace, emphasizing the shift from legacy, technology-specific competition to a dynamic environment shaped by platform convergence, emerging technologies, and evolving consumer preferences.
• Section III outlines the analytical frameworks used by both the U.S. Justice Department (DOJ) and the Federal Communications Commission (FCC) in merger review, highlighting the distinct legal standards and policy priorities each agency would bring to its assessment of the Charter/Cox transaction.
• Section IV provides a detailed evaluation of the specific competition issues raised by the proposed merger, including analysis of geographic overlap, product-market definition, and claims of merger-related efficiencies.
• Section V reviews the outcomes and lessons from recent major communications and media merger cases, situating the Charter/Cox transaction within a broader historical and regulatory context.
• Section VI concludes by assessing the likely competitive effects of the merger and offering policy recommendations, with particular attention to the appropriate application of antitrust principles and the need for consistent, technology-neutral regulatory treatment across the sector.

II. The Evolving Communications Marketplace

The communications industry has undergone—and continues to undergo—substantial transformation, marked by technological convergence and erosion of the boundaries that once defined separate markets for television, internet, and telephone services. Consumers now move fluidly among platforms, accessing a growing array of services through broadband, mobile networks, satellite connections, and streaming applications. These shifts have redefined both the structure of competition and the criteria by which market power is assessed. To understand the implications of the Charter/Cox merger, it is essential to examine how these industry changes have altered the competitive landscape and diminished the relevance of legacy regulatory frameworks.

A. Beyond ‘Cable Operator’: The Multi-Platform Reality

To understand this merger’s competitive significance, it should be noted that the term “cable operator” has increasingly been rendered meaningless. The communications landscape of 2025 bears little resemblance to the era when most regulatory frameworks governing the sector were established. In the 1990s, broadcast television, cable TV, landline telephone, mobile telephone, and information services (e.g., internet access) were all distinct technologies with distinct economic markets. Over time, those distinctions have blurred and overlapped to the point where many consumers today do not care—or don’t even know—what technology underlies their access to the internet, video programming, and voice communications.

Around the time the Telecommunications Act of 1996 was passed, cable TV was considered a “bottleneck monopoly.”[5] Cable operators had local monopolies over cable service to households, as only “one percent of communities [were] served by more than one cable system.”[6] In 2013, then-Judge Brett Kavanaugh noted the dramatic increase in competition over the intervening years:

But in the 16 years since the last of those cases was decided, the video programming distribution market has changed dramatically, especially with the rapid growth of satellite and Internet providers. This Court has previously described the massive transformation, explaining that cable operators “no longer have the bottleneck power over programming that concerned the Congress in 1992.” Comcast Corp. v. FCC, 579 F.3d 1, 8 (D.C.Cir.2009); see also Cablevision Systems Corp. v. FCC, 597 F.3d 1306, 1324 (D.C.Cir.2010) (Kavanaugh, J., dissenting) (“This radically changed and highly competitive marketplace — where no cable operator exercises market power in the downstream or upstream markets and no national video programming network is so powerful as to dominate the programming market — completely eviscerates the justification we relied on in Time Warner for the ban on exclusive contracts.”)… In today’s highly competitive market, neither Comcast nor any other video programming distributor possesses market power in the national video programming distribution market.[7]

In a 2015 report and order, the FCC adopted a presumption that cable systems were no longer monopolies and are subject to effective competition.[8] The agency cited increased competition from satellite providers (e.g., DirecTV and DISH) and telephone companies (e.g., Verizon FiOS and AT&T U-verse) as evidence that most cable operators face effective competition. Under the adopted presumption, local authorities could no longer regulate basic-cable rates unless they successfully rebutted the presumption by proving a lack of effective competition.

In 1996, cable TV had 63.5 million subscribers (about two-thirds of TV households)[9] and the fastest home-internet connection was less than 56 Kbs.[10] Direct broadcast satellite was less than two years old[11] and video streaming was virtually unheard of, with RealPlayer being released less than a year earlier.[12] Netflix would launch its DVD-rental service in 1997 and YouTube was nearly a decade away.

Today, the average fixed-internet connection has download speeds of 288 Mbps.[13] Fewer than 50 million households have a “traditional” (e.g., cable or satellite) pay-TV subscription[14] (36% of households) and 55% are “streaming only.”[15] More than 200 streaming platforms are available[16] and 17.2 million homes subscribe to a virtual multichannel video programming distributor (vMVPDs, such as YouTube TV, Hulu+Live, SlingTV, and Fubo).[17] Streaming services account for 46% of household viewing time, with YouTube and Netflix together accounting for a little less than half of that total.[18]

B. Dynamic Competition in Broadband and Mobile

The broadband marketplace has undergone a rapid evolution in recent years.[19] New technologies like fixed wireless access (FWA) and low-earth-orbit (LEO) satellite services have significantly expanded internet access and fostered intermodal competition among providers.[20] For example, 5G FWA has grown into a disruptive competitor, with 77% of operator locations now having 5G, and the North American market for 5G FWA is projected to expand significantly by 2030.[21] Similarly, satellite broadband, offered through networks like Starlink, doubled its subscriber base in 2024, becoming a more viable option, especially for rural consumers.[22] This emergence of diverse platforms means traditional wireline providers—including cable operators—no longer compete in isolation but instead face multifaceted pressures from these new forms of connectivity.[23]

This dynamic competitive environment has yielded tangible benefits for consumers. Broadband speeds have risen consistently across the industry, while prices have generally fallen.[24] The FCC, for instance, last year upgraded its fixed-speed benchmark to 100/20 Mbps and set an “aspirational goal” of 1 Gbps/500 Mbps, reflecting the advancements in available speeds.[25] Furthermore, a growing number of households are now served by multiple broadband providers, enhancing consumer choice and competitive intensity.[26] Despite some critics’ claims of limited competition, statistics indicate that a vast majority of U.S. homes have access to at least three fixed, mobile, or satellite broadband services.[27]

Beyond these emerging technologies, competition from other established wireline providers remains significant. For instance, AT&T has been aggressively expanding its fiber footprint, with plans to pass over 30 million premises with fiber by the end of 2025, and potentially 60 million by 2030 through new partnerships and acquisitions.[28] Verizon also continues to be a major competitor with its fiber buildout plans.[29] These extensive fiber rollouts directly compete with cable broadband services, especially for high-speed connectivity, thus necessitating strategic responses from cable operators to maintain their market position.

The mobile market also already presents a critical dimension of competition, as well as a significant growth opportunity for broadband providers. Cable-wireless services, such as Charter’s Spectrum Mobile and Comcast’s Xfinity Mobile, have rapidly evolved into significant competitive forces in the mobile-telecommunications market, leveraging existing cable infrastructure and bundled service offerings to attract customers, often at lower prices. This exerts considerable competitive pressure on traditional wireless carriers like AT&T, Verizon, and T-Mobile.[30]

The proposed Charter/Cox merger aims to capitalize on these mobile opportunities. Cox, which launched its mobile service nationally in early 2023, is relatively “underpenetrated” in this market, with an estimated 200,000 mobile lines, compared to Charter’s 10.4 million.[31] The combined entity would have the opportunity to extend Charter’s more favorable mobile virtual network operator (MVNO) terms with Verizon to Cox’s customer base, offering lower-priced converged fixed and mobile offerings that tend to increase customer retention and improve economics.[32] This strategic move would bolster mobile competition within Cox’s legacy footprint, directly benefiting consumers with more competitive options.

C. The Imperative of Scale in Capital-Intensive Industries

The modern communications industry, encompassing broadband and mobile services, is characterized by its inherent capital intensity, demanding substantial and ongoing investments for technology deployment and network upgrades, as ICLE reported in a 2024 white paper:

Investment and innovation do not solely come from new entrants, as incumbents often are important sources of innovation while they try to stay competitive and avoid disruption. In this way, providers compete through new product introductions and disruption, not just on price. Because of these dynamics, mergers and increased concentration can sometimes be associated with increased investment, in that they may allow firms to achieve greater economies of scale and scope.[33]

Providers must continually invest in infrastructure, such as fiber and the latest Data Over Cable Service Interface Specification (DOCSIS) standards, to deliver ever-increasing broadband speeds and capabilities, as well as to expand and enhance mobile networks. These capital requirements mean that achieving greater scale is an economic imperative to enable more efficient investment and sustain competitiveness in a rapidly evolving landscape.

This need for scale is amplified by the dynamic nature of competition across various platforms and technologies. Cable operators like Charter and Cox no longer operate in a siloed market. They face significant pressure from diverse competitors, including expanding fiber footprints from AT&T and Verizon, disruptive FWA services from T-Mobile and Verizon, and LEO satellite-broadband providers like Starlink. In this multi-platform reality, achieving greater scale allows companies to allocate resources more effectively across various regions and technologies, ensuring they can keep pace with innovation and consumer demand.

A crucial area where increased scale offers tangible economic benefits is in MVNO agreements. Cable companies typically offer mobile services by leveraging existing wireless infrastructure through MVNO deals with national carriers like Verizon.[34] Charter, with its substantial mobile presence of 10.4 million wireless lines, has demonstrated that it can compete effectively in this space. By combining with Cox, whose mobile service is relatively underpenetrated with only 200,000 customers, the merged entity can extend Charter’s more advantageous MVNO terms with Verizon to Cox’s customer base.[35] This not only would offer lower-priced converged fixed and mobile offerings to Cox customers, but also bolster overall mobile competition within Cox’s existing footprint by providing a stronger alternative to traditional wireless carriers.

Beyond favorable MVNO terms, increased scale also contributes to improved marketing and branding capabilities. The combined company, with 69.5 million passings and 37.6 million customers, would become the largest broadband provider in the country, surpassing Comcast. This expanded footprint and customer base would enhance the combined company’s ability to compete against national competitors through more effective branding and sales efforts. The planned rollout of Spectrum-branded products across Cox’s service area, including Advanced Wi-Fi and Spectrum Mobile with Mobile Speed Boost, aims to leverage this expanded scale to drive sales and reduce customer churn.[36]

Crucially, the Charter/Cox merger is largely considered a geographic expansion, rather than a consolidation that eliminates head-to-head competition. Charter and Cox do not significantly overlap within their service footprints. This distinction is vital for antitrust analysis under Section 7 of the Clayton Act, which prohibits mergers that may substantially lessen competition. In this case, the transaction aims to achieve greater scale across different regions without directly reducing the number of competitors in any given local market. This approach is distinct from more problematic recent mergers, such as Comcast’s attempted acquisition of Time Warner Cable (discussed below), which faced significant antitrust opposition due to overlapping concerns.

III. Competitive Review Frameworks: DOJ and FCC

The review of the proposed Charter/Cox transaction requires navigating two distinct regulatory frameworks, each grounded in unique statutory mandates and institutional priorities. The DOJ evaluates mergers primarily through the lens of antitrust law, focusing narrowly on whether a transaction may substantially lessen competition or create a monopoly, as set forth in the Clayton Act. In contrast, the FCC undertakes a broader inquiry guided by the Communications Act’s “public interest” standard, which allows consideration of an array of economic, social, and policy factors beyond pure competition.

This dual-track review not only shapes the substantive analysis of competitive effects but also influences the inquiry’s procedures, evidentiary burdens, and types of conditions or remedies that may be imposed on the merging parties. The following sections describe the distinct standards and practices applied by each agency as they assess major communications-industry transactions.

A. The Clayton Act (DOJ) and the ‘Substantial Lessening of Competition’ Standard

Section 7 of the Clayton Act reflects congressional intent to arrest anticompetitive conduct before it reaches full monopolization by forbidding mergers whose effect “may be substantially to lessen competition, or to tend to create a monopoly.”[37] The U.S. Supreme Court has characterized this language as creating “a relatively expansive definition of antitrust liability.”[38]

The DOJ’s merger-review process operates through a burden-shifting framework established in cases like Philadelphia National Bank.[39] When the government establishes a prima facie case showing a substantial lessening of competition through structural presumptions or other evidence, defendants bear the burden of rebutting this showing with evidence demonstrating that the merger will not harm competition.[40] The structural presumption creates a rebuttable inference of illegality when mergers significantly increase concentration in highly concentrated markets, as measured by tools like the Herfindahl-Hirschman Index (HHI). The DOJ’s 2023 Merger Guidelines establish presumptions for mergers creating firms with market shares exceeding 30%, or HHI increases of more than 100 points in markets with post-merger HHI levels above 1,800.[41]

Economic analysis plays a central role in modern merger review. Its application, however, remains “surprisingly elusive,” as it is constrained by legal precedent and statutory interpretation.[42] While agencies employ sophisticated analytical techniques—including merger simulation, diversion analysis, and natural experiments to assess competitive effects—these economic tools are constrained by the legal framework established by Section 7’s text and judicial interpretation. The Clayton Act does not require certainty of harm but rather assessment of risk based on the totality of available evidence.

The distinction between different merger types proves crucial for understanding DOJ enforcement patterns. Geographic expansion mergers, where companies operate in different markets, typically receive more favorable treatment than horizontal mergers that eliminate head-to-head competition. For example, the DOJ recently approved T-Mobile’s transaction with UScellular, which had a significant geographic-extension element.[43] Such transactions do not eliminate existing competition between merging parties, reducing the immediate competitive concerns that trigger structural presumptions. Geographic expansion allows firms to extend their market reach and increase scale economies without directly consolidating market share in existing territories, distinguishing these transactions from mergers that combine direct competitors in the same geographic markets.[44]

B. The ‘Public Interest’ Standard (FCC)

The FCC’s review of the proposed merger will proceed on a separate and fundamentally different track from the antitrust analysis conducted by the DOJ. While both agencies scrutinize the transaction’s competitive effects, the FCC operates under the Communications Act’s broad and ill-defined “public interest, convenience, and necessity” standard, as explained in ICLE’s comments in the FCC’s “Delete, Delete, Delete” proceeding:

The FCC’s transaction-review process, rooted in the Communications Act of 1934, has evolved into a complex, time-consuming, and often unpredictable system that frequently duplicates efforts already undertaken by antitrust authorities. The current dual-review system imposes substantial costs on merging parties without clear commensurate benefits. While the FTC and U.S. Justice Department (DOJ) focus narrowly on demonstrable competitive harms, the FCC employs a broader and more ambiguous “public interest” standard that allows for wide-ranging inquiries, and often demands conditions that extend beyond competition concerns. This expansive approach not only creates regulatory uncertainty but also significantly increases transaction costs and extends timelines for business combinations that might otherwise benefit consumers through enhanced efficiencies and innovation.[45]

The FCC’s authority is triggered by the companies’ need to transfer control of their various FCC licenses—a procedural hook the agency has long used to conduct comprehensive merger reviews. The process is managed by a dedicated FCC transaction team that establishes a public docket, inviting comments and petitions to deny from third parties.[46] The FCC established its public docket in the Charter/Cox proceeding Sept. 5, 2025, with a public comment period set to close Oct. 5, 2025.[47]

This public phase is critical, as it allows competitors, consumer advocates, and other stakeholders to introduce arguments and evidence that shape the scope of the commission’s inquiry, often extending it far beyond the transaction’s direct competitive effects. The commission’s initial analytical step will be to define the relevant product and geographic markets to assess competitive harm.

It is likely the FCC will determine that the transaction is not a conventional horizontal merger, as the firms’ cable and broadband service territories have almost no geographic overlap. For example, when the commission recently approved Verizon’s acquisition of Frontier, the competitive analysis focused on whether the transacting parties “currently provide, or are likely to provide, products or services that consumers view as substitutes within the same relevant geographic market.”[48] In addition, where there is no geographic overlap, the commission has indicated that no analysis of horizontal effects is necessary.[49]

This finding of almost no geographical overlap is significant, not because it exonerates the merger, but because it dictates the subsequent analytical pivot. With minimal direct horizontal harm to analyze at the local retail level, the commission could shift focus to national-level, nonhorizontal theories of harm. This dynamic was central to the reviews of prior major cable mergers, such as the abandoned Comcast/Time Warner Cable transaction and the approved Charter/Time Warner Cable deal, where concerns about national scale—not local overlaps—dominated the proceedings.

It is this focus on national harms unrelated to competition issues where the FCC’s unique public-interest standard becomes most potent, and potentially capricious. Unlike the Clayton Act standard used by the DOJ, the FCC’s framework effectively reverses the burden of proof. The merging parties must affirmatively demonstrate that their combination will produce tangible public-interest benefits that outweigh any potential harms.[50] The term “public interest” is capacious, allowing the commission to consider a wide array of policy goals beyond economic efficiency, including the promotion of localism, programming diversity, and the deployment of advanced services. This means a transaction that is competitively neutral, or even one with harms that fall short of being “substantial,” could be blocked or conditioned if the parties fail to make a sufficiently compelling affirmative case.

The FCC’s approach is not a sterile application of economic theory, but a process heavily influenced by precedent and political pressure. The conditions imposed in the Comcast/NBCUniversal and Charter/Time Warner Cable mergers—as well as the T-Mobile/UScellular transaction and Skydance’s recent acquisition of Paramount—serve as precedent, and hint at the FCC’s approach to the Charter/Cox deal.

Arguments and demands submitted to the public docket by influential third parties will also play a crucial role in defining the scope of perceived harms, and shaping the ultimate remedy. The commission will weigh the potential harms—both those identified by its staff and those advanced by outside parties—against the procompetitive justifications offered by Charter and Cox.

History suggests that the FCC is unlikely to block the merger outright. Instead, the agency can “slow walk” the process or identify potential harm to the “public interest” as leverage to extract a set of “voluntary” conditions that the companies must accept to win approval. This practice of “regulation by adjudication” allows the FCC to impose its policy preferences on the merging parties without undergoing a formal rulemaking process. Moreover, such conditions are not subject to the same cost-benefit analysis or judicial scrutiny as formal rules,[51] giving the agency immense discretionary power.

The FCC’s typical regulatory approach has become even more strategically important for the commission in the current legal environment. The Supreme Court’s recent decision in Loper Bright to overturn the Chevron doctrine has weakened the legal foundation for broad agency rulemakings by eliminating judicial deference to an agency’s interpretation of its governing statutes.[52] With its general rulemaking authority now more vulnerable to legal challenge, the FCC has a powerful incentive to use the merger-review process as its preferred vehicle for regulation.

C. The Trump Administration’s Approach to Mergers

The Trump administration’s approach to merger review represents a departure from the approach of past Republican administrations, embodying what officials have variously christened “America First Antitrust,” “MAGA Antitrust,” and “Hillbilly Antitrust.”[53] The framework combines populist skepticism of corporate concentration with nationalist economic priorities, creating a more interventionist enforcement posture than previous Republican administrations.

The DOJ’s antitrust review would be conducted under Assistant U.S. Attorney General Gail Slater’s “America First Antitrust” doctrine, which prioritizes protecting “America’s forgotten consumers” through targeted litigation, rather than broad regulatory schemes.[54] This approach views antitrust enforcement as a tool that uses courts as a “scalpel” to excise monopolistic behavior without burdening entire industries with preemptive rules.[55] The DOJ has retained the Biden administration’s aggressive 2023 Merger Guidelines, which establish lower concentration thresholds and expanded theories of harm beyond traditional price effects, including labor-market impacts and potential competition concerns.

The administration’s enforcement philosophy explicitly rejects what it sees as the consumer-welfare standard’s narrow focus on price.[56] Federal Trade Commissioner Mark Meador, for example, argues that antitrust enforcement should focus on the welfare of consumers as buyers, not corporate efficiency gains, effectively excluding most merger-generated cost savings from competitive analysis unless directly passed to consumers.[57] This doctrinal shift enables challenges based on broader theories of harm, including monopsony power over suppliers, elimination of potential competition, and labor-market-concentration effects.

FCC Chair Brendan Carr’s “Build America Agenda” prioritizes infrastructure deployment, spectrum efficiency, and national technological leadership over traditional competition concerns.[58] This framework views telecommunications mergers through the lens of enhancing American competitiveness against foreign rivals and accelerating next-generation network deployment.

Carr has also introduced an unprecedented element to FCC merger review by threatening to block transactions involving companies that maintain what he terms “invidious” diversity, equity, and inclusion (DEI) programs.[59] This represents a novel use of the commission’s public-interest authority to advance broader administration policy objectives beyond traditional communications regulation. The FCC has already demonstrated this approach’s practical impact: T-Mobile was required to modify its DEI policies to secure approval for its Lumos Fiber acquisition,[60] as well as its transaction with UScellular.[61] Verizon received approval for its $20 billion Frontier purchase shortly after announcing the elimination of its DEI initiatives.[62] This creates a new category of merger condition, where companies must align their employment and corporate-governance practices with administration priorities to obtain FCC clearance.

The DEI enforcement mechanism expands the FCC’s leverage in merger negotiations beyond traditional public-interest considerations such as service quality, network investment, and competitive effects. Companies seeking merger approval must now evaluate whether their corporate policies on workplace diversity could trigger regulatory opposition, with the FCC effectively using the merger-review process as a vehicle to reshape private-sector employment practices. This approach transforms telecommunications merger review into a broader policy-enforcement tool that extends well beyond the communications sector’s traditional regulatory boundaries.

The precedent set by the T-Mobile/Sprint merger approval demonstrates how Trump administration officials resolve tensions between anti-concentration principles and national policy objectives. That transaction combined two of the four national wireless carriers, creating presumptive competitive harm, yet received approval based on complementary spectrum assets that would accelerate nationwide 5G deployment.[63] The agencies imposed strict build-out commitments and engineered complex structural remedies to preserve market structure while achieving infrastructure-policy goals.[64]

For the Charter/Cox transaction, the DOJ could focus on potential monopsony effects in content-purchasing markets or labor-market-concentration concerns. The combined entity would become the largest buyer of video programming, potentially enabling it to extract lower rates from content creators. But it should be noted that the companies’ “no overlap” defense—that they operate in distinct geographic markets—significantly complicates horizontal-merger analysis. The administration’s retained 2023 Merger Guidelines provide tools to challenge the merger on national-concentration theories, but litigation risks remain substantial, given the lack of direct competitive overlap.

Critically, the enforcement agencies would likely pursue a negotiated settlement rather, than outright challenge. This reflects some of the recent agreements noted above, where the administration was not interested in blocking deals, per se, but instead in using potential deals as leverage to extract concessions that advance broader policy objectives. For example, Slater’s pragmatic approach—demonstrated in approving the Capital One/Discover merger that the Biden administration reportedly planned to block—suggests a willingness to accept deals with appropriate conditions.[65] The Charter/Cox merger’s alignment with infrastructure-investment goals and its potential to create a stronger competitor to dominant players like Comcast and vertically integrated platforms like Amazon and Apple provide pro-competitive justifications that may reduce the attractiveness of litigation, even if other ancillary demands are pursued.

Further, the FCC review should present few obstacles, as the companies’ goal for the merger is to advance a plan that is highly consistent with the “Build America Agenda” through promised infrastructure investments and expanded broadband access.[66] The companies have strategically emphasized commitments to accelerate fiber deployment, enhance rural-broadband service, and repatriate customer-service jobs to the United States. These promises align with Carr’s public-interest priorities.

Following the T-Mobile-Sprint template of extracting enforceable commitments in exchange for merger clearance, conditional approval represents the most probable outcome. Likely conditions could include aggressive build-out timelines with penalty provisions, price-protection commitments for existing customers, enhanced mobile-competition requirements, and elimination of company DEI policies.

The administration’s merger-review process reflects its broader philosophy of using government power strategically, rather than reflexively opposing corporate consolidation. Even under the populist “America First” framework, the Charter/Cox merger’s geographic complementarity, infrastructure-investment commitments, and alignment with national-competitiveness goals position it favorably for approval with conditions that transform potential competitive concerns into vehicles to achieve the administration’s policy priorities.

D. Regulatory Asymmetry

The modern communications marketplace suffers from fundamental regulatory inconsistencies. Traditional cable operators pursuing horizontal mergers, such as Charter and Cox, face intensive scrutiny from both the DOJ under Section 7 of the Clayton Act and the FCC under its “public interest” standard. Meanwhile, technology giants have vertically integrated across similar competitive spaces with minimal oversight.[67] This asymmetry distorts market outcomes by imposing differential regulatory burdens, rather than allowing economic efficiency and consumer preference to determine competitive success.

The regulatory burden imposed on traditional cable operators creates substantial transaction costs and uncertainty. The dual-review process requires significant legal fees, economic-consulting expenses, and executive attention diverted from core operations.[68] The FCC’s “public interest” standard proves particularly problematic because it lacks clear boundaries. Unlike DOJ analysis focused on demonstrable competitive harms, the FCC may extract concessions on matters unrelated to competitive effects. This transforms merger review into de-facto rulemaking, creating a permission-based system that chills pro-competitive transactions and investment.

Technology platforms have assembled vertically integrated ecosystems without equivalent regulatory oversight. For example, Google distributes programming via YouTube and YouTube TV and operates fiber infrastructure in select markets. Amazon has become a major content producer through acquisitions like MGM and the company’s extensive investment in original programming. This content reaches consumers through Prime Video, which accounts for nearly 4% of household viewing time,[69] some of which is consumed on Amazon’s Fire TV devices. Netflix transformed from a content aggregator into a dominant production studio with an annual content budget of $18 billion, rivaling traditional media companies.[70] This vertical integration fundamentally altered video-programming-market dynamics, creating a powerful content buyer and direct competitor to programmers and distributors.

The disparate treatment across technologies lacks coherent consumer-welfare justification. Subjecting one set of firms to intensive and costly review while allowing others to integrate with minimal oversight does not protect competition, but skews it. The result creates structural disadvantages for traditional providers based on legacy regulations, rather than competitive merit. This asymmetry distorts competition, discourages investment where it is most needed, and fails to address the competition posed by vertically integrated technology platforms.

A coherent regulatory framework would move beyond historical technology-driven distinctions to apply technology-neutral principles that reflect consumer behavior. Without such modernization, antitrust enforcement becomes arbitrary, penalizing incumbents for attempting to achieve scale and scope that unregulated rivals have already attained. Sound economic policy requires consistent treatment of similar competitive behaviors regardless of corporate heritage or regulatory classification.

IV. Potential Competition Issues in the Charter/Cox Merger

The competitive effects of the proposed Charter/Cox merger are best understood within the broader context of today’s rapidly evolving communications market. Effective analysis requires a careful examination of how the merger might affect competition across both fixed broadband and video services, considering the intersection of overlapping and non-overlapping geographic footprints, emerging technologies, and shifting consumer-demand patterns. This section assesses potential competition concerns by focusing first on the fixed and mobile broadband markets, where intermodal rivalry and expanded provider choice have redefined competitive dynamics, and then on the implications for video distribution in light of changing market definitions and increasing cross-platform convergence.

A. Fixed and Mobile-Wireless Broadband

Antitrust analysis of this transaction will likely focus on the fixed broadband market, as the deal would create the largest broadband provider in the United States. Even so, there is extraordinarily little geographic overlap between the areas served by Charter and Cox. Less than 0.1% of broadband serviceable locations in the combined footprints are served by both companies.[71] In addition, more than half of Charter locations and about half of Cox locations already are served by competing fiber providers.[72]

ICLE reported in 2024 that broadband competition is intense and dynamic.[73] Since the COVID-19 pandemic, more households are connected to the internet; broadband speeds have increased, while prices have fallen; more households are served by multiple providers; and newer technologies like satellite and 5G have expanded internet access and intermodal competition among providers. For example, from the end of 2019 through June 2025, while the overall Consumer Price Index increased by 24.3%, the index for internet services and electronic-information providers rose by just 9.0%.[74]

The International Technology & Innovation Foundation (ITIF) similarly reports that cable companies face increasing competition, such that “cable’s long-time dominance [in broadband] is fading due to new alternatives entering the market or existing companies expanding their footprint and capacity.”[75] For example, in the second quarter of 2025, Spectrum’s tally of internet customers fell by 117,000.[76]

From 2019 to 2024, the total number of subscribers to fixed wireless-access products from AT&T, T-Mobile, Verizon, and UScellular grew from less than 1 million to nearly 12 million, amounting to a 68% cumulative annual growth rate.[77] In the first quarter of 2025, 12.7 million subscribed to FWA.[78] At a March 2025 event, FCC Chairman Brendan Carr noted that “the evidence shows that when we get a new fixed wireless provider that comes in… prices decrease.”[79] In addition, Starlink’s satellite-broadband service has more than 2 million “active” U.S. subscribers,[80] while reaching 99.8% of broadband-serviceable locations in the United States. Another satellite service, Amazon’s Project Kuiper, plans to offer internet service with speeds of 100 Mbps to 1 Gbps by the end of 2025.[81]

An ICLE issue brief on T-Mobile’s transaction with UScellular reported that mobile-wireless services have been a growing segment of the mobile broadband market, with cable companies like Comcast and Charter exerting competitive pressure:

In less than a decade, cable-wireless providers have rapidly evolved from nascent entrants to significant competitive forces in the mobile-telecommunications market. Their market share has grown substantially, with recent estimates indicating they’ve captured more than half of new subscribers. This growth demonstrates that consumers view cable-wireless offerings as viable substitutes for traditional wireless services. Moreover, cable-wireless providers now cover a significant portion of the population, including within UScellular’s footprint. This indicates that their services are widely available, and they compete directly with traditional carriers in many local markets.[82]

Despite the growth of cable wireless, however, the largest providers combined still likely account for less than 5% of U.S. mobile subscribers.[83]

B. Video Services

For video services, a market definition narrowly confined to traditional MVPDs, such as cable and satellite providers, is economically indefensible and detached from current market realities. Consumers have demonstrated that they view traditional MVPDs as competitive substitutes with the vast array of over-the-top (OTT) services, including subscription video-on-demand (SVOD), advertising-supported video-on-demand (AVOD), and virtual MVPDs (vMVPDs).[84] “Cord-cutting,” the persistent, large-scale consumer migration from traditional pay-TV to these internet-delivered alternatives provides evidence of high demand-side substitutability and cross-elasticity of demand, the foundational criteria for market definition under established legal and economic principles.

Moreover, as noted above, the service footprints of the two cable operators are geographically distinct and do not overlap. A consumer in a Charter territory cannot choose Cox for cable service, and vice versa. Consequently, the transaction does not eliminate head-to-head competition between the two firms in the provision of their primary video offerings. This fact alone distinguishes the transaction from mergers that consolidate rivals within the same geographic area and immediately lessens the potential for the kinds of unilateral or coordinated anticompetitive effects that typically animate antitrust concern. The merger is a combination of two sets of non-competing local assets.

The U.S. video market is a fragmented and fiercely competitive space dominated by numerous well-capitalized global and national firms. For example, based on total television viewing time, YouTube and Netflix combined account for almost the same amount of household viewing time (21.1%) as all cable operators combined (23.4%).[85] Even in a narrowly defined “pay TV” market comprised of cable, satellite, and vMVPD providers, the merged company would have a market share of 23% of subscribers. This results in an HHI of 1,502, well below the 2023 Merger Guidelines thresholds for a highly concentrated market.[86]

The competitive discipline in the modern video market is further amplified by exceptionally low consumer-switching costs. Unlike the legacy cable model, which often involved annual contracts, professional installation, and equipment rental, consumers can subscribe to or cancel most OTT services in minutes with no long-term commitment. This consumer empowerment creates significant market fluidity, evidenced by high churn rates for SVOD services, which have grown substantially since 2019. The constant threat that a subscriber will switch to a rival service in response to a price increase or degradation in quality or content selection imposes a powerful constraint on the behavior of all market participants, including a post-merger Charter/Cox.

The transaction is thus best understood not as an anticompetitive move to consolidate market power, but as a pro-competitive response to growing competition from vertically integrated media conglomerates and global technology platforms. Companies like Disney, Warner Bros. Discovery, Google (YouTube), and Amazon leverage immense scale in both content production and global distribution, creating competitive challenges for regional distributors. By combining assets, Charter and Cox can achieve the necessary scale to negotiate for programming on more favorable terms, creating efficiencies that can be passed on to consumers. This rationale mirrors the logic accepted in prior vertical media mergers, such as AT&T’s acquisition of Time Warner, which was positioned as necessary to compete with the rising power of large technology firms in content distribution.

Consumer group Public Knowledge has expressed concerns that a larger combined distributor might use its position to foreclose independent programmers or limit content diversity.[87] Such concerns are misplaced in the current video-distribution market. To compete effectively against the vast and exclusive content libraries of services like Netflix, Disney+, and HBO Max, video distributors have a powerful commercial incentive to cost-effectively offer the most compelling and diverse array of programming possible in order to attract and retain subscribers.

Any attempts to limit choice would be self-defeating, as consumers would quickly migrate to other platforms. Moreover, programmers today have more routes to market than at any point in history. They are no longer dependent on a handful of cable and satellite operators for distribution and can reach audiences directly through their own apps or via partnerships with numerous national and global OTT services.

C. Efficiencies and Consumer Benefits

Unlike previous failed consolidation attempts in the cable industry—most notably, Comcast’s 2015 bid for Time Warner Cable—the Charter/Cox proposal combines largely non-overlapping geographic footprints. This distinction is crucial for antitrust analysis. As noted above, rather than eliminating head-to-head competition in local markets, this transaction resembles a geographic expansion that allows the combined company to achieve greater scale across different regions.

This scale matters for reasons beyond market share. The communications industry requires enormous capital investments to deploy advanced technologies. USTelecom estimates broadband investments have averaged $100 billion a year (inflation adjusted) over the past decade.[88] Charter and Cox project approximately $500 million in annualized cost savings within three years of the deal.[89] If realized, these savings could be invested in network upgrades (e.g., expanding gigabit and multi-gigabit capabilities, and accelerating deployment of the DOCSIS 4.0 internet-communications standard); product-offering innovations (such as converged mobile and broadband bundles, where a larger entity might secure better MVNO terms); and improved customer service.

In addition, the companies claim the merged firm’s increased scale will result in lower costs per passing or per customer.[90] Their consultants also claim the merger will result in lower programming-acquisition costs, but have not provided sufficient information to the public to evaluate the claim.[91]

Offloading is the practice of routing mobile-data traffic over Wi-Fi networks instead of traditional cellular networks. Charter claims that 87% of its Spectrum Mobile traffic flows over Wi-Fi access points.[92] The merger will increase the number of Wi-Fi access points available to support the company’s mobile services, increasing the availability to offload mobile traffic on to Wi-Fi and thereby reducing the amount of network access purchased through MVNO agreements.[93]

Charter and Cox claim the merger would create cost savings by eliminating Cox’s current syndication agreements regarding consumer premises equipment (CPE) and video services. Currently, Cox has a syndication agreement with a third party (reportedly Comcast) to provide equipment.[94] Charter has developed its own customer equipment for modems and routers. After the merger, Cox customers will transition to Charter’s equipment platform, removing the syndication fees Cox currently pays. The parties argue that these savings could then be passed on to customers through lower prices or better service quality.[95] These characteristics not only mitigate concerns under the traditional Clayton Act analysis, but also match the priorities emphasized by the “America First Antitrust” approach, which favors transactions that expand infrastructure, increase competitiveness, and deliver direct consumer benefits.

LightReading reports that Cox has currently uses the X1 and Xumo video platform through a syndication deal.[96] Charter and Cox’s consultants conclude this results in “double marginalization” in which the upstream technology provider sets syndication fees above its own marginal costs, and Cox, in turn, prices its video service above these syndicated input costs to earn a return.[97] Charter plans to move Cox customers to its own video platform, which bundles traditional TV with free ad-supported versions of popular streaming services like Disney+, HBO Max, and Paramount+.[98] This platform also includes a digital store where customers can easily add streaming apps and upgrade to ad-free versions if they want.

Charter argues that its larger scale after acquiring Cox will reduce equipment and service costs significantly. The company says it will reinvest these savings into network improvements and new services for customers. By bringing Cox’s operations in-house instead of relying on third-party syndication agreements, the merged company expects to operate more efficiently and offer customers better value through improved bundling options and lower operational costs.

V. Lessons from Past Merger Reviews

The regulatory and competitive landscape surrounding large-scale communications mergers has evolved significantly over the past decade, shaped both by shifting market dynamics and by changing enforcement priorities. To understand the likely trajectory and evaluation of the proposed Charter/Cox transaction more fully, it is instructive to examine recent precedent in U.S. telecommunications and media mergers. This section examines the Comcast/Time Warner Cable and AT&T/Time Warner merger reviews and draws out key analytical lessons from their outcomes and implications for future merger reviews.

A. Comcast/Time Warner Cable (2015)

In 2015, Comcast abandoned its plans to acquire Time Warner Cable Inc. (TWC) for approximately $45.2 billion.[99] The deal was scuttled following formal notification from both the DOJ and the FCC that their staffs intended to oppose the deal. The regulators argued that the combined entity’s national scale would make it an “unavoidable gatekeeper” with the ability and incentive to harm the then-nascent online video distributor (OVD) market to protect the companies’ legacy pay-TV business.[100] These theories of harm focused on the potential for the merged firm to leverage its market power in broadband distribution to raise costs for OVDs through interconnection fees and to restrict their access to programming content.

Today, the factual premises underlying these theories have been fundamentally altered. The OVD market is no longer nascent. Instead, the OVD business is a mature, dominant force in media, with streaming’s share of television viewership now eclipsing that of broadcast and cable combined. Concurrently, the broadband market has become more competitive, with fiber, fixed wireless, and satellite services providing increasingly viable alternatives to cable. Moreover, a merged Charter/Cox entity would lack the significant vertical integration into content production that was a key aggravating factor in the Comcast/TWC review.

Similar to the proposed Charter/Cox deal, the proposed Comcast/TWC transaction lacked direct, head-to-head competition between the two companies.[101] Comcast and TWC operated in almost entirely separate geographic territories, a common feature of the U.S. cable industry. Therefore, consumers in any given local market would not see the number of available cable or broadband providers reduced as a direct result of the merger. This absence of horizontal overlap meant that traditional antitrust metrics, such as local market concentration as measured by HHI, were largely irrelevant to the consumer-facing retail markets.

As a result, regulators shifted their analysis away from traditional theories of harm based on the elimination of a direct competitor. Instead, the DOJ and FCC focused on potential anticompetitive effects stemming from the combined entity’s substantial increase in national scale. The agencies advanced a “gatekeeper” theory, contending that the merger would make a combined Comcast/TWC an “unavoidable gatekeeper for Internet-based services that rely on a broadband connection to reach consumers.”[102]

Thus, rather than focusing on competition in the market for broadband service, the agencies turned their attention to competition on the broadband platform itself. The concern was not that the merged firm would raise prices for its own subscribers, but that it would use its control over access to those subscribers to harm other companies operating in upstream and downstream markets, particularly in the emerging online-video space.

In particular, the government’s case against the merger centered on the perceived threat to the then-nascent OVD services, such as Netflix and Hulu.[103] Regulators theorized that a larger, more powerful cable operator, seeing OVDs as a threat to its profitable legacy pay-TV business, would possess both an increased ability and a heightened incentive to disadvantage these emerging online competitors. Government reviewers articulated three specific economic theories of harm to substantiate this concern.[104]

1. Increased bargaining power in interconnection, in which it was argued that a larger internet service provider (ISP) could leverage its control over a greater number of broadband subscribers to demand higher fees from OVDs for interconnection (e., the physical exchange of traffic required to deliver video streams to consumers).[105] These higher costs would either be passed on to consumers through higher OVD subscription prices or would reduce OVDs’ margins, thereby stifling their ability to invest in content and innovation and ultimately limiting their growth and competitive viability.²
2. Increased bargaining power in programming negotiations, in which the merged firm would represent a much larger share of the national market for video distribution. The increased buyer power—or monopsony power—could be used to force television programmers like Disney, Viacom, or Discovery to accept restrictive contract terms.[106] Specifically, regulators feared the merged entity would demand alternate-distribution means (ADM) clauses that would prohibit or penalize programmers for licensing their content to OVDs. In this way, they could starve online rivals of the programming needed to compete effectively with the traditional cable bundle.[107]
3. Internalization of externalities. This more complex economic argument focused on the merged firm’s anticipated incentives. Under this theory, it was argued that, when a single cable operator takes an action that harms an OVD (g., by degrading its video quality or imposing a restrictive data cap),[108] it creates a positive externality for other cable operators, as the OVD is now a weaker competitor for all of them. The individual operator, however, only considers the benefit to its own business. A merger between two large operators would cause the new, larger firm to internalize these externalities and would account for the benefits that its anticompetitive actions confer on its newly acquired territories.[109] This internalization, in turn, would increase the firm’s overall incentive to engage in conduct harmful to OVDs, as it would now capture a larger share of the industrywide gains from such a strategy.

These theories of anticompetitive harm were advanced while the FCC was in the midst of drafting its 2015 Open Internet Order, also known as “net neutrality,” which mandated policies involving interconnection, blocking, throttling, paid priority, and data caps that were identified as necessary (or desired) to preserve a so-called “open” internet.[110]

An aggravating factor in the merger review was Comcast’s vertical integration into content through its ownership of NBCUniversal (NBCU).[111] This ownership distinguished the proposed merger from a pure horizontal combination of distribution assets. Time Warner Cable, having been spun off from Time Warner Inc. in 2009, did not own significant programming assets of its own. Comcast’s ownership of a major content studio created what regulators saw as a powerful, two-sided incentive to foreclose rivals. The merged firm could potentially harm OVDs, not only to protect its legacy cable-distribution business, but also to advantage its own content business.[112] This vertical integration amplified concerns raised by the programming-negotiation theory of harm, making the threat of anticompetitive foreclosure appear more concrete and credible to regulators.

The decade since the scuttled Comcast/TWC merger has been a period of substantial transformation in the media and communications landscape. The competitive dynamics and market structures of 2025 bear little resemblance to those of 2015.

• In 2015, regulators characterized OVDs as “nascent,” with entrants requiring protection from entrenched incumbents. By 2025, this narrative has been inverted. Today, streaming’s share of total television usage (46%) surpasses the combined share of broadcast and cable television (42%).[113] In 2015, around 50% of U.S. households held subscriptions to streaming services; in 2023, 83% of households subscribed to one or more of the major streaming providers.[114]
• The competitive landscape for broadband-internet access has also become significantly more dynamic. In 2015, 16% of households had no access to download speeds of 25 Mbps or higher and 45% only had access from a single provider. Today, a large majority of U.S. homes have access to at least three fixed, mobile, or satellite broadband services with speeds of 100 Mbps or higher.[115]
• Several net-neutrality orders have come and gone. Most recently, the 6^thS. Court of Appeals struck down the FCC’s 2024 effort to regulate broadband-internet providers as common carriers under the guise of net neutrality.

This evolution renders meaningless the main competitive concerns raised in the Comcast-TWC merger 10 years ago.

B. AT&T/Time Warner (2018)

The DOJ’s challenge to the AT&T/Time Warner merger in 2018 was the first fully litigated vertical-merger case in four decades.[116] The government’s ultimate failure to block that deal established a clear and demanding evidentiary standard for future challenges to a nonhorizontal merger, favoring evidence-based economic analysis over speculative theories of competitive harm.

The DOJ’s case rested on a vertical-foreclosure theory, arguing that the combined firm would have both the incentive and ability to leverage Time Warner’s seemingly indispensable Turner networks to raise programming costs for its distribution rivals, such as cable and satellite companies.[117] This argument was grounded in economic principles—chiefly the Nash Bargaining Theory—that posited that AT&T’s ownership of its own distributor (DirecTV) would reduce its downside risk from a programming blackout with a rival, thereby increasing its bargaining leverage in carriage negotiations.[118]

The U.S. District Court for the District of Columbia, in an opinion[119] that was unanimously affirmed by the U.S. Circuit Court of Appeals for the D.C. Circuit,[120] rejected the government’s theory. The court’s decision turned on the government’s failure to demonstrate a “reasonable probability” that the alleged theoretical harms would manifest in the real world.[121] Judge Richard Leon found the DOJ’s quantitative economic model to be unreliable and factually detached from marketplace realities, citing flawed inputs and a critical failure to account for the constraining effects of existing long-term contracts. The court concluded that the government’s case was built on a chain of assumptions and predictions that were ultimately contradicted by the factual evidence.

In contrast to the government’s theory, the court found the merging parties’ evidence more persuasive because it was rooted in empirical data and real-world experience.[122] AT&T presented a detailed econometric analysis of prior vertical mergers in the same industry, which demonstrated no statistically significant effect on content prices.[123] This historical data provided a counter-narrative to the government’s forward-looking, theoretical model. The court also credited testimony from industry executives, who affirmed that threatening or initiating content blackouts was a mutually destructive strategy that would remain an unattractive option even after the merger.[124] The testimony of the merging parties’ rivals, upon which the government heavily relied, was largely discounted by the court as self-interested and speculative.[125]

One factor guiding the court’s skepticism of the DOJ’s theory of harm was Time Warner’s voluntary, post-litigation commitment to submit to binding “no-blackout” arbitration to resolve any future carriage disputes with distributors.[126] This irrevocable offer was a practical, market-based solution that directly neutralized the government’s central theory of harm regarding the threat of programming blackouts. By providing a concrete mechanism to prevent the anticompetitive conduct the DOJ anticipated, the merging parties stifled one of the government’s chief concerns.

The D.C. Circuit’s affirmation of Judge Leon’s decision laid down a marker for future vertical-merger challenges. The appellate court reiterated that, because vertical mergers do not produce an immediate change in market concentration, the government cannot rely on structural presumptions of harm as it can in horizontal cases.[127] Instead, the government must make a “fact-specific” showing that the merger is likely to be anticompetitive.[128]

A key lesson of the AT&T-Time Warner merger is that enjoining a nonhorizontal merger requires more than economic theories or complaints from competitors. A successful challenge to a nonhorizontal merger demands concrete, credible, and compelling evidence that the transaction will likely cause substantial harm to competition in the actual marketplace.

VI. Conclusion and Policy Implications

The proposed Charter/Cox merger is best understood as a strategic response to sweeping changes in the U.S. communications market. Across fixed broadband, pay TV, and mobile, service providers now operate amid increasing platform diversity, rapid technological change, and more demanding consumer expectations.

Charter and Cox face significant and growing competition from fixed-wireless access, LEO satellites, and the growing fiber footprints of AT&T and Verizon, as well as from more than 200 streaming platforms. The market landscape bears little resemblance to the highly segmented, technology-specific markets of the past. With minimal geographic overlap, the merger does not remove a competitor from any local market—a fact with direct implications for how competitive risks should be weighed.

The evidence shows that the combined Charter/Cox entity largely represents a geographic expansion, rather than a horizontal consolidation. Fewer than 0.1% of broadband-serviceable locations are currently overlapped by both providers. Review of past mergers underscores that, in such cases, traditional antitrust concerns about increased local market concentration are substantially reduced or inapplicable. Instead, the key analytical question shifts to whether the transaction creates the capacity or incentives for anticompetitive behavior at a national level, especially in areas like programming acquisition or interconnection. These concerns have diminished, however, as the competitive balance has tilted toward multiplatform, vertically integrated giants, many of whom are not subject to the same regulatory scrutiny as legacy cable operators.

The efficiencies claimed by Charter and Cox rest primarily on scale: approximately $500 million in annual targeted cost savings within three years. These savings are expected to arise from reduced equipment costs, more favorable mobile-network access, programming-acquisition efficiencies, and the elimination of duplicative syndication agreements. The merged company projects that these savings will support faster deployment of advanced broadband and mobile technologies, lower average costs per customer, and improvements in service quality—outcomes consistent with the consumer-welfare standard. The ability to offload mobile traffic onto expanded Wi-Fi networks, in particular, could lower operational costs and deliver more competitive mobile offerings within Cox’s footprint.

Regulatory review of the Comcast/Time Warner Cable and AT&T/Time Warner mergers provides several lessons. First, while the agencies were focused a decade ago on the risk that national-level consolidation might create “gatekeeper” power over nascent online-video platforms, the market has since shifted. Online video is now the dominant modality, and vertical integration by technology platforms (e.g., Apple, Google, Amazon) has eclipsed that of most traditional providers. Regulatory asymmetry remains, in that cable operators navigating mergers undergo multi-agency review and bear the costs of legacy regulation, even as their principal competitors operate under few or no comparable obstacles.

For policymakers, the Charter/Cox transaction presents an opportunity to realign regulatory practice with economic reality. The consumer-welfare standard—focused on demonstrable, transaction-specific harms and efficiencies—remains the most reliable guide. Even under an “America First” antitrust approach, regulatory responses should be rooted in factual market effects, not abstract concerns about scale or size. Where claimed efficiencies are both merger-specific and verifiable, the burden should be on the agencies to demonstrate substantial, tangible harms to competition, rather than rely on speculative or politically motivated theories of harm.

The evolving role of the FCC, especially under recent “public interest” reinterpretations, deserves careful consideration. Conditions unrelated to competition—such as those advancing specific employment or corporate-governance policies—risk introducing regulatory unpredictability and could dampen incentives for investment and innovation. Such interventions, particularly when applied inconsistently across industry segments, threaten to bias market outcomes and discourage the very adaptation that consumers and the U.S. economy require.

Ultimately, a coherent regulatory framework should aim to treat functionally equivalent services the same, regardless of historical provider classification. As the communications marketplace continues to converge, policy must shift toward technological neutrality and consistent application of antitrust principles to all market players. By focusing on demonstrable effects and resisting the urge to intervene on grounds unrelated to competition, policymakers will best promote investment, innovation, and consumer benefit—a result well aligned with the economic evidence presented in this review. Importantly, that conclusion holds whether the transaction is examined under the traditional consumer-welfare standard or under the newer “Hillbilly Antitrust” framework: in both cases, the facts support approval.

* Eric Fruits and Ben Sperry are senior scholars with the International Center for Law & Economics (ICLE). Kristian Stout is ICLE’s director of innovation policy. ICLE has received financial support from numerous companies, foundations, and individuals, including firms with interests both supportive of and in opposition to the ideas expressed in this and other ICLE-supported works. Unless otherwise noted, all ICLE support is in the form of unrestricted, general support. The ideas expressed here are the authors’ own and do not necessarily reflect the views of ICLE’s advisors, affiliates, or supporters.

[1] Press Release, Charter Communications and Cox Communications Announce Definitive Agreement to Combine Companies, Charter Commcn’s (May 16, 2025), https://corporate.charter.com/newsroom/charter-communications-and-cox-communications-announce-definitive-agreement-to-combine-companies; see also Charter Communications and Cox Communications Agree to Transformative Combination, Charter Commcn’s & Cox Commcn’s (May 16, 2025), https://ir.charter.com/static-files/17f74638-d569-448c-be88-76d00f9c6fff [hereafter “Charter/Cox presentation”].

[2] Jake Neenan, Charter-Cox Merger Provides Convergence Runway, Broadband Breakfast (May 16, 2025), https://broadbandbreakfast.com/charter-cox-merger-provides-convergence-runway.

[3] Pay TV Providers Ranked by the Number of Subscribers in the United States as of 1st Quarter 2025, Statista (May 2025), https://www.statista.com/statistics/251793/pay-tv-providers-with-the-largest-number-of-subscribers-in-the-us.

[4] Charter, supra note 1.

[5] Turner Broadcasting System v. FCC, 512 U.S. 622, 661 (1994) (“Turner I”) (the “cable medium” has “special characteristics” of “bottleneck monopoly power”).

[6] Turner Broadcasting System v. FCC, 520 U.S. 180, 197 (1997) (“Turner II”).

[7] Comcast Cable Commcn’s v. FCC, 717 F.3d 982, 993-94 (2013) (J. Kavanaugh, concurring).

[8] Report and Order, In the Matter of Amendment to the Commission’s Rules Concerning Effective Competition, Implementation of Section 111 of the STELA Reauthorization Act (MB Docket No. 15-53, FCC 15-62, Jun. 3, 2015), available at https://docs.fcc.gov/public/attachments/FCC-15-62A1.pdf.

[9] Eighth Annual Report, In the Matter of Annual Assessment of the Status of Competition in the Market for the Delivery of Video Programming, CS Docket No. 01-129, Table B-1 (Dec. 27, 2001), available at https://docs.fcc.gov/public/attachments/FCC-01-389A1.pdf.

[10] Philip Graves, How System Requirements for Browsing the Internet Have Changed: Part 1: Internet Connection Speeds, GWS Media (Nov. 23, 2021), https://www.gwsmedia.com/articles/how-internet-system-requirements-have-changed (“In early 1993, the fastest available modem was capable of transferring data at a maximum speed of 14.4 kilobits per second (kbps), equivalent to 864kb per minute, or 51.84Mb per hour. The launch of the 28.8k modem in 1994 doubled this theoretical maximum. 33.6k modems followed in 1996, and eventually 56k ones arrived in 1998. …  However, in practice the maximum advertised speeds were never attained for any dial-up modems as a result of latencies in the infrastructure serving data to end-users.”); see also Eric Fruits, Kristian Stout, & Geoffrey A. Manne, The Economics of Broadband Data Caps and Usage-Based Pricing, Int’l Ctr. L. & Econ. (Oct. 23, 2024), available at https://laweconcenter.org/wp-content/uploads/2024/10/Data-Caps-2024.pdf (“In the early days of the commercial internet in the 1990s, most consumers accessed the internet via dial-up connections. These connections were slow—averaging around 56 Kbps—and content was limited. It could take a minute or more for a single image file to load.”).

[11] About Us, DirecTV, https://www.directv.com/insider/corporate/company (last visited Jul. 22, 2025).

[12] Joni Blecher, The History of RealPlayer, RealPlayer Blog (Aug. 17, 2016), https://blog.real.com/realplayer-history.

[13] Speedtest Global Index: Median Country Speeds Updated June 2025, Speedtest, https://www.speedtest.net/global-index (last visited Jul. 22, 2025).

[14] US Trad Pay TV & vMVPDs, nScreen Media, https://nscreenmedia.com/us-pay-tv (last visited Jul. 24, 2025).

[15] Eugenie Park & Colleen McClain, 83% of U.S. Adults Use Streaming Services, Far Fewer Subscribe to Cable Or Satellite TV, Pew Research Center (Jul. 1, 2025), https://www.pewresearch.org/short-reads/2025/07/01/83-of-us-adults-use-streaming-services-far-fewer-subscribe-to-cable-or-satellite-tv; see also Eric Fruits, Video Competition in 2025: It’s Literally on Heebee, Truth on Mkt. (Feb. 14, 2025), https://truthonthemarket.com/2025/02/14/video-competition-in-2025-its-literally-on-heebee (“In the face of so many streaming options, millions of consumers have ‘cut the cord’ and switched from cable and DBS satellite to streaming services over the internet. Cable subscribership peaked in 2010. Since then, the number of subscribers dropped by more than 35%. The Washington Post reports that barely half of American homes pay for live TV service from a cable, satellite TV, or an internet-delivered vMVPD (virtual multichannel video programming distributor) like YouTube TV.”).

[16] 2025 Media & Entertainment Industry Predictions Report, Alix Partners (Dec. 2024), available at https://www.alixpartners.com/media/ow1n5vey/2025-media-entertainment-industry-predictions-report.pdf.

[17] Colin Dixon, TV for the Rest of Us: How Streaming Unleashed Specialty Audiences, nScreenMedia (Jun. 3, 2025), https://nscreenmedia.com/how-streaming-unleashed-specialty-audiences.

[18] The Gauge, The Nielsen Co., https://www.nielsen.com/data-center/the-gauge (last visited Jul. 22, 2025).

[19] Eric Fruits, Geoffrey A. Manne, Ben Sperry, & Kristian Stout, Dynamic Competition in Broadband Markets: A 2024 Update, Int’l Ctr. L. & Econ. (Jun. 4, 2024), available at https://laweconcenter.org/wp-content/uploads/2024/06/Broadband-Competition-2024-Update.pdf.

[20] Andrew Long, The Proposed Charter-Cox Merger: A Pro-Consumer Response to Today’s Competitive Communications Marketplace, Free State Found. (Jun. 10, 2025), available at https://freestatefoundation.org/wp-content/uploads/2025/06/The-Proposed-Charter-Cox-Merger-061025.pdf.

[21] Eric Fruits, Ben Sperry, & Kristian Stout, The Competitive Effects of the Proposed T-Mobile/UScellular Transaction, Int’l Ctr. L. & Econ. (Dec. 16, 2024), available at https://laweconcenter.org/wp-content/uploads/2024/12/T-Mobile-USCellular.pdf.

[22] Id.

[23] See, e.g., Drew FitzGerald & Patience Haggin, Charter, Cox Merge in Megadeal Amid Escalating War With Wireless, Wall St. J. (May 16, 2025), https://www.wsj.com/business/deals/charter-communications-to-merge-with-rival-cox-in-21-9-billion-deal-c70dcff9?st=6rQTPa&reflink=desktopwebshare_permalink (“Cellphone carriers like Verizon and T-Mobile have heightened the threat with home broadband service beamed over the air. Wireless companies have racked up millions of customers with the offerings, which use 5G technology to provide internet speeds that are competitive with fixed cable lines at lower prices.”).

[24] Fruits et al., supra note 19.

[25] 2024 Section 706 Report, In the Matter of Inquiry Concerning the Deployment of Advanced Telecommunications Capability to All Americans in a Reasonable and Timely Fashion (GN Docket No. 22-270, Mar. 14, 2024), available at https://docs.fcc.gov/public/attachments/DOC-400675A1.pdf.

[26] Fruits et al., supra note 19.

[27] Id.

[28] United States Telecommunications Market Overview, Business Monitor Online (Jun. 23, 2025).

[29] Jeff Baumgartner, Verizon Doubles Down on FWA, Accelerates Fiber Buildout Plan, Light Reading (Oct. 22, 2024), https://www.lightreading.com/broadband/verizon-doubles-down-on-fwa-accelerates-fiber-buildout-plan.

[30] Fruits et al., supra note 19.

[31] Bevin Fletcher, Charter to Acquire Cox in Deal Worth $34.5B, StreamTV Insider (May 16, 2025), https://www.streamtvinsider.com/video/charter-acquire-cox-deal-worth-345b.

[32] Id.; see also Long, supra note 20.

[33] Fruits et al., supra note 19, citing Eric Fruits, Justin (Gus) Hurwitz, Geoffrey A. Manne, Julian Morris, & Alec Stapp, Static and Dynamic Effects of Mergers: A Review of the Empirical Evidence in the Wireless Telecommunications Industry, (OECD Directorate for Financial and Enterprise Affairs Competition Committee, Global Forum on Competition, DAF/COMP/GF(2019)13, Dec. 6, 2019), available at https://one.oecd.org/document/DAF/COMP/GF(2019)13/en/pdf.

[34] Long, supra note 20.

[35] Neenan, supra note 2.

[36] Fletcher, supra note 31.

[37] 15 U.S. Code § 18 [emphasis added].

[38] California v. Am. Stores Co., 495 U.S. 271, 284 (1990), (quoting 15 U.S.C. § 18 with emphasis) (citing Brown Shoe, 370 U.S. at 323).

[39] United States v. Philadelphia Nat’l Bank, 374 U.S. 321 (1963).

[40] The Use of Structural Presumptions in Antitrust (OECD Roundtables on Competition Policy Papers No. 317, 2024), available at https://www.oecd.org/content/dam/oecd/en/publications/reports/2024/11/the-use-of-structural-presumptions-in-antitrust_27777e33/3b8c6885-en.pdf.

[41] Horizontal Merger Guidelines, Dep’t of Justice & Fed. Trade Comm’n, (2023), available at https://www.justice.gov/d9/2023-12/2023%20Merger%20Guidelines.pdf.

[42] Eric Posner, What Is the Role of Economics in Merger Review?, ProMarket (Mar. 28, 2024), https://www.promarket.org/2024/03/28/what-is-the-role-of-economics-in-merger-review.

[43] Natalie Weger, Justice Department Won’t Seek Injunction for T-Mobile Acquisition of U.S. Cellular, Wall St. J. (Jul. 10, 2025), https://www.wsj.com/business/telecom/justice-department-wont-seek-injunction-for-t-mobile-acquisition-of-u-s-cellular-b9e7abac?st=81wr98&reflink=desktopwebshare_permalink; see also Fruits et al., supra note 21 (“Despite T-Mobile’s nationwide coverage, some spots served by UScellular are not well served by T-Mobile… These are areas in which the transaction would not reduce the number of competing firms, as T-Mobile would simply replace UScellular in areas where T-Mobile lacks coverage or combine assets to create even more effective competition against other providers.”)

[44] See, e.g., G. E. Hale & Rosemary D. Hale, Expanding Enterprise: Geographical Curbs on Mergers, 51 Minn. L. R. 857, 867 (1967) (“[I]t is clear that such an acquisition may result in cost savings to the combined firm. Indeed, the merger in question would presumably never have been negotiated unless some such spreading of overhead costs were envisaged. Observers have found that there are such savings in the operation of chain stores in the grocery field. In numerous other industries similar cost reduction can be achieved through nationwide promotion and distribution of goods.”).

[45] Eric Fruits, Kristian Stout, Ben Sperry & Geoffrey A. Manne, Comments of the International Center for Law & Economics Re: Delete, Delete, Delete, GN Docket No. 25-13352, Int’l Ctr. L. & Econ. (Apr. 11, 2025), available at https://laweconcenter.org/wp-content/uploads/2025/04/2025-Delete-Delete-Delete-Comments-r3.pdf.

[46] Jon Sallett, FCC Transaction Review: Competition and the Public Interest, FCC Blog (Apr. 14, 2014), https://www.fcc.gov/news-events/blog/2014/08/12/fcc-transaction-review-competition-and-public-interest.

[47] Applications Filed for the Transfer of Control of Cox Communications, Inc. to Charter Communications, Inc., Fed. Commcn’s Comm’n (Sep. 5, 2025), available at https://docs.fcc.gov/public/attachments/DA-25-810A1.pdf.

[48] Frontier Communications Parent, Inc. and Verizon Communications, Inc. Application for Consent to Transfer Control, Memorandum Opinion and Order, DA 25-421 ¶ 16 (WCB/OIA/WTB rel. May 16, 2025).

[49] Applications of AT&T Inc. and Centennial Communications Corp., Memorandum Opinion and Order, 24 FCC Rcd 13915, 13931 ¶ 34 (2009).

[50] Sallet, supra note 46 (“Fundamental is the fact that applicants have the burden of demonstrating on the public record that their proposed transaction is in the public interest.”)

[51] Though the D.C. Circuit has suggested in dicta that conditions must be “transaction-specific” and that “non-germane conditions” are “an out-and-out plan of extortion.” See Competitive Enterprise Institute v. FCC, 970 F.3d 372, 388 (D.C. Cir. 2020) (quoting Nollan v. Cal. Coastal Comm’n, 483 U.S. 825, 837 (1987)).

[52] Loper Bright Enterprises v. Raimondo, 603 US ___ (2024); see also Ben Sperry & Eric Fruits, How This Supreme Court Term Might Affect the FCC’s Digital-Discrimination Rule, Truth on Mkt. (Jul. 3, 2024), https://truthonthemarket.com/2024/07/03/how-this-supreme-court-term-might-affect-the-fccs-digital-discrimination-rule.

[53] Eric Fruits, A Hillbilly and a Hipster Walk Into a Bar, Truth on Mkt. (May 30, 2025), https://truthonthemarket.com/2025/05/30/a-hipster-and-a-hillbilly-walk-into-a-bar.

[54] Gail Slater, The Conservative Roots of America First Antitrust Enforcement, Address to University of Notre Dame Law School (Apr. 28, 2025), https://www.justice.gov/opa/speech/assistant-attorney-general-gail-slater-delivers-first-antitrust-address-university-notre.

[55] Andrew Ferguson, Keynote Speech, International Competition Network Annual Conference 2025, YouTube (May 8, 2025), https://youtu.be/yOp5__oNZ8k?si=pSitKJ3QrEvR4zsB (“My friend and colleague, Gail Slater at the U.S. Justice Department, has likened ex ante regulation to a sledgehammer, whereas ex post antitrust enforcement is a scalpel.”)

[56] Fireside with Abigail Slater, Little Tech Competition Summit, YouTube (Apr. 28, 2025), https://youtu.be/Hk6A1WcJtPs?si=EJrRe7R6v5oDTvel (“[T]he consumer welfare standard is decently broad… the variables including quality, price as a dimension of competition, innovation, things of that nature. As a matter of agency practice, often it’s been narrowed down to price… The narrow focus on price, I think, is something that is not even compatible with the consumer welfare standard construct as it exists in current precedent, and I think we need to be open… to think about it more broadly than just price.”).

[57] Mark R. Meador, Antitrust Policy for the Conservative, Fed. Trade Comm’n (May 1, 2025), available at https://www.ftc.gov/system/files/ftc_gov/pdf/antitrust-policy-for-the-conservative-meador.pdf (“Conservatives should insist that antitrust analysis credit only efficiencies that: (1) are realized in the same market as the harms they offset; (2) can only reasonably be achieved through the conduct or transaction at issue; (3) are nonspeculative (i.e., measurable in some way and likely to be realized); and (4) will directly and predominantly accrue to consumers.”)

[58] Brendan Carr, A Build Agenda for America, Fed. Commcn’s Comm’n (Jul. 2, 2025), available at https://docs.fcc.gov/public/attachments/DOC-412663A1.pdf.

[59] Jeff Green, FCC’s Carr Threatens to Block M&A for Companies With DEI, Bloomberg Law (Mar. 21, 2025), https://news.bloomberglaw.com/ip-law/fccs-carr-threatens-to-block-m-a-for-companies-with-dei-plans.

[60] Masha Abarinova, T-Mobile Updates DEI to Get Its Lumos Fiber Deal Approved, Fierce Network (Apr. 3, 2025), https://www.fierce-network.com/broadband/t-mobile-aims-high-fiber-now-lumos-bag.

[61] US FCC Approves Two T-Mobile Deals After Wireless Carrier Drops DEI Programs, Reuters (Jul. 11, 2025), https://www.reuters.com/sustainability/society-equity/fcc-approves-two-t-mobile-deals-after-wireless-carrier-drops-dei-programs-2025-07-11.

[62] Jericho Casper, Verizon Ends DEI Programs as FCC Reviews $9.6B Frontier Deal, Broadband Breakfast (May 16, 2025), https://broadbandbreakfast.com/verizon-ends-dei-programs-as-fcc-reviews-9-6b-frontier-deal.

[63] Memorandum Opinion and Order, Declaratory Ruling, and Order of Proposed Modification, In the Matter of Applications of T-Mobile US, Inc., and Sprint Corporation for Consent to Transfer Control of Licenses and Authorizations, WT Docket No. 18-197, FCC 19-103 (Oct. 19, 2019), available at https://docs.fcc.gov/public/attachments/FCC-19-103A1.pdf.

[64] Id.

[65] Robert C. Azarow, David F. Freeman, Jr., Amber A. Hay, Michael A. Mancusi, Kevin M. Toomey, & Paul Lim, Bank Regulator’s Approval of Capital One and Discover Deal Shows Path Forward for Bank M&A Deals, Arnold & Porter (Jun. 5, 2025), https://www.arnoldporter.com/en/perspectives/advisories/2025/06/bank-regulators-approval-of-capital-one-and-discover-deal.

[66] See, e.g., Public Interest Statement, In the Matter of Cox Enterprises, Inc. and Charter Communications, WC 25-233 (July 14, 2025), https://www.fcc.gov/ecfs/document/10715141122783/2 [hereafter “Public interest statement”]. See, Section IV.B. (“The Transaction Will Put America First”).

[67] Eric Fruits, Media-Ownership Regulations in a Streaming World: Time to Change the Channel, Truth on Mkt. (Mar. 5, 2025), https://truthonthemarket.com/2025/03/05/media-ownership-regulations-in-a-streaming-world-time-to-change-the-channel (“Broadcasters operate under one set of ownership regulations and cable providers operate under another set. Meanwhile, the emerging market of internet-based video distribution continues to operate almost entirely free from ownership regulations. Companies like Netflix, Amazon, and YouTube entered the market without facing the ownership limitations, public-interest obligations, or local-content requirements imposed on their legacy competitors.”).

[68] See, e.g., Fruits et al., supra note 45 (“Merging parties currently must provide substantially similar competitive analyses and market data to both antitrust authorities and the FCC. This redundancy creates unnecessary administrative burdens, without providing proportionate regulatory benefits.”).

[69] Nielsen, supra note 18.

[70] Tod Spangler, Netflix Content Spending, Set to Hit $18 Billion in 2025, Is “Not Anywhere Near a Ceiling,” CFO Says, Variety (Mar. 5, 2025), https://variety.com/2025/digital/news/netflix-content-spending-2025-ceiling-cfo-1236328510.

[71] Public Interest Statement, In the Matter of Cox Enterprises, Inc. and Charter Communications, WC 25-233, Appendix E, Declaration of Bryan Keating & Jonathan Orszag ¶ 61 (July 14, 2025), https://www.fcc.gov/ecfs/document/10715141122783/3 [hereafter “Keating & Orszag”].

[72] Keating & Orszag, supra note 70 ¶ 62 n.73 (reporting approximately 52.9 and 49.6 percent of mass market locations serviceable by Charter and Cox, respectively, are also serviceable by at least one competing FTTP provider at the end of 2024).

[73] Fruits et al., supra note 19.

[74] Consumer Price Index for All Urban Consumers: All Items in U.S. City Average [CPIAUCSL], U.S. Bureau of Labor Statistics (retrieved from FRED, Federal Reserve Bank of St. Louis, Aug. 4, 2025), https://fred.stlouisfed.org/series/CPIAUCSL; Consumer Price Index for All Urban Consumers: Internet Services and Electronic Information Providers in U.S. City Average, U.S. Bureau of Labor Statistics, https://data.bls.gov/timeseries/CUUR0000SEEE03?output_view=data (last visited Aug. 4, 2025).

[75] Ellis Scherer & Joe Kane, Broadband Convergence Is Creating More Competition, Information Tech. & Innov. Found. (Jul. 2025), available at https://www2.itif.org/2025-broadband-convergence.pdf.

[76] Press Release, Charter Announces Second Quarter 2025 Results, Charter Commcn’s (Jul. 25, 2025), https://corporate.charter.com/newsroom/charter-announces-second-quarter-2025-results.

[77] Fitch Ratings, Fixed Wireless Access Growth Disrupts U.S. Telecom Market (Mar. 26, 2025), https://www.fitchratings.com/research/corporate-finance/fixed-wireless-access-growth-disrupts-us-telecom-market-26-03-2025.

[78] Public Interest Statement, supra note 65 at 30.

[79] Brendan Carr, Free State Foundation Seventeenth Annual Policy Conference, YouTube (Mar. 25, 2025), https://www.youtube.com/live/G33c8UlQjxE?si=vxILcpPlzXcCvj3K.

[80] Starlink Network Update, Starlink, https://www.starlink.com/updates/network-update (last visited Jul. 30, 2025),

[81] Amazon Plans to Offer Satellite Internet Service in Late 2025, PYMTS (Jul. 14, 2025), https://www.pymnts.com/amazon/2025/amazon-plans-offer-satellite-internet-service-late-2025.

[82] Fruits et al., supra note 21.

[83] Fruits et al., supra note 21, Table 1. As a private company, Cox does not report mobile wireless subscribers. Even if all of its customers subscribed to mobile wireless service, however, Cox would account for less than 2% of total subscribers.

[84] Fruits, supra note 15.

[85] Nielsen, supra note 18.

[86] Pay TV Providers Ranked by the Number of Subscribers in the United States as of 1st Quarter 2025, Statista (May 18, 2025), https://www.statista.com/statistics/251793/pay-tv-providers-with-the-largest-number-of-subscribers-in-the-us.

[87] Press Release, Public Knowledge Warns $34.5 Billion Cox/Charter Merger Might Be Weaponized, Public Knowledge (May 16, 2025), https://publicknowledge.org/public-knowledge-warns-34-5-billion-cox-charter-merger-might-be-weaponized (quoting Public Knowledge Legal Director John Bergmayer: “As always with cable mergers, the question is as much a loss of opportunities for content creators and programmers to reach an audience, as the loss of choices to subscribers.”)

[88] 2023 Broadband Capex Report, USTelecom (Oct. 18, 2024), available at https://ustelecom.org/wp-content/uploads/2024/10/UST-1376-CAPEX-Report_2024_4-as-of-Oct-4.pdf.

[89] Charter & Cox presentation, supra note 1.

[90] Public Interest Statement, supra note 65 at 30.

[91] Keating & Orszag, supra note 70 ¶¶ 32-37.

[92] Public Interest Statement, supra note 65 at 38.

[93] Keating & Orszag, supra note 70 ¶¶ 42-45.

[94] Jeff Baumgartner, Charter and Cox State Their Case at the FCC, LightReading (Jul. 17, 2025), https://www.lightreading.com/regulatory-politics/charter-and-cox-make-their-case-at-the-fcc.

[95] Public Interest Statement, supra note 65 at 30.

[96] Baumgartner, supra note 93.

[97] Keating & Orszag, supra note 70 ¶ 48.

[98] Public Interest Statement, supra note 65 at 52.

[99] Press Release, Comcast Corporation Abandons Proposed Acquisition of Time Warner Cable After Justice Department and the Federal Communications Commission Informed Parties of Concerns, Dept. of Justice (Apr. 24, 2025), https://www.justice.gov/archives/opa/pr/comcast-corporation-abandons-proposed-acquisition-time-warner-cable-after-justice-department.

[100] Id. See also William P. Rogerson, Economic Theories of Harm Raised by the Proposed Comcast/TWC Transaction (2015), in The Antitrust Revolution (7^th ed., John E. Kwoka, Jr. & Lawrence J. White eds., 2018); Jon Sallet, The Federal Communications Commission and Lessons of Recent Mergers & Acquisitions Reviews, Telecomms. Pol’y Rsch. Conf. (Sep. 25, 2015), available at https://transition.fcc.gov/Daily_Releases/Daily_Business/2015/db0925/DOC-335494A1.pdf.

[101] Sallet, supra note 99 (“there was minimal horizontal overlap between the Applicants in the local markets for residential broadband and Pay TV service”).

[102] DOJ, supra note 98.

[103] Sallet, supra note 99 (“We understood that entrants are particularly vulnerable when competition is nascent. Thus, staff was particularly concerned that this transaction could damage competition in the video distribution industry by increasing both Comcast’s incentive and its ability to disadvantage OVDs and thus retard or permanently stunt the growth of a competitive OVD industry. In doing so, consumers would be denied the benefits that innovative competition could bring.”).

[104] For a critical evaluation of these theories of harm, see Geoffrey A. Manne, The FCC Distorted Market Realities to Scuttle the Comcast-TWC Merger, Truth on Mkt. (Oct. 2, 2015), https://truthonthemarket.com/2015/10/02/the-fcc-distorted-market-realities-to-scuttle-the-comcast-twc-merger.

[105] Sallet, supra note 99 (“Similarly, we also considered a national market for interconnection in which ISPs negotiate with OVDs (and their content delivery networks) over the terms by which the OVDs would reach consumers. Post-transaction, an OVD might have needed an interconnection agreement with the merged entity in order to achieve national distribution, so we also considered the ability of the merged company to impose terms that would disadvantage the OVD.”).

[106] Sallet, supra note 99 (“Alongside incentives came ability. Increased bargaining power was the central concern. The combination of distribution assets had the potential to increase the merged entity’s bargaining power in both national markets—the market where video distributors negotiate the terms and conditions to distribute video content for programmers and the interconnection market through which broadband providers provide mass-market delivery services to OVDs. Because OVDs are subject to national economies of scale, the merged company could significantly impair an OVD’s ability to compete. … But after a merger, that OVD would have to strike a bargain with only one firm, which would give that company the ability to disadvantage the OVD, or perhaps even exclude the OVD from reaching its subscribers.”).

[107] Rogerson, supra note 99 (“government reviewers were concerned that [ADM terms] could also be used simply to deny OVD competitors access to programming and that OVDs would be particularly vulnerable to such anticompetitive actions when they were just attempting to enter and had relatively small market shares.”)

[108] Sallet, supra note 99 (“we considered their competitive effect when combined with data caps and other retail broadband terms and conditions that raised the price of OVDs for consumers”).

[109] Sallet, supra note 99 (“Without the merger, a company taking action against OVDs for the benefit of the Pay TV system as a whole would incur costs but gain additional sales—or protect existing sales—only within its footprint. But the combined entity, having a larger footprint, would internalize more of the external ‘benefits’ provided to other industry members.”).

[110] Protecting and Promoting the Open Internet, 47 CFR Parts 1, 8, and 20 (GN Docket No. 14-28, FCC 15-24. Apr. 13, 2015).

[111] See, e.g., Leticia Miranda, Why Comcast Walked Away, ProPublica (Apr. 23, 2015), https://www.propublica.org/article/why-comcast-seems-to-be-walking-away (“Bloomberg complained to the FCC that Comcast placed Bloomberg TV in the outer dial away from most other business networks and its own channel, CNBC, in the lower dial with other business news where viewers would be more likely to come upon it.”)

[112] Rogerson, supra note 99 (“As part of a dispute with Netflix over interconnection fees that began in early 2013 and lasted for approximately nine months, Comcast allegedly allowed its interconnection points with Cogent and other transit providers that delivered Netflix traffic to Comcast to become congested, which severely deteriorated the ability of Comcast subscribers to view Netflix content.”)

[113] Nielsen, supra note 18.

[114] Billy Thompson, The Rise and Fall of Streaming TV?, Mich. J. Econ. (May 25, 2024), https://sites.lsa.umich.edu/mje/2024/05/25/the-rise-and-fall-of-streaming-tv; Park & McClain, supra note 15.

[115] Fruits et al., supra note 19.

[116] Koren W. Wong -Ervin, Antitrust Analysis of Vertical Mergers: Recent Developments and Economic Teachings, Antitrust Source (Feb. 2019), available at https://www.americanbar.org/content/dam/aba/publications/antitrust/magazine/archived/2019/february/antitrust-analysis-vertical-mergers.pdf.

[117] Trial Brief of the United States, United States v. AT&T Inc., (No. 1:17-cv-02511-RJL D.D.C. 2018). (“Here, the critical question is this: Would consumers quit subscribing to AT&T’s competitors and switch to AT&T if they did not carry Time Warner content, thus allowing AT&T to increase the price of that content? If so, the merger will allow AT&T to increase its rivals’ costs—and those higher costs will, in turn, be passed on to consumers.”)

[118] Trial brief, supra note 116 (“For example, post-merger, if Turner and Charter are unable to reach a deal, the merged entity now would realize a benefit in the form of increased profits for DirecTV from new subscribers to AT&T’s service gained from Charter as subscribers switch from Charter to AT&T to ensure continuity in their reception of the desired Turner content. Accordingly, the model predicts that post-merger bargaining between Turner and Charter will result in a higher price for Turner content.”)

[119] United States v. AT&T Inc., 310 F. Supp. 3d 161 (D.D.C. 2018) [hereafter AT&T District]

[120] United States v. AT&T Inc., No. 18-5214 (D.C. Cir. 2019). [hereafter AT&T Circuit]

[121] AT&T District, supra note 118 at 198-190 n. 16.

[122] See, e.g., AT&T District, supra note 118 at 240 (“I am thus left with projections of harm for the years 2016, 2017, and 2021 that all concede have not and will not occur in the real world due to Turner’s actual affiliate agreements.”).

[123] AT&T District, supra note 118 at 207, 215-218.

[124] AT&T District, supra note 118 at 218-219.

[125] AT&T District, supra note 118 at 212.

[126] AT&T District, supra note 118 at 184.

[127] AT&T Circuit, supra note 119 at 6.

[128] Id.

I. Introduction

The rapid advancement of artificial-intelligence (AI) technologies has generated tremendous interest in how to regulate this emerging field. Implicit in the question are various assumptions about how AI will be developed and deployed in the marketplace. As it turns out, this is not nearly as simple a forecast as it may at first appear. This is particularly evident when one examines the differences between open-source and proprietary approaches to commercialization,[1] and is particularly relevant in light of the growing power costs needed to run large-scale models.[2]

As AI continues to transform industries and reshape our social and economic landscape, understanding the various commercialization strategies available to AI developers is crucial for policymakers, business leaders, and researchers alike.

This paper examines the diverse commercialization models for AI development, exploring their strengths, weaknesses, and potential implications for innovation and competition. Our analysis is grounded in the recognition that AI is not a monolithic technology, but rather a diverse array of techniques and applications with vastly different market potentials and societal impacts.

We begin by defining AI in the context of commercialization, highlighting the complexity and variety of technologies that fall under this umbrella term. We then explore four essential commercialization strategies that are likely to shape how AI will be deployed: proprietary private, open-source private, hybrid private/public, and publicly directed models.

Throughout our analysis, we emphasize that no single commercialization approach is universally superior. Instead, the optimal strategy depends on the specific context, including the nature of the AI technology, target market, and broader ecosystem in which it operates. We also consider how these commercialization choices intersect with and are shaped by regulatory decisions and public policy.

II. Defining AI in the Context of Commercialization

The term “artificial intelligence” has become ubiquitous in popular culture since the November 2022 launch of OpenAI’s ChatGPT made individual “generative artificial intelligence” capabilities broadly available to the general public.^[3] Indeed, ChatGPT was in its first year the fastest-growing consumer application in history.^[4]

But what do we mean by AI? The deceptively simple term remains obscure. As Jason Potts has noted: “Generative AI is not a single technology, or even industry, and is composed of extremely complex and varied ownership and governance at each composite layer.”^[5] Instead, we must consider the diverse ecosystems of AI technologies—from core technologies through foundation models and specific applications—and how they interact with various sectors and user communities.

Some economists characterize AI software as a “general purpose technology,” with some form of the technology today widely embedded in firms, as well as their products and services.^[6] Yet until the advent of ChatGPT, many consumers were unaware of AI’s pervasiveness, as it was largely hidden in the back offices of business and government. It was only in the last few years that consumers began to interact with AI technologies regularly, such as when they accessed their bank accounts, paid bills, made purchases online, or interacted with online assistants. The introduction of highly capable large language models (LLMs) has rapidly changed public perception of applied AI’s still-incipient capabilities. Specifically, there is now widespread recognition of the potential impact these technologies could have on society.

McKinsey & Co. offers a functional definition: “AI is a machine’s ability to perform the cognitive functions we associate with human minds, such as perceiving, reasoning, learning, interacting with an environment, problem solving, and even exercising creativity.”^[7] AI itself, however, consists of multiple technologies and designs, including such methodologies and applications as machine learning, deep learning, natural language processing, computational statistics, facial recognition, and robotics (among many others).^[8] This definition is quite expansive, and is perhaps unhelpful in its tendency to give the impression that “AI” refers to a single, unified concept. It includes everything from foundational models capable of producing human-like text to algorithms that can diagnose illnesses more precisely than human physicians to “simple” algorithms that enable more personalized online experiences.

When considering commercialization models for AI, it’s crucial to recognize that we are not dealing with a monolithic technology, but rather with the intersection of a variety of techniques and technologies that often serve very different purposes. From machine-learning algorithms used in predictive analytics to natural-language processing in chatbots, from computer vision in autonomous vehicles to reinforcement learning in robotics, each AI application may require a unique commercialization approach. This diversity in AI technologies means that no single business model will be universally applicable. Instead, successful commercialization strategies must be tailored to the specific capabilities, applications, and target markets of each AI technology, while also considering the broader ecosystem in which these technologies operate and interact.

III. Commercialization Models for AI Development

There are a variety of commercialization paths for AI developers. For our purposes, we will discuss four essential commercialization strategies, and related public policies that we believe are most likely to shape how AI will be deployed: proprietary private, open-source private, hybrid private/public, and publicly directed.

A. Private Models: Tradeoffs Between Proprietary and Open Approaches

1. Private-proprietary commercialization

A fully private business model for AI production would operate on a presumption that the technology that drives the AI development is proprietary, and the firm is free to leverage this proprietary technology along a spectrum of deployments, from a fully integrated/closed ecosystem to one that is relatively open to external integrators. This approach relies on a mix of different forms of intellectual-property protections and contractual obligations. For example, copyrights, patents, and trade secrets may form the basis for the core business model, while contractual relationships with integrators and other developers could then be used to facilitate functional integration (to varying degrees) via licensing.

A fully closed business model for AI production would leverage the firm’s proprietary technology and intellectual property to create a vertically integrated ecosystem. Under this model, the firm maintains tight control over the development, deployment, and commercialization of its AI technology.[9] The primary objective would be to capture value by creating a unique, differentiated offering that is difficult for competitors to replicate.

The foundation of this model rests on various forms of intellectual-property protection. Copyrights protect the original expression of ideas, such as the source code and documentation associated with the AI system.[10] Patents provide a time-limited exclusive property right over novel, non-obvious, and useful inventions, including (in some cases) AI algorithms and architectures.[11] Trade secrets protect confidential information that derives economic value from not being generally known, such as the training data, hyperparameters, data-center design, and optimization techniques used in the AI system.[12] Building on this intellectual-property foundation, the firm can then create a fully integrated ecosystem around its AI technology. This ecosystem may include proprietary hardware (e.g., specialized AI accelerators); software (e.g., development tools, APIs, and applications); and services (e.g., cloud-based AI platforms).

A license involves a mutually negotiated, contractual agreement—often employing a nondisclosure agreement during licensing negotiations—whereby the licensor retains title to the underlying IP but assigns a right to exploit the IP to a third-party licensee.[13] The licensee can typically make, use, and sell a product using the licensed AI technologies—either through an exclusive license or nonexclusive license—for specified financial terms or by paying a stream of predetermined royalties.[14] A similar arrangement can be struck for access to proprietary software through more restricted means—for instance, for providing consumption of services via an API.

Recent inquiries from federal regulators have raised concerns about the potential implications of various proprietary business models in the AI industry.[15] There is, however, little for lawmakers and enforcers to assume ex ante when considering whether firms should pursue an open or closed proprietary strategy. As federal regulators grapple with the challenges of overseeing this rapidly evolving industry, it is crucial to recognize that the commercialization choices firms make are a reflection of the unique contingencies of their production processes. Firms often blend elements of both open and closed models to strike a balance that aligns with their goals, market dynamics, and consumer preferences.

Open approaches to AI development (e.g., offering interoperable services) offer numerous benefits, such as fostering innovation, enabling scientific research, and promoting competition in the marketplace. By making foundational AI models accessible and adaptable, open approaches can lower entry barriers and encourage a diverse ecosystem of applications tailored to various needs and challenges. This openness can drive technological advancements, price efficiencies, and ultimately benefit consumers and society at-large.

But closed approaches (also known as “walled gardens”) also have their merits. Closed proprietary development can provide enhanced security, privacy, and a more streamlined user experience. Closed models may also allow firms to maintain greater control over their intellectual property, ensuring a level of competitive advantage and providing incentives to invest in research and development (R&D). Under the right conditions, closed systems can likewise foster a healthy ecosystem of complementary products and services.

Indeed, ecosystem development is not really an either/or matte, as even closed/proprietary approaches to monetizing AI can have downstream “market making” effects on AI development:

Licenses are…a direct rent, and a way to gate or control competition through their issuance. Additionally, the terms of these licenses influence subsequent related markets, such as product embeddings, thereby also governing the development of business ecosystems, typically by restricting their growth or creating bottlenecks. A critical aspect of using licenses to manage rents and ecosystems relates to the discovery of value and the property rights that accrue from opportunities identified by third parties or users. The implication of a right, akin to an option, within the license aligns with real options theory. Effectively, the license determines where the financial value of these strategic real options is capitalized, whether in the licensor or the licensee.[16]

Moreover, whether any given ecosystem is “open” or “closed” is not a binary question:

[AI] technology is a cumulative result of decades of research and development by individuals and teams that have built extremely powerful and often valuable capabilities and products. Some of this work circulates in public (open-source code), but also in publications, open forums, institutions (universities), labour markets (hires) and financial markets (acquisitions). Still, many elements remain closed and protected with intellectual property and corporate secrecy (algorithms, private training data), and through tacit knowledge (especially in training). So, its development through innovation is both open and closed, simultaneously public and private. It is institutionally complex. [17]

The optimal choice between open and closed models depends on the relevant market participants’ specific needs and preferences. As Jonathan M. Barnett notes in the context of digital platforms, open systems may yield no net social gain over closed systems; can pose a net social loss under certain circumstances; and can impose a net social gain under yet other circumstances.[18] According to Barnett, there is a “fundamental welfare tradeoff between two-sided proprietary…platforms and two-sided platforms which allow ‘free entry’ on both sides of the market.”^[19] Consequently, “it is by no means obvious which type of platform will create higher product variety, consumer adoption and total social welfare.”^[20]

The interplay between consumer and developer preferences significantly influences the adoption of open versus closed AI models. Consumers often prioritize ease-of-use, security, and seamless integration, which can favor closed ecosystems. But some users may place greater value on the flexibility and customization options of open systems. For instance, while Linux is popular among developers for its openness, consumer-oriented operating systems like Windows and MacOS dominate the desktop market due to their user-friendly interfaces and extensive software ecosystems.^[21]

These dynamics give rise to a complex landscape in which the success of a model depends on its ability to balance the sometimes conflicting needs of both consumers and developers. Ultimately, the coexistence of open and closed models in the AI ecosystem fosters innovation and provides diverse options to meet varying user requirements. Furthermore, the distinctions between open and closed models can significantly influence competition between brands. If public policy is enacted that favors one business model over another, it would likely lead to reduced choice and lower overall benefits along this dimension.[22]

The diffusion of firms exploring different business models yields benefits in many ways. For example, when it comes to ethical AI development, various types of organizations should be encouraged to explore different approaches, and even to combine them when appropriate. Anthropic’s “Collective Constitutional AI” approach adopts a “semi-open” model that combines private elements with certain aspects of openness in order to encourage innovation, while maintaining some degree of control.[23] This model may strike a suitable balance by ensuring a degree of proprietary innovation and competitive advantage, while still benefiting from community feedback and collaboration.

Conversely, a completely open-source approach to development could yield different, and possibly superior, outcomes that address a wider range of needs through community-driven evolution and iteration. It is impossible to determine, in advance, whether an open or closed approach to AI development will inherently lead to better results in creating “ethical” AI. Both approaches have their merits, and the most effective solutions will likely incorporate elements of both.

By differentiating themselves through a focus on ease-of-use, quality, security, and user experience, closed systems contribute to a vibrant competitive landscape in which consumers have clear choices between differing “brands” of AI. In essence, codifying a regulatory preference for one proprietary business model over the other would oversimplify the intricate balance of tradeoffs inherent to AI development.

2. Private/open-source commercialization

The next flavor of private AI commercialization models is open source. This is distinct from the “open vs. closed” private models discussed in the previous subsection. In contrast to the restrictions inherent in the proprietary models discussed above, “open source” is a decentralized software-development model that encourages open collaboration among potential developers.[24] The term generally refers to source code made freely available to individuals and companies to modify and redistribute, and includes legal permission to use the source code, design documents, or product content.[25] Source code is typically released under the terms of a software license, which typically allows other users to download, modify, and subsequently publish their enhanced versions.[26]

Stable Diffusion, an open-source text-to-image generation model developed by Stability AI, is a prominent example of open-source generative AI.[27] The model was trained on a large dataset of images and their associated captions, allowing it to generate novel images from textual descriptions. The source code, model weights, and training data were made freely available, enabling developers to experiment with and build upon the technology.[28] Another prominent example is PyTorch, an open-source machine-learning framework led by Meta’s AI Research lab.[29] PyTorch provides a flexible and intuitive interface for defining and training neural networks. Other notable open-source AI projects include TensorFlow, Keras, Scikit-learn, and OpenCV.[30] These frameworks and libraries provide powerful tools to develop machine-learning and computer-vision applications, lowering the barrier to entry for AI development and enabling rapid prototyping and experimentation.[31]

While there are more than 100 Open Source Interconnection (OSI) licenses, for our purposes, we will categorize them into two typologies: permissive and copyleft.[32] An open source permissive license governs how others (i.e., those other than the copyright holder) can inspect, use, modify, distribute (and freely redistribute), or enhance software code.[33] The open-source software licensor has the option to change the terms and conditions to grant the licensee the permission and rights to use or repurpose the code for new applications or to include the code in other derivative work.[34]

Users of open-source licenses must accept the legal terms of a license, but the terms of an open-source license differ from those of a proprietary license.[35] For example, in the open source copyleft AI license, the licensor grants the licensee permission to use, modify, and share the source code, but offers stipulations against relicensing through specific terms and conditions in the agreement, including that any derivative work is required to be released under copyleft license terms and conditions identical to those of the original work.[36]

There are numerous examples of open-source licenses, including Standard License (“non-copyleft”); General Public License (“strong copyleft”); Lesser General Public License (“weak copyleft”); Eclipse License (“weak copyleft”); Mozilla Public License (“weak copyleft”); Apache License (“non-copyleft”); MIT License (“non-copyleft”); and Berkeley Source Distribution (BSD) License (“non-copyleft”).[37]

A common example of permissive licensing can be found in the Creative Commons (CC) suite of licenses. Creative Commons offers a range of standardized licenses that creators can apply to their work, each with different levels of permissiveness.[38] While not strictly open source in the software sense,[39] these licenses share similar principles of openness and accessibility. A typical Creative Commons license allows the license holder to grant broad permissions for others to share, remix, use commercially, or otherwise utilize the license holder’s work without seeking specific authorization for each use.[40] This flexibility makes Creative Commons licenses a popular choice for various types of content—from academic papers to multimedia works—fostering a culture of sharing and collaboration similar to that seen in open-source software communities.

The Creative Commons license works alongside the rules of copyright, allowing the owner to authorize a license base from more strict to more open usage of the owner’s work, and thus to choose the level of protection that best suits the owner’s needs.[41] Under the Creative Commons license, there are six primary variants: attribution, attribution share alike, attribution no derivatives, attribution non-commercial, attribution non-commercial share alike, and attribution non-commercial no derivatives.[42] There are, however, also variants of the Creative Commons license (such as “commercial use is forbidden” and “derivatives are forbidden”) that are not OSI-approved.[43]

Meta’s Llama 3 is a good example of an AI system released permissively under a quasi-open-source license.[44] Like traditional open-source software, it allows for free use, modification, and redistribution of the model and its code.[45] But it diverges from standard open-source practices by including specific restrictions—such as prohibiting the use of Llama materials to improve competing LLMs and requiring prominent attribution—that are not typically found in open-source licenses.[46]

More broadly, an open-source-specific AI license would likely incorporate elements from existing open-source software licenses, while addressing the unique characteristics of AI systems. Key components might include:

1. Granting users the freedom to study, use, modify, and share the AI system and its components, including source code, model weights, and training data;
2. Requiring any derivative works or modifications to be released under the same open-source license terms, ensuring the continued openness of the technology (copyleft provisions);
3. Permitting commercial use of the AI system, fostering its adoption and integration into real-world applications;
4. Requiring attribution to the original creators and contributors, acknowledging their work and promoting collaboration; and
5. Disclaiming warranties and limiting liability, protecting the licensors from potential legal issues arising from the use or misuse of the AI system.[47]

The development of AI systems under an open-source model offers both benefits and challenges. On the positive side, open-source AI can promote transparency, collaboration, and rapid innovation. By making the underlying technology accessible to a wide range of developers and researchers, open-source AI enables diverse perspectives and approaches, leading to more robust and adaptable systems. Open-source development can also foster trust, as the AI system’s inner workings can be scrutinized and verified by the community.[48]

But open-source AI also presents legal and economic challenges. The lack of proprietary control over the technology may discourage commercial investment, as companies may struggle to monetize their efforts.[49] A flip side of the benefits of the open nature of the software is that, in some cases, malicious actors can access and modify the technology for harmful purposes.[50] Additionally, the open nature of the development process may complicate issues of liability and accountability, as the responsibility for any the AI system’s negative consequences may be diffused among multiple contributors.

To balance these tradeoffs, a hybrid approach that combines open-source and proprietary elements may be optimal. For example, companies could release certain components of their AI systems as open source while maintaining control over others, thus enabling collaboration and transparency while preserving competitive advantages.[51]

B. Hybrid Models: The Role of Standards

The third model for commercialization comprises various forms of hybrid public/private standards. The fair, reasonable, and non-discriminatory (“FRAND”) licensing approach is one example common in some patent contexts. FRAND licensing involves a voluntary agreement between a holder of a standards-essential patent (SEP) and a manufacturer or implementer, with the agreement allowing the implementer to use the patentee’s technology in its manufactured products.[52] Established by standards-setting organizations (SSOs) to ensure technical compatibility and interoperability in their products,[53] these SEPs are “patents that are considered essential to implement a specific industry standard” and are “governed by the terms and conditions of the licensing agreement.”[54]

The fair and reasonable part of FRAND means that “the patent holder should charge a licensing fee that reflects the economic value of the patent,” while “taking into account factors such as the importance of the patent to the standard, the contributions of the patent to the product, and the prevailing licensing rates in the industry.”[55] The non-discriminatory part of FRAND means that “the patent holder should offer the same terms to all companies wanting to license the patent,” thus “preventing the patent holder from discriminating against certain companies.”[56]

As it has been traditionally understood, however, patenting the core technologies underlying AI is exceedingly difficult under prevailing U.S. law.[57] It is therefore unlikely that a true FRAND approach is possible in the United States, although that doesn’t rule out that other forms of government involvement could be used to establish a similar forced-sharing regime. For example, a weaker form of this model would be to create “standards” to guide the development and commercialization of AI technologies, either through a direct governmental standards-setting approach, or a hybrid public/private model that involves government-led and/or enforced standards development.[58]

Standards setting, either on a private basis or through government-led initiatives, can play a role in shaping development practices and ensuring safety and interoperability across technologies in a way that can potentially lead to more commercialization options as a result of elevated consumer trust.  As is always true, however, there are both benefits and costs when thinking about the way that public or public/private standards shape commercialization decisions.

These standards can not only help to organize the industry by establishing common frameworks and protocols but can also enhance public trust by providing assurances that AI development is under diligent oversight. As AI systems increasingly permeate such critical sectors as health care, transportation, and finance, the perceived need for assurance regarding their safety, reliability, and ethical implications grows. Commercialization of these technologies is necessarily shaped by these realities.

These standards can originate in a number of ways. Firms can develop standards as part of the process of instituting their own internal best practices.[59] Large professional organizations also contribute important work in this area. For example, the Institute of Electrical and Electronics Engineers (IEEE) has been working to develop an “Ethics Certification Program for Autonomous and Intelligent Systems,” the goal of which is to “create specifications for certification and marking processes that advance transparency, accountability and reduction in algorithmic bias in Autonomous and Intelligent Systems (AIS).”[60] The IEEE also has a number of similar efforts underway.[61]

In 2017, the International Organization for Standardization (ISO) and the International Electrotechnical Commission (IEC) established the ISO/IEC JTC 1/SC 42 Artificial Intelligence Committee,[62] tasked with developing universal AI standards that would apply across different industries and applications. Their current projects include creating a comprehensive interoperable framework for AI systems.[63] The Consumer Technology Association (CTA) develops trustworthiness standards to address the use of AI in health care,[64] assessing the impact of AI’s trustworthiness from multiple perspectives—including those of physicians, consumers, professional and family caregivers, public-health officials, medical societies, and regulators.[65]

Since 2019, the National Institute of Standards and Technology (NIST) has actively contributed to AI governance, recommending strategies for the federal adoption of AI to enhance both the U.S. economy and national security. A February 2019 White House executive order directed NIST to promote the use of voluntary standards in AI,[66] coordinate with standards development organizations via the National Science and Technology Council,[67] and engage with the private sector on emerging AI standards.[68]

More recently, NIST developed the AI Risk Management Framework (AI RMF), a set of guidelines to improve trustworthiness in AI design and use, emphasizing consensus-driven, outcome-focused, and nonprescriptive standards.[69] The recently proposed bipartisan U.S. Senate framework also contains many of these elements, recommending studies where laws are insufficient to capture AI systems’ potential harms but otherwise recommending federal coordination on standards with largely voluntary compliance.[70]

On the more extreme side, there are calls to create a central AI regulator, with the proposed regulator having authority to grant licenses before AI development can be undertaken, and enforce ongoing compliance.[71] This centralization aims to standardize oversight and ensure adherence to established standards across the AI industry.[72]

It’s crucial to recognize that nominally voluntary standards can effectively become mandatory in practice. This can occur, for example, when courts adopt these standards in negligence litigation as benchmarks to establish a duty of care.[73] Thus, even in the absence of any direct regulatory mandates or common agreement, standards often shape legal obligations and liabilities, influencing practices across industries by establishing a de-facto regulatory framework.

While we should not understate the benefit of some government participation in developing AI standards, the introduction of mandatory standards can carry substantial costs, depending on their scope and content. Applying the schema proposed by Knut Blind et al. to AI commercialization, we can consider how the impact of standards and regulations might vary depending on market uncertainty.[74] In markets with low uncertainty, formal standards may lead to higher innovation costs than regulations would.[75] This could be due to regulatory capture, where established firms influence standards to their advantage, potentially raising costs for competitors.[76] Conversely, in highly uncertain markets, regulations might impose higher innovation costs than standards. This may occur because regulators have less information about rapidly evolving technologies, leading to potential mismatches between regulations and technological realities.[77]

A good example of the double-edged sword of government involvement in standards setting can be seen in the campaign to require interoperability and compatibility among diverse technology providers.[78] Standardization can help to reduce fragmentation, thus facilitating smoother integration and interaction between different AI products and services. By fostering a common understanding and predictable environment, these standards might aid in accelerating technological advancements and innovation within a structured and well-defined boundary. But as noted above, a more open, interoperable business model is not a panacea, and it’s not at all clear that regulation-driven interoperability standards are the best way to reduce friction among various AI systems. There are tradeoffs in pursuing varying degrees of openness, whether imposed by fiat or market demand.

Moreover, government-led standards can restrict AI firms from fully capitalizing on unique, proprietary technologies by mandating uniformity that may dilute the distinctiveness of more innovative or advanced solutions. This can discourage investment in new technologies that don’t conform to existing standards, potentially stunting growth and innovation within the sector. Firms often leverage proprietary advantages for competitive gain and market differentiation; thus, stringent standards could limit their ability to offer something unique to the market, reducing the incentive for groundbreaking developments.

Further, adhering to mandatory AI standards can result in significant compliance costs, particularly affecting smaller firms and startups with limited resources.[79] These costs include not only the financial burden of implementing new systems and processes to meet the standards but also the ongoing expenses associated with maintaining compliance. This financial strain can divert funds from other critical areas like product development and research, potentially stifling innovation and limiting the company’s growth potential in a highly competitive market.

Standards may also significantly limit AI developers’ flexibility to engage in creative and technical decisionmaking. When standards dictate or prohibit specific aspects of AI system design and functionality, developers might find their hands tied, unable to explore alternative or potentially more innovative solutions that fall outside the set guidelines.[80] This could slow the pace of innovation in AI development, as companies might be forced to prioritize compliance over creativity and innovation.

C. Public Models of Commercialization

The fourth approach would be some form of compulsory-licensing or forced-sharing regime that creates a public commercialization model.[81]  The term “compulsory license” refers to a governmental entity that provides permission to enterprises that seek to use intellectual property without the owner’s consent, while offering financial compensation to the rightsholder.[82] Advocates for this licensing approach favor “policies that shape US markets while aligning innovation and competition policies towards creating a domestic AI innovation system that better serves the public interest.”[83]

The exact shape such a policy might take is highly contingent on the forms of intellectual property employed, and the conditions under which such use is deemed necessary. For example, the World Trade Organization’s Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS Agreement) permits member nations to institute compulsory licensing of patents under certain circumstances.[84] Compulsory licenses for patents are most often applied to pharmaceuticals and public-health-related inventions, but may apply to any patented invention.[85]

In the context of copyright, Congress has seen fit to institute compulsory licenses in certain instances, such as satellite and cable retransmission[86] and the use of musical compositions in recording “cover songs.”[87] The compulsory license is a concept that has been formally part of U.S. intellectual-property law since the enactment of the U.S. Copyright Act of 1909.[88] U.S. patent law has not paralleled copyright law, however, as compulsory licenses have been uncommon in the patent arena.[89]

Additionally, Congress is able to place compulsory-license stipulations on funding it provides to researchers and developers. The U.S. government—representing all U.S. taxpayers—has historically played a role in funding R&D to develop the nation’s computer-technology infrastructure. This extends to AI technologies and efforts to bring these AI technologies to the marketplace.[90] Moreover, public-interest advocates who push for broader “fully public” models often argue that their approach will “better align domestic investment and AI capability development with economic, societal and national security objectives.”[91]

This fully public commercialization policy might, for example, focus on a nationwide health emergency (such as developing a public-health vaccine) or some nonprofit purposes (such as meeting an environmental-policy goal). Indeed, the Biden administration indicated its willingness to use the Defense Production Act (DPA) to intervene in how AI is developed and produced.[92] Some advocates have also raised the threat of eminent domain-like actions against intellectual property.[93]

Relatedly, some recent public discourse has focused on the possibility that AI companies might obtain monopoly status, with attendant conversation about the need to act to deter this likelihood.[94] This is another flavor of treating core AI technologies as some sort of public good or essential facility that would be fully subject to governmental licensing. In many respects, such advocacy closely mirrors conversations from a decade ago about “data monopolies.”[95] But as Geoffrey Manne and Dirk Auer suggest, this characterization is overly simplistic and potentially misleading.[96] History has shown that, despite the leading tech platforms’ vast accumulations of data, their market dominance in the realm of generative AI has not been as unassailable as one might expect.[97]

The concept of market “moats” (a term used to describe barriers to entry that protect incumbents from competitive pressures) is central to this discussion. While the urge to dismantle these moats through aggressive antitrust actions is strong in some quarters, pursuing this urge would likely be a mistake. The rapid advancement and dynamic nature of AI technologies often render these moats less impenetrable than they might appear, and their existence is unlikely to be supported by rigorous empirical examinations.[98]

Moreover, using competition law to force open the business models of leading AI firms carries significant risks. Chief among these is that such policies could stifle innovation and impede the growth of the AI sector, rather than foster competitive markets. Of course, it’s not impossible that competition harms might emerge in the AI industry, as they can in any industry, but the assertions that these firms and their products should be treated as anything like an essential facility or public utility should be subject to heightened scrutiny in order to avoid high error costs and reduced innovation and consumer welfare.

A mandated forced-sharing regime would be a special (and extreme) case of the “open” business models discussed above. Thus, it’s worth considering some of the recent history of governmental interoperability requirements, and their effects on the commercialization paths that firms take.

Open-banking initiatives are a prime example. The regulatory push in the United States fundamentally aims to democratize data access within the financial sector by enabling customers to have control over their financial data, promoting a shift towards more customer-centric financial services.[99] This transition is, however, fraught with complexities that could undermine its intended benefits.[100]

One significant issue with the implementation of open banking relates to the technical and operational challenges involved in ensuring seamless and secure data sharing. The infrastructure required to support secure APIs that allow third parties to access bank data must be robust, resilient, and effective.[101] There is a risk that inconsistencies in the technological capabilities of different financial institutions and third-party providers could lead to vulnerabilities in the system that expose sensitive data or cause service disruptions. Further, in at least some implementations of open banking, obligations are placed on providers in ways that distort the competitive landscape.[102]

So-called “device neutrality” or “sideloading” represents another cautionary tale for forced openness. Article 6(1)(c) of the EU’s Digital Markets Act (DMA) addresses “sideloading”—the installation of third-party software through app stores other than the one provided by the manufacturer (e.g., Apple’s App Store for iOS devices).[103] The sideloading mandate aims to give users more choice but can only achieve this by removing the option of choosing a device with a “walled garden” approach to privacy and security, such as Apple’s iOS.[104]

By eliminating the choice of a walled-garden environment, a sideloading mandate essentially forces users to use alternative app stores preferred by app developers, who have incentives to create their own app stores or move to those with the least friction. This may also mean the app stores that invest least in privacy and security. Apple is well-known for its commitment to user privacy and security. Thus, regulations like the DMA and similar attempts in the United States[105] represent a reordering of digital-ecosystem providers that currently have the incentive to account for a broad array of consumer demands. In the place of such nuance, mandated openness overrides the preferences of a vast swathe of consumers.

As suggested above, implementing strong forced-sharing obligations could severely limit the commercialization paths available to AI developers. One of the primary challenges with compulsory licensing is the potential disruption to proprietary business models. AI firms often rely on exclusive control over their intellectual property to maintain a competitive edge and secure funding for further R&D. Mandating that these firms license their technology to third parties could dilute that competitive advantage, making it less attractive for investors to fund high-risk, high-reward AI projects. This could ultimately slow the pace of innovation, as firms might be less willing to invest in groundbreaking technologies if they are not able to fully capitalize on their proprietary developments.

Additionally, imposing forced sharing could lead to a homogenization of AI offerings, reducing the diversity and specialization that currently characterizes the AI market. AI firms tailor their technologies to specific market needs, while forced-sharing regimes could undermine this customization by requiring firms to share their innovations broadly—including potentially with competitors. This could lead to a scenario where AI technologies become more standardized and less differentiated, which might diminish the overall quality and effectiveness of AI solutions available to consumers.

Furthermore, the administrative and regulatory burden associated with enforcing compulsory licensing could create significant operational challenges for both AI firms and regulators. Ensuring compliance with licensing agreements or the requirements of a strong regulatory regime, monitoring the use of an incredibly wide scope of shared technology, and resolving disputes over access would require substantial resources and oversight. This could divert attention and resources away from innovation and toward regulatory compliance, further hindering the growth and development of the AI sector.

IV. Conclusion

The commercialization of AI technologies offers a complex landscape of opportunities and challenges. As we have explored, there is no one-size-fits-all approach to bring AI innovations to market. Instead, a variety of models—from fully proprietary to open source, and hybrid approaches in between—each offer distinct advantages and tradeoffs.

Proprietary models allow firms to maintain tight control over their intellectual property, potentially driving focused innovation and creating unique value propositions. These closed systems may, however, limit broader collaboration and slow the overall pace of advancement in the field. On the other hand, while open-source approaches could foster transparency, collaboration, and rapid innovation, they also present challenges in the difficulty inherent in finding sustainable business models and exercising appropriate quality control.

Hybrid models and industry standards attempt to strike a balance, promoting interoperability and shared progress while still allowing for proprietary advantages. These approaches, however, require careful management to avoid stifling innovation or inadvertently creating barriers to entry for smaller players.

Public models of commercialization, including compulsory licensing and government-directed development, offer potential benefits in terms of equitable access and alignment with public-interest goals. But they also risk dampening private-sector innovation and investment if not carefully implemented.

The diversity of AI technologies and their applications means that different commercialization strategies may be appropriate in different contexts. What works for a general-purpose LLM may not be suitable for a specialized medical diagnostic tool or a financial-trading algorithm.

As the AI landscape continues to evolve, it is crucial that policymakers and industry leaders remain flexible and adaptive in their views of commercialization (with perhaps the fully public models being the only ones presumed inappropriate, pending strong evidence to the contrary). Regulatory frameworks should aim to foster innovation and competition while addressing legitimate concerns about safety, privacy, and the ethical use of AI. They should avoid prematurely locking in any single commercialization model, instead allowing for experimentation and learning as the field matures.

Ultimately, the success of AI commercialization will depend on finding the right balance between openness and proprietary development, between rapid innovation and responsible deployment, and between market forces and public interest. By understanding the strengths and limitations of various commercialization models, stakeholders can make more informed decisions that will shape the future of AI and its impact on society.

[1] See infra at nn. 9-80, and accompanying text.

[2] See Lynne Kiesling, Data Center Electricity Use III: Make or Buy?, Knowledge Problem (Aug. 1, 2024), https://knowledgeproblem.substack.com/p/data-center-electricity-use-iii-make.

[3] Introducing ChatGPT, OpenAI (Nov. 30, 2022), https://openai.com/blog/chatgpt.

[4] Laurie A. Harris, Generative Artificial Intelligence: Overview, Issues, and Questions for Congress, Cong. Research Serv. (2023), at 1, available at https://crsreports.congress.gov/product/pdf/IF/IF12426.

[5] Jason Potts, Sources of Innovation in Generative AI, The Network L. Rev. (Feb. 12, 2024), https://www.networklawreview.org/jason-potts-generative-ai.

[6] Nicholas Crafts, Artificial Intelligence as a General-Purpose Technology: An Historical Perspective, 37 Oxford Rev. Econ. Pol’y 521 (2021), https://doi.org/10.1093 oxrep /grab012. It is not, however, absolutely clear how to regard “AI.” Indeed, as we note below, there is no single technology that can be called “AI.” What exists are a varied collection of techniques and technologies that add some form of intelligence to automated systems.

[7] What Is AI, McKinsey & Co. (April 24, 2023), https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-is-ai.

[8] Id.; Matt Vella, How AI Is Transforming Our World, in Artificial Intelligence: The Future of Humankind, Time (Nov. 1, 2017); Michael Atleson, Keep Your AI Claims in Check, Fed. Trade Comm’n (Feb. 27, 2023), https://www.ftc.gov/business-guidance/blog/2023/02/keep-your-ai-claims-check.

[9] For an example of this approach, see Apple Intelligence: AI for the Rest of Us, Apple, https://www.apple.com/apple-intelligence (last visited Aug. 2, 2024).

[10] See Pamela Samuelson, The Uneasy Case for Software Copyrights Revisited, 79 George Wash. L. Rev. 1746 (2011);Copyright could theoretically be invoked to protect models’ parameter-embedding weights as a means to control distribution. See Thibault Schrepel & Jason Potts, Measuring the Openness of AI Foundation Models: Competition and Policy Implications (Sciences Po Digital, Governance and Sovereignty Chair, Working Paper 11, 2024), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4827358.

[11] See generally Adam Mossoff, Why History Matters in the Patentable Subject Matter Debate, 64 Fla. L. Rev. 23 (2012).

[12] See David S. Levine & Ted Sichelman, Why Do Startups Use Trade Secrets?, 94 Notre Dame L. Rev. 751 (2019).

[13] See, e.g., Patent Licensing & Legal Options, Justia, https://www.justia.com/intellectual-property/patents/licensing (last visited Aug. 2, 2024); Nathan C. Lovette & Carlos P. Garritano, Patent Licensing: A Brief Summary, Tucker Ellis LLP (Aug. 16, 2022), https://www.tuckerellis.com/ip-tip-of-the-month-blog/patent-licensing-a-brief-summary; Proprietary Software License: Everything You Need to Know, Upcounsel, https://www.upcounsel.com/ proprietary-software-license (last visited Aug. 2, 2024).

[14] Patent Licensing and Its Types: Everything You Need to Know, GreyB, https://www.greyb.com/blog/patent-licensing-101 (last visited Aug. 2, 2024).

[15] See, e.g., Kristian Stout, ICLE Comments to NTIA on Dual-Use Foundation AI Models with Widely Available Model Weights, Int’l Ctr. L. & Econ. (2024), https://laweconcenter.org/resources/icle-comments-to-ntia-on-dual-use-foundation-ai-models-with-widely-available-model-weights.

[16] Schrepel & Potts, supra note 10, at 10-11.

[17] Potts, supra note 5, at 2.

[18] See Jonathan M. Barnett, The Host’s Dilemma: Strategic Forfeiture in Platform Markets for Informational Goods, 124 Harv. L. Rev. 1861, 1927 (2011).

[19] Id. at 2.

[20] Id. at 3.

[21] Linux adoption rates consistently trail MacOS and Windows. See Desktop Operating System Market Share Worldwide Feb 2023 – Feb 2024, StatCounter, https://gs.statcounter.com/os-market-share/desktop/worldwide (last visited Aug. 2, 2024); Nonetheless, many developers have a preferences for Linux systems. See Joey Sneddon, More Developers Use Linux than Mac, Report Shows, Omg Linux (Dec. 28, 2022), https://www.omglinux.com/devs-prefer-linux-to-mac-stackoverflow-survey.

[22] See Michael L. Katz & Carl Shapiro, Systems Competition and Network Effects, 8 J. Econ. Persp. 93, 110 (1994), (“[T]he primary cost of standardization is loss of variety: consumers have fewer differentiated products to pick from, especially if standardization prevents the development of promising but unique and incompatible new systems”).

[23] Collective Constitutional AI: Aligning a Language Model with Public Input, Anthropic (Oct. 17, 2023), https://www.anthropic.com/news/collective-constitutional-ai-aligning-a-language-model-with-public-input.

[24] Sheen S. Levine & Michael J. Prietula, Open Collaboration for Innovation: Principles and Performance, 25 Organization Science 1414 (2014), https://doi.org/10.1287/orsc.2013.0872.

[25] The Open Source Definition, Open Source Initiative https://opensource.org/osd (last visited Aug. 2, 2024).

[26] Id. at 18.

[27] Stable Diffusion Public Release, stability.ai, https://stability.ai/news/stable-diffusion-public-release (last visited Aug. 2, 2024).

[28] Id.

[29] Learn the Basics, PyTorch, https://pytorch.org/tutorials/beginner/basics/intro.html (last visited Aug. 2, 2024)

[30] See Tim Mucci, Five Open-Source AI Tools to Know, IBM (Dec. 15, 2023), https://www.ibm.com/blog/five-open-source-ai-tools-to-know.

[31] See Robert Gorwa & Michael Veale, Moderating Model Marketplaces: Platform Governance Puzzles for AI Intermediaries, 16(2) Law Innovation & Technology 9-10 (2024), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4716865 (noting how Hugging Face “is developing a business model where it can bundle additional ‘premium’ deployment features, lowering the  barrier to entry for less technical users or those simply seeking convenience. For instance, one product, ‘Inference Endpoints,’ advertises itself as a way to ‘deploy models in minutes’ on Hugging Face’s own infrastructure. Similarly, their ‘AutoTrain’ product allows one to ‘train, evaluate and deploy state-of-the-art Machine Learning models’ by simply uploading a dataset, without having to write a single line of code.”).

[32] Open Source License: Types and Comparisons, Snyk, https://snyk.io/learn/open-source-licenses (last visited Aug. 2, 2024); 5 Potential Risks of Open Source Software, Snyk, https://snyk.io/learn/risks-of-open-source-software (last visited Aug. 2, 2024).

[33] Jeanelle Horcasitas, Understanding Open-Source Software Licenses, DigitalOcean (Nov. 2, 2021), https://www.digitalocean.com/community/tutorials/understanding-open-source-software-licenses; What Is Open Source, OpenSource.com, https://opensource.com/resources/what-open-source (last visited Aug. 2, 2024).

[34] Id.

[35] Id.

[36] Id.

[37] Sylvan Leroux, Open Source Licenses Comparison [Guide], It’s Foss (Sep. 9, 2023), https://itsfoss.com/ open-source-licenses-explained; Gina Häußge, A Dev’s Guide to Open Source Software Licensing, GitHub https://github.com/readme/guides/open-source-licensing (last visited Aug. 2, 2024).

[38] Id.

[39] While some Creative Commons licenses (like CC0 and CC BY) can be considered “open” in a broad sense, they are not recognized as open-source licenses by the Open Source Initiative, the body that formally approves open-source software licenses. See About, Open Source Initiative, https://opensource.org/about (last visited Aug. 2, 2024).

[40] Michelle Kaminsky, What Is Creative Commons? 5 Frequently Asked Questions, LegalZoom (Mar. 27, 2023), https://www.legalzoom.com/articles/what-is-creative-commons-5-frequently-asked-questions.

[41] Id.; Kela Parker, Creative Commons License – Pros and Cons Explained, Studiobinder (May 28, 2023), https://www.studiobinder.com/blog/creative-commons-license.

[42] Id.

[43] See Häußge, supra note 37.

[44] Meta Llama 3 Community License Agreement, Meta (Apr. 18, 2024), https://llama.meta.com/llama3/license.

[45] Id.

[46] Id.

[47] See, e.g., Andrew M. St. Laurent, Understanding Open Source and Free Software Licensing (2004), available at https://people.debian.org/~dktrkranz/legal/Understanding%20Open%20Source%20and%20Free%20Software%20Licensing.pdf.

[48] Niklas Leicht, Given Enough Eyeballs, All Bugs are Shallow – A Literature Review for the Use of Crowdsourcing in Software Testing, Scholar Space (2018), http://hdl.handle.net/10125/50404.

[49] Josh Lerner & Jean Tirole, The Economics of Technology Sharing: Open Source and Beyond, 19(2) J. Econ. Perspectives 99, 105 (2005), (“Because firms do not capture all the benefits of the investments in the open source project, however, the free-rider problem often discussed in the economics of innovation should apply here as well”).

[50] Liam Tung, Open Source: Almost One in Five Bugs Are Planted for Malicious Purposes, ZDNet (Dec. 3, 2020), https://www.zdnet.com/article/open-source-software-how-many-bugs-are-hidden-there-on-purpose.

[51] In many cases, this will be a commercialization strategy that tries to quickly build a set of complimentary products supporting the firm’s primary revenue source. Joel West & Scott Gallagher, Challenges of Open Innovation: The Paradox of Firm Investment in Open-Source Software, 36(3) R&D Mgmt. 13-14 (2006). Apple, for example, released Swift and WebKit as open-source projects at least partly as a strategy to encourage development of software generally—and web apps, specifically—that would work well for MacOS- and iOS-based hardware products. See Open Source at Apple, Apple, https://opensource.apple.com (last visited Aug. 2, 2024).

[52] What Is FRAND Licensing?, Lerman Law Associates P.C. (Jun. 26, 2023), https://www.lermanlawpc.com/blog/ 2023/06/what-is-frand-licensing.

[53] Herbert Hovenkamp, FRAND and Antitrust, 106(1) Cornell L. Rev. (2020), https://www.cornelllawreview.org/2020/09/15/frand-and-antitrust.

[54] Garth Brian Hedenskog, What Is a Standard Essential Patent (SEP)? SHIP Global Intellectual Property (Oct. 29, 2019), https://shipglobalip.com/blog/what-is-a-standard-essential-patent-sep-.

[55] Id.

[56] Id.

[57] See Kevin Madigan & Adam Mossoff, Turning Gold to Lead: How Patent Eligibility Doctrine Is Undermining U.S. Leadership in Innovation, 24 George Mason L. Rev. 939, 955 (2017), https://ssrn.com/abstract=294343; see also Eric Schmidt et al., Final Report, National Security Commission on Artificial Intelligence, NSCAI (2021), at 201-202, available at https://www.dwt.com/-/media/files/blogs/artificial-intelligence-law-advisor/2021/03/nscai-final-report–2021.pdf.

[58] See U.S. Leadership in AI: A Plan for Federal Engagement in Developing Technical Standards and Related Tools, NIST (2019), available at https://www.nist.gov/system/files/documents/2019/08/10/ai_standards_fedengagement_plan_9aug2019.pdf; see also AI Risk Management Framework, NIST, https://www.nist.gov/itl/ai-risk-management-framework (last visited Aug. 2, 2024).

[59] Navdeep Gill, Abhishek Mathur, & Marcos V. Conde, A Brief Overview of AI Governance for Responsible Machine Learning Systems, arXiv (Nov. 21, 2022), https://arxiv.org/pdf/2211.13130.

[60] The Ethics Certification Program for Autonomous and Intelligent Systems (ECPAIS), IEEE SA, https://standards.ieee.org/industry-connections/ecpais (last visited Aug. 2, 2024)

[61] See also IEEE Standard Model Process for Addressing Ethical Concerns during System Design, IEEE SA, https://standards.ieee.org/ieee/7000/6781 (last visited Aug. 2, 2024); 7000-2021 – IEEE Standard Model Process for Addressing Ethical Concerns during System Design, IEEE Xplore, https://ieeexplore.ieee.org/document/9536679 (last visited Aug. 2, 2024), (Aiming to develop organizational standards for incorporating ethical considerations into system design); The IEEE Global Initiative on Ethics of Autonomous and Intelligent Systems, IEEE SA, https://standards.ieee.org/industry-connections/ec/autonomous-systems (last visited Aug. 2, 2024), (General project for developing ethical standards for technology).

[62] Congratulations, AI Trailblazers! Iso/Iec Jtc 1/Sc 42 Receives Prestigious Iso Lawrence D. Eicher Award, ANSI (Sep. 22, 2023), https://www.ansi.org/standards-news/all-news/2023/09/9-22-23-congratulations-ai-trailblazers-isoiec-jtc-1sc-42-receives-prestigious-iso; ISO/IEC JTC 1/SC 42 Artificial intelligence, ISO, https://www.iso.org/committee/6794475.html (last visited Aug. 2, 2024).

[63] Id.

[64] The Use of Artificial Intelligence in Health Care: Trustworthiness (ANSI/CTA-2090), Consumer Tech. Assoc. (2021), https://shop.cta.tech/products/the-use-of-artificial-intelligence-in-healthcare-trustworthiness-cta-2090.

[65] Id.

[66] U.S. Leadership in AI: A Plan for Federal Engagement in Developing Technical Standards and Related Tools, NIST 18 (2019), available at https://www.nist.gov/system/files/documents/2019/08/10/ai_standards_fedengagement_plan_9aug2019.pdf.

[67] Id. at 22.

[68] Id. at 19.

[69] NIST, supra note 58.

[70] Majority Leader Schumer Floor Remarks on the Release of the Roadmap for AI Policy by the Senate Bipartisan Senate AI Working Group, Senate Democrats (May 15, 2024), https://www.democrats.senate.gov/newsroom/press-releases/majority-leader-schumer-floor-remarks-on-the-release-of-the-roadmap-for-ai-policy-by-the-senate-bipartisan-senate-ai-working-group.

[71] See Model Legislation: Responsible Advanced AI Act, Center for AI Policy (Apr. 9, 2024), https://www.aipolicy.us/work/model.

[72] Id.

[73] See, e.g., The Tj Hooper, 60 F.2d 737 (2d Cir. 1932) (industry standards can inform the bounds of negligence even though they do not necessarily define the limits of negligence; instead, what is “reasonably prudent under the circumstances” might go beyond current practices); Daubert v. Merrell Dow Pharmaceuticals Inc., 509 U.S. 579 (1993), (consideration of whether the claimed expert scientific methods adhere to standards recognized in the relevant scientific community.); Nikolov v. Associated Env’t Servs., 52 F. App’x 975, 976–77 (9th Cir. 2002).

[74] Knut Blind, Sören S. Petersen, & Cesare A.F. Riillo, The Impact of Standards and Regulation on Innovation in Uncertain Markets, 46(1) Research Pol’y 249 (2017), https://doi.org/10.1016/j.respol.2016.11.003.

[75] Id. at 251.

[76] Id.

[77] Id. at 253.

[78] See infra, nn. 99-105 and accompanying text.

[79] See Jennifer Huddleston, The Price of Privacy: The Impact of Strict Data Regulations on Innovation and More, American Action Forum (Jun. 3, 2021), https://www.americanactionforum.org/insight/the-price-of-privacy-the-impact-of-strict-data-regulations-on-innovation-and-more/#ixzz8hmX9sfiS (Discussing how the GDPR’s privacy regulations, which generate large compliance costs, affect innovation in the EU).

[80] See, e.g., Mikolaj Barczentewicz, Does the DMA Let Gatekeepers Protect Data Privacy and Security?, Truth on Mkt. (Apr. 4, 2024), https://truthonthemarket.com/2024/04/04/does-the-dma-let-gatekeepers-protect-data-privacy-and-security (Noting how the EU’s interoperability mandates work at odds with its privacy mandates, putting firms in a bind when considering how to design their business processes. This extends to AI developers insofar as strict standards that operate at odds with each other will ultimately make certain kinds of AI behavior prohibited).

[81] We begin to strain the meaning of “commercialization” in this sense, as this would essentially turn related AI research and development into a commons. Thus, it’s perhaps better to think of this more generally as a distribution method, as the incentives to fully develop the technology that attend commercialization processes will be rather attenuated.

[82] John R. Thomas, Compulsory Licensing of Patented Inventions, Cong. Research Serv. (Jan. 14, 2014), https://crsreports.congress.gov/product/pdf/R/R43266.

[83]Marianna Mazzucato, Marietje Schaake, Seb Krier, & Josh Entsminger, Governing Artificial Intelligence in the Public Interest, Stanford Cyber Policy Center (2022); see also Bryanna Devonshire & Nicolas Harris, Government Regulation of AI – When Is It Coming?, Seacoastonline (Aug. 24, 2023), https://www.seacoastonline.com/story/business/ 2023/08/24/government-regulation-of-ai-when-is-it-coming/70671063007.

[84] Part II — Standards Concerning the Availability, Scope and Use of Intellectual Property Rights, Art. 31, WTO, https://www.wto.org/english/docs_e/legal_e/27-trips_04c_e.htm (last visited Aug. 2, 2024).

[85] John R. Thomas, supra note 82.

[86] Satellite Television Extension and Localism Act of 2010, S. 3333, 111^th Cong. (2010).

[87] 17 USC § 115.

[88] William N. Monte, Compulsory Licensing of Patents, 25(3) Info. & Comm. Tech. L. 246 (2016).

[89] Id.

[90] Funding a Revolution: Government Support for Computing Research, National Research Council (1999), https://nap.nationalacademies.org/read/6323/chapter/1#vii.

[91] Id.

[92] Kristian Stout, Biden’s AI Executive Order Sees Dangers Around Every Virtual Corner, Truth on Mkt. (Nov. 1, 2023), https://truthonthemarket.com/2023/11/01/bidens-ai-executive-order-sees-dangers-around-every-virtual-corner.

[93] See John R. Thomas, Compulsory Licensing of Patented Inventions, Cong. Research Serv. (Jan. 14, 2014), available at  https://crsreports.congress.gov/product/pdf/R/R43266.

[94] Geoffrey A. Manne & Dirk Auer, From Data Myths to Data Reality: What Generative AI Can Tell Us About Competition Policy (and Vice Versa), CPI Antitrust Chronicle (Feb. 23, 2024), https://laweconcenter.org/resources/from-data-myths-to-data-reality-what-generative-ai-can-tell-us-about-competition-policy-and-vice-versa.

[95] Id.

[96] Id.

[97] Id.

[98] Id.

[99] Required Rulemaking on Personal Financial Data Rights, Docket No. 2023-CFPB-0052, CFPB (Jun. 11, 2024), available at https://files.consumerfinance.gov/f/documents/cfpb_personal-financial-data-rights_final-rule_2024-06.pdf.

[100] Giuseppe Colangelo, Open Banking Goes to Washington: Lessons from the EU on Regulatory-Driven Data Sharing Regimes, 54 Computer L. & Security Rev., https://doi.org/10.1016/j.clsr.2024.106018.

[101] See, e.g., Ivan Bosch Chen et al., A Study on the Application and Impact of Directive (EU) 2015/2366 on Payment Services (PSD2), Publications Office of the European Union (2023), https://data.europa.eu/doi/10.2874/996945 (Estimating that TPPs spent €35 million on problems linked to accessing APIs, and €140 million on maintaining legacy systems, due to APIs not working properly).

[102] See Miguel de la Mano & Jorge Padilla, Big Tech Banking, 14 J. Competition L. & Econ. 494, 503 (2018). It’s also important to note that none of this occurs in a vacuum. In some respects, open-banking rules are broached as a way to deal with the prevalent practice of “screen scraping” in order to enable different fintech services to work. See Laurent van Huffel, From Screen Scraping to Open Banking, BAI (Oct. 20, 2023), https://www.bai.org/banking-strategies/from-screen-scraping-to-open-banking.

[103] See Barczentewicz, supra note 80.

[104] Id.

[105] See Open App Markets Act, S.2710, 117th Cong. (2022), https://www.congress.gov/bill/117th-congress/senate-bill/2710/text.

I.       Introduction

As part of the Infrastructure Investment and Jobs Act (IIJA), signed by President Joe Biden in November 2021, Congress provided $42.5 billion for broadband deployment, mapping, and adoption projects through the Broadband Equity, Access, and Deployment (BEAD) program, with the stated goal of directing the funds to close the so-called “digital divide.”[1] But actions by pole owners—such as refusing to allow broadband companies to attach their lines on reasonable and nondiscriminatory terms—threaten to slow broadband deployment significantly.

In a recent letter to Assistant Attorney General Jonathan Kanter, Sen. Mike Lee (R-Utah) put forth the argument that the U.S. Justice Department (DOJ) should take action to address abuses of the pole-attachment process by local power companies (LPCs) regulated by the Tennessee Valley Authority (TVA).[2] His concern is that such abuses threaten to slow broadband deployment, especially to rural areas served by the TVA and the LPCs.[3] Among the abuses he details are:

• Delaying or refusing to negotiate pole-attachment agreements with competitive broadband-service providers, including when the TVA LPC provides broadband service (itself or through a joint venture agreement) or is interested in doing so;
• Initially refusing to negotiate pole-attachment agreements that would enable competitive broadband-service providers to obtain permits in sufficient time to meet federal grant deadlines;
• Refusing to review pole-attachment applications on a scale or at the pace necessary to complete broadband projects in a timeframe required by federal grant programs;
• Refusing to follow the standard industry practice of approving a contractor to process pole-access applications in a timely manner when the utility’s staff is insufficient to do the work, even when the broadband-service provider is willing to pay the entire bill for the contractor; and
• Refusing to process pole-attachment applications at all, and failing to respond to provider outreach regarding the processing of applications for months on end.[4]

Section 224 of the Communications Act exempts municipal and electric-cooperative (“coop”) pole owners, such as the LPCs, from oversight by the Federal Communications Commission (FCC).[5] At the same time, the TVA’s authority over pole attachments is not subject to oversight by state governments.[6] This loophole means that it is the TVA, not the FCC, that sets the rates for pole attachments. The TVA’s rates are significantly higher than those of the FCC, [7] and the TVA’s LPCs often are able to avoid the access requirements that states and the FCC typically require.[8]

But avoiding state and FCC regulatory oversight is not the only loophole that the TVA and its LPCs can exploit: the TVA and the government-owned LPCs also may not be subject to antitrust law. These entities hold a resource critical for broadband deployment, while it is essentially impossible for private providers to build competing pole infrastructure. In situations like this, government entities that participate as firms in the marketplace—known in the literature as “state-owned enterprises” (SOEs)—should be subject to antitrust law in order to ensure access by private competitors.

Sen. Lee is correct that the DOJ should examine the practices of the TVA and its LPCs under antitrust law. Antitrust clearly applies to those LPCs that are private coops, which have no immunities. But Congress should clarify that the TVA and government-owned LPCs are likewise subject to antitrust law when they act according to their “commercial functions” or as “market participants.” They should also consider bringing the TVA and all of its LPCs under the purview of the FCC’s Section 224 authority over pole attachments.

II.     The Law & Economics of State-Owned LPCs and Rural Electrical Cooperatives (RECs)

A.    The Competition Economics of State-Owned Enterprises

SOEs’ incentives differ from those of privately owned businesses. Most notably, while a private business must pass the profit-and-loss test, SOEs often are not subject to the same constraints. This difference may manifest through setting up legal SOE monopolies against which no other firm can compete; exempting SOEs from otherwise generally applicable laws; extending explicit subsidies to SOEs, whether in the form of taxpayer-financed appropriations or government-backed bonds (which the government explicitly or implicitly promises to repay, if necessary); or cross-subsidies from other government-owned monopoly businesses.

As a result, SOEs do not need to maximize profits (with Armen Alchian’s caveat that private market participants may be modeled as profit maximizers even if that isn’t their true motivation[9]) and can pursue other goals. In fact, this is exactly why some supporters of SOEs like them so much: they can pursue the so-called “public interest” by providing ostensibly high-quality products and services at what are often below-market prices.[10]

But this freedom comes at a cost: not only can SOEs inefficiently allocate societal resources away from their highest-valued uses, but they may actually have greater incentive to abuse their positions in the marketplace than private entities. As David E.M. Sappington and J. Gregory Sidak put it:

[W]hen an SOE values an expanded scale of operation in addition to profit, it will be less concerned than its private, profit-maximizing counterpart with the extra costs associated with increased output. Consequently, even though an SOE may value the profit that its anticompetitive activities can generate less highly than does a private profit-maximizing firm, the SOE may still find it optimal to pursue aggressively anticompetitive activities that expand its own output and revenue. To illustrate, the SOE might set the price it charges for a product below its marginal cost of production, particularly if the product is one for which demand increases substantially as price declines. If prohibitions on below-cost pricing are in effect, an SOE may have a strong incentive to understate its marginal cost of production or to over-invest in fixed operating costs so as to reduce variable operating costs. A public enterprise may also often have stronger incentives than a private, profit-maximizing firm to raise its rivals’ cost and to undertake activities designed to exclude competitors from the market because these activities can expand the scale and scope of the SOE’s operations.[11]

Here, entities like the TVA and many of the government-owned LPCs that sell the electricity it produces are simply not subject to the same profit-and-loss test that a private power company would be. But even more importantly for the discussion of broadband buildout, many of these government-owned LPCs also provide broadband services (or intend to), effectively using their position as a monopoly provider of electricity to cross-subsidize their entry into the broadband marketplace. Moreover, LPCs often own the electric poles and control decisions about whether and at what rates to rent them to third parties (subject to TVA rate regulations), including to private broadband providers that may compete with the LPCs’ municipal-broadband offerings.

This raises two significant issues for competition policy:

• Because government-owned municipal-broadband providers focus on speed and price, rather than profitability, they can sometimes offer greater speeds at lower prices than private providers, deterring private buildout and competition using what, in other contexts, would be referred to as “predatory pricing” (e., the government can use its unique monopoly advantages to indefinitely set prices too low)[12]; and
• LPCs that offer municipal-broadband services can raise rivals’ costs by refusing to deal with private broadband providers that want to attach equipment to their poles (an “essential facility” or “critical input”) or by offering access only on unreasonable and discriminatory terms.

In Verizon Communication Inc. v. Law Offices of Curtis V. Trinko LLP,[13] the U.S. Supreme Court explained the reasoning behind a very limited duty to deal under antitrust law:

Compelling… firms to share the source of their advantage is in some tension with the underlying purpose of antitrust law, since it may lessen the incentive for the monopolist, the rival, or both to invest in those economically beneficial facilities.[14]

In sum, a private market participant is constantly looking to acquire monopoly power by innovating and better serving customers, and temporary monopolies—acquired through a legitimate competitive process—are not unlawful. If successful, this process provides incentive for more innovation and competition, including incentives for competitors to build their own infrastructure.

But this is not so when it comes to SOEs, which can prevent competition in a way that private market participants cannot, due to their special access to legal mechanisms like eminent domain, taxes, below-market-rate loans, government grants of indefinite monopolies, and cross-subsidies from their own monopolies in adjacent markets. As a result, SOEs possess both special ability and incentive to raise rivals’ costs through refusals to deal or predatory pricing.

Ironically, the lack of a profit motive may make SOEs uniquely positioned to harm competition. Thus, it may make sense to impose on SOEs a duty to deal on reasonable and nondiscriminatory terms when it comes to pole attachments.

B.     The Economics of Co-Ops[15]

According to the National Rural Electric Cooperative Association, the trade association for rural electricity co-ops (RECs):

• Co-ops serve 56% of the U.S. landmass and 88% of the nation’s counties, including 93% of the 353 persistent poverty counties.
• They account for roughly 13% of all electricity sold in the United States.
• More than 90% of electric co-ops serve territories where the average household income is below the national average. One in six co-op consumer-members live at or below the poverty line.
• Cooperatives serve an average of eight consumer-members per-mile of electric line, but this average masks the extremely low-density population of many co-ops. If the handful of large co-ops near cities were removed, the average would be lower.
• More than 100 electric cooperatives provide broadband service and more than 200 co-ops are exploring the option and conducting feasibility studies to do so.[16]

There are some important differences between electric co-ops and investor-owned power companies. Most importantly, co-ops are owned by their consumers. Economics helps explain why this form of organization could be pro-competitive in some situations, but the history of RECs suggests that government support and corporate rules particular to co-ops are the main reasons that we continue to rely on co-ops to distribute electricity in rural areas of the United States. As a result, RECs—especially those that distribute electricity generated and transmitted by the TVA—have incentives more like those of state-owned enterprises (SEOs) than private firms.

In other words, RECs also have the incentive and ability to act anticompetitively—e.g., by refusing to deal with private broadband providers who wish to attach to the poles they own.

1.       Why Do We Have So Many RECs?

The classic law & economics examination of firms’ choice of business organization comes from Henry Hansmann, in his book The Ownership of Enterprise.[17] He explained that the choice of ownership for any firm is driven by costs. The form that is chosen by a particular firm is that which “minimizes the total costs of transactions between the firm and all of its patrons.”[18] These costs include both transaction costs with those patrons who are not owners, and the costs of ownership, such as monitoring and controlling the firm.

Hansmann argued that the form of consumer-owned co-ops predominates in the distribution of electricity in rural areas because it is a response to the threat of natural monopoly, where high barriers to entry and startup costs suggest that one firm is likely to dominate.[19] This is particularly true in geographic areas with low population density, because the costs of building out infrastructure is extremely high per individual consumer. As such, consumers would likely be subject “to serious price exploitation if they were to rely on market contracting with an investor-owned firm.”[20] Thus, the choice is among rate regulation of an investor-owned utility, municipal ownership, or consumer ownership through a co-op.

Hansmann argued that consumer co-ops best align “the firm’s interests with those of its consumers” because they have lower overall costs than other forms of ownership in rural areas.[21] This is because electricity is a “highly homogeneous [commodity] with few important quality variables that affect different users differently.”[22] Moreover, relatively stable farm and nonfarm residential households account for the overwhelming majority of the membership and demand for electricity in rural areas, “creating a dominant group of patrons with relatively homogenous interests.”[23]

As a result, the costs of monitoring and controlling these natural monopolies are relatively lower for the consumers as owners than they would be as citizens overseeing a public utility commission in charge of regulating an investor-owned utility, or a board in charge of a municipally owned utility.

On the other hand, the history of RECs suggests that their formation and persistence may be more due to government intervention than as a market response to consumer demand. As Hansmann himself recognized, RECs received significant public subsidies in the form of below-market loans from the Rural Electrification Administration (REA), though he argues that these loans were not significant subsidies for the first 15 years); exemption from federal corporate income tax; and preferential access to power generated by the TVA.[24] On top of that, the REA essentially organized many co-ops in their early days.[25]

Nonetheless, Hansmann argues:

These subsidies have undoubtedly been important in encouraging the formation and growth of cooperative utilities, and therefore the great proliferation of rural electric cooperatives does not provide an unbiased test of the viability of the cooperative form. Evidently, however, the federal subsidies have not been critical to the success of cooperatives in the electric power industry. Even before the federal programs were enacted, there already existed forty-six rural electric cooperatives operating in thirteen different states. Also, as already noted, there was no net interest subsidy to the cooperatives for the first fifteen years of the REA. And in its early years, the REA also offered low-interest loans to investor-owned utilities that wished to extend service into rural areas, but found little interest in these loans among the latter firms.[26]

In an excellent 2018 law-review article, however, Debra C. Jeter, Randall S. Thomas, & Harwell Wells systematically detail the great lengths to which the REA went to organize co-ops in rural areas.[27] The authors convincingly argue that the co-op model was not adopted as a market response, but primarily due to the REA’s organizational efforts and the subsidies bestowed upon them.

Even if RECs were a market response to natural monopoly in rural areas at the time of their adoption, that does not mean that they would necessarily continue to be the most economically efficient model. At a given point of time, economies of scale and high costs of entry may mean that the market can only support one firm (i.e., natural monopoly). But over the last 80-90 years, underlying conditions that may have made co-ops the most efficient model may have changed. As we argued in a 2021 white paper:

[I]n any given market at a given time, there is likely some optimal number of firms that maximizes social welfare. That optimal number—which is sometimes just one and is never the maximum possible—is subject to change, as technological shocks affect the dominant paradigms controlling the market. The optimal number of firms also varies with the strength of scale economies, such that consumers may benefit from an increase in concentration if economies of scale are strong enough… And it is important to remember that the market process itself is not static. When factors change—whether a change in demographics or population density, or other exogenous shocks that change the cost of deployment—there will be corresponding changes in available profit opportunities. Thus, while there is a hypothetical equilibrium for each market—the point at which the entry of a new competitor could reduce consumer welfare—it is best to leave entry determinations to the market process.[28]

In fact, as Jeter, Thomas, & Wells go on to argue, rules particular to the co-op model make it nearly impossible to change the form of ownership through merger or acquisition.[29] These rules—adopted as part of the model acts promoted by the REA—prevent what the great Henry Manne called “the market for corporate control” that would otherwise discipline co-op managers.[30]

As has been noted by even the strongest supporters of the co-op model[31]—and seemingly undermining Hansmann’s assessment that consumer-ownership is the most effective form of organization for these entities—RECs suffer from a lack of oversight by consumer-owners, with very few ever showing up to even vote for their board of directors:

[32]

This lack of oversight from the ownership means that the board of directors can engage in all kinds of abuses, as detailed extensively by Jeter, Thomas, & Wells.[33]

Without sufficient incentives for oversight by consumer-owners or a functioning market for corporate control, there is no basis to conclude that RECs remain the best business model for distributing electricity. Their ubiquity is more due to the REA’s organizational efforts and ongoing government benefits—in the form of subsidies, tax exemptions, and preferences from the TVA—than market demand.

2.       The Competition Economics of RECs and Pole Attachments

Due to the privileged position enjoyed by RECs, particularly those that distribute electricity from the TVA, they have a unique ability and incentive to act anticompetitively toward broadband providers that want to attach to the poles they own.

Much like municipally owned electricity distributors, RECs are not motivated solely by profit maximization. RECs also have similar advantages, like access to eminent domain, below-market loans, tax exemptions, and the ability to cross-subsidize entry into a new market (like broadband) from its dominant position in electricity distribution.

On the other hand, unlike municipally owned electricity distributors, RECs can go out of business, and thus must earn sufficient revenues to remain a going concern. This means that the incentives for RECs to act anticompetitively are at least as strong as those of investor-owned firms, and may be even as strong as those of state-owned enterprises. This is especially notable, when so many RECs either have entered or are planning to enter the broadband market.

In such cases, there are strong incentives for RECs to refuse to deal with private broadband providers that are trying to deploy in—and introduce competition to—their rural areas, as Sen. Lee’s (R-Utah) recent letter to the U.S. Justice Department suggests, many of these co-ops have done exactly that.[34]

The economic logic that drives a limited duty to deal under antitrust law is that enforced sharing rarely makes sense because it reduces the incentives to build infrastructure. [35] But creating new rural infrastructure (like poles) is cost-prohibitive—at least, without the same subsidies, eminent-domain power, and other advantages that RECs have historically enjoyed. Thus, RECs may rightfully have a duty to deal with broadband providers on a reasonable and nondiscriminatory basis.

Moreover, many RECs receive little oversight from rate regulators when it comes to pole attachments. And when they do, like those RECs that distribute electricity from the TVA, the formula allows for much higher rates than the FCC would allow.[36] As a result, pole costs are much higher for broadband companies dealing with poles owned by co-ops and municipalities that are not subject to the FCC’s authority (see Figure II).[37]

III.   The Complicated Nature of Antitrust Immunities

There is, however, a complication. In his letter to the DOJ, Sen. Lee rightly complains that:

TVA’s regulatory practices enable such behavior: there is no reason why TVA’s regulation of the pole rental rates charged by its LPCs requires TVA to somehow exempt those LPCs from generally-applicable rules that protect competition by requiring pole owners to provide pole access to third parties on reasonable terms. TVA should be using its authority over LPC distribution contracts to require LPCs to offer reasonable, non-discriminatory, and prompt pole access to third-party broadband providers (particularly recipients of taxpayer-funded broadband grants) in unserved areas, rather than giving its LPCs a free pass from those requirements.[38]

Unfortunately, while Lee’s letter is addressed to the DOJ’s antitrust chief, it isn’t clear whether antitrust laws even apply to the behavior he observes. This is primarily because of two legal doctrines: federal sovereign immunity from lawsuit and state-action immunity from antitrust.

A.    Federal Sovereign Immunity and the TVA

Normally, the federal government is immune from lawsuit under the ancient (and deeply flawed[39]) doctrine of sovereign immunity, except where explicitly waived by statute. The TVA is a wholly owned corporate agency and instrumentality of the federal government. Thus, federal courts have typically found that the TVA and other federal entities operating in the marketplace are exempt from antitrust.[40] This is despite the fact that the TVA’s enabling statute states:

Except as otherwise specifically provided in this chapter, the Corporation… may sue and be sued in its corporate name.[41]

There is, needless to say, nothing in the chapter that actually says the agency can’t be sued for antitrust violations. The older cases finding the TVA to be exempt from antitrust are likely to be found wrongly decided under the logic of the U.S. Supreme Court’s most recent case dealing with TVA’s immunity from suit. In 2019, the Court took up Thacker v. TVA,[42] which asked whether the TVA was immune from lawsuits for negligence. The Court rejected the lower court’s reasoning that the TVA was immune for torts arising from its “discretionary functions,” substituting a new test as to whether the TVA was acting pursuant to its governmental function or a commercial function. As the Court stated:

Under the clause—and consistent with our precedents construing similar ones—the TVA is subject to suits challenging any of its commercial activities. The law thus places the TVA in the same position as a private corporation supplying electricity. But the TVA might have immunity from suits contesting one of its governmental activities, of a kind not typically carried out by private parties.[43]

The Court also gave examples to help distinguish the two:

When the TVA exercises the power of eminent domain, taking landowners’ property for public purposes, no one would confuse it for a private company. So too when the TVA exercises its law enforcement powers to arrest individuals. But in other operations—and over the years, a growing number—the TVA acts like any other company producing and supplying electric power. It is an accident of history, not a difference in function, that explains why most Tennesseans get their electricity from a public enterprise and most Virginians get theirs from a private one. Whatever their ownership structures, the two companies do basically the same things to deliver power to customers.[44]

The test to be applied, therefore, is “whether the conduct alleged to be negligent is governmental or commercial in nature… if the conduct is commercial—the kind of thing any power company might do—the TVA cannot invoke sovereign immunity.”[45] Here, that arguably means that, when the TVA is acting pursuant to its commercial function, it should not receive immunity from antitrust suit.

On the other hand, Congress gave the TVA broad ratemaking authority and contractual powers. One federal court (previous to Thacker) rejected an antitrust challenge to the TVA’s ratemaking formula because it was a “valid governmental action and [therefore] exempt from the antitrust laws of the United States.”[46]

As noted above, some LPCs have entered into the municipal-broadband market and act as competitors to private broadband companies who want to attach to poles owned by LPCs. Thus, even though competition economics would suggest that LPCs would have a greater incentive to raise rivals’ costs by charging a monopoly price, the TVA would likely argue that it is acting in its government function when it sets those rates.[47] If courts agree, then antitrust law would not be able to reach that problem.

Consistent with the Court’s reasoning in Thacker, however, courts could find that antitrust law reaches agreements between wholesalers (like the TVA) and retailers (like the LPCs) to charge certain rates for pole attachments to competitors in an adjacent market. This would arguably be an example of the TVA acting as any other power generator would, pursuant to its commercial function, through some type of price-maintenance agreement. As it stands, it isn’t clear which way the courts would go.

Congress should strongly consider clarifying that the TVA is not exempt from antitrust scrutiny when it acts pursuant to a commercial function, including when it sets anticompetitive rates for pole attachments that would slow broadband buildout. This clearly affects the market for access to LPC-owned utility poles.

B.     State Action Immunity and the LPCs

Even if the commercial versus government distinction is clarified with respect to the TVA, there is a further wrinkle as it relates to antitrust scrutiny of LPCs. This concerns how the TVA’s actions interact with state-action immunity in antitrust law.

Grounded in the 10th Amendment, the Supreme Court has found there is immunity from antitrust laws for conduct that is the result of “state action.”[48] This doctrine has been interpreted to immunize anticompetitive conduct pursuant to state and local government action from antitrust claims, so long as “the State has articulated a clear … policy to allow the anticompetitive conduct, and second, the State provides active supervision of [the] anticompetitive conduct.”[49] When it comes to municipalities, however, the Court has found that “[o]nce it is clear that state authorization exists, there is no need to require the State to supervise actively the municipality’s execution of what is a properly delegated function.”[50]

The Supreme Court has also left open the possibility of an exception to state-action immunity when government entities themselves are acting as market participants.[51] In one case dealing with a local municipally owned power plant in Louisiana, the Supreme Court did not grant broad immunity from antitrust laws, in part because:

Every business enterprise, public or private, operates its business in furtherance of its own goals. In the case of a municipally owned utility, that goal is likely to be, broadly speaking, the benefit of its citizens. But the economic choices made by public corporations in the conduct of their business affairs, designed as they are to assure maximum benefits for the community constituency, are not inherently more likely to comport with the broader interest of national economic well-being than are those of private corporations acting in furtherance of the interests of the organization and its shareholders.[52]

While there are a few cases applying this distinction in lower federal courts,[53] there is no Supreme Court caselaw determining how to differentiate when, for the purposes of state-action immunity, municipal corporations act as market participants versus when they act as government entities. Jarod Bona and Luke Wake have proposed applying a test similar to the one the courts use in dormant Commerce Clause cases.[54] The distinction made by the Supreme Court in Thacker and discussed above may also be applicable.

Government-owned LPCs are creatures of states or municipalities. As such, they would certainly argue they are immune from antitrust scrutiny, even when they refuse to deal with private broadband providers with whom they compete while withholding a critical input (i.e., the ability to attach to their poles). But there are two problems with this argument.

First, it seems unlikely that the LPCs could argue that they are acting pursuant to a clearly articulated policy of displacing competition when they refuse to deal with broadband providers. As Sen. Lee pointed out in his letter, there are state laws that would impose a duty to deal on reasonable and nondiscriminatory terms, but for any exemptions to that authority due to the TVA.[55] For instance, North Carolina and Kentucky require all pole owners not subject to FCC Section 224 authority to offer nondiscriminatory pole access.[56]

On the other hand, they could appeal to the TVA’s contract authority,[57] in addition to the TVA’s stated policy that its purpose is “to provide for the … industrial development” of the Tennessee Valley.[58] But even if this grants the TVA authority to regulate rates for pole attachments, it doesn’t mean the TVA has enunciated an articulable policy of displacing competition in refusing to deal with broadband providers. It also would appear to be contrary to the purpose of promoting industrial development to forestall broadband deployment in the Tennessee Valley because LPCs that also have municipal-broadband systems don’t want that competition. In other words, their refusal to deal is not protected by an appeal to any articulable policy to displace competition, either by a state or the TVA.

Second, under the caselaw that does exist, government-owned LPCs are market participants that should not receive antitrust immunity. For instance, in one case, a private arena owner challenged under antitrust law an exclusive contract between a municipal-arena owner and LiveNation.[59] The court held that state-action immunity was “less justified” because the municipality’s “entertainment contracts” reflected “commercial market activity,” not “regulatory activity.”[60] Here, the LPCs’ actions as both power companies and municipal-broadband providers reflect commercial-market activity more than regulatory activity. They shouldn’t be able to claim immunity from antitrust for this refusal to deal, any more than a private broadband provider could.

In sum, the LPCs’ anticompetitive refusal to deal appears to be separate from the rates set by the TVA pursuant to its ratemaking authority or contractual powers. They should be subject to antitrust law. But due to uncertainty in this area, Congress should clarify that LPCs are not immune from antitrust scrutiny, and consider codifying the market-participant exception to state-action immunity in antitrust statutes.

IV.   Section 224 of the FCC Act

In his letter, Sen. Lee noted that, under Section 224 of the Communications Act, “Congress determined that poles and conduits are essential facilities that lack a viable market-based alternative, which led it to require utilities to extend nondiscriminatory access to utility poles to cable operators and competitive telecommunications providers.”[61] While acknowledging that TVA distributors are not subject to Section 224, Lee argued that “the congressional conclusion that poles are essential facilities that lack a viable market-based alternative holds for all poles.”[62] Lee further noted that the “TVA’s regulation of its LPCs’ pole attachment rates also impedes competition by setting rates well above the rates set by the FCC and deemed compensatory by the U.S. Supreme Court, inflating the cost for competitive broadband providers unaffiliated with TVA LPCs to offer service.”[63]

Theoretically, government-owned LPCs and cooperative LPCs are subject to some oversight when they run services like municipal broadband, either from voters or member-owners. But it is implausible that such oversight can be truly effective, given that these pole owners are not subject to normal market incentives and have their own conflicts of interest that encourage hold-up problems. Combined with their ability to cross-subsidize operations in broadband from their electricity customers, it should be clear that these entities pose a host of potential public-choice problems.[64]

Indeed, as FCC Commissioner Brendan Carr has noted:

I continue to hear concerns from broadband builders about unnecessary delays and costs when they seek to attach to poles that are owned by municipal and cooperative utilities. Unlike what we are doing in today’s item, there is a strong argument that Section 224 does not give us authority to address issues specific to those types of poles. Therefore, I encourage states and Congress to take a closer look at these issues—and revisit the exemption that exists in Section 224—so that we can ensure deployment is streamlined, regardless of the type of pole you are attaching to.[65]

We echo both Sen. Lee’s and Commissioner Carr’s sentiments here. The FCC’s important work on this matter stands to benefit millions of Americans trapped on the wrong side of the digital divide. The co-op and municipal loophole poses a major obstacle to achieving these ends. Insofar as Congress prioritizes quick and efficient broadband buildout, the TVA and its LPCs should not be able to thwart these goals through anticompetitive rates and refusals to deal. Congress should revisit this issue and grant the FCC jurisdiction over these types of pole owners.

V.     Conclusion

Sen. Lee’s letter to the DOJ highlights issues that are extremely important to closing the digital divide. Broadband deployment could be harmed as a result of the practices by the TVA and the LPCs. If DOJ Antitrust Division chief Jonathan Kanter is serious about taking on gatekeeper power,[66] he should start here: with public entities granted a truly unassailable gatekeeper position over private markets. But even more importantly, Sen. Lee’s letter highlights the need to reform antitrust immunities that apply to SOEs. Economics suggests government monopolies are a greater harm to competition than private ones. Antitrust law should reflect that reality.

Appendix A: Sen Mike Lee Letter to DOJ

[1] 47 U.S.C. § 1702(b) (2018).

[2] See, infra, Appendix A [hereinafter “Lee Letter”].

[3] Broadband Assessment Report, Tennessee Valley Authority (Dec. 2022), https://www.tva.com/energy/technology-innovation/connected-communities/broadband-assessment-report.

[4] See Lee Letter, supra note 2, at 1-2.

[5] See 47 U.S.C. § 224(a)(1) (2018) (“The term ‘utility’ means any person who is a local exchange carrier or an electric, gas, water, steam, or other public utility, and who owns or controls poles, ducts, conduits, or rights-of-way used, in whole or in part, for any wire communications. Such term does not include any railroad, any person who is cooperatively organized, or any person owned by the Federal Government or any State.”).

[6] See Lee Letter, supra note 2, at n.2.

[7] Pole Attachment Fee Formulas Adopted by TVA and the FCC, Tennessee Advisory Commission on Intergovernmental Relations (Jan. 2017), available at https://www.tn.gov/content/dam/tn/tacir/commission-meetings/january-2017/2017January_BroadbandAppL.pdf.

[8] See Lee Letter, supra note 2, at n.4.

[9] See Armen A. Alchian, Uncertainty, Evolution, and Economic Theory, 58 J. Pol. Econ. 211 (1950).

[10] See, e.g., Jonathan Sallet, Broadband for America’s Future: A Vision for the 2020s, at 50-51 (Oct. 2019), available at https://www.benton.org/sites/default/files/BBA_full_F5_10.30.pdf.

[11] David E.M. Sappington & J. Gregory Sidak, Competition Law for State-Owned Enterprises, 71 Antitrust L.J. 479, 499 (2003).

[12] See Ben Sperry, Islands of Chaos: The Economic Calculation Problem Inherent in Municipal Broadband, Truth on the Market (Sept. 3, 2020), https://truthonthemarket.com/2020/09/03/islands-of-chaos-the-economic-calculation-problem-inherent-in-municipal-broadband.

[13] 540 U.S. 398 (2004).

[14] Id. at 408-09.

[15] This section is adapted from Ben Sperry, Broadband Deployment, Pole Attachments, & the Competition Economics of Rural-Electric Co-ops, Truth on the Market (Aug. 16, 2023), https://truthonthemarket.com/2023/08/16/broadband-deployment-pole-attachments-the-competition-economics-of-rural-electric-co-ops.

[16] See Brian O’Hara, Rural Electrical Cooperatives: Pole Attachment Policies and Issues, at 2, NRECA (Jun. 2019), available at https://www.cooperative.com/programs-services/government-relations/regulatory-issues/documents/2019.06.05%20nreca%20pole%20attachment%20white%20paper_final.pdf.

[17] Henry Hansmann, The Ownership of Enterprise (2000).

[18] Id. at 21.

[19] See id. at 169.

[20] Id.

[21] Id. at 170.

[22] Id.

[23] Id.

[24] See id. at 173

[25] See id.

[26] Id.

[27] See Debra C. Jeter, Randall S. Thomas, & Harwell Wells, Democracy and Dysfunction: Rural Electrical Cooperatives and the Surprising Persistence of the Separation of Ownership and Control, 70 Ala. L. Rev. 316, 372-395 (2018).

[28] Geoffrey A. Manne, Kristian Stout, & Ben Sperry, A Dynamic Analysis of Broadband Competition: What Concentration Numbers Fail to Capture, at 28, 32 (ICLE White Paper – June 2021), available at https://laweconcenter.org/wp-content/uploads/2021/06/A-Dynamic-Analysis-of-Broadband-Competition.pdf.

[29] Jeter et al., supra note 27, at 419-39.

[30] See Henry G. Manne, Mergers and the Market for Corporate Control, 73 J. Pol. Econ. 110 (1965).

[31] See John Farrell, Matt Grimley, & Nick Stumo-Langer, Report: Re-Member-ing the Electric Cooperative, Inst. For Local Self-Reliance (Mar. 29, 2016), https://ilsr.org/report-remembering-the-electric-cooperative/#Missing%20Members (“More than 70 percent of cooperatives have voter turnouts of less than 10 percent [] including Wilson’s Jackson Energy Cooperatives, which averages just under 3 percent turnout.”).

[32] Id.

[33] Jeter et al., supra note 27, at 397-400.

[34] See Lee Letter, supra note 2, at 1-2.

[35] See Trinko, 540 U.S. at 408-09.

[36] See Pole Attachment Fee Formulas Adopted by TVA and the FCC, supra note 7.

[37] See NCTA, Pole Attachments, https://www.ncta.com/positions/rural-broadband/pole-attachments (last accessed Sept. 4, 2023).

[38] Lee Letter, supra note 2, at 2.

[39] See Ben Sperry, When Violations of the Law Have No Remedy: The Case of Warrantless Wiretapping, Competitive Enterprise Institute (Aug. 8, 2012), https://cei.org/blog/when-violations-of-the-law-have-no-remedy-the-case-of-warrantless-wiretapping.

[40] See, e.g., Webster Cty. Coal v. Tennessee Valley Authority, 476 F.Supp. 529 (W.D. Ky. 1979) (finding the TVA is exempt from antitrust law); Sea-Land Serv. Inc. v. Alaska R.R., 659 F.2d 243 (D.C. Cir. 1981), cert. denied, 455 U.S. 919 (1982) (finding the Alaska Railroad exempt from antitrust law).

[41] 16 U.S.C. §831c(b) (2018).

[42] 139 S. Ct. 1435 (2019).

[43] Id. at 1439.

[44] Id. at 1443-44.

[45] Id. at 1444.

[46] City of Loudon v. TVA, 585 F.Supp. 83, 87 (E.D. Tenn. Jan. 30, 1984).

[47] The TVA could also argue that the rate formula for pole attachments that it sets is subject to the filed rate doctrine and thus exempted from antitrust scrutiny. The filed rate doctrine does not allow courts to second-guess agency determinations of rates. See Keogh v. Chicago & Northwest Railway Co., 260 U.S. 156 (1922). While the original case on the filed rate doctrine dealt with the literal situation of regulated entities filing rates which were approved by a regulator, courts have extended the doctrine to other situations where a regulator uses its authority to set rates. Cf. Wortman v. All Nippon Airways, 854 F.3d 606, 611 (9th Cir. 2017) (“While the filed rate doctrine initially grew out of circumstances in which common carriers filed rates that a federal agency then directly approved, we have applied the doctrine in contexts beyond this paradigmatic scheme.”) The unique situation with the TVA is that there is no clear statutory ratemaking authority over pole attachments, but they have asserted the ability to do so under their contract powers, raising the same issue of whether this is a governmental function or market function. See TVA Determination of Regulation on Pole Attachments 2 (Jan. 22, 2016), available at https://tva-azr-eastus-cdn-ep-tvawcm-prd.azureedge.net/cdn-tvawcma/docs/default-source/about-tva/guidelines-reports/determination-on-regulation-of-pole-attachments-7-12-2023.pdf. Even if the filed rate doctrine applies, though, it would not stop an enforcement action aimed at an injunction or declaratory relief by the DOJ, just treble damages sought by a private litigant. See Keogh, 260 U.S. at 162 (“[T]he fact that these rates had been approved by the Commission would not, it seems, bar proceedings by the Government.”).

[48] See, e.g., Parker v. Brown, 317 U.S. 341 (1943) and its progeny.

[49] North Carolina State Bd. of Dental Examiners v. FTC, 574 U.S. 494, 506 (2015) (internal citations omitted).

[50] Town of Hallie v. City of Eau Claire, 471 U.S. 34, 47 (1985).

[51] See, e.g., City of Columbia v. Omni Outdoor Advertising Inc., 499 U.S. 365, 379 (1991) (“We reiterate that, with the possible market participant exception, any action that qualifies as state action is ‘ipso facto… exempt from the operation of the antitrust laws…’”); FTC v. Phoebe Putney Health Systems Inc., 568 U.S. 216, 226 n.4 (“An amicus curiae contends that we should recognize and apply a ‘market participant’ exception to state-action immunity because Georgia’s hospital authorities engage in proprietary activities… Because this argument was not raised by the parties or passed on by the lower courts, we do not consider it.”).

[52] City of Lafayette v. Louisiana Power & Light Co., 435 U.S. 389, 403 (1978).

[53] See, e.g., Edinboro Coll. Park Apartments v. Edinboro Univ. Found., 850 F.3d 567 (3d Cir. 2017); VIBO Corp. v. Conway, 669 F.3d 675 (6th Cir. 2012); Freedom Holdings Inc. v. Cuomo, 624 F.3d 38 (2d Cir. 2010); Hedgecock v. Blackwell Land Co., 52 F.3d 333 (9th Cir. 1995).

[54] See Jarod M. Bona & Luke A. Wake, The Market-Participant Exception to State-Action Immunity from Antitrust Liability, 23 J. Antitrust & Unfair Comp. L. Section of the State Bar of Ca., Vol. 1 (Spring 2014), available at https://www.theantitrustattorney.com/files/2014/05/Market-Participant-Exception-Article.pdf.

[55] See Lee Letter, supra note 2, at 2.

[56] Id. at n.4; N.C. Gen. Stat. § 62-350(a) (requiring all pole owners to offer non-discriminatory pole access); 807 Ky. Admin. Regs. 5:015 § 2(1) (same).

[57] 16 U.S.C. § 831i (2018) (“Board is authorized to include in any contract for the sale of power such terms and conditions, including resale rate schedules, and to provide for such rules and regulations as in its judgment may be necessary or desirable for carrying out the purposes of this Act”).

[58] 16 U.S.C. § 831 (2018).

[59] See Delta Turner Ltd. v. Grand Rapids-Kent County Convention/Arena Authority, 600 F.Supp.2d 920 (W.D. Mich. 2009).

[60] Id. at 929.

[61] Lee Letter, supra note 2, at n.5.

[62] Id.

[63] Id. at n.3.

[64] See Vincent Ostrom & Elinor Ostrom, Public Goods and Public Choices, in Alternatives for Delivering Public Services: Toward Improved Performance (1979) (“[I]nstitutions designed to overcome problems of market failure often manifest serious deficiencies of their own. Market failures are not necessarily corrected by recourse to public sector solutions.”).

[65] Accelerating Wireline Broadband Deployment by Removing Barriers to Infrastructure Investment, WC Docket No. 17-84, Second Further Notice of Proposed Rulemaking (March 16, 2022) (Statement of Commissioner Brendan Carr), available at https://docs.fcc.gov/public/attachments/FCC-22-20A3.pdf.

[66] See, Assistant Attorney General Jonathan Kanter Delivers Opening Remarks at the Second Annual Spring Enforcers Summit, U.S. Justice Department (Mar. 27, 2023), https://www.justice.gov/opa/pr/assistant-attorney-general-jonathan-kanter-delivers-opening-remarks-second-annual-spring (“Gatekeeper power has become the most pressing competitive problem of our generation at a time when many of the previous generations’ tools to assess and address gatekeeper power have become outmoded.”).

I. Introduction

The way we enjoy our favorite shows and movies has changed dramatically from even five years ago. Streaming has expanded the amount of content available to American viewers and multiplies the ways to watch it. This has affected not just audiences, but also the marketplace, with large technology firms like Amazon, Apple, and Google and other new entrants now directly competing with traditional broadcast and cable networks and pay-TV distributors.

The economic advantages of streaming platforms initially presented compelling cost efficiencies, as they expanded available content while significantly reducing consumers’ per-hour viewing expenses relative to traditional cable packages. But the proliferation of services, the dispersion of exclusive programming, and generally growing subscription costs have narrowed this value gap. These trends have also underscored the importance of new business-model experimentation, such as optional bundling, to maintain consumer satisfaction and mitigate subscription fatigue.

While the choices available to consumers have certainly grown, the many new players in the marketplace and the disaggregation of networks, brands, and content have made the viewing experience more complicated for audiences. In the past, cable and satellite bundles gave viewers an economically efficient and convenient way to receive virtually all television content. Today, audiences might subscribe to multiple services and still be unable to receive all or even most of the popular and niche programming they desire, including content to which they previously had access.

At the same time, the sheer number of shows and outlets, and the rise in cord cutting, means that many traditional content producers and distributors are garnering relatively smaller audiences. The result has been a drop in advertising and subscription revenue, precisely when these producers and distributors face increasing pressure to invest more in order to: 1) create or acquire rights to additional content to meet growing viewer demand; 2) build out their streaming operations; and 3) increase marketing expenses to reach fractured audiences and win back viewership.

In this environment, traditional media companies have begun exploring new business models to maintain their competitiveness, seeking to invest, innovate, and collaborate in novel ways to address the market’s fragmentation. In applying regulatory, economic, and antitrust principles, policymakers should account for the increase in competition and the market’s fragmentation. The rise in competition, including from the entry of large tech companies, should assuage past concerns that traditional media companies could misuse—or even possess—meaningful market power. At the same time, any regulatory response must account for the need for flexibility to experiment with different business models—such as bundling, joint ventures, and other combinations—in order for market participants to innovate and remain competitive.

II. A New and Rapidly Changing Video Environment

The growth in video streaming and the entry of companies such as Netflix, Amazon, Google, and Apple has fundamentally altered the video market’s competitive landscape from the days when just a handful of networks and local cable companies were both the source of most popular content and responsible for its delivery. This evolution has led to a dramatic increase in competition and consumer welfare.

FIGURE 1: Number of Subscribers of Select Video Services (Q1’24, in Millions)

SOURCES: S&P Global Market Intelligence, StreamTV Insider[1]

Between 2009 and 2023, the number of original scripted shows across cable, broadcast, and streaming grew by almost 150%.[2] As of 2024, approximately 25% of total TV viewership came from services that didn’t exist a decade ago. This is before one even attempts to account for smartphone and other online viewership of content broadly available through YouTube, TikTok, Instagram, and other social-media services.[3] Only 45% of all TV viewing in 2024 came from cable and satellite services, and only about half of all TV viewing was on linear services (including online linear services).[4]

Netflix and Amazon now each have far more video subscribers than any other streaming service, including Hulu, Disney+, Paramount+, Peacock, and HBO Max. By 2026, YouTube TV is expected to surpass Charter and Comcast to become the largest linear pay-TV provider.[5]

FIGURE 2: Traditional MVPD and Online Linear Video-Service Subscriptions

SOURCE: S&P Kagan Global Forecast

As seen in Figure 2, with the rise of cord cutting, the percentage of U.S. television households subscribing to traditional MVPD services (which include cable, satellite, and telco-video services) dropped below 50% in 2022 and continues to shrink.[6]

Linear broadcast and cable-network programming that viewers historically accessed from cable, satellite, and telecom providers are now also available through multiple streaming platforms, such as Sling TV, DirecTV Stream, Hulu + Live TV, YouTube TV, and free, ad-supported streaming television (FAST) services like Tubi, Freevee, and Pluto TV. Television and film content is also available on-demand over various streaming services. In addition, mobile devices are increasingly the primary screen for video consumption, even inside the home.

As demonstrated in Figure 3, the traditional media companies are now much smaller than their newer competitors. Google, Amazon, and Apple each have a market cap of between six and 10 times the combined market caps of Disney, Comcast, Warner Bros. Discovery, and Paramount. In 2024, those big three tech platforms’ respective annual revenues were each between one and 2.2 times the combined revenues of Disney, Comcast, Warner Bros. Discovery, and Paramount.

FIGURE 3: Market Cap and Annual Revenue of Select Video Services (2024)

SOURCE: Google Finance

The new tech competitors’ business models make them particularly strong rivals in the media marketplace because they often supply their video services as loss-leaders for their other businesses, including software and hardware (e.g., Apple), and online commerce (e.g., Amazon). The companies’ size and reach also enable them to, in many cases, outbid traditional media for content. In sports programming, for example, Amazon Prime now offers Thursday Night Football, YouTube TV offers the NFL Sunday Ticket, and Apple TV+ offers Major League Baseball and Major League Soccer—all either new offerings or content that was formerly available from traditional providers.

The shift in consumer behavior—including subscriber “churn,” or frequent subscription cycling to optimize spending based on content availability—demonstrates a dynamic market environment in which traditional contractual lock-in has largely vanished.[7] This consumer-driven flexibility further underscores the diminished market power of traditional media companies and highlights the competitive discipline imposed by consumer preferences.

These changes have also affected the advertising side of the industry. Traditional media companies have always relied on advertising revenue to fund the creation and/or acquisition of compelling content, to finance innovation, and to reduce out-of-pocket costs for potential viewers. With the advent of streaming, however, advertisers now have many more options to place ads on television programming.

In its 2019 challenge to the proposed merger between broadcast companies Nexstar and Tribune, the U.S. Justice Department (DOJ) argued that the “[i]nventory of ad-supported, high-quality, long-form video on the internet is limited.… The most popular high-quality, long-form video available on the internet is provided through ad-free subscription services (like Netflix or Amazon Prime).”[8]

But streaming platforms that initially launched as subscription-only services now offer ad-supported tiers. For example, 70 million Netflix users globally subscribe to an ad-supported tier,[9] and ads are now the default for subscribers to Amazon Prime Video.[10] Overall, approximately 46% of households subscribe to at least one ad-supported tier of a streaming service,[11] and 56% of new streaming-service subscribers in the first quarter of 2024 chose an ad-supported tier.[12]

In addition, short-form content delivered via social-media platforms like YouTube, Facebook/Instagram, and TikTok compete with sitcoms and dramas for advertisers.[13] Even traditional media companies’ ability to deliver local spot advertising is no longer unique, as streaming and social-media platforms can use IP addresses to geo-target audiences for both long-form and short-form content.

The dramatic changes in the industry outlined above have greatly increased competition for users, content, and advertisers. This has eroded traditional sources of market power, such as control over the most popular mass-market content (e.g., network programming), as well as control over the limited channels of distribution (e.g., local cable distribution). This growth in competition, and the resulting reduction in potential for firms to exercise market power, will likely continue to benefit consumers.

III. Marketplace Fragmentation, Consumer Welfare, and the (In)Efficiency of Video-Content Delivery

While viewers have clearly benefited from the explosion in both content and the number of ways to watch it, some attributes of the contemporary video market actually impede the efficient distribution of content.

Until recently, the majority of viewers subscribed to one of several pay-TV packages that provided access to most of the popular and niche content available. Now, Americans who subscribe to a streaming package pay, on average, for four streaming services—for many, such subscriptions are in addition to their live TV subscription.[14] And even then, they might not have access to all the content they desire, including content that was previously available to them—either because they don’t subscribe to an additional streaming service that has exclusive content, or because they have “cut the cord,” replacing their cable or satellite service with streaming services that don’t carry programming they previously consumed.

For example, to watch all of the Green Bay Packers’ 2024 regular season games, a fan in most parts of Wisconsin would have needed access to Fox, CBS, NBC, and ABC; an Amazon Prime subscription for the Thursday Night Football game; and a Peacock subscription to watch the season opener. Then, to watch their favorite comedies and dramas, those same viewers might also have needed additional services such as Netflix, Apple TV, and Hulu.

Even viewers willing to subscribe to multiple streaming services may face frictions in accessing their preferred content, when compared to the relative ease of traditional bundled TV packages. The proliferation of separate interfaces, distinct platforms, and multiple controllers raises viewers’ practical switching costs, particularly for some elderly or less-technologically savvy viewers. As a result, subscribers who theoretically have access to a wider range of content than before may, nevertheless, incur substantial search costs when attempting to find their preferred programming or to discover new shows. Indeed, survey evidence highlights the significance of these costs. In 2023, nearly 50% of respondents told Deloitte that they would increase their viewing time on streaming-video services if finding content were easier.[15]

By the same token, increased market fragmentation requires content providers to spend more to overcome these transaction costs, as they must intensify their marketing efforts across a wider range of outlets and shows to inform audiences about content availability. Less fragmentation and more integrated offerings would decrease these transaction and search costs, enhancing efficiency for both consumers and content providers.

Relatedly, empirical evidence suggests that market fragmentation has inadvertently contributed to a modest resurgence in online piracy.[16] Troublingly, a growing number of consumers who report that they are frustrated by fragmented content offerings and rising cumulative subscription costs view unlawful access as a valid response to manage entertainment expenses.[17]

The fragmentation of outlets, combined with the proliferation of content, has diluted the audience share of programming on traditional networks, eroding their competitive positions and reducing their advertising and subscription revenue. For example, TV’s top scripted linear program in 2014, The Big Bang Theory, averaged more than 20 million total viewers. By 2024, the highest-rated scripted program was CBS’s Tracker, averaging just over 11 million viewers—a much smaller audience share.[18]

And the shrinking audience shares is happening precisely when competing successfully with new entrants would require traditional media companies to increase their investments in order to 1) create or acquire rights to additional content to meet growing viewer demand; 2) build out their streaming operations; and/or 3) increase marketing expenses to reach fractured audiences and win back viewership. These trends reinforce the urgency for media companies to innovate in user-friendly content aggregation and pricing strategies.

IV. Applying Competition Policy in the Video Sector to Enable Innovation and Competition

Technological developments have created more competition and viewer choice, but pose a new challenge: how to package all of that content and deliver it to viewers in a cost-effective and convenient way. Developing user-friendly ways to combine the vast amount of available content will be key to facilitating ongoing competition and improving the viewing experience for audiences.

Competition policy will likely play a critical role in allowing such innovation and value creation, as market participants combine and recombine to discover profitable business models that will enable continued innovation and content growth. Properly assessing these ventures will require policymakers and the antitrust agencies to recognize certain key market characteristics:

• First, technological evolution and changing consumer preferences have enabled new entrants and business models. These have, in turn, increased competition and expanded output, while significantly diminishing the market power of traditional media companies.
• Second, the current environment of abundant content is accompanied by a complex and disaggregated ecosystem of platforms and delivery mechanisms. This has increased search and transaction costs, limiting viewers’ easy access to and discovery of programming they potentially value.
• Third, the creation, acquisition, and delivery of long-form video content entails immense capital investments, which are only increasing as the growth in content and its dispersion across outlets necessitates increased spending to differentiate new content.

These market characteristics lead inevitably to a fourth consideration, the recognition of which is essential to ensure a proper assessment of business conduct in today’s (and tomorrow’s) video market:

• The ability to offer aggregated content packages is virtually essential to reduce consumer frictions, compete successfully for viewers, and therefore to draw (and keep) broad audiences. Obtaining and maintaining sufficient scale is, in turn, crucial to the subscription and advertising revenues required to fund further content creation and operations.

Partnerships, joint ventures, and combinations will thus likely play a central role in the evolving media sector. Survey evidence suggests that viewers may be nearing their limit in the number of subscriptions they will maintain, and in their willingness to spend more to gain or retain access to preferred content.[19] In January 2024, 62% of respondents told The Motley Fool they felt there were too many streaming services, up from 53% in 2022; 46% said they are subscribed to more video-streaming services than they were a year ago, down from 54% in 2022; and 37% said they subscribed to fewer services, up from 13% in 2022.[20] According to a 2023 Deloitte survey, 48% of respondents said they would cancel their favorite paid streaming video-on-demand service if monthly prices went up by $5.[21]

The fragmentation of content delivery is a challenge for consumers, and addressing this challenge is key to generating consumer benefits from the ongoing growth in content. Policymakers should take care not to inhibit providers from experimentation, and instead to allow the market to determine the industry’s structure and the content that individual streaming services provide.

The recent landmark agreement between Disney and Charter Communications illustrates the sort of innovative hybrid revenue models that can arise from combining linear television and streaming rights.[22] In describing the deal, Walt Disney Co. CEO Bob Iger and Charter Communications CEO Chris Winfrey said that it “recognizes both the continued value of linear television and the growing popularity of streaming services while addressing the evolving needs of our consumers.”[23] The partnership highlights how market participants increasingly transcend traditional business and regulatory boundaries in order to create hybrid offerings that are responsive to consumer preferences.

The need to maintain this flexibility in contracting and bundling should inform antitrust assessments of the sector, emphasizing the importance of allowing experimentation that addresses consumer demand effectively. Traditional media companies will inevitably try a range of different business models and collaborations—both vertical and horizontal—to better attract viewers and improve their operations. The content sector is in transition, and antitrust enforcement should promote, not stymie, innovation and experimentation to increase benefits for all companies and viewers in the marketplace.

Given the rapidly evolving marketplace, regulatory approaches based on traditional static equilibrium models and backward-looking market definitions are unlikely to adequately capture the competitive dynamics. In the market for both viewers and advertisers, superficially differentiated services like Google, Apple, and Amazon, on the one hand, and Disney, Comcast, and HBO, on the other, are direct and vigorous competitors. Policymakers must recognize the transient nature of apparent market positions and adopt flexible frameworks that permit novel arrangements for content creation, as well as distribution, monetization, marketing, and R&D. Such regulatory humility is essential to avoid stifling beneficial market innovations and to support sustained consumer-welfare gains in the video marketplace.

V. Conclusion

The video marketplace has undergone rapid and radical transformation since the advent of television, through phases exemplified by companies like ABC, CBS, and NBC; TBS, MTV, and ESPN; HBO, Showtime, and Cinemax; Blockbuster Video; Netflix, Amazon, and Apple; Disney+, Hulu, and Peacock; and TikTok, Instagram, and YouTube.

At every step, the market has delivered substantial consumer benefits, including greater flexibility, choice, and cost efficiencies. At the same time, these shifts have also presented new challenges. Today’s market is notably marked by fragmentation, subscription fatigue, monetization challenges, and rising capital costs.

Policymakers and industry stakeholders must embrace regulatory flexibility and innovation-friendly competition policies to foster continued experimentation. By doing so, they will better align industry practices with consumer preferences, ensuring sustained growth and consumer welfare in this dynamic market.

[1] Subscribers for Select US Video Subscription Services, S&P Global Market Intelligence (Dec. 2024); Bevin Fletcher, Amazon’s Prime Video Aims to Be One-Stop Video Entertainment Hub, StreamTV Insider (Mar. 6, 2024), https://www.streamtvinsider.com/video/amazons-prime-video-aims-be-one-stop-video-entertainment-hub (Amazon Prime Video counts reflect MoffettNathanson’s estimates of active U.S. subscriber base. Hulu counts include subscribers who have access through their Hulu +LiveTV subscription).

[2] This includes the reduction in output from the 2023 Writers Guild of America strike; between 2009 and 2022, the increase was 300%. See Lesley Goldberg, Peak TV Is Officially Over: FX Tallies 14 Percent Drop in Scripted Series for 2023, Hollywood Report. (Feb. 9, 2024),https://www.hollywoodreporter.com/tv/tv-news/scripted-originals-fall-14-percent-2023-fx-tally-1235821248.

[3] Nielsen, L+3, Total Day (6a-6a), Duration Weighted Delivery, Dual Feed Date Range: 01/01/2024-12/29/2024, 12/30/2013-12/28/2014, 2024 Share of Total TV Usage, 2014 Share of Linear TV Usage.

[4] Id.

[5] See Jeremy Goldman, YouTube TV Price Hike Follows Platform’s Massive Living Room Growth, eMarketer (Dec. 12, 2024), https://www.emarketer.com/content/youtube-tv-price-hike-follows-platform-s-massive-living-room-growth.

[6] Less Than 50% of US Households Now Subscribe to Pay TV, as Cord-Cutting Jumps More Than Expected, eMarketer (Mar. 7, 2023), https://www.emarketer.com/press-releases/less-than-50-of-us-households-now-subscribe-to-pay-tv-as-cord-cutting-jumps-more-than-expected.

[7] The ability to easily switch services gives consumers implicit bargaining power in the form of churn risk, with approximately 42% of U.S. streaming consumers regularly subscribing, canceling, and re-subscribing to services based on content availability. See Daniel Frankel, Disney Bundlers Are 59% Less Likely To Churn, Research Company Says (Chart), NextTV (Jul. 9, 2024), https://www.nexttv.com/news/disney-bundlers-59-less-likely-to-churn-research-company-says-chart.

[8] United States et al. v. Nexstar Media Grp. Inc. & Tribune Media Co., Complaint at 13-14, Case No. 1:19-cv-02295 (D.D.C. Jul. 31, 2019), available at https://www.justice.gov/archives/opa/press-release/file/1189776/dl.

[9] Brian Steinberg, Netflix Says Ad Tier Reaches 70 Million Users Globally, Variety (Nov. 12, 2024), https://variety.com/2024/tv/news/netflix-ad-tier-reaches-70-million-global-users-1236207015.

[10] Adrian Pennington, Prime Video’s Ad Tier Entry Makes Immediate Impact on Ad-Supported Streaming, StreamTV Insider (May 7, 2024), https://www.streamtvinsider.com/advertising/prime-videos-ad-tier-entry-makes-immediate-impact-ad-supported-streaming.

[11] Kevin Westcott et al., Streaming Video at a Crossroads: Redesign Yesterday’s Models or Reinvent for Tomorrow?, Deloitte 2024 Digit. Media Trends (Mar. 20, 2024), https://www2.deloitte.com/us/en/insights/industry/technology/digital-media-trends-consumption-habits-survey/2024/customization-and-personalization-lead-the-svod-revolution.html.

[12] John Koblin, What Happened to Our Ad-Free TV?, N.Y. Times (May 26, 2024), https://www.nytimes.com/2024/05/26/business/media/ads-tv-streaming.html.

[13] See Alex Sherman, TikTok and Instagram Inch Closer to the Streaming Wars as Competitive Barriers Blur, CNBC (Jul. 5, 2012), https://www.cnbc.com/2021/07/05/tiktok-and-instagram-inch-closer-to-streaming-wars.html.

[14] See Westcott et al., supra note 10. (“On average, US subscribing households spend US$61 per month on four SVOD services. Additionally, 68% of consumers surveyed pay for either a TV subscription or live streaming TV.”).

[15] Id.

[16] See, e.g., Gregory Ellis, Streaming Has a Consumer and a Piracy Problem; The Answer Lies in the Music Industry, Alliotts (Sep. 22, 2023), https://www.alliotts.com/articles/streaming-has-a-consumer-and-a-piracy-problem-the-answer-lies-in-the-music-industry; Gintaras Radauskas, With Streamers Raising Prices, Digital Piracy Is Back, Cybernews (Nov. 15, 2023), https://cybernews.com/editorial/streaming-price-fuel-torrenting.

[17] Arun Perinkolam et al., Impatience and Price Motivate Consumers to Watch Pirated Content and Borrow SVOD Passwords, Deloitte 2024 Digit. Media Trends (Oct. 8, 2024), https://www2.deloitte.com/us/en/insights/industry/technology/digital-media-trends-consumption-habits-survey/2024/password-sharing-pirated-content.html.

[18] See Tony Maglio, ‘Big Bang Theory’: How Its Ratings Are Different Than Every Other Show’s, TheWrap (May 4, 2016), https://www.thewrap.com/big-bang-theory-tv-ratings-season-9-finale-cbs; Michael Schneider, The 100 Most-Watched Telecasts of 2024, Variety (Dec. 27, 2024), https://variety.com/2024/tv/news/most-watched-shows-2024-tracker-young-sheldon-super-bowl-olympics-oscars-1236260223; Press Release, ‘Tracker’ Was Television’s #1 Show in 2023-24, CBS (May 22, 2024), https://www.prnewswire.com/news-releases/tracker-was-televisions-1-show-in-2023-24-302153267.html.

[19] See Westcott et al., supra note 10.

[20] Jack Caporal, State of Streaming 2024: Streaming Services and Consumer Sentiment, Motley Fool (Mar. 18, 2024), https://www.fool.com/research/state-of-streaming.

[21] See Westcott et al., supra note 10.

[22] See Dawn Chmielewski, Disney, Charter Deal Reshapes Media Landscape, Executives Say, Reuters (Sep. 14, 2023), https://www.reuters.com/business/media-telecom/disney-charter-deal-reshapes-media-landscape-executives-2023-09-14.

[23] Press Release, The Walt Disney Company and Charter Communications Announce Transformative Agreement for Distribution of Disney’s Linear Networks and Direct-To-Consumer Services, Walt Disney Co. (Sep. 11, 2023), https://thewaltdisneycompany.com/disney-charter-spectrum-agreement.

 

Introduction

Ex-ante regulatory frameworks for digital markets—such as the European Union’s Digital Markets Act (DMA),[1] the United Kingdom’s Digital Markets, Competition, and Consumers Act (DMCC),[2] and Section 19a of the German Act Against Competition Restraints (GWB)[3]—target specific practices deemed harmful to competition and consumers, including the oft-debated subject of “self-preferencing.”

Typically, such regimes impose prohibitions and obligations on a limited set of powerful technology firms. For instance, the DMA prohibits “gatekeepers” from favoring their own products in rankings and indexing; the DMCC bars firms with “strategic market status” from leveraging data advantages to benefit their own offerings; and Section 19a of the GWB restricts “firms of paramount significance” from privileging their own services over rivals.

Brazil appears to be following a similar path. Bill No. 2768/2022,[4] currently under congressional review, would require digital platforms with “essential access-control power” to provide “isonomic and non-discriminatory treatment to professional and end users.”

Despite these policy movements, however, neither the economic evidence nor the Conselho Administrativo de Defesa Econômica’s (CADE) legal precedents support a general presumption of harm or blanket ban against self-preferencing in Brazil. Rather than treating the conduct as inherently anticompetitive, Brazilian enforcement has to date favored a contextual, case-by-case approach. As I will try to address in this issue brief, this is likely not the correct regulatory direction to take.

The Concurrences Competition Law Dictionary offers a useful definition of self-preferencing:

Among exclusionary abuses of dominant position, self-preferencing is a conduct that falls under the more general category of abusive leverage, as it assumes an undertaking abusing its dominant position in a given market to extend that position in a closely related market (upstream, downstream or adjacent), to the detriment of competitor on the latter market.[5]

Under traditional competition law, self-preferencing is typically analyzed as a form of abuse of dominance, unilateral conduct, or monopolization. It may be understood as a specific type of leveraging or, alternatively, as a manifestation of the “raising rivals’ costs” (RRC)[6] theory. In this context, the conduct may be deemed unlawful when a dominant firm in one market favors its own products or services in a vertically related market, thereby either extending its dominance or impairing competitors’ ability to compete effectively in the second market.

While self-preferencing[7] has recently gained renewed attention in antitrust discussions surrounding digital markets,[8] the practice itself is far from new. It has long been a common business strategy in such traditional sectors as retail[9] and supermarkets,[10] where vertically integrated firms often promote their own brands or products. Moreover, many common and intuitive business practices—such as restaurants offering a “house wine,” gas stations selling their own fuel, large consulting firms providing accounting or legal services, or construction companies sourcing materials from their own subsidiaries—could be characterized, depending on how the conduct is framed by competition authorities, as tying practices or forms of “self-preferencing.”

Despite its growing prominence in discourse around digital markets, the term “self-preferencing” remains conceptually imprecise. The antitrust literature has not yet reached a conclusion on whether “self-preferencing” should be treated as a distinct category of harm or whether it is simply a new label for established theories of harm, such as leveraging, tying, or raising rivals’ costs (RRC). If self-preferencing is understood to apply across both digital and nondigital markets, then the characteristics of digital platforms—such as network effects, high economies of scale and scope, intensive use of data, low marginal costs, tipping dynamics, and significant (often collective) switching costs—raise other important questions. Do these characteristics transform the economic logic of self-preferencing and adapt it into a novel or slightly different theory of harm? Or, conversely, is the analytical core of traditional antitrust theory the same, with the only question being how to assess the practice within a given market context under a rule-of-reason analysis? Many major questions are yet to be answered either by the academic literature or competition-law decisions. Pablo Colomo describes the current use of “self-preferencing” as “an epithet capturing a diverse range of scenarios,” underscoring the conceptual vagueness that undermines its legal utility.[11]

Even under these various names and without a clearly defined legal framework, competition authorities have rarely treated self-preferencing as a priority for enforcement. This historical ambivalence raises another serious question: if the practice has not warranted strict antitrust intervention, should it now be subject to sweeping ex-ante prohibitions in the digital realm?

I. The Economics of Self-Preferencing: A Brief Review of the Literature

To understand the dynamics of self-preferencing, it is important to recognize that favoring one’s own products is often a rational and economically expected behavior. Michael Salinger,[12] former director of the U.S. Federal Trade Commission’s (FTC) Bureau of Economics, has noted that self-preferencing can stem from legitimate business strategies such as vertical or conglomerate integration aimed at eliminating double marginalization (EDM), which typically results in lower prices and procompetitive outcomes.

This rationale has been addressed in the broader economic literature on vertical integration. Nobel laureate Oliver Hart, building on the theory of transaction costs and incomplete contracts, explains that vertical integration can be understood as a consequence of the high costs of incomplete contracts:

In particular, when it is too costly for one party to specify a long list of the particular rights it desires over another party’s assets, then it may be optimal for the first party to purchase all rights except those specifically mentioned in the contract. Ownership is the purchase of these residual rights of control. Vertical integration is the purchase of the assets of a supplier (or of a purchaser) for the purpose of acquiring the residual rights of control.[13]

In this light, it is to be expected that firms vertically integrated under a single economic entity will rely on their own inputs, rather than sourcing them from competitors. Otherwise, vertical integration would become economically irrational or unsustainable, often leading to divestitures or spinoffs.

Therefore, in vertically integrated structures, some form of self-preferencing is not only natural, but is often a manifestation of efficiency. By internalizing transactions and eliminating double markups, vertical integration reduces costs, which can translate into lower prices and improved outcomes for consumers. As a result, in this context, self-preferencing tends to be the rule, rather than the exception, and its baseline economic effect is typically procompetitive.

Furthermore, this commercial behavior of “self-preferencing” one’s own services or products reflects a broader economic logic found on both the supply and demand sides. For instance, on the consumer side, Nobel-winning economists Daniel Kahneman and Richard Thaler describe how the “endowment effect”[14] illustrates how individuals tend to assign greater value to goods they own—e.g., assuming their used car is worth more than others of similar value or quality. While this behavior is rooted in psychology, it offers an intuitive parallel: just as consumers tend to “self-preference” their own possessions, it is natural for firms to “self-preference” their own products. This is even more reasonable for firms, when considering objective factors such as improved supply-chain control, better inventory management, or reduced transaction costs.

Moving beyond theoretical frameworks, it is essential to consider the empirical literature on self-preferencing. As discussed earlier, the term may encompass practices like leveraging, tying, or raising rivals’ costs in nondigital markets. But as the purpose of this brief is to assess proposals for a blanket ban on self-preferencing in digital markets, the literature review here will be limited to studies focusing on that specific context.

Geoffrey Manne has examined empirical studies on platform economics and found that self-preferencing practices—though observed in different settings—tend to produce positive welfare outcomes or, at worst, neutral or ambiguous long-term effects. On this basis, he challenges the use of presumptions in antitrust analysis of vertical discrimination:

The problem [with the intensive criticism of illegal presumption of vertical discrimination], however, is that the claims of presumptive harm from vertical discrimination are based neither on sound economics nor evidence.[15]

A more recent literature review analyzing the effects of self-preferencing by digital platforms similarly concludes that “the welfare effects of self-preferencing are contingent on the specific form of self-preferencing as well as the market environment in which it occurs.”[16] While the authors do not explicitly offer a competition-policy recommendation, the results logically point toward rejecting any blanket presumption of illegality and instead supporting a case-by-case, effects-based approach.

Along the same lines, economists Emilie Feyler and Veronica Postal reach a similar conclusion in their review of self-preferencing by pricing algorithms, a context that raises further concerns due to the delegation of pricing decisions to automated systems.[17] They observe that:

There is no consensus from the economic literature on whether procompetitive benefits or possible anticompetitive considerations prevail in the context of self-preferencing algorithms used by digital platforms. Nor is there a consensus on the welfare effects of a policy intervention to correct bias in algorithmic recommendations. Determining the net impact of self-preferencing algorithms on competition and consumer welfare requires individualized analysis accounting for the workings of specific algorithms, competitive context, and market environment.[18]

Finally, the fact that self-preferencing often produces pro-competitive effects has been acknowledged even by some of the competition authorities that are currently considering more targeted ex-ante regulatory measures for digital platforms. For example, the Australian Competition & Consumer Commission (ACCC) noted in a report on digital practices:

Although self-preferencing conduct is often benign, self-preferencing conduct that leverages market power over a key online service into a related service, which is not justified by a procompetitive rationale, can distort competition and decrease consumer welfare.[19]

In short, the economic literature makes clear that any analysis of self-preferencing ultimately hinges on the specific facts of each case, as well as the broader market context. There is no consensus—either theoretical or empirical—to supports a general condemnation of the practice. On the contrary, many studies recognize that self-preferencing can produce efficiency gains, especially when linked to vertical integration or innovation strategies.

Accordingly, the current body of economic research does not support a presumption of illegality, let alone a blanket ban on self-preferencing by digital platforms. This is not to say that self-preferencing—as a form of leveraging, tying, or raising rivals’ costs—can never harm competition or consumers. It can and should be assessed according to the legal and economic evidence of each particular case. But moving from the current analytical stage, in which the practice is presumptively legal and often efficiency-enhancing, toward strict regulatory prohibitions appears to be, at the very least, very premature.

II. Despite CADE’s Limited Case Law on Self-Preferencing, the Rule of Reason Still Reigns

Building on this economic rationale, international antitrust case law—including the approach adopted by Brazil’s competition authority, CADE—generally treats self-preferencing as presumptively lawful and subject to a rule-of-reason analysis. As will be discussed, however, CADE’s case law on self-preferencing remains notably limited. Given the distinctive characteristics of digital markets, some Brazilian antitrust scholars—such as Beatriz Kira and Diogo Coutinho[20]—have argued that CADE should not only adapt existing theories of harm to the digital context, but also develop new ones tailored to these environments. From this perspective, self-preferencing is seen as a novel theory of harm that requires further conceptual and doctrinal refinement in digital competition enforcement.

In terms of case law, the EU’s Google Shopping decision is widely regarded as the international benchmark for self-preferencing in digital markets. The core allegation in that case[21] was that Google leveraged its market power to manipulate search results, favoring its own service—Google Shopping—by limiting the visibility of competing price-comparison platforms. The case illustrates a sharp divergence in how competition authorities assess similar conduct. In 2017, the European Commission[22] fined Google €2.42 billion for abusing its dominance in the search-engine market by granting an illegal advantage to its own comparison service. By contrast, in 2019, CADE’s Administrative Tribunal[23] dismissed the charges in a split 3–3 vote, with the former president casting the deciding vote in favor of dismissal.

The key difference in approach lies in evidentiary thresholds. While the European Commission found potential anticompetitive effects sufficient to establish liability, the majority at CADE relied on Technical Note No. 34/2018[24] from its Department of Economic Studies (DEE), which concluded that Google’s conduct stemmed from product innovations aimed at improving the user experience and delivering more accurate search results. The practice was therefore judged to be procompetitive, rather than anticompetitive. The dismissal was also grounded in a lack of robust evidence and a net-effects analysis of alleged predatory innovation, which found overall procompetitive efficiencies. In short, while the EU condemned Google’s self-preferencing, CADE saw insufficient grounds for a condemnation. This raises even further doubts about whether an ex-ante prohibition on self-preferencing is necessary for Brazil as, in the benchmark case, CADE decided that the practice was not ultimately anticompetitive. This lack of international convergence is also supported by the antitrust literature.[25]

The self-preferencing discussion in Brazil has not been limited to monopolization cases; it has also appeared (albeit rarely) in vertical-merger reviews. One notable example was the Nike/SBF–Centauro transaction, in which self-preferencing concerns played a central role. The merger involved the acquisition by SBF Group—owner of the “Centauro” sporting-goods retail chain—of the entire equity interest in Brazilian Nike that was previously held by the international Nike Group. As a result of the deal, SBF became the exclusive distributor of Nike products in Brazil and took over the operation of Nike’s physical and online retail channels. According to the parties, the transaction would allow SBF to integrate Nike’s operations into its existing omnichannel retail model.

In reviewing the deal, CADE’s General-Superintendent[26]—the authority’s investigative arm—found no evidence of either total or partial vertical foreclosure (discussed here under the label “self-preferencing”). While it acknowledged Nike’s dominant position in the “wholesale market for sporting goods” (a dominant position is presumed under Brazilian competition law when a firm’s market share exceeds 20% in a defined relevant market, according to Article 36, §2º), the SG concluded that SBF would not have the ability to foreclose access to Nike products for competing retailers.

The decision was appealed[27] by Netshoes, a competitor of SBF in the retail market, which argued that vertical foreclosure (self-preferencing) remained a risk and that behavioral remedies—specifically, an obligation to distribute Nike products on isonomic and nondiscriminatory terms—were necessary. In the end, CADE’s Administrative Tribunal and the merging parties negotiated an “administrative consent decree”[28] that required SBF to ensure equal and nondiscriminatory distribution of Nike products, thus addressing the self-preferencing concerns raised during the merger review.

The SulAmérica/Rede D’Or merger was another notable vertical-integration case in which the concept of “self-preferencing” appeared. But unlike the previous case, in which the SBF Group had both physical and online retail stores selling sporting goods (Nike and many other sporting-goods brands’ products), the SulAmérica case did not involve digital markets.

The transaction involved the incorporation of SulAmérica by Rede D’Or, resulting in the unification of their operations. Rede D’Or is one of the largest private health-care networks in Brazil, offering hospital, clinical, diagnostic, and related services across several states. SulAmérica, in turn, is a major player in the Brazilian insurance market, with a strong focus on health, dental, life, and personal accident coverage. According to the parties, the deal aimed to align and expand their respective health-related ecosystems to deliver integrated care and insurance services.

The vertical theory of harm raised in the case was that SulAmérica, as a health insurer, would likely “self-preference” Rede D’Or’s hospitals and clinics by steering patients toward them and discriminating against competing health-care providers—in other words, a traditional vertical-foreclosure scenario. Notably, a technical note from CADE’s Department of Economic Studies (DEE)[29] did not refer to the conduct as “self-preferencing,” but instead used the more established categories of “input and customer foreclosure.” The merger was ultimately approved without restrictions, based on CADE’s assessment that the parties lacked both the ability and the incentive to engage in anticompetitive foreclosure of rival hospitals.

These two cases show that CADE is attentive to conduct that could be labeled as self-preferencing. As previously discussed, however, the term “self-preferencing” lacks standardized definition. In both vertical-merger cases it examined, CADE analyzed the transactions under classical vertical theories of harm—namely, input and customer foreclosures, whether total or partial. These well-established concepts have more recently begun to be described as “self-preferencing” in both digital and nondigital markets. In this sense, for instance, one could say that AT&T would “self-preference” Time Warner content on its cable TV platforms in the noted AT&T/Time Warner merger, instead of saying that AT&T would partially foreclosure Time Warner contents to AT&T’s competitors. But these are, at their core, classic vertical-foreclosure scenarios—not the exact same type of theory applied in Google Shopping. While such rebranding (from vertical foreclosure to “self-preference”) may not be inaccurate, it is difficult, if not impossible, to reconcile this form of vertical analysis with the standalone abuse-of-dominance theory that underlies calls for a blanket ban on self-preferencing.

Returning to the monopolization cases, the rule-of-reason approach, as applied by CADE in the Google Shopping case, remains the standard framework to assess self-preferencing allegations. A significant ongoing example is the MSC/Maersk[30] investigation, in which the Brazilian Association of Customs Terminals and Premises (ABTRA) accuses Maersk, MSC, and their jointly controlled terminal operator BTP of leveraging their vertical integration and cooperative agreements to self-preference and favor BTP over competing terminals at the Port of Santos. The alleged conduct involves discriminatory and exclusionary practices that allegedly grant BTP undue competitive advantages in container-handling and bonded-storage services.

At this stage, CADE’s General-Superintendent has requested a study from the Department of Economic Studies (DEE) to assess whether the conduct generates efficiencies and, if so, how they compare to the potential anticompetitive effects. The fact that this self-preferencing case is being analyzed under a rule-of-reason framework, rather than treated as per-se illegal, reinforces CADE’s preference for case-by-case evaluation over broad ex-ante prohibitions. That Brazil’s most important current “self-preferencing” case falls totally outside the digital-market context (few markets are more “nondigital” than ship transportation and container handling), however, highlights the lack of clarity and standardization around the term.

III. Conclusion

Given the current stage of the self-preferencing debate in competition law, condemning a unilateral conduct practice solely under a self-preferencing theory of harm—regardless of how the term is defined—remains a high bar for enforcement.

Regarding the antitrust duty to deal, “[c]ompetition law does not impose a general duty to share competitive advantages with rivals and does not protect the structure of the market; hence, not every exclusionary effect automatically undermines competition.”[31] Similarly, as the practice carries a procompetitive and presumably legal rationale, “firms are not under a general duty to subsidize rivals.”[32]

Second, “self-preferencing is often inextricably linked to the procompetitive benefits that come with the integration of products and services. […] Experience and economic analysis suggest that the practices that are typically labeled as forms of self-preferencing should not be deemed prima facie unlawful irrespective of their effects – whether de jure or de facto.”[33]

In Brazil, a recent study conducted by the Legal Grounds Institute[34] quantitatively analyzed CADE’s decisions involving self-preferencing allegations, offering a holistic assessment of how the authority has approached the issue. While the number of relevant cases remains limited, the study offers important insights into the ongoing political debate over ex-ante regulation. One of its primary conclusions was that the low conviction rate does not justify a general prohibition: “When we look at markets in general, there have been few convictions for self-preferencing in CADE’s jurisprudence, with a conviction rate of 27% over the past 10 years.”[35] The study also notes that CADE tends to view self-preferencing as potentially pro-competitive:

As the analysis of CADE’s case law shows, the impact of self-preferencing conduct is far from being inherently anticompetitive. Indeed, even in convictions of a particular practice of self-preferencing, there is usually divergence among commissioners, including in so-called digital markets, what shows a lack of certainty that is uncongenial to a per se regulation.[36]

Moreover, the Legal Grounds Institute study underscores that CADE’s record does not justify a regulatory policy shift in digital markets:

There is no history at CADE of convictions for self-preferencing conduct in digital markets. One may argue that this supposed lenience is the very reason for changing the approach, but the point here is that making self-preferencing ex ante forbidden may be a radical move that inverts the spectrum of a vertical practice that may bring efficiencies and benefits to consumers in different contexts.[37]

The message of this issue brief is straightforward. Despite the absence of a clear and consistent definition, self-preferencing is presumptively lawful under CADE’s approach, given its potential procompetitive benefits. The next logical step would be to build a more substantial body of case law through enforcement actions against self-preferencing, condemning such conduct where actual anticompetitive harm is proven (since, in the current scenario and according to the Legal Grounds Institute study, CADE has never convicted any self-preferencing case in digital markets).

With sufficient precedent, it may become reasonable to adopt a presumption of illegality of self-preferencing in digital markets, shifting the burden of proof to defendants or applying other procedural or competition tools to ease enforcement. But only if that proves insufficient should ex-ante regulation be considered—i.e., once it is clearly demonstrated that self-preferencing typically harms competition and consumers, and the risk of overenforcement is low.

It is important to emphasize that moving through each of these stages typically requires time, case development by competition authorities, and the support of robust economic evidence. Even if one the finds the current pace of antitrust enforcement exceedingly slow, bypassing these steps—i.e., jumping directly from the status quo stage to full ex-ante prohibition, as envisioned in PL 2768/2022 or as may potentially be proposed by the Ministry of Finance—ignores the necessary legal and economic foundations for such an extreme shift in policy. Such an approach cannot be considered sound regulatory practice.

[1] Regulation 2022/1925, of the European Parliament and of the Council of 14 Sept. 2022, on Contestable and Fair Markets in the Digital Sector and Amending Directives (EU) 2019/1937 and 2020/1828 (Digital Markets Act), Eur. Comm. (Oct. 12, 2022), at 1, https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32022R1925.

[2] Digital Markets, Competition and Consumers Act 2024, UK Gov. (May 24, 2024), at c. 13, https://www.legislation.gov.uk/ukpga/2024/13/enacted.

[3] Gesetz Gegen Wettbewerbsbeschränkungen [GWB] [Act Against Restraints of Competition], Fed. Minist. Justice (Jul. 27, 1957), BGBl. I at 1081, https://www.gesetze-im-internet.de/englisch_gwb/englisch_gwb.html.

[4] Projeto de Lei No. 2.768, Câmara dos Deputados (2022), https://www.camara.leg.br/proposicoesWeb/prop_mostrarintegra?codteor=2214237&filename=PL%202768/2022.

[5] Self-Preference, Concurr. Dict. Compet. Law (Sep. 15, 2022), https://www.concurrences.com/en/dictionary/self-preference-111802.

[6] Steven C. Salop & David T. Scheffman, Raising Rivals’ Costs, 73 Am. Econ. Rev. 267 (1983), https://www.jstor.org/stable/1816853.

[7] Guillaume Duquesne et al., What Constitutes Self-Preferencing and Its Proliferation in Digital Markets, in Digital Markets Guide – Fourth Edition (Global Competition Rev. ed., Oct. 2, 2024), https://globalcompetitionreview.com/guide/digital-markets-guide/fourth-edition/article/what-constitutes-self-preferencing-and-its-proliferation-in-digital-markets.

[8] Competition Policy in Digital Markets: The Combined Effect of Ex Ante and Ex Post Instruments in G7 Jurisdictions, Organ. Econ. Co-oper. Dev. (Oct. 4, 2024), https://doi.org/10.1787/80552a33-en.

[9] Amanda Athayde, Direito da Concorrência e Supermercados: Como Essas Plataformas de Dois Lados Podem Trazer Riscos aos Consumidores?, 16 Rev. Direito GV (Feb. 10, 2020), https://doi.org/10.1590/2317-6172201940.

[10] Carolina Saito, Self-Preferencing Apenas em Mercados Digitais?, JOTA (Dec. 21, 2022), https://www.jota.info/artigos/self-preferencing-apenas-em-mercados-digitais.

[11] Pablo Ibáñez Colomo, Self-Preferencing: Yet Another Epithet in Need of Limiting Principles, 43 World Compet. 417 (2020), https://doi.org/10.54648/woco2020022.

[12] Michael A. Salinger, Self-Preferencing, in Global Antitrust Institute Report on the Digital Economy (Douglas H. Ginsburg & Joshua D. Wright eds., 2020), at 329, 329–368, https://gaidigitalreport.com/2020/08/25/self-preferencing.

[13] Sanford J. Grossman & Oliver D. Hart, The Costs and Benefits of Ownership: A Theory of Vertical and Lateral Integration, 94 J. Pol. Econ. 691 (1986), https://doi.org/10.1086/261404.

[14] Daniel Kahneman, Jack L. Knetsch, & Richard H. Thaler, Experimental Tests of the Endowment Effect and the Coase Theorem, 98 J. Pol. Econ. 1325 (1990), https://doi.org/10.1086/261737.

[15] Geoffrey A. Manne, Against the Vertical Discrimination Presumption, Concurrences No. 2-2020 (May 2020), https://www.concurrences.com/en/review/issues/no-2-2020/foreword/against-the-vertical-discrimination-presumption-94267-en.

[16] Yuta Kittaka, Susumu Sato, & Yusuke Zennyo, Self-Preferencing by Platforms: A Literature Review, 66 Jpn. World Econ. 101191 (2023), https://doi.org/10.1016/j.japwor.2023.101191.​

[17] Emilie Feyler & Veronica Postal, Can Self-Preferencing Algorithms Be Pro-Competitive?, CPI Antitrust Chron. (Jun. 2023), at 5, available at https://www.competitionpolicyinternational.com/wp-content/uploads/2023/06/5-CAN-SELF-PREFERENCING-ALGORITHMS-BE-PRO-COMPETITIVE-Emilie-Feyler-Veronica-Postal.pdf.

[18] Id.

[19] Digital Platform Services Inquiry: Discussion Paper for Interim Report No. 5: Updating Competition And Consumer Law For Digital Platform Services, Aust. Compet. Consum. Comm. (Feb. 2022), available at https://www.accc.gov.au/system/files/Digital%20platform%20services%20inquiry.pdf.

[20] Beatriz Kira & Diogo R. Coutinho, Ajustando as Lentes: Novas Teorias do Dano Plataformas Digitais, 9 Rev. De Defesa da Concorrência 82, (Jun. 2021), https://doi.org/10.52896/rdc.v9i1.734

[21] Dario da Silva Oliveira Neto, Otávio Augusto de Oliveira Cruz Filho, & Alexandre Cordeiro Macedo, The Rule of Reason and the Fundamentals Against More Presumption-Based Illegality Legal Standards: Highlights on CADE’s Decisions on Digital Economy Issues, 12  J. Antitrust Enforc. 570 (2023), https://doi.org/10.1093/jaenfo/jnad042.

[22] Commission Decision of 27 June 2017, Case AT.39740 – Google Search (Shopping), 2018 O.J. (C 9) 11, Eur. Comm. (Jun. 27, 2017), available at https://ec.europa.eu/competition/antitrust/cases/dec_docs/39740/39740_14996_3.pdf.

[23] Paulo Burnier da Silveira & Victor Oliveira Fernandes, The Brazilian Competition Authority Decides to File Charges Against a Multinational Technology Company Due to One of Its Online Shopping Platform Features (Google), e-Competitions Bulletin, Art. No. 91181 (Jun. 2019) https://www.concurrences.com/en/bulletin/news-issues/june-2019/the-brazilian-competition-authority-decides-to-file-charges-against-a-en.

[24] Nota Técnica nº 34/2018/DEE/CADE, Processo Administrativo nº 08012.010483/2011-94, Conselho Administrativo de Defesa Econômica (Nov. 19, 2018), https://sei.cade.gov.br/sei/modulos/pesquisa/md_pesq_documento_consulta_externa.php?DZ2uWeaYicbuRZEFhBt-n3BfPLlu9u7akQAh8mpB9yODP3s4Xkowv9qF35FkSAM2hxXtRYnrpkhxBAKRhUpTIm_kb0guWVNihPtzC9415xikED6rDoAPiQUYFTrqj2ZO.

[25] Stephanie Vendemiatto Penereiro, Gustavo H. Kastrup, & Vitor Jardim Machado Barbosa, My Game, My Rules? O Enforcement Concorrencial do Brasil e do Mundo Relacionado ao Self-Preferencing, 1 Revista do IBRAC 59 (2023), https://revista.ibrac.org.br/index.php/revista/article/view/4/200.

[26] Anexo ao Parecer nº 17/2020/CGAA2/SGA1/SG/CADE, Ato de Concentração nº 08700.000627/2020-37, Conselho Administrativo de Defesa Econômica (Aug. 14, 2020), https://sei.cade.gov.br/sei/modulos/pesquisa/md_pesq_documento_consulta_externa.php?DZ2uWeaYicbuRZEFhBt-n3BfPLlu9u7akQAh8mpB9yNTvbEsQO6hM4OLUe5kpdugnCyZMLfLGz5eayQTZTy02uMnJteFo2jdtk-_FQpl06dcoGt93zn7JdCop-gADKVf.

[27] Netshoes, Recurso, Ato de Concentração nº 08700.000627/2020-37, Conselho Administrativo de Defesa Econômica (Sep. 1, 2020), https://sei.cade.gov.br/sei/modulos/pesquisa/md_pesq_documento_consulta_externa.php?DZ2uWeaYicbuRZEFhBt-n3BfPLlu9u7akQAh8mpB9yMvMUTZBztCiz5nljNFMlkJvwl3YPZynNDv5kWSSpe2bDOZrYHo1dL8Ce0Fxl3C11PFB8qSKBzf7vVP88QdTuqA.

[28] Acordo em Controle de Concentrações (ACC), Ato de Concentração nº 08700.000627/2020-37, Conselho Administrativo de Defesa Econômica (Sep. 1, 2020), https://sei.cade.gov.br/sei/modulos/pesquisa/md_pesq_documento_consulta_externa.php?DZ2uWeaYicbuRZEFhBt-n3BfPLlu9u7akQAh8mpB9yOJ5Xgeo8sm-drSUFY508gQJfKN3pzncQ4fYtcjfPxiS6altjTAqLGQxIK58F6YEAmxwhfAXx-sQcY3S1AuruPN.

[29] Nota Técnica nº 30/2022/DEE/CADE, Ato de Concentração nº 08700.003959/2022-35, Conselho Administrativo de Defesa Econômica (Nov. 7, 2022), https://sei.cade.gov.br/sei/modulos/pesquisa/md_pesq_processo_exibir.php?1MQnTNkPQ_sX_bghfgNtnzTLgP9Ehbk5UOJvmzyesnbE-Rf6Pd6hBcedDS_xdwMQMK6_PgwPd2GFLljH0OLyFT3Fr3NpoLkb1FQLNxKAd0MsaW6nmX9fL2xwRSOPcl3m.

[30] Nota Técnica nº 33/2023/CGAA11/SGA1/SG/CADE, Inquérito Administrativo nº 08700.003945/2020-50, Conselho Administrativo de Defesa Econômica (Apr. 19, 2023), https://sei.cade.gov.br/sei/modulos/pesquisa/md_pesq_documento_consulta_externa.php?HJ7F4wnIPj2Y8B7Bj80h1lskjh7ohC8yMfhLoDBLddY_X9cUsOiZo-Yts3uZC_BQ0G32xOmbb4JI-YXQ1fHyor_kWJ6kWm6VhlFtk9QIzigvujo4VOIKycL8XASUJFs-.

[31] Giuseppe Colangelo, Antitrust Unchained: The EU’s Case Against Self-Preferencing, Int’l Ctr. L. Econ. (Sep. 22, 2022), available at https://laweconcenter.org/wp-content/uploads/2022/11/ICLE-White-Paper-2022-09-22-Antitrust-Unchained-The-EUs-Case-Against-Self-Preferencing.pdf.

[32] Colomo, supra note 11.

[33] Id.

[34] Legal Ground’s Report on the Impact of Bill 2768 on Legal Certainty: Summary of the Findings, Leg. Grounds Inst. (2024), available at https://legalgroundsinstitute.com/wp-content/uploads/2024/10/FindingsReportSelf-Preferencingfinal.pdf.

[35] Id.

[36] Id.

[37] Id.

I. Introduction

The philosopher Jean-Francois Lyotard coined the term “grand narrative” (grands récits, or “grand stories” in French) to refer to the meta-discourses or totalizing narratives of modernity that provide ideologies with a legitimizing history.^[1] The prevalent (or, at least, the loudest) grand antitrust narrative of today posits that so-called “digital markets” present unique anticompetitive and evidentiary challenges that can only be adequately addressed through ex-ante rules or strong presumptions that mitigate (if not entirely overturn) the previous “dogmatic” adherence to economic analysis. This story—a subplot in a broader fairy tale about the alleged unmitigated failure of antitrust in the “neoliberal era”^[2]—functions as the lynchpin of a new antitrust ideology that seeks to redefine the role of competition and competition law in society.^[3] Some jurisdictions, especially those seeking to catch up with the European Union and the United States, appear to have swallowed this tale hook, line, and sinker.^[4]

But as narratives go, this one has a glaring problem. The supposed consensus is, in fact, more fiction than fact, and can only be sustained under a strained interpretation of events and an increasingly selective reading of the literature.

First, digital markets are not the den of anticompetitive iniquity they are sometimes made out to be. As Herbert Hovenkamp^[5] has recently noted, digital markets don’t present the symptoms typically associated with systemic monopoly, such as slow growth, sluggish investment in R&D, and collusion. In other words, this is not the industry that rational, cost-effective antitrust authorities should prioritize. Second, antitrust authorities have roughly lost as many cases as they have won in digital markets. Enforcers’ mixed track record, combined with the fact that even the victories were hard fought in court (and not without controversy outside of it), suggests that the impugned conduct was anything but unequivocally pernicious for competition. Third, despite claims that there exists a consensus that digital markets necessitate special treatment under the competition laws, or stricter carve-outs from such laws, relatively few countries to date have taken significant steps in that direction, and only four such regulatory frameworks have been put into place. And even among proponents of this new form of competition regulation, there is no real consensus about the sorts of laws that are best suited for the task.

II. There Is No Consensus that Digital Markets Are Uniquely Anticompetitive

It is often assumed that digital markets—a term generally understood to mean those core markets in which Google, Amazon, Meta, Microsoft, and Apple are active (“GAMMA”)[6]—are notably prone to anticompetitive conduct and particularly vulnerable to monopolistic outcomes (tipping, winner-take-all, etc.). A baseline problem with this narrative is that the companies typically lumped together as defining “digital markets” (in fact, nearly every modern market, from banking to groceries, includes at least some significant digital component) are actually quite heterogeneous.[7] Generalizations about the competitive landscape across dozens of products and services rendered by different companies at different points in time are thus unlikely to be useful in any meaningful sense.

As Apple CEO Tim Cook has noted:

Tech is not monolithic. That would be like saying “All restaurants are the same” or “All TV networks are the same.”[8]

For instance, while Google (Alphabet) and Facebook (Meta) are information-technology firms that specialize in online advertising, Apple remains primarily an electronics-hardware company, with around 75% of its revenue coming from the sale of iMacs, iPhones, iPads, and accessories. As Amanda Lotz of the University of Michigan has observed:^[9]

The profits on those [hardware] sales let Apple use very different strategies than the non-hardware [“Big Tech”] companies with which it is often compared.

As such, even referring to these companies as constituent of the same industry (let alone market) makes little sense. Unless, of course, one is guided by political goals outside the purview of traditional antitrust law—as, e.g., the so-called “neo-Brandeisians” are.^[10]

Take, for example, the European Union’s Digital Markets Act (“DMA”). The reason the DMA’s notion of “gatekeeper” status appears so contrived^[11] from an economic, legal, and technical perspective is because the law’s economic, legal, and technical arguments were reverse engineered to apply to a set of predetermined firms that have little in common, apart from being, for the most part, U.S.-based, technologically intensive, and vastly resourced. Another thing they share is having used innovative business models to disrupt traditional industries,^[12] in the process capturing massive rents and attracting millions of users. In other words, they are the world’s most successful “tech”^[13] companies.^[14]

But even on its own terms, the notion that there exists a consensus that these digital markets specifically are uniquely and systematically anticompetitive in a way that requires pushing the boundaries of antitrust law is belied by legal, economic, and scholarly evidence.

For one, the U.S. Federal Trade Commission’s (FTC) attempts to stifle the GAMMA firms (and, in the process, to rewrite U.S. antitrust doctrine) have been a mixed bag to date, and have met with mounting resistance.

In 2023, the FTC failed to block Meta’s acquisition of virtual-reality company Within Unlimited. As Corbin Barthold[15] pointed out, even The New York Times had to admit the case was “built upon a little-tested legal theory that the deal would hurt future competition in an untested market.”^[16] The FTC’s challenge advanced the tortured theory that a budding market was already in danger of suffering a “substantial lessening of competition” from an acquisition that could, for all intents and purposes, contribute to developing that market by providing greater scale to a firm that remained just a small (albeit important) player. This so-called “actual potential competition” doctrine, building on the false contention that the antitrust laws reflect a statutory preference for “organic growth over growth through acquisition,”^[17] would effectively allow the FTC to challenge any acquisition where the agency could show that the acquirer might hypothetically develop the acquired business independently. This, however, is a fringe reading of U.S. antitrust law that is in no way endorsed by the mainstream,^[18] and that would, if accepted, decimate M&A-driven growth and innovation across the economy.^[19]

In a similar vein, the FTC also failed to stop Microsoft’s acquisition of Activision Blizzard. In July 2023, Judge Jacqueline Scott Corley of the U.S. District Court for the Northern District of California ruled against the FTC, and denied the motion for a preliminary injunction. Contrary to the FTC’s assertions, the judge found that the deal would expand access to the Call of Duty franchise, especially since Microsoft had committed to keeping the popular action game on competitor Sony’s PlayStation platform (one of the FTC’s primary concerns). The FTC’s decision to appeal was met with strong backlash from venture-capital firms and investors, who argued in an open letter that the appeal “[threatened] the cycle of investment and entrepreneurship that drives America’s innovation economy.”^[20]  In addition, a group of 22 U.S. House Republicans sent a letter to the FTC urging it to drop the case and return to “sensible, consumer-oriented antitrust enforcement.”^[21]

The two cases’ failure momentarily foiled the agency’s aggressive attempts to recast vertical integration—which it had identified as one of the primary sources of the GAMMA firms’ success, and therefore a main source of concern—in the same light as horizontal mergers. But antitrust law has typically viewed the vertical integration as benign, and often even beneficial.^[22] As International Center for Law & Economics (ICLE) President Geoffrey Manne and former ICLE Director of Competition Policy Sam Bowman have explained:

Empirical research has consistently found vertical integration to be good for consumers through a number of mechanisms that allow for reduced costs and better product quality.^[23]

This is why antitrust merger enforcement has long relied on a fundamental distinction between horizontal and vertical mergers (or horizontal and vertical theories of harm, to be more precise): policymakers widely (and correctly) assume the former are far more likely to cause problems for consumers and competition than the latter.^[24]

Unable to persuade the courts, the FTC and U.S. Justice Department (“DOJ”) in 2023 issued draft merger guidelines that attempted to rewrite the law^[25] by, among other things, blurring the historical distinction between horizontal and vertical mergers,^[26] largely ignoring the costs of wrongly prohibiting or modifying mergers,^[27] and downplaying merger efficiencies.^[28] As many commentators have noted, however, the guidelines’ purpose is to provide a useful synthesis of established law, economic learning, and agency experience—not to create new doctrine.^[29] Far from forging a consensus, the 2023 draft merger guidelines thus far appear to have unraveled it.^[30]

There are even indications that, if they had their way, the agencies would cast all tech mergers—and perhaps even all mergers, period^[31]—as inherently problematic. For example, former Assistant U.S. Attorney General for Antitrust Jonathan Kanter and former FTC Chair Lina Khan have repeated the above-mentioned contention that U.S. antitrust law has a preference for “organic growth over acquisitions.”^[32] But, again, this interpretation of U.S. antitrust law generally, and of the Clayton Act specifically, is fanciful. As Gus Hurwitz and Geoffrey Manne have written on the 2023 merger guidelines:

Look at the antitrust agencies’ string of recent losses in major merger cases, including Microsoft’s acquisition of Activision and Meta’s acquisition of Within, to see that their views of antitrust law differ substantially from those of the courts.

The FTC has also suffered other important setbacks^[33] in its quest to realign antitrust law with the agenda to rein in U.S. tech firms. For example, the agency’s attempt to unwind Meta’s acquisitions of Instagram and WhatsApp, which it had waved through 10 years earlier, appears increasingly unlikely to succeed, given the competitive pressure exercised in the social-media space by the likes of TikTok, X.com, Mastodon, Bluesky, YouTube, Snapchat, Viber, Telegram, and Signal. The agency, nonetheless, appears undeterred.^[34] As Barthold put it:

The suit, premised on the notion that the social media market is static, has been overtaken by events.^[35]

Recent changes in the social-media market also cast doubt on the “tipping” theory often advocated by proponents of the DMA. Tipping theory suggests that competition in digital markets ceases once markets tip toward dominant players. Jinyul Ju of South Korea’s Pusan National University, who has criticized the Korea Fair Trade Commission’s (“KFTC”) platform-regulation bill, suggests that the tipping theory is “a mere hypothesis” without empirical basis. Ju also notes that:

Market definition becomes complex fact-based analyses with economic theories that from time to time lead to the intense debate in not only courtroom but also academic community. Any economic methods are contestable.^[36]

Thus, without substantial evidence, the tipping theory may risk false positives.

The Epic Games v. Apple saga put another dent in another one of the favorite dogmas of the new imaginary antitrust consensus: the idea that closed or semi-closed platforms are inherently worse for competition and consumers than open ones. The DMA elevates this dogma to the status of universal rule through the law’s interoperability requirement, laid down in Article 6(4).^[37] In 2021, however, the U.S. District Court for the Northern District of California found^[38] that Apple’s “walled-garden” model (specifically, its prohibition of third-party in-app payment systems and app stores) conferred important consumer benefits, including enhanced security and privacy. The court therefore upheld a much more nuanced understanding of the costs and benefits of each type of platform.^[39]

Other ongoing cases have also failed to curry anything even remotely close to unanimous support. The FTC’s 2023 case against Amazon,^[40] for instance, has been criticized for potentially harming consumer welfare if structural remedies are ultimately imposed in a bid to destroy Amazon’s incentive to give an “unfair” advantage to its own products on its own platform. As I argued in a piece co-authored with Geoffrey Manne,^[41] Amazon could presumably comply with any structural remedies designed to prevent it from exploiting its “anticompetitive flywheel” by shutting down its third-party Marketplace. Amazon itself would thus remain the only seller on Amazon.com—either running its logistics business solely for itself (the “Walmart option”) or offering its logistics service for off-platform sellers, as well (the “Shopify option”). This would be a huge step back for consumers and sellers currently active on Marketplace, although it may help some competitors who stand to gain from Amazon’s apparent ruin.

The case is emblematic of another questionable tenet of the “new antitrust consensus”: the idea that antitrust law should de-emphasize consumer welfare and focus on market structure, the protection of competitors, or vague notions of “fairness.” This idea is especially pertinent to digital markets, where prices are unquestionably low, and output is high. Such features mean that consumer harm is either difficult to prove, or that it does not exist.

Antitrust revisionists have sought to recast as reactionary any who would defend the hitherto uncontroversial view that antitrust law should remain tethered to robust economic analysis of effects on consumers. But this effort to shift the so-called Overton Window has been only partially successful. Contrary to the claims of consensus,^[42] a litany of voices from across the political spectrum have loudly opposed this new political philosophy of antitrust law.^[43]

Finally, not even antitrust agencies themselves are on the same page, with many current and former enforcement officials strongly disapproving of the new directions taken in competition law.

Formers senior officials at the FTC, for instance, have complained about that agency’s disregard for the law and statutory and precedential limits on the agency’s authority. In her letter of resignation,^[44] former FTC Commissioner Christine Wilson argued that “knowledgeable career staff have been scorned and sidelined,” leading to an exodus from the agency and a historical fall in staff morale.^[45] Further, according to Wilson, a “gag order was imposed on staff that prevented them from engaging in consumer and business education.” Due process, the rule of law, and precedent had apparently taken a back seat to ideology at the FTC.^[46] As Wilson wrote in a Wall Street Journal op-ed:

In November 2022, the commission issued an antitrust enforcement policy statement asserting that the FTC could ignore decades of court rulings and condemn essentially any business conduct [that] can be labeled with a nefarious adjective — ‘coercive,’ ‘exploitative,’ ‘abusive,’ ‘restrictive.”^[47]

As indicated above, many of these half-baked “you know it when you see it”^[48] theories of harm have targeted digital markets. This, indeed, has been Lina Khan’s focus since even before^[49] she joined the FTC.^[50] This approach has also extended to other areas of the FTC’s work, as reflected in sweeping revisions to several of the agency’s internal rules and procedures. Such changes, often made without public input, have included the withdrawal of enforcement guidelines and policy statements.

For example, in their dissent to the FTC’s adoption of revised Section 18 rulemaking, then-Commissioners Wilson and Noah Joshua Phillips noted that the procedural changes threatened to “undermine the goals of participation and transparency” and “limit public input,” while facilitating a biased evidentiary record.^[51]

ICLE’s Daniel J. Gilman, a former veteran FTC attorney advisor, has expressed concerns that these changes undermine the agency’s credibility and effectiveness. Furthermore, such actions “did little to signal the commission’s new view of its authority,” apart from indicating that a more expansive approach was forthcoming.^[52]

In sum, this is hardly the sort of landscape that shouts “consensus.” Even those court decisions that vindicate enforcers’ theories of harm are nowhere near the sort of victories that would justify elevating them to the level of universal principles that require no evidence of harm and admit no exemptions (see, e.g., the DMA).

III. There Is No Consensus that a Special Approach to Digital Markets Is Needed

Despite the EU’s premature vociferations claiming victory for writing the new  rulebook^[53] for digital competition regulation, the reception of that rulebook by the rest of the world has thus far been only lukewarm. To date, only a handful of countries^[54] have adopted sector-specific digital competition rules. In addition to the EU itself, Germany, Japan, and the United Kingdom have adopted regulatory regimes for digital markets that bear some resemblance to the DMA. Granted, other countries are considering the adoption of such rules (most notably, Brazil, Turkey, South Korea, South Africa, India, and Australia), but whether these will ultimately become law remains anyone’s guess.

In addition, the number of countries that have adopted ex-ante rules pales in comparison to those that have not. The United States, most notably, has rejected the path set out by the EU, as is evident from the slow death of the congressional antitrust legislative package^[55] amid growing criticism.^[56] Moreover, as Dae-Sik Hong and Daniel Sokol have pointed out:

The United States rejected such a legislative effort and its proponents have come under significant attack by academics and Congress. Likewise, most American courts have rejected this novel approach, and antitrust authorities that have brought lawsuits under such non-traditional legal theories have lost virtually every case, especially when seeking to block corporate mergers.^[57]

Other countries’ commitment to joining the alleged^[58] “global regulatory trend” are also teetering. For example, it was recently reported that India could scrap proposed legislation to regulate digital platforms,^[59] amid fierce backlash from lawyers. The South Korean government  earlier backtracked^[60] on its plans to pass the Platform Competition Promotion Act (“PCPA”), which was likewise inspired by the DMA. The South Korean government is instead contemplating a more modest—albeit still questionable—reform of its Fair Trade Act.^[61] The Philippines competition authority also recently ruled out a DMA-style bill.^[62]

Indeed, it is becoming more apparent that digital competition rules might be little more than a peculiarity of EU industrial policy. As I have pointed out previously,^[63] prior to the DMA’s adoption, many leading European politicians touted the law’s text as a protectionist industrial-policy tool that would hinder U.S. firms to the benefit of European rivals.^[64] French President Emmanuel Macron summarized it well when he said:

If we want technological sovereignty, we’ll have to adapt our competition law, which has perhaps been too much focused solely on the consumer and not enough on defending European champions.^[65]

Insofar as these goals are unique to a particular time and place (i.e., the EU in the 2020s), it is reasonable to assume they will not necessarily be shared by everyone. Sure enough, they are not. Indeed, some countries may be more interested in attracting digital platforms than in regulating,^[66] “disciplining,”^[67] or punishing them.^[68] Echoing the argument that “one size does not fit all” when it comes to digital competition regulation,^[69] Hong and Sokol note that among the reasons that ex-ante digital competition rules are inappropriate for South Korea is the marked differences between that nation’s economic, legal and regulatory context and that of the the EU:

Europe chose to regulate heavily for protectionist reasons. It lacks the tech infrastructure, innovative companies, and unicorns that are present in other vibrant economies like Korea. […] While Korea has approximately three times more unicorns than Japan, despite having a smaller gross domestic product, the adoption of a DMA-like approach may hurt Korea’s innovation advantage.^[70]

Similarly, Samir Ghandi argues that the DMA’s “one size fits all” approach would not work “for a dynamic Indian market with its own vibrant tech ecosystem.”^[71]

Other, less technologically intense countries like South Africa might have a still different set of priorities, such as attracting foreign direct investment to drive growth and the development of essential infrastructure. As I wrote in a piece with Geoffrey Manne:

Developing countries like South Africa should be especially wary of importing untested competition rules that impose government-mandated designs on the business models and user interfaces of innovative companies. It’s not trite to say that South Africa’s market is not the same as the EU’s. The consequences of unsound competition policy here may be to stymie foreign investment and domestic innovation exactly where they are needed most. […] This is a far cry from the untested, preemptive constraints contemplated by the [South Africa Competition Commission].^[72]

The point is that countries’ needs are as varied as those countries themselves. This does not preclude the possibility of common rules and standards; after all, most of the world’s competition-law systems have converged around some version of the consumer welfare standard.^[73] But one explanation for this commonality can be found in how the consumer welfare standard fares when compared to the alternatives:

The objective nature of the choice and interpretation of legal antitrust standards exists on a spectrum, and the [consumer welfare standard] conceptual congruence, measurability, and its connection to aspects that are almost universally considered to be relevant parameters of competition (price, innovation, quality) brings it closer to objectivity and further away from subjectivity.^[74]

Conversely, once it is understood that the DMA represents an attempt to pass off a sui generis, subjective policy choice as a universal regulatory paradigm, the case for harmonization quickly withers away. Clearly, not everyone is on board with trading economic performance for a set of questionable political goals.^[75] In this sense, one frequent criticism of ex-ante competition rules is that they ignore—or, at the very least significantly downplay—the effects on consumer welfare and innovation (the traditional bastions of competition policy). Instead of focusing on protecting competition to the benefit of consumers, digital competition rules commit the cardinal antitrust sin of protecting competitors. As former FTC Commissioner Maureen Ohlhausen has put it:

Some recent legislative and regulatory proposals appear to be in tension with this basic premise. Rather than focusing on protection of competition itself, they appear to impose requirements on some companies designed specifically to facilitate their competitors, including those competitors that may have fallen behind precisely because they had not made the same investments in technology, innovation or product offerings. For example, the Digital Markets Act (DMA) would force a ‘gatekeeper’ company to provide business users of its service, as well as those who provide complementary services, access to and interoperability with the same operating system, hardware, or software features that are available to or used by the gatekeeper. While this would restrain gatekeepers and presumably facilitate the interests of the gatekeeper’s rivals, it is not clear how this would protect consumers, as opposed to competitors.^[76]

This, of course, is only surprising if one falls for the story that digital competition rules—and the DMA, in particular—were ever intended to protect competition or the welfare of consumers. As I argued in a paper with Geoffrey Manne and Dirk Auer,^[77] their goal is instead to redistribute rents, protect competitors, and level down gatekeepers, even if it comes at the expense of consumers.^[78] There is no better example of this than the DMA, whose preamble explicitly disavows consumer welfare and economic efficiency as irrelevant under the new rules.

As commentators around the world have pointed out, this approach is likely to stymie dynamism in digital markets and harm consumers. As noted above, Hong and Sokol argue against introducing ex-ante digital competition regulations in South Korea, contending that such rules would stifle innovation, decrease investment, hurt startups and consumers, and jeopardize South Korea’s status as a regional leader in tech innovation.^[79] Carmelo Cennamo and Juan Santaló further argue that the DMA could produce a host of other harmful unintended consequences.^[80] For example, undermining gatekeepers’ ability to control access to their platforms could ultimately lead to lower levels of innovation. Obligations like data-sharing could reduce gatekeepers’ incentives to accumulate and process data, thereby diluting the competitive benefits and product improvements that result from such data collection.

Similarly, self-preferencing can drive value creation,^[81] and is often itself an expression of competition on the merits.^[82] There are several procompetitive reasons why firms may choose to give preferential treatment to their own downstream (or upstream) products or services, such as tighter quality control, better assurance of delivery (including better delivery tracking), lower transaction costs, and greater product differentiation. Thus, as Cennamo and Santaló have argued:

Prohibiting gatekeepers from giving their own products or services a better ranking or position on their platform than those of third parties, as the DMA does, might fail to take into account that those platform owners may also introduce their own products or services in order to steer innovation efforts towards certain areas or differentiate their ecosystem in strategic ways. This can lead to increased value for customers and overall innovation within a platform’s ecosystem.^[83]

Where gatekeepers face strong competition from other gatekeepers, the benefits of “cross-platform competition may outweigh the potential harm of reduced within-platform competition.”^[84]  This is applicable beyond the context of self-preferencing to closed and semi-closed platforms more generally. Indeed, as the court recognized in the Epic Games case, Apple’s ability to filter out harmful content on its iOS platform was an important differentiator with respect to Google’s relatively more open Android OS.[85]

While much of this discussion is rooted in theory and speculation, the DMA’s practical consequences are beginning to become apparent, as seen in delayed roll-outs of innovative products in the EU and a generally degraded consumer experience. In June 2024, for example, Apple announced it would delay the launch of three features in the EU (phone mirroring, SharePlay screen-sharing enhancements, and Apple Intelligence), citing security concerns the stem from the DMA’s interoperability requirements.^[86]

Meta similarly announced it would not launch its multimodal-AI models or products in the EU due to the “unpredictable regulatory environment.”^[87] While this may not refer specifically to the DMA (indeed, it could be referring to the EU’s GDPR or the AI Act), Meta had in 2023 already warned that “the DMA has caused, and may in the future cause, us to incur significant compliance costs and make changes to our products or business practices.”^[88] Among the obstacles Meta cited^[89] were the DMA’s large penalties (up to 20% of a company’s annual worldwide turnover in the case of recidivism, for which the bar is set extremely low) and its restrictions and requirements related to the combination of data across services, mergers and acquisitions, and product design.

Almost immediately following the DMA’s full entry into force in March 2024, Google removed useful features from its search results page, such as Google Flights and Google Maps.^[90] To avoid infringing the self-preferencing prohibition of the DMA, users in the EU are no longer allowed to access Google Maps by clicking on the thumbnail that pops up on Google search.^[91] Google explained these changes as follows:

We’ve always been focused on improving Google Search to help people quickly and easily find what they’re looking for. .. Rules that roll back some of these advances represent a fundamental shift in competition policy. We encourage other countries contemplating such rules to consider the potential adverse consequences — including those for the small businesses that don’t have a voice in the regulatory process.^[92]

This echoes arguments that Google’s parent Alphabet had made earlier. In its 2023 Form 10-K filing to the U.S. Securities and Exchange Commission (“SEC”), for example, the company lamented the uncertain regulatory landscape and contended that:

Our compliance with these laws and regulations may be onerous and could, individually or in the aggregate, increase our cost of doing business, make our products and services less useful, limit our ability to pursue certain business models, cause us to change our business practices, affect our competitive position relative to our peers, and/or otherwise harm our business, reputation, financial condition, and operating results. (emphasis added)^[93]

As Dirk Auer has pointed out,^[94] “not only is this inconvenient for consumers, but it has important ramifications for business users: Early estimates suggest that clicks from Google ads to hotel websites decreased by 17.6%^[95] as a result of the DMA.” Microsoft, another DMA gatekeeper, announced in November 2023 that it would delay the launch of some features in the EU—including Copilot, a generative AI chatbot. ^[96]

Some consumers and policymakers may be willing to accept these tradeoffs in pursuit of equity, fairness, contestability, “reining in” tech giants, or some other goal.^[97] But others, reasonably, may not. Thus, commentators from both within and outside the EU have increasingly questioned the need for rules that mechanically apply pre-set default solutions to the complex tradeoffs that have typically characterized competition-law analysis. This is of particular concern in dynamic markets driven by innovation, where uncertainty is endemic and where, except in the most egregious of cases,^[98] even the wisest enforcers can’t know a priori whether or not given conduct is procompetitive.^[99] Against this backdrop, tales of a supposed consensus in support of a special set of competition rules for digital platforms are rooted more in fantasy than in reality.

IV. There Is No Consensus About What the Correct Approach to Digital Competition Rules Should Be

Even in those jurisdictions that have taken steps to adopt “sector specific” competition rules in digital markets, there is no consensus about how these rules should be structured. To be sure, as I wrote in a recent paper with Geoffrey Manne and Dirk Auer,^[100] there are important thematic commonalities across so-called digital competition regulations. But on a legal and formal level, these approaches are vastly heterogeneous.

Digital competition rules find themselves in a “difficult epistemological situation,”^[101] where it is unclear whether they constitute competition law, sector-specific regulation (although digital markets are not sufficiently homogeneous to be considered a “sector”),^[102] or something else. Some have referred to them as the “lost child of competition law.”^[103]

Thus, some digital competition rules are an extension of competition-law frameworks and are sometimes even formally embedded into existing competition laws. In principle, where this is the case, it means that the standard goals and rationales of competition law still apply.

Germany, for instance, has amended its Competition Act by emphasizing the need to “intervene at an early stage in cases where competition is threatened by certain large digital companies.”^[104] The new rules enable the Bundeskartellamt to prohibit certain categories of conduct and impose remedies following a structural inquiry independent of any actual or potential abuse. Unlike the DMA, the Competition Act’s Article 19a permits targeted companies to justify their conduct. Contrary to the norms of competition law, however, the burden of proof rests with the defendant.

With its draft amendments to Law 4054 (Turkey’s Competition Act),^[105] Turkey has followed a similar path to Germany, although some of the new provisions go significantly further than even the DMA, partly due to their open-ended nature. For instance, the Turkish draft amendment would appear to prohibit all forms of tying and bundling, as well as potentially all exclusivity agreements. It also isn’t clear whether the prohibitions apply to all conduct by the designated digital platforms, or only to the “core platform services.”^[106] It also doesn’t contemplate any avenues for a defense—not even the narrow public-health and security exemptions laid down in the DMA.

South Korea recently scrapped plans for the PCPA—legislation inspired by the DMA.^[107] The KFTC and the government of recently impeached and indicted President Yoon Suk Yeol[108] instead announced support for amendments to the existing Fair Trade Act.^[109] Under the new rules, in cases where designated digital platforms are accused of self-preferencing, tying, or imposing most-favored nation (“MFN”) clauses or restrictions on multi-homing, the amendments would raise fines, reverse the burden of proof, and allow interim orders, including cease and desists, to be issued immediately. It also appears—although it is not certain—that the new rules would give targeted companies some leeway to mount a defense, such as by showing pro-competitive efficiencies.^[110]

There are other proposed and enacted digital competition rules that are at least nominally, competition-based, although their approaches differ. The UK’s Digital Competition and Consumers Bill (“DMCC”) allows the Competition and Markets Authority’s (“CMA”) newly created Digital Markets Unit (“DMU”) to impose “bespoke” conduct requirements on companies with “strategic market status.” This approach contrasts with the DMA, which contains (allegedly) self-executing blanket prohibitions by which all gatekeepers must abide.^[111] By contrast, under the DMCC, the DMU determines how each designated firm must conduct itself in order to achieve the law’s stated objectives of “fair dealing,” “open choices,” and “trust and transparency.” These conduct requirements must be chosen from a list of “permitted types” (e.g., prohibiting self-preferencing, or requiring choice screens).

S. 29 of the DMCC provides for a “countervailing benefits exception” to conduct requirements. But apart from the fact that the exemption sets a high bar to clear (the behavior must be “indispensable”), it also only applies once an investigation into breach of a conduct requirement is underway. As I have argued elsewhere, it is questionable how useful this defense will prove to be in practice.^[112]

India appears to be taking a middle path between the DMCC and the DMA, wherein certain firms would be designated as “systemically significant enterprises” and subject to six obligations and prohibitions, albeit with more space for customization by the enforcer. The Indian Draft Digital Competition Bill^[113] (“DDCB”) supplements the Indian Competition Act (“ICA”), but pursues different goals. The ICA’s stated goals^[114] are the protection of the interests of consumers and free trade, while the DDCB (like the DMA) pursues fairness and contestability.

Interestingly, however, the report^[115] that accompanied the DDCB’s publication stated that the bill’s goals were aligned with those of the ICA, as both supposedly aimed to promote “fairness and contestability.” This is surprising, since:

The Report of the High-Powered Expert Committee on Competition Law and Policy^[116] (“Raghavan Committee Report”), which served as the basis for the [2002] ICA, modernized Indian competition law by moving it away from the structure-based paradigm of the earlier Anti-Monopolies and Restrictive Trade Practices Act of 1969 and toward an economic-effects-based analysis. The Raghavan Committee Report was unequivocal in its support of consumer welfare as the system’s ultimate goal. Moreover, the report advised against a plurality of goals, including, specifically, “bureaucratic perceptions” of equity and fairness, which, it argued, were mutually contradictory, difficult to quantify, and potentially opposed to the sustenance of free, unfettered competition ^[117].

The difficulty of squaring the goals and rationale of digital competition rules and pre-existing competition laws shouldn’t come as a surprise, as the two are rooted in different normative visions of the role of competition and competition rules in society. At worst, these two visions are incompatible. Indeed, for digital competition rules, competition is no longer about efficiency or consumer welfare; it is instead about “fairness” and equity.

This recasts “competition” as a participatory process, where participation is more-or-less guaranteed, rather than a rivalrous one, where the overarching theme is precisely that it is not.^[118] Beyond the epistemological implications of this discrepancy, the contrived reinterpretation of competition laws to fit the new regulatory meta-narrative adds another spin on the complex relationship between competition law and digital competition regulation, with firms already operates under the looming threat of double and possibly even triple jeopardy.^[119] It suggests, in sum, that the goals of digital competition regulate may intermingle with and ultimately come to influence competition law. As I have written in the context of South Korea’s proposed reform of the Fair Trade Act:

Introducing these exogenous principles for oversight of digital platforms into the existing antitrust statute might subvert the prevailing system’s logic by emphasizing bigness, structural presumptions, and sui generis rules in ways that could eventually spread to other sectors and undermine the integrity, consistency, and predictability of the [antitrust] law.^[120]

Meanwhile, in the United States, several bills have been put forward that are formally separate from existing antitrust law, but cover some of the same conduct as would typically be addressed under U.S. antitrust law—albeit with seemingly different goals and standards.^[121] While the U.S. tech bills largely fail to describe their underlying goals, the bills’ titles and statements made by their sponsors suggest a set of overlapping concerns, such as preventing “material harm to competition” (which superficially sounds like an antitrust objective but, as the American Bar Association’s Antitrust Section has pointed out, isn’t);^[122] reducing “gatekeeper power in the app economy”; and “increasing choice, improving quality, and reducing costs for consumers.”^[123] But the measures also pursue other goals that are less obviously connected to competition, such as creating opportunities for small businesses and entrepreneurs, achieving a level playing field, and ensuring “fair” prices.^[124]

Brazil’s PL 2768,^[125] which has some of the lowest quantitative thresholds for a company to be considered a “gatekeeper” (roughly €11.6 million), pursues an expansive grab bag of social and economic goals, including freedom of initiative; free competition; consumer protection; reduced regional and social inequality; combating the abuse of economic power; widening social participation in matters of public interest; access to information, knowledge, and culture; and fostering innovation and mass access to new technologies and access models. Like the DMCC, the obligations and obligations would be tailored to each company. The provisions are broadly phrased, however, and some appear to be open to expansive interpretations. For example, Art.10(IV) prohibits gatekeepers from refusing access to business users—seemingly tout court (although Art.11 then requires enforcers to act with proportionality when establishing obligations).

Japan, whose Smartphone Act is part of an overarching policy shift “towards a new form of capitalism,”^[126] covers only four core platform services. By comparison, other digital competition rules typically cover around 10, replicating the DMA’s scope. Further, the Smartphone Act’s dos and don’ts would only apply when consumers access products or services on their phones (e.g., Google is only prohibited from engaging in self-preferencing on smartphones,^[127] but not on laptops or PCs). The Smartphone Act also allows greater scope for privacy and security exemptions. Whereas the DMA only allows for such exemptions in the case of interoperability and sideloading (the Smartphone Act also does not mandate sideloading), it appears that privacy, safety, and user protection constitute valid justifications for most types of conduct covered by the Smartphone Act.^[128]

In South Africa, the South Africa Competition Commission (“SACC”) published its Online Intermediation Platforms Market Inquiry Final Report,^[129] which calls for remedial actions against certain popular intermediation platforms. These are largely the usual GAMMA suspects, with Amazon was identified as a potential target of regulation despite not being present in South Africa at the time. These remedies would presumably be imposed by the SACC within the framework of the South African Competition Act. Unlike other digital ex-ante regulations, the SACC explicitly admits that its proposed remedies aim to redistribute wealth from the targeted digital companies to certain social groups—namely, South African companies, historically disadvantaged peoples (“HDPs”), and small and medium-sized enterprises (“SMEs”), particularly those owned by HDPs. For instance, the SACC’s report recommends requiring Google to add a South African flag identifier and South African platform filter to “aid consumers to easily identify and support local platforms in competition to global ones,” and to directly pay competitors R150 million (roughly $8 million) in “training, product support and other measures for SME and black-owned online firms to offset the competitive disadvantage faced on Google Search.”

A similar thread runs through the entire SACC report. It recasts “undistorted competition” as precisely the distortion of competition in favor of struggling small, local, and/or national competitors, and redefines competition regulation via a haphazard blend of industrial policy and wealth redistribution (e.g., Booking.com would be forced to invest in promoting SMEs and UberEats would be banned from offering deals to preferred customers that “decimate” SMEs).

There are two ways to interpret this. The first is that the South African model goes significantly further than the DMA—which, once again, cuts against the consensus narrative. The second is that the South African model is simply saying the quiet part out loud: Digital competition rules exist to redistribute rents, prop up select competitors (domestic SMEs, national champions, etc.), and “protect competition” by distorting it to hamstring incumbents.

But even if the latter is true, the fact that common themes exist across digital competition rules does not negate the many consequential differences in the design, enforcement, and structure of such rules. These discrepancies put the final nail in the coffin of the consensus narrative.

For example, will the DMCC’s “bespoke” interventions match those imposed under the DMA? Will they cover the same companies? In the few jurisdictions where privacy, security, and consumer welfare considerations still matter as defenses, will they be evaluated in the same way? Are digital competition rules part of competition law (as in Germany), or are they definitely not (as in the EU)?

These question have momentous conceptual but also practical implications: What will happen when the stated (although arguably pretextual) goals of fairness, equality, participation, and protection of SMEs (to name a few) inevitably clash with traditional notions of competition that use consumer welfare and economic efficiency as the relevant yardsticks to demarcate pro and anticompetitive conduct? What will the “consensual” solution be to ranking imponderable values across countries with vastly different strategic and policy priorities?

The truth is that nobody knows, but convergence, like consensus, appears unlikely under these circumstances.

V. Conclusion

The objective of this issue brief is not to dismantle the antitrust consensus about the need for special, sector-specific competition rules in digital markets. Rather, the goal is to argue that no such consensus exists and, by extension, that the meta-narrative on which the antitrust counter-revolution rests^[130] is grossly exaggerated, at best, and fanciful at worst. To be sure, I do not make the argument that there is a consensus in the opposite direction, either.

So why does it persist? Grand narratives will always be compelling because they tap into popular themes and prejudices to offer simple stories about (and solutions to) complex phenomena. As narratives go, the new antitrust meta-narrative is particularly difficult to dispel, because it rides a generational populist anti-capitalist wave built on a shaky but attractive ideational foundation: “techno-feudalism,” “surveillance capitalism,” “neoliberalism,” etc. These are the bogeymen of our time, and “Big Tech” embodies them all.

Here, narrative appeal trumps facts and analysis. It doesn’t matter that those brandishing the term cannot even properly define what “neoliberalism” is. It doesn’t matter that, despite the supposed hegemony of neoliberalism, government spending as a percentage of GDP has exploded.^[131] It doesn’t matter that markets are shot-through with laws and regulation.^[132] It doesn’t matter that “killer acquisitions” constitute a minuscule portion of tech acquisitions.^[133]

Meta-narratives are big stories that rely on big, sweeping claims. What they offer is a story of heroes and villains;^[134] good vs. evil; the righteous vs the wicked. And in such a story, there is no place for nuance and no need for evidence.

Ultimately, proponents of the DMA and similar regulatory regimes hope that their fantasy will become a self-fulfilling one. If they can convince the uncommitted that everyone else is already following suit, then no one will want to be left behind. But countries contemplating reform on this basis should understand that the antitrust consensus they are banking on is an illusion. By the time this illusion dissipates, it may be too late.

Grand narratives become harder to dispel with time, as they take root in the popular imagination. They are even harder to dispel once they are codified in law.

^[1] Jean-François Lyotard, The Postmodern Explained to Children: Correspondence, 1982–1985 (1992), (introducing the term “grand narrative” to describe totalizing narratives of modernity).

^[2] Thomas A. Lambert & Tate Cooper, Neo-Brandeisianism’s Democracy Paradox, 49 J. Corp. L. (2023).

^[3] Lazar Radic, Geoffrey A. Manne, & Dirk Auer, Regulate for What? A Closer Look at the Rationale and Goals of Digital Competition Regulations, 22(1) Berkeley Bus. L.J. (2024); Lazar Radic, Political Philosophy, Competition, and Competition Law: The Road to and from Neoliberalism, Part 3, Truth Mark. (Oct. 10, 2022), https://truthonthemarket.com/2022/10/10/political-philosophy-competition-and-competition-law-the-road-to-and-from-neoliberalism-part-3; Larry Kramer, Beyond Neoliberalism: Rethinking Political Economy, William & Flora Hewlett Found. (Apr. 26, 2018), https://hewlett.org/library/beyond-neoliberalism-rethinking-political-economy.

^[4] See, e.g., Aspectos Econômicos e Concorrenciais e Recomendações Para Aprimoramentos Regulatórios No Brasil, Ministerio da Fazenda (2024), available at https://www.gov.br/fazenda/pt-br/central-de-conteudo/publicacoes/apresentacoes/2024/outubro/arquivo/plataformas-digitais-concorrencia_10102024-pptx-1.pdf (reiterating many of the standard tropes of digital competition regulation—e.g., the uniqueness of digital platforms, the insufficiency of antitrust law, the pervasiveness of anticompetitive conduct, an alleged link between a country’s economic downturn and the economic dominance of a few tech companies, and asserting a supposed trend toward ex-ante regulation).

^[5] Robert Armstrong & Ethan Wu, What Big Tech Antitrust Gets Wrong: An Interview with Herbert Hovenkamp, Financ. Times (Jan. 19, 2024), https://www.ft.com/content/4eec8bc3-c892-4704-ae66-a4432c6d4fd7.

^[6] Press Release, Commission Designates Six Gatekeepers Under the Digital Markets Act, Eur. Comm. (Sep. 6, 2023), https://digital-markets-act.ec.europa.eu/commission-designates-six-gatekeepers-under-digital-markets-act-2023-09-06_en.

^[7] Amanda Lotz, Big Tech Isn’t a Monolith. It’s 5 Companies, All in Different Businesses, Houst. Chron. (Mar. 26, 2018), https://www.houstonchronicle.com/techburger/article/Big-Tech-isn-t-a-monolith-It-s-5-companies-12781761.php; Walid Chaiehloudj, On “Big Tech and the Digital Economy”: Interview with Professor Nicolas Petit, Compet. Forum (Jan. 11, 2021), https://competition-forum.com/on-big-tech-and-the-digital-economy-interview-with-professor-nicolas-petit.

^[8] Isobel Asher Hamilton, Tim Cook Says He’s Tired of Big Tech Being Painted as a ‘Monolithic’ Force That Needs Tearing Apart, Bus. Insid. (May 7, 2019), https://www.businessinsider.com/apple-ceo-tim-cook-tired-of-big-tech-being-viewed-as-monolithic-2019-5.

^[9] Lotz, supra note 7; see also Lazar Radic, Gatekeeping, the DMA, and the Future of Competition Regulation, Truth Mark. (Nov. 8, 2023), https://truthonthemarket.com/2023/11/08/gatekeeping-the-dma-and-the-future-of-competition-regulation.

^[10] Dirk Auer & Lazar Radic, The Legacy of Neo-Brandeisianism: History or Footnote?, Netw. Law Rev. (Jul. 9, 2024) https://www.networklawreview.org/auer-radic-brandeisianism.

^[11] Radic, supra note 9.

^[12] Adam Hayes, 20 Industries Threatened by Tech Disruption, Investopedia (Jan. 23, 2022), https://www.investopedia.com/articles/investing/020615/20-industries-threatened-tech-disruption.asp.

^[13] Alex Payne, What Is and Is Not a Technology Company, Arch. Today (May 7, 2021), https://archive.vn/XUh8C.

^[14] The 100 Largest Companies in the World by Market Capitalization in 2023, Statista (Jun. 2024), https://www.statista.com/statistics/263264/top-companies-in-the-world-by-market-capitalization.

^[15] Corbin K. Barthold, Failing Upward: For Lina Khan’s Federal Trade Commission, Dysfunction Is No Bar to Ambition, City J. (Aug. 17, 2023), https://www.city-journal.org/article/ftc-chair-lina-khan-fails-upward.

^[16] Ephrat Livni & Michael J. de la Merced, Microsoft’s Activision Deal Tests a New Global Alignment on Antitrust, N.Y. Times (Feb. 6, 2023), https://www.nytimes.com/2023/02/04/business/dealbook/microsofts-activision-deal.html.

^[17] Lina Khan & Jonathan Kanter, Joint Comment to Canada Competition Authority on the Ministry’s Public Consultation Paper on the Future of Competition Policy in Canada, Fed. Trade Comm. & U.S. Dep. Justice (Mar. 31, 2023), available at https://www.ftc.gov/system/files/ftc_gov/pdf/USFTC-USDOJ-joint-comment-to-Canada-Consultation-Paper.pdf.

^[18] Gus Hurwitz & Geoffrey Manne, Antitrust Regulation by Intimidation, Wall Str. J. (Jul. 24, 2023), https://www.wsj.com/articles/antitrust-regulation-by-intimidation-khan-kanter-case-law-courts-merger-27f610d9.

^[19] Alden Abbott, Chair Khan’s Latest Flawed Perspective on Mergers Ignores Empirics and Sound Economics, Truth Mark. (May 16, 2022), https://truthonthemarket.com/2022/05/16/chair-khans-latest-flawed-perspective-on-mergers-ignores-empirics-and-sound-economics.

^[20] Venture Capital Community Statement of Support for Amicus Brief Filed in F.T.C. v. Microsoft Corp. & Activision Blizzard, Inc., (Dec. 2024), available at https://blogs.microsoft.com/wp-content/uploads/prod/sites/5/2023/12/Venture-Capital-Statement-of-Support.pdf.

^[21] Letter from Members of Congress to Lina Khan, Chair, Rebecca Slaughter, Commissioner, & Alvaro Bedoya, Commissioner, Fed. Trade Comm. (Jul. 17, 2023), available at https://armstrong.house.gov/sites/evo-subsites/armstrong.house.gov/files/evo-media-document/house-letter-to-ftc-re-msft-abk-merger-final-7.17.2023_0.pdf.

^[22] Sam Bowman & Geoffrey Manne, Vertical Integration: Economies of Scope, Int’l Ctr. L. Econ. (Jul. 2020), available at https://laweconcenter.org/wp-content/uploads/2020/07/ICLE-tldr-Vertical-integration_-economies-of-scope-FINAL.pdf; Geoffrey A. Manne, Dirk Auer, Brian Albrecht, Eric Fruits, Daniel J. Gilman, & Lazar Radic, Comments of the International Center for Law and Economics on the FTC & DOJ Draft Merger Guidelines, Int’l Ctr. L. Econ. (Sep. 18, 2023), https://laweconcenter.org/resources/comments-of-the-international-center-for-law-and-economics-on-the-ftc-doj-draft-merger-guidelines.

^[23] Bowman & Manne, id.

^[24] Manne et al., supra note 22.

^[25] Daniel J. Gilman, The 2023 Merger Guidelines: What Are They Good For?, Int’l Ctr. L. Econ. (Jan. 29, 2024), https://laweconcenter.org/resources/the-2023-merger-guidelines-what-are-they-good-for.

^[26] Manne et al., supra note 22.

^[27] Richard A. Epstein, The DOJ and FTC’s Misguided Attack on Mergers, 49(2) J. Corp L. 276 (Jan. 23, 2024), available at https://jcl.law.uiowa.edu/sites/jcl.law.uiowa.edu/files/2024-01/Epstein_Final.pdf.

^[28] Luke M. Froeb, D. Daniel Sokol, & Liad Wagman, Cost-Benefit Analysis Without the Benefits or the Analysis: How Not to Draft Merger Guidelines, 97 S. Cal. L. Rev. PS1 (Aug. 10, 2023), available at https://southerncalifornialawreview.com/wp-content/uploads/2023/10/Froeb-Sokol-Wagman_Final.pdf.

^[29] Id.; see also Hurwitz & Manne, supra note 18.

^[30] Susan Dudley & Mary Sullivan, Recent Antitrust and Regulatory Changes Both Unravel the Consensus, Truth Mark. (Sep. 18, 2023), https://truthonthemarket.com/2023/09/18/antitrust-and-regulatory-changes-both-unravel-the-consensus.

^[31] Epstein, supra note 27.

^[32] Khan & Kanter, supra note 17.

^[33] Editorial Board, Another Lina Khan Theory Loses in Court, Wall Str. J. (May 15, 2024), https://www.wsj.com/articles/ftc-loses-lawsuit-welsh-carson-u-s-anesthesia-partners-kenneth-hoyt-lina-khan-9a83b653.

^[34] David Meyer, FTC Claims Facebook Withheld Information When Buying Instagram and WhatsApp, Fortune (Jun. 5, 2024), https://fortune.com/2024/06/05/ftc-facebook-withheld-information-instagram-whatsapp-meta-acquisition.

^[35] Barthold, supra note 15.

^[36] Jinyul Ju, Market Definition, Antitrust Error, and Digital Platforms in Korean Competition Law and Policy, 61(3) Pusan Natl. Univ. L. Rev. 225 (Dec. 11, 2020), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3684985.

^[37] Council Regulation 2022/1925, of 14 September 2022, on Contestable and Fair Markets in the Digital Sector (Digital Markets Act), Off. J. Eur. Union, https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32022R1925.

^[38] Epic Games Inc. v. Apple Inc., 4:20-cv-05640 (N.D. Cal. Sep. 10, 2021), https://www.courtlistener.com/docket/17442392/epic-games-inc-v-apple-inc.

^[39] For a more detailed account, see Andrei Hagiu, Proprietary vs. Open Two-Sided Platforms and Social Efficiency, Working Paper No. 06-12, Harvard Business School (May 2006), https://www.hbs.edu/ris/Publication%20Files/07-095.pdf; see also Alden Abbott et al., ICLE Brief for 9th Circuit in Epic Games v. Apple, Epic Games, Inc. v. Apple Inc., Int’l Ctr. L. Econ. (Apr. 1, 2022), https://laweconcenter.org/resources/icle-brief-for-9th-circuit-for-epic-games-v-apple.

^[40] FTC v. Amazon.com Inc., No. 2:23-cv-01495 (W.D. Wash. Nov. 5, 2024), https://www.ftc.gov/legal-library/browse/cases-proceedings/1910129-1910130-amazoncom-inc-amazon-ecommerce.

^[41] Lazar Radic & Geoffrey A. Manne, The FTC’s Gambit Against Amazon: Navigating a Multiverse of Blowback and Consumer Harm, Truth Mark. (Aug. 3, 2023), https://truthonthemarket.com/2023/08/03/the-ftcs-gambit-against-amazon-navigating-a-multiverse-of-blowback-and-consumer-harm.

^[42] Cristina Caffarra, “Consumer Welfare Is Dead”: What Do We Do Instead?—A Perspective from Europe, ProMarket (Apr. 27, 2023), https://www.promarket.org/2023/04/27/consumer-welfare-is-dead-what-do-we-do-instead-a-perspective-from-europe.

^[43] Herbert J. Hovenkamp, Is Antitrust’s Consumer Welfare Principle Imperiled?, 45(1) J. Corp. L. 65, available at https://jcl.law.uiowa.edu/sites/jcl.law.uiowa.edu/files/2021-08/Hovenkamp_Final_Web.pdf; Nicolas Petit & Lazar Radic, The Necessity of a Consumer Welfare Standard in Antitrust Analysis, ProMarket (Dec. 18, 2023) https://www.promarket.org/2023/12/18/the-necessity-of-a-consumer-welfare-standard-in-antitrust-analysis; Christine S. Wilson & Adam S. Cella, Deconstructing the Worldview of the Neo-Brandeisians Through Marxism and Critical Legal Studies, 29 Geo. Mason L. Rev. 961 (2022), https://lawreview.gmu.edu/print__issues/deconstructing-the-worldview-of-the-neo-brandeisians-through-marxism-and-critical-legal-studies; Elyse Dorsey, Geoffrey A. Manne, Jan M. Rybnicek, Kristian Stout, & Joshua D. Wright, Consumer Welfare & the Rule of Law: The Case Against the New Populist Antitrust Movement, 47 Pepp. L. Rev. 861 (2020), https://digitalcommons.pepperdine.edu/plr/vol47/iss4/1.

^[44] Christine S. Wilson, Letter of Resignation to President Joseph R. Biden Jr., Fed. Trade Comm. (Mar. 2, 2023), available at https://www.ftc.gov/system/files/ftc_gov/pdf/p180200wilsonresignationletter.pdf.

^[45] Editorial Board, The Many Abuses of Lina Khan’s FTC, Wall Str. J. (Feb. 14, 2023), https://www.wsj.com/articles/christine-wilson-resigns-federal-trade-commission-lina-khan-ftc-87328998; Shane Tews & Mark Jamison, What’s Going On at the Federal Trade Commission? (with Bilal Sayyed—Part I), Am. Enterp. Inst. (Oct. 5, 2021), https://www.aei.org/podcast/whats-going-on-at-the-federal-trade-commission-with-bilal-sayyed-part-i.

^[46] Mark A. Jamison, Why Christine Wilson’s Resignation from the FTC Matters, The Hill (Mar. 29, 2023) https://thehill.com/opinion/finance/3920706-why-christine-wilsons-resignation-from-the-ftc-matters.

^[47] Christine Wilson, Why I’m Resigning as an FTC Commissioner, Wall Str. J. (Feb. 14, 2023), https://www.wsj.com/articles/why-im-resigning-from-the-ftc-commissioner-ftc-lina-khan-regulation-rule-violation-antitrust-339f115d.

^[48] Jamison, supra note 46.

^[49] Lina M. Khan, Amazon’s Antitrust Paradox, 126 Yale L.J. 564 (2017), https://www.yalelawjournal.org/note/amazons-antitrust-paradox.

^[50] Lina M. Khan, The Separation of Platforms and Commerce, 119 Colum. L. Rev. 973 (2019), https://columbialawreview.org/content/the-separation-of-platforms-and-commerce.

^[51] Christine S. Wilson & Noah Joshua Phillips, Dissenting Statement of Commissioners Christine S. Wilson and Noah Joshua Phillips Regarding the Commission Statement On the Adoption of Revised Section 18 Rulemaking Procedures, Fed. Trade Comm. (Jul. 9, 2021), available at https://www.ftc.gov/system/files/documents/public_statements/1591702/p210100_wilsonphillips_joint_statement_-_rules_of_practice.pdf.

^[52] Daniel J. Gilman, Abby Normal, a Flood of Ill-Considered Withdrawals, and the FTC’s Theatre of Listening, Truth Mark. (Oct. 5, 2023), https://truthonthemarket.com/2023/10/05/abby-normal-a-flood-of-ill-considered-withdrawals-and-the-ftcs-theatre-of-listening.

^[53] Zach Meyers, Can the EU Set a Global Rulebook for Big Tech?, Ctr. Eur. Reform (Jun. 1, 2021), https://www.cer.eu/publications/archive/bulletin-article/2021/can-eu-set-global-rulebook-big-tech.

^[54] Thomas Graf, Jackie Holland, Henry Mostyn, & Patrick Todd, Digital Markets Regulation Handbook, Cleary Gottlieb https://content.clearygottlieb.com/antitrust/digital-markets-regulation-handbook/index.html.

^[55] Sangyun Lee, Lessons from Korea’s Roller-Coaster Ride Toward Platform (Non)Regulation, Truth Mark. (Sep. 25, 2024), https://truthonthemarket.com/2024/09/25/lessons-from-koreas-roller-coaster-ride-toward-platform-nonregulation.

^[56] Lazar Radic & Geoffrey A. Manne, The ABA’s Antitrust Law Section Sounds the Alarm on Klobuchar-Grassley, Truth Mark. (May 12, 2022), https://truthonthemarket.com/2022/05/12/the-abas-antitrust-law-section-sounds-the-alarm-on-klobuchar-grassley.

^[57] Dae-sik Hong & D. Daniel Sokol, Korea Should Prioritize Innovation, Not Misguided Platform Regulation, The Korea Her. (May 12, 2024), https://www.koreaherald.com/view.php?ud=20240512050148.

^[58] Sangyun Lee, LinkedIn (Sep. 27, 2024, 00:35:22), https://www.linkedin.com/posts/sangyunl_indian-digital-competition-law-teeters-lawyers-activity-7245289899409448960-0rtV?utm_source=share&utm_medium=member_desktop.

^[59] Charles McConnell, Indian Digital Competition Law Teeters, Lawyers Call for Rethink, Glob. Compet. Rev. (Sep. 26, 2024) https://globalcompetitionreview.com/article/indian-digital-competition-law-teeters-lawyers-call-rethink.

^[60] Chosun Ilbo, ‘Monopoly Platform’ Regulation Law Falls Away… Fair Trade Commission Cancels Plan Due to Industry Opposition, Naver News (Sep. 9, 2024), https://n.news.naver.com/mnews/article/023/0003857596?sid=101.

^[61] Kang Shin-woo, Amendment of the Fair Trade Act to Regulate Large Platforms… ‘Google, Apple, Naver, Kakao’ to Have Jurisdiction, Naver News (Sep. 9, 2024) https://n.news.naver.com/mnews/article/018/0005832606?sid=101; see also Heo Ji-hye, Platform Law that Changes Direction… Concerns Increase over Standards for Proof of Competition Restriction, Pressman (Sep. 9, 2024), https://www.pressman.kr/news/articleView.html?idxno=84619. Under the revisions, platforms must directly prove that their actions do not harm competitors, and that they benefit consumers and positively impact the market. In other words, the reforms essentially reverse the burden of proof. Critics like Hong Dae-sik warn that stringent oversight could discourage businesses to pursue new initiatives due to a lack of confidence in their ability to meet criteria. “Ultimately, if companies are not confident in the reasons they present to the Fair Trade Commission when taking certain actions, they will not take the actions.”

^[62] Charles McConnell, Exclusive: Philippine Competition Watchdog Rules Out DMA-Style Bill, for Now, Glob. Compet. Rev. (Sep. 20, 2024) https://globalcompetitionreview.com/article/exclusive-philippine-competition-watchdog-rules-out-dma-style-bill-now.

^[63] Radic, supra note 9.

^[64] Mathieu Pollet, France to Prioritise Digital Regulation, Tech Sovereignty During EU Council Presidency, Euractiv (Dec. 14, 2021), https://www.euractiv.com/section/digital/news/france-to-prioritise-digital-regulation-tech-sovereignty-during-eu-council-presidency; Lazar Radic, Digital-Market Regulation: One Size Does Not Fit All, Truth Mark. (Apr. 17, 2023), https://truthonthemarket.com/2023/04/17/digital-market-regulation-one-size-does-not-fit-all.

^[65] Barbara Moens & Paola Tamma, Macron and Merkel Defy Brussels with Push for Industrial Champions, Politico (May 18, 2020), https://www.politico.eu/article/macron-and-merkel-defy-brussels-with-push-for-industrial-champions.

^[66] Oles Andriychuk, Do DMA Obligations for Gatekeepers Create Entitlements for Business Users?, 11 J. Antitrust Enforc. 123, 123-32 (Dec. 28, 2022), https://academic.oup.com/antitrust/article/11/1/123/6964483.

^[67] Geoffrey A. Manne, Dirk Auer, & Sam Bowman, Should ASEAN Antitrust Laws Emulate European Competition Policy?, 67 Singap. Econ. Rev. 1637 (Mar. 31, 2021), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3709730.

^[68] Radic, supra note 64.

^[69] Id.

^[70] Hong & Sokol, supra note 57.

^[71] McConnell, supra note 59.

^[72] Lazar Radic & Geoffrey A. Manne, South Africa’s Competition Proposal Takes Europe’s DMA Model to the Extreme, Truth Mark. (Aug. 15, 2023), https://truthonthemarket.com/2023/08/15/south-africas-competition-proposal-takes-europes-dma-model-to-the-extreme.

^[73] Christine S. Wilson, Welfare Standards Underlying Antitrust Enforcement: What You Measure Is What You Get, Fed. Trade Comm. (Feb. 15, 2019), available at https://www.ftc.gov/system/files/documents/public_statements/1455663/welfare_standard_speech_-_cmr-wilson.pdf; Svend Albæk, Consumer Welfare in EU Competition Policy, Eur. Comm. (2013), available at https://competition-policy.ec.europa.eu/system/files/2021-09/consumer_welfare_2013_en.pdf.

^[74] Petit & Radic, supra note 43.

^[75] Dirk Auer, The Broken Promises of Europe’s Digital Regulation, Truth Mark. (Mar. 12, 2024), https://truthonthemarket.com/2024/03/12/the-broken-promises-of-europes-digital-regulation.

^[76] John Taladay & Maureen Ohlhausen, Are Competition Officials Abandoning Competition Principles?, 13 J.E.C.L. & Pract. 463 (Jul. 5, 2022), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4042226.

^[77] Radic et al., supra note 3.

^[78] Radic, supra note 9.

^[79] Hong & Sokol, supra note 57.

^[80] Carmelo Cennamo & Juan Santaló, Potential Risks and Unintended Effects of the New EU Digital Markets Act, Esade EcPol (Feb. 2023), available at https://www.esade.edu/ecpol/wp-content/uploads/2023/02/AAFF_EcPol-OIGI_PaperSeries_04_Potentialrisks_ENG_v5.pdf.

^[81] Lazar Radic & Geoffrey A. Manne, Amazon Italy’s Efficiency Offense, Truth Mark. (Jan. 11, 2022), https://truthonthemarket.com/2022/01/11/amazon-italys-efficiency-offense.

^[82] Pablo Ibáñez Colomo, Self-Preferencing: Yet Another Epithet in Need of Limiting Principles, 43 World Compet. 417 (Jul. 17, 2020), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3654083.

^[83] Cennamo & Santaló, supra note 80.

^[84] Id.

[85] Epic Games, supra note 38.

^[86] Javier Espinoza & Michael Acton, Apple Delays European Launch of AI Features Because of EU Rules, Financ. Times (Jun. 21, 2024), https://www.ft.com/content/360751cb-7a22-48e0-9b00-6a30ff41dcfe.

^[87] Jess Weatherbed, Meta Won’t Release Its Multimodal Llama AI Model in the EU, The Verge (Jul. 18, 2024), https://www.theverge.com/2024/7/18/24201041/meta-multimodal-llama-ai-model-launch-eu-regulations .

^[88] Meta Platforms Inc. Annual Report (Form 10-K), U.S. Secur. Exch. Comm. (Dec. 31, 2023), https://www.sec.gov/Archives/edgar/data/1326801/000132680124000012/meta-20231231.htm.

^[89] Id.

^[90] Oliver Bethell, Complying with the Digital Markets Act, Google: The Keyword (Mar. 5, 2024), https://blog.google/around-the-globe/google-europe/complying-with-the-digital-markets-act; Adam Cohen, New Competition Rules Come with Trade-Offs, Google: The Keyword (Apr. 5, 2024) https://blog.google/around-the-globe/google-europe/new-competition-rules-come-with-trade-offs.

^[91] @laz_radic, X.com (May 3, 2024, 5:20 a.m.), https://x.com/laz_radic/status/1786325008345645321.

^[92] Cohen, supra note 90.

^[93] Alphabet Inc. Annual Report (Form 10-K), U.S. Secur. Exch. Comm. (Dec. 31, 2023), https://www.sec.gov/Archives/edgar/data/1652044/000165204424000022/goog-20231231.htm.

^[94] Auer, supra note 75.

^[95]  Mirai, LinkedIn (Feb. 8, 2024, 12:10:54), https://www.linkedin.com/feed/update/urn:li:activity:7161330551709138945.

^[96] Previewing Changes in Windows to Comply with the Digital Markets Act in the European Economic Area, Window Blogs (Nov. 16, 2023) https://blogs.windows.com/windows-insider/2023/11/16/previewing-changes-in-windows-to-comply-with-the-digital-markets-act-in-the-european-economic-area.

^[97] Cohen, supra note 90.

^[98] Mario Monti, Why and How? Why Should We Be Concerned with Cartels and Collusive Behaviour?, Eur. Comm. (Sept. 11, 2000), https://ec.europa.eu/commission/presscorner/detail/en/SPEECH_00_295.

^[99] Geoffrey A. Manne, Error Costs in Digital Markets, GAI Report on the Digital Economy 3 (Nov. 2020), available at https://gaidigitalreport.com/wp-content/uploads/2020/11/Manne-Error-Costs-in-Digital-Markets.pdf.

^[100] Radic et al., supra note 3.

^[101] Pierre Larouche & Alexandre De Streel, The European Digital Market: A Revolution Grounded on Traditions, 12 J.E.C.L. & Pract. 542 (2021), (arguing that the DMA’s conceptual nature is in a “difficult epistemological position”).

^[102] Radic, supra note 9.

^[103] Belle Beems, The DMA in the Broader Regulatory Landscape of the EU: An Institutional Perspective, 19 Eur. Competition J. 1–29 (Jan. 2023), https://www.tandfonline.com/doi/full/10.1080/17441056.2022.2129766.

^[104] Press Release, Amendment of the German Act Against Restraints of Competition, Bundeskartellamt (Jan. 19, 2021), https://www.bundeskartellamt.de/SharedDocs/Meldung/EN/Pressemitteilungen/2021/19_01_2021_GWB%20Novelle.html.

^[105] Bahadir Balki, Nabi Can Acar, Helin Yüksel, Mehmet Mikail Demir, Seda Eliri, & Erdem Aktekin, A New Age for Digital Markets in Turkey? The Draft Amendment to the Law No. 4054 on the Protection of Competition, Kluwer Competition Law Blog (Oct. 25, 2022), https://competitionlawblog.kluwercompetitionlaw.com/2022/10/25/a-new-age-for-digital-markets-in-turkey-the-draft-amendment-to-the-law-no-4054-on-the-protection-of-competition.

^[106] Henry Mostyn, Patrick Todd, & Goksu Kalayci, Turkiye, Cleary Gottlieb (Dec. 2023), https://content.clearygottlieb.com/antitrust/digital-markets-regulation-handbook/turkey/index.html.

^[107] Ilbo, supra note 60.

[108] Jean Mackenzie & Ruth Comerford, Impeached S Korean President Charged with Insurrection, BBC News (Jan. 26, 2025), https://www.bbc.com/news/articles/cr53r1d0jz4o.

^[109] Shin-woo, supra note 61.

^[110] Lee, supra note 58.

^[111] Robert Wildner, The Digital Markets Act: What a Difference a Month Makes, Mob. Mark. (Apr. 9, 2024), https://mobilemarketingmagazine.com/the-digital-markets-act-what-a-difference-a-month-makes.

^[112] Dirk Auer, Matthew Lesh, & Lazar Radic, Digital Overload: How the Digital Markets, Competition and Consumers Bill’s Sweeping New Powers Threaten Britain’s Economy, Inst. Econ. Aff. (Sep. 18, 2023), https://iea.org.uk/publications/digital-overload-how-the-digital-markets-competition-and-consumers-bills-sweeping-new-powers-threaten-britains-economy.

^[113] Report of the Committee on Digital Competition Law, Gov. India Minist. Corp. Aff., (Feb. 27, 2024), https://www.mca.gov.in/bin/dms/getdocument?mds=gzGtvSkE3zIVhAuBe2pbow%253D%253D&type=open (hereinafter “India Digital Competition Report”).

^[114] The Competition Act, No. 12 of 2003, India Code (2003), available at https://www.cci.gov.in/images/legalframeworkact/en/the-competition-act-20021652103427.pdf.

^[115] India Digital Competition Report, supra note 113.

^[116] Indian Competition Law, Vol. I, available at https://theindiancompetitionlaw.wordpress.com/wp-content/uploads/2013/02/report_of_high_level_committee_on_competition_policy_law_svs_raghavan_committee.pdf (last accessed Jan. 30, 2025).

^[117] Radic et al., supra note 3.

^[118] Id.

^[119] Giuseppe Colangelo, The European Digital Markets Act and Antitrust Enforcement: A Liaison Dangereuse, Int’l Ctr. L. Econ. (May 19, 2022), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4070310.

^[120] Lazar Radic, Fair Trade Act Amendments Would Create Flaws, The Korea Times, (Dec. 11, 2024), https://www.koreatimes.co.kr/www/opinion/2024/12/137_384007.html.

^[121] H.R. 3849, 117th Congress (Jun. 24, 2024), https://www.congress.gov/bill/117th-congress/house-bill/3849/text; S. 2992, 117th Congress (Mar. 2, 2022), https://www.congress.gov/bill/117th-congress/senate-bill/2992/text; S. 2710, 117th Congress (Feb. 17, 2022), https://www.congress.gov/bill/117th-congress/senate-bill/2710.

^[122] Radic & Manne, supra note 56.

^[123] Radic et al., supra note 3.

^[124] Press Release, Klobuchar, Grassley, Colleagues to Introduce Bipartisan Legislation to Rein in Big Tech, Office of Sen. Amy Klobuchar (Oct. 14, 2021), https://www.klobuchar.senate.gov/public/index.cfm/2021/10/klobuchar-grassley-colleagues-to-introduce-bipartisan-legislation-to-rein-in-big-tech.

^[125] PL n. 2768/2022, Câmara dos Deputados (Brazil), (Nov. 10, 2022), https://www.camara.leg.br/proposicoesWeb/prop_mostrarintegra?codteor=2214237&filename=PL%202768/2022.

^[126] Grand Design and Action Plan for a New Form of Capitalism: 2023 Revised Version, Jpn. Cabinet Secr. (2023), available at https://www.cas.go.jp/jp/seisaku/atarashii_sihonsyugi/pdf/ap2023en.pdf; Outline of the Act on Promotion of Competition for Specified Smartphone Software, Jpn. Fair Trade Comm. (Jun. 2024), available at https://www.jftc.go.jp/file/240612EN3.pdf; @laz_radic, X.com (Aug. 14, 2024, 6:17 a.m.), https://x.com/laz_radic/status/1823665316200899036.

^[127] Simon Vande Walle, Is the EU’s Digital Markets Act Going Global? How Japan Is Crafting Its Own Version of Digital Regulation with the Smartphone Act, EU Renew (Aug. 21, 2024), https://eu-renew.eu/is-the-eus-digital-markets-act-going-global-how-japan-is-crafting-its-own-version-of-digital-regulation-with-the-smartphone-act.

^[128] JFTC, supra note 126.

^[129] Online Intermediation Platforms Market Inquiry, Compet. Comm. S. Afr. (2000-2019), https://www.compcom.co.za/online-intermediation-platforms-market-inquiry.

^[130] Lina M. Khan & Sandeep Vaheesan, Market Power and Inequality: The Antitrust Counterrevolution and Its Discontents, 11 Harv. L. & Pol’y Rev. 235 (2017), https://scholarship.law.columbia.edu/faculty_scholarship/2790.

^[131] Chris Edwards, A Century of Federal Spending, Cato at Liberty (Mar. 16, 2023) https://www.cato.org/blog/century-federal-spending-1925-2025.

^[132] The Draghi Report, for instance, found that the European Union passed 13,000 acts from 2019 to 2024, four times more than the United States. See Mario Draghi, The Future of European Competitiveness, Eur. Comm. (Sep. 2024), available at https://commission.europa.eu/document/download/97e481fd-2dc3-412d-be4c-f152a8232961_en?filename=The%20future%20of%20European%20competitiveness%20_%20A%20competitiveness%20strategy%20for%20Europe.pdf.

^[133] Marc Ivaldi, Nicolas Petit, & Selcukhan Unekbas, Killer Acquisitions: Evidence from EC Merger Cases in Digital Industries, Antitrust Law J. (forthcoming), (Mar. 27, 2023), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4407333.

^[134] Thomas B. Nachbar, Heroes and Villains of Antitrust, 18 Antitrust Source 1–12 (Jun. 2019), https://www.law.virginia.edu/scholarship/publication/thomas-b-nachbar/642681.

Executive Summary

The proposed $35 billion combination of Discover Financial Services Inc. and Capital One Financial Corp. has faced scrutiny regarding its impact on “subprime” credit-card customers. Though the deal has been approved by federal regulators, there remains the potential that states could object to the combination on grounds that it harms subprime consumers. This issue brief examines whether such consumers constitute a distinct market for antitrust purposes and concludes they do not.

A fundamental challenge is the absence of a standardized definition of “subprime.” Credit bureaus, lenders, and government agencies use different credit-score thresholds to classify consumers—e.g. VantageScore considers scores between 300-600 to be subprime, while FICO uses different terminology entirely. This inconsistency makes it impossible to clearly delineate a market segment. Additionally, credit scores are highly dynamic, with research showing that about 40% of subprime borrowers improved to higher credit tiers during the pandemic.

Consumers with lower credit scores have numerous alternatives to traditional credit cards, including secured cards, buy-now-pay-later services (used by 21% of consumers with credit records in 2022), and personal loans. These substitutes prevent the merged entity from exercising market power. Major banks like JPMorgan Chase, Citigroup, and Bank of America already serve subprime borrowers, and could expand these offerings if needed.

Rather than harming competition, a merged Capital One-Discover could benefit consumers with lower credit scores by expanding the reach of Capital One’s sophisticated analytics, which identify lower-risk individuals among those with subprime scores. While the combined company would control approximately 30% of subprime credit-card balances, this falls far short of monopoly power, given the competitive constraints from other issuers and alternative credit products.

I. Introduction

The proposed combination of Discover Financial Services Inc. and Capital One Financial Corp. has been cleared by the relevant federal regulators. Earlier this month, the U.S. Justice Department (DOJ) reportedly informed financial regulators that it does not intend to block the $35 billion deal.[1] More recently, the Federal Reserve Board of Governors and the U.S. Office of the Comptroller of the Currency both announced April 18 that they have approved the transaction.[2] Moreover, the Delaware Office of the State Bank Commissioner granted approval in December 2024,[3] and shareholders of both Capital One and Discover overwhelmingly approved the merger in February 2025.[4] Nonetheless, the potential that other state regulators or attorneys general might still challenge the transaction remains.

Under traditional antitrust analysis, Capital One-Discover barely raises an eyebrow. The combined company would become the third-largest credit-card issuer by purchaser volume, after JPMorgan Chase and American Express.[5] Given that there are thousands of credit-card-issuing banks in the United States, and even the largest issuers hold only a modest percentage of all transaction volume, any potential countervailing adverse effect on competition would likely be minor, if noticeable at all. As with its banking operations, the combined company’s scale and innovative approach toward credit cards could drive improvements—especially for those with lower credit scores—both directly for its customers and indirectly for customers of other banks, who would be driven to provide competitive offerings.

The proposed acquisition has the potential to transform competition and consumer welfare in the retail-banking market. Through synergies and cost savings, the new entity would compete more vigorously with other banks and payment networks. Not only will this better serve the public in general, by bringing together the firms’ traditional expertise in the development of innovative banking servings and credit-card markets aimed at middle-income consumers, it would also likely expand financial inclusion among underserved communities.

Nevertheless, New York State Attorney General Letitia James has raised concerns that the combined company would control roughly 30% of the subprime credit-card market.[6] While this figure might seem significant at first glance, it requires deeper analysis. The term “subprime” is commonly used in the credit-card industry to categorize borrowers who present a higher risk of default due to their credit history.[7] But a significant challenge to analyzing this market segment—especially to evaluate antitrust concerns related to mergers—lies in the absence of a uniform definition for “subprime.”

This lack of standardization is evident when examining the diverse criteria employed by credit-card issuers and government agencies to classify borrowers. Bank of America’s annual report notes that “a standard industry definition for subprime loans… does not exist.”[8] Experian reports that “subprime is a moving target. Each lender defines subprime and prime depending on their lending strategies and business goals.”[9]

Credit-card companies primarily use the credit-scoring models developed by FICO and VantageScore, both of which present different perspectives on creditworthiness.[10] VantageScore explicitly defines the “subprime” range as encompassing credit scores between 300 and 600.[11] In contrast, FICO does not use the term “subprime” directly, but instead identifies scores below 670 as “Fair” and below 580 as “Poor,” with both classifications generally considered to be subprime.[12] In addition, Experian identifies scores of 660 and lower as “nonprime,” 600 and lower as “subprime,” and below 500 as “deep subprime.”[13] The fundamental differences among service providers illustrates an inherent ambiguity in defining subprime at the foundational level of credit assessment.

Government agencies, such as the Consumer Financial Protection Bureau (CFPB) and the Federal Reserve System, also play a crucial role in understanding the credit-card market, but their definitions of “subprime” further contribute to the lack of a unified standard. For example, in a 2025 report, the CFPB described FICO scores of 300-579 as “deep subprime,” 580-619 as “subprime,” and 620-659 as “near prime.”[14] Research published by the Federal Reserve defines “subprime” as those borrowers with Equifax Risk Scores of less than 620, and “near prime” as those with scores of 620-719.[15] This inconsistency across government agencies underscores the absence of a universally accepted definition.

“Subprime” credit-card markets also lack a coherent definition because of the unusually dynamic nature of the consumers that comprise it. Consumer credit scores can change dramatically within very short periods of time. As a result, there is a great deal of churn in the individual consumers who, over time, comprise the subprime credit-card market. Thus, while certain companies may maintain a larger market share in the subprime market at a given point in time, it is only by virtue of their relative ability to attract new customers with their offerings, as existing customers either graduate out of the subprime market or drop out of the market from default.

This issue brief examines the market for credit-card borrowing, and concludes that lending to consumers with “subprime” credit does not constitute any distinct relevant market for conducting an antitrust review of the merger. Section II describes the standards for delineating a relevant market under U.S. law and policy. Section III demonstrates that demand-side substitution fatally undermines claims that “subprime” constitutes a distinct antitrust market. Sections III and IV provide evidence that demand-side and supply-side substitution further negate the claim that “subprime” constitutes a relevant antitrust market. Section V identifies ways the proposed merger will likely be procompetitive for consumers with “subprime” credit.

II. Legal Standards for Market Definition

Much of the concern about the  proposed Capital One/Discover combination has focused on potential harms to specific groups of credit-card customers, especially what has been described as the “near-prime” or “subprime” segments of borrowers with FICO scores below 660.[16] A key question for antitrust analysis is whether these consumers constitute a distinct relevant market. One critic of the merger argues that these consumers’ higher risk, as well as Capital One and Discover’s direct-mail marketing strategies targeting them, suggest they constitute a distinct “submarket.”[17] In contrast, the Bank Policy Institute reports:

No evidence has been put forth by critics of the proposed merger to define the boundaries of the subprime segment and establish that consumers in this segment are sufficiently isolated for it to be considered a distinct submarket for antitrust purposes.[18]

Broadly speaking, U.S. antitrust law provides two approaches for evaluating which products and services comprise a relevant market. The first are known as the “principal indicia,” first described in Brown Shoe v. United States. The second and more recent approach is the “hypothetical monopolist test.”

A. Brown Shoe ‘Practical Indicia’

Brown Shoe Co. v. United States established the “practical indicia” test for market boundaries, requiring evidence of distinct product characteristics, specialized vendors, or unique consumer groups.[19] This approach has long been controversial.

For example, Geoffrey Manne and Marcellus Williamson criticize Brown Shoe’s reliance on “industry or public recognition of the market as a separate economic entity.”[20] They argue that this approach is fundamentally flawed, because business people and the public often use the term “market” for various reasons unrelated to economic substitutability, which is the core of a proper antitrust market definition. They emphasize that antitrust market definition should be based on economic analysis of substitutability, not on how businesses casually use terms like “market” or “industry.” Manne and Williamson warn that relying on such noneconomic evidence can lead to overly narrow market definitions that exaggerate market power and result in erroneous antitrust decisions. Their criticism is especially apt in the context of attempting to delineate markets, “submarkets,” or “segments” based on credit score, as it was rumored that the DOJ may have been concerned that the Capital One/Discover merger would be anticompetitive in “subprime,” “near prime,” “student,” or “no-credit-history” submarkets.[21]

While companies, analysts, and commentators may casually refer to “prime” or “subprime” consumers, such shorthand does not necessarily define relevant economic markets for evaluating the competitive effects of a proposed merger. Indeed, among the business press, there is a stunning paucity of articles that describe a “subprime market” for credit cards. The most recent is an op-ed by former Federal Deposit Insurance Corp. (FDIC) Chair Sheila Bair, who concluded that barriers to entry do not explain the concentration of firms serving subprime consumers:

I suspect Capital One’s subprime market share is relatively substantial because other banks simply have less (or no) interest in serving subprime customers. Subprime lending involves higher capital requirements, greater regulatory scrutiny and more resources to underwrite and manage those accounts. Any concentrations in the subprime market are the result of banks’ conscious investment decisions, not barriers to entry.[22]

Even if one accepts, for the sake of argument, that Brown Shoe is an appropriate framework to evaluate whether there is distinct subprime—or near prime, student, or no-credit-history—credit-card market, none of these categories satisfy Brown Shoe’s principle indicia. As we discuss below, such proposed markets do not have distinct product characteristics, do not constitute unique consumer groups, and are not served by specialized vendors.

B. Hypothetical Monopolist Test in a Two-Sided Market

Credit cards are a classic example of a two-sided platform, as they facilitate transactions between merchants and cardholders. In Ohio v. American Express, the U.S. Supreme Court established that credit-card networks are “two-sided transaction platforms” where merchants and cardholders simultaneously choose to use the network.[23] The Court held that these platforms cannot be analyzed by looking at just one side of the market in isolation, as the value of the platform to users on one side depends on the number of users on the other side.

The Hypothetical Monopolist Test (HMT) evaluates whether a group of products is sufficiently broad to constitute a relevant antitrust market by asking whether eliminating competition among these products by combining them under the control of a hypothetical monopolist would likely lead to worsening terms for customers. The test is typically assessed by using “small but significant and non-transitory increase in price” (SSNIP) analysis, which asks whether a hypothetical monopolist could profitably impose a 5-10% price increase on the candidate market.

In two-sided markets like credit cards, the traditional HMT must be modified to account for the interdependence between the two sides of the platform. As the 2^nd U.S. Circuit Court of Appeals noted in the Amex case, the proper HMT analysis must “consider the feedback effects inherent on the platform” by accounting for how changes in demand on one side would affect demand on the other side.[24]

For subprime credit cards, this means examining:

1. Both sides of the platform simultaneously: The analysis must consider both merchants who accept the cards and cardholders who use them. On the merchant side, this would include merchant discount rates and interchange fees. On the cardholder side, this would include interest rates, annual fees, late fees, and rewards programs.
2. Net price, rather than one-sided price: Following Ohio v. American Express, the analysis should focus on the “net price” of transactions across the platform, not just prices on one side.
3. Cross-platform substitution: The analysis must assess whether prime credit cards or other financial products—such as personal loans, buy-now-pay-later, or secured cards—constrain the pricing of subprime cards sufficiently to prevent a profitable SSNIP. It must also assess whether merchants could steer subprime customers to other payment methods without losing significant business.

We are not aware of any published research using the HMT to evaluate whether subprime credit cards are a relevant market. While published research discusses application of the HMT to the broader credit-card market, including its complexities as a two-sided market, there is no specific mention or analysis of the subprime segment within this market. The academic literature focuses instead on defining the relevant market for credit-card services in general, often debating whether to consider the “total price,” or specific fees like interchange fees.[25]

This is likely because conducting robust HMT analysis requires detailed data on price elasticities and substitution patterns. Obtaining granular data specifically for the subprime credit-card market, separate from the broader credit-card market, might be challenging. Moreover, the credit scores used to define “subprime” are arbitrary and the boundaries of the subprime market are fluid, with borrowers frequently moving between credit-score categories over time. Thus, applying the HMT to reliably evaluate a distinct subprime market may be so complex as to be nearly impossible.

III. Demand-Side Substitution

Demand-side substitution—that is, consumers’ ability to switch to alternative products in response to price increases or reduced quality—fatally undermines claims that “subprime” constitutes a distinct antitrust market.

Citing Brown Shoe, the latest version of the DOJ/FTC Merger Guidelines, notes: “The outer boundaries of a relevant product market are determined by the ‘reasonable interchangeability of use or the cross-elasticity of demand between the product itself and substitutes for it.’”[26] Substitutability is the sine qua non of market definition: If two products are reasonably substitutable or—even better—actually substituted by consumers, then the two products are said to be in the same relevant market. Sweet onions and yellow onions are reasonable substitutes for each other, but fresh onions and frozen battered onion rings are not. For consumers with subprime credit scores, the question is what the reasonable substitutes for a credit card are.

In a 2024 earnings call, Capital One Chairman and CEO Richard Fairbank identified several substitute products that are “looking to take market share from traditional credit card players”:

Let’s also remember that consumers can choose to use another form of payment entirely, cash, debit or Buy Now Pay Later, which has exploded onto the marketplace. New fintechs are entering the payments in small dollar credit space every day, all looking to take market share from traditional credit card players like Capital One. We faced this competition for years and we’ll continue to face it in the future. It’s powerful evidence of a healthy and fiercely competitive marketplace.[27]

Evidence from the CFPB indicates that subprime borrowers routinely access credit-card alternatives, such as buy-now-pay-later (BNPL):

• Roughly 21% of consumers with a credit record borrowed using BNPL from at least one of the six firms at least once during 2022.[28]
• From 2021 to 2022, borrowers with deep subprime credit scores accounted for 45% of BNPL originations, while those with subprime credit scores were responsible for another 16% of originations.[29]

Consumers without a credit history access secured credit cards, which they can use to establish credit, increase their credit scores, and “graduate” to unsecured cards. Research published in 2024 by the Federal Reserve Bank of Philadelphia reports that about 57% of new secured card borrowers lack a credit score at origination.[30] Of those with a credit score, about half are “deep subprime” and another 25% are “subprime.”[31]

The clear, arbitrary distinction between “prime” and “subprime” products has eroded. In a letter to the CFPB, the American Bankers Association, Consumer Bankers Association, and National Association of Federally-Insured Credit Unions reported that, by mid-2022, rewards cards made up 85% of total U.S. credit-card accounts.[32] For new accounts (less than two years old), rewards cards comprised 78% of total accounts. Among subprime accounts, rewards cards made up 72% of total card volume, compared to only 42% of total volume in 2008. With rewards card available to—and adopted by—consumers with subprime credit, the presence or absence of rewards provides little information about whether a given card is a “prime” or “subprime” product.

One important consideration in evaluating this concern is that a consumer’s credit status is rarely static over time. Due to changes in income and other circumstances, a subprime borrower today may be a prime borrower next year, and vice versa. Using data from 2014 and 2015, Fair Isaac found that a “notable percentage” of FICO scores migrated up or down more than 20 points in a six-month period, with 14% of accounts decreasing by more than 20 points, and 19% increasing by more than 20 points.[33] The CFPB reports that 43% of consumers with subprime credit scores moved up at least one tier during the COVID-19 pandemic, whereas in the 10 years prior to the pandemic, only 37% moved up at least one tier.[34]

In a letter to the Federal Reserve Bank of Richmond and the Office of the Comptroller of the Currency, Capital One reported that: “Since our founding, we have enabled more than 42 million customers with subprime or no FICO scores when they opened a card with the bank to achieve prime or better FICO scores.”[35]

The 2024 Philadelphia Fed report examined the extent to which holders of secured credit cards “graduate” to unsecured cards.[36] The report found that most graduations occur between six and 12 month into an account’s life, with 33% of “unscored” borrowers (i.e., those with no credit history when the account was opened) graduating within a year, and about half graduating within 30 months.

Thus, even if a subprime or near-prime market segment could be defined, migration into and out of these segments makes it exceedingly difficult to establish a reliable market definition for antitrust analysis.

IV. Supply-Side Substitution

Supply-side substitution—the ability of firms to redeploy resources to produce substitute goods or enter new markets—further negates the claim that “subprime” constitutes a distinct antitrust market. This analysis demonstrates that credit-card issuers and financial-technology firms (“fintechs”) can rapidly adjust offerings to compete for subprime borrowers, constraining any potential anticompetitive behavior.

While the prevailing approach in U.S. antitrust analysis, as outlined in the 2023 Horizontal Merger Guidelines, prioritizes demand-side substitution in the initial market definition, supply-side considerations are not entirely excluded. Supply responses are taken into account when identifying the participants in the relevant market and when assessing the likelihood of new entry. Specifically, firms that are not currently producing the product in question, but that could rapidly enter the market without incurring significant sunk costs in response to a price increase, are considered “rapid entrants” and therefore factored into the competitive analysis. These rapid entrants are defined as firms that “very likely would rapidly enter with direct competitive impact in the event of a small but significant change in competitive conditions.”[37] Jorge Padilla explains:

Indeed, even if consumers were unable to react immediately to an increase in price, producers might be able to do so rather quickly. How? First, some of them may be endowed with assets (physical and human) that can be easily adjusted to produce substitute goods. If these producers were able to respond to a price increase by switching their production facilities to produce the goods or services subject to such price increase, then consumers would be able to avoid abuse.[38]

Capital One entered and gained its market share in “subprime” over time through its data-driven strategy. This has enabled the company to identify those lower-risk individuals in otherwise higher-risk groups, thereby serving otherwise underserved consumers while limiting default risk.[39] It’s been estimated that the combined company would account for approximately 30% of the subprime segment (Table 1). While this may be a sizable share, it is far short of a monopoly.

Table 1: Credit-Card Balances for Issuers with More Than $1B in Balances

SOURCE: Bank Policy Institute[40]

Moreover, the other three of the five largest issuers—JP Morgan Chase, Citigroup, and Bank of America—account for about one-third of subprime balances. These firms have the expertise and resources to respond to any post-merger increase in price, or diminution of quality. The fact that major prime credit-card issuers already operate in the subprime segment demonstrates that the necessary infrastructure, regulatory understanding, and risk-management frameworks are not unique to specialized subprime lenders. These firms could easily expand their subprime offerings if market conditions, such as a price increase, made doing so more attractive.

Capital One’s Fairbank noted in the previously mentioned 2024 earnings that that “any existing bank can choose where in the credit spectrum they play simply by changing their credit policy.”[41] In that same call, he claimed that BNPL “has exploded onto the marketplace.” Commenting on the pending deal between Capital One and Discover, Michael Imerman of the University of California, Irvine concluded:

As a result, the combined bank would be in a position to be more competitive against digital banks and fintech competitors that have made significant progress moving upmarket in the consumer banking space in the past few years.[42]

V. Procompetitive Effects for Subprime Consumers

Capital One and Discover have specialized in differing ways in providing services to customers with lower credit scores, and the combination will enable them to leverage their combined expertise to serve those customers better. For example, it will be able to use Capital One’s algorithms to identify Discover customers who, despite having low credit scores, are lower risk, and offer those customers loans at preferential rates.

This highlights the reality of “subprime” credit: it is not a separate credit-card “market” because, as discussed above, the individuals classified as “subprime” are dynamic; many subprime consumers improve their credit and gain access to prime cards, while those with higher credit scores may hit bumps in the road and move in the opposite direction.

Most major card issuers have focused on catering to wealthier, high-spend, and low-risk consumers who require less effort to underwrite and serve, and lower capital-reserve requirements. Capital One gained its market share in “subprime” over time through its data-driven strategy and ability to better identify diamonds in the subprime rough than its competitors. This also provides opportunities for these consumers to migrate toward a lower-risk category by gradually increasing the size of their credit lines as they demonstrate creditworthiness.[43]

The combination of Capital One and Discover will almost certainly increase access to credit for people with low credit scores, thereby enabling them to get onto the first rung of the credit ladder and to build or rebuild their credit record. It should thus be considered a good outcome for the new administration, with its increased attention to the effects of mergers on working-class Americans.

VI. Conclusion

The proposed combination of Discover and Capital One should not raise significant antitrust concerns based on “subprime” market concentration. As demonstrated throughout this analysis, the concept of a distinct “subprime” credit-card market fails to satisfy established legal standards for market definition. The fluid nature of credit scores undermines attempts to define static market boundaries. Moreover, the extensive demand-side substitution opportunities—including secured cards, BNPL services, and traditional banking products—provide meaningful competitive constraints on any potential market power in credit-card lending to consumers with lower credit scores.

From a supply-side perspective, major credit-card issuers like JPMorgan Chase, Citigroup, and Bank of America already serve subprime borrowers and could readily expand their offerings if the merged entity attempted to raise prices or reduce quality. Rather than harming competition, the Capital One/Discover merger promises to enhance financial access for consumers with lower credit scores by combining Capital One’s sophisticated risk-assessment algorithms with Discover’s infrastructure. This would enable the merged entity to better identify lower-risk individuals within traditionally higher-risk categories, potentially offering them more favorable terms than either company could provide independently.

[1] Capital One’s Discover Acquisition Gains Approval from Justice Department, GlobalData (Apr. 7, 2025), https://finance.yahoo.com/news/capital-one-discover-acquisition-gains-112243059.html.

[2] Ashley Capoot, Capital One and Discover Merger Approved by Federal Reserve, CNBC (Apr. 18, 2025), https://www.cnbc.com/2025/04/18/capital-one-and-discover-merger-approved-by-federal-reserve-board.html.

[3] Katie Tabeling, Delaware Bank Regulators Approve Capital One, Discover Deal, Del. Bus. Times (Dec. 19, 2024), https://delawarebusinesstimes.com/news/delaware-bank-regulators-capital-one.

[4] Press Release, Capital One and Discover Stockholders Approve Capital One’s Proposed Acquisition of Discover, Capital One Financial Corp. (Feb. 18, 2025), https://investor.capitalone.com/news-releases/news-release-details/capital-one-and-discover-stockholders-approve-capital-ones.

[5] Tiziana Barghini, Capital One’s Discover Acquisition Would Reshape US Credit Card Industry, Glob. Finance Mag. (Mar. 4, 2024), https://gfmag.com/banking/capital-one-acquires-discover.

[6] Caitlin Mullen, Capital One-Discover Deal Draws NY Scrutiny, Payments Dive (Oct. 25, 2024), https://www.paymentsdive.com/news/ny-ag-letitia-james-capital-one-discover-deal-antitrust-probe/731004.

[7] Tim Maxwell, The Pros and Cons of Subprime Mortgages, Experian (Jul. 11, 2022), https://www.experian.com/blogs/ask-experian/the-pros-and-cons-of-subprime-mortgages.

[8] Annual Report (Form 10-K), Bank of America Corp. (Mar. 25, 2025), available at https://investor.bankofamerica.com/regulatory-and-other-filings/all-sec-filings/content/0000070858-25-000139/0000070858-25-000139.pdf.

[9] Louis DeNicola, What Does Subprime Mean?, Experian (Jul. 9, 2022), https://www.experian.com/blogs/ask-experian/what-is-subprime.

[10] Louis DeNicola, The Difference Between VantageScore Credit Scores and FICO Scores, Experian (Mar. 31, 2023), https://www.experian.com/blogs/ask-experian/the-difference-between-vantage-scores-and-fico-scores.

[11] VantageScore 4.0 User Guide, Vantage Score (Sep. 2022), available at https://web.archive.org/web/20250108190418/https://www.vantagescore.com/wp-content/uploads/2022/09/VantageScore-4.0-UserGuide_abr_Sep22.pdf.

[12] Brianna McGurran, What Is a FICO Score, and Why Is It Important?, Experian (Jul. 24, 2024), https://www.experian.com/blogs/ask-experian/fico-score-what-it-is-and-why-its-important.

[13] Elizabeth Gravier, The 5 Credit Score Ranges You Need to Know, CNBC (Dec. 27, 2024), https://www.cnbc.com/select/borrower-risk-profiles-based-on-credit-score.

[14] Consumer Use of Buy Now, Pay Later and Other Unsecured Debt, Consum. Financ. Prot. Bur. (Jan. 2025), available at https://files.consumerfinance.gov/f/documents/cfpb_BNPL_Report_2025_01.pdf.

[15] John Driscoll, Jessica Flagg, Bradley Katcher, & Kamila Sommer, The Effects of Credit Score Migration on Subprime Auto Loan and Credit Card Delinquencies, FEDS Notes (Jan. 12, 2024), https://www.federalreserve.gov/econres/notes/feds-notes/the-effects-of-credit-score-migration-on-subprime-auto-loan-and-credit-card-delinquencies-20240112.html.

[16] CFPB provides the following definitions: superprime (800 or greater), prime plus (720 to 799), prime (660 to 719), near-prime (620 to 659), subprime (580 to 619), and deep subprime (579 or less). The Consumer Credit Card Market, Consum. Financ. Prot. Bur. (Oct. 2023), at 12, available at https://files.consumerfinance.gov/f/documents/cfpb_consumer-credit-card-market-report_2023.pdf.

[17] Shahid Naeem, Capital One-Discover: A Competition Policy and Regulatory Deep Dive, Am. Econ. Lib. Proj. (Mar. 2024), available at https://www.economicliberties.us/wp-content/uploads/2024/03/2024-03-20-Capital-One-Discover-Brief-post-design-FINAL.pdf.

[18] Haelim Anderson, Paul Calem, & Benjamin Gross, Is the Subprime Segment of the Credit Card Market Concentrated? Bank Policy Inst. (May 31, 2024), https://bpi.com/is-the-subprime-segment-of-the-credit-card-market-concentrated.

[19] Brown Shoe Co. v. United States, 370 U.S. 294 (1962).

[20] Geoffrey A. Manne & E. Marcellus Williamson, Hot Docs vs. Cold Economics: The Use and Misuse of Business Documents in Antitrust Enforcement and Adjudication, 47 Ariz. L. R. 609 (2005), available at https://laweconcenter.org/wp-content/uploads/2005/01/47arizlrev609.pdf.

[21] See also Diana Moss, The Capital One Financial-Discover Financial Services Merger: A Test for the Biden Merger Agenda?, Progress. Policy Inst. (Jun. 20, 2024), at 5, available at https://www.progressivepolicy.org/wp-content/uploads/2024/06/PPI-Capitol-One-Discover-Commentary.pdf (“Both Capital One and Discover serve the non-prime credit card lending market, where some commentators note that the merger could raise interest rates and fees, thus widening gaps in wealth and income, particularly for at-risk communities.”).

[22] Sheila Bair, How the Capital One/Discover Deal Could Boost Competition, Financ. Times (May 31, 2024), https://on.ft.com/4640E6h.

[23] Ohio v. American Express Co., 138, S.Ct. 2274, 2276-77, 585 U.S. 529 (2018) (“Respondent… Amex… operate[s] what economists call a ‘two-sided platform,’ providing services to two different groups (cardholders and merchants) who depend on the platform to intermediate between them. Because the interaction between the two groups is a transaction, credit-card networks are a special type of two-sided platform known as a ‘transaction’ platform. The key feature of transaction platforms is that they cannot make a sale to one side of the platform without simultaneously making a sale to the other. Unlike traditional markets, two-sided platforms exhibit ‘indirect network effects,’ which exist where the value of the platform to one group depends on how many members of another group participate. Two-sided platforms must take these effects into account before making a change in price on either side, or they risk creating a feedback loop of declining demand. Thus, striking the optimal balance of the prices charged on each side of the platform is essential for two-sided platforms to maximize the value of their services and to compete with their rivals.”)

[24] United States v. American Express Co., No. 15-1672 (2d Cir. 2016).

[25] See, e.g., Eric Emch & T. Scott Thompson, Market Definition and Market Power in Payment Card Networks, 5 Rev. Network Econ. 45.

[26] Horizontal Merger Guidelines, U.S. Dep’t of Justice & Fed. Trade Comm’n (2023), at 4.3, available at https://www.ftc.gov/system/files/ftc_gov/pdf/2023_merger_guidelines_final_12.18.2023.pdf.

[27] Earnings Call, Capital One Financ. Corp. (Apr. 25, 2024), https://www.sec.gov/Archives/edgar/data/1393612/000092762824000164/a425-04252024cofearningstr.htm.

[28] CFPB, supra note 15.

[29] Id.

[30] Larry Santucci, Secured Card Market Update, Fed. Reserve Bank Phila. (May 2024), available at https://www.philadelphiafed.org/-/media/frbp/assets/consumer-finance/reports/secured-card-market-update.pdf.

[31] Id.

[32] Letter from American Bankers Association, Consumer Bankers Association & National Association of Federally-Insured Credit Unions to Bureau of Consumer Financial Protection (Apr. 24, 2023), available at https://www.nafcu.org/system/files/files/CFPB-2023-0009%20Joint%20Trades%20Letter%20to%20CFPB%20re%20Consumer%20Credit%20Card%20Market.pdf.

[33] See FICO Research: Consumer Credit Score Migration, FAIR Isaac Corp. (2018), https://www.fico.com/en/latest-thinking/white-paper/fico-research-consumer-credit-score-migration.

[34] Alyssa Brown & Siobhán McAlister, Office of Research Blog: Credit Score Transitions During the COVID-19 Pandemic, Consum. Financ. Prot. Bur. (Jan. 25, 2023), https://www.consumerfinance.gov/about-us/blog/office-of-research-blog-credit-score-transitions-during-the-covid-19-pandemic.

[35] Letter from Andres L. Navarrete, Executive Vice President, Head of External Affairs, Capital One to Brent Hassell, Assistant Vice President, Federal Reserve Bank of Richmond and Jason Almonte, Director for Bank Licensing, Office of the Comptroller of the Currency (Aug. 7, 2024), available at https://www.federalreserve.gov/foia/files/capital-one-supplemental-information-20240807.pdf.

[36] Santucci, supra note 32.

[37] Merger Guidelines, supra note 28 at 4.4.A.

[38] Atilano Jorge Padilla, The Role of Supply-Side Substitution in the Definition of the Relevant Market in Merger Control, Report for DG Enterprise A/4, Eur. Comm. (Jun. 2001), https://ec.europa.eu/docsroom/documents/2658/attachments/1/translations/en/renditions/native.

[39] See Andrew Becker, The Secret History of the Credit Card, Frontline (Nov. 23, 2004), https://www.pbs.org/wgbh/pages/frontline/shows/credit/more/battle.html (“‘By identifying lower-risk individuals in high-risk groups, Capital One was able to market to reliable consumers other companies wouldn’t touch,’ says [Chris] Meyer [CEO of Monitor Networks]. In just six years, Capital One became the sixth-largest credit card issuer in the country. ‘When others were attacking the market with blunt instruments, Capital One used a scalpel,’ says Meyer.”).

[40] Bank Policy Institute, supra note 20.

[41] Capital One, supra note 29.

[42] Spencer Tierney, What the Capital One-Discover Deal Could Mean for Bank Accounts, NerdWallet (Feb. 21, 2024), https://www.nerdwallet.com/article/banking/capital-one-discover-deal-impact-on-bank-accounts.

[43] Naomi Snyder, Capital One’s Secret to Success, Bank Dir. (Aug. 15, 2022), https://www.bankdirector.com/article/capital-ones-secret-to-success.