ICLE White Paper

Digital Overreach: A Premature Turn to Ex Ante Regulation in Brazil

A Portuguese language version of this white paper is available here. 

Executive Summary

On September 18, 2025, the Brazilian government submitted Bill 4,675/2025 to the House of Representatives as part of President Luiz Inácio Lula da Silva’s “Digital Brazil Agenda.” The bill would amend Brazil’s Competition Law (Lei No. 12,529/2011) to create a new Digital Markets Superintendency (SMD) within the Administrative Council for Economic Defense (CADE), empower it to designate firms of “systemic relevance in digital markets” for up to 10 years, and impose “special obligations” drawn from a non-exhaustive statutory menu. These include prohibitions on self-preferencing, interoperability mandates, data portability requirements, mandatory merger notification regardless of existing thresholds, and broad business-user data access rights.

This paper argues that the bill is neither necessary under the current framework of Brazilian Competition Law (BCL) nor supported by compelling evidence from the antitrust literature. If enacted, it would require substantial modification to reduce its legal, economic, and institutional risks.

A.       Overarching Findings

Three findings anchor this conclusion.

1. Existing competition law already achieves the bill’s stated objectives. CADE’s December 2025 settlements with Apple and Google demonstrate that case-by-case enforcement through traditional Termos de Compromisso de Cessação (TCCs) already produces the outcomes the bill seeks to impose through ex ante These settlements were tailored to specific market conditions and allowed security-sensitive balancing that rigid rules cannot replicate. Despite CADE’s active oversight of digital markets, there remains no adjudicated finding of infringement under the BCL concerning the types of conduct the bill would regulate, weakening the evidentiary case for a sweeping new regime.
2. European evidence counsels caution. Emerging empirical evidence from the EU’s Digital Markets Act (DMA) suggests that regulatory obligations have harmed consumers, failed to shift market dynamics, and chilled innovation without delivering on contestability promises. Studies document measurable welfare losses. DMA-mandated changes to Google Maps increased search queries by 21% without increasing traffic to competing services. Chiara Farronato, Andrey Fradkin, and Alexander MacKay find that removing Amazon’s private-label products would reduce consumer surplus by 5.5%. Other evidence shows that Amazon’s post-DMA search results display fewer products (an 18% decline) and fewer sellers per product (a 7% decline). Consumer surveys report increased friction in routine online tasks, while 80% of respondents had no awareness of the DMA. Innovation costs are also significant. Apple, Meta, and Google have delayed European product launches due to compliance requirements, creating an “innovation tax” on consumers in regulated jurisdictions. At the macro level, Mario Draghi’s 2024 report finds that Europe’s productivity gap with the United States is largely explained by the technology sector and identifies excessive regulation as a key contributing factor.

3. Brazil’s institutional constraints amplify the risks. Brazil ranks 78th of 143 countries on the Rule of Law Index and 124th on ease of doing business, increasing the risks associated with broad regulatory discretion. The country already faces the “Brazil Cost,” estimated at approximately R$1.7 trillion annually (roughly 19.5% of GDP), and ranks among the most restrictive jurisdictions in OECD product-market regulation indicators. The bill also carries geopolitical risk. As the United States increasingly frames foreign digital regulation as discriminatory toward U.S. firms, adopting a DMA-style regime may expose Brazil to policy responses that lawmakers have not fully considered.

B.        Summary of Recommendations

The paper offers three tiers of recommendations, together with targeted revisions to the bill’s most problematic provisions.

1.       Tier 1: Forgo Ex Ante Regulation; Strengthen Existing Enforcement

The preferred approach is to forgo ex ante regulation and instead strengthen CADE’s existing enforcement toolkit through additional resources, streamlined procedures, and targeted market studies and sector inquiries. Brazil should allow more evidence to accumulate from jurisdictions that moved first before imposing untested regulatory burdens.

Alternative models demonstrate that broad regulation is not the only option. Japan’s Mobile Software Competition Act, effective December 2025, adopts a narrower approach focused on smartphone software ecosystems, incorporates objective-justification defenses tied to cybersecurity and system stability, does not require web-based sideloading, and explicitly protects device security and user information. A targeted model of this kind is more likely to achieve policy objectives without the collateral costs of a comprehensive regime.

2.       Tier 2: If Ex Ante Regulation Proceeds, Impose Substantial Guardrails

The bill raises five principal concerns. The following modifications would mitigate its most significant risks:

a)         Objectives and Limiting Principles (Art. 47-B)

The bill’s objectives—reducing entry barriers, protecting the competitive process, and promoting freedom of choice—lack a consumer-welfare anchor and risk justifying intervention that protects competitors rather than consumers.

Recommendations:

• Enshrine consumer welfare as the regime’s explicit limiting principle.
• Retain the three objectives only as indicators of potential harm, not as independent bases for enforcement.

b)         Designation Test (Art. 47-C; Art. 87-A)

The designation framework conflates market reach with market power and relies on structural proxies that risk capturing firms whose size reflects competitive success.

Recommendations:

• Require demonstrated, durable market power as a prerequisite for designation.
• Replace “systemic relevance” with “substantial market power” or “competitive significance.”
• Apply designation to specific services, not entire corporate groups.
• Shorten the designation period to three to five years, with mandatory periodic review.
• Remove counterproductive factors, such as “multiple digital products or services.”
• Incorporate indicators of actual competitive constraints, including potential competition and technological disruption.

c)         Institutional Design (Arts. 14-A, 14-B)

The creation of a second superintendency risks fragmentation, overlap, and inconsistent enforcement.

Recommendations:

• House the digital-markets function within the existing General Superintendence.
• Introduce complaint-screening mechanisms to deter strategic filings.
• Clarify regulatory responsibilities through statutory provisions.
• Address resource constraints explicitly.

d)         Prohibited and Mandated Practices (Art. 47-E)

The bill’s per se prohibitions and mandates risk condemning procompetitive conduct and creating tensions with security, privacy, and investment incentives.

Recommendations:

• Recognize that self-preferencing, vertical integration, and related practices can be procompetitive.
• Replace vague categories such as “predatory or abusive strategies” with effects-based standards.
• Introduce structured defenses based on security, privacy, fraud prevention, and trade-secret protection.
• Require proportionality assessments and mandatory CADE-ANPD coordination.
• Remove the blanket merger-notification requirement or replace it with a targeted call-in mechanism.

e)         Defenses (Art. 47-E, §2; Art. 87-B, §1; Art. 87-G)

The absence of an efficiencies defense creates a fundamental inconsistency with the BCL.

Recommendations:

• Adopt a meaningful efficiencies defense aligned with Articles 36, § 1 and 88, § 6 of the BCL.
• Replace “may consider” with “shall consider.”
• Define “economic justification” as requiring proportionality, net consumer benefit, and least-restrictive means, and make it judicially reviewable.
• Require reasoned economic analysis in Tribunal decisions..

3.       Tier 3: At a Minimum, Require a Regulatory Impact Assessment

At a minimum, the bill should undergo a rigorous Regulatory Impact Assessment prior to enactment. Brazil’s Economic Freedom Act (Lei No. 13,874/2019) already imposes such requirements on regulatory agencies, and Congress should apply the same standard to legislation of this magnitude. The European Commission conducted such an assessment before adopting the DMA; no equivalent preceded Bill 4,675/2025.

C.       Final Observation

Brazil has the opportunity to learn from, rather than replicate, the experience of jurisdictions that moved first. The available evidence suggests that the costs of ex ante digital regulation are real and measurable, while its benefits remain uncertain. Strengthening existing competition-law tools, investing in institutional capacity, and maintaining regulatory humility offer a more promising path for promoting consumer welfare and innovation.

I.          Introduction

On September 18, 2025, the Brazilian government submitted Bill 4,675/2025 to the House of Representatives as part of President Luiz Inácio Lula da Silva’s “Digital Brazil Agenda.”[1] The bill would amend Brazil’s Competition Law (BCL)[2] to create a new Digital Markets Superintendency (Superintendência de Mercados Digitais, or SMD) within the Administrative Council for Economic Defense (CADE). It would empower the SMD to designate firms of “systemic relevance in digital markets” (relevância sistêmica em mercados digitais) for up to 10 years and impose “special obligations.” These obligations may include prohibitions on self-preferencing, interoperability mandates, data-portability requirements, mandatory merger notification regardless of existing thresholds, and broad business-user data-access rights.[3]

This paper argues that the bill is neither necessary under the current BCL framework nor supported by compelling, undisputed evidence from the antitrust literature.[4] Three findings anchor this conclusion.

First, CADE’s December 2025 settlement with Apple shows that existing Brazilian competition law already can achieve the outcomes the bill seeks to mandate. Through case-by-case enforcement, CADE secured tailored, security-sensitive remedies—including the allowance of third-party app stores, alternative payment systems, and the removal of anti-steering restrictions—without resorting to ex ante regulation.[5]

Second, a growing body of empirical evidence, particularly from the European Union’s Digital Markets Act (DMA), suggests that at least some digital regulatory obligations have harmed consumers, failed to shift market dynamics, and chilled innovation, without delivering on promised gains in contestability.

Third, Brazil’s institutional constraints heighten the risks of broad regulatory discretion. Brazil ranks 78th of 143 countries on rule of law[6] and 124th on ease of doing business.[7] These conditions make expansive, discretionary regulation more likely to generate error costs than in the jurisdictions the bill seeks to emulate.[8]

The paper proceeds in two parts. Section II situates the bill in its Brazilian and international context and presents the affirmative case that ex ante regulation is unnecessary—or, at minimum, untested and premature. Section III offers a systematic critique of the bill’s core provisions, including its institutional design, prohibitions, and mandates.

The paper concludes with tiered recommendations. First, it proposes strengthening CADE’s existing enforcement toolkit. Second, it outlines guardrails to mitigate risks if ex ante regulation proceeds. Third, it recommends a minimum requirement of a rigorous regulatory-impact assessment.

II.      The Case Against Ex Ante Regulation in Brazil

The case for Bill 4,675/2025 rests on two implicit premises: first, that digital markets in Brazil exhibit competition problems that existing law cannot adequately address; and second, that the ex ante regulatory model developed abroad has proven sufficiently effective to justify adoption in Brazil. Neither premise withstands scrutiny.

This section evaluates those assumptions from four complementary angles.

Section II.A situates the bill within the broader global shift toward ex ante digital-market regulation. It shows that, despite frequent claims to the contrary, there is no settled international consensus on either the necessity or the design of such regimes.

Section II.B turns to Brazil’s own experience. It examines CADE’s recent enforcement actions—particularly the Apple and Google settlements—and shows that existing Brazilian competition law already can deliver the same types of outcomes the bill seeks to mandate, without imposing additional regulatory burdens.

Section II.C then assesses the growing empirical record from the European Union’s Digital Markets Act (DMA). The available evidence suggests that ex ante obligations have, in some cases, increased consumer costs, failed to shift competitive dynamics, and introduced frictions without delivering measurable gains in contestability.

Section II.D explains why these risks are amplified in Brazil. The country’s institutional constraints, existing regulatory burdens, opportunity costs, and geopolitical exposure make the adoption of a discretionary, design-based regulatory regime particularly hazardous.

Taken together, these considerations undermine the bill’s core justification. They show not only that Brazil lacks a demonstrated need for ex ante intervention, but also that the model it seeks to emulate carries significant and well-documented risks.

A.       The Rise of Ex Ante Digital Regulation—and Its Limits

Over the past decade, a series of influential policy reports argued that traditional antitrust enforcement struggles to address fast-moving digital markets. The most prominent—the European Commission’s Competition Policy for the Digital Era (the “Cremer Report”), the United Kingdom’s Unlocking Digital Competition (the “Furman Report”), and the Stigler Center’s study of digital platforms—converged on a shared diagnosis. They contended that network effects, data advantages, and tipping dynamics create structural barriers that ex post enforcement cannot remedy in a timely manner.[9] Filippo Lancieri and Patricia Sakowski later synthesized these reports, documenting the extent to which they rely on a common intellectual framework that treats digital markets as structurally prone to entrenchment.[10]

This intellectual convergence helped catalyze a wave of legislative initiatives, including Germany’s Section 19a of the GWB (2021), the European Union’s Digital Markets Act (DMA) (2022), and the United Kingdom’s Digital Markets, Competition and Consumers Act (DMCC) (2024).[11]

Supporters often portray this shift as reflecting a broad international consensus. That claim overstates the degree of agreement. As Lazar Radic explains in his analysis of the “imaginary antitrust consensus,” no consensus exists on three foundational premises: whether digital markets are inherently more prone to anticompetitive conduct, whether ex ante rules are necessary to address any such risks, or—among proponents—what an optimal regulatory framework should look like.[12] Outside the European Union, adoption of DMA-style regimes remains limited, particularly in the United States.[13]

Herbert Hovenkamp offers a related critique. He argues that antitrust’s firm-specific approach remains better suited to digital markets precisely because platforms differ in inputs, products, customer relationships, and third-party interactions. Broad ex ante rules, by contrast, risk misfiring in heterogeneous markets.[14]

To the extent a common purpose emerges from these regimes, it may not center on consumer protection. Critics argue that measures like the DMA instead aim to “level down” large firms and redistribute rents to politically favored rivals, rather than address demonstrable structural harms.[15]

Brazil’s engagement with this debate has followed a winding path. Its first major proposal, Bill 2,768/2022, combined the DMA’s rigid regulatory structure with the rhetoric of the U.S. “New Brandeis” movement, citing both the European model and the work of Lina Khan and Tim Wu.[16] That bill stalled in committee.

In January 2024, the Ministry of Finance’s Secretariat of Economic Reforms launched a public consultation on digital platform regulation, receiving 298 submissions from 72 participants.[17] The Ministry’s post-consultation report, published in October 2024, rejected the DMA’s blanket prohibitions.[18] Instead, it proposed a flexible “hybrid” model that would allow CADE to designate systemically relevant platforms and impose tailored obligations following a detailed, case-specific investigation of each firm’s business model.

That report underpins Bill 4,675/2025. Even so, it rests on questionable premises—particularly the assumption that digital markets tend toward “winner-take-all” outcomes.[19] Empirical evidence shows that tipping and entrenchment are not universal features of platform markets.[20] They must be demonstrated, not presumed.

Against a backdrop of unrelated concerns about children’s online safety, the executive branch unveiled its “Digital Brazil Agenda” on September 17, 2025, and submitted Bill 4,675/2025 the following day.[21] As noted above, the bill would amend Brazil’s Competition Law to create a digital-markets regulatory apparatus within CADE. It would establish a Superintendência de Mercados Digitais (SMD) empowered to designate firms of “systemic relevance” and impose “special obligations” drawn from a non-exhaustive statutory menu.

Mimicking foreign regimes, even if that is the goal, does not establish the existence of market failures in Brazil’s digital economy. The Ministry’s consultation record reveals sharply divided stakeholder views. Some observers—including the Coalition for App Fairness, Mercado Livre, and Match Group—argued that existing enforcement is insufficient. Others—including Amazon Brazil, Google, CCIA, camara-e.net, academic groups, and think tanks such as the Legal Grounds Institute, the Information Technology and Innovation Foundation (ITIF), the Global Antitrust Institute (GAI), and TechFreedom—countered that current antitrust tools are flexible enough to address digital markets without incurring the innovation costs of ex ante regulation.[22]

The bill also reflects broader global tensions surrounding these regimes. Its Explanatory Memorandum (Exposição de Motivos Interministerial, or EMI) claims the proposal is tailored to Brazilian conditions.[23] At the same time, it asserts that the law would place Brazil “at the vanguard” (na vanguarda) of digital-platform regulation alongside Germany, Japan, and the United Kingdom.[24] This aspiration to regulatory leadership sits uneasily with the claim that the bill reflects a cautious, evidence-based response to domestic conditions.

Ultimately, the burden rests on proponents to demonstrate a Brazil-specific problem that only an ex ante regime can solve. As the next section shows, the available evidence points in the opposite direction.

B.        Existing Brazilian Competition Law Is Sufficient

The strongest argument against Bill 4,675/2025 is empirical, not theoretical. Brazil’s existing competition-law framework has already demonstrated its ability to achieve the bill’s stated objectives.

In December 2025, CADE reached a landmark settlement with Apple through a Termo de Compromisso de Cessação (TCC). The agreement requires Apple to permit third-party app stores on iOS, enable alternative payment processors, remove anti-steering restrictions that prevented developers from informing users about lower-priced alternatives, and adopt neutral language in its communications about these changes.[25] Apple was required implement these measures within 105 days under a three-year agreement.

CADE secured a parallel outcome in the Android ecosystem. That same month, it finalized another sweeping TCC with Google.[26] The agreement prohibits Google from conditioning Google Play Store licensing on the preinstallation or preferential placement of proprietary apps such as Google Search and Google Chrome. It also bars Google from tying revenue-sharing payments to device manufacturers and network operators to the exclusion of rival search services. Google had to notify its partners of these changes and renounce incompatible exclusivity provisions within 45 days, also under a three-year agreement.

These settlements deliver the same types of outcomes that Bill 4,675/2025 would impose through its menu of “special obligations.” CADE achieved them through existing law, with several advantages over an ex ante regime.

First, the remedies were tailored to the specific facts and competitive dynamics of each case. In the Apple matter, for example, the company resisted a direct sideloading mandate on legitimate security and privacy grounds, yet still implemented substantial ecosystem changes. Case-by-case enforcement thus permits the kind of nuanced, security-sensitive balancing that rigid ex ante rules may not replicate.[27]

Second, the settlements produced concrete, enforceable commitments with clear timelines. By contrast, ex ante regimes often generate open-ended obligations that evolve through ongoing negotiation. As Europe’s experience with the DMA illustrates, such frameworks can devolve into moving targets and perpetual compliance disputes, where “conflicts of interest will be deemed resolved only when competitors are satisfied”—an inherently unstable equilibrium disconnected from consumer welfare.[28]

Given their sweeping effects, ex ante digital-market rules should be adopted only after a clear, market-wide showing that Brazil faces systemic failures that ex post enforcement cannot adequately address. In legal terms, that showing would ordinarily rest on a sufficiently developed body of case law justifying such far-reaching intervention. That condition has not been met.[29]

To be sure, the Apple and Google cases demonstrate that CADE is actively engaged in digital markets. They show sustained attention and an ability to deliver meaningful outcomes.[30] But those outcomes were achieved through negotiated settlements, not adjudicated decisions on the merits. As a result, no infringement findings exist under Brazilian competition law in digital markets for the types of conduct that an ex ante regime would regulate.

That distinction matters. Settlements may replicate the practical effects of liability findings, but they do not establish legal precedent. The absence of adjudicated violations weakens the case for expanding Brazilian competition law into a rigid ex ante framework.

Brazil thus lacks a sufficiently developed domestic evidentiary record to justify such a regime. The recent settlements reinforce this point: they show not that current law is inadequate, but that it is capable of addressing alleged concerns through targeted enforcement. An ex ante framework would become more plausible only if experience under existing law produced a robust record demonstrating persistent enforcement failures.

CADE’s own market study underscores this gap. The Cadernos do Cade: Mercados de Plataformas Digitais reflects both the agency’s technical sophistication and the limits of its enforcement record.[31] On one hand, the study documents CADE’s growing expertise across platform markets. On the other, it confirms that past investigations have resulted in settlements or dismissals, not infringement findings.[32]

This absence of convictions does not signal institutional weakness. To the contrary, CADE remains one of the world’s most respected competition authorities. The OECD’s 2019 peer review described Brazil’s regime as “largely a success” and praised CADE’s “technical capabilities.”[33] The agency continues to demonstrate proactive enforcement. In 2024, for example, it opened investigations into acquisitions of AI startups by large technology firms below standard merger-notification thresholds, using existing tools to address emerging concerns.[34]

Proponents of the bill raise two principal counterarguments. First, some note that CADE’s Google Shopping case ended in a tie vote, resolved by the president’s casting vote, and that digital investigations can take five to seven years.[35] These concerns are real. But they do not justify creating an entirely new regulatory apparatus. The Apple and Google settlements show that CADE’s toolkit has bite.[36] If enforcement speed is the issue, the appropriate response is to allocate more resources and streamline procedures—not to build a parallel regulatory regime with its own institutional costs. Europe’s DMA experience has produced the very problems it sought to avoid: lengthy proceedings, protracted compliance negotiations, and parallel antitrust investigations.[37]

Second, supporters point out that CADE itself has endorsed the idea of ex ante regulation. In its submission to the Ministry of Finance’s consultation,[38] the agency supported a complementary framework to address “dysfunctions in digital ecosystems, such as functional and distributive failures.”[39] At the same time, CADE emphasized the need to better understand the heterogeneity of digital markets[40] and highlighted the importance of a rigorous regulatory-impact analysis. Such an analysis preceded the DMA in Europe.[41] No comparable assessment preceded Bill 4,675/2025.

CADE’s institutional views merit serious consideration. But they must be weighed against the agency’s own enforcement record. An authority that has secured DMA-like outcomes through existing tools has shown, in practice, that it does not require an entirely new regulatory layer to achieve those results. Moreover, support for ex ante regulation in principle does not validate the specific design choices embedded in Bill 4,675/2025.

Those design choices raise additional concerns. The bill would create a new superintendency, introduce a non-exhaustive list of remedies, rely on unclear economic standards, and omit an efficiency defense. These features carry significant risks for Brazil’s digital economy. Evidence from other jurisdictions underscores those risks.

C.       Evidence from the DMA: Limited Benefits and Emerging Costs

Bill 4,675/2025 does not emerge in an international vacuum. The European Union’s Digital Markets Act (DMA) has been fully applicable since May 2023,[42] giving policymakers the benefit of observed outcomes, rather than theoretical projections. The emerging evidence is not encouraging.

The most granular evidence comes from Louis-Daniel Pape and Michelangelo Rossi, who examined the DMA’s restrictions on self-preferencing in the relationship between Google Search and Google Maps.[43] In January 2024, Google removed the Maps tab and clickable links from map previews on EU search pages to comply with Article 6(5) of the DMA. Using a difference-in-differences design comparing EU and non-EU countries, the authors find that mapping-related search queries in the EU increased by more than 21%. Despite this increase, overall visits to Google Maps did not change, and competing mapping services did not gain traffic. The regulation forced users to take additional steps to reach the same destination, increasing search costs without delivering meaningful competitive benefits. As the authors concluded, “these findings indicate that the DMA had weak competitive effects, highlighting Google Maps’ dominance in a market where alternatives remain limited.”[44]

Similar patterns appeared in e-commerce. Chiara Farronato, Andrey Fradkin, and Alexander MacKay conducted a field experiment on Amazon’s private-label products.[45] They found that prohibiting Amazon from offering its own brands would reduce consumer surplus by 5.5% due to lost variety. They concluded that “[c]ontrary to concerns about self-preferencing, demoting Amazon brands in search results does not increase welfare….”[46] Moreover, “[t]he consumer benefits from Amazon brands are driven both by cost advantages and by the product variety they generate… [with] value approximately 80 percent higher than the value of random products appearing in the same searches.”[47]

Additional evidence points in the same direction. Christian Peukert et al. presented preliminary findings on “Amazon’s Shrinking Shelf.”[48] Using a difference-in-differences framework, they documented an 18% decline in the number of products displayed in search results and a 7% decline in the number of sellers per product in the EU following DMA and Digital Services Act (DSA) changes.[49] These results suggest a narrowing of consumer choice.

Consumer surveys corroborated these findings. A study by the European Center for International Political Economy (ECIPE) and the European Public Policy Partnership (EPPP), based on 3,500 respondents across seven Central and Eastern European countries, found that 39% of users reported more cumbersome online tasks after the DMA, about one-third described their experience as “less seamless and more confusing,” and 80% had little or no awareness of the DMA.[50] The study found no evidence of lower prices, improved privacy, or increased contestability. Consumer behavior remained largely unchanged: more than 70% of respondents still relied on Google Search multiple times daily, and dependence on incumbent platforms remained “as strong as ever.”[51]

A separate survey by Nextrade Group of 5,000 EU consumers reported similar results. Among heavy search-engine users, 60% said routine searches took up to 50% longer, and 42% of frequent travelers reported less useful flight and hotel results.[52] These consumer costs coincided with broader economic impacts. Carmelo Cennamo, Tobias Kretschmer, Ioanna Constantiou, and Eliana Garcés estimated that DMA provisions could reduce EU service-sector revenues by up to €114 billion annually—approximately 0.64% of total sector turnover[53]—equivalent to losses of up to €1,122 per worker, depending on digital intensity.[54]

The DMA’s enforcement trajectory reinforces these concerns. In March 2025, the European Commission issued preliminary noncompliance findings against Alphabet,[55] with sanctions likely to follow.[56] In April 2025, it imposed fines of €500 million on Apple and €200 million on Meta.[57] Both companies have appealed.[58] In January 2026, the Commission opened two sets of specification proceedings against Google, covering Android and its AI services, including Gemini.[59] These developments suggest a pattern: when initial compliance proves insufficient, regulators move from setting obligations to prescribing technical implementation.[60] Enforcement thus becomes a continuous administrative process, requiring sustained oversight and significant institutional resources.

While ex ante regulation may accelerate intervention, it also increases the risk of false positives. Such regimes can impose remedies that generate “negative consequences in terms of innovation.”[61] Because the DMA relies on “detailed and absolute obligations… that do not admit of proof to the contrary,” and applied them across platforms regardless of business-model differences, it effectively “dispens[es] with the investigation itself… in a manner that is close to summary justice,” potentially penalizing conduct that benefits competition.[62]

Beyond immediate consumer effects, ex ante regulation appears to chill innovation. Apple delayed the European rollout of several features—including AirPods live translation, iPhone mirroring, and Apple Maps enhancements—citing risks to security, privacy, and user experience.[63] Meta postponed the EU launch of Threads due to uncertainty about data-integration rules.[64] Google reported that compliance with overlapping frameworks—including the DMA, Digital Services Act, and AI Act—has delayed product launches by up to a year,[65] particularly for AI-integrated services.[66] These delays function as an “innovation tax” on consumers in regulated jurisdictions—with costs that are real but diffuse, and unlikely to generate political backlash.[67]

Evidence from earlier European digital regulation reinforced these concerns. Mert Demirer, Diego Jiménez Hernández, Dean Li, and Sida Peng found that the GDPR reduced data storage by 25.7% and data processing by 15.4% among EU firms relative to U.S. counterparts, consistent with a 22.4% increase in data costs.[68] Smaller firms bore disproportionately higher burdens. The GDPR thus operates less as a neutral governance tool than as a tax on data-intensive activity.

Jian Jia, Ginger Zhe Jin, and Liad Wagman found that venture-capital investment in European technology startups declined by 26.1% in the year following the GDPR’s implementation.[69] Subsequent work found these effects persisted for at least 2.5 years.[70] The GDPR thus appears to have dampened startup financing and growth prospects.

Layering the DMA’s behavioral obligations onto this framework was bound to compound these effects. As prior research noted, “the GDPR experience demonstrates how sweeping regulation may create barriers to entry or encourage market exit among small tech firms.”[71]

The broader macroeconomic context underscores these concerns. In a widely cited 2024 report, former European Central Bank President Mario Draghi identified the technology sector as the primary driver of the growing productivity gap between the European Union and the United States.[72] He concluded that Europe lagged in the very technologies that will shape future growth.[73]

Draghi attributed this gap, in significant part, to regulatory constraints:

The problem is not that Europe lacks ideas or ambition. We have many talented researchers and entrepreneurs filing patents. But innovation is blocked at the next stage: we are failing to translate innovation into commercialisation, and innovative companies that want to scale up in Europe are hindered at every stage by inconsistent and restrictive regulations.[74]

Europe’s innovation gap is stark. Only four of the world’s 50 largest technology companies are European.[75] The region lacks counterparts to firms such as Google, Apple, Amazon, Meta, OpenAI, and Anthropic. These structural weaknesses may deepen if the DMA reduces startup exit opportunities[76] or weakens core revenue models, such as digital advertising.[77]

Causation is complex. But the correlation between Europe’s regulatory posture and its innovation deficit is difficult to ignore. Jurisdictions considering similar frameworks should weigh these tradeoffs carefully.

D.       Elevated Risks in the Brazilian Context

The foregoing evidence from Europe is concerning on its own terms. The risks of importing this model are even greater in Brazil. Institutional capacity constraints, existing regulatory burdens, opportunity costs, and geopolitical exposure all amplify the potential downsides of ex ante regulation.

Institutional capacity presents the most immediate concern. Brazil ranks 78th of 143 countries on the World Justice Project Rule of Law Index[78] and 124th on ease of doing business.[79] These rankings suggest that broad regulatory discretion is a particularly risky policy tool for Brazil. Digital antitrust regulation is especially vulnerable to capture because “the extremely high stakes will increase the incentives for incumbent digital firms to use regulation to protect the economic status quo.”[80] It also creates risks of politically motivated enforcement, reduced foreign investment, and slower long-term growth.[81] These concerns carry added weight in Brazil, which ranks 107th of 181 countries on Transparency International’s Corruption Perceptions Index.[82]

Even well-resourced regulators face capacity constraints. The European Commission—despite its institutional advantages—has filled only 19 of 80 planned DMA positions, a 76% shortfall, as Director-General for Competition Olivier Guersent has acknowledged.[83] CADE is likely to face even greater challenges, particularly because Bill 4,675/2025 does not provide for additional resources.[84]

The Ministry of Finance has stated that the proposed Digital Markets Superintendency (SMD) would be created through internal reorganization, without additional funding.[85] That claim implies one of two outcomes. Either the new regime will be under-resourced, increasing the risk of low-quality decisions, or it will divert scarce personnel from CADE’s core antitrust functions. Neither outcome is desirable. Ex ante regimes require specialized expertise that goes beyond traditional competition law and economics. Monitoring and enforcing design-based obligations demands capabilities in data science, engineering, and digital product development. A simple reallocation of existing staff is unlikely to meet these demands.

Brazil’s broader regulatory environment compounds these concerns. The country has long struggled with the “Custo Brasil,” which reflects structural inefficiencies that raise the cost of doing business, deter investment, and suppress entrepreneurship. [86] A 2019 study by Movimento Brasil Competitivo, conducted with the Ministry of Development, Industry, Trade, and Services, estimated that these inefficiencies imposed an annual burden of roughly R$1.7 trillion (about $330 billion), or 19.5% of GDP in 2022. “This means that producing and operating in Brazil costs R$1.7 trillion more per year than in countries belonging to the Organization for Economic Cooperation and Development (OECD).”[87] These costs stem not only from regulatory distortions, but also from tax complexity, infrastructure gaps, bureaucracy, and legal uncertainty.[88]

International benchmarks reinforce this diagnosis. OECD Product Market Regulation indicators for 2023–24 rank Brazil as the fifth-worst of 51 countries in economy-wide regulatory burden.[89] The OECD highlights the need to simplify administrative requirements, reduce retail price controls, and strengthen mechanisms to assess the competitive impact of regulation.[90] These priorities point toward regulatory simplification—not expansion.

A 2022 OECD review reached a similar conclusion: Brazil’s regulatory challenges reflect structural weaknesses in the design, coordination, and evaluation of policy.[91] The report emphasized persistent gaps in regulatory-impact assessment, stakeholder consultation, and ex post evaluation. These findings suggest that Brazil should rationalize its regulatory framework before adding new layers. At a minimum, any new regulation should meet a high evidentiary threshold and demonstrate clear net benefits for consumers.

Opportunity costs further weaken the case for ex ante intervention. As Mario Zúñiga argues, competition concerns are not the most pressing policy priority in Latin America. Greater gains would come from reforms that reduce entry barriers, expand telecommunications infrastructure, and improve public goods such as education and legal certainty.[92]

A January 2026 ITIF report makes a related point: countries should calibrate regulatory stringency to their level of economic development. For emerging economies, flexible and interoperable frameworks are preferable to rigid EU-style rules.[93] Brazil’s priority should be attracting investment, not imposing compliance-heavy regimes that may deter entry.[94]

Geopolitical considerations add another layer of risk. Digital-platform regulation tends to fall disproportionately on U.S.-based firms. Critics have characterized the DMA as a “non-tariff attack”[95] on U.S. technology companies.[96] Of the seven DMA-designated gatekeepers, five are U.S. firms: Alphabet, Amazon, Apple, Meta, and Microsoft.[97] Regardless of intent, ex ante regimes de facto concentrate regulatory burdens on American companies.

The geopolitical environment has shifted significantly. Since 2025, the United States has treated foreign digital regulation as a matter of trade policy. The Trump administration has characterized such measures as “unfair taxes” or “overseas extortion.” In February 2025, President Donald Trump directed the Office of the U.S. Trade Representative to consider tariffs and other responses to foreign regulatory actions deemed discriminatory.[98]

Senior officials have reinforced this position. Vice President J.D. Vance criticized the DMA, the DSA, and the GDPR as barriers to innovation.[99] In December 2025, Secretary of State Marco Rubio imposed visa restrictions on European officials linked to digital regulation, signaling a willingness to escalate beyond trade measures.[100]

Brazilian officials have downplayed these risks. Finance Minister Dario Durigan has argued that U.S. antitrust enforcement against digital platforms suggests Brazil need not fear retaliation. That comparison is incomplete.[101] Traditional ex post enforcement differs fundamentally from a standing, designation-based regime of ongoing obligations. The United States may not treat these approaches as equivalent.

Durigan also emphasized the bill’s formal neutrality.[102] But formal neutrality does not eliminate geopolitical exposure. Even country-neutral rules may disproportionately affect U.S. firms and be perceived accordingly.

The assertion that EU tech regulation is discriminatory is not merely a complaint from American boardrooms; it is supported by the design and application of the laws themselves. While European officials vehemently deny protectionist intent, claiming that the rules apply to all, the empirical reality tells a different story of targeted enforcement and “gerrymandered” criteria.[103]

Brazil retains full sovereignty over its regulatory choices. But given the current geopolitical environment, ignoring these risks would be imprudent.

A final observation on the bill’s claimed “flexibility” is warranted. As noted, the Brazilian approach is presented as more flexible and adaptive than the DMA and therefore less susceptible to its rigidity.[104] In the Ministry of Finance’s post-consultation report, the DMA serves as an important reference point, but the Ministry acknowledges that many submissions criticized the DMA model as overly rigid.[105] It therefore considers alternative approaches and ultimately recommends a “new hybrid regulatory system” that is “more flexible and adaptable,”[106] based on designating systemically relevant platforms and tailoring obligations on a case-by-case basis.[107]

That framing, however, does not resolve the underlying concerns. The bill, as submitted, still includes a non-exhaustive menu of obligations, granting CADE broad discretion to impose remedies on designated firms. Even when company-specific obligations are described as more tailored, they can, “over time,” “become a powerful force for locking in incumbents and erecting barriers to new entrants, paradoxically harming competition.”[108]

As Mario Zúñiga argues, the complexity of proving abuse-of-dominance claims serves an important function. It acts as a filter that helps agencies distinguish harmful conduct from potentially beneficial practices. This is a feature, not a bug, of case-by-case enforcement—one that an ex ante regime risks displacing.[109]

Recent experience in other jurisdictions reinforces this concern. Even where authorities formally rely on ex post enforcement, regulators have increasingly adopted sweeping remedies without establishing concrete consumer harm. South Africa’s Competition Commission has recommended that e-commerce firms separate retail and marketplace operations. Mexico’s COFECE has proposed that Amazon and Mercado Libre unbundle streaming services. The UK CMA has preliminarily advanced broad, DMA-like remedies based solely on strategic market status.[110]

In this context, “flexibility” is not necessarily a safeguard. When combined with broad discretion and weak limiting principles, it becomes a vehicle for expansive and unpredictable intervention—one that, at a minimum, undermines legal certainty and increases the risk of error.

III.     Substantive and Institutional Defects in Bill 4,675/2025

Even if one were to accept that existing Brazilian competition law is insufficient for digital markets—a premise that, as we have argued, is contradicted by CADE’s enforcement record and unsupported by empirical evidence from jurisdictions that moved first—the bill’s specific provisions would still raise serious concerns. This section therefore turns to the design and substance of Bill 4,675/2025.

The analysis does not attempt to catalog every provision. Instead, it focuses on those elements that are most consequential and that, if adopted, would require modification to better serve Brazilian consumers and the broader economy.

Five concerns stand out.

First, the bill’s guiding objectives—reduction of barriers to entry, protection of the competitive process, and promotion of freedom of choice—lack the consumer-welfare anchor that constrains enforcement under the BCL. This risks granting regulators broad discretion without a clear benchmark for evaluating outcomes.

Second, the designation criteria conflate market reach with market power, relying on structural proxies that risk capturing firms whose size reflects competitive success rather than durable dominance.

Third, the bill’s institutional design introduces governance gaps, including risks of fragmentation, resource constraints, and strategic use of the regime by private actors.

Fourth, the bill’s per se prohibitions and affirmative mandates risk condemning conduct that is often procompetitive, while creating unresolved tensions with security, privacy, and investment incentives.

Fifth, the absence of a meaningful efficiencies defense further distances the regime from the BCL and increases the likelihood of costly enforcement errors.

The subsections that follow examine each of these concerns in turn.

A.       Objectives and Limiting Principles

1.       The Consumer Welfare Standard and Brazilian Competition Law

The Brazilian government chose to frame Bill 4,675/2025 as an amendment to Brazil’s Competition Law (BCL)—similar to the UK’s DMCC and Germany’s Section 19a of the GWB—rather than as a standalone regulatory regime like the EU’s DMA. The bill thus embeds new digital-market powers within Brazil’s existing antitrust framework.

That design choice carries important implications. It situates ex ante digital regulation within the institutional and doctrinal boundaries of the BCL, importing both its strengths and its constraints. Any assessment of the bill must therefore consider its consistency with the underlying logic of Brazilian competition law.

Although the BCL does not include an explicit statement of purpose, its provisions—read systematically—reveal a coherent set of guiding principles. Article 1 establishes the law’s constitutional foundations:

This Law structures the Brazilian Competition Defense System (SBDC) and provides for the prevention (merger control) and repression of violations against the economic order, guided by the constitutional principles of freedom of initiative, free competition, the social function of property, consumer protection, and the repression of the abuse of economic power.[111]

These principles, drawn from Article 170 of the Brazilian Federal Constitution, do not provide an operational test for enforcement. They function instead as constitutional anchors that must be translated into practice through CADE’s case law and administrative decisions. The sole paragraph of Article 1 further clarifies that “the collectivity (society) is the holder of the legal rights protected by this Law.”[112]

Articles 4 and 6 reinforce the institutional structure. They define CADE as a “judicial entity… constituted as a federal autonomous agency”[113] and an “adjudicating body.”[114] This structure distinguishes Brazil from the United States. CADE acts as an administrative adjudicator, not as a prosecutorial authority that must seek judicial approval in an adversarial system.

Article 36 provides greater analytical clarity. Its § 1 introduces a central limiting principle:

Achieving dominance in a market by natural process and by being the most efficient economic agent in relation to competitors does not characterize the violation set forth in item II [“to control the relevant market for goods and services”] of the head provision of this article.[115]

This provision anchors Brazilian competition policy. Dominance is not unlawful per se. Market power—whether large or monopolistic—is permissible if it results from efficiency, innovation, or superior performance. In such cases, competitive success reflects consumer preference, not harm.

That logic implies a further requirement: enforcement must demonstrate harm to consumers. CADE must show that conduct leads to higher prices, reduced output, lower quality, or diminished innovation before it can intervene.

This framework aligns closely with the consumer-welfare standard.[116] While formulations differ in the literature, the standard consistently rejects the idea that market power achieved through superior performance constitutes a violation of antitrust law. Intervention is justified only when conduct harms consumers, not competitors.

The consumer-welfare principle is even clearer in merger control. Article 88 prohibits mergers that eliminate competition or create or reinforce dominance, [117] but it allows approval subject to a key limitation:

The mergers referred to in § 5 of this article may be authorized, provided that the limits strictly necessary to achieve the following objectives are observed:

I — cumulatively or alternatively:

1. a) to increase productivity or competitiveness;
2. b) improve the quality of goods or services; or
3. c) promote efficiency and technological or economic development; and

II — that a significant portion of the resulting benefits are passed on to consumers.[118]

Thus, even mergers that raise competitive concerns may proceed if efficiencies outweigh harms and consumers receive a “significant portion” of the benefits. As in United States v. Baker Hughes,[119] this structure implicitly shifts the burden of proof: once CADE establishes a prima facie concern, merging parties may rebut it by demonstrating verifiable efficiencies.

In practice, the BCL evaluates conduct through its effects on consumers—closer to Herbert Hovenkamp’s interpretation of the standard[120] than to Robert Bork’s total-welfare approach.[121]

CADE’s 2016 Horizontal Merger Guidelines confirm this reading. They adopt a “non-negative net effect” standard:[122] mergers must be approved if they do not harm consumer welfare and blocked if they do.

CADE’s case law reinforces the same conclusion. As Eric Hadmann Jasper observes:

An analysis of Brazilian competition law and CADE’s regulations and documents indicates, at least at this stage of the research, a diffusion of national antitrust principles/purposes (i.e., freedom of initiative, free competition, social function of property, consumer protection/consumer welfare, repression of abuse of economic power, efficiency, and protection of the competitive process) and a slight primacy of consumer welfare, at least with regard to the analysis of mergers. Finally, an examination of CADE precedents reveals a diffuse list of purposes, with emphasis on “consumer welfare” (6 mentions, including the expression “maximization of economic value to the consumer”), protection of competition (3 mentions), protection of markets (3 mentions), efficiency (2 mentions), economic welfare (2 mentions), and social effects (2 mentions).[123]

Gustavo Manicardi Schneider and Rodrigo Fialho Borges reach a similar conclusion:

Among these documents (those that defined consumer welfare), two directly adhered to Hovenkamp’s standard of consumer welfare, one directly mentioned Bork but emphasized allocative efficiency, and 12 adopted the definition of consumer welfare contained in CADE’s H Guidelines.[124]

Consumer welfare is therefore not an abstract concept in Brazilian law. It is the operative standard embedded in Articles 36 and 88, CADE’s guidelines, and its administrative practice. Enforcement focuses on consumer outcomes—not on protecting individual competitors.

2.       Article 47-B’s Departure from Consumer Welfare

Against this backdrop, Bill 4,675/2025 introduces a potential conflict. Article 47-B identifies three objectives for digital regulation: (i) reducing barriers to entry; (ii) protecting the competitive process; and (iii) promoting freedom of choice.[125]

These goals may appear unobjectionable. But they lack limiting principles tied to measurable welfare outcomes. As a result, they risk expanding enforcement beyond economically grounded standards and increasing false positives.

a)         Reduction of barriers to entry

Entry barriers play an important role in traditional antitrust analysis. CADE evaluates them in merger review[126] and unilateral conduct cases[127] as part of a broader effects analysis. High barriers may enable firms to raise prices, reduce output, or suppress innovation.[128]

The bill changes that role. It transforms barriers to entry from an analytical factor into a regulatory objective.

That shift is consequential. Under current law, CADE evaluates barriers as one element among many. It does not seek to eliminate them as an end in itself. The bill’s approach departs from this framework.

Economic theory supports caution. While potential entry can discipline firms,[129] artificially increasing contestability does not necessarily improve outcomes. In markets characterized by network effects and scale economies, intervention may simply redistribute rents without benefiting consumers:

[W]here network effects or scale economies predominate (as is always the case with digital platforms), enhanced contestability by policy is most likely to redistribute rents but not necessarily to serve consumers or create competition. ‘[I]f the market cannot profitably accommodate another entrant, due to scale economies and the nature of the oligopoly interaction, entry will be followed by some firm’s exit and another period of high prices.’ If all increased contestability does is to replace one gatekeeper with another, it can hardly be said to improve competitive outcomes.[130]

Moreover, this same economic theory suggests that artificially expanding contestability—by removing barriers to entry—in an effort to increase the number of competitors beyond an existing oligopoly (or even duopoly) “may have little effect on outcomes because such markets already behave as if they are competitive.”[131] To the extent that an explicit objective of entry-barrier reduction rests on the assumption that markets are inherently more competitive when they have more entrants, that assumption lacks clear economic justification.

While facilitating entry may improve outcomes in cases of pure monopoly—where no meaningful prospect of entry exists—the same does not hold in the typical “gatekeeper” market. In such markets, a dominant platform may hold a large and durable market share, yet still face—and at times successfully repel—the threat of entry. In those settings, artificially enhanced entry may do little to promote competition. Incumbency advantages often stem from the firm’s ability to deliver the benefits of scale or scope to consumers, rather than from constraints imposed on rivals. In such cases, policies aimed at lowering entry barriers risk supporting competitors rather than competition or consumer welfare.[132]

In sum, barriers to entry may reflect genuine market failures, but they may also arise from economies of scale, quality investments, network effects that benefit consumers, or the natural advantages of a superior product.[133] A regime that treats barrier reduction as an end in itself—without requiring that such reductions improve consumer welfare—risks mandating access, unbundling, or interoperability in contexts where the very barrier at issue underpins the product’s value.

As the foundational error-cost literature emphasizes, reducing barriers that result from efficient conduct can impose costs that are difficult to reverse, particularly in technology markets.[134]

Elevating barriers to entry from a diagnostic variable to an explicit regulatory objective therefore represents a conceptual overreach. There is no clear economic or legal justification for treating their suppression as an end in itself.

b)         Protection of the competitive process

The second objective—“protection of the competitive process”—is conceptually vague and analytically weak. Brazilian law does not define the term, and comparative experience offers little guidance.

As Nicolas Petit and Lazar Radic observe:

The “protection of the competitive process” is similarly redundant. All acts of bad conduct spelled out in antitrust statute epitomize “distortions of the competitive process”. Collusion removes independence from competitors and monopolization eliminates rivalry. Both standards add nothing to text law and are thus circular, with Herbert Hovenkamp calling them “slogans.”[135]

Conceptually appealing though it may be, “protection of the competitive process” suffers from the same measurability problem that plagues the DMA’s fairness objective.[136] Without clear limiting principles or a defined analytical methodology, the concept risks being interpreted in almost limitless ways.

This indeterminacy creates broad discretion for decision-makers. It encourages competitors to bring self-aggrandizing complaints, undermines legal predictability for businesses, and increases the likelihood of arbitrary enforcement—including enforcement that shields less efficient competitors from market exit.

Such open-endedness is particularly problematic in an administrative-adjudication setting, where clarity and foreseeability are essential to the rule of law and to effective business compliance.[137]

c)         Promotion of freedom of choice

The third objective—“freedom of choice”—is similarly problematic. It lacks a clear connection to measurable welfare outcomes and may conflict with consumer-welfare analysis:

The fatal flaw in the consumer choice standard is that it simply, indeed simplistically, rejects economic analysis of consumer preferences as the fundamental guiding principle of antitrust analysis, including the preferences consumers express in making unavoidable tradeoffs between price and nonprice values. The consumer choice standard rejects even the view that the role of antitrust is to protect the competitive process as one that produces desirable outputs (i.e., consumer welfare) in favor of an antitrust regime that analyzes nonprice competition as a standalone and inviolable virtue.[138]

While product variety and consumer choice may, in certain contexts, correlate with competitive intensity, they are not reliable proxies for consumer welfare. As a practical matter, the promotion of choice over consumer welfare is rooted in structural presumptions that assume a causal relationship between the number of firms or brands in a market and consumer welfare. But “[t]hese presumptions have no basis in modern economic theory, are not supported by empirical evidence, and are likely to provide misleading answers to the very questions concerning nonprice competition and innovation that the choice standard was designed to address.”[139]

Moreover, a market offering a wide array of products does not necessarily generate greater consumer benefit if most options are inferior or overpriced. Conversely, markets dominated by a smaller number of high-quality, efficient products may yield higher welfare, even where choice is more limited. Consumer preferences—not the sheer number of options—ultimately determine the connection between products and consumers’ needs. Competition can therefore narrow choice when consumers gravitate toward superior products, and this natural process should not be mistaken for harm.[140]

Empirical analysis of CADE’s practice supports this interpretation. Borges and Schneider find that “freedom of choice,” along with “information made available to consumers,” appears only once as a proxy for consumer welfare in CADE’s decisions, whereas “price” appears 94 times—by far the most common indicator. [141] Price, quality, quantity, and innovation remain the key measurable variables through which CADE operationalizes the consumer-welfare standard. “Freedom of choice,” by contrast, is conceptually diffuse and analytically weak.

d)         Remedying these issues

Taken together, Article 47-B’s objectives depart from the consumer-welfare framework embedded in the BCL. Even when they align in particular cases, they substitute clear, economically grounded analysis with vague and potentially misleading proxies.

Two conclusions follow. First, the objectives lack measurable standards capable of guiding decision-making. In practice, enforcers will likely revert—explicitly or implicitly—to consumer-welfare analysis. Second, because the bill amends the BCL rather than creating a separate regime, the consumer-welfare standard remains the governing benchmark.

Consumer welfare, in other words, must prevail over the bill’s alternative objectives.

The remedy is straightforward. The bill should require that any obligation imposed under the digital-markets regime demonstrably advance consumer welfare. This would preserve flexibility while ensuring that interventions target harm to consumers—not merely harm to competitors. It would also align the new framework with the core principles of Brazilian competition law and limit the risk of rent-seeking and overenforcement.[142]

B.        Designation Criteria and the Risk of Regulating Size Rather Than Harm

The bill’s designation mechanism, set out in Article 47-C, operates through two tiers: qualitative criteria and revenue thresholds.[143] The qualitative criteria include multi-sidedness, network effects, vertical integration into adjacent markets, a strategic intermediary position for third-party business users, access to significant personal or commercial data, a large base of business and end users, and a portfolio of multiple digital products or services[144]

Paragraph 2 establishes revenue thresholds of R$50 billion in global annual gross revenue or R$5 billion in Brazil, with authority granted to the ministers of finance and justice to adjust these thresholds by joint act.[145] Designation applies to the entire economic group and lasts up to 10 years, renewable through a new proceeding.[146]

1.       Regulating Size Rather Than Demonstrable Market Power

The central problem with this framework is that it targets characteristics that are not only common, but often efficiency-enhancing and procompetitive.

Multi-sidedness is the defining economic structure of platform businesses; it is a feature, not a pathology.[147] Vertical integration produces well-documented efficiencies, including the elimination of double marginalization.[148] Broad product portfolios may reflect economies of scope or dynamic capabilities that generate consumer benefits across services. Access to data often reflects product quality and consumer demand, not exclusionary conduct.[149]

By treating these features as triggers for a decade-long regulatory designation, the bill risks converting business success into a regulatory offense.

At the same time, the framework omits key indicators of actual competitive constraint. A designation regime grounded in competition economics would instead assess factors such as the presence of potential competition, the degree of substitutability, exposure to technological disruption (including AI-driven entry), constraints from offline competitors, and countervailing power from users or suppliers. These factors better capture whether a firm can exercise market power.

The bill’s terminology reinforces this misalignment. The concept of “systemic relevance” points to size and economic importance rather than competitive dynamics. This framing suggests that firms are targeted because they are large or central to the digital economy—not because they can harm competition. A more appropriate standard would require “substantial market power” or “competitive significance,” anchoring designation in established antitrust principles.

Revenue thresholds compound the problem. As William Landes and Richard Posner showed, firm size and market share are poor proxies for market power. Market power depends on demand elasticity and supply-side constraints, including the ability of rivals and potential entrants to respond.[150] Harold Demsetz likewise demonstrated that large firm size often reflects efficiency, particularly in innovation-driven sectors where scale enables sustained investment.[151]

The government estimates that five to 10 firms—overwhelmingly U.S.-based technology companies—would meet the designation criteria.[152] This underscores that the regime targets a narrow set of global firms based on size thresholds,[153] rather than demonstrated competitive harm in Brazil.[154]

The problem is further compounded by the bill’s application to the entire economic group (todo o grupo econômico), rather than to specific services where competitive concerns arise.[155] This risks regulating activities unrelated to any identified harm, imposing costs disconnected from any coherent theory of competition.

These features raise serious error-cost concerns. As Frank Easterbrook emphasized, false positives—erroneously condemning or deterring procompetitive conduct—are especially costly in dynamic markets.[156] Geoffrey Manne extends this analysis to digital platforms, where integration and coordination often generate consumer benefits, and where regulatory error can chill innovation.[157]

A more coherent approach would require a demonstration that a firm possesses “lasting market power that cannot be disciplined by actual or potential competition within a reasonable period of time.” Relevant indicia would include substantial entry barriers, evidence of consumer harm, limited competitive pressure from rivals, and weak responsiveness of market participants.

Comparative frameworks highlight these shortcomings. The UK’s DMCC requires a finding of “substantial and entrenched market power” and “strategic significance.”[158] Germany’s Section 19a GWB requires a formal determination that a firm holds a dominant position before obligations attach.[159]

The inclusion of “multiple digital products or services” as a designation factor is particularly counterproductive. It risks deterring one of the most important sources of competition—entry by large firms into adjacent markets. Cross-market competition often represents the most meaningful constraint on incumbents. Treating such expansion as evidence of regulatory concern contradicts the bill’s stated objective of reducing entry barriers.[160]

The 10-year designation period raises additional concerns. Digital markets evolve rapidly. David Evans documents substantial churn in platform leadership over periods shorter than a decade.[161] A designation lasting until 2037 could outlive the competitive conditions that justified it.

While Article 87-B, § 3 allows review of specific obligations, it does not provide for periodic reassessment of the designation itself. This rigidity contrasts with other regimes. The DMA requires review at least every three years, the DMCC sets a five-year designation period, and Germany’s regime also limits designation to five years.

Brazil’s framework is therefore broader and less constrained: it combines open-ended qualitative criteria, adjustable thresholds, and a long designation period.

Introducing stronger limiting principles would materially improve the regime. The bill should require a finding of durable market power, limit designation to specific services rather than entire corporate groups, and mandate periodic review at shorter intervals—such as three to five years.

C.       Institutional Design and Governance Risks

1.       The New Superintendency and Its Oversight Gaps

The bill creates a new Digital Markets Superintendency (SMD) within CADE, led by a superintendent nominated by the president and confirmed by the Senate, serving a two-year term renewable once.[162] The SMD’s remit is expansive. It will monitor digital markets, request information, initiate and investigate designation proceedings, impose and oversee compliance with special obligations, pursue noncompliance proceedings, and publish an annual priorities agenda subject to approval by CADE’s Tribunal and in consultation with the Ministry of Justice.[163] In practice, the SMD would serve as the administrative engine of the ex ante regime.

A closer examination of the SMD’s architecture raises broader governance concerns. These include how jurisdiction will be allocated, who may trigger proceedings, whether CADE will have sufficient resources to administer the regime, and how the new body will interact with existing sectoral regulators. The issue is not whether CADE should develop specialized expertise in digital markets—which may well be justified—but whether the bill does so in a manner consistent with CADE’s institutional design, coherent in its allocation of powers, and sufficiently constrained to avoid duplication, strategic use by private actors, and institutional uncertainty.

Four concerns arise.

a)         Institutional fragmentation

First, the creation of a second superintendency within CADE risks institutional fragmentation and mission duplication. Understanding this concern requires situating the SMD within CADE’s existing structure.

CADE is composed of three bodies: the Administrative Tribunal, the General Superintendence (SG), and the Department of Economic Studies (DEE). The SG is the agency’s operational core. It investigates mergers, cartels, and unilateral conduct, issuing decisions and recommendations that may be final in straightforward cases or referred to the Tribunal for adjudication. The SG is headed by a single General Superintendent, appointed by the president and confirmed by the Senate for a two-year term, renewable once. The Tribunal, composed of six commissioners and a president serving four-year nonrenewable terms, adjudicates complex matters, including condemnations, merger remedies, and transaction prohibitions. The DEE provides economic analysis to support both the SG and the Tribunal.

The separation between the SG and the Tribunal reflects a clear institutional logic. The SG serves as the investigative arm, while the Tribunal provides collegiate adjudication. Creating an additional investigative authority that would likewise refer cases to the same Tribunal duplicates this structure. A more coherent alternative would be to establish a specialized digital-markets unit within the existing SG.

This concern is not merely theoretical. The SG already investigates anticompetitive conduct in digital markets, as demonstrated by the Google and Apple cases. A parallel superintendency creates risks of jurisdictional overlap, inconsistent enforcement standards, and competition for resources and institutional prominence.

Comparative experience reinforces this point. The EU administers the DMA within the European Commission through coordination between DG COMP and DG CONNECT.[164] The UK houses its digital-markets regime within the CMA through the Digital Markets Unit (DMU) and a board-level committee.[165] Germany enforces Section 19a within the Bundeskartellamt through existing divisions supported by a specialized digital unit.[166] The common pattern is specialization within existing structures, not duplication alongside them.

Article 14-B, § 1 attempts to address jurisdictional overlap by transferring unilateral-conduct cases involving designated firms from the SG to the SMD. This provision seeks to ensure coherence between special obligations and conduct investigations. It does not, however, resolve the broader inconsistency between cases involving designated firms (handled by the SMD) and those involving non-designated firms (handled by the SG). That asymmetry would not arise if digital-market expertise were integrated within the SG. A unitary structure would be more coherent, more efficient, and less prone to jurisdictional friction.

b)         Vulnerability to rent-seeking

Second, the bill’s standing provisions create significant vulnerability to rent-seeking. Proceedings for designation or the imposition of special obligations may be initiated by “any interested party,”[167] by CADE’s Tribunal or SG,[168] or by certain public bodies that trigger immediate initiation.[169]

In a regime where designation carries decade-long obligations and substantial penalties, this open-ended standing creates a powerful incentive for strategic complaints. Commercial rivals, trade associations, or politically connected entities may use the process to impose costs on competitors.

This risk is not hypothetical. The consultation process already revealed that some proponents of ex ante regulation are themselves market participants with interests adverse to likely designated firms. Match Group, for example, advocated an interventionist regime targeting app-store gatekeepers, particularly Apple. Regardless of the merits of its arguments, the alignment is predictable. Firms that are large but below the designation threshold have strong incentives to support rules that impose asymmetric burdens on larger rivals.

The statutory framework should account for this dynamic. Without tighter guardrails, the regime risks becoming a vehicle for competitor-driven regulatory pressure, rather than a tool targeted at demonstrated harm to competition and consumers.[170]

c)         Resource constraints

Third, the resource question remains unresolved. The Ministry of Finance has asserted that the SMD can be funded through internal reallocations, without additional appropriations.[171] That claim is difficult to reconcile with the bill’s scope.

Administering the regime will require managing simultaneous designation proceedings, obligation-setting, compliance monitoring, and enforcement actions across multiple global firms. These tasks involve complex technical questions, including interoperability, data governance, security architecture, and algorithmic design. They require specialized expertise that goes well beyond traditional competition law and economics.

Even the European Commission, with significantly greater resources, has acknowledged capacity constraints in implementing the DMA.[172] An under-resourced regime will either produce superficial analysis or lead to prolonged proceedings—undermining the claimed advantage of ex ante regulation as a faster alternative to ex post enforcement.

d)         Unclear allocation of regulatory responsibilities

Finally, the bill’s provisions on coordination with sectoral regulators remain vague. Article 47-F calls for cooperation but does not clearly allocate responsibilities.[173]

Memoranda of understanding could partially address this ambiguity, but relying on future agreements introduces uncertainty. Greater clarity should be embedded in the statute itself. Clear lines of authority are essential to avoid duplication, conflict, and gaps in enforcement, particularly in complex digital markets where multiple regulators may claim jurisdiction.

D.       Prohibited and Mandated Practices

1.       The Perils of Per Se Rules in Complex Markets

Article 47-E(IV) creates a non-exhaustive list of prohibitions that CADE may impose on designated firms. These include self-preferencing (favoring own offers or services), tying, restrictions on third-party access, anti-steering provisions, and “predatory or abusive strategies.”[174] This subsection focuses on the three most consequential categories, with particular attention to self-preferencing, which illustrates the broader shortcomings of ex ante obligations.

a)         Self-preferencing

The economic literature does not support treating self-preferencing as a per se offense. “The notion that platform entry into competition with edge providers is harmful to innovation is entirely speculative… there is certainly no basis for a presumption of harm.”[175]

A categorical prohibition would invert the well-established principle that vertical arrangements are generally efficient.[176] Indeed,

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.[177]

In digital markets, such a prohibition would constrain platforms’ ability to refine user interfaces and integrate services, potentially foreclosing conduct that generates substantial consumer value while increasing the risk of overenforcement—a textbook Type I error.

A proper analysis must distinguish between exclusionary self-preferencing and legitimate business conduct. The former requires durable market power, substantial foreclosure, and demonstrable consumer harm. The latter includes efficiency-enhancing practices such as eliminating double marginalization, enforcing quality control, reducing search costs, and maintaining ecosystem security.[178] As the literature emphasizes, “self-preferencing by dual-role platforms is not necessarily detrimental… [and] the effectiveness of policy interventions… depends largely on the type of self-preferencing and the specific environment.”[179]

Empirical evidence reinforces this conclusion. Studies show that platform integration can expand demand and stimulate complementor activity. Empirical evidence reinforces this conclusion. Studies show that platform integration can expand demand and stimulate complementor activity. Zhuoxin Li and Ashish Agarwal find that Facebook’s integration of Instagram increased demand for photography applications more broadly, benefiting third-party developers.[180] Jens Foerderer and his coauthors similarly show that Google’s introduction of Google Photos increased user attention and demand in the photography category, leading to greater complementor innovation and entry.[181] Evidence from video-game ecosystems likewise suggests that strong first-party offerings expand installed bases and enlarge opportunities for third-party developers.[182]

Restrictions on self-preferencing can generate welfare losses. Chiara Farronato, Andrey Fradkin, and Alexander MacKay show that removing Amazon’s private-label products leads to a 5.5% decline in consumer surplus, driven largely by reduced variety.[183] Efforts to demote private labels do not generate welfare gains. Yusuke Zennyo’s theoretical work similarly finds that “a ban on one’s own content biases” may raise commissions and prices, harming both consumers and sellers.[184] Early evidence from the DMA also suggests that limiting self-preferencing increases user friction without improving contestability.[185]

Brazil’s own enforcement record points in the same direction. A 2025 report by the Legal Grounds Institute finds that CADE’s conviction rate for self-preferencing is only 27% over the past decade.[186] More importantly, there is no history of convictions for self-preferencing in digital markets.[187] As the report notes:

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, [which] shows a lack of certainty that is uncongenial to a per se regulation.[188]

The study concludes that making self-preferencing ex ante forbidden would be a “radical move” that inverts a practice often associated with efficiencies and consumer benefits.[189]

Self-preferencing thus illustrates the core problem with ex ante prohibitions: they risk condemning conduct that neither theory nor evidence supports treating as inherently harmful. Where effects are context-dependent and contested, categorical bans are ill-suited. Reclassifying a frequently efficiency-enhancing practice as presumptively unlawful risks eliminating benefits that CADE’s effects-based approach has historically recognized. The absence of convictions in Brazil reflects not underenforcement, but an evidence-based approach that requires demonstrated harm.

b)         Steering

Anti-steering prohibitions raise distinct concerns rooted in two-sided market economics. If platforms cannot prevent free-riding on their investments in discovery, trust, payments, and user experience, they may respond by raising prices on other margins or reducing investment.[190]

The CADE-Apple settlement illustrates the complexity of these tradeoffs. It imposed anti-steering remedies while accommodating Apple’s security and privacy concerns. That calibrated, case-specific approach is difficult to replicate through rigid ex ante prohibitions.

c)         ‘Predatory or abusive strategies’

The residual category of “predatory or abusive strategies” is exceptionally open-ended. Combined with a 10-year designation period and significant sanctions, this vagueness risks chilling legitimate innovation. Firms face moving compliance targets without clear benchmarks, creating an environment in which success can never be conclusively demonstrated while rivals can continually demand further intervention.[191]

The bill should replace this per se prohibition with an effects-based standard requiring a showing of concrete consumer harm, consistent with CADE’s existing practice. At minimum, the statute should explicitly recognize that vertical integration, self-preferencing, and related practices can be procompetitive.

2.       Interoperability, Data Portability, and ‘Free’ Access Mandates

Article 47-E(V) establishes a set of positive obligations, including data-transfer tools, “free and effective” interoperability, third-party app access, business-user data access, default-setting flexibility, complaint mechanisms, and non-discriminatory terms.[192] While CADE may “consider” information-security and legal obligations, this is a permissive standard—not a structured defense.[193] These obligations raise several concerns.

a)         Security tradeoffs

Mandating interoperability and open ecosystems increases attack surfaces and undermines secure-by-design architectures. The DMA experience already reveals increased risks related to fraud, malware, data-sharing, and weakened platform security protections.[194]

The CADE-Apple settlement again illustrates the superiority of case-by-case analysis. Apple was able to resist certain mandates on security grounds while still implementing meaningful changes—an outcome unlikely under rigid statutory rules.[195]

b)         Privacy and LGPD interaction

The interaction with Brazil’s LGPD raises additional risks. While the LGPD provides for data portability, it conditions that right on regulatory oversight and the protection of trade secrets.[196] Broad competition-law mandates for data access risk conflicting with privacy and confidentiality protections or creating inconsistent obligations between CADE and the ANPD.

The DMA illustrates this tension. Its data-sharing provisions may undermine GDPR objectives, creating tradeoffs between compliance regimes.[197] Brazil risks importing similar conflicts unless privacy and security considerations are made mandatory, not discretionary.

c)         Interoperability

The academic literature counsels caution. Marc Bourreau, Jan Krämer, and Miriam Buiten conclude that mandated interoperability may entrench incumbents, reduce incentives to innovate, and require ongoing regulatory oversight.[198] It may also reduce incentives to multi-home, undermining a key source of competitive discipline.[199]

The UK’s Ofcom similarly warns that interoperability mandates can weaken innovation incentives and enable free-riding.[200] As it notes, “a more closed approach may be necessary… to protect a fair return on investment.”[201]

d)         Data portability

Data portability mandates have shown limited effectiveness. Inge Graef, Martin Husovec, and Jeroen van den Boom show that GDPR portability remains subject to legal and practical uncertainty.[202] Peter Swire and Yianni Lagos argue that portability applies even absent market power[203] and effectively creates a per se rule where competition law would apply a case-by-case analysis.[204]

The OECD warns that portability may increase security risks and disproportionately burden smaller firms.[205] Universal mandates may also prevent certain business models from emerging, imposing unseen costs.[206]

e)         Remedying these issues

The bill should include explicit defenses based on security, privacy, fraud prevention, quality assurance, and trade secrets. Interoperability and data-access obligations should be subject to proportionality analysis, and coordination with the ANPD should be mandatory.

More broadly, “free” access mandates resemble compelled dealing. By forcing firms to share infrastructure, data, and services without compensation, they risk undermining investment incentives.[207] If firms cannot capture returns on ecosystem investments, they will reduce those investments—degrading consumer value.

3.       Mandatory Merger Notification for Designated Firms

Article 47-E(I) allows CADE to require designated firms to notify all mergers, regardless of thresholds.[208] While concerns about “killer acquisitions” are legitimate, this approach is disproportionate.

A blanket reporting requirement would strain agency resources and capture transactions with no competitive risk. It may also weaken startup exit opportunities, which are critical to venture capital investment. As Geoffrey Manne, Sam Bowman, and Dirk Auer explain, acquisitions play a central role in startup financing and innovation incentives.[209]

Increasing regulatory friction risks discouraging investment and entrepreneurship. As Gordon Phillips and Alexei Zhdanov show, acquisitions provide a key exit pathway that sustains venture capital activity.[210]

By increasing the friction and uncertainty associated with every acquisition by a designated firm, the bill may inadvertently deter the very acquisition activity that channels capital and talent into the startup ecosystem. “Put simply, acquisitions may offer an exit to early investors in cases where IPOs are not a realistic prospect, thus increasing the incentive to invest in startups in the first place; barriers to market exit have been known to slow investments.”[211]

Ex ante regulations on potential acquirers risk a “regulate the acquirers, harm the targets” dynamic: by constraining incumbent acquisitions, it may eliminate a crucial exit pathway for startups, dampen investor interest, and weaken the broader innovation ecosystem.[212]

Brazilian law already provides a solution. Article 88, § 7 allows CADE to review below-threshold transactions within one year. CADE has demonstrated its willingness to use this authority, including in recent AI-related investigations.[213] CADE’s 2024 probes into below-threshold Big Tech acquisitions of AI startups demonstrate that the authority is both willing and able to scrutinize transactions that fall outside the standard notification framework.[214]

The proposed provision therefore duplicates existing tools while imposing additional costs. A targeted “call-in” mechanism—triggered by specific risk factors and supported by published guidance—would achieve the same objectives with fewer negative effects.

E.        Defenses and the Absence of an Efficiencies Standard

1.       The Missing Efficiencies Defense

Article 47-E, § 2 provides that, when imposing special obligations on firms designated as having systemic relevance in digital markets, CADE “may consider” three limiting or justificatory factors: (i) information-security features, (ii) the firm’s existing legal and regulatory compliance duties, and (iii) product or service features that enhance the functionality of the firm’s ecosystem.[215] Two points follow.

First, the permissive “may consider” formulation offers little meaningful constraint on CADE’s discretion and provides limited legal certainty; replacing it with a mandatory “shall consider” requirement would serve as a more effective safeguard. Second, regarding the list of considerations, the provision omits the most fundamental defense in competition policy: an explicit efficiency justification.

Ideally, this requirement should also give rise to a substantive right. Designated firms should be entitled to challenge obligations before the Tribunal on the ground that the economic justification is insufficient, disproportionate, or fails to demonstrate net consumer benefit. Without such a right, the requirement risks functioning as a procedural formality rather than a meaningful constraint on regulatory discretion.

As discussed above,[216] the absence of an efficiencies defense conflicts with a foundational principle of Brazilian competition law.[217] Article 36, § 1 of the BCL provides that dominance achieved through the “natural process” of superior efficiency does not constitute an infringement. An ex ante regime embedded within the same statutory framework that denies firms the ability to invoke efficiency would create a doctrinal inconsistency. The same conduct could be lawful under general competition law but prohibited under the bill’s digital-markets provisions—an inconsistency that would generate uncertainty and invite strategic litigation.

To the extent the bill seeks alignment with the UK’s DMCC rather than the EU’s DMA, it adopts the less defensible approach. The DMA contains no general efficiencies defense. Recital 23 explicitly excludes economic-justification arguments from the designation inquiry and provides only narrow, non-economic mechanisms for suspension or exemption.[218] The DMCC includes a limited analogue in its “countervailing benefits exemption,”[219] but the five-part test is onerous. It requires firms to show that (i) the conduct generates user benefits, (ii) those benefits outweigh competitive harm, (iii) they could not be achieved without the conduct, (iv) the conduct is proportionate, and (v) it does not eliminate effective competition.[220] In practice, these conditions are so restrictive that the provision is “likely to be of little—and possibly no—practical value.”[221]

Brazil should, at minimum, adopt a provision analogous to the DMCC’s, but the better approach is to incorporate a clear efficiencies defense aligned with traditional competition law.

The absence of such a defense is not incidental. It reflects the underlying objectives of many ex ante regimes, which “prioritize fairness and a distorted notion of contestability over efficiency and consumer welfare,” protecting competitors rather than competition.[222] Indeed, excluding efficiencies defenses is often a deliberate design choice: allowing firms to justify their conduct on efficiency grounds would undermine a framework aimed at redistributing rents rather than promoting consumer welfare.[223]

If Brazil intends to remain consistent with the BCL, it should avoid replicating this design. A meaningful efficiencies defense should:

1. allow firms to demonstrate that their conduct generates efficiencies that outweigh potential harms;
2. align with Articles 36, § 1 and 88, § 6 of the BCL;
3. explicitly recognize privacy, security, and innovation as legitimate countervailing benefits; and
4. apply a proportionality standard rather than an indispensability test, ensuring the defense functions as a real safeguard rather than a theoretical right.

A related procedural gap appears in Article 87-G, which governs the Tribunal’s decisions. For designation proceedings, Article 87-G(I) requires only “the specification of the facts” supporting designation.[224] For obligations, Article 87-G(II) requires a description of the obligations and applicable fines, but not an explanation of why those obligations are warranted, proportionate, or beneficial to consumers.

This omission is significant. A decision that merely recites facts without explaining how those facts support the legal conclusion—or without addressing the economic consequences of the remedy—fails to provide the reasoned analysis necessary for judicial review or for an effective defense. At minimum, Article 87-G should require that the Tribunal articulate a reasoned economic analysis demonstrating that the designation or obligation is proportionate, that its expected benefits outweigh its costs, and that the chosen measures are the least restrictive means of achieving the bill’s objectives.

IV.     Conclusion and Recommendations

Bill 4,675/2025 is well intentioned and, in several respects, more carefully designed than the DMA. It remains premature, however, to adopt an ex ante regulatory regime for digital markets in Brazil. The bill’s foundational premise—that existing Brazilian competition law is inadequate—is contradicted by CADE’s recent enforcement record, most notably the Google and Apple settlements, which achieved outcomes similar to those contemplated by the bill through case-by-case enforcement. At the same time, emerging empirical evidence from the EU’s DMA suggests that this regulatory model has, at least in some instances, harmed consumers, chilled innovation, and failed to deliver on its promises of increased contestability. Brazil’s institutional constraints further amplify these risks, making broad regulatory discretion more hazardous than in the jurisdictions the bill seeks to emulate.

We offer three tiers of recommendations.

First, Brazil should forgo ex ante regulation and instead strengthen CADE’s existing enforcement toolkit. The Google and Apple settlements demonstrate that the current framework is capable of addressing digital-market concerns. If speed is a concern, the appropriate response is to invest in CADE’s resources, streamline procedures, and expand the use of targeted market studies and sector inquiries. Brazil should also allow more evidence to accumulate from jurisdictions that have already adopted ex ante regimes before imposing untested regulatory burdens on its own economy.

Alternative models underscore that broad, economy-wide regulation is not the only option. Japan’s Mobile Software Competition Act, which took effect in December 2025, adopts a narrower, more targeted approach[225] It applies specifically to smartphone software ecosystems, incorporates objective-justification defenses tied to cybersecurity and system stability, does not require web-based sideloading, and explicitly protects device security and user information. If Brazil ultimately determines that intervention is warranted, a more focused model of this kind is more likely to achieve its objectives without the collateral costs of a comprehensive ex ante regime.

Second, if ex ante regulation proceeds nonetheless, the bill requires substantial guardrails. Consumer welfare should be enshrined as the regime’s explicit limiting principle. The designation test should require demonstrated market power, not merely size, and the designation period should be shortened with mandatory periodic review. A meaningful efficiencies defense—aligned with Articles 36, § 1 and 88, § 6 of the BCL and calibrated to a proportionality standard—should allow firms to demonstrate that their conduct generates benefits that outweigh potential harms. The blanket merger-notification requirement should be removed, given the existing mechanism in Article 88, § 7, or replaced with a targeted call-in power. Designation should apply to specific services rather than entire corporate groups. The bill’s standing provisions should incorporate screening mechanisms to filter out strategic or frivolous complaints. Finally, the SMD should be integrated within the existing General Superintendence, rather than established as a parallel authority with separate leadership.

Third, and at a minimum, the bill should be subjected to a rigorous Regulatory Impact Assessment prior to enactment. Brazil’s Economic Freedom Act already imposes such requirements on regulatory agencies,[226] and Congress should treat this standard as a baseline for novel legislative interventions. The bill should not advance on the basis of political momentum tied to unrelated digital-policy initiatives. The stakes—for consumers, innovation, and Brazil’s position in the global digital economy—are too high to proceed without evidence-based deliberation.

Brazil has the opportunity to learn from, rather than replicate, the experience of jurisdictions that moved first. The available evidence suggests that the costs of ex ante digital regulation are real and measurable, while its benefits remain uncertain. A country that ranks 78th on the rule-of-law index and 124th on ease of doing business cannot afford to adopt regulatory models whose costs fall disproportionately on consumers, firms, and the broader innovation ecosystem. The more productive path is to strengthen the competition-law tools that have already demonstrated their effectiveness, invest in institutional capacity, and preserve the regulatory humility that sound law & economics demands.

[1] Projeto de Lei No. 4.675, de 2025, Câmara dos Deputados [hereinafter Bill 4,675/2025 or the Bill]. The bill forms part of Agenda Brasil Digital, a six-initiative package aimed at creating a safer, more competitive, and more innovative digital environment. Its centerpiece is the enactment of the Digital Child and Teenager Act (Lei No. 15.211/2025).

[2] Lei No. 12.529, de 30 de novembro de 2011, Diário Oficial da União [D.O.U.] de 1.12.2011 (Braz.) [hereinafter the BCL].

[3] A Bill 4,675/2025, art. 47-E. The provision sets out a non-exhaustive list of “special obligations” that CADE may impose on firms designated as “systemically relevant.” Because the list is expressly non-exclusive, CADE may impose virtually any remedy it considers appropriate. This broad discretion expands the agency’s authority and heightens the need for clear statutory limits, procedural safeguards, and meaningful constraints on its remedial powers.

[4] The paper also identifies the key modifications needed, if the bill is enacted, to reduce its legal, economic, and institutional risks.

[5] Dario Oliveira Neto & Mario Zúñiga, Apple in Brazil: Ex Post Antitrust Meets Ex Ante Ambitions, Truth on the Mkt. (Feb. 4, 2026), https://truthonthemarket.com/2026/02/04/apple-in-brazil-ex-post-antitrust-meets-ex-ante-ambitions.

[6] See World Justice Project, WJP Rule of Law Index 2025: Brazil (2025), https://worldjusticeproject.org/rule-of-law-index/country/2025/Brazil.

[7] See World Bank Group, Doing Business 2020: Comparing Business Regulation in 190 Economies (2020), https://openknowledge.worldbank.org/server/api/core/bitstreams/75ea67f9-4bcb-5766-ada6-6963a992d64c/content. Brazil ranked 124th out of 190 economies. The World Bank discontinued the Doing Business project in 2020 after a methodological overhaul and replaced it with the Business Ready (B-READY) initiative. Brazil has not been surveyed in either B-READY edition to date. See also World Bank Group, Business Ready (2025), https://openknowledge.worldbank.org/entities/publication/a35eb0e7-328a-46a3-8713-ce8ef8a7c9e3.

[8] See Mario Zúñiga, Parsing Brazil’s ‘More Flexible’ Approach to Digital Markets, Truth on the Mkt. (Feb. 5, 2025), https://truthonthemarket.com/2025/02/05/parsing-brazils-more-flexible-approach-to-digital-markets (noting Brazil’s low ranking on the Rule of Law Index).

[9] Jacques Crémer, Yves-Alexandre de Montjoye & Heike Schweitzer, Competition Policy for the Digital Era, Eur. Comm’n (2019), https://ec.europa.eu/competition/publications/reports/kd0419345enn.pdf; Jason Furman, Unlocking Digital Competition: Report of the Digital Competition Expert Panel, HM Treasury (2019), https://assets.publishing.service.gov.uk/media/5c88150ee5274a230219c35f/unlocking_digital_competition_furman_review_web.pdf; Stigler Ctr. for the Study of the Econ. & the State, Final Report of the Stigler Committee on Digital Platforms (2019), https://www.chicagobooth.edu/-/media/research/stigler/pdfs/digital-platforms—committee-report—stigler-center.pdf.

[10] See Filippo Lancieri & Patricia Sakowski, Competition in Digital Markets: A Review of Expert Reports, 26 Stan. Tech. L. Rev. 65 (2023) (synthesizing the common analytical framework across leading digital competition policy reports).

[11] Regulation 2022/1925 of the European Parliament and of the Council of 14 September 2022 on Contestable and Fair Markets in the Digital Sector and Amending Directives (EU) 2019/1937 and (EU) 2020/1828 (Digital Markets Act), 2022 O.J. (L 265) 1 [hereinafter DMA], https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:32022R1925; Digital Markets, Competition and Consumers Act 2024, c. 13 (U.K.) [hereinafter DMCC], https://www.legislation.gov.uk/ukpga/2024/13/contents; Gesetz gegen Wettbewerbsbeschränkungen [GWB] [Act Against Restraints of Competition] § 19a (Ger.), https://www.gesetze-im-internet.de/englisch_gwb/englisch_gwb.html.

[12] Lazar Radic, The Imaginary Antitrust Consensus, 9 Eur. Competition & Regul. L. Rev. 22 (2025).

[13] Id.

[14] See Herbert Hovenkamp, Antitrust and Platform Monopoly, 130 Yale L.J. 1952, 1955 (2021) (arguing for firm-specific antitrust enforcement over broad ex ante regulation).

[15] See Geoffrey A. Manne, Lazar Radic & Dirk Auer, Regulate for What? A Closer Look at the Rationale and Goals of Digital Competition Regulations, 22 Berkeley Bus. L.J. 201 (2025).

[16] See Projeto de Lei nº 2.768, de 2022, Câmara dos Deputados, https://www.camara.leg.br/proposicoesWeb/fichadetramitacao?idProposicao=2337417 (citing Tim Wu’s scholarship, the U.S. House Judiciary Committee’s report Investigation of Competition in Digital Markets, and the DMA in the bill’s justification section (“Justificação”)).

[17] See Tomada de Subsídios: Aspectos Econômicos e Concorrenciais de Plataformas Digitais, Ministério da Fazenda, Secretaria de Reformas Econômicas (2024), https://www.gov.br/participamaisbrasil/concorrencia-plataformas-digitais. See also Dario da Silva Oliveira Neto, Some Remarks on the Comments Submitted to the Brazilian Public Consultation on Economic and Competitive Aspects of Digital Platforms Held by the Ministry of Finance, Kluwer Competition L. Blog (June 4, 2024), https://legalblogs.wolterskluwer.com/competition-blog/some-remarks-on-the-comments-submitted-to-the-brazilian-public-consultation-on-economic-and-competitive-aspects-of-digital-platforms-held-by-the-ministry-of-finance.

[18] See Brazilian Secretariat of Economic Reforms (Ministry of Finance), Digital Platforms: Competition Aspects and Regulatory Recommendations for Brazil (Dec. 16, 2024), https://www.gov.br/fazenda/pt-br/central-de-conteudo/publicacoes/relatorios/sre/digital-platforms-competition-regulatory-recommendations-brazil-en.pdf.

[19] See Herbert Hovenkamp, Antitrust and Platform Monopoly, supra note 14, at 1970 (“Notwithstanding overwhelming evidence to the contrary, the market for digital platforms is often said to be winner-take-all. But this is rarely true.”).

[20] See Jonathan M. Barnett, Illusions of Dominance: Revisiting the Market-Power Assumption in Platform Ecosystems, 86 Antitrust L.J. 1 (2024); Christopher Yoo, Network Effects in Action, in The Global Antitrust Institute Report on the Digital Economy (Joshua D. Wright & Douglas H. Ginsburg eds., 2020), https://gaidigitalreport.com/2020/08/25/network-effects-in-action.

[21] Agenda Brasil Digital was unveiled at a political ceremony on Sept. 17, 2025. See also Dario Oliveira Neto & Lazar Radic, Brazil’s Digital Markets Bill: A DMA Through the Back Door?, Truth on the Mkt. (Sept. 24, 2025), https://truthonthemarket.com/2025/09/24/brazils-digital-markets-bill-a-dma-through-the-back-door.

[22] See Brasil, Relatório de Sistematização das Contribuições à Tomada de Subsídios nº 1/2024, Ministério da Fazenda, Secretaria de Reformas Econômicas (2024), https://www.gov.br/fazenda/pt-br/central-de-conteudo/publicacoes/relatorios/sre/relatorio-sre-tomada-de-subsidios.pdf/view.

[23] Exposição de Motivos Interministerial nº 00099/2025 MJSP AGU MF MGI, https://www.camara.leg.br/proposicoesWeb/prop_mostrarintegra?codteor=3003060&filename=PL%204675/2025.

[24] Id. at 3 (“This proposal introduces new measures that place Brazil at the vanguard of competition protection on digital platforms, alongside legal systems such as Germany, Japan, and the United Kingdom.”).

[25] See Ebazar.com.br Ltda. & Mercado Pago Instituição de Pagamento Ltda. v. Apple Inc. & Apple Services LATAM LLC, Processo Administrativo No. 08700.009531/2022-04, Conselho Administrativo de Defesa Econômica [CADE] (Braz.); In re Apple Inc. & Apple Services LATAM LLC, Requerimento de TCC No. 08700.006953/2025-62, Conselho Administrativo de Defesa Econômica [CADE] (Braz.). See also Conselho Administrativo de Defesa Econômica [CADE], CADE Signs a Cease and Desist Agreement with Apple (Jan. 5, 2026), https://www.gov.br/cade/en/matters/news/cade-signs-a-cease-and-desist-agreement-with-apple; Public Version of Apple TCC, https://sei.cade.gov.br/sei/modulos/pesquisa/md_pesq_documento_consulta_externa.php?HJ7F4wnIPj2Y8B7Bj80h1lskjh7ohC8yMfhLoDBLddagMC0PGtHHNJ5OLjWmXqqKoNDHq57WbWDJKa4eAKK0enGmrJphBHiCJbWXKXvI8K_s-573DMQSRm2YO5d-Z. See also Oliveira Neto & Zúñiga, supra note 5.

[26] See CADE ex officio v. Google Inc. & Google Brasil Internet Ltda., Inquérito Administrativo No. 08700.002940/2019-76, Conselho Administrativo de Defesa Econômica [CADE] (Braz.); In re Google LLC & Google Brasil Internet Ltda., Requerimento de TCC No. 08700.007062/2025-23, Conselho Administrativo de Defesa Econômica [CADE] (Braz.). See also Conselho Administrativo de Defesa Econômica [CADE], CADE Signs Cease and Desist Agreement with Google (Dec. 22, 2025), https://www.gov.br/cade/en/matters/news/cade-signs-cease-and-desist-agreement-with-google; Public Version of Google TCC, https://sei.cade.gov.br/sei/modulos/pesquisa/md_pesq_documento_consulta_externa.php?HJ7F4wnIPj2Y8B7Bj80h1lskjh7ohC8yMfhLoDBLddbIYnvUI1X8bWwIhToKTn1x5R0mu6h3LUb759k5CTmSg4i6bYBGxz4PFNami4VYRNZ3Y2xQJfx6E2IFcAjrYNqQ.

[27] See Oliveira Neto & Zúñiga, supra note 5.

[28] See Geoffrey A. Manne, Dirk Auer, Lazar Radic, Selçukhan Ünekbas & Mario A. Zúñiga, ICLE Response to First Review of the Digital Markets Act, Int’l Ctr. for L. & Econ. (Sept. 24, 2025), https://laweconcenter.org/resources/icle-response-to-first-review-of-the-digital-markets-act/ (finding that DMA enforcement has produced “lengthy proceedings, protracted compliance discussions, and frequent recourse to parallel competition-law investigations”) [hereinafter ICLE DMA Review Response].

[29] See Abbott B. Lipsky, Douglas H. Ginsburg, Alexander Raskovich & Dario da Silva Oliveira Neto, Comment of the Global Antitrust Institute on the Brazilian Ministry of Finance, Department of Economic Reform Request for Contributions: Economic and Competitive Aspects of Digital Platforms 24 (Geo. Mason Univ. L. & Econ. Rsch. Paper No. 24-11, May 2024), https://ssrn.com/abstract=4815130. The authors note that EU case law under Articles 101 and 102 TFEU aligns with DMA obligations, while Brazil lacks a comparable body of precedent. They report that, of seven Brazilian antitrust cases involving digital platforms, only three reached a decision—and all were dismissed—while four remain under investigation. No infringement has yet been found in Brazilian digital markets.

[30] See supra notes 25-26.

[31] Conselho Administrativo de Defesa Econômica [CADE], Mercados de Plataformas Digitais, Cadernos do CADE (rev. ed. 2023) (Braz.), https://cdn.cade.gov.br/Portal/centrais-de-conteudo/publicacoes/estudos-economicos/cadernos-do-cade/Caderno_Plataformas-Digitais_Atualizado_29.08.pdf.

[32] Id. at 162 (“Between 1995 and Apr. 30, 2023, 23 investigations were initiated regarding the markets examined [digital markets] in this report. Of these, 9 remained under investigation, 11 were dismissed, and 3 were closed through Termos de Cessação de Conduta (TCC). The conduct primarily involved exclusivity agreements and abuse of dominance.”). This data is likely outdated, but CADE still has not issued a fully adjudicated conviction in digital-market cases.

[33] Org. for Econ. Co-op. & Dev. (OECD), OECD Peer Reviews of Competition Law and Policy: Brazil (2019), https://www.oecd.org/daf/competition/oecd-peer-reviews-of-competition-law-and-policy-brazil-2019.htm. See also Geoffrey A. Manne, Dirk Auer & Mario Zúñiga, ICLE Comments to Brazil’s CADE on Competition in Digital Ecosystems of Mobile Devices, Int’l Ctr. for L. & Econ. (Feb. 11, 2025), https://laweconcenter.org/resources/icle-comments-to-brazils-cade-on-competition-in-digital-ecosystems-of-mobile-devices (citing the OECD’s finding that CADE “is considered one of the most efficient public agencies in Brazil” as evidence of its capacity to address digital-market challenges).

[34] See Conselho Administrativo de Defesa Econômica [CADE], The General Superintendence Is Investigating the Acquisitions of Artificial Intelligence Startups by Big Tech Companies (Aug. 26, 2024), https://www.gov.br/cade/pt-br/assuntos/noticias/superintendencia-geral-apura-aquisicoes-de-startups-de-inteligencia-artificial-por-big-techs. In merger control, CADE may review consummated transactions below the notification threshold for up to one year under Article 88(7) of the Brazilian Competition Law. See infra Section III.D.3.

[35] See Beatriz Kira, Brazil’s Fair Digital Competition Bill Offers an Alternative to Regulating Big Tech, ProMarket (Nov. 7, 2025), https://www.promarket.org/2025/11/07/brazils-fair-digital-competition-bill-offers-an-alternative-to-regulating-big-tech; Victor Oliveira Fernandes, Brazil’s Calibrated Revolution in Digital Competition, ProMarket (Nov. 12, 2025), https://www.promarket.org/2025/11/12/brazils-calibrated-revolution-in-digital-competition. For a comprehensive analysis of the Brazilian Google Shopping case, see 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 Enforcement 570 (2024).

[36] See Oliveira Neto & Zúñiga, supra note 5.

[37] Manne, Auer, Radic, Ünekbas & Zúñiga, supra note 28, at 9.

[38] See Conselho Administrativo de Defesa Econômica [CADE], CADE’s Contribution to the Ministry of Finance’s Public Consultation for Regulation of Digital Platforms 4 (Apr. 2024), https://cdn.cade.gov.br/Portal/centrais-de-conteudo/publicacoes/contribuicoes-do-cade/contribuicoes-cade-ministerio%20fazenda-ingl%C3%AAs.pdf.

[39] Id at 4.

[40] Id. at 19–20. CADE emphasizes the need for a comprehensive, technically grounded Regulatory Impact Analysis (RIA) before imposing new obligations on digital platforms. It notes that such a study should assess market characteristics—including concentration, entry barriers, innovation, and user behavior (e.g., single- and multi-homing)—as well as potential anticompetitive effects from vertical and conglomerate integration. A rigorous RIA, CADE explains, is essential to ensure effective and responsible sector-specific regulation.

[41] Id. at 20 (“This process was exemplified by the European Commission (EC) when introducing the DMA, where three approaches to regulation were examined.”).

[42] The Digital Markets Act entered into force on Nov. 1, 2022, and became applicable on May 2, 2023. In practice, however, enforcement of its core obligations began in March 2024, when the first designated gatekeepers were required to comply. This reflects the European Commission’s initial gatekeeper designations in September 2023, followed by the six-month compliance period provided under the regulation.

[43] Louis-Daniel Pape & Michelangelo Rossi, Is Competition Only One Click Away? The Digital Markets Act’s Impact on Google Maps, Mktg. Sci. (forthcoming 2026), https://doi.org/10.1287/mksc.2025.0159.

[44] Id. at 1.

[45] Chiara Farronato, Andrey Fradkin & Alexander MacKay, Vertical Integration and Consumer Choice: Evidence from a Field Experiment, Working Paper (Mar. 2026), https://alexandermackay.org/files/Vertical%20Integration%20and%20Consumer%20Choice%20-%20Evidence%20from%20a%20Field%20Experiment.pdf.

[46] Id. at 1.

[47] Id. at 35.

[48] Christian Peukert et al., Shrinking Shelf: The Effects of Digital Regulation on Amazon, Presentation at the Workshop on the Economics of the DMA (Sept. 22, 2025), https://www.tse-fr.eu/sites/default/files/TSE/documents/conf/2025/4._longtail_ec_0925.pdf.

[49] Id. at 14, 18.

[50] European Ctr. for Int’l Political Econ. & European Pub. Pol’y P’ship, What About Us? Consumer Response to the Digital Markets Act, ECIPE Occasional Paper No. 10/2025, at 14 (2025), https://ecipe.org/wp-content/uploads/2025/10/ECI_OccasionalPaper_10-2025_LY05.pdf [hereinafter ECIPE, What About Us?].

[51] Id. at 24 (“Reliance on Google, Meta, Apple, and other incumbents remains as strong as ever…”).

[52] Nextrade Group, Impact of the Digital Markets Act (DMA) on Consumers Across the European Union: Results from a Survey with 5,000 Consumers (Sept. 2025), https://www.nextradegroupllc.com/_files/ugd/478c1a_9d7c98475ce8404188d2f8dbb1c9d2ff.pdf.

[53] Carmelo Cennamo, Tobias Kretschmer, Ioanna Constantiou & Eliana Garcés, Economic Impact of the Digital Markets Act on European Businesses and the European Economy, LAMA Econ. Rsch. 45–46 (June 2025), https://www.dmcforum.net/publications/economic-impact-of-the-digital-markets-act-on-european-businesses-and-the-european-economy.

[54] Id. at 46.

[55] Press Release, Eur. Comm’n, Commission Sends Preliminary Findings to Alphabet Under the Digital Markets Act (Mar. 19, 2025), https://digital-markets-act.ec.europa.eu/commission-sends-preliminary-findings-alphabet-under-digital-markets-act-2025-03-19_en.

[56] Foo Yun Chee, Exclusive: Google Faces EU Fine Next Year for Favouring Own Services, Sources Say, Reuters (Dec. 11, 2025), https://www.reuters.com/world/google-faces-eu-fine-next-year-favouring-own-services-sources-say-2025-12-11.

[57] Press Release, Eur. Comm’n, Commission Finds Apple and Meta in Breach of the Digital Markets Act (Apr. 22, 2025), https://ec.europa.eu/commission/presscorner/detail/en/ip_25_1085.

[58] Action brought on July 4, 2025—Meta Platforms v. Comm’n (Case T-435/25), 2025 O.J. (C 5214), https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=OJ%3AC_202505214. See also Foo Yun Chee, Apple Takes Fight Against $587 Million EU Antitrust Fine to Court, Reuters (July 7, 2025), https://www.reuters.com/sustainability/boards-policy-regulation/apple-takes-fight-against-587-million-eu-antitrust-fine-court-2025-07-07.

[59] Press Release, Eur. Comm’n, Commission Opens Proceedings to Assist Google in Complying with Interoperability and Online Search Data Sharing Obligations Under the Digital Markets Act (Jan. 27, 2026), https://digital-markets-act.ec.europa.eu/commission-opens-proceedings-assist-google-complying-interoperability-and-online-search-data-sharing-2026-01-27_en.

[60] Onyeka Aralu & Dirk Auer, From Cure to Care: The DMA’s Chronic Regulation Problem, Truth on the Mkt. (Feb. 12, 2026), https://truthonthemarket.com/2026/02/12/from-cure-to-care-the-dmas-chronic-regulation-problem.

[61] Carlo Stagnaro & Christian N?sulea eds., Digital Revival? How Regulation Prevents the Rise of European Tech Leaders 38 (EPICENTER Feb. 2025), https://www.epicenternetwork.eu/wp-content/uploads/2025/02/Digital-Revival_revised_web.pdf (“The scope of remedies can shape product design, services, business models, and even firm structure. These interventions raise complex, often highly technical issues and may carry negative implications for innovation.”).

[62] Id. at 37.

[63] Apple Inc., The Digital Markets Act’s Impacts on EU Users (Sept. 24, 2025), https://www.apple.com/newsroom/2025/09/the-digital-markets-acts-impacts-on-eu-users. Apple also delayed the rollout of Apple Intelligence in June 2024, citing concerns that the DMA’s interoperability mandates could compromise device security. See Akshaya Asokan, Apple to Delay AI Rollout in Europe: Smartphone Giant Fingers Regulation Meant to Restrain Big Tech, Bank Info Security (June 21, 2024), https://www.bankinfosecurity.com/apple-to-delay-ai-rollout-in-europe-a-25598.

[64] See, e.g., Dan Milmo, Meta Delays EU Launch of Twitter Rival Threads Amid Uncertainty Over Personal Data Use, The Guardian (July 5, 2023), https://www.theguardian.com/media/2023/jul/05/meta-delays-eu-launch-of-twitter-rival-threads-amid-uncertainty-over-personal-data-use.

[65] See Egle Markeviciute, Consumer Waiting Game: Why Do Tech Products Launch Later in Europe?, Euronews (Sept. 26, 2025), https://www.euronews.com/next/2025/09/26/consumer-waiting-game-why-do-tech-products-launch-later-in-europe.

[66] See Cynthia Kroet, Google’s AI Feature on Hold in Most EU Member States Due to ‘Strict Rules’, Euronews (Apr. 1, 2025), https://www.euronews.com/next/2025/04/01/googles-ai-feature-on-hold-in-most-eu-member-states-due-to-strict-rules.

[67] See Anti-American Antitrust: How Foreign Governments Target U.S. Businesses: Hearing Before the Subcomm. on the Admin. State, Reg. Reform, and Antitrust of the H. Comm. on the Judiciary, 119th Cong. 17 (Dec. 16, 2025) (statement of Dirk Auer), https://docs.house.gov/meetings/JU/JU05/20251216/118753/HHRG-119-JU05-Wstate-AuerD-20251216-U2.pdf. Auer explains that regulatory costs are spread across millions of consumers: when users receive “a less useful list of links” instead of integrated tools, they “lose time and convenience” but often attribute the decline to the product itself, rather than regulation. This “rational ignorance” shields regulators from accountability. He adds that while aggregate welfare losses may be significant, the per-user harm is too small to generate backlash, and foreign firms’ objections are easily dismissed as the “whining of monopolists.”

[68] Mert Demirer, Diego J. Jiménez Hernández, Dean Li & Sida Peng, Data, Privacy Laws and Firm Production: Evidence from the GDPR 42 (Nat’l Bureau of Econ. Rsch., Working Paper No. 32146, Feb. 2026), https://www.diegojimenezh.com/assets/pdf/Demirer_et_al2023_Privacy.pdf.

[69] Jian Jia, Ginger Z. Jin & Liad Wagman, The Short-Run Effects of GDPR on Technology Venture Investment, 40 Mktg. Sci. 661 (2021).

[70] Jian Jia, Ginger Zhe Jin & Liad Wagman, The Persisting Effects of the EU General Data Protection Regulation on Technology Venture Investment, Antitrust Source 1, 6 (June 2021), https://www.americanbar.org/content/dam/aba/publishing/antitrust-magazine-online/2021/june-2021/jun2021-jia.pdf.

[71] Lazar Radic & Dirk Auer, A Europe Fit for the Age of Startups: Rhetoric and Reality in the EU’s Digital Package, Int’l Ctr. for L. & Econ. 17 (Aug. 1, 2025), https://laweconcenter.org/resources/a-europe-fit-for-the-age-of-startups-rhetoric-and-reality-in-the-eus-digital-package.

[72] Mario Draghi, The Future of European Competitiveness: Part A: A Competitiveness Strategy for Europe, Eur. Comm’n (Sept. 2024), https://commission.europa.eu/document/download/97e481fd-2dc3-412d-be4c-f152a8232961_en.

[73] Id. at 24 (“The key driver of the rising productivity gap between the EU and the US has been digital technology … and Europe currently looks set to fall further behind. The main reason EU productivity diverged from the US in the mid-1990s was Europe’s failure to capitalize on the first digital revolution … both in generating new tech companies and diffusing digital technology across the economy. Excluding the tech sector, EU productivity growth over the past twenty years would be broadly at par with the US. Europe is lagging in the breakthrough digital technologies that will drive future growth.”).

[74] Id. at 6 (emphasis added).

[75] See id. at 14 (“Only four of the world’s top 50 tech companies are European, and the EU’s global position in tech is deteriorating: from 2013 to 2023, its share of global tech revenues fell from 22% to 18%, while the U.S. share rose from 30% to 38%.”).

[76] See infra Section III.D.3.

[77] See Radic & Auer, supra note 71, at 42 (“To the extent the Digital Package imposes burdens on large firms and, through compliance costs and reduced exit opportunities, raises barriers for startups to scale, it may undermine the goal of fostering competitive European firms and reclaiming technological leadership.”).

[78] See World Justice Project, supra note 6.

[79] See World Bank Group, supra note 7.

[80] Joseph v. Coniglio, Lilla Nóra Kiss, Giorgio Castiglia & Hadi Houalla, A Policymaker’s Guide to Digital Antitrust Regulation, Info. Tech. & Innovation Found. (Mar. 31, 2025), https://itif.org/publications/2025/03/31/a-policymakers-guide-to-digital-antitrust-regulation.

[81] Matthias Bauer, Dyuti Pandya & Vanika Sharma, EU Export of Regulatory Overreach: The Case of the Digital Markets Act (DMA) 2 (Eur. Ctr. for Int’l Political Econ., Policy Brief No. 08/2025, 2025), https://ecipe.org/publications/eu-export-of-regulatory-overreach-dma.

[82] Corruption Perceptions Index 2025, Transparency Int’l (Feb. 10, 2026), https://www.transparency.org/en/cpi/2025.

[83] Jean Comte, ‘Grossly Understaffed’ EU Competition Enforcers Must Prioritize, Guersent Says, MLex (Apr. 8, 2025), https://www.mlex.com/mlex/articles/2322385.

[84] See infra Section III.C.

[85] See Edoardo Ghirotto, Fazenda Diz que Cade Terá Realocação de Equipes se PL 4675/25 for Aprovado, JOTA (Nov. 28, 2025), https://www.jota.info/executivo/fazenda-diz-que-cade-tera-realocacao-de-equipes-se-pl-4675-25-for-aprovado.

[86] See, e.g., Confederação Nacional da Indústria, Custo Brasil: O Vilão Invisível que Não Anda Só (2025), https://cni.portaldaindustria.com.br/custo-brasil.

[87] Observatório do Custo Brasil, Observatório do Custo Brasil: Ferramenta para Monitorar Políticas Públicas Estratégicas com Potencial na Redução do Custo Brasil e Melhoria da Competitividade, https://custobrasil.org.br.

[88] Ministério do Desenvolvimento, Indústria, Comércio e Serviços [MDIC], Agenda de Redução do Custo-Brasil, Secretaria de Competitividade e Política Regulatória, Melhoria do Ambiente de Negócios 5 (June 2025), https://static.portaldaindustria.com.br/portaldaindustria/noticias/media/filer_public/0b/79/0b7935cb-c61c-48f6-98d9-a5fd323b95b7/relatriocustobrasil012025.pdf (“O Custo Brasil é um dos principais obstáculos ao crescimento sustentável e à competitividade da economia nacional, refletindo ineficiências como excesso de burocracia, infraestrutura deficiente, insegurança jurídica e distorções regulatórias.”).

[89] Org. for Econ. Co-op. & Dev. (OECD), Key Takeaways from the 2023–2024 Update of the OECD Product Market Regulation Indicators (2024), https://www.oecd.org/en/topics/product-market-regulation.html.

[90] Org. for Econ. Co-op. & Dev. (OECD), OECD Product Market Regulation (PMR) Indicators: How Does Brazil Compare? (2024), https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/product-market-regulation/Brazil_PMR%20country%20note.pdf.

[91] Org. for Econ. Co-op. & Dev. (OECD), Regulatory Reform in Brazil, OECD Reviews of Regulatory Reform (2022), https://www.oecd.org/content/dam/oecd/en/publications/reports/2022/06/regulatory-reform-in-brazil_da75f3f8/d81c15d7-en.pdf.

[92] See Mario A. Zúñiga, Regulatory Reconquista: Ex-Ante Regulation of Digital Platforms in Latin America, Int’l Ctr. for L. & Econ. (Mar. 17, 2025), https://laweconcenter.org/resources/regulatory-reconquista-ex-ante-regulation-of-digital-platforms-in-latin-america.

[93] Ayesha Bhatti, How the Brussels Effect Hinders Innovation in the Global South, Ctr. for Data Innovation (Jan. 26, 2026), https://itif.org/publications/2026/01/26/how-brussels-effect-hinders-innovation-in-global-south.

[94] See, e.g., Geoffrey A. Manne & Dirk Auer, Brussels Effect or Brussels Defect: Digital Regulation in Emerging Markets, Truth on the Mkt. (Dec. 20, 2022), https://truthonthemarket.com/2022/12/20/brussels-effect-or-brussels-defect-digital-regulation-in-emerging-markets (“The most pressing challenge is to attract investment from international tech firms in the first place, not how to regulate their conduct.”).

[95] Robert D. Atkinson, Letter to the Trump Administration Regarding Non-Tariff Attacks on U.S. Tech Firms and Industries, Info. Tech. & Innovation Found. (July 2, 2025), https://itif.org/publications/2025/07/02/letter-regarding-non-tariff-attacks-on-ustech-firms-and-industries.

[96] Miko?aj Barczentewicz, The Digital Markets Act as an EU Digital Tax: When Compliance Costs Dwarf Regulatory Estimates, Truth on the Mkt. (July 8, 2025), https://laweconcenter.org/resources/the-digital-markets-act-as-an-eu-digital-tax-whencompliance-costs-dwarf-regulatory-estimates.

[97] Gatekeepers, Eur. Comm’n (last visited Mar. 12, 2026), https://digital-markets-act.ec.europa.eu/gatekeepers_en.

[98] See The White House, Defending American Companies and Innovators from Overseas Extortion and Unfair Fines and Penalties, Presidential Actions (Feb. 21, 2025), https://www.whitehouse.gov/presidential-actions/2025/02/defending-american-companies-and-innovators-from-overseas-extortion-and-unfair-fines-and-penalties.

[99] See Siladitya Ray, JD Vance Knocks EU’s Regulation of U.S. Tech Giants: “America Cannot Accept That”, Forbes (Feb. 11, 2025), https://www.forbes.com/sites/siladityaray/2025/02/11/jd-vance-knocks-eus-regulation-of-us-tech-giants-america-cannot-accept-that.

[100] See Kim Marcrael, U.S. Sanctions Former EU Official Over Digital-Content Law, Wall St. J. (Dec. 24, 2025), https://www.wsj.com/world/europe/u-s-sanctions-former-eu-official-over-digital-content-law-c41f574c.

[101] See Patricia Campos Mello, Governo Quer Urgência no Novo Antitruste Digital e Não Teme Reação dos EUA, Diz Dario Durigan, Folha de S. Paulo (Nov. 2, 2025), https://www1.folha.uol.com.br/mercado/2025/11/governo-quer-urgencia-no-novo-antitruste-digital-e-nao-teme-reacao-dos-eua-diz-dario-durigan.shtml.

[102] Id.

[103] Auer, supra note 67, at 18.

[104] See Zúñiga, supra note 8.

[105] See Digital Platforms, supra note 18, § 3.1.

[106] Id. at § 4.1.

[107] Id. at § 4.1.1, Proposal 1.

[108] Giorgio Castiglia, Brazil Should Avoid Rushing Into DMA-Style Regulation, Info. Tech. & Innovation Found. (Feb. 20, 2026), https://itif.org/publications/2026/02/20/brazil-should-avoid-rushing-into-dma-style-regulation.

[109] See Zúñiga, supra note 8.

[110] Id.

[111] BCL, art. 1.

[112] Id.

[113] Id. art. 4.

[114] Id. art. 6.

[115] Id. art. 36, § 1.

[116] Org. for Econ. Co-op. & Dev. (OECD), The Consumer Welfare Standard: Advantages and Disadvantages Compared to Alternative Standards (2023), https://www.oecd.org/content/dam/oecd/en/publications/reports/2023/05/consumer-welfare-standards-advantages-and-disadvantages-compared-to-alternative-standards_4de3277e/3d174fdf-en.pdf (noting that the consumer welfare standard admits “a range of alternative definitions” and “many nuanced takes,” and extends beyond price to include quality, variety, service, and innovation).

[117] See BCL, art. 88, § 5.

[118] Id. art. 88, § 6.

[119] United States v. Baker Hughes, 908 F.2d 981 (D.C. Cir. 1990).

[120] See Herbert Hovenkamp, Is Antitrust’s Consumer Welfare Principle Imperiled?, 45 J. Corp. L. 101 (2019), https://jcl.law.uiowa.edu/sites/jcl.law.uiowa.edu/files/2021-08/Hovenkamp_Final_Web.pdf.

[121] See generally Kenneth Heyer, Consumer Welfare and the Legacy of Robert Bork, 57 J.L. & Econ. 19 (2014), https://www.journals.uchicago.edu/doi/abs/10.1086/676463. Legally, this means that even if a merger increases total welfare through efficiencies, CADE must reject it unless the parties show that a significant share of those gains will accrue to consumers.

[122] Conselho Administrativo de Defesa Econômica [CADE], Guia para Análise de Atos de Concentração Horizontal (July 2016) (Braz.), https://cdn.cade.gov.br/Portal/centrais-de-conteudo/publicacoes/guias-do-cade/guia-para-analise-de-atos-de-concentracao-horizontal.pdf.

[123] Eric Hadmann Jasper, Paradoxo Tropical: A Finalidade do Direito da Concorrência no Brasil, 7 Rev. de Defesa da Concorrência 171, 186–87 (2019).

[124] Gustavo Manicardi Schneider & Rodrigo Fialho Borges, Qual Bem-Estar do Consumidor? Um Objetivo sem Significado, 188 Rev. Dir. Mercantil, Indus., Econ. & Fin. 65, 89 (2025).

[125] Bill 4,675/2025, supra note 1, art. 47-B, §§ I–III.

[126] For an explanation of why scale alone does not constitute antitrust harm, see Brian Albrecht, Scale and Antitrust: Where Is the Harm?, Int’l Ctr. for L. & Econ. (Nov. 22, 2023), https://laweconcenter.org/wp-content/uploads/2023/11/tldr-Scale.pdf.

[127] CADE has not issued formal guidelines on unilateral conduct but is expected to do so in 2026 or 2027 following an internal initiative and the hiring of an external consultant. In practice, CADE defines the relevant market and assesses dominance before evaluating conduct. Consistent with this approach, proposed guidelines by the Instituto Brasileiro de Estudos de Concorrência, Consumo e Comércio Internacional (IBRAC) include a preliminary step to assess dominance based on market definition and market-power analysis. See Instituto Brasileiro de Estudos de Concorrência, Consumo e Comércio Internacional [IBRAC], Guia de Condutas Unilaterais (2020), https://ibrac.org.br/wp-content/uploads/2024/03/IBRAC_-_Guia_de_Condutas_Unilaterais.pdf.

[128] See CADE, Guia para Análise de Atos de Concentração Horizontal, supra note 122, § 2.5.1.

[129] Marius Schwartz, The Nature and Scope of Contestability Theory, 38 Oxford Econ. Papers (New Series), Supplement: Strategic Behaviour and Industrial Competition 37, 38–39 (1986). See generally William J. Baumol, Contestable Markets: An Uprising in the Theory of Industry Structure, 72 Am. Econ. Rev. 1 (1982); William J. Baumol, John Panzar & Robert D. Willig, Contestable Markets and the Theory of Industry Structure (rev. ed. 1988).

[130] Manne, Auer & Radic, Regulate for What?, supra note 15, at 236 (quoting Schwartz, id. at 38).

[131] Id. at 236–37.

[132] See, e.g., Maureen K. Ohlhausen & John M. Taladay, Are Competition Officials Abandoning Competition Principles?, 13 J. Competition L. & Prac. 463, 465 (2022) (“Some recent legislative and regulatory proposals appear to depart from this basic premise. Rather than protecting competition itself, they impose requirements on certain firms to benefit rivals, including those that lagged in investment, innovation, or product development. While such measures may constrain gatekeepers and assist competitors, it remains unclear how they benefit consumers, as opposed to competitors.”).

[133] See, e.g., Harold Demsetz, Barriers to Entry, 72 Am. Econ. Rev. 47, 56 (1982) (“The challenge in crafting policy on entry barriers lies in distinguishing socially desirable from undesirable costs. A narrow focus on production costs overlooks investments in reputation, innovation risk, and efficient scale, while ignoring how current policy shapes future incentives.”). See also Brian Albrecht, What Is a Barrier to Entry?, Truth on the Mkt. (July 7, 2023), https://truthonthemarket.com/2023/07/07/what-is-a-barrier-to-entry (“The problem with Bain’s framework is not his definition but its failure to distinguish cause from effect and endogenous from exogenous factors. Many so-called barriers—such as product differentiation and scale—reflect firm choices and dynamic competition, rather than fixed structural constraints.”).

[134] See Frank H. Easterbrook, The Limits of Antitrust, 63 Tex. L. Rev. 1, 15 (1984); Geoffrey A. Manne, Error Costs in Digital Markets, in The Global Antitrust Institute Report on the Digital Economy 33 (Douglas H. Ginsburg & Joshua D. Wright eds., 2020), https://laweconcenter.org/wp-content/uploads/2020/11/SSRN-id3733662.pdf.

[135] Lazar Radic & Nicolas Petit, The Superiority of the Consumer Welfare Standard, EUI Dep’t of L. Working Paper No. 2024/20 (Dec. 15, 2024), https://ssrn.com/abstract=5065469.

[136] See Manne, Radic & Auer, Regulate for What?, supra note 15, at XX (observing that digital competition regimes grounded in vague goals such as “fairness” and “contestability,” untethered to consumer welfare, lack clear benchmarks and risk protecting competitors rather than competition).

[137] This concern is not unique to Brazil. Ex ante regimes globally tend to reshape competition-law concepts in subtle but significant ways, subordinating consumer welfare to broader and more abstract goals and, in doing so, transforming competition regimes into mechanisms for rent redistribution among firms. See Manne, Radic & Auer, Regulate for What?, supra note 15, at 217–28 (analyzing how ex ante regimes subvert traditional competition-law principles); id. at 228–34 (documenting rent redistribution as a common objective).

[138] Joshua D. Wright & Douglas H. Ginsburg, The Goals of Antitrust: Welfare Trumps Choice, 81 Fordham L. Rev. 2405, 2409 (2013).

[139] Id. at 2416–17.

[140] If harmful, standard competition analysis already accounts for it. See, e.g., id. at 2422 (“While incorporating product variety, quality, and innovation into welfare analysis is desirable when done correctly…, modern antitrust analysis comfortably incorporates the tradeoffs between price and quality that consumers face.”).

[141] See Schneider & Borges, supra note 124, at 88.

[142] See Thom Lambert, Rent-Seeking and Public Choice in Digital Markets, in The Global Antitrust Institute Report on the Digital Economy (Joshua D. Wright & Douglas H. Ginsburg eds., 2020), https://gaidigitalreport.com/2020/08/25/rent-seeking-and-public-choice-in-digital-markets; Manne, Auer & Radic, Regulate for What?, supra note 15, at 254.

[143] See Bill 4,675/2025, supra note 1, art. 47-C.

[144] Id. at §§ I-VI.

[145] See id.

[146] See id.

[147] See, e.g., Jean-Charles Rochet & Jean Tirole, Platform Competition in Two-Sided Markets, 1 J. Eur. Econ. Ass’n 990 (2003).

[148] See, e.g., Francine Lafontaine & Margaret Slade, Vertical Integration and Firm Boundaries: The Evidence, 45 J. Econ. Lit. 629, 633–37 (2007).

[149] See Geoffrey A. Manne & Dirk Auer, Antitrust Dystopia and Antitrust Nostalgia: Alarmist Theories of Harm in Digital Markets and Their Origins, 28 Geo. Mason L. Rev. 1281 (2021).

[150] William M. Landes & Richard A. Posner, Market Power in Antitrust Cases, 94 Harv. L. Rev. 937 (1981).

[151] See Harold Demsetz, Industry Structure, Market Rivalry, and Public Policy, 16 J.L. & Econ. 1 (1973). See also Chad Syverson, Macroeconomics and Market Power: Context, Implications, and Open Questions, 33 J. Econ. Persp. 23, 26 (2019) (“[C]oncentration is worse than just a noisy barometer of market power. Instead, we cannot even generally know which way the barometer is oriented.”); Steven Berry, Martin Gaynor & Fiona Scott Morton, Do Increasing Markups Matter? Lessons from Empirical Industrial Organization, 33 J. Econ. Persp. 44, 46 (2019) (“Within the field of industrial organization, the structure-conduct-performance approach has been discredited for a long time.”); id. at 48 (“In short, there is no well-defined ‘causal effect of concentration on price,’ but rather a set of hypotheses that can explain observed correlations….”).

[152] See Ministério da Fazenda, Governo Federal Envia à Câmara dos Deputados Projeto para Regulação Concorrencial das Big Techs (Sept. 17, 2025), https://www.gov.br/fazenda/pt-br/assuntos/noticias/2025/setembro/governo-federal-envia-a-camara-dos-deputados-projeto-para-regulacao-concorrencial-das-big-techs (“It is estimated that five to ten platforms operating in Brazil will be designated.”).

[153] See Manne, Radic & Auer, supra note 15, at 229–33 (documenting how ex ante regimes across jurisdictions aim to redistribute rents from large, predominantly U.S.-based, technology firms to domestic competitors and business users).

[154] See Oliveira Neto & Radic, Brazil’s Digital Markets Bill: A DMA Through the Back Door?, supra note 21.

[155] See Bill 4,675/2025, supra note 1, art. 87-A, § 2.

[156] See Easterbrook, supra note 134.

[157] See Manne, Error Costs in Digital Markets, supra note 134.

[158] DMCC, ch. 1, §§ 2–5.

[159] Gesetz gegen Wettbewerbsbeschränkungen [GWB] [Competition Act], as amended by Gesetz, Oct. 25, 2023, Bundesgesetzblatt, Teil I [BGBl. I] 294 (Ger.), § 19a(1)(1).

[160] See supra Section III.A.2.a.

[161] David S. Evans, Attention Rivalry Among Online Platforms, 9 J. Competition L. & Econ. 313 (2013).

[162] See Bill 4,675/2025, supra note 1, art. 14-A, § 2.

[163] See id., art. 14-B.

[164] See DMA, Eur. Comm’n (last visited Mar. 26, 2026), https://digital-markets-act.ec.europa.eu/index_en (“The European Commission is the sole enforcer of the DMA. A joint team in the Directorates-General for Competition (DG COMP) and Communications Networks, Content and Technology (DG CONNECT) is responsible for its implementation and enforcement.”).

[165] Working in Digital at the CMA, Competition & Markets Auth. (last visited Mar. 26, 2026), https://www.civil-service-careers.gov.uk/cma-working-in-digital-at-the-cma.

[166] Bundeskartellamt, The Bundeskartellamt: Organisation, Tasks and Activities (2022), https://www.bundeskartellamt.de/SharedDocs/Publikation/EN/Broschueren/Brochure_About_%20Bundeskartellamt.pdf.

[167] See Bill 4,675/2025, supra note 1, arts. 87-A–87-B.

[168] See id. at arts. 87-A, § 5 & 87-B, § 5.

[169] See id. at arts. 87-A, § 6 & 87-B, § 6.

[170] For a comprehensive analysis of the risks and consequences of regulatory extortion, see Fred S. McChesney, Money for Nothing: Politicians, Rent Extraction, and Political Extortion (1997).

[171] See Ghirotto, supra note 85.

[172] See Comte, supra note 83.

[173] See Oliveira Neto & Radic, supra note 21.

[174] See Bill 4,675/2025, supra note 1, art. 47-E, § IV(b)–(h).

[175] Geoffrey A. Manne, Against the Vertical Discrimination Presumption, Concurrences, No. 2, at 1 (May 2020).

[176] See generally Geoffrey A. Manne, Kristian Stout & Eric Fruits, The Fatal Economic Flaws of the Contemporary Campaign Against Vertical Integration, 68 Kan. L. Rev. 923 (2020).

[177] Dario Oliveira Neto, Self-Preferencing in Brazil: Should We Regulate Before We Understand?, Int’l Ctr. for L. & Econ. 4 (2025), https://laweconcenter.org/resources/self-preferencing-in-brazil-should-we-regulate-before-we-understand.

[178] See, e.g., Michael Salinger, Self-Preferencing, in The Global Antitrust Institute Report on the Digital Economy 329, 333 (Joshua D. Wright & Douglas H. Ginsburg eds., 2020), https://gaidigitalreport.com/2020/08/25/self-preferencing; Manne, supra note 175.

[179] Yuta Kittaka, Susumu Sato & Yusuke Zennyo, Self-Preferencing by Platforms: A Literature Review, 66 Japan & the World Econ. 101191, 8 (2023). See also Feng Zhu, Friends or Foes? Examining Platform Owners’ Entry into Complementors’ Spaces, 28 J. Econ. & Mgmt. Strategy 23, 27 (2019).

[180] See Zhuoxin Li & Ashish Agarwal, Platform Integration and Demand Spillovers in Complementary Markets: Evidence from Facebook’s Integration of Instagram, 63 Mgmt. Sci. 3438 (2017).

[181] See Jens Foerderer et al., Does Platform Owner’s Entry Crowd Out Innovation? Evidence from Google Photos, 29 Info. Sys. Res. 444 (2018).

[182] See Carmelo Cennamo, Yuan Gu & Feng Zhu, Value Co-Creation and Capture in the Creative Industry: The U.S. Home Video Game Market, Working Paper (2016), http://questromworld.bu.edu/platformstrategy/files/2017/06/PlatStrat_2017_paper_21.pdf.

[183] See Farronato et al., supra note 45.

[184] Yusuke Zennyo, Platform Encroachment and Own-Content Bias, 70 J. Indus. Econ. 684, 705 (2022).

[185] See Pape & Rossi, supra note 43; Nextrade Grp., supra note 52, at 2; ECIPE, What About Us?, supra note 50, at 14.

[186] See Legal Grounds Institute, Legal Grounds’ Report on the Impact of Bill 2768 on Legal Certainty: Summary of the Findings (2025), https://legalgroundsinstitute.com/wp-content/uploads/2024/10/FindingsReportSelf-Preferencingfinal.pdf.

[187] Id. at 10.

[188] Id.

[189] Id.

[190] See Rochet & Tirole, Platform Competition in Two-Sided Markets, supra note 147 (analyzing two-sided platform economics and the welfare effects of limits on cross-subsidization and steering).

[191] See Manne et al., ICLE DMA Review Response, supra note 28, at 14.

[192] See Bill 4,675/2025, supra note 1, art. 47-E(V)(a)–(h).

[193] See id. art. 47-E, § 2 (providing that CADE “may consider” information-security factors).

[194] See Kati Suominen, New Costs and Cybersecurity Challenges Flagged as DMA Compliance Starts, Ctr. for Strategic & Int’l Stud. (Mar. 22, 2024), https://www.csis.org/analysis/new-costs-and-cybersecurity-challenges-flagged-dma-compliance-starts.

[195] See Oliveira Neto & Zúñiga, supra note 5.

[196] See Brasil, Lei nº 13.709, de 14 de agosto de 2018, art. 18(V) (Braz.) [hereinafter LGPD].

[197] See Matthew Kilcoyne & Joseph v. Coniglio, Comments to European Commission, In the Matter of: Consultation on Joint Guidelines on the Interplay Between DMA and GDPR, Info. Tech. & Innovation Found. (Dec. 4, 2025), https://itif.org/publications/2025/12/04/comments-european-commission-joint-guidelines-on-interplay-between-dma-and-gdpr.

[198] Marc Bourreau, Jan Krämer & Miriam Buiten, Interoperability in Digital Markets, Ctr. on Regul. in Eur. [CERRE] (Mar. 2022), https://cerre.eu/wp-content/uploads/2022/03/220321_CERRE_Report_Interoperability-in-Digital-Markets_FINAL.pdf.

[199] See Brian Albrecht, Network Effects and Interoperability, Int’l Ctr. for L. & Econ. (May 5, 2023), https://laweconcenter.org/wp-content/uploads/2023/05/Interoperability-TLDR.pdf.

[200] Ofcom, Mandated Interoperability in Digital Markets 18–19 (Econ. Discussion Paper Series, Issue 8, Nov. 2, 2023), https://www.ofcom.org.uk/siteassets/resources/documents/research-and-data/economic-discussion-papers-/discussion-paper-mandated-interoperability-in-digital-markets?v=330343 (“There is a fundamental interaction between innovation and interoperability….”).

[201] Id. at 8.

[202] See Inge Graef, Martin Husovec & Jeroen van den Boom, Spill-Overs in Data Governance: Uncovering the Uneasy Relationship Between the GDPR’s Right to Data Portability and EU Sector-Specific Data Access Regimes, 9 J. Eur. Consumer & Mkt. L. 3 (2020).

[203] Peter Swire & Yianni Lagos, Why the Right to Data Portability Likely Reduces Consumer Welfare: Antitrust and Privacy Critique, 72 Md. L. Rev. 335, 339 (2013).

[204] Id. at 352.

[205] Christian Reimsbach-Kounatze & Andras Molnar, The Impact of Data Portability on User Empowerment, Innovation, and Competition 5 (OECD Going Digital Toolkit Note No. 25, June 2024), https://www.oecd.org/content/dam/oecd/en/publications/reports/2024/06/the-impact-of-data-portability-on-user-empowerment-innovation-and-competition_ee329380/319f420f-en.pdf.

[206] Geoffrey A. Manne & Sam Bowman, Data Portability and Interoperability: The Promise and Perils of Data Portability Mandates as a Competition Tool, Int’l Ctr. for L. & Econ. (Sept. 10, 2020), https://laweconcenter.org/resources/issue-brief-data-portability-and-interoperability-the-promise-and-perils-of-data-portability-mandates-as-a-competition-tool.

[207] See, e.g., Verizon Commc’ns Inc. v. Law Offices of Curtis v. Trinko, LLP, 540 U.S. 398, 407–08 (2004) (warning that compelling firms to share the source of their advantage may weaken incentives for both incumbents and rivals to invest in beneficial facilities).

[208] See Bill 4,675/2025, supra note 1, art. 47-E(I).

[209] See generally Geoffrey A. Manne, Samuel Bowman & Dirk Auer, Technology Mergers and the Market for Corporate Control, 86 Mo. L. Rev. 1047 (2022).

[210] Gordon M. Phillips & Alexei Zhdanov, Venture Capital Investments, Merger Activity, and Competition Laws Around the World, 13 Rev. Corp. Fin. Stud. 303, 307 (2024).

[211] Manne, Bowman & Auer, supra note 211, at 1113–14.

[212] See generally Radic & Auer, A Europe Fit for the Age of Startups, supra note 71.

[213] See BCL, art. 88.

[214] Oliveira Neto & Zúñiga, supra note 5.

[215] See Bill 4,675/2025, supra note 1, art. 47-E § 2.

[216] See supra Section III.A.1.

[217] See BCL, art. 36.

[218] See DMA recital 23 (“Any justification on economic grounds seeking to enter into market definition or to demonstrate efficiencies deriving from a specific type of behaviour by the undertaking providing core platform services should be discarded, as it is not relevant to the designation as a gatekeeper.”).

[219] See DMCC, pt. 1, ch. 3, § 29 (Countervailing Benefits Exemption).

[220] See id. § 29(2)(a)–(h).

[221] 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. of Econ. Affs. (Sept. 19, 2023), https://laweconcenter.org/wp-content/uploads/2023/09/Perspectives_4_Digital-overload_web-1.pdf.

[222] See Manne, Radic & Auer, Regulate for What?, supra note 15, at 201.

[223] See id. at 249–54 (arguing that ex ante regimes treat consumer-welfare losses as acceptable tradeoffs for competitor benefits and for supply-chain wealth transfers deemed “fair” by regulators).

[224] Bill 4,675/2025, supra note 1, art. 87-G(I).

[225] Fredrik Erixon, Andrea Dugo & Dyuti Pandya, Reviewing the Digital Markets Act: Inspirations from Japan, Eur. Ctr. for Int’l Political Econ. [ECIPE] (Feb. 2026), https://ecipe.org/insights/dma-review-inspirations-from-japan.

[226] Lei No. 13.874, de 20 de setembro de 2019, Diário Oficial da União [D.O.U.] de 20.09.2019 (Braz.), art. 5.