Stopping the Brussels Effect: Big Tech and Brazil's Digital Markets Bill

The Digital Markets Act (DMA), though favorably endorsed by many in Brussels, has its fair share of critics. Critics, however, are not united in their view of the problems of the DMA. On one side, there are critics who argue the DMA does not go far enough. Cristina Caffarra writes for Tech Policy Press that the DMA is “mostly competition on the platform, not competition to the platform. It’s not worthless, but let’s be clear that digital giants are not being disassembled here – it’s about creating competition to the services they offer.” Her argument is that expectations about what the DMA may achieve should be managed. In a previous article, I also argued that there is reason to worry that the DMA may end up further reinforcing the power of gatekeepers. Though they may be forced to open their ecosystems, gatekeepers are largely left in control of potential new industry competitors.

On the other side, critics argue that the DMA, or laws being drafted inspired by the DMA, are detrimental to innovation and are ineffective in promoting competition. Those critics often argue that existing anticompetition law suffices in addressing competition in digital markets and that over-regulation will merely scare off entrepreneurs. Google recently posted a blog warning that DMA rules hurt small businesses in an attempt to spook countries considering similar rules. On May 7, 2024, Charles McConnell reported for Global Competition Review that several Big Tech companies are attempting to dissuade Brazil from implementing the proposed Digital Markets Law Bill (PL 2768/2022), which is in large part inspired by the DMA.

Brazil’s bill is an ex-ante regulation, subjecting companies deemed as having “crucial access control power” to a variety of rules and obligations. Similar to the DMA’s conceptualization of gatekeepers, companies with “crucial access control power” can be online intermediation services, search engines, social networks, video sharing platforms, interpersonal communication services, operating systems, cloud computing services, and online advertising services with significant economic size. Though largely inspired by the DMA, legislation that is intended to complement anticompetition law, Brazil’s bill is intended to complement telecommunications legislation and is proposed to be enforced by Brazil’s National Telecommunications Agency (ANATEL).

Comparing Brazil's Digital Markets Law Bill with the DMA

Victor Oliveira Fernandes analyzes the bill in comparison with the DMA, finding that the DMA’s obligations imposed on gatekeepers are much more precisely defined and expansive compared to the Brazilian bill. While the DMA “delineates an extensive list of bifurcated obligations with differing levels of specificity, the Brazilian legislation puts forth just four broad guiding principles, leaving a large regulatory discretion to ANATEL.” The reason for this difference is explained by Congressman João Maia, who argues “that it would be wise to have a less extensive set of ex-ante measures, given the digital markets are deeply dynamic.” Fernandes points to the various obligations and explains how they easily risk underenforcement due to the vague language and lack of specified goals of the legislation.

The general obligations are set out in Article 10, stating that those platforms deemed to have “crucial access control power” must commit to transparency with the regulator, provide fair and equal service to both business and end users, ensure proper usage of data in commercial activities, and prohibits platforms from refusing to contract with business users. Beyond fines, which may reach a maximum of 2% of the conglomerate’s annual revenue in Brazil or 1% of the revenue for the duration of the violation, the proposal also grants ANATEL the power to implement mandatory measures, temporary suspension of operations, and possibly even the prohibition of a platform’s activities within the borders of Brazil.

Fernandes finds that the Brazilian proposal should learn from other digital market laws beyond the DMA, which may not be as applicable to the Brazilian context. He writes that the DMA’s “obligations reflect the European Commission’s competition case discussions. This connection simplifies the understanding of the scope of the behavioral measures intended by the law. In contrast, PL 2768's goals are dramatically opaque.” Overall, he finds that the definitions of platforms, the criteria to be designated as having "crucial access control power," and what competition means in digital markets to be all underdefined in the Brazillian bill.

Lilla Nóra Kiss similarly writes for the Information Technology and Innovation Foundation (ITIF) that the bill is underdefined. Kiss makes valid arguments regarding designation, which is far less defined than the DMA, stating that “companies with annual gross revenues exceeding BRL 70 million in services to Brazilians—less than $15 million” will be designated with no ability to contest designation decisions. Meanwhile, the DMA requires a gatekeeper to “achieve an annual turnover in the European Economic Area (EEA) equal to or above €7.5 billion in each of the last three financial years, or where its average market capitalization or equivalent fair market value amounted to at least €75 billion in the last financial year” as well as other qualitative measures.

The qualitative measures have proven important in gatekeeper evaluation, which considers whether the platform service in question is an important gateway for business users to reach end users. Both Kiss and Fernandes point out that the Brazillian threshold risks overcapture of the regulation and should be much further refined.

Industry group's arguments against the bill

While legal scholars like Fernandes sit in the camp that the legislation does not go far enough, or rather, should at least be further refined and specify exactly what abusive behavior it aims to tackle, industry groups have warned that the legislation goes too far and risks hindering the development of the Brazillian digital economy. For example, Krisztian Katona from CCIA writes that the regulation proposal is entirely unnecessary and likely to scare off investment, warning that “before regulating competition, a country needs to attract it.”

Similarly, Lilla Nóra Kiss writes for ITIF that the bill is unnecessary and the ex-ante approach is flawed. ITIF’s comments were in response to Brazil’s Ministry of Finance's public consultation on the implementation of the Digital Markets Law Bill. Respondents to the consultation included ITIF, a U.S.-based NGO, that receives support from tech companies, including Amazon, Alphabet, Meta, and Microsoft, as well as the Computer and Communications Industry Association (CCIA), whose members include Meta, Google, Apple, Amazon, and X.

Kiss argues that the bill is unnecessary because digital markets are already competitive. She assesses the marketplace “in terms of ‘eyeballs’ or ‘attention markets.’ A Sao Paulo resident who is using Twitter is not using YouTube, LinkedIn, Facebook, or TikTok. In other words, even though digital platforms serve particular niches, they compete fiercely for consumer attention.” The logic that users may view one platform at a time as a signal that the market is competitive is flawed.

Various reports and digital marketplace monitors have found that the online platform market in Brazil is concentrated, particularly the social media market. The Reuters Institute found that the top platforms in Brazil in 2023 for messaging and social media are WhatsApp (75%), YouTube (73%), Instagram (63%), Facebook (59%), X (24%), and TikTok (29%). DataReportal estimates that for January 2024, 61.8% of the “eligible” audience in Brazil (those aged 13 and above) use Facebook, and 74.8% use Instagram. Moreover, YouTube ads reached 76.6% of Brazil’s total internet user base (regardless of age).

Across these studies, it is noteworthy to see how most Brazilians report high usage of a handful of social media apps, most owned by Meta. In fact, Statista found in another survey of the most commonly found apps on smartphone users' home screens that WhatsApp, Instagram, and Facebook all scored the highest by a large margin, and they are all owned by Meta.

Moreover, Kiss argues that the current competition law, the Brazilian Competition Act of 2011, suffices to address the current digital economy. Kiss writes that the point of competition law should be to nurture innovation and protect consumer welfare, with merger review considering the effects of efficiency or technological progress in mind. Katona echoes the sentiment that the current law, and the Brazillian competition regulator, the Administrative Council for Economic Defense (CADE), already “has the analytical tools that are necessary to investigate and deter potential harms to competition in digital markets.”

However, Fernandes, who is both a professor of competition law and a Commissioner at CADE’s tribunal, illustrates CADE’s failure in defining explicit theories of harm. Fernandes points to extremely lengthy investigations against Big Tech companies, ranging from 5 to 7 years, by which time often the service under investigation has substantially changed and the investigation ends up being dismissed. Fernandes points out that CADE has been unable to implement meaningful changes in the digital economy due to the commissioner's conservative approach to evaluating theories of harm and reliance on interim measures, which do little to solve the abuse of dominance under investigation.

Fernandes finds that the existing competition law is insufficient to investigate the digital economy, finding that “it seems preferable for Brazil to adopt a new, supplementary regulatory legislation specifically tailored for digital platforms. [...] Additionally, new regulatory legislation could provide clear guidelines and criteria for assessing the behavior of dominant players in the digital market, ensuring more efficient and effective enforcement of competition law.” Fernandes’ concludes that supplementary rules would provide clear guidelines for regulators. Regulation can also provide legal certainty for businesses, rather than, as Kiss and Katona argue, “stifling innovation.”

In defense of regulation

In Brazil, providing clear rules on what might constitute abuse of dominance can give legal clarity to platforms on what behavior to avoid. As Anu Bradford writes, regulation can encourage ethical technology development rather than innovation for innovation’s sake. For example, Bradford notes that “[s]everal tech companies have acknowledged that AI regulation can serve their business interests, lending support to the notion that social innovation can also translate into market innovation. In particular, they recognize that tech regulation can enhance consumer confidence in new products, thus generating useful market innovations.” She uses the example of the EU GDPR which led to a wider recognition of privacy online and the emergence of privacy-conscious businesses.

Bradford also argues that digital regulations can protect social objectives (like privacy) and foster trade among EU Member States. Particularly, the GDPR provides a legal gateway for data flows across Europe. The idea of the DMA is also to open up digital marketplaces and foster trade. For example, forcing gatekeepers to allow interoperability is intended to encourage the development of new products and services.

Brazil’s bill appears to strive for the same goal, introducing mandates like requiring platforms to “appropriate use of data”, which suggests that a platform should avoid behaviors like hoarding first-party data across its own services. Equally, the bill prohibits designated platforms from denying business users access to its services and technology, perhaps suggesting interoperability mandates similar to the DMA.

However, critics like Kiss argue that the Brazilian bill will chill innovation, as platforms “maintaining the exclusivity of certain innovations can foster competition by encouraging companies to continuously strive for advancements, knowing that they can reap the rewards of their efforts without immediate compulsory sharing." Various researchers, like Cory Doctorow and James Bessen, have illustrated that, on the contrary, interoperability promotes innovation. The more technology is shared or can be accessed, the more it can be improved, creating new and better innovations. Bessen explicitly argues that the rise of monopolies and oligopolies, not just in tech markets but in all industries, is linked to the rise of proprietary software, which has slowed innovation and increased inequality.

This challenges the anti-regulation argument that regulation attempting to open up such propriety software “easily lead[s] to condemnation of business models and practices that provide benefits for consumers, such as lower prices and a safer user experience, among others,” as Lazar Radic and Mario A. Zúñiga, on behalf of the International Center for Law and Economics, remarked in their feedback on the Brazilian bill.

Indeed, while it may be convenient for Google users to use various integrated services within that ecosystem, more interoperable technology, such as those adopting standards, encourages a wider ecosystem of developers and increases innovation, which benefits both the consumer and a wider pool of developers. Not to mention, moving innovation out of the hands of a few tech companies lessens economic concentration and the market power and influence that comes with such a concentration.

Central to Bradford’s conclusion is that it is not regulation that hinders technological innovation in the EU, as is so often argued by the anti-regulation crowd. She lists other factors such as immigration policy; lack of integrated capital markets; bankruptcy law; and, despite continued efforts, a lack of a digital single market leading to businesses being unable to scale. A 2020 report by the OECD on Brazil’s digital economy and policy developments related to digitization equally does not suggest overregulation as preventing growth in the digital economy. Rather, the OECD finds that the digital economy is restrained by a lack of public funding for research and development; knowledge gaps as highly skilled individuals prefer academia over industry jobs; a lack of collaboration across sectors and disciplines, including technology transfer from academia to industry; and high tariffs on ICT goods.

While it is healthy for a mix of perspectives to examine incoming legislation critically, certain repeated and common arguments against regulation should also be critically examined. A more nuanced view needs to be adopted regarding new legislation beyond unfairly equating digital legislation as stifling innovation.

Equally, when drafting new legislation, though it is important to bear in mind the rapid changes that can occur in these markets, under definition or too much ambiguity is unlikely to achieve much. For example, the Brazilian bill suggests data restrictions or interoperability mandates. It uses vague language such as “appropriate use of data” without providing a definition for ‘appropriate,’ which risks consistent enforcement.

On the other side, uncritical praise of new legislation may overstate what realistically these new laws can achieve. Under the DMA, there is an opportunity to create new products, but it may still be difficult to challenge gatekeepers, particularly in their core market offerings. Taking a holistic view of how concentrated markets came to be that way, specific practices dominant companies deploy to maintain dominance as well as what effects this has beyond mere consumer welfare approaches, appears to be useful in understanding how to correct concentrated markets.

The Brazillian bill would do well to articulate specific abusive practices that regulation would correct. Rather than evaluating harms in terms of efficiency or price, considering how some platforms have come to be dominant in ways that make their products essential to modern life and how that affects citizens rather than consumers might be helpful. This might require changing mindsets away from innovation for innovation's sake as a penultimate goal.