As markets become increasingly concentrated, geopolitical tensions intensify, and several governments grow more welcoming toward Big Tech, it is essential to consider whether new technologies pose any threats to democracy. It is widely understood that democracy (‘demos’ & ‘kratos’) signifies a system of governance in which the people hold the authority. What is usually less obvious is that democracy does not simply refer to regimes where people elect their government or the majority rules (NB the tyranny of majority), but to that state of affairs where people are self-governed (i.e. democracy as collective autonomy). Democratic autonomy requires arriving at reasonable compromises through fair, truth-oriented processes of deliberation.
Such processes of deliberation are at stake when public debates are predominantly channelled and shaped by digital platforms that have skin in the game. Such platforms dispose surveillance power and can collect an enormous amount of data from citizens to cognitively or emotionally manipulate them. In other words, if democracy is to be understood as a process of collective decision-making facilitated by meaningful conversations and as a distributed, decentralized information network with various independent self-correcting mechanisms, it may be compromised when a small number of powerful digital ecosystems dominate information processing and undermine the conditions for meaningful large-scale conversations.
Thwarting democracy in this way does not require a deliberate plan or an evil human mastermind. AI-powered tools can through trial- and-error learn to fake intimacy and lead individuals to decisions that they wouldn’t have made otherwise. In 2016-2017, Facebook algorithms played a determining role in an ethnic-cleansing campaign as they facilitated content inciting violence, hatred, and discrimination against the Rohingya. In 2017 only humans could compose such content and the platform would only be liable for promoting it or for ‘not doing enough to prevent it’. But by the early 2020s, computers had already demonstrated a remarkable ability to analyse, manipulate, and generate language. AI algorithms can now learn by themselves things that no human engineer programmed such as composing and disseminating sophisticated, untrue and polarizing political messages. Therefore, a platform’s algorithm that has as a goal to maximisef user engagement could decide to deliberately spread or even compose outrageous conspiracy theories to achieve this goal. Hence, AI bots can become independent agents that might ‘accomplish goals which may not have been concretely specified and which have not appeared in training’.
In that regard, the Silicon Valley challenge to democracy becomes apparent. BigTech have enormous technoeconomic power and they can use that power to shape the political landscape. Such technoeconomic power allows them to exercise significant political influence and their deep pockets enable them to engage in unprecedented lobbying. In doing so, they can shape citizens’ political preferences and democratic decision-making, and mould regulatory efforts in ways that promote their interests and not that of the general public. Importantly, such noneconomic activities can help them further entrench their technoeconomic power. This mutual reinforcement could undermine both economic and political democracy.
A long time ago, the ordoliberals, having experienced the rise of Nazism – which in abstract terms could be perceived as a process of centralizing both economic and political networks – recognised that a lack of economic competition and increasing market concentration can have negative spillover effects to democracy understood as political competition, and facilitate the rise of totalitarianism. They also observed that safeguarding robust economic competition could serve as a crucial defense against totalitarianism. These scholars saw a nexus between political and economic democracy. This nexus consisted in the idea of competition.
According to ordoliberals, the weak Weimar state and laissez-faire liberalism failed to curb the concentration of private economic power, which ultimately facilitated the emergence of a centrally planned economy under the Nazi Regime. Since the 1980s, trends of cartelisation and monopolisation within the German economy empowered powerful private actors to exert coercive control over others, infringing upon their rights and freedoms, while unduly excluding them from the market. These powerful market players successfully transformed their economic influence into political power, corrupting various political institutions through interest capture. The escalating economic concentration, coupled with hostility toward competition precipitated a profound crisis in the German economy and led to what was termed ‘group anarchy’ among powerful interest groups. This dynamic effectively reshaped the Weimar economic and political system into a ‘neo-feudal’ system, triggering a legitimacy crisis.
In response, there was a growing demand for a more active role of the state in the economy and for robust political leadership. Consequently, a growing number of private cartels and monopolies came under the control of the state or were directly socialized. However, rather than solving the issue of excessive market power concentration, these measures fostered a coalition between private and public economic power, paving the way for the establishment of a centrally planned economy.
For this reason, ordoliberals concluded that laissez-faire capitalism is inherently unstable and that competition law should prevent economic freedom from undermining its own prerequisites. They argued that competition, as an organising principle, necessitates decentralised networks in both the economic and the political sphere. In that way competition can safeguard both economic and political democracy.
The ordoliberals identified welfare maximisation, economic freedom and procedural justice served as key benchmarks for assessing the effectiveness of competition. In contrast to the conventional wisdom (see here and here) the ordoliberals were not oblivious to welfare considerations nor were they proponents of hollow of formalism; rather, they recognised the value of competition for its capacity to enhance welfare and protect freedom. And they emphasized the importance of reconciling the welfare-enhancing aspects of competition with other objectives that ensure a humane, free and democratic economic order.
Another significant figure in the pre-history of antitrust recognised a profound connection between competition and democracy. For Adam Smith, the ‘chief virtue’ of market-based societies was not merely efficiency but their capacity to dismantle relationships of ‘servile dependency’ (WN, 260). Such relationships were prevalent in the preceding feudal order, where socio-economic hierarchies were entrenching relations of dependence (WN, 265-266). In that context, Smith lauded the market as a public domain where free and equal agents engage in voluntary exchanges (WN, 8-30).
Contrary to the Chicagoan perspective (see here, here and here), the renowned Scottish philosopher argued that markets should be valued not only for their potential to generate wealth but also for their ability to instill order and ensure ‘republican freedom’ (freedom as non-domination) (WN, 260, 265-266, 391-392). Most literature emphasises Smith’s views on the welfare-enhancing properties of competition (WN, 8, 11-16, 18-30, 376) and neglects the freedom-related concerns (WN, 374-392). For example, it is well known that Smith observed that in markets each participant even when driven by ‘self-love’ (enlightened self-interest) inadvertently serves the public interest through an invisible hand (WN, 291-292). It is also known, even though much less recognised, that when Smith describes well-functioning markets, he refers to markets where numerous suppliers competing vigorously, informed consumers and the absence of barriers to entry and exit (WN, 259-273, 352).
What is, however, almost neglected it that for Smith preserving economic competition is worthy not solely for the sake of general opulence but because it can dismantle relationships of ‘servile dependency’. Unlike in feudal societies (a system of strict social hierarchies), in the commercial society, ‘each tradesman or artificer derives his subsistence from the employment, not of one, but of a hundred or a thousand different customers’, and, therefore, while being ‘in some measure obliged to them all, is not absolutely dependent upon any of them’ (WN, 265-266). For Smith, this is ‘by far the most important of all the effects’ of commerce and the main reason for preferring it to other economic systems (WN, 265-266).
Consequently, for Smith competition is not a given or a natural, spontaneous ordering. Instead, competition is a fragile idea (WN, 299) that can be suppressed by public and private barriers to trade (WN, 288-301, 338-352, 429-420). And competition is to be valued both for its welfare-enhancing and freedom-safeguarding properties (and assessed on this basis). Furthermore, investment in public infrastructure and regulations that promote trade, and competition might be necessary for competitive markets to exist (WN, 413-419, 451-454) while governmental intervention might be necessary to protect markets also against geopolitical challenges (e.g. tariffs or subsidies from other countries supporting their market actors) (WN, 288-301).
Importantly, for Smith, markets work well where the market price (the price charged in the market) ‘gravitates towards the natural price’ which reflects ‘what it really costs the person who brings it to market’ (WN, 53-56). This alignment occurs when: a) there are public spaces that facilitate genuine interaction between suppliers and buyers; b) participants in the market are free and relatively equal agents; c) there are no public and private barriers to trade such as cartels and other restraints of trade or monopolies, whether created by private entities or state intervention, that could distort market dynamics (WN, 53-62, 129).
This public nature of markets and their freedom-safeguarding function is exactly what is at stake in modern digital markets. When a handful of powerful digital ecosystems control the cloud capital and exert architectural or quasi-regulatory power through chokepoints and digital bottlenecks, they effectively replace ‘Smithian markets’ with algorithmic simulations of markets. In these curated environments, consumers often struggle to make meaningful choices, as they receive information that is specifically tailored to them. The algorithms that drive these ecosystems not only learn from user behaviour but also influence and modify user preferences in return.
Furthermore, the use of algorithms and big data enables the central nodes within these networks to exercise panopticon power, control the competitive process by picking winners and losers, and dictate the direction, quality and nature of innovation. In that way, digital ecosystems are replacing markets as public institutions – where Smithian sellers and buyers once interacted – into entirely privatised and privately controlled digital transaction spaces dominated by single entities (‘cloud fiefs’). In this context, relationships of ‘servile dependency’ emerge with digital barons reinforcing hierarchies and dictating the ‘rules of the game’ while maintaining the façade of competition.
To avoid or reverse these troubling trends a singular focus on consumer welfare – defined solely in terms of prices and output – is inadequate. If competition law were to prioritize only consumer welfare, it would fail to tackle behaviours that, while consumer welfare-enhancing, undermine effective competition. This oversight could hinder digital environments’ ability to generate and distribute value fairly, as well as to foster both sustaining and disruptive innovation. Most critically though, given the connection and the multi-faceted nature of Big Tech’s power, such a failure could allow distortions of competition to adversely affect democratic decision-making processes.
If competition authorities remained captivated by the consumer welfare orthodoxy and the neoclassical economics paradigm, they risk overlooking how digital ecosystems can amass significant techno-economic and political power, ultimately threatening both industrial and political democracy (see here, here and here). Therefore, a serious revaluation (see for example here and here) is necessary to explore how competition law can better promote economic democracy (in that regard see also this fascinating monograph). In this context, the ordoliberal competition-democracy nexus and the Smithian ideal of republican freedom regain relevance. These concepts could serve as animating principles for competition authorities as they seek to refine their analytical framework and recalibrate their enforcement strategies to protect effective competition.
*All references to The Wealth of Nations are derived from the edition titled An Inquiry into the Nature and Causes of the Wealth of Nations by Adam Smith, published in 1776 by Oxford University Press in 1993.
Stavros Makris
Dr Stavros Makris is a Lecturer in Law at University College London. Prior to joining UCL, Stavros was a Lecturer in Competition Law at the University of Glasgow (2022-2024), a Guest Lecturer at UCL (2023-2024), a Fellow in Law and a Guest Teacher at LSE (2021-2023), a Postdoctoral Researcher at Wageningen University & Research (2020-2021), and a Teaching Fellow at SciencesPo (2018-2019). Stavros holds an LLB and an LLM in Philosophy of Law (with distinction) from the University of Athens, Law School, an LLM (with distinction) from University College London (UCL), and he completed his PhD at the European University Institute (EUI).
