The three pillars of effective European Union competition policy

Executive summary

Competitiveness is a region’s ability to achieve high productivity, attracting businesses, creating jobs and fostering innovation. It stems from efficiently using resources, is driven by competitive markets and is supported by three pillars: competition policy, procompetitive industrial policy and regulation.

Competition enforcement keeps markets competitive by preventing harmful practices such as cartels, dominance abuse and anticompetitive mergers. While there is a push to relax European Union merger laws to promote European ‘champions’, this risks inefficiency and monopolisation. Instead, expanding the EU’s market size while maintaining competition is essential, especially in digital sectors.

Procompetitive industrial policy addresses market failures and fosters well-functioning markets by targeting externalities such as worker training, R&D and infrastructure. Unlike outdated policies favouring national champions, this approach should focus on EU-level initiatives that promote competition and deepen the single market. This allows firms to achieve scale and drive innovation.

Monopolised markets, whether state-created or naturally formed, need regulation to ensure competitive outcomes. The EU Digital Markets Act (DMA) aims to enhance competition and innovation on digital platforms by mandating data sharing, non-discriminatory access and interoperability. However, resistance to compliance on the part of tech giants poses enforcement challenges. Without effective and timely enforcement, trust in the law will erode, stalling innovation and reducing Europe’s global regulatory influence.

When competition enforcement, procompetitive industrial policy and monopoly
regulation work together, markets benefit consumers, firms and workers. Effective EU policy will refine competition enforcement, regulate monopolies where necessary and redesign state aid strategically to create welfare-enhancing markets. This comprehensive approach ensures an economy that serves its people and remains competitive.

There is an undercurrent of belief in Europe that competition and industrial policy are in direct conflict, and competition enforcement is a culprit in lagging European Union competitiveness. The French and German governments, for example, in a ‘manifesto’ issued in 2019, argued for changes to European Union competition rules including merger control modifications to give companies the best chance of scaling up, and a right for EU governments meeting in the Council of the EU to override decisions of the European Commission – the EU’s competition enforcer – in certain cases (BMWK and Ministère de l’Économie, 2019). The issue was taken up by European Commission President, Ursula von der Leyen, in her 18 July 2024 pitch for reappointment for a second term:

“I believe we need a new approach to competition policy, better geared to our common goals and more supportive of companies scaling up in global markets – while always ensuring a level playing field. This should be reflected in the way we assess mergers so that innovation and resilience are fully taken into account” (Von der Leyen, 2024).

This Policy Brief clarifies the relationship between competition and industrial policy, and shows that there is in fact no tension. It explains that there is no conflict between enforcement of EU competition laws and scale (see section 2.2). A conflict between competition and industrial policy only arises when industrial policy is deployed destructively. Industrial policy is one of the tools that can create competition, though it can also be used to suppress competition, and pro-competitive industrial policy supports and complements competition enforcement. Merger control can already incorporate a loss of resilience and innovation when those factors threaten to harm consumers.

Competition policy meanwhile goes beyond enforcement and encompasses industrial policy, which can help grow market size, in turn helping firms achieve scale. Competition policy also encompasses state aid, whether that aid comes from governments inside or outside the EU. Government subsidies distort the level playing field and make it more difficult for unsubsidised but efficient firms to obtain scale. All of this matters because competition is key to achieving Europe’s goals. Firms are driven to be productive when they face vigorous competition for the business of consumers. Without improving quality, reducing prices and innovating they will not attract consumers when those consumers have other places to buy. When those firms succeed in attracting consumers, they attain scale. Those customers can be attracted from within the EU or from outside, particularly when the barriers to competing across borders are minimised.

High productivity makes a region competitive for both firms and inputs like capital and workers. The ‘competitiveness’ of a region is this productivity (Pinkus et al, 2024). It attracts the businesses that deliver the tangible outputs people want to consume and the jobs they want to fill. Such environments are also attractive for innovation and the formation of new businesses and industries. Productivity comes from using resources in the most efficient manner possible in order to produce greatest quantity and value of outputs. The best way we know to achieve that goal is well-functioning competitive markets. A major role for the state, therefore, is to create conditions conducive to such markets and protect competition within them. 

The current debate over industrial policy in Europe often seems to interpret competitiveness to be a country ‘winning’ against other countries because it hosts bigger firms. But competitiveness does not require a region’s large firms to hold the ‘number one’ position in their respective industries. Ideas like first place versus second place are imported from domains such as sport and do not translate well into an economic context. Having said that, large fixed costs and scale are important to productivity in many industries, and therefore the ability to grow in absolute size is critical for governments to facilitate, as this Policy Brief discusses. A government should want to attract and nurture such productive firms, as well as the complementary assets and input markets that tend to grow with them. But competitiveness does not involve politicians collecting large firms in the way men did in past ages with silver arm rings, ships of the line or space rockets. Instead, the best policies focus on a different goal – the well-being of society – while the birth, growth and longevity of productive firms in a jurisdiction are a result of succeeding at that endeavour.

Markets for goods, labour and capital that run smoothly, competitively and at scale will benefit every size of firm. But entrepreneurs and small firms may benefit especially because they cannot create these markets themselves. Thus, productivity reinforces itself by enabling entry and incentivising innovation, creating yet more competition and choice. Competition policy plays a pivotal role in this framework because markets will not serve consumers, current enterprises or the next generation of innovative firms unless they are competitive.

Competition policy helps a region’s economy grow in three ways. First, competition enforcement as it is carried out today protects existing competition in markets that already work reasonably well. Prohibiting an anticompetitive merger falls in this category. Second, procompetitive industrial policy enables broken or poorly performing markets to exist or perform better, become competitive and deliver good outcomes for consumers. Subsidies for green technologies and worker training in a new technology fall into this category. Regulation, the third leg of the stool, limits harms from already monopolised markets that have structures that cannot be made competitive. Regulation of a water or electric utility is a familiar example.

All three of these activities constitute competition policy, as can be seen today when the European Commission’s Directorate-General for Competition enforces Articles 101 and 102 of the Treaty on the Functioning of the European Union, the EU Digital Markets Act (DMA, Regulation (EU) 2022/1925) and state aid rules ¹ Competition enforcement is usually considered to be the process of ex-post enforcement against firms the government believes have violated competition law. . However, competition policy manifests itself in other parts of government, such as when agencies setting spectrum policy or banking regulations invite or block entrants. Competition policy operates analogously at the member-state level in local contexts.

This framing makes it clear that these three tools of enforcement, regulation and industrial policy are complements, not conflicting alternatives. Each tool fixes a different type of problem that arises in capitalism, a problem that limits the ability of competition to deliver the outcomes consumers want. Governments can repair each problem with a tool designed for that problem without disrupting competition by favouring any one undertaking. When governments use all three tools of competition policy, they serve existing and nascent firms, and deliver the productivity and employment that consumers and workers want.

Competition enforcement aims to ensure that markets that generally function well remain competitive and fair. It does this by preventing structures and conduct that could undermine competitive dynamics and that would harm trading-partner welfare. EU competition rules control the formation of cartels, abuse of dominance and mergers.

2.1 Basic competition enforcement

Explicit collusion involves arrangements to engage in price fixing, division of markets, bid rigging, predatory pricing, pay-for-delay, price parity and other practices that block competition and raise prices ² See Article 101 TFEU, https://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX%3A12008E10…. . Such practices obviously harm consumers as well as businesses buying from a cartel, or suppliers to a cartel, and no region will obtain high productivity without prohibiting this sort of conduct.

EU merger rules prohibit mergers when they may harm competition inside the EU; analysis must centre on the options available to EU consumers. Lessened competition for European consumers reduces quality or enables monopoly prices through unilateral and coordinated effects ³ See Article 102 TFEU, https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX%3A12008E…. . Allowing close competitors to merge also threatens the creation and exercise of monopsony power by increasing the bargaining power of the merged entity over its supply chain, creating entry barriers for other buyers and reducing wages and working conditions in labour markets.

The emerging narrative in Brussels argues that concerns about competition and its benefits should take a back seat to the desire to have national champions that will – after receiving large subsidies – demonstrate the competitiveness of Europe in areas including telecoms, banking and cloud computing ⁴ Emanuel Macron, the President of France, is one example. See The Capitol Forum, ‘Despite a Weakened Macron, His Fight for European ‘Champions’ Vs. Competition Enforcement Isn’t Going Anywhere, Experts Say’, 23 July 2024, https://thecapitolforum.com/weakened-macron-fight-for-european-champion…. . But what is achieved when merger control is abandoned instead of making improvements to expand or deepen the single market? Without competition enforcement, rivals will merge within a member state to achieve operating synergies. There will be a French telecom monopoly and a German telecom monopoly. There will be Dutch banking monopoly and a Spanish banking monopoly, each of which will go to their governments for subsidies and will be too big to fail. But none of those monopolists will have the scale they should have because they do not operate across a market of 27 countries with 400 million consumers – which is the market size available to telecoms and banking firms in the United States and China. A heavily-subsidised small French telecoms firm will not have low enough costs nor strong enough incentives to expand into Brazil to take share from US, Chinese or Brazilian competitors. Meanwhile French consumers must bear monopoly prices and monopoly quality. This is why continuing to enforce merger laws is critical to European productivity.

Those merger laws do not prevent European firms from achieving scale. First, firms can merge across member-state lines in industries such as telecoms or supermarkets. When undertakings already sell across the whole EU, there are two types of mergers. In the first, the relevant market is only Europe because there are no substitute products available from outside the EU. In this case, a merger between significant EU rivals can harm competition. Today that merger will be prohibited. As a consequence, those firms will have to compete vigorously on the merits ‘at home’. That competition creates the capabilities and efficiency that will allow a firm to succeed globally – as is shown by many globally successful European firms today – and raises EU competitiveness.

If, on the other hand, the market is global and foreign competitors operate in the EU, then a merger of two local firms might not lessen competition materially for European consumers. The analysis of such a merger will reflect that reality. European market shares will not be very relevant in an investigation if competition is global because they do not reflect the true state of the choices available to consumers. Such a merger will likely be permitted when competition from outside the EU is robust, for example, and the global shares of the merging firms are relatively small.

Merger control is perceived by some in Europe as being inconsistent with development of European national champions. The idea is that the combination of two medium-sized European firms can create a large European firm that will then compete with a large Chinese or American firm on a global scale, and this will bring benefits to Europe that the two medium-sized firms could not. The Alstom-Siemens proposed merger blocked by the Commission in 2019 had elements of this narrative ⁵ See Konstantinos Efstathiou, ‘The Alstom-Siemens merger and the need for European champions’, Bruegel Blog, 11 March 2019, https://www.bruegel.org/blog-post/alstom-siemens-merger-and-need-europe…. .

This idea has superficial appeal, but when confronted with the framework actually used to analyse mergers, the logic breaks down. First, current law permits European firms to work together to compete outside Europe. This is because European competition law is concerned only with protecting European consumers and not others located elsewhere. So, European competitors that want to work together to procure inputs or build a business outside Europe are free to do that. For example, European enterprises could join together to market a product in North America or procure raw materials in Africa.

Second, as described above, merger control will not prohibit mergers where competition is vigorous. If Siemens and Alstom face competition in Europe from Chinese and American firms, then global competition operates in Europe, in which case the merger would likely be approved. Alternatively, global competition is not present in Europe, in which case the European firms do not face a threat from the foreign rival, and preserving competition at home requires that they operate independently.

It may be that competition in the EU occurs between local firms today, but future expected competition will come from growth and entry of, for example, Chinese firms, and the European firms are merging because they anticipate this challenge. There is no conceptual prohibition against making an argument that competitive conditions are about to change and this is a good reason for a merger. The difficulty for the regulator is that any pair of merging parties stands to gain from making make such a claim, and therefore many will do so whether or not it is true. The claim that competition is intensifying and competitors are about to enter is made very often by merging parties. Parties have ‘cried wolf’ so often that regulators do not find the claim credible unless backed by good evidence. If the regulator has evidence that entry by foreign firms with scale advantages is happening or is imminent, this may render an EU transaction harmless. If the entry is a hypothetical idea far in the future, then it is likely the firms should wait to use this reason to justify a merger until there is better evidence of the foreign entry and its nature.

A region’s ability to care for its people depends not only on the levels of income, but also on risk. Shocks in the form of pandemics, floods and wars can cause price spikes and shortages. These harm the firms concerned, but in a competitive market, much of the harm from risk falls on consumers. Greater resilience in the face of shocks incentivises investment, allows both firms and consumers to spend less resources on insurance of all kinds and makes a region more competitive. Competition enforcers may find that a particular merger will reduce resilience and raise risk in a way that harms consumer welfare as the term is normally used. For example, the claimed efficiency of a merger could be the consolidation of the parties’ supply chains into one. If all goes well this will reduce costs, but with pandemics, floods or wars arriving over time, the reduced resilience will cause higher prices and shortages. On average, prices may end up being higher, especially when considering that a shortage means consumers must buy a different good or none at all.

While this theory of harm is also forward-looking, it is amenable to analysis using past distributions of shocks (weather, war) and modelling of supply and demand. A regulator can use economic tools to estimate outcomes and their distribution to determine if, on net, consumers are harmed, even if prices would be lower in good times. The existing frameworks for merger review can already accommodate this theory of harm.

2.2 The importance of scale

Every year a larger fraction of the economy falls into a category where scale is important to achieving productivity. This segment clearly includes digital platforms, but also increasingly includes other industries as data becomes more important in helping firms increase quality and reduce costs. When economies of scale are present, a firm’s larger scale leads to lower costs.

Minimum efficient scale (MES) refers to the smallest scale at which a company can operate while matching the lowest costs in the industry. A firm smaller than this absolute size will have higher costs than its larger competitors. The relationship between MES and the size of the EU market varies greatly across industries and over time. When MES is small relative to the size of the EU market, competition works well. When MES is large relative to the EU market, there is a conflict between sustaining competition among multiple firms and achieving efficiency, as explained above. Alternatively, if there is fair trade (without distortionary state subsidies) then the market can be global and support more than one firm at MES, restoring the role of competition.

Some believe that competition policy must be weakened to permit firms to achieve the scale necessary for commercial success ⁶ For example, see Sven Smit and Jan Mischke, ‘Scale matters more than ever for European competitiveness. Here . In particular, suppose that for European firms to succeed they must be able to grow to a larger size in Europe. The first route to growth is through a larger market in which to expand without encountering barriers. This allows for greater scale without any reduction in the number of firms. Conditional on market size, there are two ways to achieve scale. One is by engaging in exclusionary conduct so rivals do not grow. Competition laws are designed to prohibit this type of conduct because it reduces welfare. Another way to grow increases welfare – offering a superior product that consumers prefer relative to rival products. The resulting sales create low costs due to scale economies. This latter approach does not require weaker antitrust or merger control, but rather excellent products and execution on the part of the company.

The EU policy that is most helpful to firms with excellent products is expansion of the effective size of the single market and assurance of robust competition within it. A large home market becomes a place where multiple successful firms can first gain traction and scale on a path to becoming global. Because antitrust enforcement is used sparingly by European enforcers, enforcement rarely stops firms from growing and becoming large even if they engage in illegal conduct. Of course, if they have engaged in illegal conduct, antitrust enforcement is desirable. Procompetitive strategies should never be the target of antitrust enforcement. Mistakes are possible, but these cannot be numerous because of the small total number of antitrust cases brought by enforcers. Additionally, recent abuse of dominance cases (outside energy) have overwhelmingly been focused on non-EU firms, such as Intel, Google and Microsoft.

The longstanding problem of natural monopoly remains, however. If MES is large relative to the EU market and the good is not traded, there is a conflict between sustaining competition among multiple firms and achieving efficiency in Europe. Society can either accept some inefficiency due to small scale in order to achieve competition (there being none coming in from outside the EU), or achieve efficiency through a monopoly, which is then regulated. Good regulation of such ‘natural monopolies’ is better for consumers than allowing these companies to exercise their market power. While newer monopolies, such as digital platforms, are more complicated than older ones such as water utilities, there remain many tools and rich data regulators can use in the design process and in measuring outcomes. An alternative policy to regulation is to open up borders to enable a larger market, provided however, that the trade is free and fair so that it restores the role of competition ⁷ The EU Foreign Subsidies Regulation (Regulation (EU) 2022/2560) is designed to ensure competition from firms located in other countries is fair. .

2.3 Challenges in competition enforcement 

There are some challenges to regulating markets well that are either simply difficult to solve with competition enforcement or need a new tool fit for purpose if the EU wishes to solve them. Tacit collusion, for example, may be easier to create or sustain in an environment of high inflation and therefore may be a bigger problem than in past years, and yet there is no good EU-level tool to tackle it. Consumers with behavioural biases such as excess inertia or excess optimism may not be able to discipline competition in some markets because they do not choose the most competitive product. Again, there is no obvious tool for a competition enforcer to use to reform these markets so that consumers are not exploited.

Controlling mergers between small firms or startups and established incumbents has become more important to competition enforcement over time. Because some of these low-turnover mergers are competitively significant at EU level, it is crucial that the European Commission should have an accepted and settled way to obtain jurisdiction over them ⁸ The EU Court of Justice ruled out the process used by the Commission in the Illumina-Grail merger. See judgement in C-611/22 P - Illumina v Commission, 3 September 2024, https://curia.europa.eu/juris/liste.jsf?num=C-611/22&language=en. . Without such jurisdiction, established monopolists can simply acquire the rivals that threaten them, provided they identify rivals early enough when they have little revenue. An exciting disruptive entrant may well have a high market capitalisation and little revenue. A tool is needed particularly because innovative and disruptive products can be identified by the monopoly incumbent more quickly than by the regulator and – without advance notification requirements – purchased before the regulator realises there is a problem. Once the small firm has been acquired, it is in the interest of the incumbent to end the innovation competition between the two parties – to the detriment of consumers. Weak powers of review prevent the Commission from reviewing these mergers, which is a grave threat to innovation.

Competition policy protects competition in markets that generally work. When the government intervenes to fix broken or poorly performing markets and improve competitive conditions, this is industrial policy (Juhász et al, 2023). The goal of procompetitive industrial policy is to bring more valuable markets into existence and ensure they are competitive, so that they deliver good outcomes for society.

The idea of industrial policy is not new. The name we know it by is ‘state aid’ though this does not connote a strategic purpose for the aid in the way ‘industrial policy’ does. The classic, and discarded, form of industrial policy is for the government to hand out a large subsidy to a well-connected firm, which is then supposed to become a national champion. Such programmes rarely work and are a waste of taxpayer resources, in addition to distorting the single market. This is why state aid is regulated in Europe and why it is regulated by the competition authority. The old style of industrial policy – subsidies to well-connected incumbents – has cemented opposition to industrial policy in many quarters. However, there are important market failures in significant industries, and these market failures sometimes prevent these industries from existing at all or working well. The potential gains to society from a procompetitive industrial policy that fixes those externalities are significant (Goldberg et al, 2024).

3.1 Scale in the single market

Changes in the external environment make the payoff to European scale higher than in the past. The quickest solution for achieving scale for European firms is to ensure there is one market across the 27 EU countries rather than many small markets ⁹ Letta (2024) made this point. . In telecoms, the lack of common regulation of spectrum means that the economies of scale that might otherwise be achieved from operations in many member states are missing. Banking and financial services regulations prevent a true single market in capital. High-speed rail networks do not cross country borders smoothly. Energy markets are likewise separated. The reason for lack of scale is not merger control, but insufficient harmonisation within Europe.

Policies for deepening the single market are akin to a physician’s advice on weight loss: the answer is always the same, diet and exercise. While patients and politicians would prefer a quick and easy solution, the only thing that works is difficult: harmonise regulations and rules so firms can operate one business efficiently across many member states. One approach to speed up the process could be to create a new regulator and set of regulations that belong to all member states (‘a zeroth regime’), compliance with which allows a firm to operate in any member state ¹⁰ While this idea has appeared in several places badged as the ‘28th regime’, given the likely future growth in the number of EU countries, it seems prudent to place this regime in the 0th spot, rather than 28th so that its name does not need to change. . Abandoning competition enforcement will not work at all. Indeed, it would be counterproductive because industrial policy subsidies would then be spent by firms that did not feel competitive pressure to be efficient.

3.2 Examples of pro-competitive industrial policy

Pro-competitive industrial policy should be deployed at EU level to help with deepening the single market. Such a pro-competitive industrial policy is comprised of programmes that target specific externalities and articulate where the externalities come from, what harm they cause and how each externality can be mitigated or neutralised (Mazzucato et al, 2023). The discipline imposed by this explicit analysis helps prevent wasted expenditure and highlights policy tools that may be missing, but needed.

Examples of externalities are the positive spillovers from worker training, climate externalities, the gains from industry agglomeration that requires coordination among suppliers, producers, workers, and infrastructure. EU-wide policy designs that overcome these externalities could include subsidies for worker training, government support for infrastructure and zoning or coordination on a standard for a new technology. An important example of an externality is the unwillingness of private parties to invest in R&D when they cannot obtain property rights and will only capture a small share of the benefits, while much of the gains flow to competitors or society as a whole. EU-funded R&D can correct for underinvestment and stimulate entry of more innovators or catalyse an industry. However, it is important that any subsidies be competitively awarded and that conditions apply to the winners that promote competition. For example, a successful subsidised innovator might be required to license its innovation without charge to other innovators trying to build on the idea. In general, the public does not benefit when taxpayer funds create monopolies in areas that could have competition.

Another common form of externality is the need for agglomeration of an industry in one location to allow firms to benefit from shared workers, equipment, ancillary services and knowledge. A government can build infrastructure, fund universities that set up training programmes and subsidise investments of firms in that location. A third example is the insufficient incentive for manufacturers to invest in resilience. When a war or pandemic disrupts the supply of medications, losses are borne by both suppliers and consumers. However, because most of the surplus in the market is captured by consumers, manufacturers do not fully internalise the cost of supply disruptions and do not invest sufficiently to protect them. For example, society might want a secure supply of surgical masks in case of emergency. But when times are good, hospital buyers seek to buy from the cheapest supplier even if that supplier is not located in Europe. Relying on one distant supplier may cause a shortage when the next crisis hits. Buying from multiple suppliers, including one or more inside the EU, is more expensive but more secure. Pro-competitive industrial policy includes solutions for problems like these to increase resilience at EU level.

Another problem is conflicting standards in EU countries in areas including spectrum management, banking and land-use planning. For example, countries might have different regulations for electric-car charging stations, or none at all. Such disparate standards may dissuade investment by private parties as they wonder which version will become most popular. A private standard-setting organization cannot change the rules in member states. However, EU industrial policy can gather industry participants to establish common planning rules, a common engineering template to handle high voltages and common application programming interfaces (APIs) between charging points and cars. Such a scheme has the benefit of being inexpensive because the private sector is ready to do the investment; but by the same token the private actors will not invest until they have certainty about what the standards are.

In cases in which member-state regulators must harmonise local standards, a modern EU industrial policy should add conditional incentives – whether subsidies, market access or something else – to the firms participating. These subsidies should come from an EU funding source and need not be large, but it is critical that they are conditional on a company’s member state having reformed its regulations. Then an EU country that wants its firms to receive subsidies will face pressure to adopt the EU standard. If many states adopt the common template for a charging station, a construction firm would be able to tender to build the standardised charging station in thousands of locations across Europe. This would allow those suppliers to achieve economies of scale in equipment and construction. It is also important that such schemes be designed to be open in order to promote competition. For example, the APIs used by the charging station should be available to all sellers of electric vehicles in Europe, thereby encouraging entry of small new manufacturers. Furthermore, a huge network of charging points will increase adoption of electric vehicles, further incentivising investment and entry of new manufacturers.

Member states’ previous budgets for corporate subsidies can be combined and re-targeted to create a European fund for pro-competitive industrial policy. Since money is scarce, subsidies should be targeted at important externalities and satisfy many EU goals at once: the green transition, innovation, resilience and so forth. They should all come with two conditions. First, any subsidised project must help deepen the single market so that firms inside and outside the target industry have access to more scale within Europe. Second, no project may create or maintain a monopoly (or other limit on competition) when more competitive market structures are possible.

All of these schemes would create markets that were not present before or improve markets that functioned poorly. In all cases, the scheme would increase competition and output. To reiterate, procompetitive industrial policy is entirely consistent with competition-policy principles. Because the two tools are used to fix different problems and both promote competition, there is no conflict between them.

Markets that work well can be preserved through competition enforcement; markets that do not exist at all or are badly broken can be rebuilt with procompetitive industrial policy. There remains another competition problem: markets that are monopolies. Some of these monopolies were established by the state, such as those for water or other utilities. Other monopolies have arisen naturally from demand and cost conditions in that market. Still others have been created by the conduct of the firm itself in a way that could not be prevented by available antitrust enforcement. For example, a firm may engage in some conduct that helps it to obtain favourable network effects, the market ‘tips’, and then entrants can no longer gain traction, so a monopoly becomes entrenched. In all of these cases, procompetitive regulation is a tool society can use to require the regulated firm to produce (closer to) the competitive output and price, rather than the monopoly output and price. Such regulation is familiar from the telecommunications industry and similar issues such as access pricing and shared costs arise for digital platforms that are natural monopolies (or duopolies), regardless of whether or not they obtained their market power by violating a competition law.

Most regulation of utilities occurs at a local level or at the level of the member state. Some financial services are regulated at EU level. But the most recent example that is relevant for competition is the case of digital platforms. With the Digital Markets Act (DMA), the European Commission has taken on the task of creating more competition on, and for, digital platforms.

4.1 Regulation in an increasingly digitalised economy

The DMA is a prime example of procompetitive regulation, adopted after antitrust enforcement of digital platforms failed to result in more competitive market structures, despite a decade of enforcement effort. The law’s goal is increased contestability and fairness in the digital markets dominated by gatekeepers (meaning large, hard to avoid platforms) with entrenched positions.

Many of the DMA rules reflect learning from past antitrust cases, when available remedies did not sufficiently improve competition in that market. By shifting to a different instrument – economic regulation – the competition enforcer can require data sharing, the non-discriminatory access of business users to gatekeeper platforms, interoperability and the right of business users to disintermediate the gatekeeper, among others. The idea behind these rights is that enabling business users to get onto gatekeeper platforms and compete for users there – while being treated equitably by the gatekeeper – will improve competition on the platform, and that will benefit end users. In addition, certainty of the right of access to the platform and to end consumers stimulates business users to invest.

But such entry does not immediately improve competition for the platform. Instead, over time a business user that is permitted to grow and evolve, and whom the platform is not legally permitted to cut off or harm once it becomes a threat, may offer future competition for the next generation of platform. Such an evolution will lessen the harms from the current monopoly platforms and benefit business users. If core platform services comply with the DMA rules, then business users in the EU will have many more opportunities for innovation. Entry into app stores, digital wallets, messaging, gaming, entertainment content and more will be technically easier and more profitable, while the business users will be protected from discrimination and expropriation.

4.2 Challenges for regulation 

Regulators will have to be creative in order to retain the welfare-enhancing network effect characteristics of digital products while allowing competition. For example, competition in app stores will be enabled if the approval process for the app remains with the gatekeeper and all app stores are permitted to distribute any approved app. Then developers can continue to write for the popular operating system (and society gains the advantage of network effects) while having a choice of price and functionality in distribution. An additional element of this regulation would be to require that an operating system gatekeeper create a universal search function that permits a user to search their handset for an app by name in order to find out which stores carry it.

In some cases where network effects are present, it is not clear how the size of the market compares to the scale needed for efficiency and therefore whether or not that market can support more than one firm. In the case of search, Google paid large sums to keep rival search engines out of the market. If those search engines could only have offered an attractive product with a large market share, Google would not have needed to pay to prevent users from trying them. Likewise, Google testimony in antitrust proceedings in the United States asserted that the secret to Google’s search quality is the algorithm, not the data ¹¹ Stefania Palma, Stephen Morris and Michael Acton, . This type of evidence raises the possibility that more than one search engine could reach efficient scale in the market today. Similarly, advances in artificial intelligence are surely changing the optimal market structure by altering the cost structure of search, the need for data and the benefits of differentiation, so a regulator does not know today what market structure will be feasible or efficient in AI. When there is uncertainty, a good first step is to open up the market to learn what entrants and consumers will do, and then tailor further remedies in response to the resulting outcomes.

The DMA is now fully in force but not all platforms appear to be complying with it. In spring 2024, the European Commission announced noncompliance proceedings against Alphabet, Apple and Meta[12]. More such proceedings seem likely as business users and civil society absorb the platforms’ compliance reports and attempt to exercise their rights. If enforcement of the law does not compel compliance within a short period of time, the law will lose legitimacy with society. Gatekeepers will view it as optional rather than required and will simply hire large legal teams to fend off the need to change their business practices. European business users will become disillusioned about their ability to access consumers freely through the gatekeepers. Without feedback from business users the law is difficult to enforce. But business users will have no incentive to help the Commission with enforcement if their access to the platform is not improved and the gatekeeper retaliates against them for identifying noncompliance. Investment and innovation by business users will fall.

Europe cannot be the world’s regulator if its regulations do not change real-world behaviour in a reasonable time period. Lack of progress in enforcing the DMA will also raise the question of whether digital platforms are too big and powerful to have to abide by EU law. This risk is a significant threat to the goal of obtaining competitive outcomes in these very important markets.

When the three pillars of competition policy are all present – enforcement, procompetitive industrial policy and regulation of monopolies – the result is an economy that works for the people[13]. This toolkit allows a society to create and maintain the welfare-enhancing markets its firms and consumers want. Good EU policy will involve making necessary improvements to competition enforcement to keep it fit for purpose, regulating monopolies as necessary and re-designing state aid to make more strategic and useful.

Well-crafted industrial policy can be a good use of resources and it certainly does not require competition enforcement to be weakened. Indeed, the public’s money works more effectively when the firms it aids feel competitive pressure to use funds wisely. When the public projects that are given subsidies and infrastructure are chosen on the basis that they will open up and grow markets that are suffering from imperfections, and those projects are not used to anoint a winner and close those markets to others, society gains from industrial policy. This is not a subsidy race; rather, it is pro-competitive industrial policy.

When competitiveness is understood as the productivity generated by all of societies’ resources, it is straightforward to see that competition supports competitiveness. The notion that competition policy is in conflict with ‘competitiveness’ is dangerously wrong. A large single market allows European firms to reap greater returns from innovation and investment and to achieve scale in a competitive environment. Subsidies or other aid targeted to achieve several social goals at once should be the most favoured in an environment with tight budgets. Firms facing competition have an incentive to use their resources wisely and carefully, which is what the public expects from taxpayer funds. The resulting innovation and efficiency will generate European competitiveness that benefits everyone.

BMWK and Ministère de l’Économie (2019) ‘A Franco-German Manifesto for a European industrial policy fit for the 21st Century’, available at https://www.bmwk.de/Redaktion/DE/Downloads/F/franco-german-manifesto-for-a-european-industrial-policy.pdf

Goldberg, P.K., R. Juhász, N.J. Lane, G. Lo Forte and J. Thurk (2024) ‘Industrial Policy in the Global Semiconductor Sector’, NBER Working Paper 32651, National Bureau of Economic Research, available at https://www.nber.org/papers/w32651

Juhász, R, N.J. Lane and D. Rodrik (2023) ‘The New Economics of Industrial Policy’, NBER Working Paper 31538, National Bureau of Economic Research, available at https://www.nber.org/papers/w31538

Letta, E. (2024) Much more than a market: Empowering the Single Market to deliver a sustainable future and prosperity for all EU citizens, available at https://www.consilium.europa.eu/media/ny3j24sm/much-more-than-a-market-report-by-enrico-letta.pdf

Mazzucato, M. and D. Rodrik (2023) ‘Industrial policy with conditionalities: a taxonomy and sample cases’, Working Paper Series IIPP WP 2023-07, UCL Institute for Innovation and Public Purpose, available at https://www.ucl.ac.uk/bartlett/public-purpose/publications/2023/oct/industrial-policy-conditionalities-taxonomy-and-sample-cases

Pinkus, D., J. Pisani-Ferry, S. Tagliapietra, R. Veugelers, G. Zachmann and J. Zettelmeyer (2024) Coordination for EU competitiveness, study for the European Parliament Economic and Monetary Affairs Committee, available at https://www.europarl.europa.eu/RegData/etudes/STUD/2024/747838/IPOL_STU(2024)747838_EN.pdf

Von der Leyen, U. (2024) Europe’s choice: political guidelines for the next European Commission 2024-2029, available at https://commission.europa.eu/document/download/e6cd4328-673c-4e7a-8683-f63ffb2cf648_en?filename=Political%20Guidelines%202024-2029_EN.pdf