The economic and political environment in which competition law is enforced and applied has changed extensively over the last decade. Globalisation and a renewed focus on industrial policy have generated calls for more flexible competition policy. Oxera Partner Sir Philip Lowe reflects on current debates around competition and industrial policy in Europe
This article is based on a keynote speech at the Italian Antitrust Association Annual Conference, Florence, Italy, 22–23 May 2019.
Before the global financial crisis that began in 2007, there was a general transatlantic consensus that effectively functioning and competitive markets were the best instrument to deliver the goods and services that consumers need and want. At the same time, it was accepted that rules were needed to ensure free and fair competition, so that businesses competed on a level playing field and consumers were not exploited or harmed by a powerful combination of firms acting together.
On this basis, competition law was deemed the most appropriate and least intrusive way of governing markets. Cartels were rightly regarded as undermining competition and benefiting producers at the expense of consumers. They were (and still are) regarded as anticompetitive per se, and were subject to the most severe sanctions. Mergers could harm competition by strengthening the market power of some firms or by leading to the acquisition of significant market power by others. Mergers had to be vetted ex ante because it was very difficult to unwind mergers that were already promulgated.
As far as other agreements between companies (as well as unilateral conduct by dominant companies) were concerned, there was nevertheless a general acceptance that these should be looked at on a case-by-case basis, with close scrutiny but without any presumption that they were anticompetitive.
It was recognised in parallel that governments could distort markets by favouring some firms over others, whether through subsidies or through legislation. Within the EU, state aid disciplines attempt to deal with this problem. Outside the EU, there are World Trade Organization (WTO) anti-subsidy provisions, although these are more time-consuming and difficult to implement.
The consensus on the benefits of competition and free (self-correcting) markets was accompanied by strong scepticism about traditional industrial policy. Government intervention in markets to create national champions and ‘pick winners’, or to preserve local jobs, was generally derided as ineffective and wasteful. The most that could be done was to create a favourable regulatory environment for innovation and enterprise.
Today’s world is different
The concept of an industrial policy has been rehabilitated in recent years. Marina Mazzucato, of University College London, has eloquently demonstrated how the economies of countries as neoliberal as the USA base their success as much on interventionist public policies as on private enterprise.1 In particular, technologies developed for military purposes are frequently put to use in the wider economy. Mazzucato has pointed to the benefits of transformative public action to promote innovation and industrial change.
Mario Pianta has also set out the case for an active industrial policy at national and European level.2 He argues that such a policy should necessarily include strong action to correct market failures, such as those related to climate change, and help build the capabilities for industrial development—through strong support to infrastructure investments, research and innovation, and education and training. But it should also help to ensure that domestic firms benefit from a framework of fiscal, regulatory and public procurement rules that grant them the strength and protection to compete internationally.
In globalised, hi-tech, winner-takes-all markets, expecting the free market to provide all the incentives for success is simply not enough. The state has a role to play. The extent to which this intervention is compatible with state aid rules can be debated. A first reading of any text on EU state aid disciplines always starts with the general Treaty ban on state aid except under certain conditions. Yet if you look more closely at the criteria for compatible state aid, there is wide scope for an active industrial policy, especially if some categories of aid fall under the General Block Exemption. However, the process of notifying aid to the European Commission inevitably results in a delay before the aid can be approved, which can be a disadvantage where swift action is needed. Block exemptions should therefore be applied wherever possible.
In addition, confidence in the ability of markets to produce the right outcomes has been significantly undermined, in particular by the financial crisis, but also because markets are increasingly globalised. Globalisation has resulted in unequal outcomes in many of our societies. People increasingly expect governments to protect them from firms that they perceive as exploiting them—whether that is by increasing prices, lowering quality and restricting choice, or by using personal data for financial gain.
Politicians and governments have not been insensitive to these arguments. Across countries, they have been more than willing to respond to complaints about, for example, excessive electricity or petrol prices.3 They have lambasted competition authorities and sectoral regulators for their lack of action, and launched initiatives for more control of retail prices.
Such government action may not be altogether populist or irrational. Excessive pricing and exploitation are abuses that competition law enforcement aims to eliminate. However, competition authorities have always had some difficulty in dealing with complaints about excessive pricing, searching in vain for benchmarks for fixing the ‘right’ or the ‘fair’ price. The European Commission struggled for many years with abuse of dominance cases on mobile roaming charges and interchange fees. The adoption of regulation in these two areas was accompanied by a sigh of relief by many competition lawyers and law enforcers.
These examples highlight the reality that competition law may not always provide the quickest and best solution to a competition problem.
- First, pursuing a competition law case can take a long time, and frequently longer than the process of passing legislation or regulation through a national parliament. Sectoral regulators can certainly act more quickly than either of these alternatives.
- Second, a competition law case results at most in a remedy in a specific case, and sets a precedent, but does not always provide a solution that affects the market as a whole.
- Third, competition problems are rarely isolated from other public policy concerns, such as consumer protection, jobs, unfair trade practices, intellectual property protection and data privacy. Frequently, it makes sense to legislate to deal with these issues in a holistic and unfragmented way. A competition law case can certainly stimulate discussion on the need for market-wide regulation, but cannot necessarily provide the optimal answer to a competition problem, nor to a problem which involves other public policy concerns.
The interface between industrial policy and competition law
The political concern that competition policy could frustrate the pursuit of industrial policy and other public policies is not new. In the 1970s and 1980s, there were already voices in Europe calling for a much more flexible application of merger control rules.
At the same time, there have been increasing concerns regarding the impact of the acquisition of domestic firms by foreign, and sometimes state-owned, companies.4 These acquisitions can lead to a transfer of the headquarters of firms to other countries, with a consequent dilution of the influence of local policies on the merged firm. They can also result in the relocation of assets—such as R&D facilities and qualified personnel—that are seen as intrinsic to the comparative advantage of an economy internationally. In addition, some foreign direct investment may place control of infrastructures or activities that are regarded as critical to national security and the long-term competitiveness of an economy in foreign hands—and even in the hands of foreign governments. Electricity and communication networks and nuclear power stations are frequently cited as examples of strategic or critical assets. For example, in the UK there are 13 critical national infrastructure sectors, including nuclear, communications, emergency services, and health.5
In March this year, in response to these concerns, the EU institutions adopted a new Regulation which provides for closer screening of foreign direct investment in relation to national security and public issues, including provisions for exchange of information on transactions between member states, and between member states and the Commission.6 Powers to prevent a transaction, or to impose conditions on it, however, remain in the hands of national governments. At present, 14 member states have legislation which can be used for this purpose.
It is important to emphasise that this additional screening and control implies only that mergers which have been approved on competition grounds may nevertheless be prohibited or amended on other policy grounds.
Article 21(4) of the EU Merger Regulation leaves some room for this kind of exception in order to take into account issues of national security, media plurality and financial probity. Member states can raise other policy concerns, but it is within the discretion of the European Commission whether these are justified. The new Regulation on screening of foreign direct investments certainly gives member states more explicit legitimacy and legal security when imposing conditions on transactions.
However, the current debate on industrial and competition policy focuses on the possibility that a merger which should in principle be prohibited on competition grounds could nevertheless be allowed to go ahead. The main argument advanced in favour of more flexibility in merger control at the EU level and at national level within Europe is that European firms need to reach a ‘critical mass’ and to reap economies of scale in order to compete with large, often state-owned, foreign competitors on global markets. Competition authorities, according to this line of argument, should also anticipate the growing strength and penetration in EU markets of foreign firms.
Arguably, this call for greater leniency in merger control is not aimed at introducing non-competition criteria into a merger assessment, but at allowing a government, or another politically controlled agency, to correct a competition authority’s assessment of the impact of a transaction on competition.
Furthermore, the implied direction of any correction of the merger assessment would not be to impose further constraints on mergers (such as is now being proposed in the digital economy to prevent killer acquisitions or to protect data privacy) but to free them from constraints that competition authorities would normally impose on them.
The current framework for merger control at the EU level excludes this. The dominant view at both European and national level in Europe has been that competition in a firm’s home market in Europe strengthens its capacity to compete abroad. In addition, consolidation is always possible provided that there is ‘no distortion of competition to the disadvantage of consumers’. Allowing governments to wave through otherwise prohibited mergers would expose merger control to capture by business interests and diminish legal certainty and predictability.
There is admittedly some scope for government intervention of the kind envisaged in the assessment of mergers which are below the EU thresholds, but not much. Under German legislation, the Ministererlaubnis enables the relevant minister to approve mergers on public interest grounds which have been prohibited in principle on competition grounds. However, during the 45 years since the law was adopted, there have been 22 applications for merger approvals under the Ministererlaubnis, of which only nine have been successful—and not all of these are based on a revised assessment of the competitive impact of a merger. Similar provisions exist in French, Italian and UK law, but they have been used sparingly.
A question of defining markets?
Naturally, there continues to be considerable debate about market definition in merger assessment.
Sometimes mergers can achieve significant efficiencies in most national markets but threaten the strength of competition in one national market. If the market is genuinely Europe-wide or global, an intra-European merger is unlikely to be problematic from a competition point of view. But if markets are defined nationally and narrowly, transactions such as the 2000 Volvo/Scania truck merger7 could be prohibited unless a remedy existed which could resolve the problem in the specific market (in this case, Sweden) where the overlap of the two firms’ activities could not be contested by current or potential competitors.
The issue of market definition remains a live one, and it can be controversial. Product, as well as geographical, market definition can crucially determine the outcome of any merger assessment. Some updating of the Commission’s 1997 Notice on Market Definition could well be necessary to reflect current and anticipated market structures and conditions.
But are there any grounds to justify a major change in the established international consensus that competition, and in particular merger control policy, poses no major constraint on industrial policy or economic growth?
The French and German governments seem to think so. After the Commission blocked the merger of the rail businesses of Siemens and Alstom earlier this year, the two governments published a joint manifesto calling for changes in the EU Merger Regulation and/or to the Commission’s Merger Guidelines to recognise the need for further consolidation in certain sectors—given strong competition from foreign, state-owned firms.8 The manifesto proposed that mergers involving foreign, state-owned firms should be subject to particular scrutiny. It also argued that the EU needs to develop a more forward-looking and dynamic assessment of competitive pressures in markets where non-European firms are gradually acquiring a global presence. The two countries foresee procedures in which the EU’s Council of Ministers could exceptionally approve a merger in the European interest when the Commission has prohibited it on competition grounds.
Implications for competition policy
There are strong arguments in favour of an active industrial policy at European and national level. Both state aid control and competition policy need to take account of the international dimension of markets, and a dynamic assessment of competitive pressures in markets is essential.
However, that assessment cannot be based on hunch or fantasy; it must be rigorous and realistic. And it needs to be carried out by an administrative authority that is subject to control by the courts, but is free from political interference and independent of business interests.
Firms should be able to extract efficiencies from a merger. Nevertheless, in any market where there is a competition problem—because there is no realistic prospect of a viable competitor to the merged entity—every effort should be made to devise a remedy which will allow an overall transaction to go ahead but avoid harm to consumers in the specific market concerned.
Finally, if firms are faced with unfair and subsidised competition from outside the EU, perhaps the most appropriate response should be provided by trade defence mechanisms and WTO anti-subsidy procedures, rather than by weakening competition law enforcement.