EUROPEAN COMMISSION

Brussels, 22.3.2017

SWD(2017) 114 final

COMMISSION STAFF WORKING DOCUMENT

IMPACT ASSESSMENT

Accompanying the document

Proposal for a

DIRECTIVE OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL

to empower the competition authorities of the Member States to be more effective enforcers and to ensure the proper functioning of the internal market

{COM(2017) 142 final}
{SWD(2017) 115 final}
{SWD(2017) 116 final}

Table of Contents

1CONTEXT

1.1Why enforcement of the EU competition rules matters

1.2The role played by the national competition authorities

2What is the problem and why is it a problem?

2.1Untapped potential for more effective enforcement of the EU competition rules

2.2Underlying problem drivers

2.2.1Problem driver 1: Lack of effective competition tools

2.2.2Problem driver 2: Lack of powers to impose deterrent fines

2.2.3Problem driver 3: Divergences between leniency programmes in terms of summary applications, core principles, protection of self-incriminating material and interplay with individual sanctions can lead to less effective competition enforcement against cartels

2.2.4Problem driver 4: Lack of safeguards NCAs can act independently when enforcing the EU competition rules and have the resources they need to carry out their work.

2.3Who is affected by the problem?

2.4What is the EU dimension of the problem?

2.5How would the problem evolve, all things being equal?

3Why should the EU act?

4What should be achieved?

5What are the various options to achieve the objectives?

5.1Option 1: no EU action (baseline scenario)

5.2Option 2: further soft action

5.2.1Specific objective 1: ensuring all NCAs have an effective competition toolbox

5.2.2Specific objective 2: ensuring deterrent fines can be imposed

5.2.3Specific objective 3: making leniency programmes and their interplay more attractive to encourage companies to cooperate with the authorities in their fight against cartels

5.2.4Specific objective 4: ensuring all NCA have safeguards they can act independently when enforcing the EU competition rules and have the resources they need to carry out their work.

5.3Option 3: EU legislative action to provide NCAs with minimum means and instruments to be effective enforcers, complemented by both soft action and detailed rules where appropriate

5.3.1Specific objective 1: ensuring all NCAs have an effective competition toolbox

5.3.2Specific objective 2: ensuring deterrent fines can be imposed

5.3.3Specific objective 3: making leniency programmes and their interplay more attractive to encourage companies to cooperate with the authorities in their fight against cartels

5.3.4Specific objective 4: ensuring all NCA have safeguards they can act independently when enforcing the EU competition rules and have the resources they need to carry out their work.

5.4Option 4: EU legislative action to provide NCAs with detailed and uniform means and instruments to be effective enforcers

5.4.1Specific objective 1: ensuring all NCAs have an effective competition toolbox

5.4.2Specific objective 2: ensuring deterrent fines can be imposed

5.4.3Specific objective 3: making leniency programmes and their interplay more attractive to encourage companies to cooperate with the authorities in their fight against cartels

5.4.4Specific objective 4: ensuring all NCA have safeguards they can act independently when enforcing the EU competition rules and have the resources they need to carry out their work.

5.5Discarded option

6What are the impacts of the different policy options and who will be affected?

6.1Option 1: no EU action (baseline scenario)

6.2Option 2: further soft action

6.3Option 3: EU legislative action to provide NCAs with minimum means and instruments to be effective enforcers, complemented by soft action/detailed rules where appropriate

6.4Option 4: EU legislative action to provide NCAs with detailed and uniform means and instruments

7How do the options compare?

7.1Comparison of the options

7.1.1Option 1: no EU action (baseline scenario)

7.1.2Option 2: further soft action

7.1.3Option 3: EU legislative action to provide NCAs with minimum means and instruments to be effective enforcers, complemented by soft action/detailed rules where appropriate

7.1.4Option 4: EU legislative action to provide NCAs with detailed and uniform means and instruments

7.1.5Overview of how the options compare

7.2Preferred policy option

7.2.1All NCAs would have an effective enforcement toolbox

7.2.2All NCAs would be able to impose (or seek) deterrent fines

7.2.3All NCAs would have effective leniency programmes which would encourage companies to come clean across the EU

7.2.4All NCAs would have guarantees of independence and resources

7.2.5Nature of the instrument and legal basis

8Monitoring and evaluation

1CONTEXT

1.1Why enforcement of the EU competition rules matters

The EU competition rules seek to provide everyone in Europe with better quality goods and services at lower prices by ensuring that firms compete solely on their merits.

In a competitive market, the simplest way for a company to gain more market share is to offer a better price than its competitors. This is not only good for consumers - when more people can afford to buy products, it encourages businesses to produce and boosts the economy in general.

Competition also encourages businesses to improve the quality of goods and services they sell to attract more customers and expand market share. Quality can mean various things: products that last longer or work better, better after-sales or technical support or friendlier and better service. In a competitive market, businesses will try to make their products different from the rest so they are more attractive. This results in greater choice: consumers can select the product that offers the right balance between price and quality. To deliver this choice, and produce better products, businesses need to be innovative in their product concepts, design, production techniques, services, etc.

Competition within the EU also helps make European companies stronger outside the EU and able to hold their own against global competitors. Effective enforcement of the EU competition rules is therefore indispensable for the Union to both meet the challenges, and seize the opportunities, of globalisation.

The Treaty on the Functioning of the European Union (TFEU) recognises the importance of fair competition in the internal market. Article 101 TFEU prohibits agreements between two or more companies which restrict competition. The most flagrant example of illegal conduct infringing Article 101 TFEU is the creation of a cartel between competitors to fix prices, to limit production or to share markets or customers between them. Instead of competing with each other, cartel members rely on an agreed course of action, which reduces their incentives to provide new or better products and services at competitive prices. As a consequence, their clients (consumers or other businesses) end up paying more for less quality. Prohibiting a cartel therefore has a direct positive effect on consumers, business and competition as a whole.

Article 102 TFEU prohibits firms that hold a dominant position on a given market from abusing that position, for example by charging unfair prices, by limiting production, or by refusing to innovate, all of which is to the detriment of consumers' interests and may lead to the exclusion of competitors from the market by other means than competing on the merits of the products or services provided. Stopping companies from abusing their dominant position allows other companies a chance to compete fairly and to grow.

Competition enforcement is about applying rules to make sure companies compete fairly with each other. This means that businesses benefit from a level playing field, allowing them to compete on their merits and capitalise on any competitive advantage they may have. Greater competition encourages enterprise and efficiency, which in turn benefits consumers.

Competition enforcement in Europe is a vital part of the single market . The single market is recognised as Europe’s best asset in times of increasing globalisation. The Commission is determined to build on the strength of the single market by making it fairer and deeper. 1 The European competition rules are one of the defining features of the single market: if competition is distorted, Europe cannot deliver on its full potential. Competition is not an end in itself. It is an indispensable element of a functioning internal market guaranteeing a level playing field. It contributes to an efficient use of society's scarce resources, technological development and innovation, a better choice of products and services, lower prices, higher quality and greater productivity in the economy as a whole.

1.2The role played by the national competition authorities

The EU Member States are essential partners of the European Commission for enforcing the EU competition rules. Since 2004, the national competition authorities of the EU Member States (NCAs) are empowered by Regulation 1/2003 2 to apply the EU competition rules alongside the Commission. Indeed, the NCAs are obliged to apply the EU competition rules to agreements or practices that are capable of having an effect on trade between Member States. For more than a decade both the Commission and the NCAs have enforced the EU competition rules in close cooperation in the European Competition Network (ECN). The ECN was created in 2004 expressly for this purpose. The ECN is recognised as a successful and innovative model of governance for the complementary implementation of EU law at both European and national levels.

Enforcement of the EU competition rules by both the Commission and the NCAs is an essential building block for the creation of an open, competitive and innovative single market and is crucial for creating jobs and growth in important sectors of the economy, in particular, the energy, telecoms, digital and transport sectors.

Enforcement of the EU competition rules is now taking place on a scale which the Commission could never have achieved on its own. Since 2004, the Commission and the NCAs took over 1000 enforcement decisions, with the NCAs being responsible for 85%. Action by a multiplicity of enforcers is a much stronger, more effective and better deterrent for companies that may be tempted to breach the EU competition rules. The Commission typically investigates anticompetitive practices or agreements that have effects on competition in three or more Member States or where it is useful to set a Europe-wide precedent. The NCAs are usually well placed to act where competition is substantially affected in their territory. NCAs have the expertise on how markets work in their own Member State. That knowledge is of great value when enforcing the competition rules. Action at national level promotes grass root support for competition enforcement. 3

The NCAs have tackled many anticompetitive practices with a direct impact on consumers and citizens/tax payers. For example, at the end of 2014, the French NCA fined manufacturers of home care and personal care products nearly €1 billion for having coordinated their commercial policy towards supermarkets, which allowed them to maintain artificially high prices for the end consumers. The level of the fines is linked to the significance of the concerned markets, namely around €4.7 billion for home care products and €7 billion for personal care products. The practices also had an important impact on consumers as these products account for a significant share of household spending as at the time of these practices consumers in France spent an average of €190 per year for personal care products alone. NCAs have also uncovered and sanctioned a significant number of cartels in the food sector. For example, in 2014, the Greek NCA fined poultry meat producers a total of €39.9 million for fixing prices of fresh and frozen poultry meat and for allocating customers.

Enforcement by the NCAs also benefits citizens/tax payers more generally where they uncover and bring practices to an end which are at the expense of public bodies, such as municipalities or the social security system. In 2011, the German NCA imposed fines totaling €20.5 million on manufacturers of fire-fighting vehicles for having fixed the prices and carved up the German market for fire-fighting vehicles among themselves. The infringement covered more than 80% of the market and caused many municipalities considerable financial harm. In 2014, the Italian NCA imposed fines of more than €180 million on two pharmaceutical companies for having hindered the use of an inexpensive drug for treating common eye diseases among elderly people, meaning that patients were forced to buy a more expensive product. The agreement resulted in an additional expense for the national health service estimated at more than €45 million in 2012 and incurring possible future costs of more than €600 million per year.

The NCAs thus play a key role in making sure that the single market works well and fairly. This benefits both businesses which can compete more fairly on more open markets throughout Europe, as well as consumers who get a better choice of goods at lower prices. However, there is potential for the NCAs to do more.

2What is the problem and why is it a problem?

2.1Untapped potential for more effective enforcement of the EU competition rules

A key goal of the decentralised system for the enforcement of the EU competition rules that was put in place in 2004 was to ensure greater, more effective enforcement by a multiplicity of authorities through-out the EU.

The European Commission estimated the annual customer benefits 4 of only its decisions prohibiting cartels (e.g. excluding other enforcement action under Articles 101 and 102 TFEU) for the period 2010-2015. This varied between €1 billion up to €10.8 billion, depending on the year considered: 5

Model simulations show that the European Commission's cartel decisions and merger interventions have a sizeable impact on growth and jobs: GDP increases by 0.4% after five years and by 0.7% in the long run, and after ten years around 650 000 cumulated new jobs are created (the relevant simulations are based on a chain of assumptions which, however, do not undermine the usefulness of the analysis in getting a better understanding of the role of competition policy in society 6 ).

At national level, the Dutch competition authority estimated that the outcome of its actions in terms of the expected future benefits for consumers who will pay less for products and services at €260 million for 2014 (the relevant assessment appears to cover also merger control, and the relevant calculation method leads to results that might vary from one year to the next, however, it provides a useful proxy for the impact of the authority's enforcement activity 7 ).

Similarly the UK competition authority estimated that the average direct financial benefit to consumers of only its antitrust enforcement activities was £73 million averaged over a three year period. 8  These figures only take into account the direct price effect of competition interventions for consumers and the total benefits extend beyond prices and include effects on quality, choice and innovation. 9 The UK competition authority also estimated that for the period 2013 to 2016 the ratio of direct benefits to cost was £10.6 for every pound of relevant cost to the taxpayer. 10 .

Despite these positive outcomes, there is untapped potential for more effective enforcement of the EU competition rules by the NCAs.

While it is difficult to determine the effects on economic growth from a lack of effective competition law enforcement, it has been estimated that each year losses of €181-320 billion – approximately 3% of EU GDP – accrue owing to the existence of undiscovered cartels. 11 It is difficult to assess how many anti-competitive cartels are undetected in Europe because companies inevitably try to keep them secret. However, a study indicates that less than 20% of cartels in the EU between 1985 and 2010 were detected. 12 Given that there is only limited knowledge available about the rate at which cartels are discovered, this study relies on methods frequently used only to make inferences in another discipline so its conclusiveness might be considered questionable by some. Nevertheless, its results provide very useful indications about the levels of undetected cartels in Europe. If one takes 15% as the rate of detection of cartels 13 and assumes that detected cartels are not too dissimilar to undiscovered cartels, the total value drained from the European economy by such collusion could be in the range of more than 3% of 2012 euro-area GDP. 14 Cartels typically increase prices: the empirical evidence in academic literature suggests that the median cartel overcharge lies between 17 and 30%. 15 When cartels are not detected, consumers have to bear the price of this overcharge.

A primary goal of competition policy enforcement is to deter anticompetitive behaviour by companies thereby maintaining a level playing field in product markets to the benefit of the end consumer. The deterrent effect of a cartel prohibition, for example, depends on the impact of the decision by the competition authority on the existing and future cartelists' perceived likelihood of getting caught and on the size of the expected fines. Attempts have been made to get rough estimates of the deterrent effects of inter alia prohibiting and sanctioning cartels. Surveys of competition lawyers and companies in the UK and the Netherlands indicate that the number of cartels deterred per cartel detected varies between 5 and 28. 16 Keeping in mind the caveats and uncertainties in measuring deterrence (the difficultly in knowing how many undetected cartels exist), there is survey evidence suggesting that the impact of deterrence in terms of avoided consumer harm is significant. For example, one study – basing itself on the existing literature and the European Commission's competition enforcement in 2012-2014 – uses lower and upper bounds (boundary) assumptions as to the amount of avoided consumer harm which is deterred resulting from each cartel detected and sanctioned. 17  The lower bound of the total avoided consumer harm is estimated to be between 10 and 30 times the harm caused to the customers of each detected cartel. While not possible to determine precisely, there are thus indications that the untapped potential of making competition enforcement more effective is considerable.

Experience shows that equipping a competition authority with additional tools significantly contributes to untapping the potential of its enforcement activities. For example, in the 20-year period spanning from 1990 until 2010, the Spanish Competition Authority uncovered and fined a total of only around 10 cartels. The introduction in February 2008 of a Leniency Programme, as well as bestowing the Authority with increased powers with which to carry out inspections, completely changed the picture. In the 6 year period between 2010 and 2015 this figure has increased five-fold with a total of 57 cartels being fined a combined total of 1,200 million euros. 18

The potential for more effective competition enforcement is also directly experienced by consumers and companies. According to a recent survey, over two-thirds (68%) of EU citizens have experienced a lack of competition in at least one sector that resulted in problems such as higher prices, less product or supplier choice, or lower quality. 19 EU citizens point to a range of sectors in Europe where they have experienced a problem resulting from a lack of competition, notably, the energy sector (gas, electricity etc-28%), transport services (railways, airlines -23%), and pharmaceutical products (21%). 20 In addition, the Consumer Markets Scoreboard regularly identifies markets in Europe which are not functioning well for consumers, in particular in terms of choice and problems experienced by consumers. 21

In the context of the European Semester, the European Commission has also identified a number of sectors where there is still scope for improving competition. For example, a limited degree of competition in the retail sector resulting in high prices for consumer goods can be observed in Belgium, Denmark and Finland. The business services sector in Austria and Malta is also experiencing low competition. Similarly, a lack of competition in the telecom and broadband sectors results in higher prices in Croatia and Cyprus. 22 The transport sector in Croatia, Czech Republic, Denmark, Finland, Germany, and Slovenia could also benefit from increased competition.

The lack of competition on markets is tackled in two main ways: (1) by removing anti-competitive regulatory barriers and (2) through effective enforcement of competition rules. Accordingly, the European Semester focuses not just on the elimination of regulatory barriers, but also on the establishment of an efficient competition framework in all Member States enabling NCAs to tackle anticompetitive conduct by companies. 23

For example, the achievement of the EU Energy Union strategy is dependent both on removing regulatory barriers and on effective enforcement of the EU competition rules by the Commission and the NCAs. To have secure, affordable, low-carbon energy for all EU consumers and businesses, energy must flow easily between EU countries. The European Commission is currently investigating contractual restrictions which may prevent the supply of gas into other countries. 24 However, in the gas and electricity businesses there are many complex relationships which the Commission cannot solve on its own. At some levels of the market, NCAs are very well placed to investigative anti-competitive restrictions. Some NCAs are focusing on these issues, 25 but more could be done. 26

There is a clearly established link between the role and the ability of NCAs to act and fully enforce the EU competition rules and ensuring competition on markets. According to the OECD, 27 there is solid evidence in support of each of the relationships shown below:

If competition authorities are not in a position to be effective enforcers, undistorted competition in markets is not ensured and the ensuing benefits in terms of higher productivity growth, which in turn generates economic growth, are not fully realised.

Studies have shown that the inability of competition authorities to apply the competition rules can have a negative impact on growth. 28 For example, the US National Industry Recovery Act, which selectively suspended anti-trust laws in the 1930s and authorised companies to establish cartels, brought a 10% reduction in manufacturing output. 29

76% of stakeholders in the public consultation launched by the Commission in November 2015 on how to empower the national competition authorities to be more effective enforcers 30 said that the NCAs could do more to enforce the EU competition rules. The NCAs themselves repeatedly call on the Commission to help them to be more effective enforcers, because they do not always have the means and instruments they need.

The level of enforcement of the EU competition rules is very unevenly spread in the EU. Some NCAs have adopted very few decisions applying the EU competition rules since 2004. For example, between 2004 and 2015, the Irish (2), Maltese (3), Estonian (4), Latvian (5), Luxembourgish (5), Cypriot (6), Bulgarian (7), Czech (10), Finnish (14) and Polish (14) NCAs adopted less than 15 decisions, whereas the French (119), German (113), Italian (112) and Spanish (101) NCAs took more than 100 decisions in the same period.

Although the number of enforcement decisions depends on a number of variable factors, for example authorities may invest much more time in complex cases or cases that involve a significant number of market players, the above data demonstrate that there is clear scope for more effective enforcement. Even those NCAs with a high number of enforcement decisions face issues such as their inability to impose effective fines (see section 2.2.2 below). Companies have little incentive to comply with competition law when they see that infringements are not sanctioned.

The inability of NCAs to realise their full potential to be effective enforcers of the EU competition rules undermines the enforcement of one of the key chapters of the EU Treaty on the functioning of the European Union and the decentralised system set up by Regulation 1/2003. 31 It also impairs one of the main facets of the single market, which is ensuring that competition is not distorted in Europe. A lack of effective enforcement by the NCAs means that the framework conditions for efficiently functioning markets and improved competition conditions in the internal market are not ensured.

For consumers, the lack of capacity of NCAs to un-leash their full potential to enforce the EU competition rules means that they miss out on the benefits of competition enforcement, namely lower prices, better quality, wider choice and product innovation.

Stronger application of the EU competition law principles across Europe would contribute to the achievement of the overall objectives of EU economic policy. This "benefits consumers, workers and entrepreneurs, and promotes innovation and growth, by controlling and restricting unfair market practices resulting from monopolies and dominant market positions, so that every individual has a fair chance of success". 32

2.2Underlying problem drivers

Assessments of the effectiveness of competition enforcement show that this is dependent on key aspects of competition policy design. This includes not only the content and scope of the substantive competition rules themselves, but also the independence of the competition authority when it takes enforcement decisions, the powers the authority has to detect, investigate, remedy and sanction competition infringements and the enforcement capacity of the authority in terms of its resources and budget. 33 In particular, the Buccirossi study which found a positive and significant effect of competition policy on productivity, established that the strength of this effect was found to be particularly marked for specific aspects of competition policy, including the degree of independence of the competition authority with respect to political or economic interests, the scope of the authority's investigative and fining powers and the quality of the human and financial resources a competition authority can rely on when performing its tasks. 34

Regulation 1/2003 focused on giving the NCAs the power to co-enforce the EU competition rules. It did not address the means and instruments of NCAs to apply these rules. This means that although the NCAs apply the same substantive rules, i.e., the EU competition rules, the means and instruments they have to enforce depends on what is available under national law.

In 2013-2014, the Commission carried out an assessment of the functioning of Regulation 1/2003. Based on the results of this analysis, the 2014 Commission Communication on Ten Years on Antitrust Enforcement under Regulation 1/2003 35 found that the new system has considerably increased the enforcement of the EU competition rules, with NCAs now as key pillar of the system. However, it found that there is scope for the NCAs to be more effective enforcers and identified a number of areas for action to boost effective enforcement by the NCAs. It concluded that action should be taken to guarantee that NCAs:

(1)have an effective enforcement toolbox to detect infringements and bring them to an end;

(2)can impose deterrent fines on companies;

(3)have leniency programmes which work effectively across Europe; and

(4)have adequate resources and can act independently when applying the EU competition rules.

The 2014 Communication built on the Report of Five Years of Regulation 1/2003, which found that empowering the NCAs to co-enforce the EU competition rules has positively contributed to stronger enforcement. 36 However, it concluded that there is room for improvement, in particular, to ensure that NCAs have effective enforcement powers and fining tools.

By way of follow-up to the 2014 Communication, extensive data collection was carried out by the European Commission in cooperation with all NCAs on the identified areas of action to have a detailed picture of the status quo. Although there has been a good degree of voluntary action at Member State level in the direction of giving NCAs the means and instruments they need to be effective enforcers, supported by extensive soft action (see section 6.1), there are four underlying problem drivers that undermine the ability of NCAs to be more effective enforcers and the decentralized system put in place by Regulation 1/2003. Annex IV contains a problem tree outlining the problem drivers and their consequences.

2.2.1Problem driver 1: Lack of effective competition tools

It is widely accepted that to be effective enforcers, competition authorities should be equipped with operational, well-designed tools to detect infringements and to bring them to an end. 37 Investigation and decision-making powers and procedures are the main working tools of competition authorities.

In this context, the core investigation powers are the power to inspect business and non-business premises, the power to issue requests for information, the power to gather digital evidence, and the power to conduct interviews. The core decision-making powers are the power to adopt prohibition decisions, the power to issue interim measures, and the power to adopt commitment decisions. Competition authorities should have the power to impose effective penalties for non-compliance with the investigation and decision-making tools and the power to set their priorities in full.

Lack of effective investigation and decision-making tools

The fact-finding done as follow-up to the 2014 Communication has shown that there is a patchwork of powers across Europe, with many NCAs not having the powers they need. 38 The scope of NCAs' investigative and decision-making powers varies considerably, which can significantly impact on their effectiveness.

For example:

With respect to the relevant investigation powers, some NCAs (Bulgaria, Denmark and Italy) lack the fundamental power to inspect the homes of business people for evidence of infringements. 39 This is a key gap in authorities' powers, as it is all too easy for cartelists to hide evidence at home. Indeed, nowadays this tool is becoming increasingly important as the distinction between work and home becomes more blurred with modern ways of working, for example, teleworking. If a NCA is unable to obtain evidence, infringements may not be pursued in full or can even remain unaddressed.

With respect to the relevant decision-making powers, 11 NCAs cannot impose structural remedies to restore competition on markets. Major structural problems on markets may call for structural solutions to ensure that competition works effectively. 41 For example, if vertically integrated energy incumbents foreclose the downstream supply markets by refusing indispensable access to transmission capacity, the abusive conduct stems from the very structure of the companies that can leverage their control of the network to maintain their dominance downstream. In such a case the divestiture of the transmission grid ensures that the abuse could never be repeated and it creates the conditions for undistorted competition downstream. Stakeholders, particularly businesses, highlight that the lack of NCAs' power to impose such remedies is particularly a problem for companies damaged by the anticompetitive behaviour of the infringer. Access to key infrastructure can be essential for companies which have been excluded from a market. If NCAs cannot impose effective remedies, they cannot ensure competition on these markets. The infringer continues to reap the benefits of a past violation to the detriment of consumers.

NCAs also face gaps or limitations if companies do not comply with their decisions, e.g. some NCAs do not have the power to impose deterrent fines in case of non-compliance with a commitment decision. Under a commitment decision, parties make voluntary commitments to address competition concerns and the competition authority by decision makes the commitments legally binding. 42 However, 11 NCAs either cannot fine failure to comply with such a commitment decision at all or the fines are set at a level which is too low to compel compliance. This means that authorities' powers lack teeth, e.g. companies easily enter into commitments which cannot be enforced and therefore no market change ensues.

Finally, 15 NCAs do not have the full power to set their priorities and decide which cases to dedicate their (often scarce) resources. 43 Stakeholders, notably businesses, report that the lack of the power of NCAs to set their priorities in full prevents them from focusing on infringements that cause the most harm to competition.

Most NCAs (22) lack the power to collect fines from companies that are located in other Member States, when these companies do not have a legal presence on the territory of the NCA concerned. For example, one NCA has not been able to collect fines from companies based elsewhere in the EU in eight cases. This gives such companies a safe haven from fines. This is a growing issue of concern given that many companies sell over the internet to potentially numerous countries, but only have a legal presence in e.g. one Member State.

Problems for cross-border cooperation to find evidence

Gaps and limitations in NCAs' powers also present a problem for cooperation within the ECN. One of the main elements of Regulation 1/2003 is that it provides for cooperation mechanisms that allow NCAs to investigate alleged infringements beyond the borders of their Member State. One NCA can ask another NCA to carry out investigative measures on its behalf. This means that NCAs are not prevented from gathering evidence simply because it is located in another jurisdiction. However, this mechanism does not work well if not all NCAs have effective powers to carry out inspections or to request information.

Divergences in NCAs' powers, such as differences in rules on when a case becomes time-barred meaning that one authority may be able to act, but another could not, can also pose a problem for businesses operating cross-border. They want legal certainty as to whether proceedings can be brought against them.

In the public consultation, stakeholders indicated which tools they consider NCAs need (Annex V). There was also a clear demand from lawyers, business and business organisations for improvements in NCAs’ enforcement powers to be counter-balanced by increased procedural guarantees at national level, including ensuring that rights of defence can be effectively exercised (companies should receive a Statement of Objections and have effective rights of access to file) and effective judicial review.

2.2.2Problem driver 2: Lack of powers to impose deterrent fines

The ability of competition authorities to fine companies which breach competition law is a central enforcement tool. The purpose of fines is to punish companies which have infringed competition rules and to deter the same and other companies from engaging in or continuing illegal behaviour. EU law does not regulate fines imposed by NCAs 44 on companies for breaches of the EU competition rules. Each Member State has its own legal framework for imposing fines. However, in 2009, the European Court of Justice ruled that "the effectiveness of the penalties imposed by NCAs and the Commission is a condition for the coherent application of the EU competition rules". 45  

The fact-finding carried out has shown that there are a number of issues that affect the level of enforcement of Articles 101 and 102 TFEU and mean that companies can face very low or no fines at all depending on which authority acts, undermining deterrence and the level-playing field.

These issues are of two main types: (1) fines may not reflect the harm caused to competition, and (2) the nature of the fines imposed (criminal, civil or administrative).

Fines should reflect the harm caused to competition

There are differences in the methodologies for calculating fines that can have a significant impact on the level of fines imposed by NCAs. These differences mainly concern: (1) the maximum fine 46 that can be set (the legal maximum) and (2) the parameters for calculating the fine (see Annex VI).

For example, the legal maximum is calculated as a percentage of a given turnover in most Member States. There are however significant differences between Member States in the way the legal maximum is calculated in terms of the percentages applied, 47 and the turnover to which such percentages are applied. Most NCAs when calculating the legal maximum use the worldwide turnover of the corporate group that has been held liable for the infringement, but some base it solely on the national turnover or the turnover of the direct infringer. The entities for which the turnover is considered (the undertaking or the direct infringer) and whether the geographic scope of such turnover is worldwide or national make a big difference to the maximum level of the fine (depending on the size of the corporate group it can be significantly lower). In one Member State, only the direct turnover of the infringer is used and fines are limited to €16 million. For breaches of Article 102 TFEU, maximum fines of only €400 000 can be imposed. Such low legal maximums are highly unlikely to reflect the harm caused to competition and fines are likely to be under-deterrent, particularly for large multinational groups.

Such differences mean that fines vary by up to 25 times depending on which authority acts. Very low fines may be imposed for the same infringement, meaning that the deterrent effect of fines differs widely across Europe. The fines imposed may not reflect the harm caused to competition by the anti-competitive behaviour.

Another aspect which may lead to a situation in which fines do not reflect the harm to competition, is limitations regarding who can be held liable for paying the fine.

The concept of "undertaking" in EU competition law is established by the case law of the European Court of Justice. It means that different legal entities belonging to one "undertaking" can be held jointly and severally liable for any fines imposed on such "undertaking". 48 This sends a clear signal to the entire corporate group that the absence of good corporate governance and compliance with competition law will not remain unpunished. It also allows the fine to reflect the overall strength of the corporate group and not only that of the subsidiary, making it more meaningful and deterrent.

However, 5 NCAs cannot hold parent companies liable for infringements committed by subsidiaries under their control. Also, several NCAs cannot hold legal successors of an infringer (2 NCAs) and economic successors (8 NCAs) of an infringer liable for fines or there is uncertainty about whether national courts would uphold the application of these principles, despite the long established case law of the European Court of Justice. This means that companies can escape fines simply by merging with other companies or through corporate restructuring.

Moreover, some NCAs cannot effectively fine associations of undertakings, such as trade associations, either because national legislation prevents this possibility or because NCAs cannot impose fines that take into account the turnover of its members. 49 This is a problem because trade associations of e.g. lawyers, dentists etc. regularly participate in competition infringements (e.g. illegal agreements to fix the fees charged by their members) but typically have very little turnover compared to their members. NCAs need to be able to also fine the members of the association involved in the infringement to deter them. The fines imposed by NCAs without this power often have little deterrent effect and do not reflect the harm to competition.

Nature of the fines imposed can result in under-enforcement

The nature of the fines imposed by NCAs for the infringement of the EU competition rules varies across Member States. Fines can be either administrative (imposed by the NCA), civil (imposed by civil courts) or criminal or quasi-criminal (imposed mainly by criminal courts or, in some cases, by the NCA but according to quasi-criminal (misdemeanour) procedures).

In the majority of Members States fines are administrative. Civil fines 50 are imposed in three Member States. Criminal or quasi-criminal fines are imposed in five Member States: Denmark, Estonia, Germany, Ireland and Slovenia (see Annex IX).

The fact finding has shown that for the period 2004-2013 in many of those Member States in which fines are primarily criminal, EU competition law is under-enforced or, even if enforced, sanctions are seldom imposed. Ireland and Estonia 51 for example, have reported only 1 and 3 decisions respectively during this period under both Articles 101 and 102 TFEU, meaning that there is virtually no enforcement of the EU competition rules in these countries. 52 For Article 101 cases, which include the most serious competition infringements, that is hard-core cartels (e.g. when competitors fix prices, divide up markets or allocate customers between them), a fine has never been imposed in Ireland and only once in Denmark during this period.

Moreover, in (quasi) criminal systems, the data illustrate that it is much more difficult to bring cases against infringements which are not hard-core cartels. Finding competition infringements, such as whether a company abused its dominant position, involves more complex economic facts, theories and analysis and is more resource intensive. In (quasi) criminal systems during the period 2004-2013, fines for breach of Article 102 have only been imposed once in Denmark and never in Estonia, Germany and Ireland. This means companies in a dominant position in these countries have much less to deter them from abusing their dominant position and using illegal means to exclude competitors from their market.

Most stakeholders stated in the public consultation that criminal systems, and to a lesser extent also civil systems, are less suited than administrative systems for the effective enforcement of the EU competition rules. 53  

The risk of under-enforcement imposed by systems that primarily apply (quasi) criminal sanctions was recognised, for example in a report of 2015 by the OECD reviewing the competition system in Denmark. This concludes that the imposition of criminal fines on undertakings has significant implications for the effective enforcement of EU competition law. 54 The Irish Law Reform Commission has launched an issues paper for consultation 55 in which it states that civil sanctions provide advantages over criminal prosecution. There are also calls for changing from civil to administrative models in several Member States to achieve more effective competition enforcement. The Swedish government commissioned an inquiry into this issue which was finalised in June 2016. The recommendation of the inquiry is that Sweden should change from its current civil system to an administrative system 56 in which the NCA is given the power to impose sanctions directly. The inquiry mentions as the main reasons justifying this change the gains in terms of efficiency that would be obtained and the advantages in terms of better cooperation with other NCAs that would result from having a more harmonised system. The Swedish NCA also publicly advocates for the power to impose administrative fines directly. 57

2.2.3Problem driver 3: Divergences between leniency programmes in terms of summary applications, core principles, protection of self-incriminating material and interplay with individual sanctions can lead to less effective competition enforcement against cartels

Since cartels are illegal, they are generally highly secretive and evidence of their existence is not easy to find. Cooperation by cartel members with competition authorities is therefore often crucial to uncover and punish these highly detrimental practices. Leniency programmes encourage companies to come clean about cartels in return for having no fines imposed (immunity) or a reduction in fines. They are therefore a key tool for the detection of cartels.

The mere existence of a (successful) leniency programme can destabilise existing cartels. 58 Currently the Commission and all Member States except Malta have leniency programmes in place. 

However, as shown in Annex XII, the number of (summary) 59 leniency applications varies widely across NCAs. Some NCAs are much more successful in attracting such applications. A comparison between the absolute number of leniency applications and the overall level of enforcement activity per NCA shows that whilst some NCAs can rely to a significant extent on immunity applications to feed their enforcement work stream, in other Member States leniency programmes generate none or much less of the overall enforcement activity. These differences in the success of the leniency programmes may be due to problems related to the single leniency programmes or to their interplay, which undermine their (single and collective) effectiveness.

To enable the NCAs and the Commission to uncover and sanction cartels, their individual leniency programmes have to provide companies with sufficient legal certainty about whether they will benefit from immunity, whether their employees are shielded from individual sanctions and whether self-incriminating material will be disclosed outside the context of the investigation.

The efficient interplay between the different leniency programmes within the ECN is just as important for the effective enforcement of the competition rules within the ECN since cartels often extend beyond national borders and companies may have to file leniency applications to authorities in several jurisdictions. If the core principles of the different programmes are too divergent, companies do not have legal certainty about their immunity status under the leniency programme(s) of the authority/ies that will eventually deal with their case.

Divergences and weaknesses exist mainly in the following areas:

Divergences in the treatment of summary leniency applications and in core leniency features

Despite some degree of convergence achieved by the non-binding 60 ECN Model Leniency Programme (MLP) endorsed in 2006, 61 important divergences remain. For example, summary applications are still not available before some NCAs. On core leniency features, divergences continue to exist regarding which companies can benefit from leniency and under which conditions. This leads to different outcomes when it comes to deciding which companies benefit from immunity, a reduction of fines or no reduction at all and in what order their applications are assessed as adding value for the case.

Lack of protection of leniency and settlement material

Companies that choose to cooperate under leniency programmes are required to disclose their participation in a secret cartel and provide self-incriminating leniency material. In case of formal settlements, parties to the investigation are required to acknowledge their participation in, and liability, for the infringement. 62 The level of protection granted for such material varies significantly between Member States. For example, in 20 Member States leniency statements are accessible to public prosecutors and/or the police, who could use it for other purposes than for the enforcement of the EU competition rules. In 12 Member States, civil courts in proceedings other than actions for damages have access to such statements. Such access can expose the companies that choose to cooperate with the competition authorities to liability to other proceedings being brought against them.

Lack of effective interplay between corporate leniency programmes with sanctions on individuals

Many Member States foresee sanctions on individuals for their involvement in anticompetitive behaviour. However, most leniency programmes lack arrangements to protect employees of companies which make leniency applications to NCAs and/or the Commission from individual sanctions. Individuals who may be subject to criminal proceedings may be deterred from helping their employers to collect the evidence required for a successful leniency application.

All the above shortcomings may weaken incentives for potential leniency applicants to cooperate with authorities. This in turn can lead to less effective competition enforcement in the EU, as less secret cartels are uncovered. More undetected cartels activity leads to a welfare loss across the EU.

For more detail on the existing divergences between leniency programmes in terms of summary applications, core principles, protection of self-incriminating material and interplay with individual sanctions as well as the feedback from the public consultation on these issues, see Annex XIII.

2.2.4Problem driver 4: Lack of safeguards NCAs can act independently when enforcing the EU competition rules and have the resources they need to carry out their work.

Independence

It is widely accepted by organisations such as the OECD 63 and UNCTAD 64 as well as academics 65 and stakeholders that the independence of competition authorities is a prerequisite for effective competition enforcement. 66 In the area of data protection, where EU law requires national data protection authorities to act with complete independence, the European Court of Justice has clarified that the independence of a public body normally means a status which ensures that it can act completely freely, without taking any instructions or being put under any pressure. 67 The legitimacy and credibility of NCAs actions vis-à-vis stakeholders is inherently linked to their ability to act impartially, free from external influence both from the companies they supervise and from political bodies. In recent years, the European Parliament has regularly emphasised the importance of having independent NCAs. It called on the Member States to ensure sufficient human and financial resources and independence for all NCAs and on the Commission to monitor their independence. 68

Regulation 1/2003 does not provide for specific requirements to safeguard the independence of NCAs. The degree of independence of NCAs from interference by public and private bodies when enforcing the EU competition rules is solely determined by national law and differs between Member States. However, when co-enforcing the EU competition rules together with the Commission, NCAs should, like the Commission, be able to carry out their enforcement tasks free from external influence. However, not all NCAs benefit from such safeguards. For example, NCAs in only 10 Member States are explicitly prohibited from seeking or taking instructions from any public or private body.

In the absence of enforceable safeguards at EU level for NCAs to enforce the EU competition rules without taking instructions from anyone, a risk of direct or indirect influence from other public or private bodies exists, even where according to national law they are formally deemed to be independent. For example, a genuine risk of influence by other state bodies exists where state-owned companies or activities by state bodies are the subject of an investigation by the NCA or where its enforcement action would interfere with other public interests. The absence of independence, which is essential to ensure objective market supervision, can result in different treatment of anticompetitive practices according to the type of company or sector involved. The fact that NCAs lack safeguards of independence and may be subject to instructions can have a direct impact on legal certainty and predictability for companies regarding the application of the EU competition rules. It also affects the legitimacy of the application of the EU competition rules by NCAs and may compromise the willingness of companies to invest.

Resources

In order to act independently and effectively when enforcing the EU competition rules, NCAs need expert staff and sufficient resources to assess cases impartially and to defend their assessment before the courts. The NCAs have to be able to deal with powerful companies that are assisted by specialised teams of lawyers and economists. This issue was also recognised in the context of the European Semester where several Member States have been encouraged to ensure their NCAs have adequate resources, so that they could better contribute to achieving growth through more effective competition enforcement. 69  

Respondents to the public consultation also highlighted the link between having the necessary resources and being independent/effective. They considered having a safeguard of adequate and stable human and financial resources to be the most important measure to ensure the independence of NCAs. 70 In this context, stakeholders emphasise the need for specialised staff with proper expertise and measures enabling NCAs to attract and retain such people. Having insufficient resources is regarded as contributing to less effective enforcement and lower quality of enforcement decisions as authorities may not have the staff to invest in carrying out robust and detailed legal and economic analyses. For example, some NCAs do not have sufficient staff to conduct simultaneous inspections of all members of a suspected cartel but have to limit the search for evidence of anticompetitive conduct to key targets in the investigation with the risk of missing out on important evidence. One NCA had even to limit its inspection to one member of a cartel, which is not only ineffective but can also raise difficult issues such as claims of discrimination.

More generally, significant differences can be observed in resources among the NCAs in Member States with a similar GDP. NCAs in certain Members States are faced with limited human or financial resources (see Annex XIV for a comparison of the budget and staff of NCAs in Member States with a similar GDP). The following table illustrates the impact that the level of the budget of NCAs from Member States with similar GDP 71 has on their level of enforcement of the EU competition rules during the period from May 2004 until December 2014.

The NCAs in this table come from Member States with a comparable level of GDP. The table indicates that NCAs which have more limited resources (MS1 and MS2) have a significantly lower level of enforcement decisions (14 and 13 decisions respectively) compared to the NCAs with a higher level of resources (MS3 and MS4: 28 and 48 decisions respectively) in the same time period. Similar results are obtained for other groups of NCAs from Member States with a similar GDP (see the additional tables in Annex XVI). All the relevant observations presented below and in the Annex together reliably suggest a strong link between the budget available to NCAs and their level of enforcement of the EU competition rules.

2.3Who is affected by the problem?

NCAs are affected because gaps and limitations in their means and instruments undermine their ability to be effective enforcers. Many NCAs lack the necessary tools to find evidence of infringements, to fine companies which break the law, to act independently when enforcing the EU competition rules or to have the resources they need to perform their tasks. This can prevent them from taking action at all or result in them limiting their enforcement action. It also hinders them from cooperating effectively with their fellow competition enforcers in the European Competition Network. This is to the detriment of the European system of competition enforcement as a whole.

The lack of operational means for many NCAs results in untapped potential for more effective competition enforcement in Europe and thus missed opportunities for removing barriers to market entry, reinforcing the single market and creating more open competitive markets on which companies compete fairly and on their merits. This affects all businesses, irrespective of their size, including SMEs and start-ups. Businesses and consumers may particularly suffer in those countries where NCAs are less well-equipped to be effective enforcers. For example, if NCAs lack resources to pursue infringements of the EU competition rules, these stakeholders cannot reap the benefits of effective competition in the Member States concerned. Companies cannot compete fairly on the merits when there are safe havens for anti-competitive practices, e.g. because the evidence of the practices cannot be collected or because companies are not sanctioned for illegal practices. This can, for instance, prevent start-up companies from entering new markets, discouraging enterprise and efficiency.

Consumers are affected as there is scope for greater competition enforcement against anti-competitive practices which keep prices for goods and services artificially high.

Greater competition also stimulates productivity meaning that there is also untapped potential for innovation and growth in Europe.

Moreover, Regulation 1/2003 provides for a system of parallel competences where in principle any NCA or the European Commission is competent to take up a case to enforce the EU competition rules. The allocation of cases is done on a flexible basis within the ECN and means that enforcement is not hampered, e.g. because one authority lacks resources to investigate a particular case at a given moment in time. This system presupposes that for companies it does not matter which authority acts. However this is not the case when companies can have very different fines imposed on them depending on which authority takes up the case or when cases are time barred before some authorities but not others because of differences in rules on when NCAs can act. Gaps, limitations and divergences in NCAs’ means and instruments can thus result in different outcomes for companies. This can lead to costs and uncertainty for companies operating cross-border. For example, companies may think again about applying for leniency if they are not sure where they will come in the leniency queue, which can mean the difference between full immunity from fines to no reduction at all.

2.4What is the EU dimension of the problem?

Under Regulation 1/2003, the European Union decided to share its competence to enforce the EU Treaty rules on competition with the NCAs. The EU competition rules are now applied through-out Europe in 29 jurisdictions: they are enforced in 28 Member States, complemented by enforcement at the European level by the European Commission.

As explained in more detail in Section 3 below, the system of cooperation set up by Regulation 1/2003 has an obvious EU dimension with its cross-border cooperation mechanisms and enforcement actions and the inter-linkage of leniency programmes. Any gaps and limitations in the NCAs' means and instruments thus not only affect their individual capacity as effective enforcers, but that of the system as a whole.

2.5How would the problem evolve, all things being equal?

Some Member States may provide the NCAs over time with additional means and instruments to be effective enforcers. 72 However, most NCAs will continue to miss certain key tools to detect and sanction infringements or lack sufficient resources, affecting the decentralized system put in place by Regulation 1/2003.

Soft action has been used extensively to prompt voluntary action at national level and after more than a decade, the changes needed to make the decentralised enforcement system of Regulation 1/2003 work better and empower the NCAs to be more effective enforcers, are unlikely to ensue.

Section 5.2 lists the soft action that has been taken to try to support the NCAs having the means and instruments they need to enforce the EU competition rules. These measures serve as a soft framework of reference and show that there is considerable consensus within the ECN on the means and instruments authorities need to be effective enforcers. 73  

Efforts have also been made in the context of the European Semester process since 2011 and the Memoranda of Understanding of Specific Economic Conditionality with the so-called “programme” countries since 2010 to address shortcomings in several NCAs' enforcement and fining powers and their independence and resources. Both mechanisms have had mixed success, particularly in view of the fact that the EU lacks a clear legal basis to ensure the effective functioning of the NCAs. For example, the Council adopted Country Specific Recommendations addressing the degree of independence of NCAs or the level of their resources in certain Member States. However, even recurrent calls by the Council have not been effectively implemented. 74 For example, since 2013 the Austrian government has been repeatedly called on to substantially strengthen the resources of the NCA. Eventually in 2016 the staff of the authority was nominally increased by 10 additional case handlers, but no additional budget was granted to actually employ extra staff. Similarly, the Slovenian government was asked to increase the institutional independence of the NCA and to ensure sufficient budgetary autonomy. However, the legal framework has not been amended to avoid administrative investigations into the decision-making of the NCA which undermine its independence.

Another issue of concern is that achievements made to date are fragile. Even if NCAs have been granted effective means and instruments under national law, there is nothing to prevent a reversal of these changes. For example, one NCA lost the power to grant interim measures even though this tool is actively used by some NCAs to prevent anti-competitive action which could cause long-term damage to markets.

In essence, gaps or limitations in the means and instruments of NCAs to effectively enforce the EU competition rules will continue to result in NCAs refraining from enforcing or taking more limited action because they do not have the tools they need. For example, some NCAs cannot search homes and cars of businessmen for evidence of infringements, giving cartelists an easy loophole to exploit. This will continue to result in less detection of anti-competitive practices.

Some NCAs will still suffer from a severe lack of the resources they need to perform their tasks meaning they cannot be effective enforcers.

Divergences in NCAs' powers will continue to undermine the level playing field. For example, differences with respect to the main elements to be taken into account by all NCAs when calculating the fines means that companies can face very low fines for the same infringement depending on which NCA acts, undermining deterrence.

Companies which are considering reporting cartel behaviour to a number of jurisdictions in return for more lenient treatment may refrain from doing so because they lack the legal certainty they need about whether and to what extent they will benefit from this. If companies do not report secret cartels to competition authorities, detection levels fall.

A continued lack of safeguards for NCAs that they act only in the general interest of the EU without taking instructions from anyone when they enforce the EU competition rules will continue to undermine the legitimacy of competition enforcement.

Finally, the European system of competition enforcement has been designed in a cohesive way so that authorities can rely on each other to do fact-finding measures on each other’s behalf. For example, if evidence of an infringement that authority A is looking into may be found on the territory of authority B, authority A can ask authority B to conduct inspections or issue requests for information to try to find this information. But this system cannot work well where there are still gaps or limitations in authority B’s fact-finding powers, e.g. they cannot effectively gather data from mobile phones or laptops, when experience shows that evidence is often found on such devices. This means that cooperation within the ECN will continue to be less effective than it would be if all NCAs had adequate powers.

The above gaps, limitations and divergences in NCAs' means and instruments mean that the scope for more effective competition enforcement is untapped, to the detriment of the European system of enforcement as a whole.

In sum, in the absence of further action at EU level, the existing national competition frameworks will not allow the NCAs to enforce the EU competition rules more effectively across the EU. It will also mean that issues will persist, such as the broad range in the level of fines depending on which authority acts and the lack of legal certainty for companies considering reporting cartel behavior to different jurisdictions.

3Why should the EU act?

Regulation 1/2003, which enabled the NCAs to enforce the EU competition rules together with the Commission, is already in place for more than ten years. For the full potential of this decentralised system to be realised, NCAs need to be in a position to effectively enforce the EU competition rules.

NCAs are applying rules with a cross-border dimension

The EU should take action to address this, because the NCAs are applying rules which have a cross-border dimension. This means that there is an inter-linkage between enforcement action taken by a NCA in one Member State and the impact on competition in another Member State. The NCAs are obliged to apply the EU competition rules whenever trade between Member States may be affected. This criterion is easily fulfilled in the case of a cartel or an abuse of dominant position that covers two or more Member States. Action by one NCA to tackle predatory pricing engaged in by a dominant company in more than one Member State will benefit businesses present in all the Member States affected. Such anti-competitive behaviour has the aim of eliminating competitors and stopping it through enforcement action is in the interests of all affected competitors.

The effect on trade criterion can also be fulfilled in the case of anticompetitive agreements or abuses of dominance in a single Member State. For example, if a supplier sells to an exporter in the same Member State and that supplier prohibits exports to another Member State, this may have an effect on trade. Without the agreement the exporter would have been free to engage in export sales. Similarly, if companies enter into a price-fixing cartel that covers the whole of a Member State, this is normally capable of affecting trade between Member States, because such cartels typically exclude competitors from other Member States and reinforce the partitioning of the single market. Enforcement action by a NCA against such "national" cartels can thus have a positive impact on businesses from other Member States seeking to enter new markets.

Likewise, action taken by one NCA may have a positive impact on consumers in other Member States. For example, the Spanish competition authority imposed a fine on a cartel of Spanish producers of sherry destined for export under the trademarks of foreign distributors (so-called Buyers Own Brand market). The sherry producers reduced the supply of sherry for sale on this market to enable them to increase prices. These production limits were reinforced by additional anti-competitive measures, such as coordinated price increases. Because the sherry products concerned were intended for the export market, primarily to Germany, the Netherlands and the UK, it led to higher prices for sherry for consumers in these countries.

The decentralised enforcement system works well provided all NCAs have the necessary means and instruments to enforce. However, where a NCA lacks the necessary instruments and means to tackle an infringement of the EU competition rules, this can have direct consequences for businesses and consumers in other Member States. The inability of one NCA in Member State A to tackle a national wide cartel, e.g. because it lacks sufficient resources, is a problem not just for businesses and consumers in that Member State but also those in e.g. Member State B and C. However, Member States B and C are not able to address the lack of insufficient resources of the NCA in Member State A.

Ensuring that cross-border cooperation works effectively

Another issue is that only action at EU level can ensure that the system of cooperation set up by Regulation 1/2003 works sufficiently. One of the main elements of Regulation 1/2003 is that it provides for cooperation mechanisms that allow NCAs to investigate alleged infringements beyond the borders of their Member State. One NCA can ask another NCA to carry out investigative measures on its behalf to gather evidence located in another jurisdiction. As noted above in section 2.5, this mechanism does not work well if not all NCAs have effective powers to carry out inspections or to request information. Again, it is difficult to tackle this issue at national level. For example, if the NCA in Member State A needs the NCA in Member State B to gather evidence from companies located in its territory, but the NCA in Member State B does not have effective powers to gather this evidence, there is little that Member State A can do about this.

Interlinkage between competition authorities' leniency programmes in Europe

Leniency programmes are interlinked because companies regularly file applications to a number of EU jurisdictions and need guarantees of cross-border legal certainty. The experience of the last decade has shown that such cross-border legal certainty cannot be sufficiently achieved by Member States individually. Divergences in leniency programmes still lead to different outcomes for leniency applicants in terms of whether they benefit from immunity from fines or even from fines reductions at all. Companies which are considering reporting cartel behaviour to a number of jurisdictions in return for more lenient treatment lack the certainty they need about whether and to what extent they will benefit from this. EU action is needed to ensure that a leniency system is available and applied in a similar way in all Member States.

National laws can prevent NCAs from being more effective enforcers

A further reason underlining why EU action is needed is that in some Member States, national law prevents NCAs from imposing effective fines on companies for infringements of the EU competition rules. Infringing companies present in Member States where NCAs lack effective fining powers are thus sheltered from sanctions and have little incentive to act in compliance with EU competition rules. This reinforces market distortions through-out Europe and may undermine the single market. Moreover, the differences between the NCAs in the core principles for imposing fines mean that companies may face very different levels of fines depending on which authority acts. Only action at EU level can ensure that there are common core principles for imposing fines, thus, providing a more level playing field for businesses.

Measures taken to undermine the independence of the NCAs or to limit their resources necessarily emanate from the Member States themselves. For example, restrictions on independence can be motivated by the desire to exercise greater control over decision-making by the authority. Therefore it is not realistic that safeguards to prevent this at national level will always suffice. A government's ability to apply influence or pressure on a NCA may result in political considerations prevailing over sound competition enforcement based on legal and economic arguments, to the detriment of companies operating in the single market.

In sum, existing national competition frameworks will thus not by themselves allow the NCAs to enforce the EU competition rules more effectively across the EU. Moreover, the Commission cannot enforce any EU requirements regarding the investigation and sanctioning tools, resources and institutional structure of NCAs when enforcing the EU competition rules as long as such requirements do not exist. Accordingly, only an initiative at the EU level can empower the NCAs to be more effective enforcers by ensuring that they have more effective means and instruments to apply the EU competition rules.

In the public consultation, 64% of stakeholders said they would like action to be taken at both EU and national level, with 19% preferring exclusive EU action and 8% opting for solely national measures.

4What should be achieved?

Based on the problem and the four underlying problem drivers identified in section 2, the primary objective of this policy initiative is to make sure that the full potential of the decentralised system of enforcement put in place by Regulation 1/2003 is realised, by empowering the NCAs to be more effective enforcers. This would boost effective enforcement of the EU competition rules by the NCAs and the functioning of markets in Europe (general objective).

This requires the achievement of the following specific objectives:

1.    ensuring all NCAs have effective investigation and decision-making tools;

2.    ensuring that all NCAs are able to impose effective deterrent fines;

3.    ensuring that all NCAs have a well-designed leniency programme in place which facilitates applying for leniency in multiple jurisdictions; and

4.    ensuring that NCAs have sufficient resources and can enforce the EU competition rules independently.

5What are the various options to achieve the objectives?

The following policy options were considered to see if they achieve the above objectives and remedy the problem by tackling the underlying problem drivers identified in section 2:

Option 1: no EU action at all (the baseline scenario)

Option 2:    further soft action.

Option 3:    EU legislative action to ensure NCAs have minimum means and instruments to effectively enforce the EU competition rules. This would be complemented by soft action to put meat on the bones of these provisions, where appropriate. For certain limited and targeted areas, more detailed and uniform provisions may be provided for to the extent that minimum rules would not suffice.

Option 4:    EU legislative action to provide NCAs with detailed and uniform means and instruments to be effective enforcers.

These options were developed taking into account the views of stakeholders. In the public consultation, stakeholders considered that the following measures should be taken:

Ensuring that NCAs have effective enforcement tools to investigate and take decisions (87%).

Having in place effective leniency programmes which encourage companies to come clean across Europe (82%).

Ensuring that NCAs have sufficient resources to perform their tasks (85%).

Ensuring that effective fines can be imposed (80%).

Giving NCAs safeguards that they enforce the EU competition rules in the general interests of the EU and do not take instructions from anyone (76%).

In the policy options identified, there was no need to provide for a differentiated scope, e.g. to exempt or to apply a lighter regime for SMEs, because empowering the NCAs to be more effective enforcers would benefit all consumers and companies, both large and small, by boosting effective competition enforcement and creating a more level playing field. It would not impact SMEs to a disproportionate extent compared to larger companies. In principle, all companies are subject to the EU competition rules provided there may be an effect on trade between Member States. Agreements between SMEs are not necessarily capable of affecting trade between Member States, because activities of SMEs are normally local or at most regional in nature. 75  Agreements of minor importance that do not have an appreciable effect on inter-state trade and on competition are not caught by Article 101 TFEU, unless these agreements are hard-core infringements such as price-fixing cartels. 76  For Article 102 TFEU to apply a company must have a dominant position, that is to say, substantial market power, which will not be the case for many SMEs. In sum, many agreements/behavior of SMEs fall outside the scope of the EU competition rules.

The policy options for each of the specific objectives identified are as follows (see Annex X for a schematic overview).

5.1Option 1: no EU action (baseline scenario)

This is the baseline scenario, entailing no action at all at EU level to make sure the full potential of the decentralised system put in place by Regulation 1/2003 is realised by empowering the NCAs to be more effective enforcers. This essentially means keeping the status quo, taking into account the extent to which it is likely to change in the absence of EU action.

Some Member States may provide NCAs over time with some of the powers or resources they need or remove divergences in the national legal framework with regard to leniency or fines. However, this is unlikely to take place on the scale needed to make NCAs effective enforcers (see section 2.5).

5.2Option 2: further soft action

Soft action could take the form of ECN Recommendations or ECN Resolutions. "ECN Recommendations" are documents setting out the position of all ECN members and can be used as advocacy tools to influence policymakers. "ECN Resolutions" by the heads of the NCAs set their joint position to the extent possible within their national legal frameworks.

5.2.1Specific objective 1: ensuring all NCAs have an effective competition toolbox

In 2012/2013, the ECN endorsed a set of seven detailed Recommendations on key enforcement powers that NCAs need. 77 Possible areas of soft action not covered by these Recommendations could include new ECN Recommendations on issues such as the use of behavioural remedies to ensure a return to competitive conditions on markets and formal settlement procedures.

5.2.2Specific objective 2: ensuring deterrent fines can be imposed

The 2008 Key Principles for the Determination of Fines issued by the European Competition Authorities already provide general guidance on how fines should be calculated. 78 Soft action could be taken to convince Member States to apply the EU concepts of undertaking, parental liability and succession in line with the case law of the European Court of Justice, to ensure that associations of undertakings can be effectively fined and, in the case of Member States that have a primarily criminal enforcement system, to allow the imposition of administrative fines or the imposition of fines by civil courts.

5.2.3Specific objective 3: making leniency programmes and their interplay more attractive to encourage companies to cooperate with the authorities in their fight against cartels

The ECN MLP already set out the core principles of substance and procedure for effective leniency programmes, as well as on the protection of leniency materials. An extension of the ECN MLP or a separate ECN Recommendation could be envisaged to encourage the introduction of arrangements to protect employees of companies which apply for leniency from individual sanctions at national level.

5.2.4Specific objective 4: ensuring all NCA have safeguards they can act independently when enforcing the EU competition rules and have the resources they need to carry out their work.

The ECN Resolution of 2010 on the continued need for effective institutions calls for the competition authorities to be “adequately equipped for their tasks and be able to act under suitable conditions for the execution of their task, in an impartial and independent manner”. 79 Further soft action could provide for more detailed provisions on the independence and resources of NCAs.

5.3Option 3: EU legislative action to provide NCAs with minimum means and instruments to be effective enforcers, complemented by both soft action and detailed rules where appropriate

5.3.1Specific objective 1: ensuring all NCAs have an effective competition toolbox

EU action could be taken to provide NCAs with:

A minimum core set of operational investigative tools (that is, effective powers to inspect business and non-business premises, to issue requests for information and to gather digital evidence) For example, giving NCAs effective powers to gather digital evidence would ensure this option is future proof, as it would enable NCAs to access the same data as the company being inspected irrespective of how it is stored, including data stored on clouds, laptops and mobile phones. It would also allow NCAs to continue searches of large amounts of (typically digital) data at their premises to minimise the disruption caused by an inspection.

Decision-making tools (the power to adopt prohibition decisions - including the power to impose structural and behavioural remedies -, to issue interim measures and to adopt commitment decisions).

Backing up these tools with effective sanctions for non-compliance with them, e.g. the payment of a fine for failure to comply with an inspection and the power of NCAs to set their priorities in full.

Tools could also be put in place to address limitation periods and the inability of NCAs to enforce fining decisions cross-border.

The increase in the powers of the NCAs would be counter-balanced by ensuring that key procedural guarantees are in place in line with the EU Charter on Fundamental Rights, such as the obligation of NCAs to notify companies of the objections against them and by providing for effective judicial review of enforcement decisions.

5.3.2Specific objective 2: ensuring deterrent fines can be imposed

Action could be taken at EU level to provide for minimum rules to ensure that:

(a) Fines are based on the key elements widely recognised as essential for calculating a fine:

The gravity and duration of the infringement, and

The potential application of aggravating and mitigating circumstances.

This would ensure that fines are related to the infringement and to the harm caused to competition. It would be complemented by soft action on non-core aspects such as which aggravating and mitigating factors could be taken into account and how to assess the gravity of the infringement.

(b) The legal maximum is set as a percentage of the total worldwide turnover of the undertaking. This would ensure that it is set at a level which ensures deterrence.

(c) The concept of undertaking, parental liability and succession are applied in line with the case law of the European Court of Justice as explained in section 2.2.2 and that associations can be effectively fined.

(d) There exists the possibility either to impose administrative fines or to apply to a civil court for the imposition of fines. This would mean that NCAs currently operating in a primarily criminal system would be given the option of deciding, depending on the facts and circumstances of the case, whether to follow an administrative track/seize a civil court or to follow the existing criminal route.

(e) All NCAs have the power to defend their cases in court (most already can do so).

5.3.3Specific objective 3: making leniency programmes and their interplay more attractive to encourage companies to cooperate with the authorities in their fight against cartels

EU action could be taken to ensure:

The core principles of the ECN MLP are translated into law in light of experience with their application, thereby introducing binding minimum rules for leniency programmes. This would reduce the current divergences between national programmes and ensure, for example, that summary applications are available in all Member States and are applied in the same way. In particular, NCAs would have in place leniency programmes that enable them to grant immunity from fines and reduction of fines to undertakings and companies would have to satisfy core common conditions in order to qualify for leniency. Further, it would be ensured that applicants that have applied for leniency to the European Commission can file summary applications in relation to the same cartel with the NCAs and that NCAs accept summary applications with the same scope as the leniency application filed with the Commission.

Minimum rules are in place to protect employees of leniency applicants with the Commission or other NCAs from sanctions.

There are binding uniform rules for the full protection of leniency and settlement material against disclosure outside the context of civil damages actions (the latter already being addressed by the Damages Directive 80 ). EU legislative action to this end would expand the protection granted by the Damages Directive to other procedures.

5.3.4Specific objective 4: ensuring all NCA have safeguards they can act independently when enforcing the EU competition rules and have the resources they need to carry out their work.

To ensure that NCAs are protected from external influence and they have sufficient resources when enforcing the EU competition rules, minimum rules would be introduced to ensure the independence of NCAs when they enforce the EU competition rules. This would cover the following requirements:

NCAs perform their tasks and exercise their powers independently and are not subject to any instructions from any other public or private body when enforcing the EU competition rules. In particular, it would be ensured that NCAs can take decisions independently from any political and business influence and that the staff and the members of a NCAs' decision making body refrain from actions and occupations that are incompatible with the performance of their duties during their term of office and for a reasonable period thereafter.

NCAs’ board/management cannot be dismissed for reasons related to the proper performance of their powers in the application of Articles 101 and 102 TFEU.

NCAs have adequate human and financial resources to perform their tasks. 81 This would simply provide that NCAs should have sufficient financial, human and technical resources to perform their tasks and would include a list of these tasks (e.g. conducting investigations, taking decisions. and cooperating with other authorities in the ECN).

5.4Option 4: EU legislative action to provide NCAs with detailed and uniform means and instruments to be effective enforcers

5.4.1Specific objective 1: ensuring all NCAs have an effective competition toolbox

This would build on option 3 by providing NCAs with uniform (as opposed to minimum) investigation and decision-making powers, backed up by uniform sanctions for non-compliance. For example, when it comes to the power to adopt commitment decisions, it would regulate detailed issues such as having a mandatory market test for NCAs to get input from other market players on the suitability of commitments proposed by the parties under investigation to address competition concerns.

This option would also provide for a more complete competition toolbox, for instance, by including the power of competition authorities to conduct sector inquiries. This is a useful tool to allow competition authorities to assess whether there are indications to suggest that competition on market sectors of the economy is being restricted or distorted in an economic sector of the common market. Sector enquiries often provide a useful basis for NCAs to carry out enforcement. A sector inquiry can in addition provide empirical evidence that may be useful in reviewing the regulatory framework governing a sector. It would also provide for detailed procedural guarantees, such as detailed and uniform rules on access to an authority's case file and rules on the ability of complainants and third parties to intervene in proceedings.

5.4.2Specific objective 2: ensuring deterrent fines can be imposed 

Option 4 would go further than option 3 by introducing a uniform fining model through EU legislation so that only administrative fines on undertakings can be imposed. It would also provide for a uniform and detailed methodology for setting fines, determining all the parameters that are to be taken into account and prescribing how fines should be calculated and who can be fined.

5.4.3Specific objective 3: making leniency programmes and their interplay more attractive to encourage companies to cooperate with the authorities in their fight against cartels

On leniency, option 4 would involve introducing fully harmonized leniency programmes, with maximum requirements beyond the principles of the ECN MLP, to ensure that leniency applicants can file a single application with one authority that issues an immunity decision which is binding in all Member States and before the Commission (putting in place a so-called one-stop shop system). Similar to option 3, it would also include uniform and detailed rules to protect all leniency and settlement materials outside the context of civil damages actions and to protect employees of leniency applicants to either the Commission or NCAs from individual sanctions at national level.

5.4.4Specific objective 4: ensuring all NCA have safeguards they can act independently when enforcing the EU competition rules and have the resources they need to carry out their work.

In addition to the safeguards foreseen under option 3, this option would involve the introduction of uniform and detailed rules to also ensure the institutional and financial autonomy of NCAs. This would include the following requirements: (1) NCAs would be legally distinct from any other public or private body (structural independence); (2) NCAs would have full authority over the recruitment and management of staff; (3) NCAs would have a separate annual budget with full budgetary autonomy; (4) appointment of NCAs' board/management through a transparent procedure on the basis of merit.

5.5Discarded option

A further option would be for the Commission to take up cases where the NCAs are not able to act.

While it is not excluded that the Commission could do so in a very limited number of cases, intervention by the Commission on a more systemic basis is not feasible as the Commission also has limited resources to enforce the EU competition rules. Such intervention by the Commission would also directly contradict the decentralisation logic of Regulation 1/2003. By the end of the 1990s it became evident that the Commission could not meet the challenges of enforcing the EU competition rules and ensuring integrated markets on its own. The context had changed dramatically, with an EU that had increased from 6 to 15 Member States, and the prospects of more countries joining the EU in the medium term. In this situation, the Commission could not bear the responsibility for enforcing alone the EU competition rules throughout the EU. 82 Regulation 1/2003 therefore set up a system where the EU competition rules are meant to be applied effectively by a multiplicity of enforcers throughout the EU. To that end, Article 35 of Regulation 1/2003 stipulates that Member States must have NCAs in place that are empowered to apply the EU competition rules effectively.

This option was therefore discarded.

6What are the impacts of the different policy options and who will be affected?

An assessment is made of the available options as described in section 5 in relation to the general and specific objectives identified in section 4. Given that all the specific objectives share the same general objective, that is, to realise the full potential of the decentralised enforcement system put in place by Regulation 1/2003, boosting effective competition enforcement by the NCAs and the functioning of markets in Europe, they are assessed together.

The assessment of the impact of these options is to a large extent qualitative as a quantification of the effects of the proposed policy options is only partially feasible. For example, it is easy to determine that the costs and administrative burden of businesses of adapting to different procedural rules throughout the EU will be reduced, but they are very hard to quantify. This is even more true for effects on macro-economic variables like boosting economic growth and innovation and preventing harm to competition and consumers. An assessment is not made of environmental impacts, as no significant environmental impacts are expected.

6.1Option 1: no EU action (baseline scenario)

Taking no action at EU level would mean that the problem and all four problem drivers identified in section 2 would largely persist (see section 2.5 on how the problem would evolve). Most NCAs will continue to miss out on effective tools to detect and sanction infringements. Some NCAs will still lack sufficient resources and safeguards of independence. Potential applicants for leniency will continue to be dis-incentivised to apply to authorities across Europe.

6.2Option 2: further soft action

Further soft action would achieve the specific objectives as follows:

Specific objective 1: ensuring all NCAs have an effective competition toolbox

Taking further soft action on targeted specific areas which have not been addressed by the ECN Recommendations on key enforcement powers, e.g. having an ECN Recommendation on formal settlement procedures, 83 may be a useful source of best practice on these issues for national legislators. However, it will not solve the problem that soft action has not been effective so far and many NCAs lack the investigation and decision making tools they need to be effective enforcers.

Specific objective 2: ensuring deterrent fines can be imposed

Soft action could be taken to convince Member States to apply the EU concepts of undertaking, parental liability and succession in line with the case law of the European Court of Justice and to enable those Member States which operate in a primarily criminal enforcement system the imposition of administrative fines or the imposition of fines by civil courts. However, these issues are firmly established in national legal frameworks. For example, efforts in the context of the Memorandum of Understanding with Ireland to enable the Irish NCA to also impose administrative or civil fines failed because of a lack of an EU legal basis, as Regulation 1/2003 left it to national law to provide for sanctions for breach of the EU competition laws. 84  

Specific objective 3: making leniency programmes and their interplay more attractive to encourage companies to cooperate with the authorities in their fight against cartels

Soft measures could be taken to encourage the introduction of arrangements to protect employees of companies which apply for leniency from individual sanctions at national level. However, this is only likely to be effective where NCAs themselves are also responsible for imposing administrative sanctions on individuals. Ensuring that the employees of companies which apply for leniency are protected from criminal sanctions on individuals cannot be addressed exclusively through soft action, as the latter are firmly established in national legislation.

Further soft action to reduce the remaining divergences between national leniency programmes on substance and procedure and thus to improve their interplay in cross-border cartel cases appears unrealistic. The MLP has never been fully implemented throughout the EU: for example, it is not possible to get a reduction in fines in some Member States (rather only immunity from fines) and summary applications are not available in all Member States. These differences discourage companies from applying for leniency to a number of jurisdictions or from cooperating with the competition authorities at all. Moreover, the Court of Justice has recently ruled 85 that instruments adopted in the context of the ECN are not binding on national competition authorities and that thus Member States are not required to incorporate provisions of the ECN Model Leniency Programme in their leniency systems.

Specific objective 4: ensuring all NCA have safeguards they can act independently when enforcing the EU competition rules and have the resources they need to carry out their work.

Soft action could provide for more detailed safeguards on the independence and resources of NCAs. However, the 2010 ECN Resolution on the continued need for effective institutions which called for NCAs to be adequately equipped and to be able to act independently and impartially has not triggered any improvement. More detailed soft action is unlikely to achieve more change than the Resolution. Moreover, experience in the context of the European Semester and the Memoranda of Understanding with programme countries shows that soft action to ensure that NCAs can act independently and have the necessary resources when enforcing the EU competition rules is generally unsuccessful.

Impacts of option 2

Self-regulation on its own is unlikely to address the gaps and limitations that NCAs face and make the decentralised system put in place by Regulation 1/2003 work effectively. Soft action is not binding on Member States. 86  and has had minimal success to date. This option would therefore have very limited impacts.

6.3Option 3: EU legislative action to provide NCAs with minimum means and instruments to be effective enforcers, complemented by soft action/detailed rules where appropriate

This option is assessed on the basis that giving NCAs largely minimum rules (complemented by soft action and detailed rules as appropriate) to address the problem and the four problem drivers identified would boost effective enforcement of the EU competition rules (for further details see section 5).

The four specific objectives would be achieved as follows

Specific objective 1: ensuring all NCAs have an effective competition toolbox

Taking action at EU level to provide NCAs with minimum rules would ensure that NCAs have the core tools they need to detect and investigate infringements and to take enforcement decisions that stop anti-competitive practices. This would allow the NCAs to effectively enforce the EU competition rules in the digital context, for example, by enabling them to effectively collect data stored digitally, such as on mobile devices, to prove infringements.

These tools would be backed up with effective sanctions for non-compliance e.g. the payment of a fine for failure to comply with an inspection decision. This ensures that these powers are taken seriously by companies under investigation and cannot be easily ignored.

A recurrent call from some stakeholders was also for increased procedural guarantees when NCAs enforce the EU competition rules. It could be ensured that procedural guarantees are in place in line with the EU Charter of Fundamental Rights, such as ensuring that companies under investigation have the right to a formal document setting out the objections of the NCA against them so that they are provided with the information they need to defend themselves and can comment on the allegations made against them.

Specific objective 2: ensuring deterrent fines can be imposed

By ensuring that all NCAs have minimum powers so that administrative fines can be imposed or the imposition of fines by civil courts can be sought, all NCAs which currently operate within a primarily criminal system could choose which route to take (either the criminal route or to opt for the administrative route/civil court route) depending on the facts and circumstances of the case. This option would also make criminal competition enforcement systems more effective by giving NCAs the power to bring and/or defend their cases before criminal courts. NCAs are best placed to explain their decisions and this avoids the duplication of costs and effort inherent in another body defending the case.

This option would also mean that fines better reflect the harm caused to competition by: (1) setting the legal maximum of fines at a level which ensures deterrence; (2) providing for core principles for the methodology for the calculation of fines (such as the duration and gravity of an infringement), complemented by soft measures; and (3) ensuring that the concept of undertaking, parental liability and succession are applied in line with case law of the European Court of Justice.

Specific objective 3: making leniency programmes and their interplay more attractive to encourage companies to cooperate with the authorities in their fight against cartels

Translating the core principles of the ECN MLP into law thereby introduce binding minimum rules for leniency programmes, would mean that current divergences between national programmes would be reduced and for example, summary applications would be available in all Member States and applied in the same way. Uniform rules would be put in place to ensure the full protection of leniency and settlement material outside the context of civil damages actions. This would preserve incentives for companies to provide information to competition authorities. Moreover, in order to ensure that companies are incentivised to apply for leniency, mechanisms could be put in place to protect employees of leniency applicants with the Commission or other NCAs from sanctions.

Specific objective 4: ensuring all NCA have safeguards they can act independently when enforcing the EU competition rules and have the resources they need to carry out their work.

By introducing minimum rules on the independence of NCAs, they would be protected from external influence when enforcing the EU competition rules, for example, by explicitly excluding instructions from any other body or prohibiting the dismissal of the board or management for reasons related to their individual decision-making. In addition, the introduction of an explicit requirement to make available adequate and stable human and financial resources would ensure that they will always have the necessary administrative capacity to effectively enforce the EU competition rules.

Impacts of option 3

It is not meaningful to set numerical figures on the extent to which effective enforcement of the EU competition rules would increase as a result of giving NCAs minimum powers to be effective enforcers. The number of cases brought depends on a number of variable factors such as the availability of information to the authority to detect infringements ex officio or as a result of a leniency application. Other factors also play an important role, for example, authorities will invest much more time and resources in complex cases which can set an important precedent for the market and have a much greater multiplier effect than more routine, less-resource-intensive cases. The scope of cases can also vary significantly, e.g. cases involving a large number of companies take more time to process.

Stakeholders who favour taking legislative action at EU level think that it would have a positive or very positive impact on: (i) the effective enforcement of the EU competition rules (92%) and (ii) legal certainty for businesses (85%). They consider that costs for businesses would decrease (52%) and cooperation in the ECN would be enhanced (83%). Taking action at EU level would have a positive impact on the legitimacy of decisions taken by NCAs (83%) and on the investment climate and economic growth (79%). 87  

6.4Option 4: EU legislative action to provide NCAs with detailed and uniform means and instruments

This option would provide a more complete harmonisation so that NCAs would have detailed and uniform means and instruments. It would fulfil the specific objectives as follows:

Specific objective 1: ensuring all NCAs have an effective competition toolbox

This would mean giving NCAs a uniform and complete set of investigation and decision-making powers, backed up by uniform sanctions for non-compliance. It would also provide for uniform and detailed procedural guarantees. Accordingly all NCAs would have an identical comprehensive toolbox with a high level of protection of procedural rights. Most feedback from stakeholders is that giving NCAs minimum effective investigative and decision-making tools would be sufficient to meet the objective of ensuring they are more effective enforcers. For example, obliging Member States to have a formal market test to assess the appropriateness of commitments proposed by parties under investigation would impinge on NCA's flexibility to get the same input by potentially swifter means. Similarly, designing detailed rules on access to the NCAs' case file would be very difficult as there are very different systems in place at national level. This is also the case when it comes to rules on the participation of complainants and third parties in competition proceedings where there are a large variety of systems in place ranging from the very informal, to systems with detailed procedural steps and rights and obligations for these actors. It is not apparent that having uniform and detailed rules on access to file or regulating the role played by complainants and third parties would lead to more effective enforcement. The fact-finding and the public consultation have not given rise to significant concerns in this respect.

Indeed, NCAs' tools should not be addressed in cases where there is not a clear need for action. For example, although 89% of stakeholders replied in the public consultation that sector inquiries are a tool that NCAs need to effectively enforce (and some NCAs face issues which make conducting such inquiries more difficult 122 ), most NCAs currently do not experience serious issues with the functioning of this instrument. EU legislative intervention on this aspect therefore does not seem merited.

Having uniform and detailed rules on NCAs' toolbox would prevent Member States from providing for more far-reaching powers or stronger procedural guarantees. It may also create rigidities by not leaving enough flexibility to Member States to adapt and change the toolbox as needed.

Specific objective 2: ensuring deterrent fines can be imposed

Introducing a uniform fining model through EU legislation so that only administrative fines on undertakings can be imposed (and no longer civil or criminal fines) would create a more level field for companies in terms of the sanctions to which they are exposed. However, this option would constitute undue interference in the design of Member State competition enforcement systems as this is a fundamental aspect of how enforcement systems are set up. The risk of exposure to criminal enforcement may also prompt companies to comply with the competition rules, so removing this possibility would not be in the interests of overall enforcement.

Having a uniform methodology for setting fines which fixes all the parameters (e.g. value of sales, gravity, duration) that are to be taken into account when imposing fines and prescribes how they should calculate such fines (for example, by determining the actual value which should be attributed to each of the parameters) would mean that the companies would have much greater predictability about the level of fine they would face, irrespective of which NCA acts. This would reinforce the single market. However, it would leave NCAs with limited flexibility to set fines in light of the facts of the specific case. Providing for uniform and detailed rules would also remove the flexibility needed to adapt the rules to new developments.

Specific objective 3: making leniency programmes and their interplay more attractive to encourage companies to cooperate with the authorities in the fight against cartels

Putting in place a one-stop-shop for leniency applications would go a long way towards ensuring a consistent application of leniency criteria, especially as the Commission would be best placed for dealing with all applications in a centralised manner. It would also be the least resource-intensive and most transparent way for applicants to file for leniency. However, depending on the set-up of the one-stop shop, this option would have major drawbacks. First, it would presuppose a full harmonisation of the leniency conditions, which would amount to a far-reaching interference in Member States' systems. For example, such full harmonisation would prevent Member States from devising a more generous leniency programme as regards certain conditions. More lenient conditions may be warranted in some countries where leniency programmes are still novel and a firm culture of cooperating with the NCA has not yet been established. A fully centralised system would also run counter to the current successful system of decentralised enforcement of the EU competition rules, parallel competences and flexible case allocation set out in Regulation 1/2003. Furthermore, a fully centralised system would place a very significant procedural burden on the Commission, which could have repercussions on the Commission's ability to deal with the substance of its own cartel cases and thus lead to under-enforcement at European level. The reduction in the burden for companies, which is the main advantage of all one-stop-shop solutions, does not appear to outweigh these disadvantages.

Specific objective 4: ensuring all NCA have safeguards they can act independently when enforcing the EU competition rules and have the resources they need to carry out their work.

While most respondents to the public consultation support the introduction of the independence safeguards foreseen under option 3, several stakeholders also plead for additional independence safeguards already in place for sectoral regulators. 123  They favoured in particular the introduction of budgetary autonomy and transparent appointment procedures for NCAs' management/board. The downside of this approach is that it would interfere more in the right of Member States to design their enforcement systems in light of their national traditions and specificities. It is not clear that these disadvantages would be outweighed by the benefits these uniform and detailed safeguards on institutional and financial autonomy could bring.

Impacts of option 4

Option 4 is likely to result in all of the impacts set out for option 3, but for some aspects to a somewhat greater extent. For example, by providing for more uniform rules, companies would be treated more consistently through-out Europe and the single market would be more complete. However, it is not evident that putting in place such more detailed and uniform rules would necessarily result in significantly more effective enforcement of the EU competition rules than by giving NCAs a minimum core set of means and instruments. Moreover, it would involve greater intrusion in Member States' legal traditions and specificities, preventing Member States from providing for further-reaching procedural guarantees.

7How do the options compare?

7.1Comparison of the options

The options are compared against the baseline scenario based on their contribution to the general and specific objectives set out in section 4, as well as their main impacts as analysed in section 6.

7.1.1Option 1: no EU action (baseline scenario)

The baseline scenario would not entail additional costs or obligations, but it is highly unlikely to achieve the policy objectives. Some Member States still may make some changes, even in the absence of EU law, however this is unlikely to take place on the scale needed. NCAs would continue to apply the same substantive EU rules on the basis of incomplete national means and instruments.

The baseline scenario would not be in line with stakeholders’ expectations given that 80% of respondents to the public consultation indicated that action should be taken to empower the NCAs to be more effective enforcers. 124  

7.1.2Option 2: further soft action 

Soft action would be most respectful of rules at national level and would carry negligible costs. However, soft action has all the drawbacks of option 1 and does not provide a sound legal basis to ensure that all NCAs have the necessary means and instruments to be effective enforcers.

Soft action has already been used extensively to address the problem and the four problem drivers. It has prompted a certain level of change and has been useful to illustrate to other stakeholders, including policy-makers, the consensus that exists among competition authorities in the EU on the means and instruments NCAs need. It may also lead to some further changes in certain Member States, although this will be limited given that the soft measures have been in place for a number of years, without achieving the aim of fully realising the potential of the decentralised system put in place by Regulation 1/2003 by making NCAs more effective enforcers of the EU competition rules (see section 6.1).

Soft law does not allow for converging EU law concepts to be clarified via the case law of the European Court of Justice, as national courts will continue to give different interpretations of issues such as the notion of undertaking. It also does not enable the Commission to take enforcement action (in the form of an infringement action) if soft measures are not respected.

Exclusive soft action would not be in line with the expectations of the broad majority of respondents to the public consultation, who prefer EU legislative action (79%) to be taken either exclusively or in combination with soft action (28% and 51% of respondents respectively). 125  

7.1.3Option 3: EU legislative action to provide NCAs with minimum means and instruments to be effective enforcers, complemented by soft action/detailed rules where appropriate 

Putting in place a minimum core set of means and instruments, if calibrated correctly, would allow NCAs to more effectively enforce the EU competition rules.

This calibrated approach would allow distinguishing between areas where only high-level principles need to be harmonised and those where, in view of the nature of the perceived gaps, more detailed rules are required to reap added value in terms of competition enforcement. This fine-tuning would limit interference in the Member States' legal systems to the extent that this is strictly necessary to boost effective enforcement, in line with the principles of proportionality and subsidiarity as laid down in Article 5 TEU.

For example, regarding the calculation of fines, establishing a minimum set of aspects that should be taken into account when determining a fine, such as the duration and gravity of an infringement, would go a long way towards ensuring deterrent fines across Europe, even if the more detailed methodology for their application were left for the NCAs. These measures could be complemented by soft measures on non-core aspects.

Another example is the nature of the fines imposed. In order to address the issue that there is little or no enforcement of the EU competition rules in (quasi) criminal systems for competition enforcement, particularly when it comes to fining abuses of dominance, it is necessary to grant NCAs the option of choosing another route than the (quasi) criminal route on a case-by-case basis. Attempts to address this through soft action have not succeeded and after 12 years it is highly unlikely Member States will act, meaning that safe havens from fines will continue to persist. This would directly contradict the decentralised system put in place by Regulation 1/2003 whereby the EU competition rules should be effectively enforced by NCAs in all EU Member States. Conversely, it does not seem necessary or proportionate to prescribe the alternative route to be chosen (administrative or civil courts) or to impose the abolition of the (quasi) criminal route altogether, which might still be appropriate and more deterrent for hard-core cartels in certain Member States. This approach also ensures that the choice of those Member States which have opted for a judicial model of competition enforcement is fully respected.

In some limited cases it may be necessary, however, to provide detailed and uniform safeguards to ensure NCAs are effective enforcers, e.g. to provide for full protection against the disclosure of leniency statements and settlement submissions to preserve incentives for companies to provide information to NCAs.

Such a calibrated approach would not be a radical departure from, but a logical evolution of, general EU law requirement that Member States must provide for effective procedures and sanctions for the enforcement of EU rules. According to the Court of Justice of the Europe Union, national law must ensure that EU competition law is fully effective. 126  The Court has also held that detailed national procedural rules for the functioning of NCAs must not jeopardise the attainment of the objective of Regulation 1/2003, which is to ensure that Articles 101 and 102 TFEU are applied effectively by those authorities. 127

In terms of the benefits it would bring, NCAs would become more effective enforcers. Only action at EU level can ensure that fines can effectively be imposed on companies by tackling national provisions which prevent the imposition of effective fines on companies for infringements of the EU competition rules. It is also the only means to address the wide variations in levels of fines imposed by different authorities and ensure that there are common core principles for imposing fines, thus, providing a more level playing field for business and ensuring deterrent fines across the EU. This is also the case for leniency programmes where companies regularly file applications to a number of EU jurisdictions and need guarantees of cross-border legal certainty. Such cross-border legal certainty cannot be sufficiently achieved by Member States individually.

It would also ensure that the necessary safeguards are in place to ensure that the NCAs enforce the EU competition rules without taking instructions from anyone and have the necessary resources to perform their tasks. All NCAs would have minimum powers needed to investigate and take decisions to effectively enforce the EU competition rules.

Having minimum rules would reduce divergent outcomes for companies and make the application of the EU competition rules by NCAs more predictable and consistent across the EU. The competition authorities in the EU would be able to cooperate better with each other, as it would ensure that they are all in a position to effectively assist each other, e.g. by gathering evidence. The credibility of the ECN would thus be reinforced.

Giving NCAs the means and instruments to be more effective enforcers would boost effective enforcement of the EU competition rules. This would result in more open competitive markets, where companies compete more fairly on their merits and enabling them to generate wealth and create jobs. Greater competition boosts productivity - a key driver for economic growth, leading to greater prosperity for EU citizens, as well as long-term employment growth. Competition enforcement helps to remove barriers to enter markets. The single market would be reinforced and be fairer. Businesses would benefit from a more level playing field and consumers would profit from a more equivalent level of enforcement through-out the EU. More effective competition enforcement protects European consumers from business practices that keep the prices of goods and services artificially high and enhances their choice of innovative goods and services. As explained in more detail in Section 6.3 above, even a relatively small increase in the effectiveness of competition policy enforcement would give a significant boost to productivity. Option 3 thus actions two levers: First, it maximises the increase in effectiveness with a minimum of interference in national specificities by limiting the most detailed rules to the areas where they are most needed to improve effectiveness. Second, the effectiveness thus achieved translates into a significant increase in TFP, whose growth has been low in the EU over the last ten years. This approach is thus fully in line with the proportionality principle.

When it comes to costs, this would largely entail implementation costs of the legislative initiative for public administrations. These costs would be variable depending on the extent to which rules that NCAs are empowered to be effective enforcers are already in place, although all NCAs have the basic framework in place. It is therefore more a matter of filling certain gaps and removing specific limitations depending on the Member State concerned. The implementation costs for the public purse would be negligible compared to the benefits realised by making NCAs more effective enforcers, e.g. consumers pay lower prices for goods and services, and thus better 'value for money'.

Companies would face limited initial adaptation costs in terms of familiarisation with the new rules (which would vary depending on the Member State in which they operate). Overall, the costs for businesses involved in cross-border activities to adapt to different legal frameworks would reduce, reinforcing the single market.

7.1.4Option 4: EU legislative action to provide NCAs with detailed and uniform means and instruments

This option would aim at a higher level of harmonisation. It would bring limited additional benefits relative to option 3, but at the same time entail greater interference in national legal systems and traditions.

Providing for detailed and uniform means and instruments would create the best possible level playing field for the effective enforcement of the EU competition rules by NCAs, and thereby strengthen the single market. This would be to the benefit of businesses operating cross-border: uniform rules means lower costs in terms of becoming aware of different laws in different national frameworks and is also likely to lead to more predictable outcomes and increase legal certainty.

However, it is not evident that giving NCAs more detailed and uniform means and instruments would necessarily result in significantly more effective enforcement of the EU competition rules than by giving NCAs minimum tools and safeguards. There is of course some added enforcement value in going for a high level of harmonisation. For example, the NCAs might receive even more leniency applications, leading to the detection of secret cartels, if the criteria leniency applicants must fulfil were fully convergent and applied by a central authority. Fines would be even more homogenous and deterrent if the more detailed methodology for fine-setting were regulated at European level. However, there is not a linear relationship between having more regulation and the effectiveness of the NCAs. The real step forward in terms of NCAs' effectiveness is between option 2 and 3, not between 3 and 4.

However, the main drawback of a high level of harmonisation is that it would take limited account of national specificities, and would mean greater interference in national legal systems and traditions than in option 3. For example, setting up a one-stop-shop for leniency applications would require the full harmonisation of the conditions and thresholds to be met for leniency to be granted across Europe. This would interfere in Member States' systems at a time when leniency programmes are still novel in some countries and have not reached a stage of sufficient maturity to merit a broader review. Similarly, having detailed and uniform rules on the parameters for setting fines would remove the flexibility NCAs need to adapt the rules to new developments. Allowing NCAs to only impose administrative fines, and thereby abolishing the criminal route in Member States where this is a deep-seated tradition would not respect the right of Member States to design their enforcement systems in light of their national specificities. Setting uniform rules, would also prevent Member States from providing for further-reaching powers or procedural guarantees.

On balance, since the advantages of option 4 in terms of creating a more level playing field and thus boosting effective enforcement are not substantial, especially not as compared to the real step forward made between option 2 and 3, option 3 and its calibrated approach are more in line with the proportionality and subsidiarity principles than option 4.

7.1.5Overview of how the options compare

The table below provides an assessment of options 2, 3 and 4 against the baseline scenario based on a number of criteria, namely effectiveness, coherency of the options with other EU policies, costs, benefits, subsidiarity and proportionality. Annex XI also contains a table containing an overview of how the different options compare with the baseline scenario in terms of achieving the general and specific objectives.

Summary table – Assessment of options

7.2Preferred policy option

The preferred policy option best suited to achieve the general and specific objectives without disproportional interference in the Member States’ legal systems and traditions is option 3. EU legislation is the only effective instrument to provide NCAs with the means and instruments necessary to boost their enforcement of the EU competition rules. Having primarily minimum rules as described in section 6.3 strikes the right balance by achieving the set objectives at the lowest possible cost, without unduly interfering with the national legal traditions and specificities.

Uniform and detailed safeguards would only be used in limited cases where minimum rules would not suffice to meet the objective of boosting effective enforcement of the EU competition rules by NCAs. For example, when it comes to protecting corporate leniency statements and settlement submissions, it may be better to follow the precedent of the Damages Directive 2014/104/EU and provide for full protection against their disclosure to preserve incentives for companies to provide the necessary information to competition authorities.

Soft action could be used in a number of areas where it is useful to build on basic provisions in the EU legislative instrument, while leaving sufficient flexibility to adapt to both national specificities and new developments. For example, soft action could be used to put 'meat on the bones' of minimum rules regarding fines set out in EU legislation. Such soft action has the benefit of allowing more detailed solutions to be put in place, which can easily be adjusted to new developments. Complementary soft action would ideally be developed in the ECN. This would entail little costs as one of the key roles of the ECN is to develop policy on the co-enforcement of the EU competition rules and foster a common approach across Europe. The ECN is a tried and tested forum for devising soft action which has the support and buy-in of all NCAs.

Option 3 would also best meet the expectations of stakeholders, including many key stakeholders such as NCAs, business and consumer organisations, of which the large majority (79%) in favour of EU action, expressed a clear preference for legislative action. It would also meet the majority view (51%) of stakeholders in favour of action at EU level as they preferred a mixture of EU legislative and soft action, as opposed to 28% in favour of exclusive legislative action (see section 7.1.2 for details per type of stakeholder). Such broad support for EU legislative action covers all four specific objectives. Moreover, stakeholders' backing for EU legislative action is often even higher when considering their replies for each specific objective. For example, on the competition toolbox, 89% opt for some type of EU legislative action. This option would be most in line with the initial feedback from Member State ministries, who did not object/were favourable to EU action to ensure a common competition enforcement area in Europe, provided it does not unduly interfere with national traditions. For example, when it comes to fines the main feedback from the Member States was that it would be useful to have common rules in place to ensure that effective fines can be imposed, provided there is room for further guidance to be given by the Court of Justice of the European Union on issues such as the notion of undertaking, parental liability and economic and legal succession. There should also be room for soft action, such as ECN Recommendations, to ensure further convergence in a flexible way on issues such as fining methodologies.

Stakeholders who favour taking action at EU level think that it would have a positive or very positive impact on: (i) the effective enforcement of the EU competition rules (92%) and (ii) legal certainty for businesses (85%). They consider that costs for businesses would decrease (52%) and cooperation in the ECN would be enhanced (83%). Taking action at EU level would have a positive impact on the legitimacy of decisions taken by NCAs (83%) and on the investment climate and economic growth (79%). 128  

Annex III sets out the practical implications of the preferred policy option for NCAs, business, consumers and public administrations.

The preferred policy option would boost effective enforcement of the EU competition rules by the NCAs by closing the gaps and remove the limitations that NCAs face which prevent them from enforcing more effectively. NCAs would have the capacity they need to fulfil their tasks under Regulation 1/2003. Putting guarantees in place that the NCAs can effectively enforce the EU competition rules would bolster public confidence in the national frameworks for ensuring compliance with the EU competition rules and enhance the legitimacy of the European system of competition enforcement as a whole.

The preferred policy option would achieve the specific policy objectives as follows:

7.2.1All NCAs would have an effective enforcement toolbox

All NCAs would have the tools they need to be able to investigate infringements and to take decisions to bring them to an end. NCAs would no longer be prevented from detecting infringements because of restrictions, for example, in their power to gather evidence. NCAs would be able to set their enforcement priorities in full and choose which cases to dedicate their scarce resources, e.g. to tackle cases which may bring significant consumer benefits or those which may have a strong precedent value for businesses. Their investigation and decision-making powers would be meaningful because they would be backed up by deterrent fines. Companies would think twice about not complying with a NCA decision. At the same time, it would be ensured that adequate procedural guarantees are in place in line with the EU Charter of Fundamental Rights.

7.2.2All NCAs would be able to impose (or seek) deterrent fines

All NCAs would be able to impose deterrent fines on companies for breaches of the EU competition rules. This would reduce wide variations in fines depending on which authority acts. It would also prevent companies escaping fines altogether, for example, by restructuring. NCAs would be able to impose (or seek) fines for the entire duration of an infringement and not just part of it, meaning that the fines better reflect the harm caused to competition. NCAs which currently can only impose primarily criminal fines, meaning that they have low levels of enforcement and fines imposed, particularly in abuse of dominance cases, would be given the option of imposing administrative fines or apply to civil courts for fines to be imposed, depending on the facts and circumstances of the case. Giving those NCAs this possibility means that they are no longer limited in terms of what cases they can take (now usually limited to the more straightforward cases such as hard-core cartels) and the fines they can impose or seek. This option would also improve criminal enforcement by enabling NCAs to bring and defend their cases in court. By making fines more deterrent and effective, companies contemplating breaking the EU competition rules would be much less likely to decide that the potential gains outweigh the risks. Complementary soft action could be taken on certain non-core aspects of the methodologies for imposing fines to develop an even more effective and coherent fining policy across the EU on a flexible basis. This could for example cover which aggravating/mitigating factors should be taken into account and how to assess the gravity of the offence.

7.2.3All NCAs would have effective leniency programmes which would encourage companies to come clean across the EU

All NCAs would have effective leniency programmes in place adapted to their needs, with a coherent EU system to deal with applications to multiple jurisdictions. This would increase legal certainty for companies as to their place in the leniency queue and would enable them to know whether they are eligible for immunity from fines or potentially no reduction in fines at all. With such a system in place, companies will be incentivised to apply for leniency. This will help NCAs to detect cartels which are often difficult to find out about due to their secret nature. Putting in place arrangements to ensure that employees of companies that make leniency applications to NCAs and/or the Commission are protected from sanctions will also ensure that companies are incentivised to apply for leniency. Without these guarantees, the legal risks for the individuals involved may have a chilling effect on the willingness of companies to come clean about their involvement in cartels. This prevents NCAs from becoming aware of the most pernicious forms of competition infringements.

7.2.4All NCAs would have guarantees of independence and resources

All NCAs would have guarantees that when they enforce the EU competition rules they act for the common good and they neither seek nor take instructions from any public or private body. This would give NCAs a shield to protect them from interference in their decision-making and allow them to act impartially on the basis on the facts and law. Trust in enforcement by NCAs would be enhanced. It would also mean that NCAs have a basic guarantee that they are not prevented from performing their tasks because of a complete lack of resources. This would best meet the views expressed by Member State ministries to date, who did not object to a basic hook, which does not prevent them from taking into account cyclical fluctuations.

7.2.5Nature of the instrument and legal basis

When taking EU legislative action, a choice has to be made between either a regulation or a directive. A regulation is directly applicable, and thus automatically deemed to be enshrined in Member State law, there is no need for implementing legislation. However, it leaves Member States very limited or no scope to adapt to their national specificities. A directive is only binding as to the result to be achieved, which gives Member States a choice as to the form/method to achieve this goal. Directives can set minimum standards which do not prevent Member States from having in place provisions which go further. It also allows for more uniform and detailed requirements to be put in place where appropriate.

The aim of the preferred policy option is to enhance effectiveness, while not imposing one size fits all so as to allow taking into account Member States’ legal traditions and institutional specificities. Accordingly, a regulation would not seem an appropriate instrument, but rather a directive would be the best way of ensuring that NCAs have the guarantees they need to be more effective enforcers, without unduly interfering in national specificities and traditions. A directive would also better reflect the expectations of stakeholders, 65% of whom in the public consultation said they prefer action to be taken at both EU and national level. In terms of the legal basis for a directive, Article 103 TFEU enables the EU to adopt legislation that gives effect to the EU competition rules. Article 114 TFEU allows for the adoption of legislation which supports the functioning of the internal market.

8Monitoring and evaluation

This section describes the monitoring and evaluation that could be applied to assess the impact of the preferred option, namely, a combination of EU legislative action in the form of a directive, complemented by soft action. The monitoring and evaluation is designed to cover the four specific objectives indicated in section 4.

As a first step, the Commission would try to refine and update the quantitative baseline at the moment when the new legislation has been adopted by the legislator.

As a second step, the monitoring would be focused on the implementation of the directive by the Member States in the run up to the date for transposition. The directive could require Member States to communicate to the Commission the preliminary text of the national provisions implementing the directive and any explanatory documents, thereby allowing the Commission to assess whether there has been adequate implementation. As a second step, the monitoring would be focused on both the transposition and the practical application by Member States and NCAs of the provisions of the legislative initiative, as well as the implementation of soft law measures, to see whether the four specific objectives have been achieved. The monitoring would also cover the impact of the initiative in monetary terms on NCAs.

The monitoring would be based mainly on indicators measuring the availability and application of the measures in the directive and, where possible, on quantitative/monetary indicators that could help to measure the impact of such measures and which can serve as basis for the evaluation. Annex XV provides an overview of the proposed indicators that could be used to assess whether each of the four specific objectives have been met. The sources of information for an assessment based on these indicators would range from fact-finding within the ECN to input received from stakeholders.

Complementary soft action would be developed within the current framework of the ECN, which has a successful track record for reaching consensus on such measures. This forum could also be used to discuss possible difficulties faced at national level in the implementation of the directive, learn from each other on potential solutions and exchange experience on the implementation of both the soft action and the legislative measures. The ECN framework would therefore be a key tool providing useful information to monitor the preferred option, which would complement the monitoring based on the core indicators.

During the first step of implementation of the Directive, monitoring would take the form of a report taking stock of the implementing measures adopted by each Member State. During the second step of application of the Directive, every two years there would be a monitoring of the values of the indicators and how they have evolved over time per Member State.

On the basis of the evidence gathered through the monitoring exercise, an evaluation would be carried out to assess the performance against each of the four objectives, as well as the possible existence of unintended/unexpected effects. While monitoring would be partly a continued process during the period of implementation of the directive, before carrying out a useful ex-post evaluation a reasonable period of time after its transposition would be needed to assess the impact of the changes in practice and whether the four objectives have been met. An ex-post evaluation would therefore be carried out after 5 years from the date of its transposition. This would also allow time for the complementary soft measures to be developed in the ECN and to be applied in practice.

The monitoring and evaluation process would provide useful information for any potential future modifications.

(1)  See the current priorities of the European Commission: https://ec.europa.eu/priorities/index_en.

(2)  Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty (OJ 2003, L 1, p.1).

(3)  EU citizens who have heard of a case being taken by a competition authority are more likely to think that it was their NCA who took the decision, see Flash Eurobarometer 403, "Citizens Perceptions about Competition Policy", March 2015.

(4)  The approach followed to benchmark the observable customer benefits from stopping a cartel (prevented harm) is broadly in line with the OECD Guide for assessing the impact of competition authorities' activities (April 2014), available at https://www.oecd.org/daf/competition/Guide-competition-impact-assessmentEN.pdf. It consists of multiplying the assumed increased price brought about by the cartel (called the “overcharge”) by the value of the affected products or markets and then by the likely duration of the cartel had it remained undetected. A 10% to 15% overcharge is assumed. This is conservative when compared to the findings of empirical literature which report considerably higher median price overcharges for cartels. In order to estimate what the likely duration of the cartel would have been if it had continued undetected, a case-by-case analysis was carried out. This analysis focussed on the particular circumstances of each case and an assessment of key indicators, including the specific market conditions, the lifespan of the cartel, the ease of reaching and renewing cartel agreements as well as the potential reactions of outsiders (such as new entrants). The cartels are classified into three categories: "unsustainable", "fairly sustainable" "very sustainable". It is assumed that the cartels in the first category would have lasted one extra year in the absence of the Commission's intervention, the cartels in the second category 3 years, and the cartels in the third group 6 years. The assumptions concerning the likely duration of the cartels are made prudently to establish a lower limit rather than to estimate the most likely values. Finally, the estimates obtained are also conservative because other consumer benefits, such as innovation, quality and choice are not taken into account. The methodology followed by the Commission is not connected to the size of GDP but rather reflects the number of competition enforcement decisions taken per year. This explains the fluctuations in the Commission's yearly estimates. Similarly, the order of magnitude of customer benefits in a given year/period estimated by NCAs which carried out such an exercise is based on the number of competition enforcement decisions taken in that Member State.

(5)  Annual Activity Reports of the Directorate-General for Competition of the European Commission for the years 2010-2015.

(6)  See Dierx A., Heikkonen J., Ilkovitz F., Pataracchia B., Ratto M., Thum-Thysen A. and Varga J. (2015), "Distributional macroeconomic effects of EU competition policy – A general equilibrium analysis", paper to be published in a World Bank-OECD publication on Competition Policy, Shared Prosperity and Inclusive Growth. This paper tries to bridge the gap between the microeconomic estimates of the customer savings associated with important merger and cartel interventions and the longer term macroeconomic effects of these interventions. It also attempts to measure not only the direct effects of competition policy interventions but also their deterrent effects. Finally, it sheds some light on the distributional impact of competition policy. This objective is ambitious and the simulations are reliant on a chain of assumptions, going from the calculation of customer savings and the approximation of the deterrent effects to the specification and calibration of the general equilibrium model. However, these assumptions do not undermine the usefulness of the analysis in getting a better understanding of the role of competition policy in society.

(7)   https://www.acm.nl/en/publications/publication/14113/2014-ACM-Annual-Report/ . This assessment appears to rely on the amount of fines the authority imposed and/or collected as well as the remedies it imposed on certain merger plans in 2014 so it is expected to vary significantly from one year to the next.

(8)  See the Impact Assessment 2015/2016 of the UK Competition and Markets Authority, https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/537539/cma-impact-assessment-2015-16.pdf.

(9)  These estimates necessarily rely in part on assumptions. The UK competition authority explains that its impact estimations are conducted immediately after cases are completed and are therefore based only on information available during the case and on assumptions regarding the expected impact of its interventions. On this basis the estimates are considered to be "ex ante’ evaluations. The authority considers that in general, the assumptions it applies are cautious and hence the estimates are conservative. For example, it excludes impacts from a number of cases where consumer benefits were difficult to quantity in a sufficiently robust manner. Further, its estimates exclude its compliance work, international activities, and advocacy to government for policies that support competition because the benefits of these are difficult to quantify due to the nature of the work.

(10)  See https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/539987/cma-annual-report-and-accounts-2015-16-web-accessible-version.pdf . The authority undertakes these assessments itself, with subsequent review by external academics. Its methodology is based on that developed by the predecessor to the current UK competition authority (the Office of Fair trading), validated by successive independent academic reviewers. The current authority, the Competition and Markets Authority, considers that this is consistent with approaches now regarded by the OECD as international good practice.

(11)  European Parliament resolution of 19 January 2016 on the Annual report on EU Competition Policy (2015/2140(INI)), introduction, letter G. See further Mariniello, M. (2013) "Do European Union Fines Deter Price Fixing", breugelpolicybrief, Issue 2013/04.

(12)  Ormosi, P. L. (2014) "A tip of the iceberg? The probability of catching cartels", Journal of Applied Economics, Vol. 29(4) 549.

(13)  A 15% probability of detection appears to correspond to the average of estimations by several researchers and is used by most authors in the relevant literature (see Combe, E and C Monnier (2009), “Fines against hard core cartels in Europe: the myth of over enforcement”, Cahiers de Recherche PRISM-Sorbonne Working Paper.

(14)  See Mariniello, M. (2013) "Do European Union Fines Deter Price Fixing", breugelpolicybrief, Issue 2013/04.

(15)  See Connor, J. M. & Bolotova, Y. (2006) "Cartel overcharges: survey and meta-analysis", International Journal of Industrial Organisation, 24(6), 1109-1137 and Smuda F. (2014), "Cartel Overcharges and the deterrent effect of EU competition law", Journal of Competition Law & Economics, Vol. 10(1), pp. 63-86.

(16)  London Economics (2011), "The impact of competition interventions on compliance and deterrence", OFT Report No. 1391, December, and SEO (2011), "Anticipating cartel and merger control", SEO-report.

(17)  See Dierx A., Heikkonen J., Ilkovitz F., Pataracchia B., Ratto M., Thum-Thysen A. and Varga J. (2015), "Distributional macroeconomic effects of EU competition policy – A general equilibrium analysis", paper to be published in a World Bank-OECD publication on Competition Policy, Shared Prosperity and Inclusive Growth.

(18)  See https://www.cnmc.es/Portals/0/Ficheros/download/cnmc_competition.pdf , p. 22.

(19)  Flash Eurobarometer 403, "Citizens Perceptions about Competition Policy", March 2015.

(20)  EU citizens have also experienced competition problems in the telecommunications and internet sector (18%), food distribution (14%) and financial services (12%).

(21) Consumer Market Scoreboards available at: http://ec.europa.eu/consumers/consumer_evidence/consumer_scoreboards/index_en.htm .

(22)  See the Digital Economy and Society Index (DESI) and in particular the connectivity dimension: https://ec.europa.eu/digital-single-market/en/fixed-bb-price-desi-indicator-1d1.

(23)  See the Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, 2014 European Semester: Country-specific recommendations-building growth, COM(2014) 400 final, which states: "An efficient competition framework is a key aspect for the functioning of markets in goods and services. Progress has been mixed in this area".

(24)  In April 2015, the Commission sent a statement of objections to Gazprom over concerns that the latter is breaking EU antitrust rules by pursuing an overall strategy to partition Central and Eastern European gas markets (see press release in Case 39816 at http://europa.eu/rapid/press-release_IP-15-4828_en.htm). In December 2015, the Commission accepted commitments by Bulgarian Energy Holding to open up the Bulgarian wholesale electricity market (see press release in Case 39767 at http://europa.eu/rapid/press-release_IP-15-6289_en.htm ).

(25)  For example, following a complaint filed by Direct Energie, the French NCA adopted interim measures ordering the French incumbent gas operator GDF Suez to provide access to information contained in its database with respect to customers under the regulated gas tariffs which was considered to be essential for the successful entry of new operators and to allow them to make competitive non-regulated offers and therefore for the development of the energy market (Decision of the French NCA 14-MC-02 of 9 September 2014). In the UK, the UK NCA has been carrying out a detailed investigation of the supply and acquisition of energy in the British market (see final report of 24 June 2016 available at https://assets.publishing.service.gov.uk/media/5773de34e5274a0da3000113/final-report-energy-market-investigation.pdf ) and is now in the phase of implementing the remedies.

(26)  See the speech of Commissioner Vestager of 10 June 2016 in which she stated that all competition authorities in Europe have to play their part to make energy markets competitive: http://ec.europa.eu/commission/2014-2019/vestager/announcements/making-energy-markets-work-consumers_en.

(27)  OECD, 2014, Fact-sheet on how competition policy affects macro-economic outcomes.

(28)  For example, Taylor, J. E. (2002) "The output effects of government sponsored cartels during the New Deal", the Journal of Industrial Economics, 50(1), 1-10 and (2007).

(29)  Taylor, J. E. (2007) "Cartel Code Attributes and Cartel Performance: An Industry-Level Analysis of the National Industry Recovery Act", Journal of Law and Economics, 50(3), 597-624.

(30)  For all references to the public consultation, further information is provided in Annex II.

(31)  The European Court of Justice has ruled that procedural rules at national level should not jeopardize the attainment of the objective of Regulation 1/2003, which is to ensure that Articles 101 and 102 TFEU are applied effectively by the NCAs, Case C-439/08 Vebic ECLI:EU:C:2010:739.

(32)  European Parliament resolution of 19 January 2016 on the Annual report on EU Competition Policy (2015/2140(INI)), introduction, letter E.

(33)  See for example, Borrel, J-R and Jiménez, J. L., (2008) "The Drivers of Antitrust Effectiveness", Instituto de Estudios Fiscales.

(34)  Buccirossi, P., Ciari, L., Duso, T., Spagnolo, G., Vitale, C., (2013) "Competition Policy and Productivity Growth: An Empirical Assessment, Review of Economics and Statistics", 95(4), 1324-1336. The study assessed 12 OECD countries, nine of which are EU Member States.Indicators developed in OECD Working Papers which measure the strength of competition law and policy include powers to investigate, powers to sanction and remedy and the independence of the institutions enforcing competition law, Alemani, E., Klein, C., Koske, I. Vitale, C. and Wanner, I. (2013). "New Indicators of Competition Law and Policy in 2013 for OECD and non-OECD countries", OECD Economics Department Working Paper no. 1104 See also Hoj, J. (2007) "Competition law and policy indicators for the OECD countries", OECD Economics Department Working Paper no 568.

(35)  Communication from the Commission to the European Parliament and the Council, Ten Years of Antitrust Enforcement under Regulation 1/2003: Achievements and Future Perspectives, COM(2014) 453 and the accompanying Staff Working Documents: Ten Years of Antitrust Enforcement under Regulation 1/2003, SWD(2014)230 and Enhancing Competition by the Member States’ competition authorities: institutional and procedural issues, SWD(2014) 231, http://ec.europa.eu/competition/antitrust/legislation/regulations.html.

(36)  Communication from the Commission to the European Parliament and the Council, Report on the functioning of Regulation 1/2003 COM(2009) final and the accompanying Staff Working Paper SEC(2009) 574 final, http://ec.europa.eu/competition/antitrust/legislation/regulations.html.

(37)  See, e.g. the Guidance of the International Competition Network on Investigative Process: http://www.internationalcompetitionnetwork.org/uploads/library/doc1028.pdf , and the Antitrust Enforcement Manual of the International Competition Network: http://www.internationalcompetitionnetwork.org/working-groups/current/cartel/manual.aspx.

(38)  In November 2012, the ECN made a detailed inventory of the investigation and decision-making procedures for competition enforcement which exist in the Member States: http://ec.europa.eu/competition/ecn/documents.html . The Reports provided a clear overview of the status quo for the first time. Further fact-finding was carried out in the ECN by way of follow-up to the 2014 Communication. Annex XVI illustrates that for a sample of 5 competition tools, 25 NCAs are lacking at least one of these powers.

(39)  The ECN Recommendation on Investigative Powers, Enforcement Measures and Sanctions in the context of Inspections and Requests for Information (available at http://ec.europa.eu/competition/ecn/recommendation_powers_to_investigate_enforcement_measures_sanctions_09122013_en.pdf) recommends that all authorities have effective powers to inspect non-business premises.

(40)  For example, in a cartel among 15 commercial banks in Portugal, using forensic IT techniques was the only possible way to find evidence for the Portuguese NCA. Similarly an increasing number of Commission decisions rely on information which was exclusively exchanged by digital means.

(41)  The ECN Recommendation on the Power to impose Structural Remedies (available at http://ec.europa.eu/competition/ecn/structural_remedies_09122013_en.pdf ) provides that is desirable that all authorities have the choice and the power to impose both behavioural and/or structural remedies.

(42)  The ECN Recommendation on Commitment Procedures ( http://ec.europa.eu/competition/ecn/ecn_recommendation_commitments_09122013_en.pdf ) recommends that all authorities can impose effective fines for non-compliance with a commitment decision and have effective means to compel compliance, e.g. the imposition of periodic penalty payments set at an appropriate level.

(43)  The ECN Recommendation on the Power to set Priorities (available at http://ec.europa.eu/competition/ecn/recommendation_priority_09122013_en.pdf) advocates for authorities to have greater flexibility to choose which cases to investigate. 8 NCAs are obliged to investigate cases which are not a priority and 15 NCAs cannot reject complaints which are not a priority without doing a detailed investigation on substance.

(44)  For the purpose of this impact assessment, references to fines imposed by NCAs is shorthand for fines imposed by the authorities designated by the Member States for the enforcement of Articles 101 and 102 TFEU, whether of an administrative or a judicial nature.

(45)  Judgment in Inspecteur van de Belastingdienst v X BV, C-429/07, EU:C:2009:359, paragraphs 36-39. The case concerned the question whether fines imposed by the Commission can be wholly or partially deductible from Dutch taxes. The ECJ ruled that the effectiveness of the Commission’s decision by which it imposed a fine on a company might be significantly reduced if the fine could be deducted from tax. The Court stated: "To dissociate the principle of prohibition of anticompetitive practices from the penalties provided for … would therefore deprive of any effectiveness the action taken by the authorities responsible for monitoring compliance with that prohibition and punishing such practices. Thus, the provisions of Articles 81 EC and 82 EC [now Articles 101 and 102 TFEU] would be ineffective if they were not accompanied by enforcement measures provided for in Article 83(2)(a) EC [now Article 103(2)(a)]" (Article 103(2)(a) provides for the imposition of fines and penalty payments to ensure compliance with the EU competition rules).

(46)  No issues have arisen demonstrating that there would be a need to have a minimum level of fine that can be set. In some cases, e.g. in areas where the legal situation was unclear, competition authorities need to have the ability to impose symbolic fines, which would be prevented by having a minimum level of fine.

(47)  While many NCAs apply a percentage of 10%, in other Member States the percentages applied are lower (up to 5%) for less serious infringements. Similarly, in one Member State, a cap of 5% is imposed on the turnover of the direct infringer only for vertical anti-competitive practices between companies operating at different levels of the supply chain i.e. agreements between a manufacturer and its distributor and abuses of dominant position contrary to Article 102. In another Member State, the cap is generally set at 10% for competition infringements, but for the specific case of cartels, the cap is 10% for each year of infringement up to a maximum of 4 years: this means that the maximum can reach 40% for cartels lasting 4 or more years. Moreover, these amounts can be doubled for cartels in cases of recidivism (that is, if a company has already been found to have breached competition law), with the result that the legal maximum could potentially reach 80% of worldwide turnover.

(48)  Case C-97/08 P AkzoNobel NV v Commission [2009] ECR I-8237. It has to be shown that the parent company exercises decisive influence over the subsidiary that committed the infringement.

(49)  In 1 Member State it is not possible to fine associations of undertakings and in 9 Member States the fine can only be based on the turnover of the association, and not the turnover of its Members.

(50)  For the purpose of this impact assessment, the term “civil fines” is shorthand for fines imposed by a court in civil proceedings.

(51) The Irish NCA stated in the public consultation that criminal sanctions are neither appropriate nor practicable in relation to non-hard core competition law infringements (for which it considers civil procedures would be more suitable) and that it does not in practice pursue criminal prosecutions in such cases. Estonia also confirmed in its reply to the public consultation that the fact of having a criminal (misdemeanour) system makes competition enforcement more complicated.

(52)  There are other Member States that have reported a low number of cases, but these are small countries whose NCAs had insufficient resources.

(53) Regarding criminal fines, 56% of the respondents consider that it is a problem that some NCAs impose only/primarily criminal fines vs. 9% that disagree with this conclusion. Regarding civil fines, the percentage of respondents considering that it is a problem that some NCAs impose only/primarily civil fines decreases to 45%, vs. 21% that disagree with this conclusion. In both cases, the support to the conclusion that there is a problem tends to be higher amongst companies/SMEs and consultancy/law firms, and lower amongst public authorities and industry associations.

(54)  DAF/COMP(2015)1/FINAL.

(55) Issues Paper on Regulatory and Corporate Offences: www.lawreform.ie.

(56)  Government committee of inquiry on enhanced decision-making in the Swedish Competition Authority, ("En utökad beslutanderätt för Konkurrensverket, Betänkande av Utredningen om en utökad beslutanderätt för Konkurrensverket", June 2016) which concluded that moving to an administrative system would make the Swedish competition enforcement system more effective through fast high-quality decisions of the NCA, greater certainty for potential leniency applicants and better cooperation by the Swedish NCA in the ECN.

(57)  The Swedish NCA issued a report in December 2013 advocating for a change from its civil system to an administrative system (Konkurrensen i Sverige 2013, Rapport 2013:10).

(58)  A study has shown that the introduction of the leniency programme in the US resulted in a 40% reduction in the number of detected cartels: Miller, N. H. (2009) "Strategic Leniency and Cartel Enforcement, American Economic Review", Vol. 99(3) 750. The number of cartel discoveries increases around the date of the introduction of the leniency programme and then falls below pre-leniency programme levels, consistent with enhanced cartel detection and deterrence.

(59)  Rather than having to file complete leniency applications with all NCAs with (potential) jurisdiction to take actions against the cartel, a summary application system allows companies to file a leniency application to these NCAs on the basis of more limited information where a full application has been given to the Commission.

(60)  In the judgment in DHL Express (Italy) S.r.l. and Others v Autorità Garante della Concorrenza e del Mercato and Others, Case C-428/14, EU:C:2016:27, the Court held that soft instruments adopted in the context of the ECN are not binding. As a result, Member States are not required to incorporate provisions of the ECN Model Leniency Programme in their leniency systems and, they are not precluded from adopting rules not present in that model programme or which diverge from it.

(61)   http://ec.europa.eu/competition/ecn/documents.html . The Model Leniency Programme was revised in 2012.

(62)  A formal settlement is a simplified procedure which results in the faster handling of the case and in a reduction of the fines. In order to benefit from this procedure, the companies involved have to acknowledge their participation in the infringement.

(63)  In 2013, the OECD presented its new competition law and policy indicators which measure the strength and scope of competition regimes. Independence of competition authorities is one of the indicators as there is broad consensus among OECD countries that it constitutes good practice for competition regimes. See E. Alemani and others, "New Indicators of Competition Law and Policy in 2013 for OECD and non-OECD Countries" (2013) OECD Economics Department Working Papers, No. 1104, OECD Publishing. In an OECD survey on competition policy, “greater independence” was the factor most frequently identified as likely to lead to better promotion of competition law’s objectives. OECD, Global Forum on Competition, "The Objectives of Competition Law and Policy: Note by the Secretariat" (2003) Session 1, Doc. No. CCNM/GF/COMP(2003)3, 8.

(64)  See UNCTAD, "Independence and accountability of competition authorities" (2008), Note by the UNCTAD secretariat referring to other international organizations including the World Trade Organization, World Bank, International Monetary Fund and the Organization for Economic Cooperation and Development.

(65)  A. Mateus, "Why should national competition authorities be independent and how should they be accountable?" (2007) 3(1) European Competition Journal 17; and G. Monti, "Independence, Interdependence and Legitimacy: The EU Commission, National Competition Authorities, and the European Competition Network" (2014) EUI Working Paper No. 2014/1, 4-5.

(66)  The European Court of Justice has also recognised the link between guaranteeing the independence of national supervisory authorities and ensuring the effectiveness of such supervision. See judgment in Commission v Germany, C-518/07, EU:C:2010:125, para 25.

(67)  Ibid., para 18.

(68)  European Parliament Resolution of 11 December 2013 on the Annual Report on EU Competition Policy (2013/2075 (INI)); European Parliament resolution of 10 March 2015 on the Annual Report on EU Competition Policy (2014/2158 (INI)); European Parliament resolution of 16 January 2016 on the Annual Report on EU Competition Policy (2015/2140 (INI)).

(69)  Austria, Latvia, Luxemburg, Malta and Slovenia, see http://ec.europa.eu/europe2020/index_en.htm .

(70)  96 respondents supported the importance of this measure, which was considered as the first most important measure by 50%, the second by 28%, and the third for the remaining 22%. This support was particularly strong amongst consumer and non-governmental organisations, and public authorities (both NCAs and particularly ministries), while it was somehow lower amongst companies/SMEs, industry associations and consultancy/law firms.

(71)  GDP is used as a proxy for the size of markets.

(72)  For example, on 12 May 2016, the Latvian Parliament adopted amendments to the Competition Law which among others enable the NCA to impose penalties on undertakings which do not provide information requested by it or provide false or misleading information.

(73)  In 2009, a Report was made assessing the level of convergence with the ECN MLP. Similarly, ECN Reports of 2012 provided an EU-wide overview of the different procedures for enforcing the EU competition rules. The reports are available at: http://ec.europa.eu/competition/ecn/documents.html .

(74)  See Country Specific Recommendations addressed to Austria (2013 – 2015) and Slovenia (2012 – 2014), http://ec.europa.eu/europe2020/index_en.htm .

(75)  Commission Guidelines on the effect on trade concept contained in Articles 101 and 102 TFEU, OJ 2004, C 101, p.81, para. 50.

(76)  Commission Notice on Agreements of Minor Importance, OJ 2014, C 291, p.1.

(77)  The ECN endorsed Recommendations on (1) Investigative Powers, Enforcement Measures and Sanctions in the context of Inspections and Requests for Information; (2) The Power to collect Digital Evidence, including by Forensic Means; (3) Assistance in Inspections conducted under Article 22(1) of Regulation 1/2003; (4) The Power to set Priorities; (5) The Power to grant Interim Measures; (6) Commitment Procedures and (7) The Power to Impose Structural Remedies. The Recommendations are available at: http://ec.europa.eu/competition/ecn/documents.html.

(78)  http://www.autoritedelaconcurrence.fr/doc/eca_principles_uk.pdf.

(79)  http://ec.europa.eu/competition/ecn/ncas.pdf.

(80)  Article 6(6) of the Directive 2014/104/EU of the European Parliament and of the Council of 26 November 2014 on certain rules governing actions for damages under national law for infringements of the competition law provisions of the Member States and of the European Union, OJ 2014, L 349, p.1.

(81)  Where NCAs have other competences, such as consumer protection or regulatory functions, Member States have to ensure that sufficient budget and staff be assigned to competition enforcement.

(82)  See Commission's "White paper on modernisation of the rules implementing Articles 85 and 86 of the EC Treaty" and the explanatory memorandum of the proposal for the Regulation implementing Articles 81 and 82 of the Treaty (OJ C 365 E, 19/12/2000 P.0284-0296).

(83)  A formal settlement is a simplified procedure which results in the faster handling of the case and in a reduction of the fines. In order to benefit from this procedure, the companies involved have to acknowledge their participation in the infringement.

(84)  Instead of civil fines, the Government proposed legislation in September 2011 to double jail terms for cartel offences from five to ten years., see "Boom, Bust and Bailout: A Tale of Modern Irish Competition Law" by Philip Andrews, available at http://kluwercompetitionlawblog.com/2012/07/12/boom-bust-and-bailout-a-tale-of-modern-irish-competition-law/.

(85)  Judgment in DHL Express (Italy) S.r.l. and Others v Autorità Garante della Concorrenza e del Mercato and Others, C-428/14, EU:C:2016:27.

(86)  See DHL Express (Italy) S.r.l. and Others v Autorità Garante della Concorrenza e del Mercato and Others, Case C-428/14, EU:C:2016:27.

(87)  The results were in general similar also per type of stakeholder, with smaller support coming normally only from non-governmental organisations and industry associations. However, even for these two groups of stakeholders, the most frequent alternative replies were that the effects would be neutral or that they had no opinion, (very) negative effects being supported generally by less than 7%.

(88)  The best evidence for the effectiveness of competition law enforcement tends to be that based at the level of the enforcement itself. Competition authorities and academics have published a large number of ex-post studies of the results of enforcement actions, which were surveyed by the OECD in 2013: "Evaluation of competition enforcement and advocacy activities: The results of an OECD survey".There are also meta-studies which have sought to measure the effectiveness of competition enforcement across a large number of cases. For example, Dutz, M. & Vagliasindi, M. (2000), "Rules versus implementation: determinants of competition policy effectiveness in transition countries", EBRD, London used data on a number of transition economies to show that better implementation of competition law leads to better competition. In a previous study Dutz and Hayri had also found a positive link between measures of competition law effectiveness and GDP growth (Dutz M. & Hayri, A. (1999). "Does More Intense Competition Lead to Higher Growth?", Policy Research Working Paper No.2320. World Bank).

(89)  Keith N. Hylton and Fei Deng, (2007) Antitrust Around the World: An Empirical Analysis of the Scope of Competition Laws and Their Effects.

(90)  Arnold, J. M., Nicoletti, G., & Scarpetta, S. (2011) Regulation, Resource Reallocation and Productivity Growth. European Investment Bank Papers,16(1), 90-115; OECD’s project of 2013 on Supporting Investment in Knowledge Capital, Growth and Innovation.

(91)  A number of surveys suggest that the deterrent effect of antitrust enforcement may be substantial. See Baarsma B., Kemp R., van der Nol, R., and Seldeslachts J. (2012), " Let's not stick together: anticipation of cartel and merger control in the Netherlands ", De Economist, which estimated on the basis of a survey of lawyers and other advisors that for every sanction decision taken by the Dutch competition authority there are almost 5 other cases in which a prohibited act has been terminated or modified in response to advice on competition law. A study commissioned by the former UK competition authority, the Office of Fair Trading (OFT): London Economics (2011), "The impact of competition interventions on compliance and deterrence", OFT Report no 1391, found, based on a survey of more than 800 companies and a small number of law firms, that for each abuse of dominance case, 12 potential infringements are deterred. For cartels 28 potential agreements were deterred and for commercial agreements (anti-competitive agreements between companies which are not cartels), the ratio was 1:40.

(92)  See Nickell, S. (1996) "Competition and Corporate Performance", Journal of Political Economy 104(4), 724-746 which found that the most competitive firms experienced productivity growth rates 3.8-4.6% higher than the least competitive. See also Disney, R., Haskel, J., & Heden, Y. (2003). "Restructuring and productivity growth in UK manufacturing". The Economic Journal, 113(489), 666-694; Blundell, R., Griffith, R., & van Reenen, J. (1999). "Market Share, Market Value and Innovation in a Panel of British Manufacturing Firms. Review of Economic Studies", 66(3), 529-54; Januszewski, S. I., Köke, J. & Winter, J. K. (2002). "Product market competition, corporate governance and firm performance: an empirical analysis for Germany". Research in Economics, 56(3), 299-332. 

(93)  See OECD, 2014, Fact-sheet on how competition policy affects macro-economic outcomes.

(94)  See Ahn, S. (2002). "Competition, Innovation and Productivity Growth: A Review of Theory and Evidence". OECD Economics Working Paper No. 317. A study carried out by Petit L., Kemp R. and van Sinderen J. (2015) "Cartels and productivity growth: an empirical investigation of the impact of cartels on productivity in the Netherlands", assessed the impact of cartels on total factor productivity (TFP). TFP is a measure of the output of a company, sector or total economy that cannot be explained by the amount of inputs used in production and whose level is determined by how efficiently and intensely the inputs are utilized and is an indicator of competitiveness. The results showed that the entry and presence of a cartel had a negative impact on TFP and it was estimated that cartels had a negative impact on TFP of between 2% to 3% during the period covered.

(95)  Bucirossi, P., Ciari, L., Duso, T., Spagnolo, G. & Vitale, C. (2011) "Measuring the deterrence properties of competition policy: the competition policy indices" Journal of Competition Law & Economics, 7(1), 165-204.

(96)  Buccirossi, P., Ciari, L., Duso, T., Spagnolo, G., Vitale, C., (2013) "Competition Policy and Productivity Growth: An Empirical Assessment", Review of Economics and Statistics, 95(4), 1324-1336. Similarly, Dutz, M. & Hayri, A. (1999), "Does more intense competition lead to higher growth?" Policy Research Working Paper No. 2320, World Bank, found a strong correlation between long-run GDP growth, and effective enforcement of competition rules, on the basis of a cross-section of 52 countries.

(97)  The figure is illustrative to make concrete the effects of the elasticity between total factor productivity growth and the aggregate Competition Law and Policy Index the authors of the study constructed.

(98)  To put it simply, if an economy is able to produce more with the same amount of inputs, its TFP increases. To illustrate the importance of TFP, an annual growth of TFP of 1% would mean that an economy using the same amount of input resources would increase its production with around 10.5% over ten years. If the growth of TFP is only 0.5% the increase in production would only be 5.1% higher.

(99)  The authors recognise that there is scope for further refinements, such as expanding the study to more countries, to a longer period, and that the CPI could also be improved if more detailed information would be available (such as the sanctions actually imposed). On the other hand, econometric tools have been used to support the conclusion that the link between competition policy and TFP growth is of a causal nature, and the findings prove to be robust to several checks such as the use of various measures of productivity, different aggregation techniques of the CPI and the use of several subsamples. (see Buccirossi, P., Ciari, L., Duso, T., Spagnolo, G., Vitale, C., (2013) "Competition Policy and Productivity Growth: An Empirical Assessment", Review of Economics and Statistics, 95(4), 1334-1335.

(100) 0.004976 x 0.09 = 0.0004478 ~0.0448%.

(101)

     See Clougherty, J. A. (2010), "Competition Policy Trends and Economic Growth: Cross-National Empirical Evidence", International Journal of the Economics of Business, 17(1), 111-127. The study found that increasing competition policy funding by about €60 million would result in economic growth of 0.84%.

(102) For example, this work relies, amongst other, on estimates regarding customer savings from important 2012 merger and cartel decisions by the European Commission, on estimates regarding the reduction in prices resulting from competition policy enforcement decisions in the market concerned and also on estimates regarding the duration of the price reduction.

(103)  See Dierx A., Heikkonen J., Ilzkovitz F., Pataracchia B., Ratto M., Thum-Thysen A. and Varga J. (2015), "Distributional macroeconomic effects of EU competition policy – A general equilibrium analysis", paper to be published in a World Bank-OECD publication on Competition Policy, Shared Prosperity and Inclusive Growth.

(104)  Barone and Cingano demonstrate that productivity growth in manufacturing is harmed by regulations that reduce competition in services (especially financial services and energy provision). (Barone, G., & Cingano, F. (2008). "Service Regulation and Growth: Evidence from OECD Countries", Bank of Italy Temi di Discussione (Working Paper) No, 675).Studies carried out by Bourlès, R., Cette, G., Lopez, J., Mairesse, J., & Nicoletti, G. (2013). "Do product market regulations in upstream sectors curb productivity growth? Panel data evidence for OECD countries. Review of Economics and Statistics, 95(5), 1750-1768", and Forlani, E. (2010). "Competition in the Service Sector and the Performances of Manufacturing Firms: Does Liberalization Matter? CESifo Working Paper No. 2942", reaches similar results.

(105)  Gutmann, J., & Voigt, S. (2014). "Lending a Hand to the Invisible Hand? Assessing the Effects of Newly Enacted Competition Laws", (February 8, 2014).

(106)  Borrell, J. R., & Tolosa, M. (2008). "Endogenous antitrust: cross-country evidence on the impact of competition-enhancing policies on productivity". Applied Economics Letters, 15(11), 827-831. The authors found that the impact of antitrust enforcement on total factor productivity is positive and statistically significant, implying that competition policy effectiveness raises productivity. The estimates suggest that increasing the average antitrust effectiveness in one standard deviation would increase average total factor productivity by 28%. The study assesses the combined effect of both competition and trade policy, as examining competition law alone over-states its effect on productivity growth.

(107)  Aghion, P., Bloom, N., Blundell, R., Griffith, R., & Howitt, P. (2005). "Competition and Innovation: an Inverted-U Relationship". The Quarterly Journal of Economics, 120(2), 701-728. Other studies showing the relationship between competition and innovation are Polder, M. and Veldhuizen, E. (2010), Grünewald, O. (2009) or Carlin, W., Schaffer, M., & Seabright, P. (2004). On empirical work, see Aghion, Philippe, Stefan Bechtold, Lea Cassar and Holger Herz (2014), "The causal effects of competition on innovation: Experimental evidence", Harvard University Working Paper, February.

(108)  Ahn, S. (2002), "Competition, Innovation and Productivity Growth: A Review of Theory and Evidence". OECD Economics Working Paper No. 317, Gilbert, Richard J. (2007), "Competition and innovation", Competition Policy Centre, UC Berkeley.

(109)  Ahn, S. (2002), op. cit. See also for example, a study by the Directorate-General for Competition of the European Commission on "The Economic Impact of enforcement of competition policies in the functioning of EU energy markets" (2015), which found that the Commission's decision finding an abuse of dominance by E.ON lead to a reduction in prices for both wholesalers and retailers to the benefit of consumers. See also the Note by the UNCTAD Secretariat (2014) "The benefits of competition policy for consumers".

(110)  See Connor, J. M. & Bolotova, Y. (2006) "Cartel overcharges: survey and meta-analysis", International Journal of Industrial Organisation, 24(6), 1109-1137, Oxera and a multi-jurisdictional team of lawyers led by Dr Assimakis Komninos (2009) "Quantifying antitrust damages", (Study prepared for the European Commission) available at http://ec.europa.eu/competition/antitrust/actionsdamages/quantification_study.pdf , p. ix, and Smuda F. (2014), "Cartel Overcharges and the deterrent effect of EU competition law", Journal of Competition Law & Economics, Vol. 10(1), pp. 63-86. A study by Connor, J.M. (2014), "Price-fixing overcharges", found that the median average long-term overcharge for all types of cartels over all time periods is 23%.

(111)  Connor, J. M. & Bolotova, Y. (2006) "Cartel overcharges: survey and meta-analysis", International Journal of Industrial Organisation, 24(6), 1109-1137.

(112)   http://ec.europa.eu/atwork/synthesis/amp/doc/comp_sp_2016-2020_en.pdf .

(113)  See the Impact Assessment2015/2016 of the UK Competition and Markets Authority (https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/537539/cma-impact-assessment-2015-16.pdf) , which estimated the benefits of the Competition and Markets Authority’s work averaged over a 3-year period and the ratio of these benefits to costs. The above estimates do not include many other benefits stemming from competition law enforcement, such as better product quality and wider choice, the deterrent effect of the enforcement actions, or other effects of competition policy such as productivity gains or impact on jobs.

(114)  The UK competition authority explains that its impact estimations are conducted immediately after cases are completed and are therefore based only on information available during the case and on assumptions regarding the expected impact of its interventions. On this basis the estimates are considered to be "ex ante’ evaluations. The authority considers that in general, the assumptions it applies are cautious and hence the estimates are conservative.

(115)  "Does competition kill or create jobs?", OECD Global Forum on Competition, DAF/COMP/GF(2015)9.

(116)  See also Dierx A., Heikkonen J., Ilzkovitz F., Pataracchia B., Ratto M., Thum-Thysen A. and Varga J. (2015), "Distributional macroeconomic effects of EU competition policy – A general equilibrium analysis", paper to be published in a World Bank-OECD publication on Competition Policy, Shared Prosperity and Inclusive Growth, who estimate that enforcement of the EU competition rules by the European Commission has a sizeable impact on the creation of new jobs (they estimate around 650 000 after ten years).

(117)  "Does competition kill or create jobs?", OECD Global Forum on Competition, DAF/COMP/GF(2015)9, paragraph 78.

(118)  See Dierx A., Heikkonen J., Ilkovitz F., Pataracchia B., Ratto M., Thum-Thysen A. and Varga J. (2015), ibid, whose simulations show that competition policy has important redistributive effects. This supports the view that competition policy interventions, by lowering prices and increasing the quality and variety of products, are particularly beneficial for the poorest in the society.

(119)  Hausman, J. A., & Sidak, J. G. (2004). "Why Do the Poor and the Less-Educated Pay More for Long-Distance Calls?" Contributions in Economic Analysis & Policy, 3(1).

(120)  Urzúa, C. M. (2013), "Distributive and regional effects of monopoly power", Economia Mexicana NUEVA EPOCA, 22(2), 279-295.

(121)  "Market power and wealth distribution", John Davies and Sean Ennis, OECD, presentation at the conference on "Looking beyond the direct effects of the work of competition authorities: deterrence and macroeconomic impact", Directorate-General for Competition of the European Commission, Brussels, 17/18 September 2015.

(122)  For example, some NCAs cannot issue binding and enforceable requests for information to get the data they need.

(123)  See for example, Article 3 of Directive 2002/21 of the European Parliament and of the Council of 7 March 2002 on a common regulatory framework for electronic communications networks and services, OJ 2002, L 108, p.33 as amended by Directive 2009/140 of the European and of the Council of 25 November 2009, OJ 2009, L 337, p.37; Article 35 of Directive 2009/72 of the European Parliament and of the Council of 13 July 2009 concerning common rules for the internal market in electricity and repealing Directive 2003/54, OJ 2009, L 211, p.55; and Article 55 of Directive 2012/34 of the European Parliament and of the Council of 21 November 2012 establishing a single European railway area, OJ 2012, L 343, p.32. Recently, detailed independence safeguards have been enacted in the new General Data Protection Regulation, see Articles 51 - 54 of Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC, OJ 1995, L 281, p.31. The Commission also proposes to introduce a legal EU requirement of independence of audio-visual regulatory bodies, see Commission proposal for a Directive of the European Parliament and of the Council amending Directive 2010/13/EU on the coordination of certain provisions laid down by law, regulation or administrative action in Member States concerning the provision of audio-visual media services in view of changing market realities, COM(2016) 287 final.

(124)  8% of respondents (strongly) disagreed, 10% had a neutral opinion, and 2% indicated that they did not know/not applicable. Per group of stakeholders, the (strongly) disagree option was supported by higher percentages in the case of companies/SMEs (11%, but there was still more than 77% support) and industry associations (28%, but still 61% support).

(125)  At the level of the individual groups of stakeholders, a lower level of support come mainly from non-governmental organisations and industry associations, which nevertheless still gave a total support to EU legislative action (either alone or combined with soft action) of around 50%. Regarding the rest of the stakeholders, the support was in all cases above 70-75%, although with some differences regarding the preference between exclusively legislative action and legislative combined with soft action. For example, NCAs preferred exclusive EU legislative action (56%), with 40% favouring legislative combined with soft action), and Ministries preferred EU legislative action combined with soft action (50%) over exclusive EU legislative action (25%).

(126)  Case C-557/12, Kone AG v. ÖBB-Infrastruktur AG, EU:C:2014:1317, at para. 32.

(127)  Case C-439/08, Vlaamse federatie van verenigingen van Brood- en Banketbakkers, Ijsbereiders en Chocoladebewerkers (VEBIC) WZW, EU:C.2010:739, at paras 56 and 57.

(128)  As already indicated in section 6.3, these results were in general similar also per type of stakeholder, with lower levels of support coming normally only from non-governmental organisations and industry associations. However, even for these two groups of stakeholders, the most frequent alternative replies were that the effects would be neutral or that they had no-opinion, (very) negative effects being supported generally by less than 7%.

EUROPEAN COMMISSION

Brussels, 22.3.2017

SWD(2017) 114 final

COMMISSION STAFF WORKING DOCUMENT

IMPACT ASSESSMENT

Annexes

Accompanying the document

Proposal for a

DIRECTIVE OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL

to empower the competition authorities of the Member States to be more effective enforcers and to ensure the proper functioning of the internal market

List of Annexes

{COM(2017) 142 final}
{SWD(2017) 115 final}
{SWD(2017) 116 final}

List of Annexes

Annex I - Procedural information

Annex II - Stakeholder consultation

Annex III - Who is affected by the initiative and how

Annex IV - Problem tree

Annex V - Replies of stakeholders on the tools NCAs need to effectively enforce

Annex VI - Issues related to legal maximum, fines calculations and who can be fined

Annex VII - Results of hypothetical cartel case for a duration of 3.75 years

Annex VIII - Article in German press

Annex IX - Imposition of sanctions in civil and criminal procedures

Annex X - Policy options

Annex XI - Achievement of objectives

Annex XII - Number of leniency and summary leniency applications submitted before NCAs

Annex XIII – Divergences between leniency programmes

Annex XIV - Budget and staff of NCAs

Annex XV - Core indicators

Annex XVI – Costs/benefits analysis of the preferred option

Annex XVII – Glossary of terms

Annex I - Procedural information

1. Initiative key information

Leading DG: DG Competition

Agenda planning reference: 2017/COMP/001

Initiative title: Legislative proposal - Enhancing competition in the EU for the benefit of businesses and consumers – Reinforcement of the application of EU competition law by national competition authorities

Expected adoption of legislative proposal: 1st quarter 2017

2. Reports on the functioning of Council Regulation 1/2003

During 2013 and 2014 DG Competition conducted an assessment of the functioning of Council Regulation 1/2003.

As part of this new assessment, DG Competition examined a range of areas that either were not addressed by Regulation 1/2003, were addressed in a general way but a need for a detailed response has subsequently arisen in practice, or have emerged as new issues.

Based on the results of this assessment, the Commission adopted in July 2014 the "Communication from the Commission - Ten Years of Antitrust Enforcement under Regulation 1/2003: Achievements and Future Perspectives" 1 ("the Communication"). The Communication concluded that the enforcement of the EU competition rules had considerably increased as a result of the achievements of the Commission, the ECN and the national competition authorities (NCAs). The guidance provided by the Commission to stakeholders, NCAs and national courts and the cooperation within the ECN had favoured the coherent application of the EU competition rules throughout the EU and boosted the enforcement of EU competition rules. However, the Communication also concluded that it was important to build on these achievements to create a truly common competition enforcement area in the EU, in particular by:

- further guaranteeing the independence of NCAs in the exercise of their tasks and that they have sufficient resources;

- ensuring that NCAs have a complete set of effective investigative and decision making powers at their disposal;

- ensuring that powers to impose effective and proportionate fines are in place in all Member States;

- ensuring that well-designed leniency programmes are in place in all Member States and consider measures to avoid disincentives for corporate leniency applicants.

This exercise built on the previous "Report on the functioning of Regulation 1/2003" 2 of 2009 which had found that the new competition enforcement system had positively contributed to a stronger enforcement of the EU competition rules, but that some aspects merited further evaluation, in particular, with respect to making NCAs' enforcement tools and fining powers more effective.

3. Evidence used to support the Impact Assessment

By way of follow-up to the Communication, extensive fact-finding has been carried out by DG Competition in cooperation with NCAs on all the objectives identified by the Communication in order to have a detailed picture of the status quo.

In addition to the Communication on ten years of Regulation 1/2003 and the Report on five years of Regulation 1/2003, noted above, the fact-finding built on the following Reports:

- Investigative Powers Report (31 October 2012), which provides an overview of investigative procedures within the ECN. 3  

- Decision-Making Powers Report (31 October 2012), which provides an overview of decision-making powers within the ECN. 4  

- Report on the Assessment of the State of Convergence with the ECN Model Leniency Programme (15 October 2009). 5

- Several publications on the impact of competition, such as the OECD report "Fact-sheet on how competition policy affects macro-economic outcomes" (October 2014)

Fact-finding within the ECN

For the preparation of this Impact Assessment, DG Competition relied on fact-finding carried out by three ECN working groups: the Working Group on Cooperation Issues and Due Process, the Cartels Working Group and the ad-hoc Working Group on Fines and Related Issues. Detailed questionnaires have been sent on the different issues raised in the Communication.

Moreover, the fact-finding done at the level of the respective working groups was further discussed and reviewed at a higher level in the context of the ECN Plenary meetings, and finally at the highest level during the ECN Directors-General meetings, attended by the Heads of the NCAs and DG Competition.

On the basis of all the information gathered, the Commission has decided to carry out an Impact Assessment in order to define in more detail the scope of the identified problems and the objectives to be achieved and assess the different policy options to address them.

4. Organisation and timing

4.1. Inter-Service Steering Group

An Inter-Service Steering Group (ISSG) was set up in October 2015. In total four meetings have been held on the following dates: 22 October 2015, 19 May 2016, 23 June 2016 and 14 July 2016.

The following Directorates-General and services participated: BUDG, CNECT, ECFIN, ENER, ENV, FISMA, GROW, JUST, MARE, MOVE, OLAF, TRADE, LS and SG.

The feedback received from these Directorates-General and services has been taken into account in the draft Impact Assessment Report.

The ISSG approved the Inception Impact Assessment that was published in November 2015 and the draft Impact Assessment Report.

The minutes of the meeting held on 14 July 2016 were submitted to the Regulatory Scrutiny Board.

4.2. Consultation of the Regulatory Scrutiny Board

This draft Impact Assessment Report was submitted to the Regulatory Scrutiny Board on 31 August 2016 and a meeting took place on 28 September 2016.

The table below provides an overview on, and, where necessary, brief explanations about, the changes introduced in the revised draft Impact Assessment Report after the meeting on 28 September 2016 and the main recommendations for improvements of the Regulatory Scrutiny Board.

The Regulatory Scrutiny Board issued a subsequent opinion on 9 December 2016, giving a positive opinion, with a recommendation to improve the following aspects:

The Regulatory Scrutiny Board also recommended the addition of a glossary of acronyms, issues and expressions used. A glossary of terms has been added in Annex XVII.

Annex II - Stakeholder consultation

I.    Report on the Contributions to the Public Consultation on Empowering the national competition authorities to be more effective enforcers of the EU competition rules

Introduction

A public consultation 6 on empowering the national competition authorities to be more effective enforcers of the EU competition rules was launched on 4 November 2015 and ran until 12 February 2016.

The public consultation follows up the Commission's Communication on Ten Years of Regulation 1/2003 7 , which identified a number of areas of action to boost the powers of national competition authorities ("NCAs") to enforce the EU competition rules. The objective of the public consultation was to get feedback from a broad range of stakeholders on their experience/knowledge of issues that NCAs may face having an impact on their ability to effectively enforce the EU competition rules and what action, if any, should be taken in this regard.

The public consultation followed the Commission's minimum standards and has taken the form of an EU Survey which was split into two parts, a first one with general questions seeking input from non-specialised stakeholders, and a second one for stakeholders with a deeper knowledge/experience of competition matters. This second part addresses four key issues:

A.resources and independence of the national competition authorities;

B.enforcement toolbox of the national competition authorities;

C.powers of national competition authorities to fine undertakings; and

D.leniency programmes.

The public consultation page and the general questions were available in the following EU official languages: bg   cs   da   de   el   en   es   et   fi   fr   hr   hu   it   lt   lv   mt   nl   pl   pt   ro   sk   sl   sv   to encourage input by consumers and SMEs and to allow the public at large to contribute. The detailed sections of the open public consultation questionnaire were only available in English but answers could be provided in all EU official languages.

Validity of the public consultation: assessment of its weaknesses and strengths

The main weakness of the questionnaire, which has been raised by some stakeholders, is that it is rather long. This, together with the inherent complexity of the issues it tackles, might have dissuaded some stakeholders from replying. To address this issue, the questionnaire contains a shorter section with general questions aimed at all (including non-specialised) stakeholders.

Another weakness is that it was not possible to translate the entire questionnaire in all official languages. To encourage wide participation in the public consultation the introductory sections and the section with the general questions, which covered the essence of the main issues covered by the questionnaire, were translated into all official languages. Over 40 participants opted to exclusively use this option.

The public consultation had however several features that counterbalanced, at least partially, the weaknesses referred to above.

First, although the public consultation has been officially open for participation for about 12 weeks, in practice stakeholders could provide input for around 16 weeks. This has allowed stakeholders willing to participate ample time to do so.

Moreover, respondents had for almost every question the possibility to add additional comments clarifying or expanding their replies and to attach supporting documents. Replies in the form of a position paper as opposed to through the questionnaire were also accepted.

Finally, in order to promote participation as much as possible, we encouraged NCAs to bring the the public consultation to the attention of their respective national consumer and business associations. This was complemented by initiatives by the Commission to promote awareness of the public consultation by reaching out to organisations with a pan-European dimension such as the European consumer organisation BEUC and Business Europe, as well as through participating in conferences at national level.

Summary of the general questions

There have been 181 replies from a wide variety of stakeholders, ranging from private individuals, law firms and consultancies, companies and industry associations, consumer organisations, academics, non-governmental organisations, think tanks and trade unions to public authorities, including a number of Ministries and NCAs, from within and outside the EU. This is a very good response rate for a public consultation in the competition field.

The results show that for the majority of respondents NCAs are effectively enforcing EU competition rules. There is however a 31% of respondents that considers that this is not the case.

Are EU competition rules effectively enforced by NCAs?

However, a wide majority of respondents consider that NCAs could do more to enforce EU competition rules than they currently do:

Could NCAs do more to enforce EU competition rules?

Respondents also consider that the following measures would help NCAs to be more effective:

Other issues raised by stakeholders

A majority of stakeholders (59%) also consider that other actions should be taken to boost the effectiveness of the NCAs. There is in particular a consistent demand from lawyers, business and business organisations that any enhancement of NCAs' enforcement powers is counter-balanced by increased procedural guarantees, including ensuring that rights of defence can be effectively exercised by having greater transparency of investigations and effective judicial review (e.g. companies should receive a Statement of Objections and have effective rights of access to file).

Other issues raised are the request of greater coherency within the ECN in the application of the EU competition rules, the recognition of Legal Professional Privilege (LPP) for in-house lawyers and of compliance programmes as a mitigating factor for fines, that NCAs should be able to defend their cases in court, a more consistent application of the effect on trade criterion or the abolition of the power of NCAs to apply stricter rules on unilateral conduct.

The questionnaire has also sought views from stakeholders on whether action to boost enforcement by NCAS should be taken and, if so, who should take action. The graphs below show the results which indicate that a wide majority of stakeholders supports that action should be taken and that such action should preferably be a combination of EU and Members States action. 9

Should action be taken? Who should take action?

Per group of stakeholder, the replies are as follows:

Should action be taken?

Who should take action?

Respondents also consider that, if EU action were to be taken, it should preferably take the form of a mix of legislative and non-legislative action.

What type of EU action is most appropriate?

Per group of stakeholder, the replies are as follows:

What type of EU action is most appropriate?

Finally, the overall view of stakeholders is that taking action at EU level would have a (very) positive impact on various aspects, as shown in the table below:

Impact of EU action

Summary of results of the detailed questions

A. Resources and independence of the national competition authorities

A wide majority of stakeholders agree with the findings of the Commission's Communication on Ten Years of Regulation 1/2003 that it is necessary to further guarantee the independence of NCAs and that they have sufficient resources when enforcing the EU competition rules.

Per group of stakeholder, the replies are as follows:

Is it necessary to further guarantee the independence of NCAs and that they have sufficient resources when enforcing the EU competition rules?

Many respondents consider that the following measures are needed to ensure NCAs' independence when they enforce the EU competition rules (stakeholders were asked to identify and rank the three measures they considered to be of most importance): 

(*)e.g. budgetary autonomy and transparent appointment procedures for NCAs' management

The majority of stakeholders prefer action to be taken at both EU and national level on resources (59%) and on independence (54%).

In terms of those who consider that EU action is appropriate, approximately 43% consider that a mixture of legislative and soft action is the best solution.

Per group of stakeholder, the replies are as follows:

Who should take action?

Ind = Independence and **:Res = Resources

What type of EU action is most appropriate?

B. Enforcement toolbox of the national competition authorities

A lack of effective powers for NCAs is considered by stakeholders to be a problem, firstly, in terms of the effective enforcement of the EU competition rules (e.g. NCAs may refrain from taking action/carry out more limited action/take action which does not meet the desired objective), and secondly, for cooperation within the ECN (e.g. it can impinge on the ability of NCAs to carry out inspections etc. on each other's behalf under Article 22 of Regulation 1/2003).

Divergences in NCAs' powers is seen as a problem in terms of legal certainty for business (63%), costs for business (62%) and for cooperation in the ECN, e.g. different rules on what evidence can be gathered on behalf of another NCA (57%).

The table below shows the investigation and decision-making tools stakeholders think that NCAs need to have in order for them to be effective enforcers of the EU competition rules.

A majority of stakeholders consider that ensuring that the NCAs have an effective toolbox should be addressed by a combination of EU and national action.

Those who consider EU action appropriate prefer a mixture of legislative and non-legislative action (48%), with a smaller number opting for exclusive legislative action (41%).

Per group of stakeholder, the replies are as follows:

Who should take action?

What type of EU action is most appropriate?

C. Powers of national competition authorities to fine undertakings

The public consultation has covered three main issues: the nature of the fines imposed by NCAs (criminal, civil or administrative); who can be fined (concept of undertaking, parental liability and succession); and fines methodologies/legal maximum of the fines.

The graphs below show to what extent stakeholders considered that there are problems in the three areas identified:

Is it a problem that some NCAs impose only/primarily criminal / civil fines?

       Criminal fines                    Civil fines