Compliance is the outcome of interactions between firms and competition authorities. For a competition authority to design compliance policy instruments requires understanding the origins of harmful conduct so that firms’ incentives could be changed. This includes adjusting instruments in terms of information gathering, rules and procedures and sanctions. It also needs addressing agency issues and adapting to the economic context. These points are illustrated with examples derived from the French Autorité practice in the field.

By Marianne Faessel & Henri Piffaut[1]


What is the purpose of competition policy? An oft misconception is to confuse the end with the means. To find an infringement of competition rules and impose fines is not an end in itself, although it has a clear impact. In Europe at least, fines do not bring reparation. They are set in such a way that any company about to adopt a possibly illegal behavior would internalize the risk of being fined and rationally opt for a non-infringing behavior (i.e. deterrence). The potential decision and fine are the means to adjust the incentives of companies when they decide on their behavior on the market. To put it differently they aim at ensuring compliance with competition law. The end is a functioning of the market that maximizes consumer welfare.[2]

Most agencies have developed compliance policies. By so doing they use various tools to create incentives for firms to comply with the rules: when deciding on adopting a given behaviour, based on the information available, a firm balances the expected benefits with the expected costs. In turn, many firms have, as part of wider compliance programs, developed one dedicated to competition rules.[3] Compliance has a meaning in various fields such as medicine, psychology, physics or law. In general legal terms, compliance means conforming to a rule, such as a specification, policy, standard or law. Regulatory compliance describes the goal that organizations aspire to achieve in their efforts to ensure that they are aware of and take steps to comply with relevant laws, policies, and regulations. Hence compliance has two dimensions: how competition agencies seek to influence the incentives for firms to enter into anticompetitive conducts and how firms internalize these incentives. As in medicine, compliance is not just a policy objective for a competition agency or just an internal effort by firms, it relies on interactions between agency and firms.

The need for competition policy to address compliance is exemplified by two challenges that competition agencies currently face. First, they are criticized for having failed to develop a system that would tame the digital giants: they seem immune to existing instruments. Second, there appears to be an information gap where categories of companies do not seem to be sufficiently aware of competition law and possible liabilities, for instance SMEs and professions. That raises the question of how to best design competition instruments to ensure compliance, communicate about them and how to help firms and industries to develop their own compliance programs.


Designing a proper regulatory setting to address a given issue depends on the decision making of a company before it adopts a given behavior, the credibility and ability (in terms of resources and powers) of a regulator to identify, investigate and sanction that behavior, and the ability and incentives of third parties to provide information to the regulator.

To internalize the costs and benefits of a given behavior under a principle-based regime, a firm must (i) be aware that it might undertake the conduct (i.e. to have in place some control mechanism, systems and data flows, over agents decisions); (ii) understand what rules may be infringed; (iii) evaluate the likelihood that an infringement may be found by a competition authority; and (iv) anticipate the consequences (cost, changes to behavior) it may incur as a result of the finding of an infringement. Such a finding has a probabilistic nature and its realization depends on the instruments that a competition authority has developed. A competition authority has to be able to (v) provide and inform on rules and actions; (vi) identify the behaviors and focus on the most harmful; and (vii) prove that a given behavior infringes rules and adopt a finding with sanctions (fines and/or remedies) which may or may not be sustained if appealed. Finally, there must be some (viii) redress mechanism so that harm that has been inflicted be corrected.

To influence companies’ incentives, competition policy relies on information, rules and procedures, and enforcement powers. Competition authorities need to develop instruments that impact all of these eight steps.

When designing a compliance policy, a competition authority faces three obstacles: a double information asymmetry (how clear are rules to firms? And how detectable are bad behaviors?), a credibility issue (how to ensure that firms perceive a high enough probability for the authority to identify and sanction without support from market participants?), and how to balance obligations and principles to comply with and how to set the possible rewards and sanctions so that possible agency issues are minimized.

The information asymmetry aspects underline the importance for competition authorities to communicate on their instruments and actions. Often this may require that communication is adapted to the various targets: large firms and small firms do not have the same access to information and internal skills to anticipate issues. For instance, the Autorité has developed dedicated tools to inform SMEs and professions of their obligations under competition rules. The other type of information asymmetry touches upon the ability to gather information for competition authorities. That requires investigative powers, resources and incentives for economic agents to come up with information.



Competition authorities must have and show an ability to uncover and sanction bad behaviors ex officio. However, many cases would come too late to their attention, after damages have been done. They have therefore developed policies that provide incentives for individuals, parties or third parties to come up with information on potentially unlawful conducts. In order to obtain information on potentially unlawful conducts, competition agencies typically aim at providing gains to informants that go beyond their potential costs. This is the reasoning behind leniency policies. Also, some agencies have started programs where they pay individual informants. For instance, in the U.S., the Commodity Futures Trading Commission has awarded almost $200m to a former Deutsche Bank employee who raised concerns about the manipulation of the Libor interest rate benchmark.[4] In France, the Autorité has developed such leniency program,[5] and updated its fining policy.[6] As for third parties, tools include the possibility to claim damages but often protection against retaliation is key. Ensuring confidentiality is a prerequisite. In other instances, authorities rely on an obligation to inform. For instance, mergers have to be notified with extensive information on the markets concerned or in France, supermarket chains have to inform on joint purchasing structures.[7]

But that relies on the authority credibility in detecting, on its own, possible infringements and enforcing its rules. The credibility of competition authorities depends on various parameters. That means having investigative powers that allow to gather information. That means using these powers and in particular, developing an ability to make ex officio cases so that there is an incentive for firms to provide initial intelligence and later to cooperate with the authority. And that requires resources to analyze and draw findings out of the information gathered: competition agencies must have the infrastructure and staff necessary to implement their policies. Finally, that means communicating on their rules and actions. In that regard, the Autorité is pursuing efforts to develop tools to detect on its own possible infringements.



Although they bind the agency, competition policy instruments range from soft (not binding) to hard (binding) with many steps in between them. Guidelines bind the competition authority, regulations bind both agencies and companies, commitment decisions bind both the agency and the parties; and infringement decisions, which may include some forward-looking obligations, bind the parties. Regulations, however, apply to a set of companies, targeted by the regulation that matches its scope and detail behavior that would raise concerns as opposed to those that would not and those where further investigation would be required.

The timing aspect of competition instruments is much more complex than an ex anteex post dichotomy. Think of cartels. The rule is clear, competitors shall not agree on prices or on quantities. It has been confirmed in numerous infringement decisions and court rulings. The rule acts as an ex ante principle, and the infringement procedure is no different to what would happen were a rule included in an ex ante regulation be alleged to be infringed. A firm can internalize competition rules in its decision making only if rules, procedures and sanctions are known in advance. It is perhaps more relevant to think of the timing of setting out the rules and possible sanctions when the rules are infringed, so that companies can decide on their future behavior. All instruments are by nature forward looking (or in the case of infringement decisions have a forward-looking component by setting precedent) since they (also) aim at influencing future behavior of companies. A block exemption sets out for the future which behaviors are permissible, should be avoided or are in a grey zone that would require a case specific investigation. It brings some level of legal certainty prior to adoption of a behavior but there is always the possibility of a finding of infringement ex post facto. Although the regulator has no ex ante knowledge of the behavior. Similarly, guidelines set out principles in advance, notably on how an assessment may be conducted and with them some legal certainty without limiting the possibility of finding infringements. Even an infringement decision provides an assessment mostly based on the facts existing at the time of the action that is being under review, not at the time of the adoption of the decision. This is because the aim is really to have competition enforcement risk internalized ex ante by companies. Clearly an infringement decision will also aim at remedying the specific competitive harm caused by the behavior at stake. So, this second parameter should be seen in terms of degree of guidance.

Lastly, the possible costs of finding an infringement matter for firms. They include possible remedies, fines and damage actions. Competition authorities aim at making them predictable for firms.

In general principle-based regulation (such as the principles set out in Articles 101 and 102 accompanied by Regulation 1/2003) is preferred over ones with some ex ante obligations. This is because the costs of ex ante obligations are much higher. They require some standardization of the behavior of companies and their interactions with the regulator. In order to regulate specific business decisions, regulators generally require information in a pre-set format and regular interactions with the company’s staff. That may create barriers for smaller companies and regulatory capture. They may decrease the incentives to innovate because of the cost of the regulatory review, the lower competition threat, and a reluctance for the standard that governs the interactions to have to change. These drawbacks do not exhibit the same intensity for all regulations. They are probably the most intense for industry specific regulation (like for telecoms) and less so for regulation that applies to some infrequent business decisions, such as control over foreign investment for instance.

However, in some instances, relying only on principle-based regulation is inefficient. Merger control is a good example. Absent an authorization-based merger control regime, mergers with harmful effects may still, after their implementation, be found to harm competition. However, there are two structural issues that may make such a system suboptimal. First, competition agencies may have an identification problem: how do they know that a merger took place and, in addition, may have caused competition harm? Second, even assuming that a competition agency was aware of the merger and has found competition harm but post completion of the transaction, how can it remedy it? That is the famous problem of “unscrambling the eggs” and more generally how to bring an industry to reach another equilibrium. That led to a notification and authorization-based regime where mergers cannot be implemented before a green light has been granted. Obviously, that relies on the pre-selection of mergers that are the best candidates for such a regime and let the other ones go through. In some jurisdictions there is still a possibility to challenge mergers after they have been consummated. Experience has shown that some mergers not caught by turnover thresholds, notably involving digital platforms, could raise significant competition issues that would be hard to remedy ex post. This has led to the revival of Article 22 of the merger regulation, championed by the Autorité, and information provisions on acquisitions in the Commission’s draft Digital Markets Act. Ultimately, this should lead firms, including the tech giants, to internalize possible merger control over such transactions.

Another example relates to the telecom markets where sector specific regulation has been put in place that relies on information provided by telecom operators and imposes numerous conditions on access and pricing. Such sector specific regulation addresses the inability of generic competition rules to design and monitor access markets that allow competition to develop. This is in part due to identification problems mentioned above but also to the working of such industries that are affected by network effects (how to undo competition harm). That is also part of the challenge that large digital platforms raise: their complexity raise identification issues and the working of network effects may make undoing the harm of a past behavior nearly impossible. The current discussions on a Digital Markets Act try to address these points.

Regulations however generate costs. Enforcement requires resources and manpower both by companies and enforcers. In addition, they have distortive effects: they increase the cost of doing business, distort the incentives to invest and innovate and create uncertainty. Regulation does not operate under perfect information with infinite resources either from the standpoint of regulators of that of companies. That leads to error costs issues. Companies may adopt over-cautious or over-risky conducts. Regulators may produce badly designed rules that create false positive and false negative situations. These costs are particularly relevant for platforms as they may lead, through the working of network effects, to unfairly either strengthening or weakening platforms. For that reason, authorities try to make information gathering, rules and procedures as proportionate as possible to the objective of maintaining competition.



On the one hand competition authorities cannot directly measure the success of their policies: they can measure their enforcement success but it does not necessarily reflect their success in altering firms’ incentives. The number of adopted decisions or the amount of fines imposed are not necessarily informative on the effectiveness of competition policy. In a way one could even argue that the more there are infringements the more it would show that they failed in changing compliance incentives. The reasons for that can vary: lack of clarity of when infringement can occur, too low level of sanctions, lack of market monitoring, inefficient procedural rules, etc. Competition agencies such as the Autorité constantly work on improving these aspects, stating policy objectives, trying to develop ex post assessment tools, and other actions either through their own efforts or thanks to feedback from stakeholders.

On the other hand, firms themselves face significant agency issues. To engage into potentially anticompetitive conduct or infringe procedural rules is rarely the outcome of corporate level decisions. Most often such actions result from manager decisions. Managers may see benefits from higher profits from anticompetitive conducts (higher bonuses) but not necessarily the costs. Firms may not be able to identify anticompetitive conducts implemented by their managers and would have to bear the consequences of possible sanctions. They are perhaps most vivid with digital platforms where their business relies on regulating environments for users to interact while at the same time maximizing profits through their own interactions with users. That is a general challenge for firms and that has led to governance and compliance policies that try to minimize the extent of the agency issue. Their objective is twofold: to change managers’ incentive to enter into anticompetitive conducts and to identify and report ongoing anticompetitive conducts. Many leniency applications have been triggered by due diligence reviews at times of changes in internal governance either because of merger and acquisitions, governance demands from shareholders or simply changes in personnel.

Competition authorities try to solve this firm’s agency issue by creating incentives for firms to adopt and implement compliance policies and by directly altering managers’ incentives. The following describes the Autorité approach in that regard.



Some competition agencies have chosen to create a link between the existence of a compliance policy and possible fines: either such policy would lead to lower fines or the absence of such policy would increase possible fines. In France, the Autorité adopted back in 2012 a policy[8] of reducing penalties for companies that set up competition compliance programs. The objective of that policy was to encourage companies to put in place structural and sustainable tools at a time when compliance programs were not widely deployed. The Autorité issued its 2012 so-called “Framework Document on competition compliance programs” after a broad public consultation. It not only gave guidance on key points for an effective compliance program. It also provided that a company could obtain an up to 10 percent reduction in their fine if it did not challenge the objections raised against it and committed to implement a compliance program for the future.

When the Autorité adopted a new settlement procedure in 2017, it decided to revise its policy regarding compliance programs by withdrawing the 2012 Framework Document.[9] This policy had drawbacks. It created an incentive not to implement a compliance policy before a finding of infringement. Firms with a compliance program would be fined more than those without, ceteris paribus. There was therefore a perverse effect in rewarding companies that had not made an effort in this area. In the press release that was then published to inform the public of this change, the Autorité emphasized the fact that companies have had time, over the last years, to internalize competition law and implement compliance programs. It therefore decided that commitments offered by companies to set up such compliance programs could no longer justify a reduction in the penalties incurred for competition law infringements.

From that time on, the Autorité implemented a new approach to compliance by actively disseminating practical tools and guidance for stakeholders such as papers, in-depth studies and guides that explain the Autorité’s decision-making practice in a detailed and pedagogical manner. These resources, available on the Autorité’s website, cover a large range of topics, including: gun jumping (2018), loyalty rebates (2018), algorithms and competition (2019), competition for SMEs (2020), competition policy and the digital economy (2020), behavioural commitments (2020), online commerce and competition (2020), and professional bodies (2021). The Autorité also completed a major overhaul of its mergers guidelines in 2020. These contributions show significant efforts to reach out to firms of all kinds.

In an effort to further respond to the needs of stakeholders, the Autorité decided to join a working group, formed in 2020 at the initiative of in-house counsels, attorneys, and members of trade associations with a view to discuss and identify good compliance practices. One important finding of the group was that companies felt the need to have further guidance on how to implement compliance programs. This led the Autorité to draft a new Framework Document on competition compliance programs. It launched a public consultation for a two-month period on October 11, 2021. The final text of the new guidance is expected to be published early 2022.

With this document, the Autorité wishes to contribute even more to developing a culture of compliance among all firms, whatever their size and industries. The new document points out that there is no one size fits all program and that a successful program must be based on five elements: (i) a public commitment of the company to support the compliance program; (ii) the appointment of persons responsible for implementing the compliance program; (iii) ensuring information, training and awareness at all levels of the company; (iv) the implementation of monitoring and warning mechanisms; and (v) the implementation of a system to monitor the treatment of alerts.

In parallel to this new guidance, and the continuing development of soft law via the release of regular publications, the Autorité is also keen to promote compliance at a larger level through a more and more interactive communication notably via using new tools on its website. These tools are very effective to help disseminate a culture of compliance in a world where most people run after time. The Autorité also continually seeks to improve its traditional means of communication. For instance, its press releases now often include compliance text boxes that summarize lessons learned from the case at stake.

Designing compliance policy instruments requires understanding the origins of harmful conduct so that incentives could be changed. This includes addressing agency issues and adapting to the economic context. The aim is to dynamically adjust such instruments to ensure maximum compliance at minimum costs.

1 Marianne Faessel is an Adviser to the President of the Autorité de la concurrence. Henri Piffaut is a vice president at the Autorité de la concurrence. The views expressed in this article are those of the authors and do not represent those of the French Autorité de la concurrence.

2 One may object that European competition law is not just about consumer welfare and concerns the competitive process and market integration as well. That does not change the argument even though the rules to internalize for companies may be slightly more complex.

3 OECD (2021), Competition Compliance Programmes, OECD Competition Committee Discussion Paper.

4 Financial Times, October 21, 2021.

5 Its latest communication on leniency dates back to April 3, 2015.

6 It adopted a new procedural notice on the method for determining fines on July 30, 2021.

7 Law of August 6, 2015 pour la croissance, l’activité et l’égalité des chances économiques.

8 Framework Document of February 10, 2012 on Competition Compliance Programmes.

9 Communiqué of October 19, 2017 on the settlement procedure and compliance programmes.