Western Balkans and the Design of Effective Competition Law: The Role of Economic, Institutional and Cultural Characteristics

In recent years, a vast number of countries worldwide have adopted competition laws, including antitrust provisions and merger control. This phenomenon is linked to the belief that competition is beneficial for economic welfare, and that competition law is capable of protecting competition in the marketplace. Literature shows that the effectiveness of competition law in a given country may depend on the economic, institutional and cultural characteristics of the country itself. This paper contributes to this literature by assessing what should be the optimal design of competition law in Western Balkan countries. First, we identify the “policy choice set”, the essential variables that define a competition policy regime. Second, we review the economic, institutional and cultural characteristics that should shape the optimal policy design. Finally, we examine the Western Balkan economies to describe where they stand, in terms of these characteristics, and derive policy implications on how their competition policy should be designed and implemented. Several conclusions emerge from our analysis. The existence of high barriers to entry and poor institutional quality points to the importance of an institutional set-up where the independence and transparency of the competition authorities is maximised within the context of an administrative model; also, no sector or enterprise, including SOEs, should be excluded from competition law enforcement, and competition law provisions should ensure that the voice of the competition authority is heard whenever new legislation that potentially affects competition is introduced. In terms of enforcement, while the role of advocacy emerges as crucial, along with the prosecution of entry-foreclosing abuses, a more lenient approach to merger control can be suggested, in the form of high notification thresholds.


Introduction
In 2001 the International Competition Network was created by the competition authorities of 14 jurisdictions, and today it has 138 members. This rapid growth was due to the introduction of competition law in many countries that previously lacked one. The adoption of anti-monopoly regulation was spurred by the transition from a planned to a market economy, in some areas, and by the general belief that competition could be one of the main drivers of better economic performance. Indeed, there is extensive economic literature that shows that competition can foster productivity growth. 1 More focused literature deals with the relationship between competition policy and economic performance. 2 These contributions indicate that competition policy does play a significant role. However, they also point out that the mere existence of rules intended to protect competition does not suffice to generate the results aimed at. What is needed is a "good" competition policy. So, the policy issue becomes what features a competition policy regime should have in order to effectively pursue its intended goals. Existing literature also shows that the effectiveness of competition policy depends on other characteristics of the given country. 3 Thus, the previous policy question cannot be addressed without considering the wider context in which the anti-trust regime is set. The general normative statement is that the desirable features of a "good" competition policy regime can only be identified taking into consideration the economic, social, cultural, and institutional characteristics of the specific country. This statement needs to be further detailed and limited but growing literature provides attempts to refine the general recommendation. 4 This paper aims at contributing to this research agenda.
In order to establish the desirable characteristics of a competition policy regime, given the relevant specific characteristics of a country, one has to perform two operations. First, one must identify the "variables" that warrant a policy decision: we may call it the choice set. It can be thought of as the menu that lists the many models of competition policy regimes from which the decision-maker has to choose. These regimes vary along various dimensions concerning the substantive rules, the institutions entrusted with their enforcement, and the way this enforcement is conducted. Different models can be built by combining these elements in different ways.
The second task is to identify the exogenous characteristics that influence the ability of the previously identified models of competition policy regime to achieve its objectives. These can be thought of as fixed factors that are likely to alter the performance of the regimes in the choice set and are linked to some crucial economic characteristics of the country, its institutions and cultural factors. Once these factors are identified, one has to ascertain which policy model is more likely to perform better, i.e. to better promote and protect competition.
We conduct these analyses in the next two sections of the paper. In Sect. 2 we describe the choice set, i.e. the main elements of a competition policy regime and the alternatives that are available to the policy maker. Section 3 examines the economic, institutional and cultural factors that need to be considered when selecting the model that is predisposed to perform better. In doing so, we focus in particular on those factors that are more likely to characterise emerging economies. This section has normative content and provides some suggestions on what we believe are the best policy choices under the described conditions. Section 4 presents some indicators reflecting the existing economic, institution and cultural characteristics of the countries in the Western Balkan region. 5 The purpose of this section is to show that the previous recommendations are indeed relevant for the design of the competition policy regime in these countries. The analysis looks at central and south-eastern European countries belonging to the EU as comparators, as well as at some more advanced jurisdictions, to give a sense of the magnitude of the existing gap for the relevant identified characteristics.
Section 5 concludes and suggests initiatives that can be undertaken to improve the effectiveness of competition policy in the West Balkan region.

Models of Competition Policy Regime
A competition policy "regime" can be thought as a combination of characteristics that concern the content of the prohibitions or prescriptions set in the law, the institutions entrusted with their enforcement and the way they perform their task. Many combinations exist and competition policy regimes vary significantly across jurisdictions. They have some common traits and a convergence process occurred on some elements. In particular, substantive rules tend to cover similar threats to competition, although the language of the provisions and their interpretation may differ. In a nutshell, antitrust rules prohibit agreements that may distort competition, abusive conducts undertaken by dominant firms, and many jurisdictions prescribe that mergers are ex-ante scrutinised to prevent those that may substantially lessen competition. The exact scope of these rules varies across countries and may even change over time. In the following we set aside this aspect and focus on some of the many other dimensions that shape a competition policy regime.
For the sake of explanation, these dimensions can be roughly divided in two groups. A first group encompasses the choices concerning the institutional set-up; the second group relates to how the main institutions exert their powers, something we may refer to as the "implementation". The institutional set-up includes the following three elements: (1) the position of the competition authority in relation to other public bodies; (2) the scope of the rules whose enforcement is attributed to the authority; and (3) the powers attributed to it. The implementation group includes: (1) the type of analysis used to interpret the substantive rules; (2) the use of sanctioning powers; and (3) the way the agency sets its priorities and goals, and the instruments used to pursue them.
In the following we briefly present the main alternatives that are available for each of these elements. 6

Institutional Set-Up
The institutional set-up of a competition policy regime concerns many factors. We focus on three main aspects that we think are particularly relevant. These are: the general model adopted to enforce competition rules, and the independence and accountability of the competition agency; the scope of the enforcement powers attributed to the competent authority; and the investigative and sanctioning powers that support and complement the enforcement activity.

The General Model and the Independence and Accountability of the Competition Authority
Two basic institutional models can be adopted for the enforcement of competition law. 7 The first one is the administrative model where an administrative authority is responsible for the investigation of cases and makes enforcement decisions that are then subject to judicial control. The administrative model is the most common within the EU. The second option is the judicial or prosecutorial model. In this instance, the administrative authority performs the investigation and then brings the case before a court. The court is responsible for making a decision on substance and on sanctions, or in regards to the imposition of sanctions only. In some jurisdictions, the administrative model has been amended to reap some of the benefits of the separation between prosecution and adjudication, which is typical of the judicial model. They have adopted a so-called "dual administrative model" where one body investigates the case and a different institution is responsible for making the decision. 8 The choice between the two models affects some important aspects that are generally related to independence and accountability. However, while the choice between the two models bears important implications, the overall level of independence and accountability of the competition law enforcement authority in the two institutional models depends on the more general characteristics of the institutions in a country, in particular on the overall quality of the institutions.
Where the judiciary enjoys a great degree of autonomy from other public powers and from private interests, the judicial model guarantees the maximum level of independence of the decision-making body. Yet, this may be achieved only at the expenses of other desirable features. First, to preserve their independence from other powers, courts are less accountable to the general public than administrative authorities, whose leadership is politically appointed. Second, courts generally lack personnel with economic expertise and therefore are not well positioned to conduct more complex economic analyses. Third, even in the judicial model there exists an administrative agency that to a large extent decides which cases to probe and which type of evidence to collect, thus reducing the scope of the independence in competition law enforcement.
The degree of independence that a competition authority enjoys is a trait that distinguishes competition policy regimes within the administrative model. Given the considerable powers that a competition agency exercises in any case, it is important to decide how it should be positioned in the sphere of public powers. 9 At one extreme of the spectrum we may attribute the role of competition agency to the executive branch. This is, for instance, the choice that has (partly) been made in the USA, where one of the two antitrust agencies is a division of the Department of Justice, and at the EU level, where the European Commission (i.e. the Union's executive body) is responsible for the enforcement of competition law. At the opposite extreme, we find agencies that enjoy a maximum level of autonomy, which is guaranteed by formal statutory independence from other public bodies, the lack of supervisory power of other governmental institutions, and the availability of adequate financial and human resources over which the agency has full control, so as to enjoy organisational and financial independence, too. In between there are various options in which the competition authority is subject to general instructions by the government or parliament, or to various degree of supervision, that may include guiding the authority's activities, giving instructions on some general aspects of the law or in regards to the budget or pertaining to wider policy matters.
As the degree of independence of the competition authority varies significantly within the administrative model, so does the degree of the authority's accountability. Especially when the competition authority has a very large degree of independence it becomes important to decide if and which measures should be adopted in order to ensure that it remains accountable to the citizens. This can be achieved by imposing some transparency obligations, and by requiring the authority to submit periodical reports to the parliament or to the government. In some cases, this can be coupled with the obligation to submit plans for upcoming years.

Scope of the Enforcement and Other Powers
An obvious important decision to be made in designing a competition policy regime concerns the extent of the enforcement powers attributed to the competition authority. This relates to three different areas: (1) the scope of the competition law, i.e. the conducts that the competition authority can scrutinise to verify their compatibility with the provisions of the law; (2) the exclusion of certain sectors (e.g. defence), or typologies of firms (SOEs), or transactions (e.g. mergers below a certain threshold) from the reach of competition rules or of some of them; and (3) the combination of competition law enforcement with the enforcement of other legislation.
Competition law generally contains three sets of prohibitions, concerning anticompetitive agreements, abusive practices by dominant firms and the ex-ante control of mergers. It is quite striking how the substantive rules of competition law tend to converge on a global scale, notwithstanding the many differences that exist in terms of economic systems and legal and cultural backgrounds across countries. To be clear, the interpretation of these rules is largely influenced by local specificities, an aspect to which we return later. Yet, a less widespread consensus concerns the desirability of including merger control among the enforcement powers of the agency. Historically, even in the most advanced antitrust jurisdictions, a merger control regime was introduced only decades after the first enactment of the competition law. However, countries that have introduced competition law more recently have set up a merger control system from the very beginning, but even in these cases there is some room of manoeuvre as the scope of the merger regulation may be defined by adequately setting the thresholds that trigger the powers of the competition authority.
In many jurisdictions some economic activities are exempted from the applicability of competition law. The rationale for this exemption is that these activities pursue more general interests and require an organization that is not compatible with competition. Moreover, the state assigns to some bodies the objective of pursuing these general interests and does not want other institutions to interfere with the decisions they have to make, aimed at achieving the assigned objectives. This approach makes perfect sense. However, it can be easily distorted to protect vested interests (i.e. rents) in activities that can be efficiently and effectively performed in competitive markets. Thus, the extent of these exclusions is an important policy choice that needs to be made based on careful assessment. Moreover, competition authorities might still act as advocates of the competition principles, even if the law prevents them from enforcing competition rules against the institutions entrusted with these general interest objectives. 10 This is part of the more general advocacy powers that the authority typically can exert to induce lawmakers and policymakers to avoid unnecessary restrictions of competition.
Finally, competition agencies may combine the competition law enforcement task with the enforcement of other rules. For instance, some authorities also act as sectoral regulators, or have supervisory powers over public procurement. More frequently, competition authorities have also the power to enforce consumer protection legislation, which include rules banning misleading advertising and unfair commercial practices.

Investigative and Sanctioning Power
In order to ascertain antitrust infringements, the competition authority needs a complex set of powers. These relate to: the ability to collect evidence of illegal conduct and data to inform the assessment of the likely impact of a firms' behaviour on the functioning of markets; and the possibility to impose remedies that restore competition and sanctions that confer deterrence properties to the enforcement activity.
Investigative powers range from simple requests for information to far-reaching powers such as the ability to inspect business and non-business premises (dawnraids). The effectiveness of inspection powers may be further guaranteed by ancillary powers such as the power to seal premises, to collect evidence stored on digital media and to impose sanctions against non-compliant firms. These powers are especially needed to uncover secret violations such as cartels. However, they may be crucial also for abuse of dominance cases where business plans may prove essential to establishing a proper theory of harm.
Once the authority has collected the required evidence, it needs powers to make its enforcement effective. These include the obvious power to adopt prohibition decisions. In some cases, the authority may be given the power to mandate interim measures to avoid irreparable harm to competition or to impose behavioural or structural remedies, as the mere termination of the illegal conduct may not lead to a well-functioning competitive market.
All these powers are pointless if not backed up by an effective sanction system. Sanctions may range from administrative fines on firms to criminal sanctions on individuals. They may stem from the violation of the substantive prohibition of the competition law or from non-compliance with all the other powers attributed to the competition authority, such as the refusal to respond to requests for information, or violations of interim measures, or the unjustified opposition to inspection.
All the powers briefly described so far seem indispensable for the effective enforcement of competition law so that it appears that there is little choice to be made in this respect. However, they should not be taken for granted, as in many jurisdictions the competition authority lacks some or many of them. So, it is worthwhile understanding why and in which circumstances giving the authority a less ample set of power is the proper decision.

Implementation of the Rules
Once the body responsible for the enforcement of competition law if identified, the general boundaries of the substantive provisions and the powers that can be exerted by the authority are defined, the competition policy regime depends on the way all this is put into practice. We refer to this aspect as the implementation of the rules, which comprises the main discretionary decisions that the authority has to make to turn the law into a policy.

Interpretation of the Substantive Rules
Across jurisdictions, antitrust provisions share the characteristics of being framed in very general terms. The general prohibitions need to be translated in a clear division between licit and illicit behaviour. In the past decades, an effect-based approach has gained popularity. This requires that each conduct is assessed taking into account the legal and economic context in which it takes place, so that a thorough economic analysis can reveal whether it is likely to impair competition or to foster certain efficiencies that justify its adoption. All conducts that do not alter competition or that have efficiency effects that outweigh the anticompetitive effects should be deemed compatible with competition law. Hence, the prohibitions will catch only those conducts that clearly have a potential negative impact on competition and welfare. This effect-based analysis intends to replace a more formalistic approach that aims to classify conducts as prohibited or permitted according to some prominent formal characteristics.
Although the effect-based approach has gained momentum, it would be inappropriate to claim that it has completely superseded the formal approach. Indeed, competition authorities, in all jurisdictions, continue to rely on some formal analyses as well as on precedents. This is understandable, also given that it responds to the need of the business community to have rules that are predictable.
A more nuanced way to describe the options available to a competition authority is to classify them according to the presumptions that are set up and the role that they play. In this respect we can have per se rules (generally, per se prohibitions), which establish that a certain conduct (e.g. a cartel) is prohibited irrespective of any economic analysis; when a per se rule is defined, the conduct is presumed to negatively affect competition and this presumptions cannot be rebutted; similarly, a per se legality rule establishes the presumption that the conduct does not restrain competition and, again, such a presumption is unrebuttable. 11 At the other extreme there is the rule of reason; such a rule starts with a presumption of legality and requires that the enforcement authority proves, through a full-fledged economic analysis, its anticompetitive effects and the failure of the possible efficiency benefits to balance out the social cost of an impaired competition. 12 These presumptions are frequently established through case law. However, competition authorities have other instruments to do so: guidelines, notices, etc. These soft-law instruments have the advantage of being more concise, general and widely known than a specific enforcement decision; moreover, they grant the authority more flexibility because they can revise their position if the need arises, whereas they cannot intervene on their past enforcement decisions.
A competition authority can use these soft-law instruments to clarify presumptions that it will apply in its enforcement activity, the type of evidence it will take into consideration to overcome these presumptions, and the required standard of proof. Hence, they will define the complexity of the economic analysis that needs to be performed to prove either the illegality or the legality of certain practices. Softlaw instruments can cover a number of subject matters, such as the definition of the relevant market, vertical agreements, cooperation agreements, unilateral practices, horizontal and vertical mergers, and so on.
For all these dimensions, the choice set is defined by the policy decisions concerning the adoption of per se rules or rules of reason, and whether the competition authority has to define its position by issuing guidelines or it can rely on its case law.

Sanction Policy
In the administrative model, competition authorities may enjoy some degree of discretion in the imposition of sanctions. This discretion can be exerted both in the definition of the harshness of the punishment and in the identification of the circumstances that warrant a more lenient or a more severe attitude. The authority may also identify cases in which the sanction can be strategically traded against some form of collaboration from the parties.
Again, all these aspects of the policy, within the limit defined by law, can be specified through soft-law instruments (e.g. guidelines). The guidelines can cover the general criteria by defining the value of the fine according to the gravity of the infringement and its duration. Moreover, they may define the aggravating circumstances that lead to an increase in the penalty and the alleviating circumstances that allow the party to pay a lower fine. As for the strategic use of the sanctioning power, this may concern: (1) the set up of a leniency programme to obtain information on secret cartels from one or more of the cartel members; 13 (2) special discounts granted to firms that adopt effective compliance programmes in order to encourage their introduction; (3) the adoption of commitment decisions where the authority foresees the possibility of swift resolution of its competition concerns and the opportunity to save resources for other enforcement activities; (4) the definition of settlement procedures so that the competition authority can avoid its decision being challenged before a court.

Priority Setting
Competition authorities, as any organization, are resource constrained. Hence, they cannot pursue all cases and need to prioritise their activities. Priority-setting may concern three dimensions. First, competition authorities have to decide whether they want to focus their activity on the anticompetitive conducts undertaken by firms or on the rules set by public authorities that distort competition. The former objective is pursued through the enforcement of the antitrust prohibitions; the latter through the exercising of its advocacy powers. Second, they may choose the types of infringements that they want to concentrate on. Third, they may want to identify sectors that deserve special attention.
Priority setting can be done through formal strategy and planning documents, which can be even submitted to a public consultation, or through informal statements by the authority's board. In some cases, for example in Turkey, priorities are defined by law.
To summarise: in this section we have identified several elements that define the characteristics of a competition policy regime and that can be fine-tuned to take into consideration the specificities of a country so as build a more effective policy. These elements are presented in Table 1. In the previous section we defined the main dimensions that shape a competition policy regime. We now turn to the assessment of the characteristics of emerging economies 14 that affect the optimal design of competition policy. We distinguish three main types of characteristics: economic, institutional and cultural. In the next section we will look at the Western Balkan countries through the lens of the identified relevant characteristics in order to derive the implications for the optimal policy mix that such countries should adopt.

Economic Characteristics
Several economic characteristics of emerging economies can be identified as having an impact on how competition policy should be designed and implemented. We will focus on three that appear to be particularly relevant: the existence of widespread barriers to entry and government interference in the economy; the sectoral composition of output and domestic consumption; finally, the role of the informal (shadow) economy.

Barriers to Entry and the Role of the Government
Emerging economies, especially those with a legacy of central planning or in general those without a tradition of well-functioning markets, are characterised by the presence of strong regulatory and economic barriers to entry, which impede the creation of well-functioning markets. Regulatory barriers have to do with existing laws and regulations that limit entry and operations in markets, such as licensing restrictions, trade rules or more general red-tape regulation that affect the possibility to open and run new businesses. 15 Also-and this is again especially true for formerly centrally planned economies-poorly planned or implemented privatisation processes have led to the entrenchment of legal monopolies instead 14 The literature that explores the link between economic development and the optimal design of competition policy refers in some cases to the notion of emerging economies, in others to that of developing or industrialised economies. We believe that adopting one terminology or another makes little difference, so we will use the different terms interchangeably and we will focus on the description of the economic, political and cultural characteristics that we believe should be taken into consideration when developing the optimal competition policy design. 15 Industrial policy, defined as the set of policies aimed at encouraging the development and growth of part of or the entire manufacturing sector, as well as other sectors of a country's economy, may also play a role in distorting competition and entry into the market. For a recent review on industrial policy and its rationale see Warwick (2013).
of open and contestable markets, contributing to the creation of artificial barriers to entry by granting dominant players exclusive access to essential inputs. The existence of regulatory barriers to entry in many developing jurisdictions, especially those emerging from a past of central planning, is combined with a still pervasive role of the state in the economy. The government still retains significant shares in different economic sectors, which tends to alter the playing field especially when inappropriate regulation provides privileges to state-owned companies, for example through exemptions from anti-trust laws enforcement. 16 Even when the legislation does not explicitly foresee exemptions for state-owned companies, political constraints may exist that prevent competition authorities to intervene against companies where governments' stakes are high.
Economic barriers relate to two separate aspects: first, the existence of poor framework conditions that limit the potential for entry and competition (e.g. poor infrastructure, underdeveloped financial markets, geographic barriers); 17 second, the conduct adopted by dominant companies, which actively pursue strategies to foreclose entry of competitors and take advantage of weak enforcement of competition legislation.
The presence of high economic and regulatory barriers to entry and counterproductive government regulation, from the welfare point of view, especially towards SOEs, has important implications for the optimal design of competition policy. Using the classification previously introduced in this paper, we believe that the features of the competition policy regime that should be affected are the independence and accountability of competition authorities; the scope of their enforcement power, the interpretation of substantive rules and the priority setting.
In particular, three main considerations seem to apply. First, in terms of institutional set-up, the significant role of the government in the economy and of possibly inappropriate regulation (especially towards SOEs) has important implications. In order to build trust with investors and the business community, competition agencies should not be perceived as favouring state-owned businesses, implying that rules should be defined to maximise transparency and reduce discretion. Transparency of the decisions and of the decision-making process (e.g. by means of publicly available guidelines that define in detail how the competition agency operates, assesses cases, and sets fines), as well as the choice of a legal standard that maximises certainty (even at the cost of a more reasonable effects-based approach) are paramount in ensuring trust in the private sector and also ensuring appropriate investment levels in sectors where the state retains an important role. We will return on this aspect later in this section. 16 In the words of Gal and Fox (2015), state and state-complicit restraints are among the major impediment to competition, which "clog the pathways for initiative on the merits, sometimes almost fully, while reciprocally raising prices". Because of this, the authors emphasize that the law should cover state-owned enterprises and state officials who facilitate illegal cartels or bidding rings by conduct outside the course of their duties. 17 A detailed description of the economic features that characterise emerging economies and shape competition policy effectiveness and design can be found in Gal and Fox (2015).
Second, in the presence of significant legal barriers to entry, the definition of the scope of enforcement power should attribute significant advocacy powers to competition authorities, imposing on policymakers the obligation to fully motivate their decisions when they diverge from the recommendations provided by the competition authority. This would imply that any legislation or regulation potentially affecting competition is scrutinised by competition authorities, whose expert opinion has to be taken into consideration by legislators. Similarly, when it comes to setting its priorities, competition authorities may choose to prioritise advocacy over enforcement, or at least the relevance of advocacy should not be underestimated. In countries where well-functioning markets are yet to be formed and significant barriers exist, market creation becomes a more important objective than the protection of competition. 18 Advocacy, both private and public, serves the function of promoting reforms that eliminate or reduce barriers to entry and operations, and creates the appropriate "demand" conditions for more open and contestable markets, by showing the benefits of higher competition for consumers.
Finally, still in terms of scope of enforcement power and priority setting, the existence of significant economic and regulatory barriers to entry should arguably lead to the prioritisation of abuses of dominance cases, 19 especially those where the potential for foreclosure is greater, precisely because tackling barriers to entry should 18 This conclusion is widely supported in literature. Budzinky and Beigi (2013), "Competition Policy Agendas for Industrialising Countries", Ilmenau Economics Discussion Papers, No. 81, underline that while in advanced industrialised countries, sustainable competitive markets exist and "just" need to be preserved or optimised (promoting competition), in industrialising or developing countries, the objective should be to establish working and accepted competitive markets (generating competition). In this context advocacy is identified as the most suitable tool in the hands of competition authorities to help build the market economy by informing the Government's economic policies. 19 The case for prioritising abuse of dominance cases in the presence of significant barriers to entry does not apply only to exclusionary conduct. Indeed, as discussed in Motta and de Streel (2006), "Excessive Pricing and Price Squeeze under EU Law", in Ehlermann and Atanasiu (eds), European Competition Law Annual 2003: What is an Abuse of a Dominant Position?, Hart, 91-125, the authors argue that under two main conditions the prosecution of excessive (exploitative) practices makes sense and the benefits may outweigh the costs of an otherwise dangerous antitrust instrument. The first necessary (but not sufficient) condition is the presence of high and non-transitory barriers to entry. In such cases, it is extremely unlikely that market forces would be able to challenge the dominant firm and that the abusive practices would be self-correcting. The second necessary condition is dynamic and it is that the monopolistic position should be due to current or past exclusive or special rights. In such cases, the traditional argument that monopoly prices should be accepted, to generate incentives to invest, is weaker and non-competitive prices may in principle be tackled with antitrust enforcement, especially in the absence of other effective instruments. be considered the priority for the competition law regime. 20,21 This does not imply that cartels should be considered less relevant, given the impact they have on consumers, especially in developing jurisdictions where basic goods that represent high proportion of the consumption basket of poorer segments of the population are usually the object of price-fixing arrangements. However, it does imply that when market creation is considered a priority, given the development status of a country, merger control should not absorb significant energies by competition authorities, and legal standards and presumptions should be set in such a way that mergers, which in general contribute to the natural dynamic process of market creation, should be prohibited only in very specific cases (for example when they consolidate existing dominant positions that have been acquired to privileged access to essential inputs). 22

Sectoral Composition of Output/Domestic Consumption
A second relevant feature to look at is the sectoral composition of the gross domestic product and of domestic consumption. For many developing jurisdictions it is typical to observe specific patterns, where, for example, agriculture and food-processing industry still represents a high share in domestic value added, and food products represent a major share in the average consumption basket. The specific sectoral composition of output/consumptions bears direct implications on the 20 An interesting assessment of the importance of prosecuting abuse of dominance, as a priority, can be found in Gal and Fox (2015). In their view, the "availability of the abuse of dominance prohibition is one of the most important weapons in the antitrust arsenal of developing countries to open up closed markets and thus help make markets work where they have never worked before; where business has been tied up in privilege and cronyism and dominant firms have blocked paths [. . .]." The presence of barriers to entry, especially linked to the legacy of centrally planned economies or more generally to the presence of formerly legalised monopolies, justifies an approach where targeting potentially foreclosing strategies by incumbents becomes a strategic priority for antitrust enforcement. 21 The case for prioritising abuse of dominance cases in the presence of significant barriers to entry does not apply only to exclusionary conducts. Indeed, as discussed in Motta and de Streel (2006). The authors argue that under two main conditions the prosecution of excessive (exploitative) practices makes sense and the benefits may outweigh the costs of an otherwise dangerous antitrust instrument. The first necessary (but not sufficient) condition is the presence of high and non-transitory barriers to entry. In such a case, it is extremely unlikely that market forces would be able to challenge the dominant firm and that the abusive practices would be self-correcting. The second necessary condition is dynamic-that the monopolistic position should be due to current or past exclusive or special rights. In this case, the traditional argument that monopoly prices should be accepted, in order to generate incentives to invest, is weaker and non-competitive prices may in principle be tackled with antitrust enforcement, especially in the absence of other effective instruments. 22 An interesting discussion about the relevance of merger control in developing jurisdiction can be found in Begovic (2017). In the paper, the author provides convincing arguments explaining why in the early transition towards market economies merger control should not represent an enforcement priority and notification thresholds should be set relatively high.
implementation of competition law, in regards to the way competition authorities may set their priorities. First of all, competition agencies should focus their implementation efforts on sectors that have a large share in the domestic economy and represent a significant portion of domestic consumption, or that have a significant weight in the production structure. 23 This is important not only because focusing on relevant sectors will have the largest effect on the economy's performance (in terms of efficiency and productivity gains), but also because it will help the competition authority to build its reputation as an agency that focuses on addressing issues that matter the most to the citizens and consumers. Second, competition agencies, whose financial and human resources are generally scarce, should specialise in terms of sectoral knowledge in those areas that also appear to be most relevant for the national economy.

Informal Economy
Much of the trade in developing countries is conducted by means of the informal sector, where traders are unrestrained by legal obligations such as registration and licensing requirements, as well as health and safety policies, taxation and labour laws. Informality poses challenges to the implementation of a sound competition policy, because of the competition that the informal sector exerts on the formal one. Indeed, the prevalence of informality in markets renders it difficult to assess the existence and impact of anti-competitive conducts, the size of the relevant market, the levels of concentration, the market shares attributable to incumbent players, and the extent to which firms hold dominant positions therein. In essence, the very basic tools of competition law are difficult to implement when informality is prevalent, and this should be carefully considered in the enforcement of all aspects of competition law. 24

Institutional Characteristics
Besides the economic characteristics of developing jurisdictions, institutional and governance features have significant impact on the way competition policy works and the way it should be designed. It is clear that competition policy does not operate in isolation, and it is unlikely to work effectively if overall institutions do not function properly. The effectiveness of the courts system is particularly relevant, 23 As suggested by Evenett (2015), one needs to be careful to potential endogeneity issues. For example, the reason why a sector may not be particularly relevant for an economy may have to do precisely with the lack of competition and high barriers to entry. 24 For a discussion about the impact of informality on the enforcement of competition law, see du Plessis et al. (2011). as courts have the authority to review agency decisions (in administrative systems) and make both first and second instance decisions (in prosecutorial systems). This implies that the degree of independence and effectiveness of the judiciary plays a key role in ensuring the effective enforcement of competition law. The role of state/ governments also has significant implications on competition law enforcement, both directly, through limitations and exemptions often granted to state-owned enterprises, but also (what matters most here) through indirect government interference in the competition authority's decision making, which limits their de facto independence.
A number of developing jurisdictions, in particular those that emerge from central planning, are characterised by weak rule of law, widespread corruption and economic structures, where resource allocation is not always driven by market mechanisms. On the contrary, resources are often allocated by governments in non-transparent ways to connected business groups and oligarchs. 25 Weak rule of law manifests itself through an incompetent, ineffective and non-independent judiciary, as well as through regulatory capture by vested interests that influence the activities and priorities of government bodies and de jure independent agencies. Weak rule of law is compounded by corruption, which is usually indicated as one the main business obstacles in internationally accredited surveys.
Institutional quality does not have only to do with corruption and vested interests. Even in countries where these phenomena are perceived to be less pervasive, the institutions that support the market require time to set up and start working effectively. This is particularly true for competition agencies, which generally operate with scarce financial and human resources, struggling to recruit people with adequate skills due to low salaries (lower than other civil service jobs) and a low reputation of the agency, at least initially. 26 Rather than being regarded with pessimism and leading to far-reaching negative conclusions about the desirability of competition policy in the first place, all these 25 The range of relevant institutional characteristics of developing jurisdictions is surveyed in Gal and Fox (2015) who describe the phenomenon of "missing or deficient institutions", which are taken for granted in the industrialized world. The authors stress the role that financial institutions, economic laws and a functioning court system play in guaranteeing the functioning of market economies. While financial institutions provide opportunities for the credit necessary for entering the market, the enforcement of contracts and property rights by the courts enables market players to engage in long-term trading. In the survey, another example of deficient institutions focuses on the inefficiency of the executive branch, which is a feature in many developing jurisdictions. 26 Several papers insist on this argument. Among others, du Plessis et al. (2011) stress how, even assuming that a developing country could fund the drafting of a suitable competition law policy, its adoption would be rendered useless without properly trained personnel to enforce the legislation in question. The authors rightly remark that many developing jurisdictions often suffer from a severe shortage of trained professionals, especially economists (with quantitative skills) that are qualified to assess the complex competition law concepts. Moreover, there is a lack of available training courses and seminars and scarce funding available to facilitate such training. To the extent that competition law policies are misapplied owing to a failure to understand the underlying concepts, the authors conclude, the adoption of the legislation would be without benefit, and could in fact be detrimental to the economy. aspects need to be properly acknowledged in the design and implementation of competition law. In particular, we believe that the institutional characteristics of developing jurisdictions should lead to several policy design choices.
The first consideration refers to the institutional set-up and the independence/ accountability of competition authorities. As discussed in Sect. 2 of the paper, lack of judicial independence points to the choice of an administrative model where, in order to protect as much as possible the competition agency from the undue interference of the government and of private interests, maximum independence should be granted by the legislation to the competition authority, which should not formally be part of the government and should not receive any instruction from the government itself. Having a de jure independent authority does not ensure an impartial enforcement of competition law-after all, competition authorities can be expected to reflect the positive and negative aspects of the jurisdictions in which they operate. For this reason, once again, it is fundamental that competition agencies operate with full transparency and that their discretion is limited. Transparency concretely means that all decisions should be made available to the public, that the agency should publish annual reports that fully explain its activities and the costs of its activities, as well as its strategy and priorities going forward, be open to scrutiny by third parties and stakeholders, including international organisations. Transparency also means that public and comprehensive secondary legislation should be available, especially in the form of guidelines that clearly describe the way the competition agency operates, including how it assesses cases, conducts investigations, and set fines.
Limiting the discretion of competition agencies is seen as of paramount importance in the context of a weak institutional environment. This limits the possibility of capture of competition agencies ensuring consistency and credibility of its decision making. Simplicity and predictability should be favoured, even if economic theory tells us that a more complex and case-by-case approach is expected to deliver the maximum welfare in normal conditions. In essence, as already highlighted, simplicity and predictability amount to the choice of a per se standard, whereby conducts are identified by clear and well-defined parameters and are allowed or prohibited depending on their characteristics, and not on a case-by-case effects based analysis.
The issue of discretion and transparency becomes more controversial in relation to the investigative power attributed to the competition agency. It may be argued that in a context of potential capture of the agency by the government or vested interests, the faculty to conduct unannounced inspections into business premises (dawn raids) should be limited, as the instrument could be abused, for example, to damage companies in favour of government clientele. This risk, the argument goes, cannot be mitigated by a strict authorisation process under the control of the judiciary, given that the judiciary in weak institutional environment is also subject to capture by vested interests. While we think that these arguments have some merits and we understand the choice adopted by certain competition agencies (e.g. Georgia), where the existence of cartels can only be proved by means of indirect economic evidence, we do not think that this should be the choice taken by developing jurisdictions. As the history of Western Europe and US has proven, the best way to tackle hard-core cartels, is through intrusive investigative techniques, coupled with stiff fines and effective leniency programmes. On the other hand, proving the existence of cartels through indirect economic evidence is an extremely complicated exercise, which requires deep expertise and extremely good data -precisely what is missing in developing jurisdictions. When good data and sophisticated modelling is not available, the analysis usually turns out to be poor, which in turn undermines the credibility of competition authorities. Overall then, the policy rule that we suggest is to attribute clearly defined investigative powers (in terms of conditions under which inspections can be conducted) to the competition authority, which should ensure its accountability through thorough reporting of its activities and of its decisions.
In regard to the sanction policy, a nuanced approach is advisable. High sanctions are necessary for fighting hard-core cartels. They are also needed to increase the effectiveness of a leniency programme, as the advantage of revealing the existence of a cartel is strictly related to the sanction that a cartelist may expect to pay. However, for other potential infringements, at least in an initial phase of operation, a competition authority may prefer to impose less severe sanctions. This is especially appropriate if the enforcement of the antitrust prohibitions is guided by simplified rules that aim to limit the discretionary power of the enforcer and increase the predictability of the interpretation of the law. Indeed, applying simple rules that are largely based on presumptions has the advantage of limiting the risk of the enforcement being distorted by inappropriate interests, but at the same time increases the possibility that some benign behaviours are prohibited and that harmful ones escape prohibition. In other words, simple rules may be over-inclusive, as they may prevent conduct that is competitive, or under-inclusive, as they may allow anticompetitive conduct. In these cases sanctions need to be lower in order to preserve the optimality of the enforcement system. 27 As mentioned, the effectiveness and independence of the court system is crucial in ensuring the sound implementation of competition policy. In this respect, while the issue of court independence is a complex one, which cannot simply be addressed through competition law, the creation of highly specialised review courts is expected improve the quality of the judicial review, while keeping low the risk of capture by vested political and economic interests.

Cultural Characteristics
Along with economic and institutional characteristics, cultural aspects may have a profound influence on the effectiveness of competition policy and should be taken into consideration when designing and implementing competition law. In developing jurisdictions the main relevant feature is the lack of a competition culture.
Especially in former centrally planned economies, competition is still not fully appreciated as a "positive" mechanism that is capable of favouring an efficient allocation of resources, promoting investment in better products and services, and ultimately delivering more and better jobs. As described extensively in literature, 28 in developing jurisdictions it is difficult to find absolute belief in decentralised economic power, and rules cannot work if the mind-set of a majority does not value the rules and their underlying mentality. In essence, a certain spirit of competition is a precondition for a working, sustainable and accepted competitive market economy. In many cases, competition is perceived with diffidence, as a negative force that may weaken societies by pitting people against each other, at the expense of a more egalitarian and cooperative growth model. Most importantly, the idea that the free market forces, combined with serious ex-post prosecution of unfair and abusive practices, can protect consumers from excessive prices is still difficult to accept. In many cases, the government is still perceived as the sole institution capable of protecting the poor through ex-ante regulation and price control. However, this type of intervention often leads to de facto government promoted cartels, which are unlikely to improve consumer welfare, as argued by competition sceptics.
How should this negative cultural bias against competition be reflected in the design and implementation of competition policy?
A natural response, in relation to the priority setting strategies of competition authorities, lies again in advocacy, focused on the private sector, in this context. Competition authorities should invest in educating the public, in particular the consumers and the media, of the benefits of competition. Pointing to the experience of other countries, telling concrete stories of how competition leads to more and better choices for consumers, competition authorities can contribute to the gradual creation of appropriate demand conditions for more competition and less economic control. 29 A second aspect to stress is that competition authorities need to implement the right strategy to rapidly build their reputation. This implies to select relevant and easy cases, quick wins that can contribute to affirming the role of the competition agency as an institution that is sided with consumers and with the most dynamic and innovative firms in the country. The strategy should involve building alliances with societal players that are receptive to the message, such as (foreign) investors, nascent consumer associations and international organisations. A competition culture cannot be created overnight, but the experience of several countries confirms that much can be achieved by an active and smart competition agency. In regards to the scope of enforcement power, the need for the competition authority to quickly build a 28 See, in particular, Budzinky and Beigi (2013). 29 Besides the fundamental role of competition advocacy, competition authorities should actively engage in providing guidance to businesses, about how to comply with competition norms. Businesses need guidance whenever a country lacks the appropriate competition culture. As Budzinky and Beigi (2013) notice, lacking spirit of competition implies that a considerable part of the competitors in the market are insecure or ignorant about how to behave in competition and compete on the merits. reputation of "being on the side of consumers" may suggest attributing to competition authorities functions that go beyond core antitrust and merger control, and encompass consumer protection. This is usually a controversial argument, given that lack of resources and adequate skills, combined with multiplicity of functions, may lead the competition authority to focus excessively on "easy" consumer protection cases that have immediate media and political visibility. This in turn would prevent the build-up of credible skills to prosecute hard-core antitrust infringements. We believe there is merit in this critique, but overall we lean towards the combination of antitrust and consumer protection responsibilities, especially in the early stages of competition law enforcement where pro-competition culture is very weak. This is not necessarily the case of Western Balkan countries, as we will see in the next section.

Emerging Economies and the Shape of Competition Policy
We now turn to the analysis of the Western Balkan economies, in order to confirm the relevance of the analysis conducted so far and to derive policy implications in terms of competition law institutions and enforcement. In the previous section we defined multiple characteristics that should be taken into consideration when designing and enforcing competition laws. We will now analyse the Western Balkan economies through these lenses. Naturally, a choice has to be made as to which variables one needs to look at, with no ambition to be exhaustive. Our assessment will mainly look at variables and data collected by international organisations such as the World Bank, the IMF and the EBRD. Furthermore, a choice of comparators needs to be made in order to quantify the gap between Western Balkan countries and both peer and more advanced jurisdictions (in terms of economic development). In our analysis, we would normally include as comparators Croatia and Slovenia, both former Yugoslav countries falling outside of the Western Balkan group; Bulgaria, Romania and the Czech Republic, central and south-eastern European countries that have completed the EU accession process; and the more advanced jurisdictions of Germany, Sweden and the US. These reference points should give us a clear picture of the positioning of the Western Balkan countries vis-à-vis countries that are at least in principle at different stages of the transition process towards well-functioning market economies. However, the comparators may differ according to the variable used and the availability of data, which often acts as a constraint to the analysis.

Economic Characteristics
In the previous section we identified several economic characteristics as essential to shaping competition law design and implementations. In particular, we focused on barriers to entry and the role of the state, the sectoral composition of the output, and the level of informality. We now present some data to depict the Western Balkan economies.

Barriers to Entry, Economic Activity and the Role of the State
As discussed, there are several dimensions that contribute to create barriers to entry and economic activity. We choose here to focus on five indicators that we consider relevant. First, we use the World Bank's Doing Business survey and provide the overall rank of the "ease of doing business" in the Western Balkan countries and the selected comparators. The overall ease of doing business rank captures several dimensions, including regulatory barriers to start (and close) a business, as well as barriers linked to access to electricity, dealing with construction permits, trading across borders, etc. 30 We then look at two specific economic (non-regulatory) barriers to entry and economic activity, namely the quality of the overall logistics infrastructure and the breadth of financial banking markets, as a proxy for access to finance. We measure the former using the World Bank's Logistic Performance Index, 31 a survey based measure of the quality of the overall logistic system in a country, while for the latter we use a simple measure of private credit (by banks) to the GDP, acknowledging the fact that this may be a crude measure of access to finance, especially for more advanced jurisdictions that have access to deeper financial and capital markets. 32 Also, in order to measure the extent to which barriers to entry may be caused by conducts pursued by incumbent dominant operators, we use the World Economic Forum GCI (Global Competitiveness Index) of the 30 For a description of the data and the methodology used by the WB Doing Business indicators, see http://www.doingbusiness.org/methodology 31 For a description of the underlying variables and methodology used to measure the quality of logistics, please see http://lpi.worldbank.org/. The LPI is based on a worldwide survey of operators on the ground (global freight forwarders and express carriers), providing feedback on the logistics "friendliness" of the countries in which they operate and those with which they trade. They combine in-depth knowledge of the countries in which they operate with informed qualitative assessments of other countries where they trade and experience of global logistics environment. Feedback from operators is supplemented with quantitative data on the performance of key components of the logistics chain in the country of work." The index is measured on a 1-5 scale, where higher numbers correspond to perceived better quality of performance in logistics. Data on Kosovo is missing. 32 Data on private sector credit to GDP come from the World Bank's World Development Indicators.
perceived "extent of market dominance". 33 Finally, our descriptive section emphasised the role of the state in the economy, as a key economic as well as an institutional factor that should be taken into consideration. In this respect, while a comprehensive measure of state presence in the economy is missing, we look at the quality of government regulation overall, by using the World Bank's Worldwide Governance Indicators of "regulatory quality", which in principle should capture perceptions of the government's ability to formulate and implement sound policies and regulations that permit and promote private sector development. This indicator notably includes, among other aspects, the burden of government regulation, the scope of state subsidies and administered prices. The graphs presented Figs. 1, 2, 3, 4, and 5 provide, even taking account all the limitations associated with the data, 34 an overall consistent picture. Western Balkan countries, with the notable exception of Macedonia (especially due to lower taxes and low administrative barriers to start a business), rank below EU countries in central and south-eastern Europe in terms of ease of doing business, which is a sign of higher barriers to entry and economic activity. As expected, the gap separating them form more advanced jurisdiction is very big, which clearly suggests that if we 33 Naturally, dominance is only a necessary but not sufficient condition for the existence of abusive foreclosing conduct. As such, the data provided can only indicate whether existing conditions can be conducive to potential anticompetitive entry-deterrent strategies by incumbent dominant companies. WEF data on Kosovo is missing. 34 There is an intense debate over the validity of Doing Business data, which led to a thorough review initiated by the President of the World Bank in 2012. For a review of the criticisms addressed to the Doing Business methodology see http://www.doingbusiness.org/~/media/WBG/ DoingBusiness/Documents/Methodology/Open-Letter-Review-of-the-Arguments-on-DB.pdf? la¼en. Similarly, WEF Global Competitiveness Index has been criticised essentially for being based solely on market participant's perceptions, which in turn can be affected and distorted by cultural bias which would undermine the credibility of the proposed rankings.
believe in the theoretical considerations presented in Sect. 3, the design of competition law and its enforcement should take these aspects into consideration, implying a degree of divergence from what is commonly considered "best practice". The gap is confirmed when we look at other aspects measuring overall barriers to entry, including the quality of logistics and level of development of the financial (banking) markets, however the picture is less univocal for these. In particular, while there is a clear infrastructural gap in terms of infrastructure between the Western Balkan countries and developed countries, the gap compared to central and south-eastern EU countries is less pronounced, with the exception of the Czech Republic, for logistics, and Croatia, for access to finance. In terms of access to finance, there is significant variability within the group of Western Balkan economies, with Albania and Kosovo showing significantly lower private sector to the GDP ratios than other countries in the Region. Figure 4 further contributes to the understanding of economic barriers in the market, this time linked to the existence of dominant positions in the market, which may lead to potential abuses and foreclosure of competitors and potential entrants. Again, while the graph suggests the existence of heterogeneity among Western Balkan countries, with Serbia ranking below its peers and Macedonia closer to EU countries such Romania and Bulgaria, on average, the intensity of the perceived dominance is significantly higher in the Western Balkans than in EU members in central and south-eastern Europe, and the gap appears stronger when advanced jurisdictions in Europe and with the US are used as comparators.
The final graph in Fig. 5 shows the overall regulatory quality as measured by the Worldwide Governance Indicators, which, as mentioned, captures aspects of state intervention in the economy and the overall quality of regulations. The graph confirms the analysis so far, with a significant gap between Western Balkans and both EU members in central and south-eastern Europe, (particular large in the cases of Kosovo and Bosnia and Herzegovina), and with the more developed countries included in our sample.

Sectoral Composition of Output
In terms of the sectoral composition of the output, aggregate data is reported in Fig. 6. We included a simple repartition of economic activity across three categories: industry, agriculture, and services. Naturally, a more disaggregated analysis would provide more insight, but even at this level, at least two observations can be made. First, compared to more advanced jurisdictions, agriculture still plays a significant role in Western Balkan countries, accounting between 5% and 20% of the GDP. Second, there is significant heterogeneity within the analysed group of countries, with economies such as those of Montenegro and Kosovo largely based on services, while in others, such as Serbia, industry features more prominently. In line with what was discussed in the previous section, the conclusion we should make by looking at this data is that antitrust enforcement should be prioritised accordingly, and appropriate sector assessment skills should be developed within competition agencies,

Informal Economy
Informality is certainly a pervasive phenomenon in all developing jurisdictions, and Western Balkan countries are no exception. For obvious reasons, measuring informality is a difficult task, and while several attempts have been made to get a sense of the size of the shadow economy, a comprehensive database that includes all the countries explored in this paper is not available. We rely on different data sources to provide an overall assessment of the phenomenon, namely: the BEEPS database assembled by the EBRD and the World Bank, which surveys domestic firms inquiring to what extent informality is perceived as an obstacle to doing business, on a scale from 1 to 4; the database provided by Schneider (2007), which contains a measure of the share of the informal sector in the overall economy and includes more advanced jurisdictions although the data is not up to date; and the ILO dataset, which contains only data for a small subset of the countries we analysed. 35 Several conclusions could be derived by analysis of these data sources: first, the level of informality in the Western Balkan countries is significantly higher than in more advanced jurisdictions. Broadly speaking, according to Schneider (2007), there is an average difference of 20 percentage points in the share of the informal economy between countries such as the UK and Germany and countries in the Western Balkan region. Second, there is heterogeneity in terms of informality both within the group of Western Balkan countries and when looking at central and south-eastern-eastern European countries currently members of the EU, as reported in Fig. 7 (2017) BEEPS database. For Kosovo, Macedonia, and to a lesser extent Albania, informality appears to be a stronger obstacle to business than in Bosnia and Herzegovina, Serbia, and Montenegro. In turn, informality appears to be more of a constraint in EU countries, such as Bulgaria, Romania, and the Czech Republic, with a much more positive perception in EU countries from the former Yugoslavia (Croatia and Slovenia). Finally, a note of caution should be attached to this data, as confirmed by the ILO database: the gap in informality between countries like Serbia and Albania appears to be much greater than what the BEEPS data suggests, as measured by the estimated share of employment in the informal economy.

Institutional Characteristics
We now turn to the quality of institutions in the Western Balkan countries. Several aspects and indicators could naturally be used to provide an assessment of the gaps existing between Western Balkans and the chosen comparators. We focus on three indicators, which in our view can provide an overall understanding of the institutional environment, but can also shed light on two specific aspects, namely the quality of the judiciary and its perceived degree of independence, and the effectiveness of corruption control. As indicated in Figs. 8,9,and 10, for this purpose we use three indicators: first, the Worldwide Governance Indicators measuring the "overall rule of law"; second, the World Economic Forum GCI index of perceived "judicial The overall picture is consistent with the one described for the economic characteristics, with a significant gap emerging between Western Balkan countries and the more developed jurisdictions of Germany, Sweden, and the US. However, two aspects need to be emphasised, which differentiate the narrative in terms of institutional quality compared to the one on barriers to entry and other economic characteristics. First, in terms of within country heterogeneity in the Western Balkan region, it appears from these indicators that Macedonia and Montenegro stand out as the countries with more effective institutional environment, at least in terms of perceptions. Second, the gap between the Region and EU members of central and south-eastern Europe is evident only in relation to Slovenia and the Czech Republic, while Romania, Bulgaria, and to a lesser extent Croatia, seem to be closer to the Western Balkan region and are far from Western Europe and the US. Overall, if we believe in the theoretical framework provided in the previous section, both the Western Balkan countries and the EU members in central and south-eastern Europe should be carefully considering the institutional and enforcement patterns of their competition laws, possibly diverging in significant terms from the trend currently observed in more advanced jurisdictions.

Cultural Characteristics
We finally look at the cultural characteristics of the Western Balkan region, in particular analysing the perceived attitudes towards the market economy and competition as a stimulus to growth and progress. We saw in the previous section that the lack of a strong competition culture should in principle shape the enforcement priorities of competition authorities, which should favour advocacy and "quick wins", as well as the institutional characteristics of the competition law, which may attribute more visible consumer protection roles to the agencies. The two indicators we used are taken from the Life in Transition Survey (LITS), published by the EBRD. The survey contains two relevant questions addressed to a sample of about two thousand individuals per country. The first relates to the overall attitude towards the market economy; specifically, individuals are asked whether they believe that "the market economy is always preferable to a planned economy", whether they think that "under certain circumstances the planned economy could be preferable" or whether they are "indifferent" to having a planned rather than a market economy. Responses are provided below, in Fig. 11, where the chosen comparators are different to the ones used previously, as Germany and Italy are the only available developed economies. The second indicator, also taken from the LITS, is specific to the notion of whether competition is good or bad, from a cultural point of view. Individuals are asked to rank, on a scale from 1 to 10, whether competition is definitely good as it brings out the best in people (with the most pro-competition answers scored 1) or whether it is bad because it undermines social stability and disrupts social relationships (with the most anti-competition views scored 10). We report in Fig. 12 the percentage of respondents who provided scores between 1 and 3, thus expressing an overall pro-competition attitude.
The results are very interesting and somewhat surprising, and surely would deserve a deeper analysis, which is beyond the scope of this paper. We limit ourselves to two observations. First, the overall pro-competition culture seems to be strong in the Western Balkan countries, broadly in line with other central and south-eastern European countries, and more interestingly-significantly higher than in both Germany and Italy. On the other hand, the overall attitude towards the market economy is more heterogenous, with high shares of pro-market respondents in Albania, Kosovo and Macedonia, and much lower shares in other former Yugoslav countries, including the Western Balkan countries as well as Croatia and Slovenia. Interestingly enough, a strong divergence can also be observed between Italy and Germany, with Italians apparently much less favourable towards the market economy. While it is difficult to reconcile these conflicting results, the pattern that seems to emerge for the Western Balkan countries is an overall belief in the merits of competition, while on the other side the perception is that under certain circumstances (possibly in some specific and essential sectors), a planned or regulated system should be adopted. This is indeed consistent with the patterns observed in  some developing jurisdictions, where the overall introduction of competition legislation is accompanied by significant exemptions granted to sectors that are considered key or strategic for national interests, independent of whether competition could or could not be introduced to the market.

Conclusions
In this final section we can try to assemble the pieces of information disseminated throughout the paper and harvest the results of the analysis. To do so we reverse the direction of our journey and start by summarising the main findings on the economic, institutional and social characteristics of the Western Balkan countries. These characteristics are then considered in order to identify the features of the competition policy regime, which seems more in tune with the broader context.
In terms of economic characteristics, a prominent feature that emerges from our analysis is the existence of high and pervasive barriers to entry. These are due to several factors: intrusive and inadequate regulations that limit the scope and dynamism of entrepreneurial initiatives, large-scale direct involvement of the government in many sectors, and the persistence of several markets in which, as a legacy of the previous economic regime, the dominant players may act strategically to prevent the entry of new competitors and defend their market power.
As for the sectoral composition of the economy, we notice a considerable level of heterogeneity across the Region; hence, we cannot offer conclusions of general validity. However, there are clear signs that the informal economy is sizeable in all Western Balkan countries. This suggests the importance of carrying out more focused analyses of each country to identify the sectors that have greater impact on the economy and consumer welfare, and in which informal activities are more pronounced.
The available evidence on the institutional characteristics of Western Balkan countries shows that in all of the vested interests have a strong ability to influence the decision-making process of the government and other public institutions. Of course there are no countries for which the opposite can be said and, furthermore, the situation in the countries under study does not differ significantly from that observed in central and south-eastern European countries that have already joined the EU. However, the gap with respect to more advanced jurisdictions is noticeable. This suggests that some attention needs to be given to this aspect.
Finally, the evaluation of certain indicators of relevant cultural characteristics, seems to provide striking support of the principle of competition and an overall positive attitude towards the market economy.
The design of competition policy in the Western Balkan countries has to take into considerations the factors just summarized. Some robust recommendations are applicable for the definition of the institutional set-up. Overall, the administrative model seems appropriate, given the low level of independence of the judiciary. However, within this model effective measures need to be adopted to fence the competition authority from the interference of the government and interest groups. This can be achieved also by imposing a high level of transparency of the activities carried out by the authority and making it accountable to the general public. For instance, the competition authority may be required to publish and fully explicate all its decisions (not only the decisions to open and close investigations, but also those dismissing complaints), or to publish a periodical report in which it describes past activity and sets the priorities for future actions.
As for the scope of antitrust rules and enforcement powers, it is strongly advisable to eliminate any formal exclusion of SOEs and other public entities from the application of competition law, as they may be a major force against a full deployment of competitive initiatives. Moreover, the competition authority should have the ability to have its voice heard. Its opinions and recommendations cannot replace the normal functioning of the democratic political process. However, the legislator and other decision-makers have to be forced to evaluate the implications of their decisions on competition and be asked to show that other general interests have to prevail and that they cannot be pursued by adopting measures with less anti-competitive impact.
As for the inclusion of ex-ante merger control regulation, we think that international experience has proven that an important gap exists if this is not part of competition law. However, we think that the reach of the merger regulation can be fine-tuned by setting relatively high thresholds that trigger the obligation to notify a merger. This would also allow the competition authority to save resources that can be used more productively in other tasks.
Combining competition law enforcement and other responsibilities, such as sectoral regulation or the enforcement of consumer protection rules, has pros and cons. In the case of the Western Balkan countries the cons seem to prevail. Sectoral regulators are more exposed to the risk of capture, a risk already significant in these countries. Additionally, it is not clear whether the competition authority actually needs the power of enforcing consumer protection rules to build its reputation of being on the consumers' side. Indeed, the available evidence indicates that consumers in the Region already perceive competition as a mechanism capable of defending their interests. Of course, there is a possibility that the used indicators overstate the ability of citizens to fully understand what competition is and the channels through which it fosters consumer welfare. Therefore in this aspect our conclusions have to be taken with extreme caution.
To complete the picture of the institutional setup we have to add a few considerations on investigative and sanctioning powers. We believe that a competition authority will be armless without the possibility of using effective instruments to collect direct evidence of illegitimate conducts and if it does not have the power to inflict adequate sanctions. Moreover, the quality of its decisions may be jeopardized if its decisions have to rely solely on indirect and circumstantial evidence, leading to the deterrence power of sanctions being further diminished by the possibility of higher rates of legal errors. However, in a context in which these powers may be bent toward the protection of vested interests, some prudence is wise. Moreover, this would call for even stronger transparency and accountability requirements. Some clear indications seem applicable also with respect to the implementation side of the competition policy regime. First, competition authorities in the Western Balkan countries should devote considerable energy to their competition advocate role, as inappropriate regulations appear to be the main source of obstacles to competition. This can be done by performing market inquiries and by acting as an advisor and a constant watchdog of the policy-makers. Sectors that have a significant weight, either in household consumption or in production, should be the preferred target of these activities.
Second, it seems advisable to focus enforcement efforts on abuse of dominance cases especially in markets where ill-conceived liberalization policies have led to the creation of dominant firms that could impede or retard full-scale entry of new competitors.
Third, competition authorities should be encouraged to clarify the interpretation of the applicable rules through soft-law instruments. This approach also has the advantage of making the enforcement less vulnerable to the action of special interest groups. A fast way to achieve this result would be to anchor the interpretation of the applicable rule to the case-law that has been developed in other jurisdictions, such as the EU.
Fourth, it would be advisable to ask the competition authority to state the main priorities of its action in publicly available documents. Of course, this should not compromise the ability of the enforcer to intervene when unexpected news uncovers the existence of serious anti-competitive conducts in sectors that were not prioritized, nor should it oblige the competition authority to reveal information that may put at risk the effectiveness of its investigations. However, an indication of the lines of action that the authority intends to follow increases its accountability and helps improve its performance.
Fifth, at least in an initial stage, competition authorities should use their sanctioning power with caution. Of course, hard-core cartels have to be harshly punished; severe sanctions may also be levied against firms that infringed on competition rules in prioritized sectors, if such priorities had been disclosed to the public. In all other cases more moderate sanctions are justified.
All these suggestions do not form a recipe that guarantees the perfect competition policy. Other ingredients need to be added. The quality of the human resources is of paramount importance, as is the quality of the judicial system and of other public institutions in general. However, this is true in any case: any recommendation in this respect would state the obvious.
To move the analysis one step forward one would need to check the actual features of competition policies in the Western Balkan countries and verify to what extent they match the ones that we believe are desirable and which aspects can be improved. We set this as the subject of future research. Other scholars and practitioners are invited to join. Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made. The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.