Bernhard Duijm
In several countries there has been a trend to transform highly regulated markets into competitive markets. Most economists agree that in many cases the introduction of competition in markets previously dominated by state monopolies or vertically integrated firms cannot be achieved simply by abolishing the old anticompetitive regulations. Often, pure market forces would not lead to the development of competition even if the anti-trust laws were strictly applied. Therefore, the process of liberalisation of these sectors should be accompanied by a new kind of regulation policy, not aiming at preventing competition, but at making competition possible and fostering it.
This competition-orientated economic regulation1 may include obligations for the incumbent firm to give access to certain facilities to new entrants in the market at prices (or more general: conditions) fixed or at least permitted by a regulator. While competition-orientated regulation policy and competition policy share the same objective, namely fostering competition, the methods of both fields of policy differ considerably. The usual goal of competition policy is to prevent anticompetitive behaviour and concentration processes which are expected to reduce competition. With the exception of merger control, the instruments of competition policy are applied ex post. They are used in a manner to forbid certain economic activities (including doing mergers and acquisitions that would lead to a dominant position or strengthen an already existing dominant position).
On the other hand, the instruments of regulation policy are applied ex ante. Usually, they prescribe the regulated firms to act in a specified manner, or in other words, they fix some of the variables or the way of setting the variables indicating market performance, such as prices and quantities for example, the prices that a privatized airport charges may be set in advance by the regulator.
Of course, competition policy is also a kind of regulatory policy and is sometimes called ex-post regulation policy. For the sake of clarity, in this paper the term "regulation" is used in the sense of ex-ante regulation only.
In a market economy, market performance is usually the outcome of market forces. In such a system, the replacement of the market forces by the settings of a regulator2 can be justified only in certain circumstances and where other policy instruments would be ineffective, e.g. in addressing particular market failures or in achieving specified policy objectives. In order to prevent any misuse of regulation policy, there is general agreement that regulators should be politically independent. This requirement is fulfilled when there are independent regulatory agencies separate from the government. Autonomous ministerial agencies linked to a ministry may be able to act with a substantial degree of independence (OECD/IEA, 2001, p. 4). However, a minister acting directly or indirectly as a regulator may use his regulatory power to achieve non-competition goals, e.g. reducing regulated postage rates during an election campaign.
In respect of the regulator's sectoral scope of authority there are three main approaches (Estache and de Rus, 2000, p. 45):
These institutional designs differ with regard to the sectoral breadth of authority; which of them is optimal depends on various determinants. Industry- or sector-specific regulators might be superior when regulations in different sectors or industries differ considerably. Main advantages of a single regulatory agency are seen in savings from "economies of scope" and in the avoidance of regulatory inconsistencies across industries (Hewitt, 1999, pp. 196 ff.).
However, in the view of most economists, even well designed regulation is inferior to a functioning market. Not least because of regulatory failure, regulation should be limited to specific situations and activities, as when natural monopolies or other structural factors impede the development of a functioning competitive market. Therefore, even in regulated industries, not all economic activities should be subject to ex-ante regulation, for example most of the non-aeronautical activities of airports. The rules of the competition law(s) are, of course, applicable to those activities, and - depending on the legal system - even to activities permitted by the regulatory agency.
Thus, some industries are subject to regulation rules as well as to general competition law and perhaps to sectoral competition law. Then the question arises which authority should be in charge of competition law enforcement in regulated industries. As there is interaction between competition and regulation policy (or in other words: ex-post and ex-ante regulation) the conduct of competition policy in a certain industry may influence the conduct and even the need for (ex-ante) regulations in this industry and vice versa. The assignment of authority in the field of competition policy is therefore a very important political decision.
For the sake of simplicity, let us first assume that there are one or several regulatory agencies separate from the general competition authority. The second assumption is that there is only a general competition law; sector-specific competition laws do not exist. There are at least five main ways of assigning the power to apply the competition law.
In practice, the situation might be more complex. The kind of assignment may not only differ from industry to industry. It might be different also within one industry depending on the kind of anti-competitive activity concerned. Generally, regulatory agencies are more involved in mergers and acquisitions than in the application of rules concerning anti-competitive behaviour of firms.4 The official institutional design may lie between some of the five "pure" ways of assignment described above. And the institutional design practised in reality may be different from the one laid down in laws. This might be especially true in the field of cooperation between the regulatory and the competition agency.
There is a more general aspect one has to deal with when the assignment of competencies is discussed: can general competition law be applied to issues laid down in a (specific) regulation law or not? Or in other words: does the application of a (specific) regulation law exclude the application of the general competition law or are both laws equally applicable?5 If the latter is the case then all regulatory decisions can be supervised by the agency in charge of competition law enforcement.
One possible advantage of combining regulation and competition policy enforcement within one agency may be the avoidance of inconsistent decisions. As both fields of policy deal with promoting and protecting competition - although those sub-goals have different weights in both policies - a decision taken on the base of the appropriate regulation law might counteract decisions based on competition law and vice versa. These inconsistent decisions can be avoided in cases where a single agency is in charge of applying the relevant regulation laws and general competition law.
Inconsistencies of regulation and competition law enforced by two different agencies may cause severe problems for the regulated firms. The same or similar activities which are permitted by the regulator under regulation law may be forbidden by the competition agency under competition law. But even the simple fact that regulators and the competition agency have concurrent competencies for the same activities may lead to higher costs for the firms.
It can be assumed that every regulator is reluctant to leave a case to the competition agency when he is empowered to deal with that case under the relevant regulation law. If he has the authority to apply the competition law equally, he will probably choose the most appropriate regime (regulation law or competition law) to solve the problem (Hewitt, 1999, p. 185). If he does not have these concurrent powers he must use regulation law, even in cases where the application of competition law would be sufficient. The result is that detailed rules (based on regulation law) have to be observed by the regulated firms with negative consequences for competition.6 Regulatory interventions may also affect investment and hence economic growth in the long run, especially when many indivisibilities in investment exist, e.g. in increasing the capacity of airports by constructing a new runway.7
This argument of over-regulation is based on a static point of view. It can be modified slightly for a dynamic point of view. Many regulations are presumed to be temporary. They are expected to be applied only until a functioning competition has developed. This process might be speeded up by technical progress, international integration of markets8 and so on, but also by the conduct of regulation policy. In order to ensure that regulations remain only temporary, many laws contain obligations for regular assessments concerning the need for continued regulation and in some cases the need for the regulatory agency itself (Haffner, 2000, p. 212). Such a regulation policy cannot be evaluated continuously, but in intervals of for example three to four years. Making regulation superfluous as soon as possible is not in the interest of the regulator when the abolition of regulation is connected with the abolition of the regulatory agency. Therefore, the regulator might be inclined to retain more regulations than necessary in order to demonstrate the need of regulations and the need for the regulatory agency.
Empowering the regulatory agency with competition law enforcement in the relevant industry may act as an incentive to reduce regulations sooner and faster, for under this institutional design the agency does not lose "responsibilities", as long as there are reasons for maintaining the agency. Only the instruments adopted have to be changed.
Another advantage of combined competencies is the possibility of synergies. Within an agency, information gathered for regulatory decisions are available for competition decisions, while the exchange of information between different agencies might be hindered. There might only be a few people with in-depth knowledge of a certain industry. Then, scarce human resources can be used in a more efficient manner when regulation and competition law enforcement is combined. Other possibilities of saving costs may also exist.
Combining regulatory and competition law competencies can have some serious drawbacks.
The conduct of regulation policy might influence the conduct of competition policy concerning the non-regulated activities of the regulated companies as well. For example, a regulator being used to set prices according to a certain rule when applying the relevant regulation law might ask for detailed commitments before giving his consent to a merger of two firms.9 This can lead to a permanent supervision of the behaviour of the merged firms which is not intended by competition law. The differences in the conduct of regulation and competition policy then diminish by implicitly expanding regulations to the non-regulated areas of economic activities. But here, the de facto ex-ante regulation (instead of a pure ex post regulation) might lead to a reduction in competition.
This "bias" towards regulatory measures may be reinforced by a least two factors:
Compared with the competition agency, a regulatory agency may show a less strict or even lax attitude towards actual or potential violations of anti-trust law. This can be explained by the fact that the regulatory agency might resort to the instruments of regulation law in cases where the anti-competitive behaviour leads to unwanted market performance. The result of the inappropriate application of competition law is more regulation than necessary. By doing so, the regulation agency can demonstrate the supposedly still existing need for regulation and for its existence.
Therefore, there are reasons for applying sunset legislation with respect to temporary regulation. But combining regulation and competition policy enforcement might make it more difficult to put through sunset legislation as competition law enforcement is a permanent task to be pursued also after all regulations of an industry have been abolished. It is hardly acceptable that a separate "regulatory" agency should exist any longer in order to apply general competition rules to an industry when no specific ex-ante regulations are applicable any more. As one can assume the employees of the regulatory agency are reluctant to give up their jobs, especially when some of their tasks, namely applying the competition law, still exist, they will continue to regulate ex ante as long as possible. An agency which has only explicitly temporary regulatory tasks can be dissolved with less resistance. In this way of looking at the assignment, combining regulation policy and competition policy within the regulatory agency will lead to a rather long life of the latter.
A division of tasks between the regulatory agencies and the competition agency giving the latter the exclusive power to enforce competition law might be advantageous for several reasons (Duijm, 2000, pp. 16 ff.).
If the enforcement of competition law and the enforcement of regulation laws are separated, co-operation and co-ordination of the agencies involved is absolutely necessary in order to avoid inconsistencies. Proper co-operation makes it easier for all agencies responsible for an industry or a sector to fulfil their tasks. But even if a regulator has combined competencies for a regulated industry or sector, cooperation with the competition agency is necessary to avoid inter-industry or inter sector distortions of competition resulting from considerable differences in applying competition law.
Co-operation and co-ordination may include the exchange of information and the common definition of subjects relevant for conducting competition policy, like the definition of markets and of market power. A far-reaching kind of co-operation is realised when the regulatory agencies are required to consult with the competition agency before they can apply competition law, and vice versa the competition agency must request a (non-binding) opinion from the appropriate regulatory agency before taking action against firms in a regulated industry or sector.
Whether and to what extent the above-mentioned arguments and other arguments in favour or against combining or separating regulation law and competition law enforcement are valid depends on many factors, which cannot be analysed here, except for one. One important factor is the institutional design of regulatory agencies concerning their sector responsibilities, whether there are (many) industry-specific regulators or (a few) sector-specific regulators or one general regulator. Generally, the narrower the industry coverage of a regulatory agency the higher the probability that the regulator is captured by the regulated firms. So, an industry-specific regulator is more likely to be captured by the firms than a sector-specific one, for in the latter case there are probably different interests among the regulated industries, which cannot be taken into account by one regulator conducting a unique regulation policy in all industries he is responsible for. Another reason for industry-specific regulators to be more prone to be captured is the fact that persons preparing and taking regulatory decisions will be well acquainted with the regulated firms after some time because, within the agency, there is no possibility for job rotation. Job rotation would effect a regular change of the industry an employee of a regulator deals with. When the captured regulator is in charge of the enforcement of regulation and competition law he will probably conduct both, regulation and competition policy, more in the interest of some or all regulated firms than in the interest of the public. Multi-sectoral or general regulators are expected to be the least capture-prone agencies for they are confronted with a wide range of divergent interests. From this point of view, giving the general regulator the competency of competition law enforcement seems to be the least harmful alternative with regard to a "proper" competition policy.
One has to ask whether this institutional design is really ideal for preserving a proper competition policy.
First of all: does it make sense to create one single regulatory agency in order to enforce different regulation laws? Should not the - in most cases - already existing competition agency10 be assigned with conducting competition and regulation policy? Such an approach is followed for example in Australia where the Australian Competition and Consumer Commission (ACCC) is the competition authority and a multi-sector regulator in charge of inter alia electricity, gas, telecommunication and airports on the national level11 (OECD/IEA, 2001, pp. 40 ff.).
Here, the question of assignment of competition policy competencies, the question of combining them with, or separating them from, regulatory competencies, is easily answered for there are no alternative agencies. This kind of approach had been favoured by the German Monopolies Commission until recently. It opposed the creation of a Regulatory Authority for Telecommunications and Posts and wanted the regulatory competencies in these industries to be assigned to the Federal Cartel Office. According to the last biennial report - published some months ago (July 2002) - the Monopolies Commission has changed its opinion. Due to the fact that regulatory measures - taken to create and promote competition - seem to be necessary for a longer time than originally expected, the Commission argues for a separation of regulation and competition policies. The Monopolies Commission fears that the conduct of competition policy will be negatively affected if the Federal Cartel Office should gain regulatory competencies, e.g. in the field of the gas market (Monopolkommission, 2002, p. 50*). To avoid a fragmentation of regulation policy it proposes the creation of a general regulation agency for all network infrastructure industries which should be in charge of the economic regulation, whereas technical regulations should be assigned to industry- or sector-specific agencies.
This accepted, the question of assignment of competition policy competencies still remains: should the general competition agency (in Germany: Federal Cartel Office) retain competencies in applying competition Law (in Germany: Act Against Restraints of Competition) to all industries or only to the non-regulated industries while the single regulation agency is in charge of applying it to the regulated industries?
There is no reason to expect different approaches to competition law enforcements across the regulated industries when a single regulatory agency is empowered to apply this law. This is in contrast to a set of sector- or industry-specific regulators where even closely related industries might be treated unequally. But one problem remains even if there is only one regulator - he might apply competition law towards regulated industries in a different way than the competition agency does towards non-regulated industries. If the regulator applies competition law in a regulatory manner, his competency in competition policy has more and farther-reaching effects on competition compared to - small - industry-specific regulators doing so. As even a politically independent central bank cannot ignore fiscal policy in designing the monetary policy, the general competition agency cannot ignore the competition law enforcement in the regulated industries. Otherwise, serious distortions of the level playing field throughout the economy may arise. Assuming he same number of regulated industries, the balance of powers in applying competition law is more affected to the disadvantage of the competition agency when there is only one general regulator than when there are many - smaller - regulators with competition policy competencies. This change in the balance of powers will remain for a long time, although the general regulator will probably be less reluctant to end regulation (and at the same time his competition policy competencies) in certain industries.
Another argument put forward in favour of combining regulation and competition law enforcement is less valid in the case of a general regulator specific knowledge of a certain industry. This special knowledge can be found in industry-specific agencies, especially when they are also in charge of technical regulations. However, it is doubtful whether there are industries with exceptional production processes or other peculiarities justifying a special approach to competition policy (Behrends, 2001, p. 84 f.). Even when one ignores this objection, there are still some disadvantages in empowering industry-specific regulators with competencies in competition policy:
It can be expected that industry-specific agencies would be very reluctant to phase out a special regulatory regime because this phasing out would lead to the end of their existence. The reluctance tends to prolong the separated enforcement of competition law in this industry.
Furthermore, the number of decisions based on competition law made by one industry-specific agency is probably relatively small.13 This small number creates uncertainty for the firms involved in competition law cases. Furthermore, there is high probability that overlapping competencies exist, resulting from economic activities of firms operating in different industries, for example in the electricity and in the gas markets. Firms operating in both markets are in the interest of customers who want "one-stop energy shopping" (Hewitt, 1999, p. 197). Technical progress may lead to other converging markets, e.g. the communication and the broadcasting market. When there are different agencies for both markets uncertainty concerning the responsibility arises. If two or more agencies have equal competencies, probably a forum-shopping arises.
Thus, in the case of industry-specific regulators with competencies in competition policy, it is not the individual power of the regulatory agencies which might give rise to concern for the general competition agency, but the potential differences and fragmentation in competition law enforcement.
Therefore, in the case of a general regulator as well in the case of industry-specific regulators there are - albeit different - reasons to give the general competition agency some power to control the actions of the regulators concerning competition law enforcement. For example, the regulatory agency could be obliged to request the opinion of the competition agency when it wants to apply competition law. A publication of the opinion forces the regulator to explain his competitive decisions if not in line with the policy of the competition agency.14
There is little need for the general competition agency to supervise or control competition law enforcement by regulators when the regulatory agencies are organised on a sectoral level. On the one hand, the individual coverage of industries of a certain agency is small enough not to impede the general competition agency in conducting competition policy as intended and not to distort the economy-wide level playing field seriously. On the other hand, the coverage of industries is large enough to resist attempts of capture from single industries and to facilitate consistent conduct of competition policy across closely related industries of the sector, provided that all industries of the relevant sector are under the regulatory and competitive responsibility of the appropriate agency. The last-mentioned condition is a serious drawback of this institutional design - the agency would have to be empowered to apply competition law to certain industries only because other industries, belonging to the same sector, are regulated. Since, in many countries, not all industries of a sector are regulated (e.g. often the oil market in the energy sector is not regulated) this would reduce the competencies of the general competition agency, when a regulatory agency is created for the network-based gas and electricity market and effect a change of the responsible competitive agency for some industries, which might lead to a change in competition policy in these industries. Therefore, this institutional design is a viable solution only for sectors where all or most industries are regulated.
Up to now, the analysis was made under the assumption that no industry- or sector-specific competition laws or rules exist. However, industry- or sector-specific laws often do not only contain regulatory rules, but also competitive rules.15 These specific competition rules might be either equally applicable with the general competition law or might overrule the latter. They might be enforced by the general competition agency or by the regulator(s). When the specific competition rules are enforced by the general competition agency, the results discussed above have not to be altered significantly.
When they are applied independently by the regulatory agency, most of the problems concerning competition law enforcement by regulators are reinforced an even greater resistance of regulators against phasing out can be expected, as there are special competition policy (permanent) tasks fulfilled exclusively by them; overlapping applicability of several specific competition rules; divergent application of similar rules, causing uncertainties; and so on.
Some of these problems can be overcome by concurrent competencies of the general competition agency - applying general competition law - and the regulators - enforcing specific competition rules. But this will probably lead to forum shopping if it is not ensured that the strictest policy - in the sense of the most pro-competitive policy - is put through. But uncertainties remain: will a permission of one authority be repealed by another authority? Therefore, specific competition rules - enforced by regulators - call for co-operation between them and the general competition agency.
If specific competition rules or laws exclude the applicability of the general competition law, the need for co-operation also exists. The objectives of specific competition rules and the objectives of general competition law are essentially the same. Instruments, definitions and so on should be used in a similar manner to prevent competitive distortions. There is no place for an uncontrolled competition of competition policies within one economy, although some procedural rules might differ according to different market structures, e.g. the existence of market entry barriers.
Many countries are undertaking regulatory reforms, some of them, like Ireland, in a public consultation process (Department of the Toaiseach, 2002). One of the key problems in all the countries is the relationship between regulation and competition and the relationship between competition and regulatory authorities. According to the OECD, there are an increasing number of combined regulation/competition laws being tried in the OECD member states, most of them assigning sector specific regulators with regulatory and competition policy tasks (Hewitt, 1999, p. 186).
Such developments have been observed in Germany, where in the telecommunications industry certain competencies concerning competition policy were transferred from the Federal Cartel Office to the Regulatory Authority for Telecommunications and Posts. In the UK, the Office of Gas and Electricity Regulation (OFGEM) was established in 1999. Thus, a sector regulator was created in place of the former industry-specific regulatory agencies for gas supply and electricity regulation. OFGEM has concurrent power in applying the competition act in cases concerning anti-competitive behaviour, but not with regard to mergers (OECD/IEA, 2001, pp. 86 ff.). In Germany, the government gave up plans to extend the federal railroad authority with primary technical regulatory functions into an economic regulator with considerable competencies in the area of competition policy. Nevertheless, the authority is now empowered to investigate ex officio in cases of alleged discriminatory access to the railroad networks.
However, other developments can be observed as well. In Italy, up to 1997, the Broadcasting and Publishing Authority was in charge of enforcing general competition law in these industries. This assignment was repealed in 1997 and the general competition authority has assumed responsibility for these industries. A new Communication Regulatory Authority was established, which has no competencies for competition law enforcement; it can express its opinion on competition policy issues (Wise, 2001, p. 101). This can be regarded as a separation of competencies and as a division of labour according to the supposed relative advantages of competition and regulatory agencies.
In the Netherlands, the regulator for the electricity market was created as a chamber of the general competition agency, but for the telecommunication sector an independent regulatory agency was established, which has to co-operate closely with the general competition authority. This can be seen as an institutional design where the regulators are linked to the competition agency (OECD, 1999, pp. 224 and 232).
It is interesting that in the UK the Financial Services Authority has no important competition policy competencies - contrary to the other sectoral regulators (Hewitt, 2000, p. 38). The same is true for Germany.
So, rather divergent developments in assigning powers concerning competition policy can be observed.
What do all these considerations suggest when the institutional design of competition policy in the field of airports is concerned?
First of all, airport regulation policy and airport competition policy seem to be highly mutually dependent. With the exception of very large cities being served by two or even more airports, airports are considered as more or less "natural" monopolies. As they have significant potential to exercise market power they are (still) government owned and/or regulated. However, they are not always monopolies. There are passengers travelling from points located between airports who may consider different airports as substitutes. The same is for true for passengers who are not time-sensitive (for example tourists). Although inter-airport competition is often still not intensive, it might become so when infrastructure is improved, for example by connecting airports and cities by a network of high-speed trains.
Actual and potential inter-airport competition is endangered by collusion and cartels among airports within a region. Individual market power is then transformed into collective market power. Regulation in order to avoid misuse of market power will become a permanent task if restrictions of (potential) competition cannot be prevented. The same is true in the case of mergers of privately-owned airports within a region or when an incumbent airport establishes a new one (for example by purchasing a former military airfield).
Regulation of airports (for example the allocation of take-off and landing slots) influence the competition in the airline industry, especially the effects of mergers and alliances in this industry. It can act as an important barrier to entry for new domestic or foreign airlines. In order to overcome these barriers to entry, airlines might purchase other airlines to obtain their slots. In addition, mergers and alliances between two or more airlines influence inter-airport competition. As travellers prefer connecting flights of the same airline to those of distinct airlines,16 a merger of airline X serving the route from airport A to airport B and of airline Y serving the route from airport B to airport C might make airport A a closer substitute for airport D, from which direct flights to airport C exist. But of course airline mergers and alliances can also reduce inter-airport competition.
From the interdependences between regulation and competition in the field of "airport-policy" it follows that a high degree of co-operation between the authorities in charge of competition law and regulation law enforcement is necessary. Combining regulatory and competition policy competencies concerning airports might be the best institutional design.
But which authority should be in charge of these combined competencies? As we have seen, inter-airport competition has effects on inter-airline competition and vice versa. Effective competition in the market of air transport requires functioning competition in both markets - airline services and airport services (OECD, 2000b, p. 228). This high mutual dependence of the two markets makes an "air transportation authority" superior to a narrowly defined (industry-specific) "airport regulatory authority". Although the air transportation market is a specific market in that sense that other forms of transportation are no (close) substitutes, assigning regulatory law and competition law enforcement to the general competition authority might be advantageous. As mergers and especially alliances between airlines are often across the border they may be subjected to several anti-trust jurisdictions and reviewed by different national authorities. While some international co-operation agreements between competition authorities exist, facilitating consistent decision-making, such an international network of agreements between transportation authorities would have to be developed.17
However, drawbacks of this institutional design may also exist. In Germany, the state anti-trust authorities are in charge of applying most parts of the GWB (Gesetz gegen Wettbewerbsbeschraenkungen, Act Against Restrictions of Competetion) as long as restrictive business practises do not have inter-state effects. The state anti-trust authorities are not regional branches of the de facto politically independent FCO, but in most länder (states) parts of the state ministries of economics and therefore politically not independent. Thus, in many cases, the realisation of this design would lead to a shift of competencies within the länder governments from the ministries of transportation (up to now in most länder responsible for the de facto regulation of airports) to the ministries of economics. Regardless of which ministry is in charge, its policy might be influenced by the ministry of finance, representing the interests of the state as owner of the airports. In the run-up to privatisation, fiscal interests of a state can dominate the economic interests of the whole economy. Other authorities, for example the FCO, might pursue these economic interests in a better way.18
This example shows that an evaluation of institutional designs requires a very broad assessment of a wide range of inter alia political, constitutional, juridical, and economic aspects.
Why do differences in the institutional design of regulatory and competition agencies exist? Why do they exist between countries and within countries?
There are at least two answers. First, not all governments tried to reform regulation in a systematic manner. In Germany, the energy bill did not contain rules concerning the assignment of agencies in order to circumvent the second chamber of parliament, where the government had no majority (Bundesrat, 1998, pp. 124 f.). The Dutch parliament opposed a telecommunication authority being integrated into the competition authority (OECD, 1999, pp. 224, 232). Thus, even if the optimal institutional design had been known, there was - for political reasons - not always the possibility to realise it.
But - and this is the second answer - the ideal institutional design is not yet known. Which of the "models" described above is ideal depends on many factors, varying from country to country and on time. I will mention just a few factors:
The answers to these and similar questions are different in different countries. In a negative way, some institutional designs can be excluded as obviously inferior to others, e.g. dozens of regulators, whether with or without power to enforce competition law. But a positive selection is impossible. A "good" regulatory and competition policy can be achieved by various designs, which may sometimes combine some of the described models, e.g. a general regulatory agency with no competition policy competencies and very few sectoral regulators with competencies in competition policy.
Whichever policy is adopted, an intensive co-operation of the authorities involved and a clear-cut pro-competitive regulation is essential to foster competition throughout the economy.
1 Aspects of technical regulations will not be discussed in this paper
2 Regulator, regulatory authority, and regulatory agency are used as synonyms. For the institutional design of regulatory agencies see OECD/IEA (2001, p. 23).
3 In this paper competency in competition policy and regulation policy means the power to apply the competition law or the laws concerning economic regulations (regulation law). Usually, both types of law give considerable discretionary powers to the agency applying the law. Competency in the sense of making laws is not discussed, although the agencies in charge of conducting policies might play an important role in designing the legal framework. See OECD (1997, pp. 19 ff.) for the role of competition agencies as advocates of deregulation. Of course, courts also have important competencies concerning competition policy, but they are omitted in this paper.
4 Especially in the US, some regulatory agencies have the power to approve mergers and acquisitions; see Victor (1998, p. 520) and Calvani and Robb (1998, pp. 515 ff.). However, there are important deviations from this rule.
5 In Germany, this question is discussed especially with regard to the telecommunications sector; see Moschel (2001).
6 See Boyfield (2000, p. 38) for examples of over-detailed regulations.
7 See Nicoletti (2001, pp. 22 ff.) for network industries.
8 For example, international liberalization of the airline industry might increase the number of potential origin airports within a certain region (especially within border zones), thus intensifying competition between airports and reducing the number of airports that can be regarded as a natural monopoly.
9 For other examples see Hewitt (1999, p. 184).
10 Möschel (2002), pp. 684 f. argues that in many countries, competition agencies other than ministries are relatively new bodies while some industry-specific supervisory or regulatory agencies have been existing for a long time. This is true for industries where private firms are active - sometimes alongside public enterprises (banking, assurance etc). However, it is questionable whether these regulators are pro-competition. In the case where all enterprises are (still) public-owned (in many countries: airports) generally some kind of an "internal regulation" is applied by the (local) government which acts as the owner and - more or less directly - as the regulator. In the process of privatisation and liberalisation explicit regulation might be necessary. In order to protect the conduct of regulatory policy against particular interests of the public owner (e.g. revenues from privatisation) a new institutional design for regulation is desirable.
11 However, there are state-based industry- and sector-specific regulators.
12 See Hewitt (1999, pp. 190 ff.) for this example and other cases.
13 This argument is less valid when the regional scope of the agency in charge is large, e.g. in the case of an EC regulator in place of national regulators in the EC member states. But it is not clear whether centralisation of regulation enforcement is in accordance with the principle of subsidiarity. For an efficient regulation the regulator might have to rely on information and complaints from the consumers, which tend to be more easily available in the case of national regulators.
14 This kind of co-operation is adopted between the Italian Central Bank as the regulator of the banks and the Italian Anti-trust Authority; see Wise (2001, p. 100).
15 See Koenig and Kuhling (2001) for the German electricity and railroad industries.
16 See OECD (2000a, p. 130) for some reasons.
17 See OECD (2000a), p. 136 for the AA/BA alliance; there was no international arrangement involving the US Department of Transportation.
18 The predominance of fiscal interests (revenues from privatisation) could be observed in the regulation of the German telecommunication and postal industries.
Behrends, Sylke: Neue Politische Ökonomie, München 2001.
Boyfield, Keith: The Politics of Regulation, London 2000.
Bundesrat: Stellungnahme des Bundesrates, in: GWB - Gesetz gegen Wettbewerbsbeschränkungen, Wirtschaft und Wettbewerb, special issue, Düsseldorf 1998, pp. 120-130.
Calvani, Terry and Kevin Robb: Market Liberalisation and Competition Policy, in: International Business Lawyer, Vol. 26 (1998), pp. 514-517.
Department of the Taoiseach: Towards Better Regulation, Dublin 2002.
Duijm, Bemhard: Fachbehörden als Wettbewerbshüter?, in: ifo-Schnelldienst, Vol. 53 (2000), No. 12, pp. 3-10.
Duijm, Bernhard: Wettbewerbssicherung nach Privatisierung und Deregulierung: Aufgabe allgemeiner Wettbewerbs- oder sektoraler Regulierungsbehörden?, in: Hartmut Berg (ed.), Deregulierung und Privatisierung: Gewolltes - Erreichtes - Versäumtes, Berlin 2002, pp. 9-27.
Estache, Antonio and Ginés de Rus: The Regulation of Transport Infrastructure and Services: A Conceptual Overview, in: A. Estache, G. de Rus (eds.), Privatization and Regulation of Transport Infrastructure, Washington, D.C. 2000, pp. 5-50.
Haffner, Robert C.G.: Policy Issues in Telecommunications Reform: The Netherlands, in: OECD (ed.), Privatisation, Competition and Regulation, Paris 2000, pp. 207-213.
Hewitt, Gary: The Relationship between Competition and Regulatory Authorities Background Note, in: OECD Journal of Competition Law and Policy, Vol. 1 (1999), No. 3, pp. 177-219
Hewitt, Gary: Bank Mergers - Background Note, in: OECD Journal of Competition Law and Policy, Vol. 2 (2000), No. 4, pp. 13-66.
Koenig, Christian and Jürgen Kühling: Institutionelle Regulierungsordnung in der Eisenbahn- und Energiewirtschaft - sektorspezifische Regulierungsbehörden oder Bundeskartellamt?, in: Wirtschafl und Wettbewerb, Vol. 51 (2001), pp. 810-820.
Möschel, Wernhard: 1st das Verhältnis von TKG und GWB neu zu durchdenken?, in: Kommunikation und Recht, Vol. 4 (2001), pp. 619-622.
Möschel, Wernhard: Das Verhältnis von Kartellbehörde und Sonderaufsichtsbehörden, in: Wirtschaft und Wettbewerb, Vol. 52 (2002), pp. 683-688.
Monopolkommission (Monopolies Commission): Netzwettbewerb durch Regulierung. 14. Hauptgutachten der Monopolkommission gernäß § 44 Abs. 1 Satz 1 GWB 2000/2001 - Kurzfassung, 2002.
Nicoletti, Giuseppe, Regulation in Services: OECD Patterns and Economic Implications, OECD Economics Department Working Papers No. 287, Paris 2001.
OECD: The Role of the Competition Agency in Regulatory Reform, OECD Proceedings, Paris 1997.
OECD: The Relationship between Competition and Regulatory Authorities, in: OECD Journal of Competition Law and Policy, Vol. 1 (1999), No. 3, pp. 169-176 and 220-246.
OECD: Airline Mergers and Alliances, in: OECD Journal of Competition Law and Policy, Vol. 2 (2000a), No. 2, pp. 127-225.
OECD: Competition Policy and International Airport Services, in: OECD Journal of Competition Law and Policy, Vol. 2 (20006), No. 2, pp. 227-261.
OECD/IEA: Regulatory Institutions in Liberalised Energy Markets, Paris 2001
Victor, Paul A.: Interaction Between Competition and Regulatory Regime in Deregulation, in: International Business Lawyer, Vol. 26 (1998), pp. 518-521.
Wise, Michael: Review of United States Competition Law and Policy, in: OECD Journal of Competition Law and Policy, Vol. 1 (1999), No. 1, pp. 9-70.
Wise, Michael: Review of Competition Law and Policy in Italy, in: OECD Journal of Competition Law and Policy, Vol. 3 (2001), No. 2, pp. 65-122.