GETTING THE BALANCE RIGHT

��

��

Kuantan, Malaysia, 5-6 September 1998

��

Session on the Interrelationship of Competition Policy and Regulation

��

��

(Davies, Ward & Beck �� Toronto)

��

Introduction

��

"The list of things that one can 'demand' of an economic system is limited only by the human imagination, itself a fairly outrageous thing"

George J. Stigler

It is probably safe to say that in most industrialized countries, competition is now accepted as the best available mechanism for maximizing the things that one can demand from an economic system. Economic regulation is increasingly perceived to be at the opposite end of the spectrum, i.e., it tends to lead to disappointing, sub-optimal outcomes, leaving many people with reduced real income and lower standard of living. Unlike competition, economic regulation imposes costs on society in terms of its administration, its distorting effect on economic efficiency and, as we have seen in recent years, the significant time, effort and expense associated with its removal.

Studies consistently demonstrate that deregulation has been accompanied by large price reductions to consumers and substantial improvements in quality and service. In one recent review of a number of such studies relating to the deregulation of the natural gas, long distance telecommunications, airlines, trucking and rail industries in the U.S., it was reported that real prices dropped at least 25% and sometimes close to 50% within ten years of deregulation in those industries. The annual value of consumer benefits from such deregulation was estimated to be approximately US$5 billion in the long distance telecom industry, US$19.4 billion in the airline industry, US$19.6 billion in the trucking industry, and US$9.10 billion in the railroad industry. "Consumers gained substantially �� not just because of rate reductions, but also because of improvements in the quality of service. The only declines in service quality attributable to deregulation or regulatory reform occurred when regulation previously limited customer choice, forcing consumers to pay premium prices for gold-plated service." Moreover, "[c]rucial social goals like airline safety, reliability of gas service, and reliability of the telecommunications network were maintained or improved by deregulation and customer choice".

Unfortunately, many transitioning and developing countries continue to be highly regulated, with large state-owned sectors and inefficient firms or oligopolies operating in markets insulated by various types of barriers. A strong competition policy can not only help to deliver lower prices to consumers and greater allocative, dynamic and productive efficiency, to the benefit of all citizens in such countries, but it can also help to create better conditions for democratic institutions. In short, "the democratization of political systems and decentralization of economic decision making are mutually reinforcing processes".

In the wake of disappointing experiences with regulation, it is often forgotten that competition and regulation have the same ultimate goals, namely: to prevent the exercise of market power and facilitate the efficient allocation of resources. Where competition is unlikely to produce this result, some sort of regulation has typically been called for, either as a substitute for competition or as a means of "holding the fort" until the competition arrives. This suggests that there is a need to coordinate regulatory policy and competition policy in a way which recognizes their mutually reinforcing nature and optimizes economic welfare.

Changes in technology, the reduction of barriers to international trade and the increasingly important impact of economic thinking on policy making have recently led many countries to reconsider the need for various regulatory regimes in their economies. The deregulation of the natural gas, telecommunications, transportation and, most recently, electricity industries, provides ample evidence of the deregulation epiphany which has grasped numerous countries over the last 20 years. Now that we have had some experience with regulation, deregulation, regulation and competition, it is useful to step back, revisit the underlying rationale for regulation, examine the experience and develop some basic principles for (i) reducing the distorting effect on competition of regulations in those sectors that warrant continued regulation; and (ii) deregulating sectors that no longer warrant regulation.

A major theme of this paper will be that regulatory frameworks which unnecessarily prevent or limit competition need to be reassessed with a view to eliminating, restructuring or amending them so that they either no longer prevent or limit competition, or do so only to the minimum extent necessary. Another major theme is that the transition towards competition in those sectors which no longer require full scale regulation should be swift and include standards and milestones. It is recognized that the extent to which each nation may be able to make progress in deregulating and introducing competition into its domestic economy will be a function of its unique political and economic realities.

Part I of this paper provides a brief discussion of the goals of competition policy and notes that in some cases those goals may be better achieved through means other than competition, including regulation and even private cooperation. Part II summarizes the principal alternative justifications that have been given for economic regulation, identifies sectors that are often subjected to such regulation and provides examples of the types of policy instruments through which such regulation has been effected. Part III discusses ways that regulatory regimes, even in a natural monopoly context, can impose unnecessary limitations on competition and lead to sub-optimal results in terms of efficiency and overall economic welfare, particularly in industries in transition towards deregulation. Part IV then provides some suggestions for striking the right balance between competition and regulation, in order to maximize economic efficiency and the standard of living of affected citizens.

��

1. The Goals of Competition Policy

Despite the fact that the first competition law was enacted over a century ago, the goals of competition policy have not been well articulated. Indeed, there is a continuing debate regarding those goals. These include the protection of competition as an end in itself, the promotion of economic efficiency, the prevention of wealth transfers from consumers to producers, the protection of personal autonomy, equality and liberty, the protection of free markets and democracy against Marxism and totalitarianism, the protection of small business, the protection of workers and even the protection of the environment.

Indeed, the purpose clause in section 1.1 of the Canadian Competition Act includes a potpourri of often conflicting objectives. It states:

The purpose of this Act is to maintain and encourage competition in Canada in order to promote the efficiency and adaptability of the Canadian economy, in order to expand opportunities for Canadian participation in world markets while at the same time recognizing the role of foreign competition in Canada, in order to ensure that small and medium-sized enterprises have an equitable opportunity to participate in the Canadian economy and in order to provide consumers with competitive prices and product choices.

Similarly, the contradictory nature of the underlying goals of U.S. antitrust policy has led one observer to note. "Antitrust policy cannot be made rational until we are able to give a firm answer to one question: What is the point of the law �� what are its goals?" Everything else follows from the answer we give. Is the antitrust judge to be governed by one value or by several? Only when the issue of goals has been settled is it possible to frame a coherent body of substantive rules." The significant shift that has occurred in the hierarchy of these contradictory goals in the U.S. is perhaps nowhere better reflected than in the shift in merger policy from the days of the Warren Court, when a highly structuralist approach oriented towards the preservation of small and independent businesses was adopted, to the welfare economics oriented 1992 Horizontal Merger Enforcement Guidelines. Similarly, there was a substantial, ideologically driven, shift in vertical restraints policy from the days of United States v. Arnold, Schwinn & Co., which treated all vertical non-price restraints as being per se illegal, to the 1985 Vertical Restraints Guidelines, which severely narrowed enforcement policy towards vertical restraints, and which ultimately were withdrawn after they were rejected by the U.S. Congress.

Despite the continuing debate about the ranking of the underlying goals of competition/antitrust policy in North America and elsewhere (for example, the European Community), there is a growing consensus in nations that have significant experience with competition laws that the paramount objective of such laws and the broader competition policies of which they form a part is to promote vigorous competition (i.e., rivalry) in the marketplace. In turn, competition is desired in order to maximize economic efficiency, including allocative efficiency (which is maximized when there is an optimal allocation of resources in a market), dynamic efficiency (which is maximized by the optimal introduction of new products or new production processes over time) and productive efficiency (which is maximized when products are produced at lowest possible cost). Optimizing these types of efficiency is generally recognized to be the best way of maximizing total welfare in an economy, i.e., the sum of consumer's surplus and producers' surplus across all markets, thereby maximizing the average standard of living of people within the economy.

Private restraints that create, enhance or facilitate the exercise of market power are generally prohibited or at least subjected to review by competition laws, because they may lessen or prevent rivalry and ultimately reduce allocative, productive and/or dynamic efficiency. It will often be the case, however, that even in the absence of private restraints, a market will not be able to support competition. Where a "market failure" exists which prevents competition from serving the critical purpose of facilitating efficient price and non-price behaviour in the marketplace, some sort of regulation has typically been warranted. This, and other justifications for regulation will be discussed in the following section.

It may be noted that competition laws and, more broadly, competition policy, are increasingly reviewed as being poor tools for pursuing non-efficiency objectives. That is to say, society can "get a bigger bang for its buck" by using other industrial policy tools to promote objectives such as regional development, employment, small businesses, wealth redistribution, etc., than by using competition policy to pursue those types of goals. Moreover, pursuing non-efficiency objectives such as fuller employment, the preservation of small and medium-sized businesses and vaguely defined international competitiveness can be counterproductive, because, in the long run, a market power/efficiency standard is more likely to lead to positive results in terms of these objectives.

In addition, whereas the potential market power and efficiency implications of proposed initiatives or past conduct can be objectively assessed in terms of producers' surplus and consumers' surplus, the weighting of non-efficiency goals is entirely subjective. The same is necessarily true regarding any balancing of social and political effects against market power/efficiency effects, because these two classes of effects are not commensurable on any standard and cannot be assessed pursuant to operational rules. As a result, it is difficult, if not virtually impossible, to achieve consistency over a period of time with such a policy. This in turn leads to substantial costs associated with uncertainty. Furthermore, as predictability decreases and the decision-making process in increasingly perceived to be somewhat arbitrary, public confidence in a competition policy is difficult to maintain. Finally, the costs of administering a broad "public interest" standard are probably significantly higher than the costs of administering a market power/efficiency standard.

��

2. Alternative Justifications for Regulation

We now consider alternative justifications for regulation and evaluate these justifications from an economic efficiency perspective. Rationales that have been offered for regulation can be grouped into the following categories:

• to address market failure
• to advance the "public interest"
• to advance special interests
• to assist in the transition to a competitive market

1. Market Failure

The most broadly accepted rationale for regulation is that it is necessary to address situations of where the market is unable or is unlikely to be able to deliver goods and services to consumers in an efficient manner, i.e., because competition cannot be sustained in the industry in question. These types of situations are referred to as "market failure". For example, there might be externalities, barriers to entry, few buyers, few sellers or imperfect information. Where these conditions exist, markets may not work and regulation may be appropriate. Each situation needs to be assessed individually.

One of the classic and generally recognized examples of market failure is public goods. To the extent that the transaction costs that would be associated with charging a price for a public good likely would exceed the benefits that are derived from charging the price (particularly if few people are willing to pay), or to the extent that it may be impossible to exclude individuals from enjoying the benefit of the good, the government must assume responsibility for production of the good and recover its costs through the tax base. Examples of public goods include national defence, parks, public schools, flood control protection, lighthouses, and road construction and maintenance (although toll highways that are constructed and operated by the private sector are increasingly appearing). These are goods or services for which the cost of extending the service to an additional person is zero. The form of "regulation" typically adopted in respect of public goods is for the government to assume responsibility for deciding what is to be produced, thereby effectively preventing private businesses from making those decision, although assistance is typically requested from the private sector pursuant to calls for tenders or other contracts for the supply of the goods and services required (e.g., to supply the army or to build the park, school, lighthouse, road, etc.).

In addition to public goods, another form of market failure may arise when economies of scale and scope create barriers to entry and thereby prevent competition from developing or being sustained in an industry. When there are economies of scale or scope, such that only one firm can survive, a natural monopoly exists. However, given that a natural monopolist could easily exercise market power by extracting substantial prices for its product(s) from consumers, and may have little incentive to behave efficiently, governments often regulate natural monopolies. This is particularly so where the monopoly occurs in respect of necessities (that is, widely consumed goods which are highly demand inelastic, such as water, natural gas distribution and electricity distribution), because wealth transfers from consumers to the monopoly supplier have the potential to be very large. Historically, the regulation of natural monopoly was thought to allow the realization of the economies associated with the economies of large scale production while avoiding the "deadweight" loss associated with monopoly pricing.

There are a number of alternative forms of regulation that may be used to regulate natural monopolies. These fall into two broad groups, (i) cost of service or rate of return regulation, which permits the regulated entity to obtain a predetermined rate of return on its capital (as defined by its rate base), and (ii) performance based regulation, such as price caps, which set a maximum price that a regulated entity can charge, while permitting the entity to retain any profits that can be realized through cost reduction initiatives. The latter approach is designed to decouple costs and rates so that carriers have a strong incentive to minimize costs and no incentive to inflate their costs or to shift them between regulated and unregulated activities. Revenue sharing is somewhat of a hybrid of rate of return regulation and performance based regulation, in that it permits the regulated entity to retain a share of the returns that exceed its allowed rate of return, thereby providing a greater incentive to pay greater attention to costs.

It may be noted that the forces of globalization and technology are narrowing the class of industries considered to be natural monopolies, as markets that formerly were thought to display natural monopoly characteristics (e.g., electricity generation, electricity retailing, natural gas retailing, local and long distance telecommunications, rail transportation and even postal services) are opened up to competition. As one observer has noted, this suggests that experience, rather than theory, is more instructive about whether a particular industry or market is a natural monopoly.

Another type of market failure relates to information. Consumers who are uninformed or misled may under or over consume. In egregious cases, they may be defrauded of substantial sums of money or even their life savings. Similarly, where incumbent suppliers are able to exploit informational asymmetries between themselves and potential new entrants, they may effectively exercise market power by deterring entry, thereby putting themselves in a position to exercise market power. To enable the market to function efficiently and prevent against fraud, regulations such as product labelling laws, deceptive marketing practices laws and even criminal laws relating to fraud are often enacted.

An additional category of market failure relates to externalities. Externalities occur when the costs and benefits of producing and consuming certain products are not fully considered or internalized in the production and consumption calculus. Environmental legislation is the most obvious example of regulation dealing with externalities. A less obvious example is zoning regulation, which prohibits activities which may generate externalities from taking place in close proximity to inconsistent land uses. An increasingly common type of externality is the network externality. Demand side network externalities can occur where there are enormous benefits to being a member of a network or standard. As the network or standard is embraced by more people or organizations, its value to existing members rises. Supply side network externalities can occur when the cost of providing services to additional consumers reduces the overall cost of the network. As an existing network grows, potential suppliers of competing other networks are often unable to generate or maintain enough sales to compete with the "first mover" or the growing network. The dominance of VHS over Beta is the most cited example of this type of externality, while the strong rise of Microsoft's Windows computer operating system at the expense of Apple's operating system appears to be poised to be a more recent case. In emerging network industries, governments are still wrestling with how to approach this problem �� i.e. through competition law or by leaving the market to decide winners and losers. To some extent, this is reflected in the recent Microsoft case.

2. Public Interest

Another rationale for regulation that can be related to market failure is to promote what is perceived to be the public interest. Regulations related to health and safety, the environment, labour, food and drugs, transportation (e.g., airline, trucking and rail services), securities, financial institutions, insurance, government procurement, health care and investment often are supported by reference to "public interest" considerations. These types of regulations can be effected through a variety of instruments, including legislation which establishes a licensing regime, prevents or requires certain types of behaviour, imposes foreign ownership restrictions, or imposes product or technical standards. Alternatively, a national or sub-national government may simply assume full responsibility for the sector. Examples of this include health care and air transport, which are delivered by the public sector in some jurisdictions.

3. Special Interest-based Regulation

Often, regulatory regimes are imposed because special interest groups are able to influence the government, usually in the name of "public interest", to implement such regimes. The anticipated benefits to the special interest groups from such regulation provide such groups with a strong incentive to lobby for these types of regimes. In short, this form of regulation often is nothing other than a means by which special interest groups use the political process to extract wealth from other groups in society. Examples of regulatory regimes that are justified by the need to protect one or more special interest groups include supply management schemes (which are common in the agricultural sector �� for example, the dairy, poultry, pork, grain and fruit industries), labour codes, investment and procurement laws which impose restrictions on foreign firms, licensing regimes which make it difficult for foreign and other firms to enter markets (these types of restrictions are common in the transportation and financial services sectors), product and technical standards which have a similar effect (these types of restrictions are common in the construction sector) and foreign ownership restrictions (such as those that exist in cultural industries and some of the industries noted above).

4. Transitional Regulation

Another rationale for regulation is to facilitate the transition of industries from regulation to competition. The forces of globalization and technological change have necessitated ongoing and sometimes profound changes to regulated industries. In industrialized nations, many industries, to varying degrees, have already begun the process of deregulation. As previously noted the most significant include telecommunications, airlines, natural gas, trucking, railroads and, more recently, electricity and financial services. In some former central command economies and certain other developed countries, deregulation is moving more slowly, even though it arguably may be more pressing.

The ultimate objective of deregulation is to permit market forces to perform the functions that regulators and regulations have attempted, often unsuccessfully, to achieve. An industry's adjustment to deregulation will be characterized by intensified competition and increasing operational liberty that will cause the industry to become more technologically advanced, to adopt more efficient operating and marketing practices, and to respond more effectively to marketplace signals. Through this process, prices, rather than arbitrary and subjective factors, become the primary mechanism involved in determining entry, investment and the ultimate allocation of resources. This ultimately leaves consumers and society as a whole far better off.

Deregulation, however, involves complicated issues that require very careful consideration to ensure that the benefits of deregulation are not lost, for example by inadvertently enabling deregulated entities to establish private restraints in the place of the public restraints that have been removed. In addition, during the transition to open markets, the situation can be politically fragile, because if the transitory framework fails to produce expected benefits on a timely basis or produces short-term harm to stakeholders with political influence, policy makers may be pressured to re-regulate or to limit deregulation. Thus, a close monitoring of the transition can be critical, both to preserving the appropriate industry framework as well as the necessary support of politicians and other stakeholders. The critical questions become, therefore what is the role of competition and what is the role of the regulator and regulations in this transition?

Sorting out the roles to be played by competition and by regulatory authorities during the transition requires striking a balance between sending appropriate investment signals to potential entrants and mitigating the exercise of market power of incumbent entities. Where to draw the line depends in large measure on the speed with which new rivals can gain a foothold in the deregulated industry, and their incentives to do so.

��

2. Ways in which Regulatory Regimes Can Impose Unnecessary Limitations on Competition and Reduce Efficiency

Having identified the various justifications that have been advanced in support of regulation, as well as some of the policy instruments that are frequently used to effect the desired regulatory goal, we now discuss ways that regulatory regimes, even in a natural monopoly context, can impose unnecessary limitations on competition and reduce efficiency, particularly in industries in transition.

1. Natural Monopolies

Regulating to address natural monopolies and other types of market failure can be consistent with maximizing allocative, dynamic and productive efficiency, provided that the regulation does not unnecessarily prevent or limit competition, or reduce efficiency. Whether the regulation is consistent with the maximization of wealth and efficiency will depend upon the type of regulation adopted, its costs and its scope.

With respect to the type of regulation, a consensus appears to be emerging that cost of service/rate of return types of regulation reduce the incentives of regulated entities to minimize their costs. In fact, this approach to regulation "creates incentives for the access provider to deliberately choose an inefficient mix of productive inputs", because this will expand its rate base and increase the level of profits the entity is permitted to receive. "Consequently, the rate of return approach tends to be associated with over-capitalisation and over-investment in infrastructure facilities." In some cases, this over-investment has led to the existence of huge "stranded" costs, i.e., costs that could not be recovered in a competitive environment, subsequent to deregulation. Rate of return regulation also provides incentives for the regulated entity to cross-subsidize its competitive businesses by shifting costs to its regulated business. In addition, in an environment where a regulated entity also competes in competitive markets, rate of return regulation gives rise to very complex problems, "as the regulator becomes embroiled in increasingly adversarial battles over service pricing, costing and cost allocation issues". Other shortcomings associated with rate of return regulation relate to the fact that historical costs must be used in determining the rate base, there are difficulties associated with determining a fair rate of return, and difficulties are often encountered in developing a rate schedule when there are different categories of customers. To the extent that performance based regulation such as price caps avoids these problems and provides strong incentives to minimize costs, it is more desirable, from the perspective of allocative, dynamic and productive efficiency.

With respect to cost, a regulatory regime will not be efficiency enhancing if the economic costs associated with operating and complying with the system exceed the economic costs associated with not regulating.

With respect to the scope of regulation, many regimes have gone further than necessary in regulating natural monopolies by extending the regulatory framework beyond the segment of the industry that is a natural monopoly. For example, the telecommunications industry has been pervasively regulated despite the fact that only the local loop was arguably naturally monopolistic. Even this assumption is now questionable, as cable companies and suppliers of wireless technologies are poised to enter local telecommunications markets in several jurisdictions.

The electricity industry provides another recent example. In the past, virtually the entire industry around the world was regulated. However, deregulation of the generation and retail stages of the industry in many jurisdictions over the last few years has confirmed that competition can thrive in those segments, provided that they are structurally separated from the transmission and distribution functions (which continue to display natural monopoly characteristics) and provided that certain other basic steps are taken. A similar experience has occurred with the natural gas industry in a number of jurisdictions, where the retail and production stages have been significantly deregulated.

Another "natural monopoly" industry in which regulation arguably has gone further than necessary is the postal industry. A number of observers, including the Canadian Competition Bureau, have suggested that there is substantial scope for competition to emerge in this industry, including in the first class letter business.

In addition to extending the scope of regulatory regimes beyond the natural monopoly aspects of the industry, another way that competition can be adversely affected in regulated industries is by permitting blanket or overly broad exemptions from domestic competition laws for such industries. Furthermore, jurisprudential "state action" or "regulated conduct" doctrines can adversely impact upon competition by providing an exemption from domestic competition laws for conduct that is simply permitted or tolerated by a regulator, for example, pursuant to its power of forbearance. This can lead to situations where conduct is neither required by law, actively regulated nor subject to domestic competition laws. Unfortunately, it is not clear what action can be taken to address these types of situations, other than perhaps limiting the scope of such exemptions by legislation.

Another way that competition can be harmed in respect of activities that are not within the ambit of a "natural monopoly" is by failing to prevent the extension of market power by the regulated entity into upstream, downstream or adjacent unregulated markets. For example, by failing to require the establishment of structurally separate affiliates and establishing complementary protections to ensure that full separation is maintained between the regulated entity and its affiliates operating in other markets, policy makers can leave substantial scope for anti-competitive cross-subsidization to occur. Although such separation may lead to a sacrifice of efficiencies, the difficulties associated with attempting to implement satisfactory accounting rules and procedures to ensure proper cost allocation between competitive and regulated aspects of the regulated entity's business are such that any benefits to be gained by not requiring structural separation are typically not worth the significant risk to competition that can arise under an accounting rules approach, particularly in fragile, emerging markets.

2. Other Public Interest Regimes

As with regulatory regimes directed toward natural monopolies, other types of public interest regulatory regimes may or may not increase economic efficiency, depending upon their scope and cost. Broadly speaking, regimes which prevent or otherwise hinder the development of competition with respect to prices, service, quality, variety, advertising and innovation are likely to reduce efficiency and the economic welfare of consumers. However, they may promote other public interest goals that may be considered to be more important. Examples would include environmental and health and safety laws, laws directed towards the prudential aspects of the financial services and insurance industries, securities laws and regulations directed towards minimum quality standards, such as exists in the professions (e.g., law, medicine, dentistry, accounting, land surveying), the construction industry (e.g., building codes) and the food and drug industry. To the extent that such regulatory regimes promote confidence in the standards of the goods or services that are purchased, they promote efficient consumption of those goods and services, thereby increasing allocative efficiency and encouraging innovation. These positive effects may outweigh the above-noted negative effects on competition and efficiency.

However, even these types of regulatory regimes can prevent or lessen competition unnecessarily, for example, by erecting barriers to entry by foreign or other competitors. For example, certain types of technical product standards in the construction industry sometimes appear to have been designed with the express purpose of eliminating legitimate competition from foreign suppliers who make their products according to different, albeit equally safe and technically sound, specifications. Similarly, entry requirements into many professions, including the legal profession, often impose inordinate impediments to entry by qualified members of the same profession in other jurisdictions.

Perhaps the most anti-competitive aspects of public interest regulatory regimes are prohibitions or restrictions on price competition and/or new entry (whether from the same jurisdiction or otherwise). For example, prior to 1978, the U.S. airline industry was subject to both maximum and minimum fare regulation by the Givil Aeronautics Board ("CAB"). The CAB also controlled entry into the industry as a whole and on city-pair routes. The net effect of this regulation was to maintain a government enforced carted that kept average fares far above competitive levels. In addition, the CAB often tried to keep fares on low-volume, short-haul routes below the actual cost of service, by subsidizing them with higher-than-competitive fares in the long distance and high-volume markets.

Similarly, prior to 1980, the U.S. interstate trucking industry was regulated by a regime which maintained restrictive rules on entry by new carriers and even on the ability of existing carriers to provide service to new routes or for different commodity classes. As a result, regulated rates were maintained far above long-run incremental costs, particularly for less than truckload services.

Other examples include the rail industry, the long distance telecommunications business, and the production of natural gas in the U.S. Canada has had similar experiences with virtually all of these industries, as well as trucking and airlines. However, the effects of regulation arguably were more severe because the corresponding regulatory regimes in Canada lasted longer than in the U.S., thereby placing participants in those industries at a serious disadvantage to their U.S. counterparts when competing for certain types of business. As with the U.S. experience, deregulation of these industries in Canada was followed by substantial price reductions, service improvements and innovation.

In addition to the foregoing, some types of public interest regimes substantially impair competition by explicitly prohibiting entry by foreign rivals, and/or by limiting foreign ownership of domestic firms to a relatively small percentage. These types of restrictions continue to exist in a significant number of Canadian industries, including the financial services sector, the airlines sector and the telecommunications sector.

3. Special Interest Regimes

As noted in Part II嚗嚗�, the raison-d'etre of virtually all special interest based economic regulation is to protect the markets of the special interest groups from competition from other competitors. For example, many aspects of regulation in the transportation sector are designed to limit competition between the various modes of transportation (e.g., rail versus truck, or intercity bus versus air). In other industries, the domestic or sub-national producers are protected from foreign competitors or from competitors in adjacent sub-national jurisdictions. The Canadian cultural sector (e.g., book publishing and distribution, film production and distribution, broadcasting) is one of the more high profile examples of the former, while provincial marketing boards in the agricultural industry are glaring examples of the latter. Procurement policies of governments at all levels fall into one or both of these categories. In addition, barriers to competition from foreign competitors are often erected by tariff or non-tariff barriers (e.g. quotas, voluntary export restraints, technical standards, anti-dumping or countervailing duties), while barriers to competition from competitors in other sub-national jurisdictions of the domestic economy often are erected through the licensing and accreditation process. At a more local level, the taxi industry is an example of special quasi-interest regulation that has left virtually everyone worse off, with the exception of the holders of scarce permits, who are able to obtain huge rents for such permits.

Other types of special interest "regulation" that can have significant adverse effects on competition include the export cartel exemptions in various domestic competition laws (including those of Canada and the U.S.), and provisions in the competition laws of other nations (such as Germany and Japan) that apparently authorize rationalization cartels, crisis cartels and/or recession cartels. Indeed, weak and over-zealous enforcement of competition laws can also have anti-competitive consequences, in the former case by permitting private restraints to flourish, and in the latter case by chilling potentially procompetitive behaviour. In both cases, economic efficiency is reduced.

As previously noted, the net effect of special interest economic regulation generally is to transfer wealth from consumers and/or potential competitors to the entities benefiting from protection. To the extent that these regimes maintain prices at supra-competitive levels, retard innovation and seriously limit rivalry, they reduce allocative, dynamic and productive efficiency.

4. Transitory Regimes

Perhaps the most significant way in which competition and economic efficiency can be impeded in respect of transitory regimes is by leaving the regulator with excessive power. This can arise by express design or by leaving the regulator with too much residual power or discretion and by not placing a time limit or other objective benchmark on the duration of the transitory regime. Alternatively, the objective benchmark can be set too high, for example by requiring a finding of no ability to exercise any market power whatsoever, before the regulatory regime terminates.

Unfortunately, politicians sometimes have little incentive to limit the power of regulators. As one observer has remarked, politicians may be more concerned about short-term complaints from constituents than the long-term benefits which are likely to result out of the transition to competition.

Regulators are notorious for being unable to resist the temptation to elaborately plan either the structure of markets or the transition process. A case in point is the telecommunications industry in Canada. The Canadian Radio-television and Telecommunications Commission ("CRTC") has had a very heavy hand in managing the shift to competition in the telecommunications industry over the course of the last decade. As Janisch has noted, its approach to deregulation "required a massive shift in regulatory resources from traditional concerns of monopoly regulation to new concerns for 'fair' competition on an 'even playing field" Notwithstanding that this process has dragged on for approximately a decade, the CRTC's Chair, Francoise Bertrand, commented last year that she still sees "a need, for some time, for an expert referee to ensure that the competition [in the industry] is sustainable and to see that the interests of consumers and citizens are protected". This led Janisch to reply: "There is little recognition [in Ms Bertrand's statement] that it is the 'invisible hand' of competition, not a regulatory referee, which should be protecting citizens and consumers."

One of the most potentially serious shortcomings associated with leaving a regulator broad scope to "manage" a transition to competition is that the regulator has a basic conflict of interest: the sooner the transition is complete, the sooner the regulator and staff at the regulatory agency must look for new work. Moreover, regulators do not have a competition mind set. Their natural instincts are to carefully manage and control, rather than to leave things to uncertain market forces. This can colour the entire transition process, with the result that their "management" of the transition has the ironic result of delaying that transition. Indeed, regulation also winds up occupying a larger portion of field, at the expense of competition policy. In short, not enough of the industry gets deregulated on a timely basis. This is in part attributable to the fact that, "regulators have a tendency not only to want to set the rules of the game, but to referee it in a fashion that all participants come out as winners".

The fundamental problem with such an approach is that it is antithetical to competition, which has winners and losers. In this sense, competition is not a "fair" process, at least not from the perspective of the losers. This was recognized long ago by the Supreme Court of Canada, when it observed, with respect to the predecessor of the current Competition Act:

Speaking broadly, the legislation is not aimed at protecting one party to the agreement against stipulations which may be oppressive and unfair as between him and the others; it is aimed at protecting the interest in free competition.

In another old case, it was observed that competition provides:

�� the incentive to struggle which is one of the conditions of advance, not only in the realm of nature, but in the realm of industry. Without that incentive, as the naturalists point out, men would sink into indolence and the more gifted would not be more successful in the battle of life than the less gifted. It follows that there should be open competition for all men and that the most able should not be prevented from succeeding best.

In short, the only "fairness" issue that is relevant in a transition to competition is to ensure that there is a level playing field for all competitors and potential competitors, in the sense that competitors of the deregulated entity (or entities) do not face anticompetitive vertical or horizontal restraints, including exclusionary or predatory behaviour by incumbent dominant firms, or discriminatory conditions with respect to access to essential bottleneck facilities, such as local telecommunications switching networks, electricity transmission grids or distribution networks, slots at airports, rail lines, automated teller networks and pipelines.

�� marketplace innovation cannot be managed. By that, I mean that the notion of introducing just "enough" competition in a marketplace to stimulate innovation is not where we should be going. Even if it were possible, which is not likely, the costs of error are tremendous. Quite simply, the more intense the marketplace rivalry, the greater will be innovation and its resultant economic benefits.

In addition to harming consumers by impeding the ability of new entrants to enter and rapidly expand in a deregulated market, attempts to "manage" the transition to competition can harm consumers by preventing them from obtaining the full benefit of competition from incumbent firms, if those firms are forced to compete "with one arm behind their back" until a regulator decides to free them from their shackles. These types of conditions are often imposed at the behest of new entrants, who have a strong incentive to lobby for asymmetrical industry conditions. For example, by denying incumbent telephone companies the same pricing and marketing flexibility enjoyed by their rivals in the Canadian long distance market, consumers were deprived of the benefits of more vigorous and innovative price and non-price competition.

The bottom line is that competition can be harmed by not clearly circumscribing the powers and responsibilities of the regulator, articulating clear rules (including with respect to the maintenance of confidential information) and committing to as short a transition to competition as possible.

This is not to suggest that every industry is a good candidate for a short transition to competition. In fact, the shirt to competition can be severely impeded by moving too quickly. This can occur by not ensuring that a minimum level of competitive disciplines, together with an effective and vigorously enforced domestic competition law to protect those disciplines, are in place prior to removing regulatory controls over the behaviour of incumbent firms. In short, removing such controls prior to taking whatever steps are necessary (e.g., divestiture, removal of impediments to entry and/or active encouragement of entry) to create competitive conditions, and to protect those conditions (e.g., by active enforcement of an effective domestic competition law), can give rise to the very real risk that incumbent firms will be able to replace the regulatory restraints that have been removed with private restraints, thereby impeding the development of competition and enabling them to exercise significant market power vis a vis customers and suppliers.

��

2. Suggestions for Striking the Right Balance between Competition and Regulation where Ongoing or Transitory Regulation is Maintained

1. General

An evaluation of the goals of competition policy reveals that competition is not an end in itself, rather, it is a means to achieve other goals, including the optimal allocation of resources in society, the optimal introduction of new products and processes and the optimal level of productive efficiency. The pursuit of these objectives in an economy generally increases the standard of living of people in the economy. However, the scope for competition to realize these objectives can be limited, for example, due to market failure, the legitimate need to sacrifice a certain amount of competition in order to attain other public interest objectives, the realities of special interest polities in a democracy, the failure to limit the role of regulators and regulation in industries in transition to competition and by not moving to competition swiftly enough in such industries, or by deregulating too swiftly.

As a general principle, and subject to political realities, competition should be introduced to all activities that are not natural monopolies, to the maximum extent possible. This includes minimizing any restriction or distortion of competition to achieve environmental, social or other public interest objectives. The implications of this general principle are discussed in parts (b) �� (e) below.

Second, existing regulations and laws which adversely impact upon competition should be revisited from the perspective that competition is preferable to regulation, apart from in markets characterized by natural monopolies, such that if a valid justification for continued regulation cannot be articulated, regulation should not be maintained. In short, the onus should not be to demonstrate the benefits of competition. The onus should be to demonstrate the continued need for regulation, or at least the competition impeding aspects of desirable public interest regulation.

Third, the enabling legislation or regulation of the regulatory regime should include, as one of its objectives, the promotion of economic efficiency.

Fourth, an effective domestic competition law must be adopted and a commitment made to its vigorous enforcement by an independent agency established for that purpose. Implicit in this is that the agency be provided with sufficient funding to fulfill its mandate. An effective competition law is critical to ensuring that the benefits of deregulation are not undermined by private anti-competitive conduct, and anything less than strong enforcement of that by an independent competition law enforcement authority can leave significant scope for anti-competitive behaviour to flourish, thereby seriously impeding the shift to competition. As has been pointed out elsewhere:

Competition is the cornerstone of a market economy and competition laws are the main policy tool for maintaining or achieving competition. The theory underlying competition laws is based on the long-standing belief that market forces are, potentially at least, far more powerful guardians of the social welfare than any other form of regulation. Thus, they should be used whenever possible to regulate economic activity.

Although some have suggested that there may be some merit in delaying the introduction of a domestic competition law until a nation reaches a certain stage of economic development, or in delaying the repeal of an industry exemption from the domestic competition law until the industry "gets on its feet" after deregulation, this would simply give former state enterprises and other deregulated firms the opportunity to engage in a broad range of anti-competitive conduct that would seriously impair the development of competition and push back the point in time at which competition is able to deliver the benefits discussed above. To minimize the scope for this to occur, domestic competition laws should be enacted as quickly as possible and contain as few industry and other exemptions as possible.

We recognize that there may initially need to be a number of broadly defined public interest or special interest exemptions from the domestic competition law in order to build the necessary political support for the law and for a commitment to strong enforcement of the law. As with the corresponding exemptions in the laws of many jurisdictions that already have competition laws, we are hopeful that many of those exemptions can be eliminated or narrowed as quickly as possible, as political realities permit. Indeed, one potentially promising prospect for achieving that goal is to seek an agreement in a multi-lateral or pluri-lateral context on either (i) the basic rules for coverage and the rules for exemption; (ii) the initial sectors to be covered by domestic competition laws, with a schedule for adding additional schedules over time; or (iii) specifically identified exemptions, or a specific number of exemptions (which any nation can allocate as it sees fit) from domestic competition laws. As discussed elsewhere, our initial sense is that option (iii) may have the best prospects for achieving practical results in the near to mid-term. We support the efforts of the OECD, CLP and Trade Committee to promote the extension of the scope and coverage of competition laws, i.e., by revisiting the rationale for sectoral exemptions and state enterprise exemptions. This exercise obviously would be more helpful if it were more broadly embraced by non-OECD nations.

Fifth, the domestic competition agency should be permitted to intervene in all regulatory proceedings, make submissions to all regulators, and participate in the process of developing laws or policies that have the potential to impact adversely upon competition, to ensure that competition policy implications of such laws or policies are fully understood. In Canada, the Director of the Competition Bureau has a statutory right to make submissions to all federal regulatory agencies in respect of competition whenever such representations are, or evidence is, relevant to a matter before the agency. He has a similar right to appear before provincial agencies, with the consent of the agency. Over the years, he and his predecessors have helped to substantially narrow the extent to which federal and provincial regulations and other laws restrict competition in various sectors (e.g., transportation, financial services, energy, agriculture, trade, telecommunications), and to assist with the transition to competition in deregulating industries. For example, one observer has noted that where the Director's interventions before the CRTC "focussed (sic) on the anti-competitive effects of specific policies and provisions, they were usually influential, often decisive�� [and that] the Director played a key role in the development of effective procedures and standards for non-discriminatory access" to bottleneck facilities.

In addition to these general suggestions, we have provided below additional suggestions with respect to regulatory regimes pertaining to natural monopolies, other public interest regimes, private interest regimes and transitory regimes.

2. Natural Monopoly Regulatory Regimes

Our first suggestion with respect to these types of regulatory regimes is that they be revisited with a view to reassessing whether the industry or market in question is likely to remain a natural monopoly in that jurisdiction in the foreseeable future. Where the experience in other jurisdictions has demonstrated that the industry or some part of it is no longer a natural monopoly, we strongly suggest that consideration be given to developing a plan to deregulate the industry or the part of it that is no longer a natural monopoly. If some parts of the industry (for example the production of equipment) do not display any natural monopoly characteristics, consideration should be given to removing all artificial barriers to competition into, and otherwise deregulating those aspects of, the industry immediately, or in any event as quickly as possible.

Second, if the market in question continues to display natural monopoly characteristics, consideration should be given to creating competition for the market, for example, by auctioning off the right to be the monopolist supplier. We recognize that this may not be possible in markets where the existing supplier has made substantial infrastructure investments.

Third, cost of service/rate of return and hybrid approaches to the regulation of natural monopolies should be replaced with performance based regulation.

Fourth, where prices in one part of an industry (for example, long distance telephone services) have artificially subsidized prices in another part of the industry (for example, local telephone services), prices in the subsidized part of the industry should be adjusted to reflect their underlying costs. In short, suppliers should recover the costs of each product through the prices charged for that product. This will have the salutary effect of eliminating a strong disincentive to efficient entry by new competitors in the market for the supply of the subsidized product.

Fifth, artificial restrictions on the establishment of affiliates of the regulated entity in upstream, downstream or adjacent unregulated markets should be removed and replaced with measures to ensure that the regulated entity cannot engage in anti-competitive cross-subsidization. At a minimum, these measures should include structural separation between the regulated and non-regulated activities of the regulated entity (i.e., by creating separate affiliates for the purposes of conducting regulated and non-regulated activities), together with a number of complementary measures to ensure that structural separation is in fact maintained. Although this may result in some loss of economies of scope, this cost probably is well worth incurring in order to protect competition and achieve the benefits that it may offer in fragile, emerging markets.

Structural separation is required because accounting and costing rules are not sufficient to ensure that costs are not being misallocated between competitive and regulated activities. As the Director of the Canadian Competition Bureau recently noted a context where the concern was cross-subsidization between vertically related entities: "Without structural separation, a vertically related monopolist may have incentives to discriminate against its competitors in unregulated markets or to shift costs between regulated and unregulated business to give itself a competitive advantage. Such discrimination may be subtle and difficult to detect (for example, it may take the form of delays or reduced quality of service). Shifting of costs may also be complex and hard to detect or prevent." For this reason, in a recent independent review of Canada Post's activities, it was found that even though Canada Post had adopted a rigorous and sophisticated activity based costing system, "there is still considerable latitude to exercise discretion in deciding what should or should not be included in the definition of [long-run] incremental costs" (to be attributed to competitive and regulated activities). Moreover, it was found that "Canada Post has certain common costs which are impossible to attribute to any specific product or product line in a way that everyone will agree is 'accurate'".

In addition to structural separation, various other complementary requirements are necessary because, as the CRTC has noted, "the mere existence of separate corporate charters, under a structural separation approach, would not diminish the incentive and the ability" of the regulated monopolist to cross-subsidize or otherwise confer benefits upon its unregulated affiliate. Moreover, additional requirements are necessary to ensure that it can "be demonstrated from the public record that cross-subsidization �� is not occurring". Complementary requirements typically should include prohibiting the regulated entity and its competitive affiliates from:

1. amalgamating, transferring assets or leasing assets to each other;
2. having an interest in the same facilities or other assets;
3. jointly marketing any services;
4. jointly paying or incurring costs or expenditures;
5. jointly providing any services;
6. using common facilities or assets in the operation of their businesses, on any basis (except where the affiliate requires access to an essential facility, in which case the terms of such assess should respect the general principles outlined below);
7. furnishing to or receiving from one another any administrative, marketing, financial, accounting, technical, research and development or other services or assistance, on any basis (subject to possible exception permitting joint research and development of technology), including but not limited to providing the services of their employees to one another on any basis; and
8. providing any information to one another regarding the other's operations that has not been publicly disclosed.

Of course, it would also be necessary to require the regulated entity to provide sufficient information to the regulator to the regulator to satisfy the regulator that the complementary measures have been respected. This should include a full right of access to the regulated entity's books.

Sixth, where the regulated entity controls network or bottleneck facilities to which third parties must have access in order to compete, the access rules should be clear, transparent and mondiscriminatory. As a basic starting point, access to the essential facilities should be open and the "access price should be based on the cost of providing the service". In most cases, to ensure a level playing field, this should be the long-run incremental costs associated with the provision of the services, although there may be situations in which other definitions of "cost" (e.g., short-run incremental costs, directly attributable costs or fully distributed costs) would be appropriate. In addition, there should be a procedure for resolving disputes between the regulated entity and third parties regarding issues related to access, and that procedure should ensure swift resolution of disputes. Measures should also be adopted to protect the confidentiality of any competitively sensitive information that the regulated entity might otherwise be in a position to learn about its rivals in upstream, downstream or adjacent markets. (This may require the establishment of a separate affiliate through which rivals can place their orders.) In addition, provision should be made to ensure that rivals have adequate notice of changes to essential facilities that may adversely impact upon their competitiveness.

Seventh, where the regulated entity is a state enterprise, consideration should be given to introducing some private capital into the ownership structure, to increase the likelihood that the entity will act in an efficient manner.

1. Other Public Interest Regulatory Regimes

Our first suggestion with respect to other types of public interest regulatory regimes is that they be revisited with a view to determining whether there is still a need for regulation. In addition, where the regulatory regime involves a state enterprise, an assessment should be made of whether there is still a need for the government to supply the product in question. In the words of the Honourable Paul Martin, Canada's Finance Minister: "If the government does not need to run something, it should not��"

Second, if there is still a need for regulation, the competition limiting aspects of that regulation should be revisited with a view to reassessing whether those aspects of the regime are truly necessary to enable the objectives of the regulatory regime to be achieved. In short, the regulatory regime as a whole should minimize the extent to which competition is restricted or distorted in order to achieve the environmental, social or other public interest objectives of the regime. To the extent that aspects of the regime which restrict or distort competition are not necessary to the attainment of the paramount objective of the regime, those aspects should be eliminated as quickly as possible. (See discussion in part (e) below.) In conducting this exercise, it will be important to ascertain whether the regime restricts foreign investment or limits foreign competition or other forms of new entry or expansion, either explicitly or indirectly (for example, through licensing requirements or technical standards that are difficult for foreign or other potential entrants to meet).

2. Special Interest Regulatory Regimes

In the economic area, these types of regimes generally are designed to protect the economic interests of special interest groups which have successfully lobbied politicians to shield them to some extent from competition. As noted in Part II (c) above, the net effect of such regulation is to transfer wealth to the protected special interest groups from consumers, who pay higher prices, and/or potential competitors. Society as a whole suffers in terms of allocative, productive and dynamic efficiency.

The optimal solution would be to eliminate such regulatory regimes completely. However, to the extent that this may not be politically feasible in the short term for many regimes in many jurisdictions, we suggest in the alternative that these regimes be revisited with a view to ascertaining whether their competition restricting or distorting aspects can be reduced over a defined time period that is as short as possible.

3. Transitory Regulatory Regimes

Our first suggestion for increasing competition and efficiency in markets that are in transition from regulation to competition is to minimize the transition period. In some situations, impediments to competition can be removed virtually overnight, whereas in other situations it will be entirely appropriate to move more slowly in order to avoid undermining the paramount objective of achieving conditions that are conducive to the maintenance of long-term competition. In any event, the reliance on market forces should be maximized at every stage of the transition process, and restrictions or other limitations on competition should be removed as soon as they are no longer required for the purposes of the transition to competition. As suggested earlier, every effort should be made to avoid "over managing" the transition to competition.

Second, the transitory regime should be sufficiently flexible to be able to adjust to changing market conditions. For example, if technological advances occur more quickly than anticipated, the timetable for the transition can be reduced. Conversely, if competition is not developing as quickly as hoped, it should be possible to adjust the transition process to take account of that fact.

Third, to the extent possible, regulation of new entrants should be avoided.

Fourth, market based pricing should be adopted as soon as possible.

Fifth, where foreign competition and new entry cannot be relied upon to created an effectively competitive market within a reasonable time frame (say, 2-4 years), consideration should be given to restructuring the industry prior to deregulation.

Sixth, a clear time limit, in the form of a sunset clause in the enabling legislation or regulations of the regime, should be placed on the transition period as a whole or on certain aspects of it. Where clear time limits are not appropriate, clear milestones should be established in the legislation or regulations for either the termination of the transition, the termination of certain aspects of the transition, or the obligatory (as opposed to the permissive) forbearance of the regulator when such milestones are reached. For example, a clear milestone would be that the regulator should forebear, or the entire transition regime be dismantled, when the deregulated entity's market share reaches a certain level (for example, 35-40%). A less clear milestone would be to require forbearance or termination of the transitory regime when "effective competition" has emerged, as defined by an inability of (i) a firm to materially raise prices or materially reduce the non-price benefits of competition, unilaterally, or (ii) several firms to achieve such a result, interdependently. In addition, all market participants should be notified when those milestones have been reached and there should be a fixed period of time (for example two years) within which to review any decision to forebear from regulation.

Seventh, where the milestones approach is adopted, consideration should be given to having someone other than the regulator determine when those milestones have been reached. This will avoid placing the regulator in a conflict of interest of determining when its own mandate should terminate or be narrowed.

Eight, consideration should be given to including "carrots and sticks" in the transitory regime, to provide incentives for the incumbent firm(s) to reach the milestones as quickly as possible.

Ninth, consideration should be given to the extent to which a voluntary code of conduct can eliminate, reduce the need for, or reduce the scope of the regulatory regime. In most cases, an important component of such a code would be oversight by an independent party.

Tenth, a level playing field should be created between the entity that is being deregulated and its new rivals (see, for example, the discussion in Part IV(b) above with respect to cross-subsidization and access principles, and the discussion in PART III (d) with respect to ensuring that the playing field in not tilted against the firm that is being deregulated).

Eleventh, there should be no arbitrary restrictions against bundling of regulated and non-regulated services by the incumbent, so long as those services are also offered to the market on an unbundled basis at prices which a meaningful number of customers are likely to find attractive.

Twelfth, consideration should be given to exploring the extent to which alternative dispute resolution may be used to reduce the role of the regulator in the market or industry.

Finally, to the extent possible, anti-competitive conduct should be dealt with by the domestic competition law, rather than by the regulator, although it may be appropriate to provide for a regulator to address the exercise of certain forms of market power, such as raising prices or reducing service, if the competitive forces in the market are not yet sufficient to discipline such behaviour.