Australian Competition Policy: Deregulation or Reregulation?

PREFACE

In the 1990s, "competition policy" has been at the forefront of microeconomic reform in Australia.  The most prominent dimension of contemporary competition policy involves the introduction of competition to areas from which it was previously absent, in particular much public infrastructure and some of the professions.  Microeconomic reform in Australia has previously promoted competition by withdrawing government regulation, a pioneering example of which occurred in the finance industry.

In contrast, current competition policy proceeds by way of introducing new regulations, particularly over core service features that are thought to constitute natural monopolies.  As well, by declaring an infrastructure facility "essential", the regulators may require its owners to allow potential competitors access to the facility on terms that may be negotiated or, failing that, arbitrated by the Australian Competition and Consumer Commission (ACCC), the most powerful of the regulatory agencies.

A further dimension of competition policy concerns mergers and takeovers.  The ACCC is also the body with regulatory oversight of mergers.  Increased regulation of mergers was in principle agreed to by the government lowering the threshold of possible prohibition of a merger to one where the "market power" of the new entity is likely to result in a "substantial lessening of competition".  The previous test required "market dominance" before the ACCC could block the merger.

This paper is concerned mainly with interventions affecting corporate takeovers or mergers on the one hand, and access arrangements for "essential facilities" on the other.  The underlying rationale of both these sets of regulations is that competitive markets may be impaired by the presence of monopoly.  Markets may not bring efficient production and price levels if they are dominated by firms whose position is protected by factors like large sunk costs or increasing returns to scale.  In such cases of "market failure", government intervention is allegedly justified to ensure that some competition takes place and consumers are not exploited by monopoly producer power.

We are critical of this argument.  We characterise it as a static analysis that proceeds from an unattainable ideal of "perfect competition".  We stress the dynamic nature of markets, which ensures that monopoly power is constantly vulnerable to the innovations of entrepreneurs motivated by the search for temporary super-profits.  Regulatory attempts to correct natural monopoly, we believe, are likely to reduce welfare by dampening the self-correcting nature of markets and by weakening the property rights that underpin the profit motive.  Finally, we set out New Zealand's so-called "light-handed" approach to regulation, and compares it with the current Australian approach.

Much is at stake in the resolution of this argument.  If the critics of current Australian trade practice policy are correct, it threatens microeconomic reform by undermining rather than liberating the sources of wealth creation.  This collection carries the debate to the frontiers of knowledge and analysis of the subject.

FROM PLANNING TO REGULATION:  TOWARDS A NEW DIRIGISME?

INTRODUCTION

For many years I have been a student of what I have elsewhere termed the "dirigiste dogma" (Lal, 1997a).  With the worldwide discrediting of planning after the events of 1989, the dirigiste impulse has not been stifled but merely transformed from planning that sought to supplant the price mechanism to regulation that seeks to supplement it.  The intellectual basis for both sorts of dirigisme is the same.  Dirigistes spanning the political spectrum have rallied around the banner of bureaucratic regulation to correct various forms of alleged market failure.  These relate to externalities, in particular those having to do with the environment and various forms of monopoly.  Having dealt with the former elsewhere (see Lal, 1994, 1995) I shall be mainly concerned with the latter.

This new metamorphosis of the dirigiste beast is one which has plagued the US since World War II:  bureaucratic regulation of the market.

NOTIONS OF COMPETITION

The common intellectual basis for the justifications provided for planning and regulation is linked to a subtle but important shift that has occurred in economists' notion of competition from that advanced by the classical economists from Adam Smith to J.S. Mill.  Modern mainstream economists' intellectual moorings are provided by the so-called Arrow-Debreu theory of general equilibrium, which, it is asserted, gives precision to the classical economists' claims about the virtues of the market (see Arrow and Hahn, 1971:  vi-vii).  But as Blaug (1987:433) points out, one needs to note:

the subtle but nevertheless unmistakable difference in the conception of "competition" before and after the "marginal revolution".  The modern concept of perfect competition, conceived as a market structure in which all producers are price-takers and face perfectly elastic sales curves for their outputs, was born with Cournot in 1838 and is foreign to the classical conception of competition as a process of rivalry in the search for unrealised profit opportunities, whose outcome is uniformity in both the rate of return on capital invested and the prices of identical goods and services but not because producers are incapable of making prices.  In other words, despite a steady tendency throughout the history of economic thought to place the accent on the end-state of competitive equilibrium rather than the process of disequilibrium adjustments leading up to it, this emphasis became remorseless after 1870 or thereabouts, whereas the much looser conception of "free competition" with free but not instantaneous entry to industries is in evidence in the work of Smith, Ricardo, Mill, Marx and of course Marshall and modern Austrians.  For that reason, if for no other, it can be misleading to label classical economics as a species of general equilibrium theory except in the innocuous sense of an awareness that "everything depends on everything else".

It is equally surprising that the "Chicago school", as Kirzner (1994:103) has noted, "maintains that the competitive market economy displays systematic regularities only to the extent that it can be reasonably fitted into the perfectly competitive mold.  Subsequent [to Frank Knight] generations of Chicago theorists would maintain that as a matter of fact the real world competitive market can so be fitted".  Thus self-proclaimed mainstream theorists on both sides of the market-dirigiste divide now use the Arrow-Debreu model as their paradigm.

From this theoretical perspective, the two so-called Fundamental Theorems of Welfare Economics are derived, which, theorists assert, establish the superiority of a market economy (see, for example, Dasgupta, 1980;  Hahn, 1984;  Sen, 1983).  Allegedly, if one or the other condition for the existence of the utopian state of perfect competition is not met, there is "market failure" and therefore a prima facie case for government intervention.  This justification for dirigisme is bizarre (Lal, 1983, 1987).  To compare "competition" in any actual market economy with an unattainable ideal is, to use Demsetz's (1969) useful phrase, a form of "nirvana economics".  It is child's play to show that because of incomplete markets, external effects and the existence of public goods, "market failure" defined as deviations from the perfectly competitive norm is ubiquitous;  but it does not follow that massive corrective public action is needed.

Yet "market failure" was the intellectual rationale of the planned economy.  In a famous debate that occurred in the 1930s, the pro-planners Oskar Lange and A.P. Lerner argued that (a) because of the ubiquitous imperfections in most markets, no market economy could ever in practice attain the utopian norm of perfect competition, and (b) through computations simulating the outcome of a perfectly competitive economy, the planners could compel the production of the resulting quantities of inputs and outputs (or legislate their optimal relative prices).  A planned economy could, thus, achieve nirvana.  Friedrich Hayek and Ludwig von Mises pointed out that, although such a form of planning might be theoretically feasible in a world where information about resources, technology and the myriad actual and possible production processes and tastes of consumers could be costlessly acquired by the central planning authority, in the real world it would be impossible.  The market-based price mechanism is essential because it makes use of the division of knowledge which is unavoidable in any real-world economy (see Hayek, 1935).

The failures of centralised planning are now well known, and one hopes that the events of 1989 have buried the notion of planning.

REGULATING MONOPOLIES

Monopoly and Competition

But now a more ancient debate concerning monopoly and how best to deal with it is likely to promote a new dirigisme.  This was the basis of the vast dirigiste regulatory framework built up over the years in the United States, that supposed beacon of the free market, and was also constructed in Thatcher's Britain.

Socialist thinkers propound the popular view that a market economy will inevitably be dominated by monopolies.  This view continues to resonate, not least in many supposedly market economies.  But is it right?  An important paper by my colleague at the University of California, Harold Demsetz, is useful in setting the record straight.  As he notes, there have been two systems of belief about monopoly.  One view, stemming from Adam Smith, is that monopoly is necessarily underwritten by government action which prevents potential rivals from competing.  The other view is that monopoly arises without government intervention because of the theoretical model of monopoly, which provides an analysis of a case where there is only one firm in an industry in contrast with the atomistic case of perfect competition.  This view in turn has led to the belief that monopoly is significantly correlated with market concentration.  But as Demsetz notes:  "the monopoly model assumes that monopoly power exists, it does not explain how monopoly power is exercised and maintained".  In particular, there is "no good explanation ... provided for how present and potential rivals are kept from competing without some governmentally provided restrictions on competitive activities" (Demsetz, 1988:94).  The usual culprits, economies of scale, indivisibilities of capital, and advertising as sources of barriers to entry are acquitted, while the empirical evidence in support of the view based on Bain's supposed demonstration of a positive correlation between profit rates and measures of market concentration is shown to be at best shaky if not non-existent on the basis of more recent research.

A similar view, that the degree of market concentration does not imply that market prices and outputs will necessarily diverge from the competitive outcome, is also stressed by the recent theory of contestable markets (see Baumol, Panzar, Willig, 1982).  Even with scale economies which limit the number of firms that can service a particular market, as long as potential rivals can contest the "monopoly", the single eventual incumbent's pricing and output policies need not diverge from those under competition.  The only rent such a "monopolist" can acquire is in the form of the sunk costs of firm-specific assets essential for production.

All this suggests that, appearances to the contrary, the old Smithian view that monopolies ultimately depend on government support is valid.  In the absence of such public protection, even in industries where, depending upon "scale economies", for example, only one firm survives, there is no necessary presumption that its behaviour will be monopolistic.

This of course means that regulations designed to increase competition, like antitrust legislation in America, are unnecessary.  Worse, because of the evidence of the capture of the regulatory agencies by the companies being regulated (see the essays in Stigler, 1988), for well-known reasons of political economy, there is the clear danger that such regulations instead of promoting competition create the very government-mediated barriers to entry which nurture monopolies.

The basic reason for this is that efficient economic performance does not depend upon only one type of competition, namely, the imitative output competition emphasised by theories of perfect competition.  Equally important is innovative competition, particularly of the creatively destructive kind emphasised by Joseph Schumpeter.  Whereas, for imitative output competition, efficiency does require a large number of firms, innovative competition most likely does not.  Much innovation has the hallmarks of a race in which the winner takes all.  As Demsetz (1995:139) notes, "the competitive intensity of [such] a contest is not always increased by adding more contestants".  What matters is the quality of the contestants and the size of the prize.  The existence of patents and other devices to prevent imitative competition -- at least for a time -- to allow the winners in innovative competition to secure a big payoff for their innovative effort shows that, in a dynamic market economy, there may be many dimensions of competition.  Given this and the resulting incommensurability of different dimensions of competition relevant for the efficient functioning of a dynamic economy, no single measure of competitiveness, such as market concentration, can be used to judge the dynamic efficiency of an actual market economy.

"Rate of Return" and "Price Cap" Regulations

Nor will "rate of return" or "price cap" regulatory formulae necessarily ensure competition in the large.  For once there are scale economies, prices can no longer equal marginal costs and perfect competition is impossible.  Competition will not be merely imitative but have some of the elements of a contest, in which some agents will lose and others win.  It would be inappropriate to judge the intensity of competition of such a contest by the ex post rate of return of the winner.  For as Demsetz (1995:139) notes, "if one were to gauge competitive intensity by the rate of return on investment made by winners in a lottery game, the rate of return would be quite high, but a negative return is obtained if the calculation includes the wagers made by losers".  So if the rate of return criterion is used to judge the competitiveness of a particular industry, the calculation should ideally also include the costs incurred by those who competed to become incumbents but lost.  If, moreover, the decision on incumbency depends on government favours, then the cost would also have to include the "rent-seeking" costs of all the contestants associated with competing for political favour.  This inclusive rate of return need not be above some competitive norm.  But, in practice, it will be impossible to calculate.

NATURAL MONOPOLIES:  REGULATION OR AUCTIONS?

But what of natural monopolies?  Most infrastructure services have elements of natural monopoly, which is why they were often nationalised.  But, with growing fiscal constraints and the well-known inefficiencies associated with public enterprises, there is a welcome move globally to privatise them.  Will this not inevitably lead to these natural monopolies being used by private producers to exploit consumers?  Should not these utilities therefore be regulated?

"Competition for the Field" vs "Contestable Markets"

The industrial organisation school at UCLA ⁽¹⁾ has provided a distinctive and important answer to this question, which unfortunately is not as well known as the various dirigiste regulatory regimes currently being touted by mainstream theorists.  The basic idea has been labelled "competition for the field" by Harold Demsetz, following a distinction that Edwin Chadwick in the nineteenth century made between it and "competition within the field".

"Competition for the field" differs from the later notion of "contestability", which is concerned with competition between an existing incumbent and potential entrants to the natural monopoly.  By contrast, competition for the field is competition for becoming an incumbent in the first place.  This has important consequences for the price-output configuration and hence the competitive efficiency of the economy.  In the theory of contestable markets it has been shown that, in equilibrium, the only rents the incumbent of a natural monopoly can acquire are the incumbent's sunk costs, associated with the transition costs which a new entrant would have to incur in moving in and out of the monopoly.  If an outsider could enter and exit a market without incurring any transition costs, then the natural monopoly would be perfectly contestable, and, despite economies of scale and scope, the incumbent insider would not be able to garner any rents.  But as there are unlikely to be many natural monopolies in which these transition costs are insubstantial, according to contestability theory insiders would usually be able to extract rents equal to these transition costs from consumers. ⁽²⁾

The situation is very different from the viewpoint of competition for the field.  Here the competition takes place before production begins, with would-be natural monopolists competing for the right to serve the market in which each rival could serve the market at the lowest cost, adopting the best technology.  In this competition for the field, as Demsetz (1968) showed in his famous essay "Why Regulate Utilities?", the potential rents of the natural monopoly would be competed away with the best bid among the rivals being accepted by the community for becoming the incumbent of the natural monopoly.  Thereafter there would be a distinction between insiders and outsiders, and substantial transition costs for the latter -- in sharp contrast with the conclusions of contestability theory.  For without these entry barriers, the potential cost reductions associated with scale economies may not be realised by the successful incumbent.  The frequency of competition for the field, or (equivalently) the length of a franchise to the natural monopoly, will depend upon the particular supply and demand conditions for the output of the natural monopoly.  Also, there is no reason why there should not be contractual conditions attached to the possibility of renegotiation of the terms of the franchise before its expiry.  In fact, given uncertainty on this account, the rivals bidding for the franchise will take account of these renegotiation costs in their bids.  Similarly, if there are likely to be future cost reductions caused by technical progress, which would lead to future rents for the incumbent, these too would be taken into account in the rivals' bids for incumbency if they can be forecast, and the best bid again will involve the whittling away of these potential future rents.

Positive or negative windfalls, which are the result of unavoidable uncertainty, need not be inefficient.  For instance, even in the near-perfect markets for commodities, economic agents suffer positive and negative windfalls all the time, but this does not provide a case for regulation.  However, in the case of natural monopolies, as these windfalls could continue for some considerable period of time, there could be political pressure for their curtailment if they are positive, and the danger of bankruptcy for the incumbent and hence of a disruption of supply if they are negative.  This would provide a case for some renegotiation clause in the contract granting a franchise to a natural monopoly.  But what cannot be laid down is some ideal form of contract.  For given the ubiquitousness of imperfect information and the associated uncertainty, agents can only search for the best available mutually advantageous contract.  In Hayek's felicitous phrase, the market is par excellence "a discovery process".

Game Theory

In contrast with this UCLA view on regulation of natural monopolies, there is an emerging technocratic view.  This is based on the frail framework of non-co-operative game theory. ⁽³⁾  As the leading lights of game theory recognise, it is of very limited practical relevance because of the plethora of Nash equilibria which can be generated (Binmore, 1990;  Kreps, 1990).  Though of use in training the intellectual muscles of the young, in my view it has not as yet yielded any robust policy-relevant results. ⁽⁴⁾

On Privatising Infrastructure Services

How then in practice should the current and future provision of infrastructural services in electricity, natural gas, water, sewerage, roads, telecommunications, be dealt with?  Though there are some important differences between these different "utilities", they have one common feature.  The natural monopoly element in their provision consists essentially of the "networks" they use to "ship" their products.  They provide common "transportation" facilities for all possible users rather than being dedicated to individual ones. ⁽⁵⁾  Thus an electricity grid, a gas pipeline, a system of telephone lines, water and sewage pipelines, railway track and of course roads are "networks".  All other aspects of the provision of the services of these utilities can be made competitive by allowing multiple users of these networks to service consumers.

Consider the provision of electricity or gas.  There are three stages, each of which should ideally have a separate company:  the production stage, the transmission stage through the common "network", and, finally, the distribution stage to consumers.  There is no reason why the first and last of these stages should not be competitive.  If rival firms are free to produce electricity as they see fit, and to service users on the common "network", there is no intrinsic reason why the production and distribution of electricity need require regulation.  It can be produced and distributed like any other commodity by competing firms. ⁽⁶⁾

This leaves the "common" network.  Here there are two choices.  The first is for it to be communally owned and financed through taxation, but built and run through a franchise given to the bidder who offers to build and supply the network and its services at lowest cost to users.  The services of the network would then be available to any user at a fixed fee, or, if administrative costs are high, could be free.  This is the solution adopted for most public roads in many countries.

The second is a purely privately financed alternative.  Consider electricity.  First, each regional grid is set up as an independent private time-bound franchise.  This franchise is then auctioned to the bidder who offers to execute the quantity-quality terms of the franchise at the lowest cost to users during the fixed period of the franchise.  The bidder who bids most for the existing grid whilst meeting the other franchise conditions obtains the franchise for the stipulated period.

At the end of the franchise there are two options.  One is for the grid to return to the "community" which then auctions a new franchise for the grid as before.  This reversal of the assets in the expanded "network" to the community has been adopted by China in its foreign direct investment projects.

The other option is for the incumbent of the grid to obtain the highest price anyone is willing to pay for the grid, subject to the new price-quality and expansion conditions.  Of course, the incumbent would also be able to participate in the bidding procedure.

There are a number of reasons for preferring this latter option rather than have the network's capitalised value revert to the community at the end of the franchise.  As can readily be shown, in the second form of contract where the incumbent recoups the capitalised value of the grid from the highest bidder for the new franchise, the price charged to users of the network, and hence the price to final consumers, will be lower than with the first option where the grid reverts back to the community.  Of course, what the consumer gains through lower prices, he loses through the loss of tax revenue which would accrue if the grid reverted back to the community.  But if, for reasons of what may cryptically be called political economy, the social value of a dollar of tax revenue is less than one dollar, in contrast to what I believed in my misguided youth, ⁽⁷⁾ consumers may be better off getting their dissipation of the potential rents from the natural monopoly through a reduction in prices than through the government budget.

A second reason for preferring the second option where the incumbent "sells off" the grid to the highest bidder after the end of his franchise is that this reduces the time inconsistency in his investment decisions which could arise with the other option of the grid reverting to the community.  For in this latter case, the incumbent would have an incentive to underinvest in both maintenance and expansion towards the end of his incumbency, and thus run down the assets of the natural monopoly.  This would be avoided if he could obtain the capitalised value at the end of his incumbency of the assets he bought, maintained and created during his franchise.

Finally, as the incumbent will usually be a private firm, the trading of its shares on the stock market would permit takeovers by other private firms, which could prevent any monopoly developing on the networks even during the franchise period, as has so often happened under regulation.

Now move to the next stage:  the transmission of electricity at the intra-regional or local level.  The same scheme would be applicable.  Intra-regional or local franchises would be set up and auctioned on the same principle as the regional grid.

With any user of the "network" having access to it at the fixed fee determined in the auction for the "network", any company could set itself up, without any government regulation, to sell electricity to consumers.  With the generation of power privatised, these distributing companies would be able to purchase electricity from the cheapest source given the varying demands for power.  In fact, as has happened in the UK, a spot market for delivery of power by competing generators would develop.  These generators could also end up specialising, with some finding it profitable to provide base-load and others peak-load power.  There would be no need for government intervention of any sort in either the production or distribution of power.

Similar schemes can be set up for all the other infrastructural services, which do not therefore need to be funded from tax revenues.  This would also prevent the regulatory jungle and rent-seeking that the botched privatisation of utilities in the UK has promoted (see Robinson, 1993;  Beesley, 1994).

Finally, it may be noted that in many countries the contracting out of the provision of many local public services such as garbage collection, which was pioneered in the UK, is now growing.

CONCLUSION

Planning was promoted by an articulate clerisy.  Another form of dirigisme promoted by the current technocratic "best and the brightest" could once again blight the economic prospects of many countries.  In this paper I have attempted to show, first that with so much contemporary theory, in Peter Bauer's sage words, "the emperor's new clothes are of the finest hue but there is no emperor within" (1987:36);  and second, that, as regards the problem of natural monopolies in the provision of infrastructural services, there are simple ways to avoid the new dirigisme of regulation which, as in the past, the siren voices of the clerisy are promoting.

REGULATING MERGERS AND ACCESS TO ESSENTIAL FACILITIES

REGULATION AND SUSTAINED ECONOMIC PERFORMANCE

Many areas of government intervention in business decisions have been much reduced over recent years.  "Social" regulation covering standards, pollution and the like has tended to mount;  but the "economic" regulation of businesses through tariffs and subsidies, and directions to offer services has been much reduced.

Two areas where "economic" regulation has been increased, normally on the grounds that it is necessary to promote competition, are access to monopoly services and control over mergers.

Over the longer term, successful economic performance requires market competition with established property rights.  Competition means a ceaseless striving to steal a march on rivals by cost-cutting and better pleasing the customer.  Established, secure property and contract rights offer the incentive of personal gain from searching out new and changing needs of consumers and continuously seeking ways to meet these more cheaply.  The socialist economies collapsed under the weight of bureaucratic controls and lack of incentives that are the inevitable corollary of attempts to improve upon the outcomes of market processes.

Stable institutions with the ultimate backing of law are essential to sustained growth in living standards.  Government intervention, whether through owning businesses, directing resources into favoured areas, or reviewing commercial decisions will detract from and possibly arrest this process.

At issue is the scope of government-backed rule of law.  As stable property rights and competition are the twin engines of prosperity, is there a role for government to intervene to promote competition?  If such a role exists, it is necessary:

• to examine where there might be a conflict between the exercise of private property rights and the assurances of workable competition;
• to determine, if such a conflict occurs, the criteria for overriding property rights;  and
• to establish institutional arrangements that generate the minimum of waste and paperburden costs in addressing the issue.

While competition undoubtedly generates increased efficiency, will government intervention select the appropriate competitive model?  And will it lead to the diversion of entrepreneurial energies into avenues that are unproductive?

There are two areas where well-intentioned regulatory measures could frustrate the efficiency of the economy.  The first concerns the conditions required for access to "essential" facilities:  wires, pipes, roads and ports.  The second concerns conditions placed on businesses wishing to merge.  The latter are of two forms:  those which are publicly specified and those which are negotiated with the regulatory authorities and agreed to under the duress firms face in having to consummate a deal quickly or see the value of the merger reduced.

COMPETITION REFORM IN AUSTRALIA

Breaking-up Government-Owned or Government-Mandated Monopolies

With the reforms to competition policy introduced in 1995, previously sheltered industries in Australia have experienced a massive increase in productivity.  Gas and electricity have doubled their labour productivity.  Part of this has been attributable to privatisation, but a major part was due to corporatisation, which placed the businesses themselves on a footing akin to private ownership.

Much of the stimulus to the productivity improvements that have taken place is directly attributable to the reform of competition laws, in particular Part IIIA of the Trade Practices Act.  This has required government businesses to act in a competitive manner and so has placed pressure on them to reduce costs.

At the heart of the competition policy reforms is the requirement that infrastructure -- rail, telecos, gas and electricity lines, port facilities -- is opened to all suppliers.  In the final analysis, this access is established at a price and subject to conditions that government bodies deem fair and reasonable.

There are several such bodies.  The National Competition Council (NCC) determines whether a facility should be "declared" essential as an open access facility.  The Australian Competition and Consumer Commission (ACCC) supervises the rules under which access is granted and can provide operators of these facilities immunity from legal challenge if it can be satisfied that the terms offered are reasonable.  State bodies like the Independent Pricing and Regulatory Tribunal (IPART) in New South Wales and the Office of the Regulator-General (ORG) in Victoria control pricing within those two States.  Industry-specific bodies, including the National Electricity Code Administrator (NECA) and the National Electricity Market Management Company (NEMMCO), operate the market rules for electricity under the broad direction of the ACCC.

The reforms that preceded the creation of these bodies have brought an unlocking of productivity gains.  These "essential facilities" over which regulatory control is maintained comprise a considerable part of the total product, in the case of electricity some 40 per cent of the final product cost.  They also have a considerable bearing on the remaining 60 per cent, since new lines linking existing supply sources are often an alternative to increased generation.

The Control of Natural Monopolies or Essential Facilities

The notion of essential facilities is both elusive and subject to change.  In the US, the regulatory authorities have ceased controlling AT&T as it has become clear that there is ample competition in what was once thought to be an archetypal essential facility.  By contrast, in Australia the ACCC is urging Telstra to offer access to its network on terms it does not favour.

The most entrenched monopolies -- perhaps the only ones with durability -- are those supported by government.  The central purpose of the Australian competition reforms was to smash these.  In part, this meant hiving off the clearly contestable part (for example, generation of electricity).  What is left is a set of residual apparent monopolies covering wires, pipes, ports and roads.  Most of the recent policy debate has focused on the price and access conditions for the use of these facilities.

With perfect knowledge, the wise and incorruptible bureaucrat could devise a transmission system that would prevent monopolistic waste and could also bring about a great many of the dynamic gains achieved with commercial rivalry.  But these conditions are not present.  Producers and carriers will be reluctant to reveal to competitors and customers alike the extent of their costs;  and they too have imperfect knowledge of these.  Buyers will seek to keep options open to the maximum degree, and will not reveal the full extent of their demand, their alternative means of having it supplied and their preferred means of supply.

While government is at liberty to insist on certain access rules for its own facilities, it must be careful not to impose these on private facilities that are already in existence and that were built under different contractual arrangements.

Requiring private firms to grant access to competitors or others will diminish their incentive to build a facility or will distort the nature of that facility if built.  Well-established property rights are, however, consistent with open access at a commercial price.  For example, it is generally agreed that patents which assign ownership to ideas and products encourage innovation for the benefit of all.  Access of such ideas and products can then be arranged under mutually accepted licensing agreements.

The requirement for open access to these facilities might have perverse effects both on the competitive process itself and on economic efficiency.  Some entrepreneurs will want to have greater control of the sources of supply and the throughput of the facility than would be permitted under open access.  Either they would build a suboptimal facility so that only their own booked capacity is transmitted, or they may be discouraged from building any facility, with consequent loss of additional supply to the market.

Using Competition to Bring About Optimal Access Conditions

With regard to wires, ports and pipelines, we can say with some certainty that the lowest consumer prices and the optimum production rates will be achieved if there are many customers competing for the capacity and many independently-owned producers vying to supply those customers.

The Hilmer recommendations (Hilmer et al., 1993) rightly focus upon the importance of competition in bringing about a more efficient and productive economy.  Government should do everything possible to prevent its own agencies and institutions from inhibiting this process.  This means abandoning exploitative monopolies in the form of utilities and outlawing procedures that create barriers to commercial entrants or prevent the full force of competition.

The codes covering access and pricing to gas and electricity networks are subject to requirements on price and access which presume that they are monopolies.  Yet recent events have demonstrated the potential for active competition in this area of supply.  In Victoria, rival electricity distributors are planning to drive new lines into each other's territory.  Further evidence of the potential is observable in the skill that AGL has shown over many years in setting its New South Wales pipeline charges at a level that allows it profitably to ward off rival facilities.  AGL has responded to competitive threats by reducing prices in areas where those threats have greatest potential.

The nightmare for a utility business is that if it adopts too hard-nosed an approach to pricing and service, it will call forth competition and leave the existing asset "stranded".  Fear of having "stranded" assets means that little by-pass is likely to eventuate.  But the control over excess prices that competition brings does not require that a competitor physically emerges.  Contestability for the market is quite adequate.

Competition or contestability is much superior to regulation.  Indeed, the regulator's role is to make judgements that, in his view, correspond to those that would emerge in a competitive market.  The problem with a regulated price is that it is likely to bring distortions.  If set too high, and the facility is indeed a monopoly, excessive prices will shift customers towards activities and expenditures that offer less value than would be the case with market-determined prices.  Of course, if the facility is not a genuine monopoly, prices set too high are irrelevant because competition will force them down to market-determined levels.  If prices are set too low, competition will be pre-empted and the facility owner will have inadequate incentive to properly maintain and expand the system.

A well-structured competitive regime on wires and pipes would ensure that all barriers to entry were removed and that the incumbent businesses were unable to block new rivals.  The New Zealand regime operates very successfully in this way without any price regulation (the breakdown of electricity supply into Auckland in early 1998 was unrelated to this feature).

If any price cap is to be set, it should apply only to those areas that were connected at a subsidised rate for community service reasons.  Ideally, a subsidy should be paid from general revenue but it could be made a condition of sale where the facility is privatised (with the government therefore accepting a lower sale price).  For the rest, the initial price should be established as if contracted at present levels, with downward pressure on incumbents' prices provided by rival suppliers.  After all, customers have willingly connected at the present price and suffer no disadvantage from the status quo.

No facility -- at least no facility unprotected by government franchise -- has untempered monopoly powers.  Many facilities can be by-passed and almost all others supply products, like gas, that compete with electricity.  That facilities have an element of natural monopoly is not cause of itself for the suppression of property rights.  Nor is it incompatible with the concept of access for others' product.  Where excess capacity exists, access can be marketed at a price which reflects the tremendous level of capital and expertise necessary to construct a large-scale system.  The alternative to a market based on the assignment of property rights may be significant under-investment in the gas sector, at a significant loss to suppliers and consumers.

Auctioning the Right to a Monopoly Facility

Should it be determined that only one pipeline facility is economic and there are many suitors for that pipeline, one option is to auction its approval.  The selected supplier is the one offering the lowest charges or other benefits.

This is an option favoured by the Gas Code.  While more market oriented than most options, there is seldom a situation where several suitors would be available unless the opportunity to build the pipeline has been previously suppressed by government regulation.  And if a proposal is to be opened to tender by those who did not spot the opportunity in the first place, the incentive to search for new ways to profitably meet consumer requirements will be reduced.

CONTROL OF MERGERS

The Regulatory Framework

The other arm of competition policy covers takeovers and mergers.  These fall under the control of the ACCC.

The form of natural monopoly that was the harbinger of contemporary merger laws was monopoly obtained by entrepreneurial excellence.  The US Sherman Act 1896 was targeted against the Rockefeller dominance of kerosene as much as against railroads.  In the case of the former, the monopoly lasted for only a few years before new oil finds in Texas and Iran brought a great many new competitors.

The pattern of constant technological change and increasing openness that characterises the world economy calls into question regulations that combat merger activity.  Markets are constantly redefining themselves, with market players spilling over into areas that would once have been considered unrelated diversifications.

Earlier concerns that a business could obtain a stranglehold on a market have been blunted by the ability of rivals to move into each other's territories and international firms to offer services once the preserve of locally domiciled concerns.  These same pressures are forcing suppliers to look at economies in production, marketing and systems to obtain savings so that they can better compete.

Draft Merger Guidelines were published by the ACCC's predecessor, the Trade Practices Commission, in November 1992 (TPC, 1992) following a change in the law that widened the ambit of oversight from one that focused on dominance to one that looked more closely at the ability of merged entities to exercise any form of market power.  The Draft Merger Guidelines specify "safe harbours" where mergers can proceed without ACCC scrutiny.  They also outline the criteria and procedures necessary to obtain ACCC clearance of mergers and takeovers.  The Industry Commission (1996) considered these Draft Merger Guidelines to be too onerous, time-consuming, excessively focused on consumer price benefits and likely to inhibit efficiency-enhancing mergers.

In performing its duties, the ACCC must be alive to the twin building blocks of efficiency:  competition and property rights.  It must be conscious of the incentive which firms have to seek higher profits -- economic "rent" to some -- and that such activity is harmful only where there is a monopoly that cannot be challenged.  Monopolies of this nature are very rare indeed.  It used to be said that IBM was in such a position but the US Justice Department abandoned its decade-long litigation against the firm only when it was clear that its monopoly had disappeared under the avalanche of new rivals.  As the Industry Commission (1996) points out, even firms in a dominant position must behave as though they are subject to competition as long as the market is open to competitors.

Some Issues in Australian Controls of Mergers

Although fewer than 10 per cent of the 150-or-so mergers examined each year are opposed in a formal sense, others "voluntarily" change the terms and conditions from those initially preferred.  Still other possible acquisitions are not contemplated due to the known hostility of the ACCC (Industry Commission, 1996:15).  In part, these perceptions stem from some high-profile decisions of the ACCC that have indicated a forceful ideological position regarding mergers.  The ACCC has tended to oppose mergers even where strong efficiencies have been documented and the reduction in competitive pressures would have been minor;  in addition, the ACCC has used its authorisation powers over mergers to force their proponents to provide rivals with greater competitive opportunities.  Its approach to three recent merger proposals is examined below.

Westpac/Bank of Melbourne

The ACCC received a lengthy submission seeking clearance for a merger between the two banks on 15 April 1997, some twelve days after the merger proposal was announced.  It announced its decision on 25 July 1997.  During this time the customer uncertainty at the Bank of Melbourne (BML) resulted in a considerable loss of business and the two entities were pressed to agree to the Commission's requirements.

In its examination of the proposed merger, the ACCC addressed six features of the banking market:  deposits, home loans, personal loans, small business banking, credit cards, and transaction accounts.  Each of these market segments was addressed from the ACCC perspective of whether the segment was a State or national market.  The criteria used were the ACCC "safe harbour" rules which allow a merger to proceed if:

• the merged entity comprises less than 40 per cent of the market;  and
• the combined market power of the four largest firms is less than 75 per cent of the market or where the four largest firms had more than 75 per cent of the market, the merged firm had less than 15 per cent.

In terms of overall matters, the existence of four major banks and a great number of smaller entities competing in the various segments should have been assurance enough of continued robust competitive conditions, as the Bank of Melbourne's market share was less than 2 per cent.  In different segments, however, especially on a State basis, that share would be much more significant.

By considering the market as State-based, the ACCC managed to define two of the six segments as crossing its threshold.  Thus, although the two entities each only had 9 per cent of the Victorian deposit market, this was sufficient to cause "concern".  Similarly, although the merged entity had only 20 per cent of the Victorian transaction accounts, "concern" was again triggered.

On the basis of these findings, the ACCC obtained section 87 undertakings, which included:

• maintaining significant local decision-making autonomy;
• maintaining BML's extended trading hours;
• preserving the entitlement of existing transaction account customers to fee exemptions (subject to certain qualifications);  and
• granting access to their electronic networks by new and small Victorian competitors for their Victorian customers for a reasonable period.

These measures represent an imposition on the parties of the ACCC's preferred competitive paradigm.  They cannot be said to represent the outcome of normal market processes.

It is hardly the role of the ACCC to dictate to businesses how they should manage their hierarchies.  Requiring local autonomy might offer some comfort for the Melbourne-based staff of BML but management decision-making is surely the function of the firm's executive.  It is not apparent that the ACCC has expertise in corporate organisational arrangements and would surely not claim such qualifications in respect of a commercial organisation.

It is also not clear why the ACCC should require the former BML branches to remain open on Saturday mornings.  Longer banking hours offer benefits to consumers but these are traded off by business organisations conscious of costs.  It might have been no less appropriate for the undertaking to require all the merged branches to open on a Saturday morning or open for even longer hours.

Similarly, requiring the entity to maintain particular conditions for small depositors is an unfortunate intervention in business activities.  Such decisions should be based on commercial considerations.  If the ACCC thinks them to be appropriate it should require all financial institutions to follow them.

Perhaps the most egregious undertaking is that which requires the payment system to be opened to other competitors on terms that are not considered commercial by the owners.  On one plane, payment systems so opened allow more competitors and bring lower costs.  But there are considerable investments in devising such systems.  If owners of a successful system are forced to make it available to competitors, this reduces the incentives to take the risk of building it in the first place.  Firms will await the activities of their competitors and free ride (or cheap ride) on systems that turn out to obtain strong market acceptance.  Innovation will be the casualty.  In this and other respects, the ACCC makes the mistake of assuming a system is in place and then seeking that its owners allow full use of it by others.  The short-term gain to competition is a long-term loss to incentive.

Ampol/Caltex merger

In its consideration of this merger, the ACCC forced the parties to offer undertakings on:

• sale of certain terminals in major cities;  this provision was made even though there were already terminals in Sydney, Melbourne, Brisbane and Perth independently-owned by major businesses and the cost of new terminal construction is only $10-20 million;
• to supply at least a billion litres of petrol per annum on reasonable commercial terms to independents;  and
• to sell at least 35 metropolitan and 15 country sites.

There were concerns ⁽⁸⁾ expressed about:

• the high level of concentration in refining;
• refinery exchange agreements (swaps) and borrow-and-loan arrangements between the majors that allowed each a presence in States where they have no refinery capacity;
• vertical integration of the majors;  and
• the subsidisation by the majors of low retail margins through high refining margins.

High concentration.  The ACCC has recognised that Australian refineries are old and under-sized.  Minimum efficient size is thought to be at least 200,000 barrels per day and may be closer to 300,000.  The largest Australian refinery is 120,000 barrels per day and the total demand would accommodate little more than one world-scale refinery.  In fact, the concentration in refining is clearly too low for an efficient industry.  This is a matter that has recently been highlighted by rumoured merger talks between Shell and another major.

Refinery exchanges and borrow-and-loan arrangements.  The Commission's objection to these seems to be twofold.  First, they fuel suspicion of a cartelised industry.  Second, they are not normally available to businesses other than those with refining capacity in Australia.

As for the first of these objections, it would seem that if a cartel were operating, it has been highly ineffective.  The profits of the oil industry are very low.  Moreover, the classic signs of a cartel -- stable market shares -- are absent.  The borrow-and-loan arrangements have been acknowledged by the ACCC's consultant, C.E. Hyde, as contributing to greater competition by allowing a presence in each market (ACCC, 1996a).

As for the second objection, the rationale behind the co-operative arrangements needs to be understood.  If the swaps were to be arranged through the financial system they would attract a taxation penalty.  Firms without an ability to offer reciprocal arrangements offer no advantage to those with refining capacity.

Vertical integration.  The decision on whether firms seek to control their production through to retailing is one based on their views of efficiency.  Firms that adopt such an approach often see benefits of ensuring a consistent consumer image of the firm, to allow more effective integration of promotional campaigns, and so forth.  Commonly, such strategies are combined with franchising to fuse the overall policy with a greater enterprise which is often shown by small business.

The majors' subsidisation of low retail margins through high refining margins and strategies to constrain independent operators.  The ACCC has apparently come to a view that the majors act in collusion to pursue particular strategies designed to squeeze out opposition.  Professor Allan Fels, Chairman of the ACCC, stated that the proposed merger would result in a substantial lessening of competition, would bring "unilateral" price increases and facilitate "tacit and possibly explicit collusion", and higher retail prices.  He went on to state that there was little if any prospect of independents supplying petrol, ⁽⁹⁾ an astonishing statement since it was made at a time when independents (mainly supermarkets) were making massive incursions into many markets across the world.  The subsequent entry of Woolworths into the Australian market was claimed as an outcome of the ACCC's market engineering.

The ACCC justifies its position in the petroleum industry by (unsubstantiated) claims of collusion.  Its determination concerned a market where there are clear inefficiencies caused by inadequate scale, a long history of price regulation and regulation of vertical integration, and very low returns on investment.  It is also a market that has seen the exit of some of the world's largest businesses as a result of poor profits.

The subsequent entry of independents using imported product was greatly facilitated by the regulatory arrangements forced on the incumbents, in particular the limitations on ownership of retailers, a limitation exacerbated by the ACCC determination, and by requirements to supply competitors on terms and conditions not necessarily to the liking of the refiners.

The ACCC in effect tilted the playing field in favour of new competitors which did not even have a claim to legitimacy for such treatment on the grounds that they are small businesses.

Australis/Foxtel

On 14 October 1997, the ACCC announced that it would block the proposed merger between Foxtel and Australis Media.  Australis Media had secured access to a key library of Hollywood movies but had done so at a high price.  It had been attempting to merge with Foxtel (a joint Telstra/Murdoch business) for the previous two years but the ACCC had prevented this.

The ACCC claimed that the merger would have given the entity a high market share.  The outcome of the decision was to place Australis Media in bankruptcy.  The loss of a significant market player occurred in any event.  But the shareholders in the company and what the Chairman of the ACCC referred to in a press release as American junk bond holders (who had $US375 million at stake) sustained losses.

The Chairman of the ACCC was reported as saying it was not the ACCC that caused Australis to have problems but the fact that it paid $184 million to acquire pay-television licences and extremely high prices for their film material.  Professor Fels was reported as saying (Australian, 6 May 1998) that with a combined Foxtel/Australis "there was a real possibility that Optus (the main rival) would have disappeared from the scene".

The upshot is that the ACCC made a determination that the structure of the market required it to take action to ensure that one player, Optus, was not swamped by another.  Determining that specific firms should continue in business is a very bold decision for a regulator to make.  Sacrificing the interests of one set of shareholders to those of another requires the wisdom of Solomon.  Moreover, the market for pay-TV is one with a great deal of fluidity.  The content competes with other outlets (free-to-air TV, video stores and the general entertainment industry) and is characterised by very rapid market and technological change (it is likely that means of fast downloading film material through the Internet will provide meaningful competition before any of the present participants are operating profitably).

Different Policy Approaches to Mergers

Competition is not an end in itself:  it is valued because economic experience tells us that, as a rule, competition is the best way to maximise the community's welfare through enhanced efficiency.  Mergers allow economies to be made.  They allow sharing of production and marketing overheads and the rationalisation of facilities.  They therefore make possible improved efficiency.

In the US, antitrust activity has undergone a transformation.  In the past, the focus was on firms that dominated industries, although some decisions shifted from the targeting of firms that dominate industries to one that examines the likely impact on consumers.  Many merger proposals have been disallowed which have covered some very minor parts of an industry.

In general, the US approach has focused on the degree of concentration in markets.  A measure of market power, the Hirfindahl-Hirschman index, first applied by the United States Department of Justice following its 1986 report on gas pipelines (see Laine, 1995) has been used.  This takes as a proxy the existence of four similarly sized firms as providing a low risk of the exercise of market power.  The ACCC's "safe harbour" approach mirrors this.

However, this approach was largely abandoned soon after its adoption.  The US regulatory authorities took a relaxed view of all types of mergers following the growth in the influence of contestability theory which emphasised that any market power was likely to be transitory because it would attract competitors.  Such an approach is favoured in this paper.

More recently there has been a further development.  The 1998 Economic Report of the President includes a major chapter on merger policy that indicates a more selective and possibly more interventionist approach.  An article, "The Economics of Antitrust", in The Economist (2 May 1998) offers some insights into why this may be so.  The article points out that markets such as that for soft drinks, dominated by just two firms, are intensely competitive, while firms in others where there are apparent high degrees of rivalry and few entry barriers can be interpreted to exhibit market power.

In line with the Economic Report of the President, the US Department of Justice has now moved to a model that examines the case for blocking mergers on the basis of the likely effects on price.  By examining scanner-based price information in markets where two competitors are close substitutes, and comparing the outcomes in regions where they both compete with outcomes where that competition is absent, the likely immediate effects on consumers can be seen.  In one case, concerning the merger of two firms that were market leaders in white bread, the merger was disallowed on these grounds.  In another case, Bell Atlantic and Nynex, a merger was permitted to proceed even though the combined firm would have a dominant share in a major market because an examination of the likely actions of rival firms indicated that the dominance would be subject to challenge.

While the "safe harbour" basis of action adopted by the ACCC is clearly highly inflexible (and recognised as such by the ACCC, which supplements it with other criteria) its replacement by an approach which examines likely price outcomes is also unsatisfactory.

Regulatory measures to forestall price increases in markets may detract from efficiency.  Where two firms are engaged in cut-throat competition, the outcome will often be that one is driven out of business or exits the particular market segment.  A merger of two such firms can avoid this outcome and restore normal profits in a less socially-wasteful manner.  Accordingly, it is incorrect to oppose a rationalisation of two business entities on the grounds that this might bring increased prices.  If lower prices are merely the result of a continued existence of surplus labour and capital forcibly retained within a production facility, improved resource allocation can be brought about by their shift to other activities.  In this sense, preventing the industry rationalisation is akin to imposing a high tariff to arrest a domestic industry contraction.  Although a tariff is designed to bring increased prices, and measures to prevent a rationalisation are designed to bring reduced prices, both can have the effect of establishing a larger industry than economic fundamentals would prescribe.

Hence, in effect, actions to prevent rationalisations, especially where there are few entry barriers, are actions to force an enlarged industry and lower prices.  With respect to the latter, it is generally recognised that imposing price reductions on competitive businesses (or foreclosing price increases) will detract from efficiency:  for example, rent control has been demonstrated to result in a reduction in rentable properties.  Similarly, preventing a merger on the grounds that prices would subsequently rise seeks to lock firms into unsustainable positions, forcing them to treat their capital and marketing assets as sunk.  It is, in short, a veiled form of price control.

CONCLUDING COMMENTS

Deregulation has been accompanied by a considerable reregulation of certain aspects of business.  This process may be leading us into areas of inefficient resource allocation.

The most significant anti-competitive activity we need to fear in the medium term is that arising from government-protected entities.  The most important competition policy issue is therefore the elimination of all (unjustified) legal restrictions and interventions in the market which give certain entities (whether wittingly or not) supernormal profits protected from the otherwise inevitable erosion of entry or threat of entry by other firms.  At the very least, there needs to be a critical examination of specific cases and an exit from oversight where competition or contestability is possible.

Enforcing open access on transport/transmission facilities and setting a regulated price on that access are likely to curtail the incentive to build new ones and give inadequate incentives for the maintenance of existing ones.  Other than where some NIMBY interest prevents any possibility of rival provision -- and it is difficult to envisage cases other than some airports that might fall into this category -- the authorities should take the view that the facility is contestable.  Regulation should be confined to requiring a price level that has been set in place by the de facto contract that prevailed prior to competition being allowed.  Competition should then be left to erode those prices that are excessive.

This treatment of mergers and competition policy generally derives from two basic precepts:

• that competition brings dividends in terms of forcing increases in (allocative, productive and dynamic) efficiency and
• that individual property rights protected by a stable legal system are essential if firms and individuals are to have the incentive to seek out gains from trade and innovation.

There are differences of view about the need for merger and competition commissions and the powers of those commissions.  These stem from a divergent emphasis on the importance of, on the one hand, the capabilities of regulatory oversight to provide superior outcomes, and, on the other, the power of markets to automatically correct serious deviations from a stereotype of "perfect competition".  This spontaneous correction is caused by the attraction of third parties to profitable opportunities.

As Deepak Lal (1997b) points out, the notion of a market equilibrium whereby there would be no economic rents or excess profits stemming from "market failure" is relatively modern.  There are myriad situations where firms extract higher profits from serving particular groups of customers or where they are able to price-discriminate to squeeze out more profits.  In one sense, these activities can be attributed to some degree of market power.  But to attempt to dismantle all such hurdles in pursuit of the chimera of perfect competition would extend the role of the regulatory authorities into requiring one price only to be charged by hairdressers, airlines and building owners.  In doing so and denying sellers the right to exploit different market segments resulting from differing demand profiles, the regulatory authorities would require that some demand that could profitably be met goes unsatisfied.

Workable competition is the best we can ever achieve and, indeed, it is the lure of earning supernormal profits that provides the stimulus to innovation and optimum usage of labour and equipment.  If mergers result in improved profits, which are their proponents' clear goal, this will often mean higher prices.  Attempting to ensure that all players remain in an industry means it is ossified, frozen in a time when production and market characteristics allowed more players than can profitably compete at present.

The one unambiguous role for government is to ensure that its laws do not prevent new players from contesting a particular set of demands.  Beyond that, the attraction of profitable opportunities and the desire of consumers to obtain good value will conspire to prevent sustained price gouging on the part of suppliers.

TAKING DYNAMICS SERIOUSLY IN COMPETITION REGULATION

INTRODUCTION

The rhetoric that accompanied the appearance of the Hilmer Report (Hilmer et al., 1993) suggested that the document would herald a new era of efficiency and dynamism through the medium of a national competition policy.  Although not a document that expounded privatisation of public utilities, the Hilmer Report, along with a renewed emphasis by State and federal governments on economic performance, has encouraged a widespread and serious questioning of the role of government in service delivery and the scope for wealth-improving economic regulation.  The policy shifts that have occurred reflect a general awareness on the part of governments and policy-makers of the need for greater reliance on market mechanisms, rather than on the traditional statist mechanisms, in the provision of infrastructure and similar services.

The general trend in favour of markets is the result of an increased recognition that many government-controlled industries have now become sites of rapid technological change and innovation.  The flexibility, dynamism and rewards for risk-taking behaviour which the unimpeded market permits are recognised as providing the better institutional framework for such fast-changing industries.  Alternatively, public ownership and the need to limit the inherent liabilities which necessarily remain with taxpayers, however unwilling, are a constant friction against innovation and risk taking.

Even within industries that are not subject to rapid changes, the benefits for consumers of promoting competition among firms striving for market share are now better understood.  The emergence of competition policy in Australia has, however, been accompanied by increasing regulatory activity on the part of the Commonwealth and State governments.  For example:

• new access regime provisions have been included in Part IIIA of the Trade Practices Act (TPA) (dealing with essential facilities) and in Part XIC (dealing with telecommunications);
• State-based regulators with price-setting powers have emerged;
• the National Competition Council (NCC), and the Australian Competition and Consumer Commission (ACCC) have been established;  and
• a Payments System Board has been established within the Reserve Bank with powers to regulate and mandate standards and access in the payments system.

Even this brief list is enough to indicate that, since the release of the Hilmer Report, the level of government oversight of market behaviour and of top-down price regulation has increased.  What is more, most of these seemingly anti-market legislative initiatives have been implemented ostensibly in the name of fulfilling the aims of Hilmer, of opening up industries to market forces and putting faith in market outcomes as opposed to government-controlled outcomes.  What the government frees up with one hand it seems to regulate again with the other.

The main purpose of this paper is to try to explore what it might mean for a national competition policy were we to take seriously the presumption in favour of competition, and, in particular, of dynamic efficiency.

One issue that arises from such an exploration is a scepticism about whether true dynamic efficiency can ever be promoted through trade practices legislation.  Furthermore, even if this were possible, the good intentions of those who draft such reforms may well be undone by those entrusted to enforce the trade practices legislation.  This is not meant to be a simple-minded criticism of bureaucrats or regulators.  Rather, it refers to an inherent administrative bias, independent of the intentions of administrators, in favour of over-regulation, when the dictates of dynamic efficiency require, if anything, a winding back of top-down regulation.

The paper analyses the concept of dynamic efficiency and discusses its role in promoting optimum investment, risk-taking behaviour and innovation.  The paper distinguishes dynamic efficiency from static efficiency and briefly discusses the trade-off between the two.  It further indicates how, from the perspective of enhancing dynamic efficiency, bureaucrats are inherently biased in favour of regulation rather than non-interference in market outcomes.  This is illustrated in case studies analysing merger policy and the characteristics of a merger policy that took dynamic efficiency seriously.  The main conclusion is a proposal for reforming the trade practices legislation, at least as applies to mergers, by making the pursuit of dynamic efficiency the paramount goal of policy.

Government intervention in the marketplace is not automatically bad from the point of view of dynamic efficiency.  Indeed, the marketplace could not exist unless government provided the public good of property and contractual rights enforceable through the public good of a legal system.  These are the classic examples of readily accepted government "intervention" in the "market".  There is thus a natural role for government intervention and regulation in people's private affairs.  The issue is to determine the extent of that role:  that is to say, to distinguish between good and bad regulation.

STATIC EFFICIENCY AND COMMUNITY WELFARE

Efficiency in its static sense is the optimal use of resources at any given time.  If a good can be produced at less cost than currently, then it is efficient to do so.  Similarly, if individuals value a good differently, it is efficient to allow trade to take place, so that the individuals most willing to pay for a good obtain it.

Economic efficiency is desirable because it maximises the community's welfare.  Economic theory has identified the conditions of efficiency in an ideal market structure known as perfect competition.  In this market structure all (single-product) firms in all markets set price equal to marginal cost.  When this occurs, there is no waste in the economy and welfare is maximised in a single-period setting.  Indeed, marginal-cost pricing is the classic static condition for efficiency in a market.

The diametrically opposite case to perfect competition is perfect monopoly.  Monopolies are not efficient from a static viewpoint.  They set prices higher than marginal cost by limiting output.  Since monopolies could produce more output at a cost less than some people in the market would be willing to pay, they are inefficient from the community's point of view.  Although a monopolist maximises profit to the firm, the overall size of the social surplus is reduced by the restraints the monopolist imposes on output.

Economists use the tools of game theory to analyse those intermediate market structures which lie on the continuum between the two limit cases of perfect competition and perfect monopoly.  Such analysis is much more complex than in these two cases because competition occurs not simply with respect to price, but also with respect to a range of factors such as service, quantity, marketing, innovation, R&D, organisational structure, product development and so forth.

Modern game theory attempts to model the continual second-guessing that goes on among firms profitably competing in an industry.  As such, it is an attempt to capture the dynamic aspects which characterise all markets.  For this reason, the study of oligopolistic market structures is also much more difficult than the study of perfect competition or perfect monopoly.  This explains the scant reference in basic economic texts to game-theoretic models of economic behaviour.  This oligopolistic market structure is, however, far more compatible with a dynamic efficiency view of markets than either perfect competition or perfect monopoly.  Inherent in the model are rigorous dynamic competition on many fronts besides price and, by definition, monopoly rents and above-cost pricing.

Dynamic Efficiency and Community Welfare

Efficiency is the fundamental objective of competition, not for its own sake, but because promoting efficiency is the surest way to maximise the community's welfare.  The community's welfare, just like the concept of efficiency, has both a static and a dynamic aspect.  Whereas static efficiency maximises welfare in any given single period, dynamic efficiency maximises (the discounted net present value of) welfare over time.  Dynamic efficiency is important because we live in a world where time is a significant variable in our decision-making processes.  We are interested in our happiness not just here and now, but also in the future (suitably discounted).

But what is dynamic efficiency?  Whereas static efficiency is the optimum use of resources in a given single period, dynamic efficiency is the optimum use of resources over time.  In concrete terms, ensuring dynamic efficiency means ensuring that efficient investment decisions are made.  Investment needs to be optimal in both its quantity and its timing in order to maximise society's welfare over time.

Trade-offs

There is a potential trade-off between the two goals of static efficiency and dynamic efficiency.  In seeking to maximise single-period social surplus, we can detract from the maximisation of the discounted net present value of the social surplus as required for dynamic efficiency, and of course vice versa.

That such a trade-off probably exists can be seen in the area of patents and R&D, a field of economics virtually unintelligible from the perspective of static efficiency.  A unifying conceptual thread runs through the seemingly disparate topics of dynamic efficiency, innovation and economic growth.  Markets exist through time, and market participants must continuously deal with change.  A dynamic vision of markets emphasises those aspects of market behaviour which a static approach overlooks.  It sees markets as a process of continuous experimentation, change, innovation and (dynamic) competition.  Under this paradigm, any dominant position which a firm may have achieved cannot last long, because new firms, new procedures, and new technologies are always about to appear.

In this conception of competition, even the idea of natural monopoly makes little sense.  The barriers to entry that might exist at a particular time will be worn away by innovative technological change.  The most obvious recent example of this is IBM, a firm which attracted years of antitrust interest before sinking of its own accord.  A similar fate will doubtless befall Microsoft should it fail to retain its lead in innovation.  Until then, applying antitrust proceedings to Microsoft, as is currently occurring, can be considered akin to penalising a competitive winner for being and staying the best.

A corollary of this trade-off between static and dynamic efficiency is that there would appear to be an inherent tension between the market structure required by static efficiency considerations (namely, perfect competition) and that required by dynamic efficiency considerations (namely, oligopolistic competition).  This tension is again best illustrated by the case of patents.  A patent is a government-sanctioned temporary monopoly for the purposes of ensuring optimal innovation within the economy.  Optimal innovation is necessary to ensure that the social surplus is maximised over time.  Without patents, there would be an under-supply of innovation and, given that technological change is the engine of productivity growth, a considerable diminution in community welfare.  The case of patents therefore highlights the link between innovation and monopolistic profits.  However, even in this instance, the scope for entry into innovative activity means that resources will be attracted into areas earning superior returns.  In this way, excessive rent is competed away.

The existence of monopoly rents, therefore, is the Rorschach test for economists and their approach to economic efficiency. ⁽¹⁰⁾  Whereas monopolies and monopolistic profits are irreconcilable with static efficiency, monopolistic profits are the rewards for innovation and successful risk-taking from a dynamic point of view.  The monopoly rents that such a structure makes possible for industry participants can be seen as the reward for innovation.  Such innovation manifests itself in numerous dimensions besides the special case of R&D, such as firm structure, product development, successful marketing campaigns and the like.

If one were to accept that there is an efficiency policy trade-off whenever assessing firm behaviour within a market, the question which naturally arises is:  what decision-rule should a regulator adopt when faced with a choice between the two goals of static and dynamic efficiency?  The answer is that dynamic efficiency must be chosen.  The benefits to community welfare of greater productivity growth and technological advancement in an economy exceed by orders of magnitude the benefits that might flow at any particular time due from the maximisation of static social surplus.

Given this inconsistency between the two concepts of efficiency, the policy goal of trade practices legislation ought to be solely the encouragement of competition for the purposes of promoting dynamic efficiency.

What then are the implications of this claim for trade practices policy and enforcement?

Can Dynamic Efficiency Be Operationalised for Policy Purposes?

The world was a very different place when the TPA was enacted in 1974.  Import barriers and transport costs were much higher, the information and digital revolution was only beginning, the world economy was less open and financial markets more hamstrung, and the economy less outward-looking and less dynamic generally.

Perhaps most important, the economic sub-discipline of industrial organisation at that time was very under-developed.  Economic theorists have become interested in industrial organisation only in the last twenty-or-so years, rejuvenating it and placing it on a firmer theoretical foundation.  Given the state of economic theory at the time, and the more static nature of the economy, it is perhaps little wonder that the TPA reflects a static efficiency viewpoint.

In a more dynamic world, however, a bolder trade practices policy is arguably required.  Many aspects of the current Act would not survive if it were revised in the light of dynamic efficiency as the goal of competition policy.  For example, it would not give paramount importance to the concept of market definition.  But it is more difficult to speculate on what such a rewritten Act would contain rather than what it would not contain.  It is fair to say that it would be a rather minimalist document, perhaps embodying no more than the traditional common-law restrictive practices, or pared down to only "per se" prohibitions.

Furthermore, even if we could enact an Act which satisfactorily embodied the goal of dynamic efficiency, the incentives facing bureaucrats entrusted with enforcing such an Act are such that they would be biased towards undermining the policy goal of the Act, regardless of their intentions.

WHY TRADE PRACTICES BUREAUCRATS ARE BIASED IN FAVOUR OF REGULATION

The incentives facing trade practices bureaucrats are such that, even if they accepted the dynamic nature of the world, and took seriously their role in promoting dynamic efficiency, they would still be biased in favour of an over-regulatory and interventionist approach because of two factors:  asymmetry in errors, and risk aversion of decision-makers.

Experimentation and risk-taking are at the heart of dynamic competition, and are especially evident in oligopolistic market structures.  If a practice is prohibited, the option of experimenting is foreclosed for the period of the prohibition.  According to William Brock (1989), this situation should be analysed along lines similar to the cost-benefit analysis of irreversible decisions in environmental policy.  In environmental policy the option value of an environmental resource (say, a natural rain forest) increases as:

1. the underlying variance of uses for the resource increases;
2. the length of the "life" of the option increases;  and
3. the discount rate on the future decreases.

With respect to the first point, clearly the greater the change an industry is undergoing, the greater will be the variance of potential value created by a practice whose prohibition is sought.  In a fast-changing environment, regulators should therefore be "bolder" or more tolerant in allowing practices that in quieter times they might not have allowed.

With respect to the second point, legal and government decisions, once enacted, are notoriously difficult to reverse:  once a practice is prohibited, it is likely to remain so for a very long time.  Furthermore, the mere existence of an ACCC-like regulator is likely to have a "chilling" effect on innovative practices.  Even where practices have finally been authorised or permitted, there is likely to be a considerable "delay" cost involved.  This can be frequently observed in merger and take-over policy.

The third point needs no amplification.

The cost of an over-regulatory (prohibitive) approach lies therefore in the lost option value of the prohibited practice.  For any given decision by a regulator, this cost is likely to be high due to both the unobservability of certain errors and the conservatism (risk aversion) of regulators.  If a regulator permits a practice (say, a merger) and prices rise, then such a detrimental effect is immediately observable and the regulator's error of judgement available for all to see.  If, on the other hand, a banned practice would have been of lasting benefit to the community, this would not be observable, for the error takes the form of an opportunity forgone.  The regulator's error of judgement in this case is invisible.

Economists call the former error a Type I error, the latter a Type II error.  These two types of errors are two sides of the same coin.  An attempt to avoid a Type I error increases the likelihood of a Type II error, and vice versa.

From the standpoint of dynamic efficiency, Type II errors are probably more costly than Type I errors.  But as Type I errors are observable and Type II errors are not, in practice a natural bias sets in among regulators in favour of preventing Type I errors.  Put simply, decision-makers try to make mistakes that are difficult to detect. ⁽¹¹⁾  In the context of economic reform, or merger policy, this bias works to the advantage of the status quo.  Being risk averse, the regulator will inevitably err on the side of caution and favour merger prohibition.  But this bias simultaneously tends to maximise Type II errors, thus harming dynamic community welfare.

In the enforcement of Part IV-type provisions, competition policy should be solely concerned with enhancing social welfare through the promotion of dynamic efficiency.  This should be the ACCC's primary goal and even raison d'être.  The above analysis indicates that current competition policy needs to be reassessed to determine the welfare-reducing impact such policy has on market activity.  As Brock (1989:11) has stated:

I believe that the [ACCC] should be held accountable for the economic consequences of their decision.  Commissions tend to be punished more when they lose welfare unconventionally than they are rewarded when they gain welfare conventionally.  But they tend to lose little when they lose welfare conventionally.  This leads to a bias against experiments, especially economic efficiency-enhancing experiments where the benefits do not diffuse rapidly and visibly across a broad set of people. [emphasis added]

Regulators should make more Type II errors than they do.  The trick is to identify a promising institutional arrangement which would permit such an otherwise unlikely event occurring.  Alternatively, an arrangement should be put in place which allows for experimentation, with the scope to learn about the nature of the Type II errors and with the possibility of reversing decisions which, in the light of experience, are found to be wrong.

CASE STUDY:  MERGERS

To bring the foregoing discussion into focus, consider a current example:  bank mergers.  In Canada, banks are merging because of increased competition from US banks.  Australia has the "six pillars" policy whose abolition was recommended by the Wallis Report but deferred by the Treasurer for political reasons. ⁽¹²⁾

From the dynamic standpoint, innovation and experimentation in organisational structure and management are wholly legitimate responses to a changing environment.  There are many reasons why two firms (in the same industry or not) may want to merge:

• to reduce transaction costs;
• to realise cost savings through rationalisation;
• as a means of acquiring capital;
• to appropriate complementarities and economies of scope (and possibly of scale) where these cannot be realised through arm's-length arrangements;  and
• as an alternative to a firm ceasing to trade when an industry is undergoing adjustment to the long-run supply equilibrium.

A rough rule of thumb in orthodox antitrust policy (though not in strict economic analysis) is that vertical mergers are acceptable, but horizontal mergers are not.  From a static point of view the argument against horizontal mergers seems obvious enough:  they increase concentration within an industry, making it easier, and hence more likely (though not inevitable), for explicit or implicit collusion to take place.  But if the ACCC were to realise the importance of dynamic efficiency, it could well let horizontal mergers proceed unless the harm to static efficiency requirements were overwhelming.  Would the ACCC's approach be less regulatory as a result?

The ACCC's actions are clearly governed by more than pure economic theory and efficiency considerations.  Even if the ACCC wanted to allow bank mergers, the political consequences would be such as to encourage an overcautious (that is, prohibitive/regulatory) decision.  Thus, Luke Woodward, a Senior Assistant Commissioner at the ACCC, has stated that:

The ACCC naturally accepts the proposition that decisions it makes now may have long term effects but it is far less comfortable with the idea that we should make predictions about the long term structure of the Australian banking industry.  We do not plan to be either short termist or short-sighted, but the sheer number of factors that could affect the nature of the development of banking and of financial services more generally requires that the ACCC necessarily take a conservative view and base its judgments on the actual structures it is faced with, taking into account, to the extent it is able, emerging trends and likely developments.  (Woodward, 1996:18;  emphasis added)

Discussing the home loan market and the factors which would need to be taken into account in predicting the future in that market, Woodward goes on to say:

Each of these factors on its own could significantly change market structure and so market definition and the ACCC will necessarily focus on current market conditions rather than gamble on future developments.  (Woodward, 1996:19;  emphasis added)

A merger prohibited is an opportunity for experimentation forgone.  The financial sector, both nationally and globally, is undergoing a period of rapid change.  Under the analysis presented in the previous section, this means that the "option value" of a lost period of experimentation due to merger prohibition is far greater today than it would have been fifteen-or-so years ago, when the financial system was protected by government regulation.

Naturally experimentation is a risky activity, and not every experiment, by definition, can be successful.  An acceptance of the inherent dynamism of the world entails a certain sanguineness about business failure, certainly a greater level of sanguineness than the typical risk-averse regulator could permit himself.

REMEDIES?

The ideal policy recommendation for dealing with mergers which arises out of the above analysis is the abolition of section 50 of the TPA and a total tolerance of mergers and take-overs.  Since that is never going to happen, it is necessary to turn to some second-best solutions.

Currently, for all practical purposes, the burden of proof rests with firms seeking authorisation for a merger to show that the proposed merger would not lead to a substantial lessening of competition.  This is putting the cart before the horse.  Given the benefits of experimentation to the economy, placing the burden of proof on the firms that wish to merge favours the status quo and may thus deprive the economy of valuable experimental outcomes and encourage the ACCC to intervene out of cautiousness.  The burden of proof must therefore be shifted to the ACCC.  Section 50 and the corresponding enforcement provisions of the TPA ought to be changed to permit mergers as of right, subject to the ACCC being able, within some pre-specified period (say, six months or a year after the merger or take-over was effected) bringing an action in court for the divestiture of the new agglomeration. ⁽¹³⁾  In such cases the onus would of course lie with the ACCC, as the applicant party, to show on the balance of probabilities that the new agglomeration had resulted in a decrease in the social surplus or failed by reference to whatever other criteria had been specified.

Naturally there are many possible objections to this proposed change.  It is difficult and expensive to unscramble the egg.  Furthermore, firms anticipating a divestiture order may engage in activities that raise the costs of divestiture as a shield against remedial intervention by the regulator.  Such objections, although worthy, should not deter us from contemplating change.  As we have discussed, the present situation is not free of shortcomings.

The proposed change to the merger rules absolves the ACCC from the responsibility of administering competition rules which are not suited to a dynamic world.  By redefining the regulatory task, it offers a possible remedy to the bias in risk-taking that is inherent in the current approach.  The current dilemma for the ACCC, of speculating somewhat fruitlessly on what the future might bring as a token to market dynamics, is resolved.  Under the proposed change to merger rules, the ACCC's regulatory task is to act only if events prove that a merger was anticompetitive.  That is, the role of the ACCC is to remedy Type II errors.

The regulatory burden on industry would not be totally removed.  Mergers would carry the risk of subsequent divestiture, should evidence accumulate against them.  But what is important for all participants in the economy is that the possibility of experimentation and learning would not be lost.  That is a prize worth pursuing.

THE ACCC's ATTITUDE TO MERGERS

INTRODUCTION

Over the more than twenty-year life of the Trade Practices Act, mergers have probably received more publicity than most other matters.  They have also featured prominently in litigation undertaken by the ACCC and its predecessor, the Trade Practices Commission.

Given the emphasis on mergers in recent years, it is somewhat surprising that early antitrust legislation lacked specific provisions against mergers.  In Australia, the Trade Practices Act passed in Australia in 1965, and its 1971 successor, lacked a specific mergers provision.  It was not until the 1974 Act was passed that this was rectified.  Essentially, early legislation intended to deal with trade practices focused on conduct and did not seek to limit future problems by considering the implications of structural changes resulting from mergers for conduct.

WHY THE FOCUS ON MERGERS?

Contrasting with this, mergers have become the focus of attention over the last twenty years or so because the link between conduct and mergers has been recognised.  This is highlighted by the structure-conduct-performance (SCP) approach to competition.  Essentially, this assumes that the structural features of a market provide a relatively stable environment within which conduct occurs.  Nevertheless, while the SCP paradigm assumes a causal relationship from structure to conduct to performance, in practice the interrelationship between them is also recognised.

The Trade Practices Tribunal in Queensland Co-operative Milling Association Ltd and Defiance Holdings Ltd (QCMA) 1976 indicated that whether firms compete in a market is very much a function of the structure of the relevant market.  The production and selling decisions of firms are constrained to a greater or lesser extent, according to the structural features of the market.  For example, a firm may currently be the only major participant in a particular market and so appears able to restrict output and raise prices and hence profits.  But if barriers to entry are low, increased profits will signal an opportunity to potential new entrants who, upon entering the market and increasing supplies, will cause the price to fall.  Thus, the single seller will have a relatively short period during which it possesses market power.

Market structure may change through time and this may have implications for the competitive process.  Such changes may result from firms which are very successful in the market driving less successful competitors out of business.  This is simply the outcome of rivalrous behaviour which is the competitive process in operation.  Alternatively, structural changes may result from new entry (a more competitive industry) or from exit (a more concentrated industry) in response to changing market conditions.  Mergers have a variety of possible purposes but may result in a reduction in competition by reducing the number of sellers competing in the market and, in some cases, by raising the barriers to new entrants, for example by gaining control of an essential raw material or a particularly favourable location.  In the absence of a mergers provision in the trade practices legislation, if collusive conduct is a breach of the Act, then a merger could provide a means to the same end;  the acquisition obviates the need to enter into an agreement.

THE COMPETITION TEST

The nature of the competition test included in section 50 of the Trade Practices Act (TPA) has changed several times since 1974.  The Act initially prohibited mergers likely to result in a substantial lessening of competition in a market for goods or services in Australia.  In 1976 the Trade Practices Review Committee was established to review the TPA.  This took place at a time when the economy was in recession and unemployment was relatively high.  It was suggested that too many mergers were being caught by section 50, resulting in high compliance and administrative costs, while weakening the effectiveness of the market for corporate control by sheltering less efficient industries, thus discouraging economies of scale and other rationalisation benefits.

In 1977, the test changed to one whereby a merger breached the Act if it made a firm dominant or strengthened existing dominance in a substantial market.  Consequently, so long as there are at least two relatively evenly matched participants in a market, it is unlikely that a merger would be caught by the dominance test.  Yet a merger which results in increased concentration in a market may facilitate collusion.  Where collusion is overt and can be proven, it can be dealt with under the relevant section(s) of the TPA.  However, where there are only a small number of market participants (such as when a merger reduces the number of participants from three to two), an agreement may not be necessary for co-ordination to occur.  Such conscious parallelism does not usually represent a breach of the TPA, but it may substantially lessen competition within the relevant market.

An example of the problems associated with the dominance test was the acquisition by Amcor and Visy Board of 50 per cent each of the only remaining Australian corrugated fibreboard manufacturer, Smorgon, in 1989.  This did not lead to dominance by either company as they each increased their market share proportionately;  but it did substantially lessen competition.  Other acquisitions which were not opposed under the dominance test but which caused competition concerns included Tubemakers' take-over of McPherson's (1991);  Coles' take-over of Myer (1985);  Ansett's acquisition of East West Airlines (1988);  and the sale of the Herald and Weekly Times to News Ltd (1986-87).  In each case, as there was at least one other substantial remaining market participant, dominance was not created or strengthened.  It is somewhat surprising to realise that, under a dominance test, the Commission would not look at, for example, mergers between NAB and Westpac or Optus and Vodafone or between the major oil companies.  As the economy has become more concentrated and where imports are small, concern about mergers which substantially lessen competition rises.

From at least the 1980s, consideration was given to returning to the substantial lessening of competition test.  Two parliamentary committees examined the appropriateness of the dominance test.  While the majority of the Griffiths Committee (1989) did not recommend change, two of its members favoured change on the grounds that a corporation can be in a position to engage in anti-competitive conduct without dominating the market.  In December 1991, the Senate's Cooney Committee tabled its report recommending a change.  As a consequence, in January 1993 the test was changed back to one relating to the substantial lessening of competition in a substantial market.  The latter phase distinguishes the current test from that which applied between 1974 and 1977.

Since 1993, then, the Act has prohibited mergers or acquisitions which have the effect or likely effect of substantially lessening competition in a substantial market.  Of course, mergers may be authorised if there is a sufficient public benefit.  Moreover, section 87B is available for undertakings to overcome the anti-competitive effect of mergers where appropriate.

Where governments privatise, they normally refer questions about the competitive effect of acquisitions to the Australian Competition and Consumer Commission (ACCC).  In addition, the Commission believes that section 50 generally applies.  Scrutiny of privatisations has become a significant part of the Commission's mergers work.

The change in the law from a test of dominance to a test of substantial lessening of competition seems to have occurred satisfactorily.  Moreover, as more and more mergers in deregulating sectors come before the Commission, it is increasingly evident that the change in the test was necessary if deregulation is to work properly.

The Commission has published guidelines (ACCC, 1996b) which do not differ very greatly from those in North America and Europe and in some cases have led them.  There have been some updates in the guidelines covering such issues as the role of efficiency considerations in relation to section 50 and in relation to "globalisation" arguments that anti-competitive mergers may be justified even if the merger is anti-competitive in the Australian market.

THE ACCC APPROACH TO MERGERS

I would now like to comment briefly on the approach that the ACCC follows when assessing merger proposals.

As a guide for industry, the ACCC published its Merger Guidelines in 1996 setting out the process for, and issues relevant to, its administration of the merger provisions.  The guidelines do not bind the Commission, but they provide parties with an indication of what the Commission considers when investigating mergers and, importantly, indicate to industry what the Commission is looking for in a submission outlining a proposed acquisition.

The guidelines provide a five-stage process for the Commission's assessment of substantial lessening of competition, as follows.

First, the market is defined.  In establishing the market boundaries, the Commission seeks to include all those sources of closely substitutable products, to which consumers would turn in the event that the merged firm attempted to exercise market power.

Market concentration ratios are assessed.  If the market concentration ratio falls outside the thresholds, the Commission will determine that substantial lessening of competition is unlikely.

Potential or real import competition is examined.  If import competition is an effective check on the exercise of domestic market power, it is unlikely that the Commission will intervene in a merger.

In the fourth stage of its merger assessment, the Commission looks at the barriers to entry to the relevant market.  If the market is not subject to significant barriers to new entry, incumbent firms are likely to be constrained by the threat of potential entry to behave in a manner consistent with competitive market outcomes.  A concentrated market is often an indication that there are high barriers to entry.

Finally, the Commission looks to other factors which are outlined by the legislation (section 50(3)).  They include whether the merged firm will face countervailing power in the market;  whether the merger will result in the removal of a vigorous and effective competitor;  or whether the merger is pro-competitive, not anti-competitive.

MERGERS STATISTICS

The ACCC has recently published a detailed statistical analysis of mergers (ACCC, 1998).  The statistics show that in the latest year available, 1996-97, that only about 5 per cent of mergers were opposed and, of those, some were later not opposed once satisfactory undertakings had been given.  In some respects the 5 per cent figure overstates the extent of the Commission's opposition because many mergers do not raise competition issues at all and are not considered by the Commission.  On the other hand, there may be some mergers that are not brought forward to the Commission because the nature of the section 50 prohibitions is well known.  It is not the Commission's impression, however, that business people are shy in coming forward to sound it out about possible mergers, even impossible-looking mergers.  There have not been many of these.

CRITICAL MASS ARGUMENTS

Business people frequently raise the question of whether or not the merger provisions of the Trade Practices Act prevent the mergers necessary for Australian firms to be of the size necessary to take part in global markets.

The answer to this is:  rarely, if ever;  and, if so, then in circumstances where it is on balance undesirable because of the anti-competitive effect in the Australian market.  The fact is that the Commission has not in the last seven years opposed mergers where imports make up more than 10 per cent of the relevant market (this is not a rigid rule but it is a fact of history).  In other words, the Commission has not opposed mergers in sectors already exposed to international trade competition.  It is in this sector that the argument for firms needing to be large to take part in world markets is most relevant.  Moreover, even where there is no important competition, the Commission opposes relatively few mergers, and, where it does, some of them can be resolved by undertakings.

It is often argued that Australian industries need to develop the "critical mass" necessary to compete internationally.  But I think it is important to point out that obstacles to export growth may face industry participants of all sizes.  It is not apparent that, simply by entering a collaborative arrangement like a merger or joint venture, a participant's ability to compete internationally is enhanced.  Size is often not necessary to enhance the ability to compete on world markets.  It has been convincingly argued that, in many cases, domestic rivalry rather than national dominance is more likely to breed businesses that are internationally competitive.  A recent report to the government which reviewed business programmes in the context of an increasingly competitive global market noted that a lack of domestic competition was one of a number of impediments to building globally sustainable firms in Australia.

While size may not be necessary to enhance export opportunities, correct and complete market information is crucial.  Small and medium-sized enterprises may be disadvantaged when it comes to having access to adequate information -- something that is often claimed to be an advantage of operating under a single-desk system.  Continuing improvements in information technology and electronic commerce, however, suggest that this is likely to be less of an issue in the future.

It is sometimes argued that Australian primary industry producers benefit from having a single-desk policy because single-desk buyers overseas are able to use their market power to force the price down.  But there is less and less use of single-desk buyers overseas.  For example, as the recent report into the Barley Marketing Act by the Centre for International Economics (CIE, 1997) explains, China now has a general policy of privatisation and market reform as it prepares to join the World Trade Organisation.  Therefore, the previous central buying agency in China, COFCO, has lost its dominating influence for buying barley, and other Australian statutory marketing organisations (other than the Australian Barley Board, ABB, that is) now do not deal with COFCO.  Another example is that in Saudi Arabia and South America, the ABB's single desk sells directly to grain traders (who, in Saudi Arabia, then on-sell to the GSFMO).

A common argument for retaining single-desk marketing is that it is needed to counter the effects of export subsidies by the United States and European Union.  Yet there has been a scaling down of subsidies by the US and EU following the Uruguay Round agreement.  For example, there have been no subsidised US barley exports under the Export Enhancement Program since July 1995.  Also, single-desk selling does not actually affect the level of subsidies:  export subsidies may or may not occur regardless of whether Australia has single-desk marketing arrangements.  It may in fact be the case that a single desk is likely to be an inappropriate response to export subsidies;  for example, if a statutory marketing authority genuinely price discriminates, this implies that it restricts supplies to premium markets, which means more, not less, product is likely to be sold to non-premium markets including those affected by export subsidies.

It is also sometimes argued that there may be instances of domestic countervailing power that require a single desk as a counterweight.  For example, there is a small number of buyers in the domestic malting barley market (contrast this with the domestic feed grain market where there are large number of players and little scope for power to be exercised by private traders or users).  However, the CIE (1997) has indicated that maltsters are prepared to pay premiums for the qualities of malting barley they require;  and, in order to obtain surety of supplies, some growers are keen to deal directly with maltsters to obtain these premiums.  This suggests that there is no case for anti-competitive arrangements, even in areas where there may be some domestic countervailing influence.  Trade practices law should be able to deal with any anti-competitive market power problems (on the side of buyers) should they arise.

AUTHORISATION

If a merger is anti-competitive, authorisation is possible on public benefit grounds.  Since 1993, the Act has explicitly stated that export generation, import replacement or contributions to the international competitiveness of the Australian economy are public benefits.

Clearly the framework of the Act is not an obstacle to allowing Australian firms to merge to achieve the scale necessary for international competitiveness providing there is a sufficient public benefit.  There are in fact many cases where authorisations have been permitted.  Over half the number of authorisations have in fact been successful.  A number of them have related to cases where the merger would cause a substantial reduction in competition in Australia but would bring international-type benefits.  The ACCC's recent publication on Exports and the Trade Practices Act (ACCC, 1997) provides a number of case studies including Dow Chemical/Huntsman Chemical, July 1996;  Chemcor/Hoechst Plastics, March 1997;  ICI Australia/Auseon, May 1997;  Du Pont/Ticor, 1996.  It identifies the kinds of arguments which the Commission considers most relevant to claims for mergers that will enable Australian firms to take part in world markets, even where the effects may be anti-competitive in the home market.  There are, of course, instances in which the trade-off of loss of competition in the home market versus benefits to Australia from a firm playing a role in world markets is unfavourable in terms of the public interest;  and in some cases mergers create monopolies or "home champions" in the home market.  They are not necessarily firms well prepared to compete in world markets, as Michael Porter (1990) has demonstrated.

DEREGULATING SECTORS

The real agenda of merger policy relates largely to the deregulating sectors of the economy.  Deregulation gives rise to circumstances in which mergers are likely to occur.  Some mergers are necessary for efficiency and should not be blocked.  Others are sought to undo the pro-competitive effects of deregulation and may need to be opposed.  As noted earlier, the present test is superior to the dominance test in dealing with those matters.

In recent years State, Territory and Commonwealth governments have initiated various pro-competitive reforms, involving horizontal and vertical disaggregation of government-owned monopolies, corporatisation or privatisation, and the removal of various restrictions on the operation of free markets.  These initiatives were given further impetus by the Competition Principles Agreement (COAG, 1994), whereby all governments agreed to a systematic review of all legislation restricting competition.

As a consequence of this, the assessment of privatisation proposals has become a much more significant part of the Commission's work in recent times.  In many cases involving individual asset sales, a number of bidding consortia require individual consideration.  The Commission's role is to ensure that the acquisition of an asset does not result in a substantial lessening of competition in a market.  In assessing privatisations, the Commission considers the existing interests of all bidders.

In the great majority of cases, bidders for privatised assets do not raise competition concerns and therefore do not raise problems under the Act.  But in the case of certain asset sales in the Victorian electricity sector the Commission did object to some bidding consortia.  The Commission did so where it took the view that the interests of certain consortia parties would have raised potential competition concerns through horizontal linkages in the Victorian electricity generation sector.

In the case of the privatisation of Hazelwood power station, for example, the Commission raised its concerns with one bidding consortium.  No further action was taken as the structure and composition of the consortium were changed during the course of the sale process in such a way that the Commission's initial concerns were no longer relevant to the bid.  In another case involving the sale of Loy Yang A power station, the Victorian Government sales group required bidders to give certain undertakings addressing the Commission's competition concerns about board representation and information flows.

In performing its assessment of any proposed acquisition, one of the matters which the Commission must take into account is the likelihood that the acquisition or merger would result in the acquirer being able to significantly and sustainably increase prices or profit margins.

Network industries can differ from others in that market power and the associated ability to increase prices are not always proportional to the amount of capacity controlled by any particular organisation.  Market power can also arise through technical characteristics;  for example, at peak periods gas or hydro generators with the ability to "ramp up" quickly may have greater market power than base-load generators with larger capacity.

The Commission has also focused on mergers and acquisitions within the electricity industry because of concerns arising from the fact that the sector is not subject to the competitive discipline of import competition and because of the lack of direct substitutes for electricity.

In its authorisation determination on the National Electricity Code, the ACCC expressed concern over the structure of the market in a number of jurisdictions.  The issue of market structure is not only crucial at the commencement of the National Electricity Market (NEM) but will be of continuing interest, particularly in respect of possible re-integration of firms participating in the NEM.  Concerns also include possible mergers within each segment of the market, arrangements whereby NEM participants operate in upstream or downstream sectors (such as a generation company also operating a retailing business) and merger proposals between different energy suppliers (such as an electricity industry participant buying a gas industry participant).

CONVERGENCE

The last issue of convergence is one that the ACCC is likely to have to consider in assessing mergers and acquisitions in the utilities sector in the future.  There have, for example, been recent reports in Australia of joint ventures and acquisitions involving telecommunications companies and energy distributors and retailers.  The development of multi-utility service provider companies is a logical further step.

Convergence raises challenges to effective competition policy, in terms of possibilities for regulatory "bypass" and for incumbents if the policy approach and the manner of regulation are uneven across the different industry sectors.  It can also be argued that convergence has the potential to create substantially resourced business units holding market power.  At the same time, however, convergence may lead to industry growth and diversity and therefore to greater competition between products and greater choice of suppliers for customers.  There are arguments for convergence in terms of economies in carrying out common functions:  for example, integrated billing for energy and reduced consumer transaction costs.  These benefits are likely to be maximised by having an integrated regulator which takes a consistent approach across industry sectors.

REAGGREGATION

Another issue that the ACCC is likely to have to consider in the near future is that of reaggregation of utility companies.  Consider, for example, the possibility that in Victoria the five power generation companies seek to merge or to take over (or to be taken over by) the distribution companies.  Of course, some cross-ownership restrictions are built into Victorian law (until around 2002).  If these mergers went ahead they could undo the pro-competitive effects of the Victorian divestiture of the former State Electricity Commission of Victoria.  Likewise, when deregulation gives rise to the replacement of State by national markets, firms often manoeuvre and merge in order to cope with the new situation.  Again, sometimes there are considerable efficiency gains, but at other times considerable anti-competitive effects.

To take the energy industry as an example, several kinds of mergers may arise for consideration in future:  first, horizontal mergers within a State, for example between power generators or distributors located in the same State;  second, vertical mergers between, for example, generators and distributors in the State;  third, conglomerate mergers between different utilities, for example between gas and electricity utilities, in the distribution and/or retail field;  fourth, interstate mergers combining some or all of the above elements.  These matters will fall to be assessed under section 50 of the Trade Practices Act.  In assessing them, one background factor worth noting is that the ownership structure of the energy industry and some other deregulating industries has been greatly affected by public ownership arrangements over the years.  The ownership pattern which might have emerged in a privatised market subject to competition laws was not present owing to the preference of most governments for the public utilities to have both horizontal and often vertical integration.  Clearly, the deregulation of current public utilities brings advantages compared with the artificial integration established by governments.  For example, the Victorian disaggregation of the electricity industry would seem to represent an improvement over the pre-existing monopoly arrangements.  However, it is not especially likely that an initial disaggregation will yield the optimum ownership patterns in the industry.  In free markets, reliance is placed on the workings of the capital markets to achieve more efficient ownership arrangements and on competition policy to make sure that those arrangements are not anti-competitive (unless they can be shown to be in the public interest).  The present Victorian electricity market starts without the benefit of these processes unfolding over the years.  It is quite likely that restructuring pressures will arise to create more efficient arrangements.  The possible efficiency benefits of such mergers will need to be recognised and accepted under the Trade Practices Act.  Equally, however, it will be important to ensure that mergers are not simply anti-competitive and designed to undo the pro-competitive effects of deregulation.

These kinds of considerations apply to all mergers in sectors of the economy undergoing deregulation.

DAIRY INDUSTRY

Another interesting case at present concerns the dairy industry, where the Commission has opposed few mergers in recent years, knowing that many of the mergers have been driven by the need to restructure in the light of deregulation of the dairy industry.  In the latest case concerning the proposed acquisition of Pauls (formerly QUF) by National Foods, the proposed acquisition was judged by the Commission to be anti-competitive in the Victorian market for fresh milk and also in the national market for yoghurts and dairy desserts.  Undertakings were provided by National Foods to overcome the ACCC's concerns.

The ACCC assessed this proposed acquisition against the backdrop of an industry that has been going through deregulation, in the downstream sector, since 1990 in various States.  Western Australia was deregulated in January 1990, Tasmania in July 1993, South Australia and Victoria in January 1995, and New South Wales and the Australian Capital Territory were deregulated at the beginning of July 1998.  At the moment, the deregulation of the Queensland dairy industry has been put on hold.

In the past, the Commission has defined the market for fresh milk as being State (such as Victoria) or regionally based (such as Northern NSW/South-East Queensland).  This has reflected the high degree of State-based regulation but also the commercial limits to shipping a product of high bulk/low value and physical limits to shifting a product with a limited shelf-life.  Even with most of the States deregulating, it is unlikely that a wider market definition would be adopted because of the commercial and physical limits to moving packaged fresh milk.

National Foods' bid for Pauls was ultimately unsuccessful, however.  Parmalat made a counter-bid and is in the process of taking over Pauls.

MERGER POLICY

Merger policy makes an important contribution to the achievement of a competitive and productive Australian economy.  Regulation of anti-competitive mergers is an important part of national competition policy.  Trade practices merger law conforms with the principles of national competition policy agreed to by all Australian governments when the Hilmer Review was established.  These principles included:

1. No participant in the market should be able to engage in anti-competitive conduct against the public interest;
2. Conduct with anti-competitive potential said to be in the public interest should be assessed by an appropriate transparent assessment process, with provision for review, to demonstrate the nature and evidence of the public costs and benefits claimed.  (See Hilmer et al., 1993, Terms of Reference.)

Merger policy is not some necessary evil.  Rather, it has a positive contribution to make to Australia's international competitiveness.  If mergers are allowed to occur without the application of competition law, then our exporters and import competitors will be supplied uncompetitively and inefficiently and their capacity to compete in world markets will be hindered.

A general point which needs to be made about mergers is that most of the matters that receive detailed consideration from the ACCC are mergers which are borderline cases.  Critics could sometimes argue that there is inconsistency in decisions.  While I do not agree with this, the significance of the criticism must be placed in context.  When a series of close mergers is considered by the Commission, it is not so difficult to mount a case of apparent inconsistency.  Often there will be very similar structural circumstances, but the Commission will go in different ways depending on the weight accorded to particular factors.  The fact that the Commission has to make difficult on-balance decisions about a few borderline mergers each year does not mean that there is no general consistency in the application of the Act to the vast majority of mergers which it must consider.

There are always some contentious decisions.  One which has been discussed in recent years has been the Commission's treatment of the mergers in the paint industry.  The Commission was informed in August 1995 of the proposed merger of Wattyl and Taubmans.  It advised the companies that it would oppose the merger.  They eventually sought authorisation the following April and the Commission refused the authorisation.  Just as the Tribunal hearings began, Taubmans was acquired by a new entrant to the market.

There has been some comment on the delay in the matter.  It was certainly not the fault of the ACCC.  The question of paint mergers has been discussed very extensively during the Cooney Senate inquiry into mergers;  and the Commission's concerns about paint mergers had been made very clear.  When the Commission was informed in August 1995 of the proposed merger, it promptly advised the parties of its very considerable concerns.  They chose, however, not to go to court or to seek authorisation until April in the following year.  It is not surprising that in this time another buyer emerged.

Throughout the period from August until April, the Commission continued to make its attitude clear.  At one stage there was a serious proposal to resolve the issue by Wattyl selling at least one of its major brands.  It is understood that this proposal was recommended by management to the board of Wattyl, but the board rejected the compromise.

There are two possible views of the merger between Wattyl and Taubmans.  The first was that the combined entity would have a market share of around 45 per cent, approximately equal to the share of Dulux.  It was argued that a new, larger, stronger firm could compete more vigorously against its big competitor.  The other view was that the competition between Wattyl and Taubmans was very important for the promotion of competition generally in the paint market and that their merger would create a much quieter state of affairs in which two big firms coexisted comfortably.  The Commission's market enquiries tended to suggest that the latter was more likely to occur.

Was the outcome a good one?  There is a strong argument that the final outcome has been highly desirable.  Following the acquisition of Taubmans, a merger occurred between Taubmans and Bristol, as a result of which the paint industry now has three strong national competitors.  Had the Commission not opposed the merger, there would have been only two competitors and it is the Commission's view that they would not have competed as hard as is likely in the present situation.

UNDERTAKINGS

Section 87B has become a very important part of the Trade Practices Act.  It has, however, attracted greatest attention in relation to its use in merger situations even though in fact the Commission is very sparing in its use of undertakings to resolve merger questions.

The Ampol/Caltex merger of 1995 provides the best-known example.  The Commission formed the view that the merger was likely to substantially lessen competition and so advised the parties.  They sought reasons for the Commission's decision and then suggested undertakings which would neutralise the anti-competitive effects of concern.  After much consideration and negotiation, the Commission accepted undertakings and the merger went ahead.  The ACCC did not see itself as engaging in social engineering, even in this case.  The parties had sought to merge and in doing so to engineer an outcome in which the petroleum products market would be much less competitive than in the past.  The Commission needed to be satisfied that the undertakings balanced or neutralised the anti-competitive effects.  Whether this is called engineering or not is a semantic matter.  The fact is that the Act clearly contemplates that undertakings will be used in these situations.  The gain is that mergers can go ahead and realise many of their benefits.

The question of whether undertakings should be negotiated publicly is sometimes raised.  The ACCC's preference is that undertakings should normally be made known publicly before being accepted so that there is a full opportunity of assessing their likely effects on the marketplace, aided by players currently involved in the marketplace.  There is, however, opposition by firms which want to make undertakings confidentially.

There are some circumstances in which the Commission may accede to such requests.  These include cases where the ACCC is reasonably well informed about the industry's history and circumstances, as it was in the dairy industry where it has considered a range of mergers in recent years.  Two merger proposals which would almost certainly have been unable to go ahead had the Commission not agreed to undertakings given in private were the National Foods case, discussed above, and the eventually aborted Wesfarmers attempt to acquire ICI's Australian assets.  The Commission is very hesitant indeed about agreeing to undertakings that are given privately, but it does not rule them out totally.

CONCLUSION

Experience in the years since 1974 with the mergers provisions of the Trade Practices Act provides a degree of certainty in terms of the process.  Through its 1996 Merger Guidelines, the Commission has sought to identify "safe harbours" for potential merger partners, as well as to highlight the structural features of a market which may result in difficulties.

In conclusion, I would like to reiterate that the merger laws and the Commission's administration of them are consistent with enabling Australian firms to realise greater international competitiveness.

NATURAL MONOPOLY REGULATION IN NEW ZEALAND

INTRODUCTION

In this paper I propose to discuss:

• the role of the Commerce Commission;
• New Zealand's "light handed" regulatory approach, with specific reference to natural monopoly/utility regulation;
• outcomes of New Zealand's "light handed" regulatory approach;  and
• some of the differences between New Zealand's regime and Australia's.

THE ROLE OF THE COMMERCE COMMISSION

The Commerce Commission is an enforcement agency established under the Commerce Act 1986.  Its primary role is to bring about awareness, acceptance and compliance with the Commerce and Fair Trading Acts.

The Fair Trading Act promotes fair trading by prohibiting misleading and deceptive trading conduct.  The Commerce Act aims "to promote competition in markets within New Zealand" by:

• attacking anti-competitive behaviour whether by groups of competing firms or by dominant firms;
• monitoring "monopoly" pricing;
• assessing business acquisitions to prevent the acquisition or strengthening of a dominant position in a market;  and
• imposing price control in circumstances of restricted competition.

The Commerce Act provides the rules for New Zealand businesses facing an open competitive economy, and has broad coverage.  It applies to almost all markets, irrespective of the industry or corporate form involved, whether firms are publicly or privately owned, and whether or not they are utilities.

As an enforcement agency, the Commission has a range of enforcement tools available to it.  These include education, administrative settlement, and prosecution through the courts.  Private court action may also be taken under most parts of the Act. ⁽¹⁴⁾  Only the courts can make findings on whether or not there has been a breach of the Commerce Act and impose penalties.

It is important to note that the Commission, as an enforcement body, has no role as an arbiter or mediator of disputes involving practices which may breach the Act.  The Commission is also not an industry-specific regulator or a market designer.  It does not have predetermined ideas about the numbers or relative strengths of competitors that should be in particular markets.

ANTI-COMPETITIVE BEHAVIOUR

In "promoting competition" through attacking anti-competitive behaviour, the Commerce Act prohibits restrictive trade practices (RTPs) which adversely affect competition.  Restrictive trade practices which contravene the Act include:

• arrangements, contracts and understandings between competitors that substantially lessen competition in a market (section 27);
• arrangements between competitors that reduce the competitiveness of another rival (section 29);
• price fixing (section 30);
• a company, dominant in a market, using its position to prevent competition in either that market or another market (section 36).

The Commission enforces the RTP provisions of the Act by court action or through administrative settlements.  The penalties for a breach contained in the legislation are significant (up to NZ$5 million for an organisation and up to NZ$500,000 for an individual).  The Commission, however, would like to see the Courts imposing much higher penalties for breaches of the Act. ⁽¹⁵⁾

An RTP resulting in a loss of competition can be authorised by the Commission provided it can be shown that the public benefits exceed the detriments stemming from the loss of competition.  Authorisation gives protection to the trade practice from legal action under the Commerce Act.

Given the characteristics of natural monopolies, section 36 of the Commerce Act is of particular relevance.  Based on section 46 of the Australian Trade Practices Act 1974, it prohibits a person who has a dominant position in a market from using that market power for the purpose of restricting the entry of any person into that or any other market, preventing or deterring any person from engaging in competitive conduct in that or any other market, or eliminating any person from that or any other market.

Section 36 is not directed against the existence of monopolies, but against the conduct of the monopoly;  that is, the misuse of its market power.  The section does not prevent a dominant firm from using its market power for purposes other than restricting competition.  For example, the charging of prices above the competitive level is not, in itself, prohibited by the Commerce Act.

Behaviour which may constitute an abuse of a dominant position and therefore breach section 36 of the Act includes:

• charging below cost ⁽¹⁶⁾ for a good or service for the purpose of eliminating competition;
• restricting the supply of a good or service;
• requiring customers to deal exclusively with it;  and
• refusing to deal with suppliers who sell to competitors.

With the exception of behaviour which falls within section 36, restrictive trade practices are authorisable on application to the Commission, the test being whether detriments to competition are outweighed by benefits to the public, derived from the practice.

BUSINESS ACQUISITIONS

Section 47 of the Commerce Act prohibits business acquisitions if, as a result of the acquisition, a firm would acquire a dominant position in a market, or a firm's dominant position in a market would be strengthened.  In this section, as in section 36, the meaning of "dominant position" is important and there has been considerable discussion in the courts about the meaning of the phrase, and how the presence or absence of dominance may be determined.

Section 3(9) of the Commerce Act defines a person as having a dominant position in a market if:

... that person as a supplier or acquirer ... of goods or services is ... in a position to exercise a dominant influence over the production, acquisition, supply, or price of goods or services in that market ...

In determining whether a person has such a dominant influence, regard is paid to the market share, technical knowledge, access to materials and capital of that person;  the extent to which that person is constrained by the conduct of competitors or potential competitors in that market;  and the extent to which that person is constrained by the conduct of suppliers or acquirers of goods or services in that market.

In 1996 the Court of Appeal approved the following dominance standard, which had earlier been adopted by the High Court:

... dominance involves more than "high" market power;  more than mere ability to behave "largely" independently of competitors;  and more than power to effect "appreciable" changes in terms of trading.  It involves a high degree of market control. ⁽¹⁷⁾

If there is a risk of dominance arising from a business acquisition, a clearance or authorisation can be sought from the Commission, which, if approved, gives protection to the proposed acquisition from challenge under the Act by either the Commission or other parties.  Business acquisitions which do not give rise to dominance concerns can be given clearance by the Commission.  The Commission will grant an authorisation if it is satisfied that the detriments arising from the dominance gained or strengthened as a result of the acquisition are outweighed by benefits to the public.

Section 47 does not apply to business acquisitions where there is a bare transfer of market dominance;  that is, where dominance is transferred from one party to another without being strengthened in the process.

NEW ZEALAND'S "LIGHT HANDED" REGULATORY REGIME

New Zealand's approach to competition regulation, often described as "light handed" regulation, has attracted considerable international attention.  Many countries these days seem to be calling their regulatory regimes "light handed", including Australia.  I would, however, claim (without the slightest suggestion of bias) that New Zealand has the original and only truly "light handed" regime.

The concept of "light handed" regulation emerged from the policy debate concerning the deregulation and privatisation of the state-owned telecommunications monopoly during the late 1980s.  Following this, and as part of wider economic liberalisation policies, other utilities in New Zealand (which were generally statutory monopolies under state ownership) were also progressively reformed.  Reforms included:

• the removal of nearly all statutory monopoly rights so as to expose utilities to competition;
• the corporatisation (and in some cases, privatisation) of numerous state trading departments, structuring them as limited liability companies with independent (primarily private-sector) boards of directors and commercial objectives;
• their restructuring to isolate the natural monopoly elements from the more contestable parts of the industries;  and
• the removal of social service obligations or their explicit funding by government rather than by cross-subsidy with profits earned by the business in non-contestable markets.

The "light handed" approach, which has been in operation for about ten years, followed a long period of heavy-handed industry-specific forms of regulation with high levels of trade protection, considerable regulation of the production sector, a major role for government in the provision of services, price control and many legislative monopolies.  The overall thrust of the "light handed" policy is to achieve efficiency and to encourage competition in New Zealand markets.

"Light handed" regulation provides an attractive, less economically distorting alternative to heavier forms of regulation with the associated industry-specific regulatory bodies and higher compliance costs.  For example, direct regulatory control imposed by an industry-specific regulator can generate its own inefficiencies, including the costs of operating the regulatory body, the information supply costs imposed on the regulated firms, and the compliance costs arising from the distortions caused by imperfect regulation.  The possibility of "regulatory capture" is often noted as another possible concern;  that is, where the regulator is captured by the regulated with the monopoly firms influencing the regulator to their own advantage.

As stated, the "light handed" regulatory regime in New Zealand has broad coverage, with the Commerce Act 1986 applying to most markets, irrespective of the industry involved.  As such, utility regulation in New Zealand is approached in a characteristically straightforward manner, using the same set of competition principles used to regulate other industries, together with some additional light-handed elements.

Broadly, "light handed" regulation consists of three main elements.  First, in some utility industries, the contestable and non-contestable/natural monopoly businesses within the utility have been separated, either by separate ownership or by the accounting "ring fencing" of the two businesses (for example, the separation of Trans Power, the natural monopoly electricity transmission system, from the electricity generator).

Second, there is a reliance on general competition law, as expressed in the Commerce Act 1986.  As I outlined, under section 36 of the Act, dominant firms must not use their dominant position in a market to restrict entry, prevent or deter competitive conduct, or eliminate competitors.  Utilities should also not engage in trade practices which substantially lessen competition, or in business acquisitions which lead to the acquisition or strengthening of a dominant position.

Third, industry-specific information disclosure regulations require the annual disclosure of accounting and other information by incumbent operators of "essential facilities" with market power.  These regulations are designed to make transparent the operations of companies possessing market power, and to encourage self-regulation.

Stronger action is possible, given sufficient justification, through the provision for the introduction of price control by the government, as provided for in Part IV of the Commerce Act, or other forms of regulation.

The application of "light handed" regulation to utilities or "vertically integrated natural monopolies" (which includes telecommunications, electricity and natural gas reticulation) starts with the recognition that not all parts of an incumbent's business are natural monopolies.  The central aim is to encourage competition in those related markets where entry is possible, and to ensure that entrants into those markets are not deterred by the market power of the integrated incumbent in the essential facilities services market.

The main regulatory problems posed by these utilities arise where access to the incumbent's essential facility network is necessary for new entrants to compete in the provision of upstream or downstream services.  These utilities have an obvious incentive to preserve their monopoly power and profit by hindering access, either by an outright refusal to supply, or by setting an access price and terms high enough to render entry unattractive.

There is also a group of utilities organised around natural monopoly facilities which are not vertically integrated.  These include electricity transmission, ports and airports.  Such utilities may have the horizontal market power to extract excess profits from the sale of the final good or service.

To date, the Commerce Commission has had to resolve three main competition issues involving utilities, namely, business acquisitions, the price and other terms of access to network facilities, and complaints of monopoly pricing.

Briefly looking at access issues, I would point out that complaints about access to a utilities network can range from an outright refusal to negotiate to issues about the appropriate access price and terms.  For example, important case law has resulted from a dispute over the attempts by Clear Communications (the entrant) to negotiate an interconnection agreement with Telecom (the incumbent) to use Telecom's local loops in order to provide a local telephone service.

Telecom put forward the Baumol-Willig rule as the way of setting the interconnection price.  That rule states that monopolists are entitled to provide services to competitors at the same price they implicitly charge themselves, including any monopoly profits.  The Privy Council sanctioned this rule in 1994.  Critics of the rule, however, have argued that it has undermined the effectiveness of section 36 and the government has issued a policy statement that it did not consider the Baumol-Willig rule to constitute a satisfactory form of access pricing.

A number of interconnection agreements have been negotiated to date (for example, between Telecom and Clear, BellSouth, and Saturn Communications).  Significantly, the interconnection terms are more favourable than Baumol-Willig prices.

The Commission has also received complaints from those involved in the supply and distribution of electricity, alleging inability by one company to gain access to another's network.

PRICE CONTROL

The threat of price control is an important and necessary part of the "light handed" regulatory regime.  Part IV of the Commerce Act provides for the introduction of price control by the government in circumstances in two instances:  where goods or services are, or will be, supplied or acquired in a market where either limited competition exists or competition is likely to be lessened;  and it is necessary or desirable for the prices of those goods or services to be controlled in the interests of consumers.

The Commission can recommend the imposition of price control or, alternatively, may be asked by the government to report on the necessity or desirability of such a step.  While there are no goods or services subject to price control at the present time, the Minister of Commerce has asked the Commission to report on the desirability of price control relating to international airport, airfield services charges.

Once a good or service is declared as a controlled good or service, a supplier of that good or service may apply to the Commission for an authorised price.  It is an offence to supply at any price more onerous than the authorised price.  If a supplier does not apply for a price to be authorised, the Commission can initiate the price authorisation process itself.  The Commission may authorise a "maximum, actual, or minimum price, as the case may require".  The Commission can also accept price undertakings as an alternative to setting an authorised price.  Price authorisations are subject to appeal to the High Court.

The broad criteria the Commission must take into account in setting an authorised price are set out in the Act (section 73):

• the extent to which competition is limited or likely to be lessened in respect of the controlled goods or services;
• the necessity or desirability of safeguarding the interests of users, or consumers or, as the case may be, suppliers;  and
• the promotion of efficiency in the production and supply or acquisition of the controlled goods or services.

It probably goes without saying that for the threat of the imposition of price control to be credible, the government needs to be prepared to use the price control provisions of the Act when the situation demands it.

OUTCOMES OF NEW ZEALAND'S "LIGHT HANDED" REGULATORY REGIME

It is argued by some that New Zealand's approach involves regulation that is too light-handed, with the result that firms with market power operate without sufficient restraint, and to the detriment of their customers and of efficient production.  However, the Commerce Commission considers that the policy, while still being developed and refined, has had some important successes for the economy in terms of improved efficiency and increased competition.  Of course, it would also be fair to say that the policy has encountered some as yet unresolved difficulties.

In relation to the telecommunications industry, Telecom had its statutory monopoly right over the provision of telecommunication network services removed in 1989.  In line with the country's "light handed" regime, there is very little in the way of industry-specific forms of regulation apart from specific information disclosure requirements, and no specific telecommunications regulatory body.

The Telecommunications (Disclosure) Regulations 1990 require Telecom to provide details of its local access-related services, including interconnection agreements, any substantial discounts offered, and financial accounts for activities related to local access.

The Commerce Commission believes that the telecommunications industry has made positive gains since deregulation, not least in terms of its competitiveness.  There has been significant new entry, a general downward trend in prices, a wide range of new products and services have been introduced, and the industry has become more customer-orientated with improvements in service quality.  Perhaps the major outstanding issues are those of number portability, local loop access arrangements and issues related to industry convergence.

A number of recent studies reflect reasonably favourably on the New Zealand approach in comparison with overseas alternatives.  For example, Professor Henry Ergas (1996) concluded that, in relation to telecommunications, the New Zealand regime compared favourably with the more heavy handed Australian regime.  Economists De Boer and Evans (1996) favourably compared Telecom's productivity improvements with those of British Telecom following privatisation.

Turning to the electricity industry, deregulation within the "light handed" regulatory framework has seen some benefits, particularly for industrial and large commercial users, and technological advances and service improvements.  But there is considerable debate as to the level which these benefits have filtered through to residential and small customers.

Again, the information disclosure requirements (specifically, the Electricity (Information Disclosure) Regulations 1994) play an important role in encouraging self-regulation in the industry by facilitating the comparison of prices, costs and other performance indicators across power companies.  Local electricity supply companies must publicly disclose separate accounting information on their generation, distribution and other activities such as retailing.  The information includes prices, line charges to all consumers, and other key contractual conditions.  Consumers' bills were also unbundled so as to reveal the separate line and energy components.

There are, however, some shortcomings in the present information disclosure regime.  For example, the ability to make inter-company performance comparisons is hindered by the diversity of power companies, and the flexibility within the regulations for companies to define their businesses and to allocate assets and costs between them.

ELECTRICITY REFORMS

These, and related concerns, are being addressed in the government's recently announced electricity reforms, which I will just touch on here.  Broadly, the reforms are aimed at providing more competitive prices and a choice of electricity suppliers for smaller customers.  However, some argue that the reforms are a backward step towards direct intervention by the government, and will not remedy the distortions that exist in the industry.

The package of reforms has two main elements.  The first is the reform of the government's major generation business, with the Electricity Corporation of New Zealand (ECNZ) to be split into three separate generation companies which, along with Contact Energy, will be permitted to compete in the retail energy market.  These generation companies will remain in state ownership.

Second, ownership of electricity lines and businesses is to be split from the competitive activities of electricity retailing and generation.  Electricity supply companies have a choice on how they implement ownership separation.  They can set up a separate trust by 1 April 1999 to own and run whichever business they do not wish to keep, or sell whichever business they do not wish to keep by 31 December 2003.  Companies choosing the second option will have to set up separate companies in the meantime and run them as if they had separate owners, including having separate boards and management.

Other reforms include stronger/enhanced information disclosure requirements, and a requirement that the industry adopt within twelve months low-cost arrangements to enable smaller consumers to change retailers.

The reforms also create new roles for the Commerce Commission, including the enforcement of the ownership separation rules and of the requirements that owners who have not yet complied with the ownership separation rules must operate at arm's length.  The Commission is also empowered to issue exemptions from the involvement rule for specific investors.

The Electricity Industry Reform Act 1998, which gives effect to these reforms, was enacted by Parliament on 8 July 1998, but (apart from a number of specific provisions) does not come into force until 1 April 1999.

DIFFERENCES BETWEEN NEW ZEALAND'S COMMERCE ACT 1986 AND AUSTRALIA'S TRADE PRACTICES ACT 1974

The New Zealand Commerce Act is similar in several important respects to the Australian Trade Practices Act 1974.  For example, the stated objectives of the two statutes are comparable:  the Commerce Act aims to promote competition in markets in New Zealand and the Trade Practices Act aims to enhance the welfare of Australians through the promotion of competition and fair trading.  But there are some important differences between the two statutory regimes.

First, while section 36 of the Commerce Act is similar to section 46 of the Trade Practices Act (on which it was based), the two provisions are not identical.  Most significantly, different threshold tests apply:  use of a "dominant position in a market" in New Zealand, and taking advantage of "a substantial degree of power in a market" in Australia.  "Dominance" connotes a greater degree of market power than is required by a "substantial degree of market power".

Second, different threshold tests apply in relation to mergers:  the acquisition or strengthening of a "dominant position" in New Zealand, and substantial lessening of competition under the Australian legislation.

Third, as previously discussed, Part IV of the Commerce Act enables goods or services to be placed under direct price control, the "threat" of which is aimed at constraining the pricing behaviour of dominant firms.  There is no counterpart in the Trade Practices Act, although the Australian Competition and Consumer Commission (ACCC) does have price surveillance functions under the separate Prices Surveillance Act 1983.

Fourth, with the exception of price fixing and resale price maintenance, the Commerce Act does not single out particular trade practices for attention in the way that the Trade Practices Act does.  For example, the Australian legislation includes specific provisions dealing with price fixing, resale price maintenance, primary boycotts, secondary boycotts, and exclusive dealing.

Fifth, the Commerce Act does not contain any provisions matching the access regime provisions in Part IIIA of the Trade Practices Act, aimed at regulating access to certain "declared" essential facilities with natural monopoly characteristics.  Nor does the Commerce Act contain any specific provisions relating to the telecommunications industry, as is now provided for in the Australian legislation (Part XIC of the Trade Practices Act).

Sixth, the institutional arrangements to administer the Acts are similar in the two countries, with the Commerce Commission performing similar roles to the ACCC.  In New Zealand, however, appeals are heard by the High Court, with the assistance of lay members, in contrast to the specialist Australian Trade Practices Tribunal with its Federal Court judge and lay assessors.

CONCLUSION

In conclusion, I would like to reiterate that the Commission's role is somewhat different from that of most overseas regulatory bodies, including Australia's, in that the Commission enforces general competition law under a "light handed" policy regime.

The area of utility regulation is particularly difficult to monitor and enforce due to the natural monopoly characteristics of networks which give rise to issues of dominance and access.  The Commission's involvement with utility industries occurs only when an issue arises which is at risk of breaching the Commerce Act.

As stated, the Commission is of the view that enforcement of this general legislation, together with information disclosure requirements and the threat of more heavy-handed regulation through price control, has had important successes for the economy.

NOTE

This paper has drawn on the following publications by Alan Bollard, the previous Chairman of the New Zealand Commerce Commission:

• A Brief Summary of Competition Policy in New Zealand (1997).
• Operationalising Economic Liberalisation:  Regulating Competition in New Zealand (1997).
• A Regulator's Perspective:  Monitoring Utility Sector Competition (1997).
• Utility Regulation in New Zealand (1997).
• (With M. Pickford), "New Zealand in Utility Regulation 1997", in I. Lewington (ed.), Economic Regulation of Utilities and Network Industries Worldwide, Centre for the Study of Regulated Industries, London.

---

REFERENCES

Arrow, K. and F. Hahn (1971), General Competitive Analysis, Holden-Day, San Francisco.

Australian Competition and Consumer Commission (ACCC) (1996a), Inquiry into the Petroleum Products Declaration, volume 2, AGPS, Canberra:  Appendix O.

------ (1996b), Merger Guidelines, Canberra.

------ (1997), Exports and the Trade Practices Act, Canberra.

------ (1998), The ACCC's Approach to Mergers:  A Statistical Analysis, Canberra.

Bauer, P. (1987), "The Disregard of Reality", Cato Journal, 7, pages 20-42.

Baumol, W., J. Panzar and R. Willig (1982), Contestable Markets and the Theory of Industry Structure, Harcourt Brace Jovanovich, New York.

Beesley, M. (ed.) (1994), Regulating Utilities:  The Way Forward, Institute of Economic Affairs, London.

Berg, S. (1995), "Reconciling Social Issues and Infrastructure Provision:  The US Experience", Infrastructure Pricing -- Policy Forum 1995, Bureau of Industry Economics, Canberra.

Binmore, K. (1990), Essays on the Foundations of Game Theory, Blackwell, Oxford.

Blaug, M. (1987), "Classical Economics", in J. Eatwell, M. Milgate and P. Newman (eds), The New Palgrave:  A Dictionary of Economics, Macmillan, London, volume 1.

Brock, W. (1989), The Antitrust Debate in New Zealand, New Zealand Business Roundtable, Wellington.

Buchanan, J., and C. Stubblebine (1962), "Externality", Economica, 29, pages 371-84.

Centre for International Economics (CIE) (1997), Review of the Victorian and South Australian Barley Marketing Act 1993, Canberra.

Council of Australian Governments (COAG) (1994), Competition Principles Agreement, Canberra.

Dasgupta, P. (1980), "Decentralisation and Rights", Economica, 47, pages 107-23.

De Boer, D., and L. Evans (1996), "The Economic Efficiency of Telecommunications in a Deregulated Market:  The Case of New Zealand", The Economic Record, volume 72(216), pages 24-35.

Demsetz, H. (1968), "Why Regulate Utilities?", Journal of Law and Economics, 11:  reprinted in H. Demsetz, The Organisation of Economic Activity, Blackwell, Oxford, 1988, volume 2.

------ (1969), "Information and Efficiency -- Another Viewpoint", Journal of Law and Economics, 12, pages 1-22, reprinted in H. Demsetz, The Organisation of Economic Activity, Blackwell, Oxford, 1988, volume 2.

------ (1988), "Two Systems of Belief About Monopoly" in H. Demsetz, The Organisation of Economic Activity, Blackwell, Oxford, volume 1.

------ (1995), The Economics of the Business Firm:  Seven Critical Commentaries, Cambridge, Cambridge University Press, 1995.

Ergas, H. (1996), Telecommunications Across the Tasman:  A Comparison of Regulatory Approaches and Economic Outcomes in Australia and New Zealand, Centre for Research in Network Economics and Communications, University of Auckland, Auckland.

Gilbert, R., and D. Newberry (1994), "The Dynamic Efficiency of Regulatory Constitutions", Rand Journal of Economics, 25, pages 538-54.

Hahn, F. (1984), "Reflections on the Invisible Hand", in F. Hahn, Equilibrium and Macroeconomics, Blackwell, Oxford.

Hayek, F. (ed.) (1935), Collectivist Economic Planning, Routledge, London.

Hilmer, F., M. Rayner and G. Taperell (1993), National Competition Policy:  Report by the Independent Committee of Inquiry, AGPS, Canberra.

Industry Commission (1996), Merger Regulation, Information Paper, Canberra, June.

Katz, M., and C. Shapiro (1985), "Network Externalities, Competition and Compatibility", American Economic Review, 75, pages 424-40.

Kay, J. (1994), "Regulating Networks", in M. Beesley (ed.), Regulating Utilities:  The Way Forward, Institute of Economic Affairs, London.

Kirzner, I. (1994), "The Ethics of Competition", in H. Siebert (ed.), The Ethical Foundations of the Market Economy, Mohr, Tubingen.

Kreps, D. (1990), Game Theory and Economic Modelling, Clarendon Press, Oxford.

Laffont, J.-J., and J. Tirole (1993), A Theory of Incentives in Procurement and Regulation, MIT Press, Cambridge, Mass.

Laine, C. (1995), "The H-H Index:  A Concentration Measure Taking the Consumers' Point of View", Antitrust Bulletin, Summer, pages 423-32.

Lal, D. (1980), Prices for Planning, Heinemann Educational Books, London.

------ (1983), The Poverty of Development Economics, Institute of Economic Affairs, London.

------ (1987), "Markets, Mandarins and Mathematicians", Cato Journal, 7, pages 43-70;  reprinted in D. Lal (1994), Against Dirigisme, ICEG, ICS Press, San Francisco.

------ (1989), The Hindu Equilibrium, Clarendon Press, Oxford.

------ (1993), The Repressed Economy, Edward Elgar, Aldershot.

------ (1994), Against Dirigisme, ICEG, ICS Press, San Francisco.

------ (1995), "Eco-Fundamentalism", International Affairs, 71, pages 515-28.

------ (1997a), The Poverty of "Development Economics" (second edition), Institute of Economic Affairs, London.

------ (1997b), "From Planning to Regulation:  Toward a New Dirigisme", Cato Journal, volume 17(2), pages 211-27.

Liebowitz, S., and S. Margolis (1994), "Network Externality:  An Uncommon Tragedy", Journal of Economic Perspectives, 8, pages 133-50.

Porter, M. (1990), The Competitive Advantage of Nations, Macmillan, London.

Robinson, C. (1993), Energy Policy:  Errors, Illusions and Market Realities, Institute of Economic Affairs, London.

Sen, A. (1983), "The Profit Motive", Lloyds Bank Review (January), pages 1-20.

Stigler, G. (ed.), (1988), Chicago Studies in Political Economy, Chicago University Press, Chicago.

Trade Practices Commission (TPC) (1992), Merger Guidelines:  Draft for Comment, Canberra.

Walker, J., and L. Woodward (1996), "The Ampol/Caltex Australia Merger:  Trade Practices Issues", Trade Practices Law Journal, 4(1), pages 21-48.

Woodward, L. (1996), "Mergers in the Finance Industry", IIR Finance Law Conference, Sydney, 18-19 March.

---

ENDNOTES

1.  As Demsetz (1995:144n) notes:  "as a small act of institutional immodesty, I note that the profession has allowed the University of Chicago to appropriate to itself the efficiency doctrine of antitrust.  The offering of this doctrine in a substantive, analytical way originated at least as much from work done at UCLA as from that done at Chicago".  Along with Armen Alchian and Ben Klein, Demsetz has been the leader of this industrial organisation school at UCLA.

2.  I have found this theory particularly useful in thinking of the natural monopoly which is the state.  In Lal (1989) I develop a model of the predatory state in which contestability plays a central role.  The model is used to explain the rise and fall of empires in India over the millennia.

3.  See for instance Gilbert and Newberry (1994), which also has references to this literature.

4.  But see Laffont and Tirole (1993) for an attempt to provide a textbook for the dirigiste technocratic regulator!

5.  See Kay (1994) for this illuminating characterisation of the natural monopoly element of utilities.  But I do not subscribe to the technocratic regulatory conclusions of his argument.

6.  Recently, Katz and Shapiro (1985) have advanced the notion of "network externalities", which are claimed to lead to market failure.  But most of these, for instance in computer networks and telecommunications, are examples of pecuniary externalities, as rightly emphasised by Liebowitz and Margolis (1994).  But as Buchanan and Stubblebine (1962) pointed out, such pecuniary externalities are Pareto-irrelevant and do not constitute examples of market failure.  See also Lal (1994:  Ch 11).

7.  See Lal (1980), and Lal (1993:  Ch. 12), for why I now believe the shadow price of public funds is likely to be less than unity.

8.  See Hearings of the HoR Standing Committee on Financial Institution and Public Administration with Professor Fels, Hansard, 21 April 1997;  Walker and Woodward (1996).

9.  Ibid.

10.  The original Rorschach test was devised as a tool for the analysis of personality, calling for responses to ink blots and drawings.

11.  This point was made by Sanford Berg (1995) at a Bureau of Industry Economics forum.  A comment from the audience offered a corollary to the effect that when mistakes are made decision-makers are difficult to detect!

12.  The "six pillars" policy derives from a policy position of the Commonwealth Government, announced in 1990, and reiterated in subsequent years, that mergers between any of the four major banks or two or three major life insurance institutions would not be permitted.

13.  The ACCC has the power to bring divestiture suits under section 81 of the TPA, though these proceedings are related to a breach of section 50.

14.  For example, only the Commission can apply to the Court under section 80 (pecuniary penalties for restrictive trade practices) and section 83 (pecuniary penalties for prohibiting business acquisitions).

15.  In relation to restrictive trade practices, the Australian Trade Practices Act provides for penalties of up to a maximum of A$10 million for companies and A$500,000 for individuals.

16.  Arguably below average variable cost, or average total cost, or at a level somewhere in-between.

17.  Port Nelson Ltd v Commerce Commission [1996] 3 NZLR 554.