Viewpoint: Canadian competition policy: progress and prospects Thomas W. Ross
Sauder School of Business, University of British Columbia
Abstract. This paper discusses the state of competition policy – in particular the economics of competition policy – in Canada today and considers its prospects going forward. It argues that: (i) the importance of competition policy has become accepted widely in Canada and indeed throughout much of the world; (ii) competition policy design and enforcement is in general well done in Canada; (iii) economists, including many Canadians, have played a central role in the development of an efficient and effective competition policy in Canada and elsewhere; and (iv) competition policy in Canada is today facing very serious challenges, and economists should be concerned. La politique de concurrence au Canada: progre`s et prospective. Ce me´moire discute de l’e´tat de la politique de concurrence – en particulier l’e´conomie de la politique de concurrence – au Canada aujourd’hui et examine ce qui l’attend. On sugge`re que (i) l’importance de la politique de concurrence est devenue largement accepte´e au Canada et dans le gros du reste du monde, (ii) le dessein et l’application de la politique de concurrence sont en ge´ne´ral bien faits au Canada; (iii) les e´conomistes, et plusieurs Canadiens parmi eux, ont The author is UPS Foundation Professor of Regulation and Competition Policy, Sauder School of Business, University of British Columbia. This paper is adapted from a State of the Art Lecture of the same title presented at the annual meetings of the Canadian Economics Association at Carleton University in May 2003. I am very grateful to Curtis Eaton for inviting me to deliver the lecture and the managing editor of this journal, Michel Poitevin, of suggesting I turn it into a piece for the Canadian Journal of Economics. Over the years, a great many people have contributed to the development of the ideas expressed her – too many to list. I would, however, like to mention a few who provided helpful comments or suggestions as I was preparing this lecture and article: Gwill Allen, Zhiqi Chen, Jeff Church, Marc Duhamel, David Krause, Frank Mathewson, Margaret Sanderson, Guofu Tan, Peter Townley, and Ralph Winter. I am also very grateful to Jennifer Ng and Ann-Britt Everett for their assistance pulling this together and to the Social Sciences and Humanities Research Council of Canada and the UBC Centre for the Study of Government and Business for research support. Email:
[email protected] Canadian Journal of Economics / Revue canadienne d’Economique, Vol. 37, No. 2 May / mai 2004. Printed in Canada / Imprime´ au Canada
0008-4085 / 04 / 243–268 /
#
Canadian Economics Association
244
T.W. Ross
joue´ un roˆle central dans le de´veloppement d’une politique de concurrence efficace et effective au Canada et ailleurs; et (iv) la politique de concurrence au Canada fait face a` des de´fis tre`s se´rieux ces temps-ci et les e´conomistes devraient s’en inquie´ter.
1. Introduction These are active and important days for competition policy in Canada. Legislative amendments in the 1980s and more recently have provided our Competition Bureau with increased powers to investigate and punish anticompetitive agreements and acts, and to block mergers that would lessen competition.1 Over the same period we have seen a considerable increase in merger activity and in the discovery of major international cartels. Increased concentration in many industries has introduced further concerns over the possibly anticompetitive behaviour of some of our largest firms. Competition law is inherently a form of economic regulation, with the implication that these increased powers must be wielded with a sound appreciation for the economic effects of contemplated interventions. A strong competition policy must then be based upon legislation that reflects current economic thinking about competition and efficiency. It must, additionally, be enforced by an independent agency of skilled professionals with the resources they need to make what are often difficult decisions. It was a very great honour for me to be asked by the incoming president of the Canadian Economics Association, Curtis Eaton, to present a State of the Art Lecture on the topic of Canadian competition policy at the Association’s 2003 annual meetings. It was a particular pleasure to deliver the lecture at Carleton University, my academic home for eight years and the place that provided me with a terrific opportunity to learn about competition policy in Canada from a great set of experienced colleagues and with such ready access to the Competition Bureau. This paper records and builds on that lecture. It does not, however, provide a fully detailed account of how we got where we are or of all the challenges that are presenting themselves these days.2 In this sense, it offers a somewhat idiosyncratic view of its subject – one I hope readers will find interesting nonetheless. 2. Progress and prospects: four themes For those readers not immersed in this subject, perhaps we should begin with a working definition. Competition policy (or antitrust) is a set of laws and enforcement/adjudication institutions that protect and enhance the competitiveness 1 Prior to 1996 the Competition Bureau was known as the Bureau of Competition Policy. 2 For the most comprehensive treatment of Canadian competition policy available today, I enthusiastically recommend the new text by Trebilcock et al. (2003). For a shorter treatment of the modern development of competition policy in Canada, the interested reader may want to refer to Ross (1998).
Canadian competition policy
245
of markets.3 While many motivations have driven the development of competition policy around the world and over time, the prime support for competition policy today (among economists at least) comes from our belief that competitive markets are the best guarantors of efficiency and generators of wealth.4 Thus, competition is seen as a means to an end, that end being efficiency. While a particular nation’s competition law may have dozens of sections, the main thrusts of modern antitrust lie in three areas. All involve mechanisms to limit the creation and abuse of market power. i) Controls on agreements between competitors to lessen competition (i.e., ‘cartels,’ ‘collusion,’ ‘price-fixing’). In Canadian law, these sections are often referred to as the ‘conspiracy provisions’ of the Competition Act. Their purpose is to prevent the creation of market power by contract or agreement of otherwise independent competitors. ii) Controls on abusive behaviour by large firms. These provisions limit the freedom of large firms to take actions to enhance or maintain market power. Many practices that would appear pro-competitive or benign when undertaken by firms without market power may in other situations help firms with large market shares to ‘lock in’ their market power – or even extend it. iii) Controls on mergers with the potential to create market power. In contrast to the two previous items, merger control is meant to prevent competition problems before they develop by blocking the build-up of market power through mergers and acquisitions. While the large majority of mergers will present no competition issues, some will raise the possibility that market power will be created in one or both of two ways: the merged firm might be itself large and powerful enough to profitably raise price post-merger (e.g., if the merging firms were each other’s best substitute);5 or the loss of one independent decision-making firm might be enough that the remaining firms can more successfully collude, either explicitly or tacitly.6 In addition to these broad areas, many competition laws will include provisions related to vertical contractual restraints (such as exclusive dealing, 3 Competition policy in Canada used to be referred to as ‘anti-combines policy’ or simply ‘combines policy,’ reflecting its nineteenth-century origins as an attack on the cartel-like ‘combines’ (similar to what Americans called ‘trusts’) that had developed in a number of Canadian industries. For much of its history, the Canadian competition law was known as the Combines Investigation Act. However, with the renaming of the law as the Competition Act in 1986, this usage is fading, and Canadians now typically use the international term ‘competition policy’ and the American term ‘antitrust’ interchangeably. 4 Other motives have included a desire to prevent redistributions of income from consumers to producers, a desire to control inflation, a desire to support small business. For a review of the historical development of the support for competition policy in Canada, see Gorecki and Stanbury (1984). 5 This is referred to as a lessening of competition through the unilateral exercise of market power. 6 This is referred to as a lessening of competition due to increased interdependence (or ‘coordinated effects’).
246
T.W. Ross
exclusive territories and resale price maintenance). There may also be sections that control some aspects of firms’ pricing decisions (e.g. tied selling, predatory pricing and price discrimination). Some laws, such as those of Canadian and the United States, even add consumer protection provisions (e.g., misleading advertising, pyramid selling) to their competition laws, though these are often administered quite separately, as if by a separate body. There are four principal themes I would like to develop in this paper. Together I hope they persuade the reader that competition policy is a field that needs good economics and economists and that it is a field that rewards them with rich opportunities to make significant contributions to the efficient functioning of a modern market-based economy. I hope to also establish the point that the role of sound economics in Canadian competition policy is not assured these days. My four themes are as follows: 1. Competition policy is important – and almost ‘sexy’ – not just for business leaders such as Bill Gates (of Microsoft) and Jack Welch (of General Electric), who see themselves and their companies as victims of meddlesome regulators, but for the economy as a whole, producers and consumers alike. 2. By and large we do competition policy well in Canada – but we could be doing it better – and there may be storm clouds on the horizon. 3. Economists have done a great deal to improve competition policy in Canada and elsewhere, even as their own views have evolved. 4. Competition policy in Canada and elsewhere is facing some significant challenges.
3. Theme 1: Competition policy is important While I cannot pretend to be a scholar of western economic history, a little will be useful here. As the industrial revolution gave rise to modern capitalintensive manufacturing and more efficient distribution methods (e.g., rail roads), economies of scale became an important determinant of firm size – and in many cases minimum efficient scale was large relative to the size of the market. An important consequence was the increasing concentration of industries on a regional and national level. While concentration can bring concerns about market power, concentration by itself is no danger when barriers to entry are low – particularly when there are successful, efficient firms just on the other side of the border ready to serve your domestic market. Unfortunately for competition in Canada, Sir John A. McDonald’s National Policy Tariff of 1879 – an attempt to stimulate the development of a Canadian manufacturing sector through protection by high tariff barriers – blocked much foreign competition from entering concentrated Canadian markets. Protected by these barriers, domestic firms began to cooperate to achieve higher prices – creating ‘combines’ in Canada (‘trusts’ in the United States) to further control competition between cartel members. Significantly,
Canadian competition policy
247
collusion was in general not illegal under common law. (For more on this history see Trebilcock et al. 2002.) Something of a populist revolt in both the United States and Canada produced demands for controls on collusion – resulting in the first modern antitrust statutes in 1889 in Canada and 1890 in the United States. The first Canadian law was aimed squarely at price fixing, but amendments in later years added provisions for mergers and the abuse of dominance as well as some pricing practices like resale price maintenance. The history of Canadian competition policy has since see some successes but many difficulties. In the early years, the problems were largely related to the absence of a proper enforcement machinery; more recently a series of court decisions made enforcement extremely difficult. (For more on the history of enforcement see Ross 1998.) As a result there were loud and frequent calls for reform which ultimately led to substantial amendments in the 1970s and 1980s. The principle pieces of antitrust law in Canada now are the 1986 Competition Act and Competition Tribunal Act. There have been some amendments since 1986 (and more are contemplated) but the major features of the law remain as they were in 1986. The Competition Bureau is the agency of the Canadian federal government charged with enforcing the provisions of the Competition Act. The head of the Bureau is now called the ‘Commissioner of Competition’ and is frequently referred to in the business press as one of the most powerful men/women in Ottawa.7 It is important to recognize that competition law is framework law – that is, it is a law of general application, applying to (almost) all industries in all parts of the country. As a form of regulation of business activity it contrasts with the sort of industry-specific regulation we have experienced in areas such as telecommunications, broadcasting, energy, natural gas, and transportation. Economists see many advantages to ‘regulation by competition’ as supported by competition policy. Such regulation is less interventionist, less costly, and significantly less vulnerable to capture. I dared to suggest above that competition policy has more recently become ‘sexy’ (not a term used often to describe what economists do) for a few reasons. First, during the 1980s and 1990s there was a very significant wave of mergers that consolidated a number of important industries.8 Though it has slowed more recently, the wave has left a number of industries with levels of concentration that concern consumers and policy makers, sparking much discussion. Second, in this period there has also been observed a considerable amount of deregulation and privatization. Moving these firms and industries from alternative methods of regulation effectively expands ‘beat’ of the competition 7 Prior to March 1999 the Commissioner of Competition was known as the ‘Director of Investigation and Research.’ 8 Industries in which there has been significant consolidation by merger or exit would include financial services (with the large banks acquiring almost all of the trust companies and the largest investment dealers), airlines, popular media (newspapers, cable, radio, television), forestry, and various retailing fields.
248
T.W. Ross
TABLE 1 Large Canadian fines for price fixing: some examples Case
Firm
Fine
Bulk vitamins (1999)
Hoffman-LaRoche BASF Rhoˆne-Poulenc
$48,000,000 $18,000,000 $14,000,000 $95,500,000
UCAR Inc. SGL Carbon
$11,000,000 $12,500,000 $23,500,000
ADM Ajinomoto
$14,000,000 $ 3,500,000 $17,570,000 $ 4,700,000 $ 2,900,000 $ 2,000,000 $ 1,900,000 $11,575,000
Vitamin total Graphite electrodes (1999/2000) Graphite electrode total Lysine (1998) Lysine total Citric acid (1998)
Haarmann & Reimer Hoffman-LaRoche ADM Jungbunzlauer
Citric acid total Compressed gas (1991) Compressed gas total
Union Carbide Canadian Liquid Air Liquid Carbonic
$ $ $ $
1,700,000 1,700,000 1,700,000 6,460,000
cops. Consider areas that had previously been exclusively regulated by other means that are now being opened up to competition and regulation by competition law – transportation, telecommunications, financial services and energy. Significantly, what were probably the two largest Canadian competition cases of the last ten years – the two proposed bank mergers of 1998 and Air Canada’s acquisition of Canadian Airlines International in 1999 – arose in industries that had previously been extensively regulated. Third, and maybe most important, there have been some terrifically high-profile cases in each of the main areas of antitrust enforcement. With respect to collusion, some significant home-grown cartels in areas such as compressed gas, have been completely overshadowed by the uncovering of huge international price-fixing conspiracies in a large number of fields, including citric acid, lysine, and vitamins. Prosecution of these cartels has produced record fines in a number of jurisdictions and jail sentences for a few business people in Canada and (mostly) the United States. The fines in the vitamin price-fixing case, for example, approached $100 million in Canada alone, and, because this was a global conspiracy, the firms have had to pay very significant fines in the United States (US$910 million), Europe (855 million euro), and elsewhere as well.9 Table 1 provides further details about 9 The costs for the vitamin companies do not end with these fines. In both the United States and Canada the law provides for private parties damaged by price fixing to sue cartel members to recover damages. Both countries also permit class action lawsuits, facilitating the collection of damages even for relatively small-scale purchasers. A number of private actions are underway in both countries.
Canadian competition policy
249
TABLE 2 Some global cartels: estimated overcharges and fines to date Firms fined (or given leniency) Lysine 1992–95 Citric Acid 1991–95 Vitamins 1989–99 Sorbates 1979–96 Methionine 1989–99
Est. overcharge Global fines (millions $US) (millions $US)
ADM, Ajinomoto, Kyowa Hakko, etc. 240 ADM, Roche, Bayer, Jungbunzlauer, etc. 700 Roche, BASF, Aventis, Takeda, etc. 7,500 Aventis, Daicel, Nippon Gohsei, etc. 1,200 Degussa, Nippon Soda, Aventis 1,580
202 239 1,745 136 125
SOURCE: Connor (2001)
some of the very large fines that have been secured by Canadian authorities in recent price-fixing cases. Table 2, reproduced from Connor (2001), provides data on several of the global conspiracies including the total fines the firms have paid and the estimated overcharge during the cartel period (which could serve as a guide to the damages recoverable where such actions are permitted). While mergers in Canada – even in concentrated markets – will seldom generate popular interest beyond the business press, recently there have been a number of proposed and completed transactions that caused great concern and stimulated considerable debate. Chief among these are the aforementioned two large proposed mergers of major Canadian banks in 1998 and the 1999 acquisition of Canadian Airlines International by Air Canada. Public feelings about these deals were strong, and the debates that surrounded each helped to introduce ordinary Canadians to the function and purpose of competition law and to the Competition Bureau. With respect to cases involving allegations that a dominant firm abused its position to enhance or extend its market power, the leading case by far – some have called it the most important antitrust case in at least a generation – was that of the U.S. Department of Justice against Microsoft. The growing importance of personal computers in our daily lives and the large share of that business controlled by Microsoft guaranteed public interest as the Department of Justice made its case for behavioural controls or even structural remedies (i.e., an ATT-style ‘break-up’) to control Microsoft’s market power.10 Interest in competition policy is by no means limited to North America and Europe. Indeed, there has been an explosion of growth internationally in the field. Where 20 years ago there would have been only a handful of jurisdictions with 10 The initial decision in the Microsoft case included an order to break up Microsoft: U.S. v. Microsoft, Memorandum and Order, 7 June 2000, Civil Action Nos. 98–1232, 98–1233 (TPJ) (D.D. C. 2000). On appeal, the D.C. Circuit Court of Appeal allowed Microsoft’s appeal in part and remanded certain issues to a new District Court judge: U.S. v. Microsoft Corporation, No. 00–5212 (D.C. Cir. 28 June 2001). These cases and the field generally have captured the attention of media beyond newspapers and magazines. A major motion picture released in 2001, titled Antitrust, involved a plot in which a computer company is under investigation by antitrust authorities. The lysine conspiracy was detailed in a best-selling book by Eichenwald (2000), which some recent reports have suggested may also be made into a movie.
250
T.W. Ross
TABLE 3 International Competition Network (ICN) membership, 2003 ALBANIA ANDEAN COMMUNITY ARGENTINA ARMENIA AUSTRALIA AUSTRIA BARBADOS BELGIUM BRAZIL CANADA CHILE COSTA RICA CROATIA CYPRUS CZECH REPUBLIC DENMARK EFTA Surveillance Authority ESTONIA EUROPEAN UNION FINLAND FRANCE GERMANY GREECE HUNGARY ICELAND INDONESIA IRELAND ISRAEL ITALY JAMAICA JAPAN KENYA KOREA, REPUBLIC OF LATVIA
LITHUANIA MACEDONIA MALTA MEXICO NETHERLANDS NEW ZEALAND NORWAY PAKISTAN PANAMA PERU PHILIPPINES POLAND PORTUGAL ROMANIA RUSSIAN FEDERATION SLOVAK REPUBLIC SLOVENIA SOUTH AFRICA SPAIN SRI LANKA SWEDEN SWITZERLAND TAIWAN THAILAND TUNISIA TURKEY UKRAINE UNITED KINGDOM UNITED STATES UZBEKISTAN VENEZUELA YUGOSLAVIA ZAMBIA
enforced competition laws, the industry has become global. Table 3 provides a list of countries with national competition agencies that have joined the new International Competition Network.11 It is clear that competition policy has stretched across the world – virtually every region of the globe is represented in this list which includes wealthy nations, transition economies, and less-developed countries.12 11 It is my understanding that several more countries are preparing to join the ICN in the near future. 12 The point that competition policy is important is starting to sink in with Canadian business, and with good reason: (1) the fines are getting big, and many companies get in trouble because of the actions not of senior officials, but rather of those several steps down the ladder; (ii) jail terms have been used and will likely become more prominent; (iii) today’s competition law touches many firms every year, most through merger review; (iv) increased concentration in many industries means that the Competition Bureau will be paying more attention, looking for evidence of abusive behaviour; (v) the Bureau has shown a willingness to take on the largest firms in the country, and it seems to enjoy fairly broad public support when it does so; and (vi) further amendments are under consideration that will broaden the Bureau’s and the Commissioner’s powers.
Canadian competition policy
251
4. Theme 2: By and large we do competition policy well in Canada It has not always been a smooth process, but I would say that we now have one of the most modern, economically sound competition laws in the world. In keeping with economists’ vision for antitrust, its basic thrust is to promote efficient markets by protecting competition (not competitors). There are, however, elements of small-business protection in our law that may not serve competition, and some of this concerns me, a point to which I will return later. Our appreciation of competition and distaste for monopoly are for a host of reasons familiar to economists everywhere: much theoretical and empirical evidence indicates that with monopoly we (typically) get higher prices and lower outputs (with the resulting deadweight loss), higher costs, and reduced rates of innovation. Replacing a law that had been rendered unenforceable, no one can argue that the Competition Act does not have teeth. In addition to the numerous fines for price fixing discussed above, Bureau actions have been responsible for the restructuring or abandonment of a several mergers.13 The Bureau has also resolved a number of cases in the area of abuse of dominant position, including the important Interac case, which succeeded in opening up the electronic bank machine network to more financial services firms.14 In the Competition Bureau, Canadians have a professional enforcement agency, and, importantly, one that is independent and that defends its independence vigorously. Recent concerns about this independence based upon the ‘interference’ into competition matters by ministers from other departments in the bank and airline mergers (from the departments of Finance and Transportation, respectively) are somewhat exaggerated. The other departments became involved because of a view that there were issues of public interest beyond those related to competition that needed to be considered. We might properly ask whether the other considerations are valid in reviewing mergers such as these, and we might decide they are not. However, the Bureau’s right to speak to competition issues directly and to present its views publicly was never questioned. It was unfortunate that these two sets of mergers arose so close together in time – there are not many industries in which this sort of extra-Bureau review would take place. The result was 13 A few examples: the bank mergers proposed in 1998 were abandoned, in part because of concerns expressed by the Bureau; conditions were attached to the acquisition of Canadian Airlines International by Air Canada; and a number of mergers involving retailing (e.g., Imperial Oil - Texaco, Loblaws- Provigo, TD Bank - Canada Trust) were permitted conditional on the sale of selected outlets/branches in concentrated local markets. 14 This was resolved with a consent order. Canada (Director of Investigation and Research) v. Bank of Montreal et al., Competition Tribunal Reasons for Consent Order No. CT-95/02 (1995). Bureau victories in contested abuse of dominance cases include those involving NutraSweet and Laidlaw: Canada (Director of Investigation and Research) v. NutraSweet Company, (1989), 28 C.P.R. (3d) 316 (C.T.), and Canada (Director of Investigatioon and Research v. Laidlaw Waste Systems, (1991) 40 C.P.R. (3d) 289 (C.T.).
252
T.W. Ross
damage to the Bureau’s reputation for independence. This all said, there is reason to provide greater certainty that the Bureau’s independence will be maintained, a point that has been made by others. The Bureau resides in the federal government department Industry Canada and operates, given its considerable responsibilities, with a relatively modest budget and limited complement of staff.15 In the 2001–02 fiscal year, the Bureau received funding totalling about $51.9 million ($7.5 million from fees collected, the rest from the government) and employed about 380 full-time staff.16 Professional Bureau staff are in trained in law, business and economics; skills certainly appropriate to their assignments, and in general they are a hard working and committed group who do want to get the policy ‘right.’ The Bureau is well respected internationally, as evidenced by calls for assistance they receive from many foreign agencies working to draft or enforce new competition legislation – a recent list of countries asking for assistance includes: Mexico, Panama, South Korea, Brazil, South Africa, Taiwan, Indonesia, and Vietnam. While we have a modern competition law, it is not without its flaws. However, the recent battles over amendments have me concerned; after years of struggling to get any changes to Canadian competition law, we have an amendment a year now. Of some of these I certainly approve; however, some seem to be precipitous reactions to Bureau losses or difficulties. For example, we have special competition rules for airlines now, and an amendment is working its way through Parliament that would remove the efficiency defence in mergers, something most economists felt was a cornerstone of Canada’s merger law.17 5. Theme 3: Economists have made competition policy better Economists and economic theory have made two broad types of contributions towards better competition policy in Canada and elsewhere. First, they gave it a focus. In the earlier days of antitrust there were many pressures to use the law in the service of several goals, sometimes simultaneously – for example, as a weapon to fight big business, or to protect domestic firms from international competition, or simply to attack large firms for being large. Eventually, thanks in large measure to the work of economists, competition policy in Canada, the United States and many other places found a focus and purpose that found wide agreement: make the economy more efficient through the protection and 15 In addition to its primary role in the enforcement of the Competition Act, the Bureau has responsibilities related to the enforcement of the Textile Labelling Act, the Consumer Packaging and Labelling Act, and the Precious Metals Marking Act. 16 Of course, in some years the Bureau has more than paid for itself in the fines it has collected (which do not stay with the Bureau) largely in price-fixing cases. 17 The process by which this amendment to the efficiency defence came up is particularly troubling, given its significance – it was proposed as a private member’s bill (C-249) with little of the kind of prior consultation we had come to expect for such changes.
Canadian competition policy
253
enhancement of competitive market forces. That is, we came to recognize that the goal should be the protection of competition (not necessarily competitors) and that the ultimate concern was efficiency.18 The second type of contribution made by economists had to do with providing the tools needed to determine what kinds of behaviour and market structure do serve competition and efficiency. It has been through careful economic modelling that we have gained a better understanding of whether a given action or structure is good or bad for wealth creation and efficiency. Economic modelling has replaced evaluation based upon intuition, suspicion, and ideology. Interestingly, economists’ views about monopoly and competition policy have evolved over time. A review of this history might suggest something like four phases through which economic thinking has moved. Phase I: Strict laissez-faire While economists since at least Adam Smith, over 200 years ago, recognized that firms, given the opportunity, may not act in a wholly competitive way, the leading economists of the day did not present a united front in support of competition laws in the late nineteenth century. Stigler (1982) points out that several famous economists were lukewarm or cold on the need for an antitrust law – feeling that concentration and collusion was somewhat inevitable because of economies of scale and that collusion would be very unstable. Phase II: Deeply suspicious Beginning early in the twentieth century, economists began to see monopoly as an evil to be addressed at every turn by competition law. Out of this attitude was born the ‘inhospitality tradition,’ which viewed every action by large firms that did not fit the textbook model of perfect competition as an example of the exploitation and abuse of market power or an attempt to create a monopoly. The Canadian antitrust law of the period was relatively ineffective, but in the United States during this period there was some aggressive antitrust enforcement, leading to, for example, the breakup of American Tobacco and Standard Oil. Phase III: Chicago School Beginning in about the 1950s, members of the Chicago School of economics began to attack the very weak underpinnings of the inhospitality tradition, arguing that we should not condemn a practice without justification – and that this justification should come from sound economic analysis, including a theoretical model and empirical testing. Celebrated contributions here came 18 This said, there are differences between jurisdictions with regard to how they would deal with cases in which competition is reduced while efficiency is enhanced – as in the case of a merger that builds market power but dramatically reduces production costs at the same time.
254
T.W. Ross
from such Chicago-associated scholars as George Stigler, Aaron Director, Lester Telser, Howard Marvel, Richard Posner, Frank Easterbrook, and Robert Bork.19 As a result, a number of practices that had been viewed as anticompetitive were given new respect by Chicago economists who showed how they could in fact contribute positively to economic welfare.20 Phase IV: ‘Post-Chicago’ In the 1970s, we entered a period in which developments in the theoretical side of industrial organization gave us new tools with which to understand the behaviour of firms in markets – in particular, we were introduced to game theory and to the analysis of markets functioning with imperfect information.21 Using models more sophisticated in these ways than the models used by the early Chicago School, the pendulum has swung back against some of the practices about which Chicago had given us comfort. In my mind this does not diminish the contributions of the Chicago School which I think are undervalued in much of current discussion of ‘post-Chicago’ antitrust. 5.1. Chicago and ‘post-Chicago’ antitrust22 It is fashionable now to say that antitrust has gone beyond the Chicago School – that we have a new way of thinking about competition policy. I think that is wrong. In my mind, modern antitrust is build on a foundation laid by the Chicago School – while subsequent research may have changed some of the details, it has not altered the basic principles. My list of the important, enduring lessons of the Chicago School would include the following. i) Respect the power of the market Unregulated, free markets are still the best generators of wealth and prosperity, and we should be very careful before we intervene. Free markets will correct many competition problems on their own (e.g., through entry). Most of the really durable market imperfections (related to the state of competition) are the products of government actions. ii) Before you intervene, have a theory It is easy to forget that antitrust before the Chicago School was largely atheoretical. The Chicago approach was to use the power of simple price theory to try to understand the causes and consequences of firm behaviour. Chicago had some views about how to approach theoretical work, as well. 19 20 21 22
Interestingly, the last three on this list were formally trained as lawyers. Some of these contributions will be cited below. Both of these areas were to provide eventual Nobel Prize winners (in 1994 and 2001). For reasons that will be clear below, I do not really like the term ‘post-Chicago,’ because it might suggest that the Chicago School teachings are no longer relevant – and this I do not believe. Nevertheless, I concede that the term has come into general use as a label for much recent work in the field, so I will use it here as well.
Canadian competition policy
255
Chicago School theoretical models tended to be as simple as possible to illuminate the problem at hand, and there was a strong bias towards models that were empirically testable and were, in fact, tested. iii) Respect the limitations of enforcement officials and courts The Chicago School recognized that, in antitrust as well as regulation, governments and regulators do not operate without imperfections of their own. Even a wellcrafted law poorly enforced may do more harm than good. For example, overly aggressive antitrust enforcement risks chilling the climate for tough-but-fair competition. The twin requirements of detailed market information and economicanalytical sophistication may be beyond what we can expect from enforcement officials and the judiciary in the most complex cases (about which, to be honest, skilled economists may disagree). As a consequence, it is important not to expect too much of those charged with enforcing and adjudicating our competition law – while case-by-case review might always seem to be the theoretically correct approach to complicated competition issues, in the real world of imperfect enforcement this might not be second-best. There is therefore a considerable value in filters that allow us to quickly and easily screen out cases that should raise no antitrust concerns so that business can proceed under those conditions with reasonable certainty that they will not be found in violation of the law. iv) Protect competition not competitors There was a tendency before the Chicago School to associate competition with the number of competitors – a legacy of the early (atheoretical) structureconduct-performance school of industrial organization. Chicago reminded us that in good vigorous competition some firms will win and some will lose – and the losers will exit. Indeed, steady flows of entry and exit from a market can be a sign of healthy competition. Thus, as an example, even if the practices of a dominant firm or the exclusivity arrangements between vertically related firms are seen to disadvantage rivals, the relevant test for public policy purposes is whether or not competition and efficiency are served. Antitrust law should not be about propping up weaker competitors. If I could sum up the contributions of what has been called the ‘postChicago antitrust,’ it would be this: get the right model. As suggested above, Chicago School models tended to be the simplest possible to make progress on the particular question at hand. This would seem like the right way to start, and we should not be surprised that subsequent research, using more complicated (but often more realistic) models has pointed to different conclusions in some cases. Chicago models were built on strong assumptions to simplify the analysis, and interesting new insights have been obtained by relaxing some of those assumptions. To point to just three examples: our understanding of predatory pricing, tied selling, and collusion has been much improved by further theoretical work.
256
T.W. Ross
5.2. Some Important Recent Contributions Let me elaborate briefly on some of the areas of competition policy to which economic theory and empirical work have given us a better understanding of what firms are up to and help us to sort the good from the bad. This review will be highly selective. I am pleased to be able to report that many significant contributions in the field have been made by Canadians and I will take this opportunity to highlight some of those.23 5.2.1. Predatory pricing Once widely regarded as common, views changed dramatically when John McGee (1958) debunked the Standard Oil story that had suggested that the American oil monopoly had been built through predation.24 Deprived of their best piece of ‘anecdotal evidence,’ economists became very suspicious that predation would ever be a rational strategy. Why should a firm sell at prices below its own cost – a very expensive strategy – when it could just buy out its rival? We have a number of answers to this question now, most based upon models of imperfect information. For example, even if predatory pricing in one market is very costly, it could be worthwhile if a firm develops a ‘‘reputation for toughness’’ that frightens away other potential entrants (see, e.g. Milgrom and Roberts 1982). Also, as in Poitevin (1989), if the predator has significantly deeper pockets than the victim, a threat to predate might persuade the rival to exit or sell out at a low price. Despite this progress, predatory pricing still presents one of the most bedevilling problems in competition policy: When is a price so low it is predatory, and when is it just good, tough competition? 5.2.2. Collusion and Interdependence Understanding the conditions that support collusion is most important (perhaps surprisingly) for merger law. For it is when we review mergers that we have to predict the degree of competition post merger (and whether the merger will lead to a substantial prevention or lessening of competition). The economics literature that began with Stigler’s (1964) seminal contribution has attempted to determine conditions under which explicit or tacit collusion is more likely to be observed. Stigler’s work emphasized the essential Prisoner’s Dilemma nature of the one-shot cartel game, while newer research has studied collusion using supergame models in which repeated play provides a credible punishment mechanism in support of collusion. Into these models 23 In my definition of ‘Canadian’ I include scholars at Canadian universities (regardless of their actual citizenship) and Canadian citizens working outside the country. I will focus here exclusively on contributions by economists, but it should be noted that there are a number of lawyers (in academia and private practice) producing important scholarly work on competition law as well. 24 This is not to suggest that there were no anticompetitive practices associated with the building of the Standard Oil empire. See Granitz and Klein (1996) for an interesting recent contribution.
Canadian competition policy
257
imperfect information (about costs or demand) has been introduced to produce equilibria in which cartels break down and reform. Particularly notable here is the work by Green and Porter (1984). There is another literature that has looked at a number of practices that firms may adopt (even unilaterally) that could have the effect of reducing competitive tension in a market – practices such as most-favoured-customer clauses or meetingcompetition clauses. Salop (1986) provides a general discussion of these strategies and Chen (1995), in this journal, contains an interesting model that focuses on meeting-competition clauses. Salop (1986) explains that various practices can be useful to firms looking to support tacit collusion by improving information flows (to facilitate early detection and punishment of cheating) or by changing the payoffs that firms gain from cheating. For example, most-favoured-customer (MFC) clauses, when added to sale contracts make it very costly for firms to make selective discounts to steal business from rivals – since the lower price must be made available to all customers who have MFC clauses in their contracts. Chen (1995) shows that the low price guarantees implicit in meeting- and beating-competition clauses may or not reduce competition. Games in which firms first commit to their ‘meeting’ or ‘beating-competition’ strategies, then pick prices can have multiple Nash equilibria. Their effect on prices (and whether firms will choose to use them) will depend in part on the solution concept used. Collusion is possible, since they can remove the incentive for a firm to try to steal a rival’s customer – the customer will be bound to report any price to its current supplier and let that supplier match it and keep the customer’s business.
5.2.3. Exclusive Dealing, Resale Price Maintenance, and other Vertical Restraints Exclusive dealing involves the establishment of an exclusive relationship between firms vertically related in the distribution channel (e.g., a manufacturer and a retailer).25 Resale price maintenance (RPM) is the practice by a manufacturer of dictating the price (or a minimum or maximum price) at which its product must be resold. Initially, exclusive dealing was viewed suspiciously, since it prevented rival firms from working with one or the other of the firms in the relationship. Thus, for example, a manufacturer could not have its product distributed by retailers committed to other manufacturers through exclusive dealing arrangements. If broadly practised in an industry, this could indeed create problems. That is, if every manufacturer and every retailer in some industry were locked into exclusive dealing arrangements, a new entrant manufacturer would not be able to secure any distribution, and a new retailer would not get supply. 25 The exclusivity can run either or both ways. That is, the retailer can commit to not selling the products of other manufacturers and/or the manufacturer can commit to not selling its product to other retailers.
258
T.W. Ross
RPM (of the minimum-price variety, in particular) was viewed suspiciously because it seemed to limit the ability of retailers to charge lower prices and thereby to limit competition at the retail level. However, thanks to much important work by (among others) Lester Telser (1960), Howard Marvel (1982), Frank Mathewson and Ralph Winter (1984, 1986, 1987), Scott Masten and Edward Snyder (1993), Mike Whinston (1992), and others, we have a better picture of why firms might enter into these arrangements.26 In some cases the restrictions protect firms’ specialized investments. Marvel (1982) described the famous Standard Fashions case in which Standard Fashions spent a great deal of money designing dress patterns, only some of which would turn out to be successful. The problem it had was that the successful patterns could easily be copied by firms not spending the development money. To protect the higher prices it needed to recover its development costs, Standard Fashion required department stores to not carry the patterns of copy-cats (to stop the retailer from switching customers to the lower-priced copy-cats). Similarly, if a manufacturer builds traffic for a retailer with an expensive advertising campaign, it will not want to grant the retailer the ability to switch consumers to no-name rivals who did nothing to build traffic but who charge slightly lower prices. Mathewson and Winter (1984, 1986, 1987) have made a number of important contributions to this literature, studying the efficiency properties of a number of vertical restraints in environments in which price is not the only retail variable about which manufacturers care. RPM has been seen as a way to provide incentives for retailers to compete on service levels when that is more important (on the margin) to manufacturers than competition on price. While the original insight here is attributed to Telser [1960], Winter [1993] provides a very nice treatment of the trade-off manufacturers face when trying to provide incentives for retails to offer both high levels of service and low prices, and demonstrates that under reasonable conditions, price competition will be too intense relative to service competition. 5.2.4. Economic theory and empirical testing It goes without saying among economists that the best empirical work is shaped by solid theory. The theory-based approach favoured by empirical economists has also greatly advanced competition policy. For example, estimates of demand elasticities and marginal costs are married with oligopoly models to predict the effect of a particular merger. These merger simulations are becoming common elements of merger review in the United States and Canada as the techniques for conducting them improve. Significant contributions (with Canadian connections) to the literature here include papers by
26 Mathewson and Winter (1998) have a particularly nice summary of the literature with respect to resale price maintenance.
Canadian competition policy
259
Capps, Church and Love (2003), Hendricks and McAfee (2000), and Pinske and Slade (forthcoming). In an interesting recent example, which appeared in this journal, McIntosh (2002) simulated the effects of the proposed large bank mergers, generating results that suggested customers would have benefited were the mergers allowed to proceed. The significant amount of available data generated by public auctions and the bidding for public contracts has provided rich opportunities for testing the advancing theory of auctions. Predictably, the potential for collusion in auctions and bidding rings has drawn attention from antitrust scholars and enforcement officials. For example, interesting work by Robert Porter and others has examined data from auctions in an attempt to determine if the bidding was competitive or collusive (see, e.g., Porter and Zona 1999). Finally, significant empirical progress has been made towards testing the available theories of cartel stability. While most cartels are secretive enterprises (owing to their illegality), making it difficult to get the information necessary to study their stability, a body of interesting work has focused on cartels from time periods or in situations where they were legal, and data therefore more readily available.27 For example, in Dick (1996) the author studies the stability of legal cartels registered under the Webb-Pomerene Export Trade Act in the United States which permits cartels for export, provided there is no harm to domestic competition. He finds that these cartels were more stable when they sold to smaller buyers, when members represented a very large fraction of the market, when demand was stable, and when they involved cooperation on more than just price fixing.28 6. Theme 4: Competition policy faces significant challenges Canada has a modern competition law, significant scholarly activity and experienced, professional enforcement. As well positioned as Canada may be to emerge as a world-leader in the design and implementation of competition policy, I have concerns about our system going forward. First, the expanding scope for competition law and limited Bureau resources are making the job more difficult all the time.29 In addition, significant changes 27 In some countries (e.g., Canada and the United States) cartels operated solely for the purpose of serving export markets may be legal. 28 For example, some of the cartels set up common sales agencies, some undertook joint promotional activities, and some negotiated with foreign governments and agencies on behalf of their members. 29 Stanbury (1998) makes a good case that the Bureau has insufficient resources to do its job. While its budget has grown somewhat from the end date of Stanbury’s data set (budget year 1995/96), so have its responsibilities. In addition to a number of very high-profile and expensive cases in recent years and a merger wave that has only recently (and perhaps temporarily) slowed, the Bureau has been handed responsibility for the enforcement of three additional regulatory acts (Consumer Packaging and Labelling Act, Textile Labelling Act, Precious Metals Marking Act) and been given additional authority to regulate telemarketing (under the Competition Act).
260
T.W. Ross
in the structure of the Canadian economy present new challenges: an antitrust policy designed for a twentieth-century economy will not match the needs of the twenty-first century. Deregulation, privatization, globalization, network externalities, e-commerce, new organization forms including new types of strategic alliances and joint ventures – all have implications for the appropriate design and enforcement of competition policy. As a result of their limited and uncertain resources, the Bureau has been forced to pick the cases with most significant public benefits, and pass on many others that might have been ‘winnable.’ This is not entirely a bad thing, of course; an agency with too many resources and underemployed staff could make much economic mischief. However, when the Bureau comes up against something really big, like the proposed bank mergers or the Air Canada/ Canadian merger, the resources are not always there to do the job demanded of them. The result is that some good cases may be abandoned, and others may be less well prepared than they should be. The shortage of resources means things slow down, too, as when work on cases is put on hold until more resources can be found to support them. In some cases, special financial allocations above the base budget have been made to a starving Bureau, and, while certainly welcome, such a method of funding threatens the Bureau’s independence. A government unhappy with the Bureau’s unwillingness to ‘play ball’ on a controversial file might feel less inclined to make supplementary allocations regardless of the merit of the request.30 Let me offer a personal view of the most pressing challenges, beyond the financial, facing the Bureau at this time. 6.1. Cartels, strategic alliances and joint ventures Recognizing the potential benefits of strategic alliances and joint ventures and seeing that current laws on cooperation between competitors may be ‘chilling’ legitimate cooperative activity, many antitrust agencies are looking to change their enforcement approach or even their laws to accommodate these horizontal arrangements if they are not anticompetitive. To this purpose, the Bureau issued a Strategic Alliances Bulletin in 1995, which they have suggested they may revise to provide greater clarity. The American and European agencies have recently issued their own guidelines. These guidelines stress the acknowledged benefits of certain agreements between competitors, for example research and development joint ventures that share skills and spread risks; code-sharing alliances between airlines that allow them to provide better and expanded services to passengers; joint operating agreements that allow 30 The Commissioner can, on occasion, find him/herself in conflict with other government departments or even cabinet ministers. For example, the last Commissioner and the Minister of Transport did not see eye to eye on the ways to address problems in the Canadian airline industry.
Canadian competition policy
261
independent newspapers to share the considerable fixed costs of running modern large-scale printing presses; and international alliances that permit firms to export more quickly and easily into foreign markets with the assistance of local partners. At the same time, the guidelines try to warn prospective alliance partners when the authorities will see the potential for anticompetitive harm in their arrangements. It is important to note that Canada may get a new law on agreements between competitors. An earlier private member’s bill (Bill C-472) took up this challenge (but was not enacted) and reform to the Competition Act’s conspiracy provisions continue to be debated.31 The bill, and similar proposals, would have created two tracks for agreements between competitors – the naked price-fixing type agreements will actually be easier to prosecute (with the need to establish that the agreement lessened competition ‘unduly’ removed), while those agreements that have other intentions and effects will get a much more sympathetic hearing, being evaluated rather like we currently review mergers. As mentioned, this bill did not pass, but the House Industry Committee has encouraged the government to proceed in this direction. There has since been further study of the two-track model and the government discussion paper describing a series of possible amendments and being circulated for public comment now includes such a provision (Canada 2003). 6.2. Private enforcement, what is the right role? In contrast to the United States, where most antitrust activity involves private cases in which the government agencies are not involved, Canada has relatively little private enforcement. Private actions do exist in Canada, and their frequency is increasing even as their scope is limited by legislation that permits private parties to seek (single) damages only when they result from a violation of one of the Competition Act’s criminal provisions. With this increased private activity have come calls for expanded rights of private access to the Competition Tribunal. Indeed, recent amendments now permit private parties to approach the Competition Tribunal directly to seek relief (but not damages) under some of the civil provisions. We will need to decide how far down this ‘American path’ we want to go. On the one hand, we take some pride, and may enjoy some considerable advantages, from having a less litigious society. On the other, when the Bureau does not have the personnel or resources to take on all the meritorious cases with which it is presented, there is an ‘access to justice’ issue that argues that private parties should have some way of pressing claims on their own.32 This will 31 There has been a healthy professional debate about the best approach to horizontal agreements generally, including my own efforts with Tim Kennish (Ross 1991; Kennish and Ross 1997) and Warner and Trebilcock 1993). 32 There is also an argument that those who suffer harm from anticompetitive acts have a right to restitution.
262
T.W. Ross
be the source of some debate as the current reform provisions do contemplate a wider scope for private enforcement (including damages). An important concern will have to be that we avoid opening the door to the use of competition laws as a strategic competitive weapon, as is often alleged to be the case in the United States. 6.3. Changing goals There is reason to worry that the focus we have been able to put in Canada on the use of competition laws to promote efficiency may be at risk. First, there are concerns about how the law will treat efficiencies created by merging partners and how they are to be traded off against the higher prices that might follow an increase in market power. The view until recently at the Bureau has been to consider shareholders and consumers as social equals for the purpose of adding up the costs and benefits of mergers – but this has changed. A current private member’s bill in Parliament would remove the efficiency defence in mergers and replace it with a section indicating that the Tribunal ‘may’ consider efficiencies as a ‘factor’ in the determination of whether a merger lessens competition. Second, there has always been an element of small business protection in the Act and to the extent provisions that protect the competitive process help small business that is appropriate and consistent with a focus on efficiency. However, there are those who would use the Competition Act as a club to beat large firms when small firms are threatened by changing economics in their industry. The Act should not be about protecting competitors: it should be directed at protecting competition. This point combined with the previous one related to the expanded role for private actions together point to a more general concern. Now that we have a law with teeth, we need to worry about the use of antitrust by powerful groups to serve goals other than competition and efficiency. This brings me to the last concern on my list. In my view, it is the most important – if it is not addressed, I am not optimistic the rest can be. 6.4. The quality of economics done at the Bureau In part because of current salary levels and conditions of employment, the Bureau is having trouble getting and holding on to good economists. While there have been and continue to be some very talented economists working for the Bureau, there are not enough. Of the 383 authorized full-time staff (reported in the 2001/02 Annual Report) only a few are economists with advanced graduate (i.e., doctoral) training. For an agency charged with such broad-based economic regulation, this is just not satisfactory. For a comparison, I have borrowed a table from Sanderson (2003), reproduced as table 4, which shows the percentage of total staff and of professional staff represented by economists in a number of competition agencies. Canada’s numbers are
Canadian competition policy
263
TABLE 4 Competition agency staffing, selected countries 2001–02
Country
Total Economists Economists Total agency professional Economists as a per cent as a per cent of total employment staff employed of total staff professional staff
Australia* 382 Belgium 42 Brazil* 398 Canada 383 Czech Republic 113 Denmark 140 European Union 570 Finland 60 France 110 Germany 298 Hungary 111 Japan 576 Kenya 37 South Korea* 444 Mexico 207 The Netherlands 200 Poland 219 Slovak Republic 73 Sweden 110 Switzerland 45 Taiwan 223 Thailand 129 Turkey 307 UK: OFT 461 United States: FTC 1049 United States: DOJ* 801 Venezuela 18
297 35 167 254 100 101 260 40 40 165 85 unknown 24 259 152 110 210 51 85 40 223 129 141 106 595 567 16
96 16 60 13 38 55 70 20 5 49 39 80 5 51 38 35 51 24 47 20 38 51 44 40 71 57 6
25.1 38.1 15.1 3.4 33.6 39.3 12.3 33.3 4.5 16.4 35.1 13.9 13.5 11.5 18.4 17.5 23.3 32.9 42.7 44.4 17.0 39.5 14.3 8.7 6.8 7.1 33.3
32.3 45.7 35.9 5.1 38.0 54.5 26.9 50.0 12.5 29.7 45.9 unknown 20.8 19.7 25.0 31.8 24.3 47.1 55.3 50.0 17.0 39.5 31.2 37.7 11.9 10.1 37.5
*Figures are from 2000. SOURCE: Sanderson (2003)
remarkably low both in terms of absolute numbers of economists and in the economists’ share of total agency staff.33 Conceding that the Bureau serves a law enforcement function that demands a high level of legal and related expertise, it is nevertheless essential that it have the skills necessary to get the economics right. It is not unusual for there to be a certain tension in an antitrust agency between the economists and the lawyers, indeed this can be healthy. But there must be balance, and my concern is now that this balance has been lost. While it is true that the Bureau can hire outside economist-consultants for cases or selected policy work, this is no substitute 33 To be fair, some of the differences in these numbers could be driven by different definitions of just who is an ‘economist.’ While a doctoral-level education might be required to earn this designation in Canada, a Masters degree or even an undergraduate degree in economics might be enough in some other agencies.
264
T.W. Ross
for a strong internal capability needed to advise on general policy matters, case selection, and to screen and work with the outside experts. Related to the problem of the low number of professional economists on staff is the following concern. Economics does not have the profile within the Bureau it deserves, with the risk that legal issues and the ‘winnability of cases’ can come to completely dominate Bureau decision making, at the expense of analyses that determine whether taking cases is in the public interest by furthering competition and efficiency. The Competition Act is a very broad piece of economic regulation – its ‘Purpose Clause’ (section 1.1) makes this quite clear. To apply this regulation well requires a high level of economic, as well as legal, sophistication. The economic knowledge to be applied to competition cases comes in two forms. First, there is the basic application of wellknown ‘textbook’ theories. Then there is the development of new theories or explanations when the textbook theories do not quite fit or when it is not clear which theory applies in a given situation. While many Bureau officers will do some of each of these kinds of work, the latter is often what is passed over to the very small economics group in the Bureau or to outside consultants. I think the Bureau needs more internal capability for this second kind of work and a mechanism for assuring the quality of all the economics work, whether it is provided internally or externally. Even a quick review of recent important antitrust cases in Canada and elsewhere demonstrates that the level of economic sophistication required for optimal enforcement is high and rising. (i) For example, the Microsoft and Interac cases highlighted the importance of network externalities. (ii) The airline predation cases in Canada, the United States and Europe challenge us to determine where the appropriate line should be drawn to separate predatory behaviour from that which is merely aggressively competitive. (iii) The Superior Propane case forced us to consider how much weight to attach to efficiencies in a merger in which competition is substantially lessened. (iv) The proposed bank mergers and a number of recent European merger cases have pushed us to refine our theories of non-collusive interdependence. (v) The Nutrasweet, Nielsen and Laidlaw cases demonstrated the possible role of contracts as behavioural barriers to entry.34 (vi) The proposed General Electric – Honeywell merger raised issues related to bundling and ‘portfolio effects.’ (vii) Cases like Staples in the United States and the large proposed bank mergers in Canada point to the increased use of econometrics to understand the potential effects of mergers (Staples, Inc. v. the Federal Trade Commission, 970 F.S. 1066, [D.D.C] 1997). Beyond cases, the Bureau is charged with serving as a ‘champion’ for competition within government – providing valuable consumer-friendly input related to regulated industries and industries in which there are significant 34 The Nielsen case is Director of Investigation and Research (Canada) v. The D & B Companies of Canada Ltd., Competition Tribunal and Orders No. CT-94/1 (1995).
Canadian competition policy
265
government-imposed barriers to entry (such as foreign ownership limits). Finally, the Bureau plays a central role in the design of its own legislation. The thought of a Bureau under-skilled in economics taking on these imposing responsibilities is indeed a troubling one. I have no reason to believe that careful economic analysis of cases and policy has been de-emphasized through any conscious decision of senior Bureau management; indeed, I have known a number of past directors/ commissioners who clearly appreciated the importance of getting the economics right. Much of the problem may be due to high rates of turnover, leaving too few senior economists at the Bureau to provide the necessary leadership. Whatever the cause, there is a very real danger that the Bureau will follow the European Commission’s example and become embarrassed by the poor quality of its economics. The Commission has just had three merger decisions in a row reversed on appeal by the Court of First Instance, in part because of questions about the quality of its economic analysis. (For a discussion of the Commission’s recent problems, see Kuhn 2003). From the Superior Propane case, there is evidence that the Bureau has already started down this sad path.35 Reviews of the Competition Bureau by the international publication Global Competition Review, have also called into question the quality of its economic analysis.36 To its credit, the European Commission has recognized its problems and taken strong steps to bolster its economics capability. It is time for the Bureau to reinvest in economics as well. In recent memory all the regular directors/commissioners have been lawyers; indeed, it seems that the accepted wisdom (at least among the Canadian competition bar) is that the position requires a lawyer’s training and skills. This is far from clear to me, and in fact many competition agencies in other jurisdictions, such as the United Kingdom, Australia, New Zealand, and even that of the European Commission, have been headed by economists. Certainly one significant step towards raising the profile of economics at the Bureau would be to appoint an economist as Commissioner. While we should not give up hope of this happening, I am not optimistic that it will be soon.37 A second-best strategy would start with the appointment of a chief economist (CE).38 In my model, the CE would be the second-in-command of the Bureau, reflecting the significance attached to the economics function. His or
35 For a discussion of an important error made by the Bureau in the case, see Mathewson and Winter (2000). 36 However, we should not make too much of these reviews. Little is known of how scientific and careful these surveys are. 37 In November 2003 the government announced that it had appointed Sheridan Scott, a communications lawyer, to be the next Commissioner of Competition, effective early in 2004. Ms Scott had been serving as Bell Canada’s head of regulatory affairs. 38 I first made this suggestion in an appearance before the House of Commons Standing Committee on Industry in May 2000.
266
T.W. Ross
her principle responsibility would be to ensure the quality of the economic analysis going into Bureau work More specifically, the CE could: 1. be the main adviser to the Commissioner on economic matters as they relate to cases and policy, reviewing and monitoring the quality of work done by Bureau economists; 2. work to recruit new economists of the highest calibre for the Bureau and mentor and supervise those economists working in the Bureau; 3. work to keep the economics staff up to date with respect to new learning in antitrust economics; 4. be a champion for economics at the Bureau – fighting to see that the resources and people are in place to do work of the highest quality, and to make sure that the economic analysis of a case gets the attention it deserves; 5. be a connection to the external economics community. In this capacity, the CE would be able to encourage research into important questions and to build networks of external advisers from the academic, consulting, and business communities in much the same way past commissioners have done with the legal community. It is interesting to note that a significant part of the European Commission’s recent effort to improve the quality of its economics involves the creation of a position for a ‘Chief Competition Economist,’ who will report directly to the Director-General of Competition and will have a dedicated staff of 10 specialized economists.39 7. Conclusions There is much that is interesting for economists about competition policy: it is important to a modern economy; done well it uses the best economic theory and the most advanced empirical techniques; and it is constantly providing new challenges as businesses invent new strategies and market structures continually evolve. Canada has the potential to be a world leader in competition policy (and to a certain extent we are already) – we have a strong law, an international orientation, capable enforcement, and skilled academic economists working on questions of domestic and international importance. However, the role of economics and economists is not assured. As economists, we may have to fight to see that competition policy in Canada continues to recognize that ours is a foundational discipline upon which good competition policy is built. We must resist the tendency in some quarters,
39 On 16 July 2003 the Commission announced the appointment of the well-known antitrust economist, Lars-Hendrik Ro¨ller, from Humboldt University in Berlin, to a three-year term. In the press release announcing this appointment, Competition Commissioner Mario Monti was quoted as saying, ‘The appointment of a Chief Economist forms an integral part of my commitment to strengthen further the economic underpinnings of our competition analysis.’
Canadian competition policy
267
whether deliberate or not, to view economics as a service function to be employed only when convenient in the support of actions taken for other reasons. We can hope that the new Commissioner will take a fresh look at the role and status of economics in the Bureau and will see the need for some significant reinvestment.
References Bork, Robert (1978) The Antitrust Paradox (New York: Basic Books) Canada (2003) ‘Options for amending the Competition Act: fostering a competitive marketplace,’ Discussion Paper, June Capps, O., Jeffrey Church, and H.A. Love (2003) ‘Specification issues and confidence intervals in unilateral price effects analysis,’ Journal of Econometrics 113, 3–31 Chen, Zhiqi (1995) ‘How low is a guaranteed low price?’ Canadian Journal of Economics 28, 683–701 Connor, John (2001) Global Price Fixing: Our Customers are the Enemy (Norwell, MA: Kluwer Academic) Dick, Andrew (1996) ‘When are cartels stable contracts?’ Journal of Law and Economics 39, 241–283 Director, Aaron, and Edward Levi (1956) ‘Law and the future: trade regulation,’ Northwestern University Law Review 51, 281–96 Easterbrook, Frank (1981) ‘Predatory strategies and counterstrategies,’ University of Chicago Law Review 48, 263–337 — (1984) ‘The limits of antitrust,’ Texas Law Review 63, 1–40 Eichenwald, Kurt (2000) The Informant (New York: Broadway Books) Gorecki, Paul, and Stanbury, W.T. (1984) The Objectives of Canadian Competition Policy 1888–1983 (Montreal: Institute for Research on Public Policy) Granitz, Elizabeth, and Benjamin Klein (1996) ‘Monopolization by ‘‘raising rivals’ costs’’: the Standard Oil case,’ Journal of Law and Economics 39, 1–47 Green, Edward, and Robert Porter (1984) ‘Noncooperative collusion under imperfect price information,’ Econometrica 52, 87–100 Hendricks, Ken, and R. Preston McAfee (2002) ‘A theory of bilateral oligopoly with applications to vertical mergers,’ mimeo, University of Texas – Austin, June Kennish, Tim, and Thomas Ross (1997) ‘Toward a new Canadian approach to agreements between competitors,’ Canadian Business Law Journal 28, 22–68 Ku¨hn, Kai-Uwe (2002) ‘Reforming European Merger Review: Targeting Problem Areas in Policy Outcomes’, 2 Journal of Industry Competition and Trade, 311–64 Marvel, Howard (1982) ‘Exclusive Dealing,’ Journal of Law and Economics 25, 1–25 Masten, Scott, and Edward Snyder (1993) ‘United States versus United Shoe Machinery Corporation: on the merits,’ Journal of Law and Economics 36, 33–70 Mathewson, Frank, and Ralph Winter (1984) ‘An economic theory of vertical restraints,’ RAND Journal of Economics 15, 27–38 — (1986) ‘The economics of vertical restraints in distribution,’ in New Developments in the Analysis of Market Structure, ed. J. Stiglitz and G.F. Mathewson (Cambridge, MA: MIT Press) — (1987) ‘The competitive effects of vertical agreements: comment,’ American Economic Review 77, 1057–62
268
T.W. Ross
— (1998) ‘The law and economics of resale price maintenance,’ Review of Industrial Organization 13, 57–84 — (2000) ‘The analysis of efficiencies in Superior Propane: correct criterion, incorrectly applied.’ Canadian Competition Record 20, McGee, John (1958) ‘Predatory price cutting: the standard oil (N.J.) case,’ Journal of Law and Economics 1, 137–69 McIntosh, James (2002) ‘A welfare analysis of Canadian chartered bank mergers,’ Canadian Journal of Economics 35, 457–75 Milgrom, Paul, and John Roberts (1982) ‘Predation, reputation and entry deterrence,’ Journal of Economic Theory 27, 280–312 Pinske, Joris, and Margaret Slade ‘Mergers, brand competition and price of a pint,’ forthcoming, European Economic Review Poitevin, Michel (1989) ‘Financial signaling and the ‘‘deep-pocket’’ argument,’ RAND Journal of Economics 20, 26–40 Porter, Robert and Douglas Zona (1999) ‘Ohio school milk markets: an analysis of bidding,’ RAND Journal of Economics 30, 263–88 Posner, Richard (2001) Antitrust Law, 2nd ed. (Chicago: University of Chicago Press) Ross, Thomas (1991) ‘Proposals for a new Canadian competition law on conspiracy,’ Antitrust Bulletin 36, 851–82 — (1998) ‘Introduction: the evolution of competition law in Canada,’ Review of Industrial Organization 13, 1–23 Salop, Steven (1986) ‘Practices that credibly facilitate collusion,’ in New Developments in the Analysis of Market Structure, ed. J. Stiglitz and G.F. Mathewson (Cambridge, MA: MIT Press) Sanderson, Margaret (2003) ‘Strengthening the economic team within a competition agency,’ paper prepared for the Global Competition Forum of the International Bar Association, Mexico City, 26 June Stanbury, W. T. (1998) ‘Expanding responsibilities and declining resources: the strategic responses of the Competition Bureau,’ Review of Industrial Organization 13, 205–41 Stigler, George (1964) ‘A theory of oligopoly,’ Journal of Political Economy 72, 44–61 — (1982) ‘The economists and the problem of monopoly,’ American Economic Review 72, 1–11 Telser, Lester (1960) ‘Why should manufacturers want fair trade?’ Journal of Law and Economics 3, 86–105 Trebilcock, Michael (1986) The Common Law of Restraint of Trade (Toronto: Carswell) Trebilcock, Michael, Ralph Winter, Paul Collins, and Edward Iacobucci (2002) The Law and Economics of Canadian Competition Policy (Toronto: University of Toronto Press) Warner, Presley, and Michael Trebilcock (1993) ‘Rethinking price-fixing law,’ McGill Law Journal 38, 679–723 Whinston, Michael (1990) ‘Tying, foreclosure, and exclusion,’ American Economic Review 80, 837–59 Winter, Ralph (1993) ‘Vertical control and price versus nonprice competition,’ Quarterly Journal of Economics 108, 61–76
