Case Documents

JAN-27-1900 15:14 P.02 CT - 2003-005 THE COMPETITION TRIBUNAL IN THE MATTER OF the Competition Act, R.S. 1985, c. C-34; as amended; IN THE MATTER OF an application by Canadian Waste Services Holdings Inc., Canadian Waste Services Inc. and Waste Management Inc. under s. 106 of the Competition Act;

AND 1N THE MATTER OF the acquisition by Canadian Waste Servi~~JJ),f,pL the Ridge Landfill in Blenheim, Ontario from Browning-Ferris Industri+ Ltd. · i "r'r'"' "·" ' I u ··''····;,, '·"" - .· :c~ r) ; .<. [ BETWEEN: c ., ~ . '.l l 'I 'r '. CANADIAN WASTE SERVICES HOLDINGS INC., CANADIAN WASTE SERVICES INC. and WASTE MANAGEMENT, INC. Applicants

-and-COMMISSIONER OF COMPETmON Respondent Affidavit of Michael R. Baye Sworn September 26, 2003

I, MICHAEL R. BAYE, of the City of Bloomington, in the State of Indiana, one of the United States of America. MAKE OATH AND SWEAR:

1. I am the Bert Elwert Professor of Business Economics and Public Policy at Indiana University. My fields of specialization within economics include industrial organization, microeconomic theory, and game theory. From 1988 to the presen~ I have served intermittently as a consultant to the Antitrust Division of the United States Department of Justice on a variety of mergers within the waste industry.

2. I was retained by the Commissioner of Competition to examine the competitive effects of the acquisition by Canadian Waste Services Inc. ("CWS") of Browning-Ferris Industries Ltd.,s

09/26/2003 FRI 09:48 [TX/RX NO 6425] 141002

JAN-27-1900 15:15 P.03 RESPONSE 2 ("'BFIL") Ridge Landfill (the "Ridge,1 on the markets for waste generated in the Greater Toronto Area ("GTA") and the Chatham-Kent area ("CK.A"), and gave expert evidence in proceedings before the Competition Tribwial in proceedings brought by the Commissioner relating to that acquisition (in file number CT 2000/002).

3. I have been retained to examine the economic relevance of the "changes in circumstances" alleged by CWS in this proceeding, and in particular whether any of these alleged changes in circumstances would lead me to change my opinion regarding the anticompetitive effects ofCWS' acquisition of the Ridge landfill.

4. I attach as Exhibit "A" to this affidavit my report setting out my opinion on the economic relevance of the "changes in circumstances" alleged by CWS.

5. I attach my curriculum vitae as Exhibit ''B" to this affidavit. SWORN BEFORE ME at the City of ) ·Bloomington, in the State of Indiana, one ) of the United States of America, this ) ~ b day of September, 2003. ) ..... ..:· · ..- ::·:~~··-.... ) ) -_><~~:h R. am~ ) MICHAEL R. BA YE · · · . -~~sici .. n .... er, ) etc. :. .. . . '. . .. . : . : ) '. ""- ~,.-;- ........... ..: ,.. • .. _ 0 M.; -- 0 : 0 • , ---- --~_ ............: .....· OM . .. : - --· ~... · ... · - ,., ... ·' "'···~ ..... ... :~. . 09/26/2003 FRI 09: 48 [TX/RX NO 6425] I4J 003

JAN-27-1900 15:15 IN THE MATTER OF the Competition Act, R.S. 1985, c.C-34; as amended;

IN THE MATTER OF an application by Canadian Waste Services Holdings Inc., Canadian Waste Services Inc. and Waste Management Inc. under s. 106 of the Competition Act;

AND IN THE MATTER OF the acquisition by Canadian Waste Services Inc. of the Ridge Landfill in Blenheim, Ontario from Browning-Ferris Industries Ltd.

BETWEEN: Canadian Waste Services Holdings, Inc., Canadian Waste Services Inc., and Waste Management, Inc.

Applicants -and-Commissioner of Competition Respondent

Affidavit of Micha.el R. Baye FRASER MILNER LLP 1 First Canadian Place 100 King Street West. Suite 3900 Toronto, ON MSX 1B2

Donald B. Houston Tel: 416·863-4620 Fax: 416-863-4592

AFFLECK GREENE ORR LLP Banisters & Solicitors One First Canadian Place Suite 840, P .0. Box 489 Toronto, Ontario MSX lES

W. Michael G. Osborne Tel: (416) 360-5919 Fax: (416) 360-5960

DEPARTMENT OF JUSTICE Place de Portage, Phase 1 SO Victoria Street, 22nd Floor Hull, Quebec KIA OC9

Andre Brantz Tel: (819) 997-3325

Fax: (819) 953-9267

Solicitors for the Respondent,·the Commissioner of Competition

TOTAL P.04 09/26/2003 FRI 09: 48 [TX/RX NO 6425] l4J 004

P.04 CT - 2003-005 THE COMPETITION TRIBUNAL

Exhibit A: Report of Michael R. Baye, September 26, 2003 I. Qualifications 1. I am the Bert Elwert Professor of Business Economics and Public Policy at Indiana University. I received my B.S. degree in economics from Texas A&M University in 1980 and a Ph. D. in economics from Purdue University in 1983. I received a Fulbright award during 1985-1986 to conduct research and to present lectures on spatial pricing at Erasmus University Rotterdam. 2. I have taught Ph. D. level courses in industrial organization at Texas A&M University, Penn State University, Indiana University, and the New Economic School in Moscow. In addition, I have held visiting appointments at a number of other universities, including Cambridge, Oxford, and Bonn. My fields of specialization within economics include industrial organization, microeconomic theory, and game theory. 3. My research on the economics of pricing has been published in the American Economic Review, Journal of Political Economy, Econometrica and other leading economics journals. I am the author of a best-selling textbook, Managerial Economics and Business Strategy (McGraw-Hill, 4 th edition, 2003), which has been translated into several languages and is used around the world at both the undergraduate and graduate levels. I am the editor of Advances in Applied Microeconomics, and also serve on the editorial boards of the Economics of Governance and the Journal of Public Policy and Marketing. 4. I have been involved in this case since the fall of 1999, when I was retained by the Canadian Competition Bureau to examine the competitive effects of the acquisition by Canadian Waste Services Inc. (CWS) of Browning-Ferris Industries Ltd. (BFIL). Since that time, I have provided both written reports and oral testimony before the Tribunal in hearings regarding Report of Michael R. Baye, September 26, 2003 Page 1 of 20

the initial complaint as well as in the remedy hearing. In addition, from 1988 to the present, I have served intermittently as a consultant to the Antitrust Division of the United States Department of Justice on a variety of mergers within the waste industry. II. Scope 5. The Bureau has asked me to examine the economic relevance of the “changes in circumstances” alleged by CWS, and in particular whether any of the following allegations by CWS would lead me to change my opinion regarding the anticompetitive effects of CWS’ acquisition of the Ridge landfill: (1) delays in the expansions at the Warwick and Richmond landfills; (2) the absence of any price reductions within southern Ontario for commercial, institutional and industrial (ICI) waste from the Greater Toronto Area (GTA); and (3) the absence of any reductions in shipments of ICI waste from the GTA to Michigan. III. Summary of Opinions 6. None of these allegations change my fundamental conclusion that disposal prices for ICI waste generated in the GTA and Chatham-Kent Area (CKA) critically depend on whether the Ridge is owned and operated by CWS or an independent competitor. These markets are spatial oligopolies in which a handful of landfills price discriminate among customers based on how close a given customer is to alternative facilities. Giving CWS full control over pricing strategies at the Ridge would eliminate competition between the Ridge and CWS landfills, thereby enhancing CWS’ market power in these markets and permitting it to charge significantly higher tipping fees than would otherwise prevail in these markets. 7. From an economic perspective, it is inappropriate to view changes in variables that CWS can manipulate – such as capacity, prices, and waste flows – as “changes in circumstances.” Doing so creates perverse incentives for firms involved in antitrust litigation to behave Report of Michael R. Baye, September 26, 2003 Page 2 of 20

opportunistically, and will lead to moral hazard problems that undermine future antitrust policy. 8. While the likelihood of expansions at Richmond and Warwick were central to my opinions regarding a likely substantial prevention of competition in the GTA market, anticipation of these expansions played no role in my opinions regarding the CKA market, nor are they central to my opinion that the acquisition of the Ridge by CWS would substantially lessen competition for ICI waste from the GTA. 9. CWS does not allege in its Application that its proposed expansions at Richmond and Warwick will not take place. Even if this was the case, one is left with the fact that control of the Ridge by CWS would substantially lessen competition in both the GTA and CKA markets. Even in this scenario, divestiture of the Ridge remains the appropriate remedy. 10. The economic reasons underlying these opinions are summarized below. IV. Economic Critique of CWS’ Arguments A. It is Inappropriate to View Changes in Endogenous Variables as a “Change in Circumstances” in an Attempt to Overturn a Divestiture Order 11. Economists call a variable that is influenced by the actions of an interested party an endogenous variable. For example, the number of times a motorist is involved in an accident is endogenous, since it is influenced by his efforts to drive safely. This does not mean that a motorist has complete control over accidents; even a driver who is careful to obey all traffic laws may have an accident due to factors outside of his control. However, a driver who takes unsafe actions – such as speeding or driving while intoxicated – is more likely to have an accident than one who obeys traffic laws. 12. As I noted in the Remedy hearing, Report of Michael R. Baye, September 26, 2003 Page 3 of 20

“The level of capacity in southern Ontario is an endogenous variable influenced by the behavior of CWS.” Report of Michael R. Baye, Remedy Hearing, Paragraph 15a, 23 May 2001. 13. This circumstance has not changed; as a general matter, time and money are the main requirements for overcoming obstacles that get in the way of CWS’ proposed expansions at the Warwick and Richmond landfills. 14. Furthermore, landfill expansions are not the only economic means by which CWS can create additional capacity for ICI waste from the GTA. I noted in my Reply in the Remedy hearing that: “CWS can increase its own capacity for ICI waste from the GTA by taking in lower volumes of non-GTA waste or by shipping some waste from its own collection operations to landfills that it owns in Michigan… the increase in market power that would result if CWS acquires the Ridge would likely provide it with an incentive to do so.” Michael R. Baye’s Reply to the Expert Report of Christopher Vellturo, 13 June 2001, Paragraph 5b. 15. In the remedy hearing, CWS proposed a remedy that would have constrained to some extent its pricing behavior until such time that there was no longer any excess capacity in the market. They proposed gaining full control of the Ridge once a capacity shortfall was established. 16. I pointed out in my Remedy Report that, since the amount of capacity in the market is strongly influenced by the actions of CWS, it would be inappropriate for the Tribunal to base antitrust decisions on ex post (future) levels of capacity. In particular, I noted that doing so Report of Michael R. Baye, September 26, 2003 Page 4 of 20

would lead to moral hazard 1 and give CWS sharp incentives to engage in opportunistic behavior to distort observed capacity levels so that it could ultimately gain full control the Ridge and raise prices. 17. In that report, I specifically noted if the Tribunal adopted CWS’ proposed remedy, it would be providing CWS with strong incentives to delay proposed expansions, speed up the closure of its existing landfills, or manipulate waste flows among its various disposal facilities in order to “demonstrate” to the Tribunal that its concerns regarding a substantial prevention of competition – which arise when there is excess capacity – were no longer relevant. Once released from antitrust scrutiny, it would then be able to readjust waste flows, follow through with planned expansions, and raise prices to take advantage of the market power gained through complete control of the Ridge landfill. 18. Ultimately, the Tribunal opted against CWS’ proposed remedy in favor of complete divestiture of the Ridge. From an economic perspective, however, CWS’ current allegations that “circumstances have changed” place the Tribunal and the Commissioner in exactly the same position that would have occurred had the Tribunal accepted their proposed remedy in the first place. In effect, CWS argues in its current Application that, since there is no excess capacity in 2003, it should be permitted to gain full control of the Ridge. 19. For reasons similar to those articulated in my 2001 report, economic theory indicates that it is inappropriate to view changes in endogenous variables – variables that are strongly influenced by CWS – as a change in circumstances. 1 Moral hazard is the economic term referring to a situation where one party takes a “hidden” or “unobservable” action that benefits him or her at the expense of another party. For example, in insurance contracts, the probability of loss is typically endogenous and depends on unobservable effort expended by the insured to avoid a loss. In this context, moral hazard occurs if the insurance contract reduces the incentive of the insured party to take actions to avoid a loss: Hidden actions (reduced effort) benefit the insured at the expense of the insurance company. See, for instance, Michael R. Baye, Managerial Economics and Business Strategy, 2003, Chapter 12.

Report of Michael R. Baye, September 26, 2003 Page 5 of 20

20. A decision by the Tribunal to view changes in prices, waste flows, or delays in capacity expansions as “changes in circumstances” would have broad implications well beyond this case. If the Tribunal establishes a precedent for viewing changes in endogenous variables as changes in circumstances, it would create perverse incentives for future firms subject to divestiture orders. Economic theory indicates the resulting moral hazard would not only undermine future antitrust policy, but potentially lead to countless rounds of litigation. Future firms subject to hold separate orders would have sharp incentives to behave opportunistically and to strategically manipulate business decisions in the short-run. 21. These perverse effects occur regardless of whether CWS has behaved opportunistically in this case. 22. To illustrate, imagine that the Tribunal in this case establishes a precedent for viewing changes in endogenous variables as changes in circumstances. Further suppose a future Tribunal finds that the price charged by a hypothetical profit-maximizing firm would fall if it required merging parties to divest an asset. Since the price charged by the hypothetical firm is an endogenous variable – that is, within its control – it would have an economic incentive to delay divestiture of the hold-separate asset through legal proceedings, temporarily raise price above current levels, and then file a new application before the Tribunal that “circumstances have changed” because prices have increased, not decreased as predicted by the Tribunal. This strategy – if such a precedent were established in the present case – would enable the hypothetical firm to circuitously gain control of the asset, thereby gaining market power through opportunistic litigation. 23. More generally, when factors contributing to an antitrust authority’s divestiture order are endogenous to the parties – capacities, prices, qualities, sales volumes, and so on – permitting Report of Michael R. Baye, September 26, 2003 Page 6 of 20

the parties to claim that ex post changes in these variables are “changes in circumstances” would create perverse incentives, undermine antitrust policy, and result in socially inefficient litigation. 24. It is important to stress that I am not asserting – and I have no knowledge – that CWS has taken deliberate actions to delay the expansions at Richmond or Warwick. Except for the potential impact on the current proceedings, their incentive to do so was largely eliminated when the Tribunal chose not to adopt CWS’ proposed remedy. 25. However, the economic theory of moral hazard – which is the heart of the incentive issue – indicates that asymmetric information makes it difficult to know the extent to which observed changes in endogenous variables (such as capacity, prices, or waste flows) are due to strategic actions by an interested party or due to exogenous factors. Indeed, moral hazard stems from the problem of unobservable actions, and it is well-known in the economics literature that parties benefiting from such actions have a strong incentive to “hide” them from those adversely affected. B. Ex Post Delays in the Richmond and Warwick Landfill Expansions are Not Economically Relevant “Changes in Circumstances” 26. It is common knowledge that the market for disposal services entails numerous uncertainties. As I indicated in my initial report of 11 October 2000, these risks include not only uncertainties which affect the time and money required to obtain approvals from various government authorities, but a variety of other legal, political, and environmental risks. Numerous events can alter the political and economic landscape, thereby speeding up or delaying attempts to create new landfills or expand existing sites.

Report of Michael R. Baye, September 26, 2003 Page 7 of 20

27. As I noted in Paragraphs 95 and 97-98 of my initial report, these uncertainties affect market structure by making it difficult for new entrants to enter the market for disposal services. It is well-documented in the economics literature that such barriers to entry enhance the ability of a merged entity to exercise the market power gained through an acquisition. 28. Viewed in this context, delays in the Richmond and Warwick landfill expansions are not changes in the economic environment. Rather, they are manifestations of the uncertainties that make entry into disposal markets difficult, contribute to CWS’ market power, and enhance CWS’ ability to profitably raise prices if it gained control of the Ridge landfill. 29. More broadly, in the face of uncertainty, ex post realizations almost always differ from ex ante expectations. For example, relative to expectations formed in 2001, ex post increases in 2003 southern Ontario capacity for ICI waste from the GTA have occurred due to realizations of greater capacity at the Petrolia landfill, while ex post decreases in 2003 capacity have occurred due to delays in CWS’ proposed expansions at the Richmond and Warwick sites. Differences in expected and realized values of random variables are an artifact of uncertainty. 30. It is well-documented in the economics literature that optimal decisions in the face of uncertainty are forward looking, not backward looking. A classic example of suboptimal backward-looking behavior is the insurance fallacy. This occurs, for instance, when an individual decides not to renew his homeowner’s policy today, simply because his house did not burn down during the past year. 31. Likewise, economic theory indicates that it would be a mistake to permit CWS to gain control of the Ridge today, simply because additional capacity was not realized during the past year. Report of Michael R. Baye, September 26, 2003 Page 8 of 20

32. When I wrote my initial report in 2000, it was not certain that additional capacity would be available at the Richmond and Warwick landfills. However, based on evidence at the time, both the Tribunal and I concluded that capacity expansions were likely at these two disposal facilities. Indeed, the expectation – based on CWS documents – was that the expansions would be operational by 2003. 33. Ex post, the expansions were not operational by 2003. This does not imply that the expansions are unlikely, nor does it imply that CWS should be permitted to control the Ridge. In light of inherent uncertainties, economic theory indicates that delays in and of themselves are not changes in circumstances. 34. Economic theory dictates that optimal decisions with respect to any future uncertainties in capacities at the Warwick and Richmond landfills be forward looking. As noted in Section A above, a change in an endogenous variable should not be viewed as a change in circumstances. However, if the delays to the expansions are going to be considered, the relevant economic question today is not whether there were past delays, but whether it is no longer likely that the expansions will take place. 35. While I have no specific expertise to aid the Tribunal in determining the ultimate timing of future expansions, CWS’ pleadings in this case do not allege that the expansions at Richmond and Warwick have become unlikely. With regard to timing: a. In paragraphs 55 and 56 of its current Application, CWS anticipates that the Richmond landfill expansion could be operational in 2006 or 2007. b. In paragraphs 69 and 70 of its current Application, CWS anticipates that the Warwick landfill expansion could be operational in 2005 or 2006.

Report of Michael R. Baye, September 26, 2003 Page 9 of 20

36. While there is uncertainty regarding the timing of these expansions, based on forward looking expectations, the decision horizon today regarding the Richmond and Warwick expansions appears similar to when I filed my first report in late 2000. 37. More generally, the point is not the exact date at which additional capacity would be available, but that looking forward from today (late 2003), it has not become unlikely that additional capacity will be available at the Richmond and Warwick landfills. 38. While uncertainties inherent in the market have delayed expansions at Richmond and Warwick, from an economic perspective, these delays do not alter my opinion on the matter; nor do they change optimal policy in this case. By arguing that ex post delays in the Warwick and Richmond expansions are a change in circumstances, CWS is committing the insurance fallacy. 39. Using the optimal forward looking approach, the data provided by CWS in its Application (and discussed in Section C below) implies that the economic basis for divestiture is stronger in 2003 than it was in 2001. At the time of the Tribunal’s original decision, successful expansions at both the Richmond and Warwick landfills were necessary in order for excess capacity to emerge in the market. Today, due to the unanticipated expansion at the Petrolia landfill, only one of the expansions is required to generate excess capacity in the market for ICI waste from the GTA. C. Ignoring Endogeneity, Moral Hazard, and Opportunism, Data Provided by CWS in its Application Indicates That Divestiture Remains Necessary and Appropriate 40. Based on data in CWS’ Application in the present case, and considering the economic impact of location and increased concentration on tipping fees in a spatial oligopoly, it is my opinion that divestiture of the Ridge remains necessary and appropriate. Report of Michael R. Baye, September 26, 2003 Page 10 of 20