PUBLIC THE COMPETITION TRIBUNAL IN THE MATTER OF the Competition Act, R.S.C. 1985, c. C-34; as amended; IN THE MATTER OF an Application by the Commissioner of Competition for an order under section 92 of the Competition Act;
AND IN THE MATTER OF the acquisition by United Grain Growers Limited of Agricore Cooperative Ltd., a company engaged in the grain handling business.
BETWEEN: THE COMMISSIONER OF COMPETITION Applicant
-and-UNITED GRAIN GROWERS LIMITED AFFIDAVIT OF HALLDOR P. PALSSON \ t 1. I, Halldor P. Palsson, of the City of Ottawa, in the District of Ottawa-Carleton, in the Province of Ontario, Public Servant, MAKE OATH AND SAY: 2. I am a senior economist in the Enforcement Economics Division of the Competition Bureau at Industry Canada. I was assigned to be the investigative economist in the above captioned matter. In the course of my duties I analyzed the competitive process in the grain-handling industry in local markets in Western Canada and in certain canola seed purchasing and -1-
·:"'· Respondent l : t··~~~ • '" •. ' . ' • ', .. j I . . . ' '-~·; "' ''. ~, . I ~',t.~ :; ·:.~_,: .. :}. .'. .·~-· ._·\ :' .. . : _.-~.-.. ~:: r DEC 1 7 2001 (/ t .c : l:: .. ~- ;., ... , 1rd2.J
processing markets. Counsel for the Commissioner of Competition asked me to undertake an economic analysis of how UGO and Agricore interact in these local markets in Western Canada. Attached hereto as Exhibit "A" is a true copy of the Report prepared pursuant to the aforesaid request. 3. I have a Ph.D. in economics and my area of specialization is industrial organization. I have studied issues of competition and market power in a number of industries. Included as Appendix "A" to the aforesaid Report is a true copy of my Curriculum Vitae.
SWORN BEFORE ME, at the City of Hull, in the Province of Quebec, this 10th day of December 2001.
( ~ · ~ COMMISSIONER OF OATHS HALLDOR P. PALSSON -2-
) ) ) )
THIS IS EXHIBIT:.. ... 8. ....................................T O THE Exhibit ''A'' -PUBLIC ~h.i:..t.I.b~~·"""""'f!{'l!ii.'l;s~·s.a.iJ.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.:~:~ swoj) B~FOE M E THIS ..... 1..0. ...t .h ......... DAY Report by Halldor P. Palsson, ;~'~'~ . . 6;;Jf.::~~:=~····-·-·-· COMMISSI ER FOR OAT~~ i·at.••• .• c.i,1e a ser!J1 "'"' ,. ~s· -~~"' .,,i': '%~ December 10th, 2001 ~Cf ' '?o ,. ! JOSEE BRUNET l: °! II 96 746 ~~ •«? ~: •< •. o 0 :, , ,J, . ;;_"• ~~· Qualifications and Introduction .. ,.,.aa;cia11e ~11~•· ''•" ..... 1. I have been an economist with the Economic Policy and Enforcement Division of the Competition Bureau (the "Bureau") since 1992. In my position I am responsible for providing economic analysis of cases at the Competition Bureau, including, in this case, the likely competitive impact in Canada of the acquisition by United Grain Growers ("UGO") of Agricore Cooperative Ltd. ("Agricore") (the "Acquisition").
2. I have a Ph.D. in economics and my area of specialization is industrial organization economics. I have studied issues of competition and market power in a number of industries including, most recently, the cement industry. A copy of my C.V. is attached hereto as Appendix "A".
3. In conducting my analysis in this case, I examined: (1) the records of the Mergers Branch of the Competition Bureau pertaining to its review of the Acquisition, and (2) economic literature relevant to the grain industry. In addition, I participated with Bureau competition law officers in meetings and interviews with several industry participants and consulted with Dr. Daryl F. Kraft and Mr. John DePape, who did work on behalf of the Bureau which is more fully described below. These sources have been used to form my opinions and draw my conclusions on how markets relevant to these products operate in Western Canada.
4. In analyzing the competitive effects of the Acquisition, I proceeded in four steps. First, I defined the relevant product and geographic markets, having particular regard to features of market supply, as well as demand. Second, I examined the position of Agricore United in
the relevant markets. Third, I examined the likely impact on competition of the Acquisition, and, in particular, the likely effect on prices if Agricore were to, in effect, exit from the market as a competitive alternative as a result of the Acquisition. Finally, I considered possible remedies.
5. Pursuant to the terms of a Merger Agreement between UGG and Agricore dated July 30, 2001, UGG and Agricore agreed to merge by way of a court-approved plan of arrangement ("Plan of Arrangement") under section 192 of the Canada Business Corporations Act. The Plan of Arrangement provided that UGG would acquire control of all business assets of Agricore. These assets included:
(a) whole or partial interests in Port Terminal facilities in Vancouver, Prince Rupert and Thunder Bay;
(b) whole or partial interests in Western Canadian Primary Grain Elevator facilities;
(c) Agro-business interests (crop inputs supplies and services); and (d) .a 16.67% interest in CanAmera Foods Limited Partnership ("CanAmera").
6. The Acquisition was completed on November 1, 2001 and as of that date UGG and Agricore have been carrying on business as Agricore United.
A. Primary Grain Elevators I. Industry Overview Introduction 7. Pursuant to the Acquisition, UGG acquired all of Agricore's primary grain elevator assets. 8. The grain industry in Western Canada has a number of elements and various participants. 2
They include: (a) farmers, who produce grain; (b) grain handling companies such as Agricore United (and prior to the Acquisition, UGG and Agricore) who purchase grain from farmers, either as agents of the Canadian Wheat Board ("CWB") or on their own account, at the grain handling companies' primary grain elevators which are located across the Prairies. There are two kinds of primary elevators - traditional wooden elevators and high through-put elevators ("HTP"). HTPs have substantially greater capacity than traditional elevators. Each of these two types of elevators is described in greater detail below;
(c ) the CWB, which is, by law, the only purchaser of wheat and barley, that is either to be exported from Canada or for domestic human consumption. Grain meeting that description is referred to as "CWB Grain" - all other grain is referred to as "non-CWB grain." (hereinafter, where no distinction is required between CWB grain and non-CWB grain, it will be referred to simply as "grain"). Grain handling companies merchandise all non-CWB grain that they purchase;
(d ) the railways (i.e., Canadian National Railway and the Canadian Pacific Railway) both of which transport CWB and non-CWB grain from primary elevators to, among other places, port terminals located in Vancouver, Prince Rupert and Thunder Bay;
(e) port terminals, where grain from the Prairies is delivered for storage, in some cases "cleaning," and ultimately, for shipping; and
(f) vessels onto which grain is loaded for export. 9. The grain industry in Western Canada comprises production regions in the three Prairie Provinces and the Peace River region, which is an area that traverses Northern Alberta and British Columbia.
10. As noted above, grain handling companies, including Agricore United (and prior to the 3
Acquisition, UGG and Agricore), purchase grain from farmers either on their own account or as agents of the CWB.
11. Grain handling begins when a producer's grain is transported from the farm to a primary elevator where the grain is weighed, graded and dockage (i.e., foreign material in the grain such as dirt and straw) is assessed. At a primary elevator, grain handling comprises receiving, grading, possibly cleaning, elevation, storage and loading grain onto rail cars. The farmer is then issued a cheque for the grain delivered, based upon the then current market price for the grade, less charges levied for delivery (if made by the grain company), elevation, dockage and cleaning (if applicable).
12. Elevators are licensed by their static storage capacity in tonnes. The amount of grain received at a given elevator is referred to as that elevator's handle. The number of times an elevator ships or handles the equivalent of its total storage capacity in a year is referred to as the number of "turns" or the "tum ratio." In 1992-93, the average tum ratio for primary elevators was 5. That figure increased to 5.92 in 1996-97, but then declined to approximately 5.5for1999-00. In a 1999 study, Trimac Consulting Services Ltd. estimated that in 1998-99, the 186 HTPs for which it had data averaged 7.3 turns, with approximately 10% achieving more than 20 turns.
13. According to the Grain Handling Infrastructure Rationalization Study (1998), Western Canadian primary elevator storage capacity peaked in 1970 at about 11 million tonnes. The number of elevators, deli very points (stations where more than one elevator could be located) and storage capacity has declined sharply over the years.
Crop Year Primary Elevators Delivery Points 1970 4,947 1992-93 1,500 1997-98 1,027 2000-01 681 4
Storage Capacity (tonnes) 1,899 11,000,000 967 7,200,000 811 6,600,000 NIA 6,260,730
14. As of July 11, 2001 there were 681 licensed primary elevators still in operation across Western Canada. Pre-merger, the principal grain handling companies in Canada were: Agricore, UGG, Saskatchewan Wheat Pool ("SWP"), Pioneer Grain (the grain handling arm of James Richardson International ("JRI")), Cargill Limited ("Cargill"), N.M. Paterson & Sons Limited ("Paterson"), Louis Dreyfus Canada Ltd. ("Louis Dreyfus") and Parrish & Heimbecker, Limited ("P&H"). The table below depicts the breakdown of elevators owned by the various grain handling companies:
Grain Co Storage % Storage Elevators % Elevators (tonnes)
Agricore 1,432,340 22.9 UGG 751,020 12.0 SWP 1,517,200 24.2 Pioneer Grain 561,740 Cargill 479,430 Paterson 290,040 Louis Dreyfus 259,860 P&H 251,110 Others 717,990 11.5 I Total I 6~260~130 I 100 I Inland Terminals 15. There are ten farmer owned and operated inland terminals, most of which are located in Saskatchewan. These facilities, which are included in the "others" category in the table above, provide grain handling services and are in direct competition with the major grain handling companies. However it should be noted that most of these inland terminals have some form of affiliation with a major grain handling company.
Producer Cars 16. In addition to grain-handling companies, farmers also have the option to use "producer cars" to ship their grain to port. Provided that a farmer has sufficient quota left on his delivery
5
207 30.4 87 12.8 151 22.2 9.0 78 11.5 7.6 45 6.6 4.6 48 7.0 4.2 11 1.6 4.0 23 3.4 31 4.5 681 I 100 I
contract with the CWB (which is entered into each crop year and uses an acreage quota system, based on CWB sales), he or she can apply to the Canadian Grain Commission ("CGC") for a producer car. By using a producer car, farmers save on elevation fees which could amount to approximately $1,000 on a carload of grain. Historically, farmers have filled producer cars one at a time, using an auger system to transfer the grain from a truck to a grain car. Recently, however, in an attempt to obtain a Multi-Car Incentive ("MCI") rebate (as described below in paragraph 24), a group of farmers have invested in a 25 car siding with approximately 3,000 tonnes of condominium storage (i.e. the farmer owns the grain and merely stores it on site). However these facilities are not licensed elevators and therefore they do not provide all of the services of a grain handling elevator such as blending and in some cases cleaning. Historically, only approximately one percent of all grain produced in Western Canada has been shipped by producer cars, however in light of recent developments to ship multi-car producer cars, it is anticipated that producer car shipments could double or triple.
Regulatory Environment 17. The grain handling industry is regulated by the CGC and the CWB pursuant to the Canada Grain Act and the Canadian Wheat Board Act, respectively.
18. The CGC is responsible for ensuring that grain produced in Canada meets certain quality standards. CGC inspectors monitor grain quality and enforce standards in respect of the grain delivered to primary elevators and port terminals.
19. The CWB is by law the sole purchaser and seller of CWB grains (i.e., wheat and barley for export and domestic human consumption). Grain handling companies purchase CWB grains from farmers as agents of the CWB at prices fixed periodically by the CWB. The majority of all non-CWB grains (i.e., grains such as canola, peas and lentils) are purchased at primary elevators by grain handling companies at market prices.
6
Changes in the Industry 20. The grain handling industry has undergone significant structural and regulatory changes in recent years. Many older, wooden, primary elevators have been closed and in their place, HTP elevators have been constructed.
21. In 1980 the average primary elevator in Western Canada had about 3,500 tonnes of storage and about 10 rail car spots. By comparison an HTP typically has 15,000 to 40,000 tonnes of storage and 50 tolOO rail car "spots." The term "spot" refers to the capacity of a facility to accommodate rail cars at a given time. For example, a facility with a 50 car "spot" could accommodate 50 rail cars at one time.
22. It is generally accepted in the industry, with certain limited exceptions, that the storage space for a given elevator should be approximately four times the total capacity of the rail cars that fill that elevator's car spot. One rail car holds approximately 90 tonnes of grain. Therefore, for example, an elevator with a 25 car spot loading 2,250 (90 x 25) tonnes of grain at one time would require approximately 9,000 tonnes of storage. In keeping with the foregoing, an elevator with a 50 car spot would require approximately 18,000 tonnes of storage space and a 100 car spot about 36,000 tonnes.
23. HTP elevators invariably have a larger number of rail car spots and are capable of handling much greater volumes of grain than traditional primary elevators. HTPs also have to draw grain from a larger area than the traditional primary elevators.
24. The incentives to build new HTP primary elevators are best understood by examining the basic economics of primary elevators. The commodity mix in the draw area and the MCI rebates offered by the railways drives the operation of a primary elevator. MCI rebates are offered by railways to grain handling companies based on their ability to provide, within a set period of time following the delivery of empty rail cars, loaded blocks of 25, 50 and 100 rail cars for transport from individual elevators. In order to obtain the rebate, the loaded
7
block of cars, whether 25, 50 or 100, must also be unloaded within a fixed period of time. Since the supply of grain cars can be a bottleneck in the system, the loading and unloading time constraints are intended to expedite the handling of rail cars so as to minimize their turnaround time. The MCI rebate scheme is set out in the following table.
Rail Incentives Rail Car Block Rail Incentive 25 to 49 $1 per tonne 50 to 99 $4 per tonne 100 $6 per tonne 25. To illustrate certain differences in the economics of HTP and traditional primary elevators, I constructed an example, based on information I obtained from interviews with, and submissions from, industry participants depicting how major revenue and cost components of each type of facility are viewed by the industry.
26. In my example, the tariff on CWB wheat is assumed to be $12.25 per tonne and trucking costs to move the grain from a farm to a primary elevator are assumed to be $3.50 to $4.00 per tonne. Trucking costs are somewhat lower for smaller elevators since they generally draw grain from the immediate vicinity of the elevator. The average discounted cost of elevation services on CWB wheat to producers is $8.25 to $8.75 per tonne (i.e., tariff less net trucking costs). Operators of HTP elevators tend to offer greater trucking incentives in the form of rebates to farmers and, because of their relatively high number of car spots, are able to obtain higher MCI revenues from the railways. An HTP generally has lower variable cost because it is more efficient than older elevators. Although fixed costs tend to be higher for HTPs on a per tonne basis, they are similar to traditional elevators since they handle a greater volume. For illustrative purposes, a comparison of the typical revenue and cost per tonne for traditional and HTP primary elevators, operating at average levels, is set out in the following table.
8
Incentive Conditions Load Time Unload Time 10 Hours 48 Hours 10 Hours 48 Hours 24 Hours 48 Hours
Small Elevator <25 cars ($ per tonne) Tariff CWB Wheat $12.25 Net Trucking (3.50) Variable Cost (3.25) Rail Rebate Cleaning Margin Gross Margin $5.50 Fixed Cost (4.00) Depreciation (1.25) Margin $0.25 27. Other structural changes to the industry include the abandonment of certain secondary rail-lines by the railways, a process that has been both hastened by, and contributed to, the closure of a number of older primary elevators. These changes have resulted in a ?rain-handling network, which, in economic terms, is more efficient (i.e., the cost of transporting grain from the grain elevator to port terminals has been reduced). The logistics of handling and transportation from primary elevator to port terminal to vessel position is referred to as "pipeline management" in the grain handling industry.
CWB Business and Grain Car Allocation 28. The CWB recently adopted a tendering system pursuant to which grain handling companies can tender to supply grain and ship to a specified destination. Rail cars come with successful tenders. During the current crop year, the CWB will tender a minimum of 25% of its grain handling requirement to grain handling companies, rising to a minimum of 50% for the 2002-03 crop year. The allocation of rail cars for CWB non-tendered requirements among the grain handling companies is based on: (1) an 18-week running average of CWB grain through-put at each primary elevator; and (2) the balance of outstanding CWB quota from farmers who last delivered to the grain company's elevators and are assumed to continue to do so.
9
NewHTP>lOO Range of Opinion ($per tonne) ($ per tonne) $12.25 (4.00) $0-$5.00 (2.75) 2.00-4.00 0 4.00 0-6.00 0 3.50 0-3.50 $13.00 (4.00) 2.00-5.00 (3.50) 1.00-7.00 $5.50
Transportation 29. As noted above, incentives in the form of rebates are sometimes offered by grain handling companies to farmers as a means of encouraging them to transport grain significant distances to the relevant company's primary elevators. Most of the grain that is delivered to primary grain elevators is then transported via rail to domestic users (e.g. at flour mills, barley malters, feed mills, feed lots and oilseed crushers), US consumers or port terminals for export to other countries.
30. A key factor in ongoing rationalization of traditional primary grain elevators, is the adoption of MCI rates by the railways (described in paragraph 24 above). MCI rates which range from $1-$6/tonne are a substantial fraction of the rail rate to Vancouver, which typically ranges from $28-$45/tonne depending on the point of origin in Western Canada.
II. Relevant Product Market 31. The major grain crops in Western Canada are wheat, canola and barley, which together account for about 90% of the total grain production. Other crops grown in Western Canada are flax, oats, rye and specialty crops which include canary and mustard seed, lentils and field peas.
32. The purchasing and handling of grain is a candidate relevant product market. In performing my review in this case, I considered smaller product markets. For example, I considered whether CWB grain and non-CWB grain should be considered two separate product markets for the following reasons. First, the distinction between CWB grain and non-CWB grain is important because the grain handling companies purchase CWB grain as agents of the CWB, whereas they purchase non-CWB grain on their own account. Thus, the handling services for milling wheat, durum wheat, malting barley and feed grains delivered for the CWB's account or CWB grain is a candidate relevant product market. Second, the grain handling companies, not the CWB, market and sell non-CWB grains, such as canola, feed barley and
10
feed wheat and other grains (e.g. mustard and canary seed, flax, oats, rye, peas, lentils and dry beans).
33. Notwithstanding the CWB/ non-CWB distinction, the firms in the industry define themselves around purchasing and handling all types of grain and their primary elevators receive, store and ship grain in bulk. Grain handling companies have considerable flexibility to move from one grain crop to another.
34. In view of the foregoing, I concluded that the product market in this case should be defined as the purchasing and handling of grain. In summary, this best accords with trade views and practices in light of the fact that every company handles most types of grain. In addition, production substitutability is high in that primary elevators generally handle both CWB and non-CWB grains. The production facilities for handling different types of grain are essentially the same.
35. The purchasing and handling of grain is a distinct product market without practical substitutes. The purchasing and handling of grain differs from the purchasing and handling of all other agricultural commodities in their physical characteristics, means of production, uses, and pricing.
36. On the supply side, farmers are tied to grains. Due to the length of growing seasons, and the suitability of grains to certain climates and regions, it is my view that grain farmers would not switch to the production of other agricultural commodities in sufficient numbers to prevent a small but significant decrease in price they obtain for their grain products.
37. In my opinion, the purchasing and handling of grain constitutes the relevant product market within the meaning of the Competition Act (the "Act").
11
III. Relevant Geographic Market 38. Grain flows from producers on their farms to primary elevators, from which it moves by rail to domestic or US purchasers or to port terminals for export to other countries.
39. Farmers typically haul grain by truck to nearby primary elevators. Delivery of grain from the farm to elevators is a relatively costly and time consuming exercise. As a result, farmers generally sell and deliver their grain within a limited geographic area surrounding their farms.
40. Grain trading companies generally purchase grain from farmers at primary elevators. The geographic area from which a primary elevator receives grain is limited by transportation costs and is known as the "draw area" for that facility. The size of the draw area for any given primary elevator varies depending on factors such as the crop yield and the number and location of competing grain elevators. Draw areas expand and contract in response to normal economic fluctuations in crop supply, crop demand, and transportation costs.
41. In its 1997 review of the proposed acquisition of UGG by the Alberta Wheat Pool and the Manitoba Pool Elevators, the Bureau concluded that draw areas for UGG' s primary elevators were a 30 mile radius around each UGG elevator in Alberta and Manitoba.
42. In connection with the Commissioner's inquiry in this case, UGG submitted market share data based on 60 and 90 mile circles around UGG HTP facilities.
43. UGG arrived at these figures through the analysis of their hauling data. UGG compared the average delivery distance of a grain delivery in the 1996-97 crop year to the average distance for the 1999-00 crop year and found that it had doubled to [ ] kilometres. UGG's data indicates that 59% of deliveries to their primary elevators are made from within [ ] miles (approximately []kilometres) of any given elevator. In addition, UGG found that 82% of all deliveries were within [ ] kilometres or approximately [ ] miles of the elevator.
12
44. To assist it in this matter, the Competition Bureau hired Dr. Daryl F. Kraft, the Head of the Department of Agribusiness and Agricultural Economics at the University of Manitoba. Dr. Kraft was assisted by Mr. John DePape, a grain handling consultant.
45. Dr. Kraft and Mr. DePape used Census Consolidated Subdivision ("CCS") areas as defined by Statistics Canada as a starting point for their review. Dr. Kraft and Mr. DePape used a 30 mile radius as the draw area for elevators with less than 50 car spots. For larger elevators with 50 or more car spots, they assumed that 70% of their draw was from within 30 miles and 30% from 30-60 miles. This reflects the ability of operators of larger facilities to draw grain from greater distances as a result of being able to obtain MCI rates from the railways, part of which, in tum, can be offered to farmers as trucking incentives. Any elevator that cut into a CCS was assigned a market share in the CCS in proportion with its reach. This was done to capture the market presence of elevators that are just outside a given circle, that may draw from inside that circle. With these modifications the market shares assigned to companies by Dr. Kraft and Mr. DePape were similar to those calculated by UGO for their HTP elevators.
46. I also asked Dr. Kraft and Mr. DePape to do a local market analysis on all Agricore HTP elevators as these are the grain handling competitive alternatives which would be eliminated as a competitive alternative as a result of the Acquisition. Post-merger market shares, based on capacity were calculated in a 60 mile radius surrounding each of Agricore's 38 HTP elevators.
47. In examining the issue of geographic market definition, I also reviewed submissions from grain companies and the CWB. The grain companies were in general agreement that sourcing grain much beyond 100 kilometres on a regular basis was not possible in light of trucking costs. However, if a specific type or grade of grain was required to complete a particular rail shipment, sourcing it beyond 100 kilometres was an option.
13
48. Based on the submissions of UGO, the review of Dr. Kraft and Mr. DePape and industry interviews, I conclude that as a general rule the draw area for primary elevators is typically a radius of 50 to 100 kilometres.
49. For many primary elevators, draw areas overlap. Prior to the Acquisition, UGO and Agricore operated a number of primary elevators that had overlapping draw areas. They therefore competed with one another for the purchase and handling of grains from the same producers.
50. In some of these overlapping draw areas, UGO and Agricore were two of a small number of competing grain handling companies. If Agricore United were permitted to retain Agricore' s facilities in these local or primary draw areas, in my opinion, it would be in a position to unilaterally decrease prices paid to farmers because transportation costs would preclude them from selling to purchasers outside the draw areas in sufficient quantities to prevent the price decrease.
51. In my opinion, focusing the local market analysis on Agricore' s elevators is appropriate since these were the alternative purchasers of grain that were eliminated as a competitive alternative as a result of the merger. It is also my opinion that each such draw area for a primary elevator is a relevant geographic market within the meaning of the Act.
14
IV. Industry Concentration 52. UGG and Agricore are two of a small number of grain handling companies competing to purchase and handle grain in the following local geographic markets:
(a) the draw areas for primary elevators in the vicinity of the Agricore HTP at Dauphin, Manitoba;
(b) the two draw areas for primary elevators centred around Agricore' s HTPs at Star and Legacy Junction, Alberta (Edmonton area). The draw area for the Agricore HTP at Star includes primary elevators at Westlock and Gaudin (near Fort Saskatchewan). The draw area for the Agricore HTP at Legacy Junction includes primary elevators at Killam and Bawlf; and
(c) the draw areas for primary elevators in the vicinity of the Agricore HTP at Rycroft, Alberta (Peace River Area).
53. A combination ofUGG and Agricore substantially increased concentration in already highly concentrated grain purchasing and handling markets.
54. In the draw areas for primary elevators in the vicinity of Dauphin, Manitoba and in the Edmonton area, the post-merger market share of Agricore United is approximately 50% to 55%, while in the Peace River Region, the post-merger market share of Agricore United is approximately 60% to 65%.
V. Section 93 Factors Acceptable Substitutes 55. In my view there are no acceptable substitutes for primary grain elevator purchasing and handling services. While there are other facilities that can legally receive grain, such as process elevators (e.g. at flour mills and barley malters), the vast majority of the grain
15
received at such facilities is sourced directly from primary grain elevators and not from farmers' operations. In my opinion producer cars cannot effectively compete with grain handling companies since they do not provide all of the services of a grain handling elevator such as blending, cleaning or storage.
56. Primary grain elevators located in the US are not potential substitutes for primary grain elevator services in Canada in view of the prohibitive transportation costs associated with shipping grain from the geographic markets of Peace River, Edmonton or Dauphin over the US border and on to US primary elevators.
Barriers to Entry 57. In my opinion the barriers to entry into primary grain elevator handling services are high because of sunk costs.
58. The capital costs for construction of a new HTP elevator facility are estimated to be in the range of about $10-$15 million, depending on the configuration adopted and the size of the storage built. New elevators are usually built to load 50 or 100 rail cars and therefore qualify for railway MCI rebates.
59. There are few, if any, economically viable possible alternative uses for an HTP grain elevator other than grain handling and storage. For smaller older primary elevators the cost of demolition is estimated to range from $30,000-$100,000 while the net salvage value of a new HTP is about $1 million. The capital invested in a new grain elevator is therefore almost entirely sunk cost.
60. In my opinion the existing capacity of incumbent grain companies is more than sufficient to handle all grain crops grown in Western Canada. In my opinion, existing excess capacity of incumbent grain companies is a barrier to entry because it represents a sunk cost.
61. An operator of a primary elevator also requires access to port terminals on commercially 16
competitive terms. 62. Regulation is not a significant barrier to entry. The approval and construction of a new primary elevator facility can be completed in about one year.
Removal of a Vigorous and Effective Competitor 63. Agricore has been a strong competitor to UGG in providing primary grain elevator handling services in local markets in the areas of Peace River, Edmonton and Dauphin.
64. In the relevant markets, the acquisition of Agricore by UGG will result in significantly less grain handling choice for farmers. This will allow Agricore United to exercise market power, resulting in higher handling fees and lower grain prices.
Effective Remaining Competition 65. In my opinion if Agricore United is permitted to retain all the Agricore and UGG primary grain elevators in the affected areas of Peace River, Edmonton and Dauphin, the remaining companies will not be effective competitors for the purposes of eliminating the substantial lessening of competition. UGG and Agricore were two of a small number of grain handling companies competing to purchase and handle grain in those areas. In my view, without the divestitures contemplated in the Draft Consent Order ("DCO") filed by the Commissioner as part of his Application herein, the remaining third party grain-handling companies cannot be relied upon to prevent the substantial lessening of competition arising from this merger.
Foreign Competition 66. For the reasons set out in paragraph 56 above, US primary grain elevator facilities do not compete in the affected local markets.
17
VI. Anti-competitive Effects 67. In my opinion it is unlikely that UGG' s exercise of market power will be prevented by: (1) new entry, (2) farmers transporting their products to more distant markets, or (3) any other countervailing competitive force. It is also my opinion that it is unlikely that Agricore United's exercise of market power in any of the relevant geographic markets would be thwarted by significantly increased purchases of grains by processors or other buyers. The purchase decisions of these buyers are based on factors other than small but significant changes in grain prices, such as supply and demand conditions in their selling markets. In my opinion, without the remedy contemplated in the DCO the Acquisition would result in substantial lessening of competition for a significant period of time.
VII. Remedy 68. The DCO agreed to by the Commissioner and UGG contemplates the divestitures set out in Schedule "A" of the DCO. The primary elevator assets to be divested are situated in the following local market areas:
(a) in the vicinity of Dauphin Manitoba, (b) in the vicinity of Agricore's HTP at Star which includes primary elevators at Westlock and Gaudin (near Fort Saskatchewan) and in the vicinity of Agricore' s HTP at Legacy Junction which includes primary elevators at Killam and Bawlf; and
(c) in the vicinities of Rycroft and Fabler, Alberta. 69. In my opinion these divestitures will remove the substantial lessening of competition arising from this transaction in these local markets.
VIII. Alternatives to the Settlement 70. As an alternative to the DCO, contested litigation was considered as a means of seeking 18
divestitures in the relevant markets. However, I believe that the proposed divestitures should result in sales of the relevant primary elevators to effective competitors to prevent any possible exercise of market power in the markets concerned. I also believe the DCO provides a timelier and more certain outcome for Agricore United and producers of grain.
B. CANOLA PROCESSING I. Ownership Structure 71. Pursuant to the Acquisition, UGG acquired Agricore's interest in CanAmera which is a leading Canadian manufacturer and marketer of canola oil, and, is one of the largest canola processors in Canada. By virtue of its interest in CanAmera, Agricore (and now Agricore United), could nominate a representative to sit on CanAmera' s Board of Directors.
72. Agricore United has 16.67% interest in CanAmera. SWP has a 33.3% interest and CSY Agri-Processing, Inc. and its subsidiary Central Soya Company, Inc. effectively hold the remaining 50%.
73. The actual governance of CanAmera is administered through CF Edible Oils Inc. As a result of its shareholder interest, Agricore's CEO has traditionally been one member of the six person CF Edible Oils Inc. Board and Agricore has had an observer at certain committee meetings where detailed operational information is provided, discussed and commercial decisions are taken.
74. Archer Daniels Midland Company Ltd. ("ADM") is also a major canola oil seed processor and is a direct competitor with CanAmera. Pre-merger ADM had a 42% ownership position in UGO while post merger it holds 19% of the common shares of Agricore United which could, at ADM's option and subject to certain conditions, ultimately rise to 45%. ADM also has the right to nominate two representatives to the Agricore United Board of Directors. ADM also has the right to nominate one of four members to a Grain Operations Committee established by UGO. Further, the agreement establishing that Committee provides that ADM
19
shall have " ... substantial influence over the operating units of UGG that procure, transport and market grain ... ". 75. Through its Board representation and the Grain Operations Committee, ADM could receive competitive information concerning the operations of CanAmera as well as have the opportunity to influence CanAmera and take competitive advantage of commercially sensitive information which could result in a substantial lessening of competition for canola purchasing and processing.
II. Industry Overview 76. Canola seed processing results in two products: (1) a dry protein meal used in livestock and pet feed; and (2) a canola vegetable oil used as a major ingredient in numerous food products. As described below, the crushers purchase seed from grain handling companies, who themselves have purchased it from farmers.
77. Canola seed processors generally have a limited amount of storage capacity and therefore they tend to deal with grain handling companies who have an efficient and responsive delivery system. While direct purchases by processors from the farmers are utilized, the logistics of timely and efficient delivery make direct sourcing a minor proportion of their requirements.
78. CanAmera is viewed as a dominant participant in the North American canola processing industry. Together CanAmera and ADM account for approximately 65% of the North American canola processing market. Other significant participants in the market include Cargill and JRI which operates its canola processing business through its subsidiary, Canbra Foods Ltd.
79. CanAmera operations consist of: (a) a soya bean and canola crushing plant in Hamilton, Ontario; 20
(b) canola crushing plants at Fort Saskatchewan, Alberta; Nipawin, Saskatchewan; and Harrowby and Altona, Manitoba;
(c ) edible oil refineries at Toronto, Ontario; Montreal, Quebec; Ni paw in, Saskatchewan; Altona, Manitoba; and at Wainwright, Alberta; and
(d) packaging plants in Oakville, Ontario; and Edmonton, Alberta. 80. ADM has oilseed processing facilities at Lloydminster, Alberta; Windsor, Ontario and Velva, North Dakota.
III. Relevant Product Market 81. Canola purchasing and canola processing are relevant product markets. Canola differs from other types of grains in terms of customary uses and pricing.
82. Canola processing, which includes crushing, extraction and refining is a capital intensive industry. Canola seed processing results in two products, a dry protein meal used in livestock and pet feed and a canola vegetable oil which is used in salad dressings, margarine, cookies and other types of bakery products. Canola processing consists of crushing canola seed to extract a crude oil and further refining it into an edible vegetable oil. Canola meal is the solid portion of the canola seed remaining after the oil is removed.
83. Canadian crushers purchase just under 40% of total Canadian canola seed production, the balance is exported, primarily to the US, Mexico, Japan and China.
84. ADM is a leading exporter of Canadian grown canola. ADM's share of total Canadian canola seed exports is approximately 50%. AD M's significant position in both domestic and foreign canola purchases makes it unlikely that a small but significant price decrease by domestic canola crushers would be defeated by an increase in canola exports.
85. Canola oil is Canada's most popular all-purpose vegetable oil. This is due in large part to 21
the perceived health benefits associated with its use, as compared to other oils. Canola oil is often recommended by nutrition experts over other oils as it has the lowest level of saturated fat. 86. Canadians are the largest per capita consumers of canola oil in the world. Canola oil accounts for approximately 78% of total Canadian production of edible oils, including approximately 8 8 % of salad and cooking oils, 71 % of shortening oils, and 53 % of margarine oils.
IV. Relevant Geographic Market 87. Canola seed production requires a climate with cool nights, therefore the crop is primarily grown in the northern regions of the US and in Canada. For this reason canola processors are located in these same regions.
88. Canola processors purchase the bulk of their canola seed requirements from grain handling companies. The geographic market for grain handling companies is local as discussed in Section 'A' which deals with primary grain elevators. Some of the grain handling companies are integrated into canola processing (i.e., Cargill and JRI) while UGO, Agricore and SWP have either equity or contractual ties to ADM or CanAmera.
89. In respect of canola processing, it is my opinion that the geographic market is regional encompassing the major grain growing regions in Western Canada and the Northwestern US. This is also reflected in the processing plant locations. The major competitors in the canola processing industry have plants located in both the US and Canada.
90. The sales of processed canola products from both Canada and the US are distributed throughout North America establishing that the geographic market for the outputs from canola processing is large, often North American.
91. Since canola oil is a major food product and a major ingredient in other food products, a key 22
component of the canola processing business is to have plants situated in various locations to effectively service the food industry.
V. Industry Concentration 92. CanAmera and ADM are the dominant suppliers in the market for canola processing services in North America. In respect of canola oil processing, CanAmera and ADM account for at least 46% and 22% of the North American market, respectively. Cargill and JRI are the only other two major participants in this market and would account for approximately 20% and 12% of the market, respectively. These same market shares are reflective of the situation that exists in Canada. There is only one other competitor, a small farmer's co-operative canola processor located in Montana which would represent an insignificant share of the market.
VI. Section 93 Factors Acceptable Substitutes 93. Soya bean oil, which like canola oil is extracted through crushing, is substitutable for canola oil in some applications. Canola oil sells at a premium over soya bean oil. Each oil has its own distinct nutrient and fatty acid components. For example, as noted above canola is recognized as having the lowest level of saturated fat as compared to all other oils and, as such, is viewed by many as a more healthy oil alternative. The use of each of these two oils therefore depends on the desired benefits and functionality required.
94. Other edible oil products include, olive oil, com oil, palm oil, sunflower oil and peanut oil, all of which have higher levels of saturated fat than canola oil and are priced differently than canola oil. In addition, the use of peanut oil in North America has declined significantly in recent years due to common concerns relating to food allergies and its serious consequences on health, particularly among children.
23
Barriers to Entry 95. The barriers to entry into the canola processing and purchasing market are high and include the following:
(a) significant sunk costs for required multiple plant locations; (b) excess capacity in the market; and (c) specialized knowledge for operations and commodity trading. Remaining Competition 96. The dominant industry participants in canola processing in North America are ADM and CanAmera. Cargill and JRI together represent less than 35% of the market.
Foreign Competition 97. CanAmera, ADM, Cargill and JRI are the only major canola processors in North America. US plants are located in North Dakota, Montana and Minnesota. There is no foreign competition which originates from outside the US and Canada.
98. As noted above, canola oil and canola meal which are processed in Canada are sold in both Canada and the US. However, while canola meal is processed in the US and sold into Canada, little if any canola oil is processed in the US and sold into Canada.
VII. Anti-Competitive Effects 99. This transaction will reinforce ADM's already significant position with respect to both domestic and export purchases of canola seed which in my view would likely result in lower prices being offered to canola seed producers. ADM would be in a position to substantia11y lessen competition by influencing canola seed prices, grade assessment, trucking allowances and other terms.
24
100. CanAmera and ADM are the largest canola processors in North America. Together they account for approximately 65% of the canola oil processing market both in Canada and the US. Absent some safeguard, the Acquisition could result in ADM being in a position to receive commercially sensitive information concerning the operations of CanAmera and, indirectly, being able to influence the output pricing decisions of CanAmera. These circumstances would in my opinion, likely result in a substantial lessening of competition for canola purchasing and processing.
VIII. Remedy 101. The DCO agreed to by the Commissioner and UGO provides that Agricore United shall keep all non-public information it receives regarding CanAmera, which is obtained as a result of Agricore United's direct or indirect shareholdings in CanAmera, confidential and separate from ADM (including the ADM nominee's to Agricore United's Board of Directors). The DCO also provides that Agricore United shall not appoint any director, officer or employee of ADM as a nominee to CanAmera's Board of Directors. Finally, Agricore United's Grain Operations Committee shall exclude canola seed processing from the scope of its mandate.
102. In my opinion these remedies will remove the substantial lessening of competition arising from this transaction in the canola purchasing and processing markets.
IX. Alternatives to the Settlement 103. As an alternative to the DCO, contested litigation was considered with a view to seeking divestiture of UGG's shares in CanAmera. However, I believe that the safeguards proposed in the DCO should result in ADM being unable gain access to information which would likely increase its ability to exercise market power in the relevant markets.
25
APPENDIX ''A'' CURRICULUM VITA~'i~~{'=,;~~::;:.;f: :::~ .. ---»AY
Name: Halldor P. Palsson COMM.iSSiO~J~-• •• ~e . a \'J :i ' s er1 . 1 . 1 .. • •"'ss~ ~«-• 'If{} '?~.-.. "f§:j ~y ~ c::i 1 ., Address: 633 Island Park Cr. : <.5 JOSEE SRlJNE 1> Ottawa, Ontario ·~ #96 746 :: "•' KlY 3P4 <? ' &•k ··~ •, e , : ,; J ,,u,d,i.c,i.a,i.1,~ . .~ · -~ .: < e~ t 't. ,#;. E-mail: palsson.halldor@ic.gc.ca Phone: Office (819) 953-4256 Fax (819) 953-6400 Home (613) 728-0577
POST SECONDARY EDUCATION: 1991 Ph.D. in Economics, Carleton University, Ottawa, Ontario, Canada. Thesis Title: Population Dynamics and Extinction: An Application to the Fin Whale Stock Off Iceland.
Fields of Specialization: Industrial Organization, and Public Finance. 1982 M.A. in Economics, University of Waterloo, Waterloo, Ontario. 1980 B.B.A. Honours in Business Administration with an Economics Minor. Wilfrid Laurier University, Waterloo, Ontario.
PROFESSIONAL EMPLOYMENT: Apr 92 to present Industry Canada, Competition Bureau, Economic Policy & Enforcement (ES-05). Responsible for providing economic advice and analysis in support of competition investigations and court cases. This includes litigation support, the supervision of contracts with outside experts and support staff and testifying. Participate in the settlement of cases through negotiations with private sector
-1-
THIS IS ccPJiifr.. m f.Lm m-·-7>TO THE AFFIDAVIT OF. .. .H.A.LL:Do.R .... J?.A .. t:'l:l./..Sa.tiL .................................................................................... ffi"'"'"""""-·
corporations. Develop interventions to Federal and Provincial regulatory bodies and legislative committees on competition matters. Other duties include research on competition policy and related issues.
Feb. 88 to Apr. 92 Department of Fisheries and Oceans, Ottawa, Economic Analysis and Statistics Division, Economic and Commercial Analysis Directorate. (ES-02 to ES-04) Developed the economic arguments for the Canada-France Maritime Boundary Arbitration case over St. Pierre and Miquelon. Completed a review of factory freezer trawler policy. Designed and implemented projects on the effectiveness of fishery regulation enforcement to structure requests to central agencies for more resources for DFO. Represented DFO at the OECD fisheries committee and liaised with the U.S. Department of Commerce on methodologies to calculate producer subsidy equivalents for fishing industries. Conducted research on fisheries management regimes and related work on property rights versus effort regulation. Designed and supervised projects for junior economic staff and summer students.
Oct 87 to Feb. 88 Revenue Canada Taxation, Ottawa, Forecaster, Forecasting and Research Section, Finance Directorate. (ES-03) Worked on the Ideal Tax Administration project on the effectiveness and need for additional auditors for tax enforcement purposes. The economics of crime literature as it relates to tax evasion was surveyed and applied to Canadian tax evasion. Previous work on tax compliance and estimates of tax evasion generated within Revenue Canada were critically reviewed and further studies proposed.
May to Sept 1986 Department of Fisheries and Oceans, Ottawa, Economic Analysis and Statistics Division, Economic and Commercial Analysis Directorate. Completed a cost benefit study of the factory freezer trawler policy for Canada. Supervised the in-house data work for a model of the fishing industry that was built by outside consultants.
May to Sept 1985 Department of Fisheries and Oceans, Ottawa, Economic Analysis and Statistics Division, Economic and Commercial Analysis Directorate. Worked on a final demand model for groundfish in the United States market. Conducted a study of the Icelandic fishing industry and the role of government subsidies on the industry.
June 1983 to June 1984 Research assistant to Professor Kanta Marwah Carleton University. The estimation and simulation of various foreign exchange models
-2-
were the main tasks. Sept 1981to1982 Research assistant to Professor Jim Brox Department of Economics, University of Waterloo. Duties were computer work and data
management for Canadian macro models and tutoring the 4th year forecasting class in Troll.
Feb to Sept 1981 Project Accountant for Eaglebrook Investments Ltd of Toronto. Had responsibility for the day-to-day financial management of four residential developments in Pickering.
EXPERT WITNESS TESTIMONY: 1. Qualified as an expert economist witness in the analysis of industry structure and market competition before the Canadian International Trade Tribunal (CITT) in: Dumping of Refined Sugar Originating from the United States, Denmark, Germany, Netherlands etc. February 1996. My expert report was Appendix I: An Analysis of the Price Impact of Dumping Duties on the Canadian Refined Sugar Market. Counsel was Mr. James Sutton of the Department of Justice.
2. Qualified as an expert economist witness at the CITT in Certain Prepared Baby Foods Originating in or Exported from the United States of America, September 1998. My export report was: Exhibit "C" to the submission of the Director of Investigation and Research, August 10, 1998. Counsel were Mr. Simon V. Potter and Brenda C. Swick-Martin of Ogilvy Renault.
3. Qualified as an expert witness at the CITT in the matter of the Review of Dumping of Refined Sugar Originating from the United States, Denmark, Germany, Netherlands etc. My expert report was Exhibit J-3 Review RR-99-006 (Refined Sugar): The Impact of Dumping Duties on the Canadian Refined Sugar market 1995 to 2000. September 2000. Counsel was Ms. Josephine Palumbo of the Department of Justice.
4. The Commissioner of Competition and Lafarge S.A Application for a Consent Order. Affidavit and Report of Halldor P. Palsson Ph.D., June 14, 2001.
OTHER SKILLS: Good computer skills and knowledge of econometric packages for PCs.
Languages: Icelandic (native), French (BBB), Danish and some Spanish.
-3-
PUBLICATIONS: Kleit, Andrew N. and Palsson H. (1999) Horizontal Concentration and Anticompetitive Behaviour in the Central Canadian Cement Industry: Testing Arbitrage Cost Hypothesis, the International Journal oflndustrial Organization, Vol 17, (1999) pp. 1189-1202.
Kleit, Andrew N. and Palsson, H. (1996) Is There Anticompetitive Behaviour in the Central Canadian Cement Industry?: Testing Arbitrage Cost Hypothesis, Canadian Journal of Economics, Vol XXIX, No 2 (May, 1996) pp. 343-356.
Lane, D. E. and Palsson, H. P. (1996) Stock Rebuilding Strategies Under Uncertainty: The Case for "Sentinel Fisheries". pp. 61-90 in D.V. Gordon and G.R. Munro [ed.] Fisheries and Uncertainty: A Precautionary Approach to Resource Management, University of Calgary Press, 1996.
Lane, D. E. and Palsson, H. P. (1994) Stock Rebuilding Strategies Under Uncertainty: The Case for "Sentinel Fisheries". Canadian Journal of Economics, Special Issue, Part I, Vol. XXIX (April 1996) pp. S151-S156.
Palsson, H., Lane, D.E. and Kaufmann, B.(1993) Bioeconomic methods for determining TACs. pp. 357-369. In S.J. Smith, J. J. Hunt and D. Rivard [ed.] Risk evaluation and biological reference points for fisheries management. Can. Spec. Puhl. Fish. Aquat. Sci. 120.
R.W. Crowley and Palsson, H. (1992) Rights Based Fishing in Canada, Marine Resource Economics Vol. 7, No 2, (September) pp. 1-21
R.W. Crowley and Palsson, H.(1992) "Modeling Offshore Fishery Regulation Enforcement In Canada" American Journal of Mathematical and Management Sciences, Vol 12, No 1-2, pp. 153-190.
Edwin G. West and Palsson, H. (1988) Parental Choice of School Characteristics: Estimation Using State-Wide Data, Economic Inquiry, Vol. XXVI pp. 725-740
K. Marwah and Palsson, H.(1988)"Direct Interventions, Interest Rate Shocks and Monetary Disturbances in the Canadian Foreign Exchange Market: A Simulation Study", Ch 20 in Homa Motamen (ed) Economic Modelling in the OECD Countries (Chapman and Hall, London 1988) pp. 407-455
-4-
K. Marwah and Palsson, H. ( 1988) The Tracks of the Managed Exchange Rate of the Indian Rupee with Monetary Shocks and External Disturbances, in M. Dutta (ed) Asian Industrialization: Changing Economic Structures. (JAi Press Inc., Greenwich, Conn.).
TEACHING EXPERIENCE: Winters 1993-95 Government Policy Toward Business 330A, a third year course in industrial organization and competition law. Economic analysis of post-1986 Canadian competition law cases were emphasized.
Fall 1990 Microeconomics for public administration students, Carleton University. 1985-1987 Sessional lecturer at Carleton University, Ottawa, Ontario. Winter 1987 Intermediate Macroeconomics, 212 a second year course, two sections. Closed economy macro with emphasis on consumption, investment and the demand for money.
Summer 1986 Intermediate Macroeconomics. Economics 213 is an introduction to policy and the open economy.
Winter 1986 Intermediate Macroeconomics and Microeconomics. Fall 1985 Intermediate to Microeconomics. MEMBERSHIPS: Canadian Economics Association. Resource Modeling Association. Scientific Committee of the International Whaling Commission 1991-92.
Other: Vice President of the Chess Federation of Canada 2000-2001 Secretary of the Chess Federation of Canada 1999-2000 -5-