CT-88/l THE COMPETITION TRIBUNAL IN THE MATTER OF an application by the Director of Investigation and Research under Section 64(1) of the Competition Act, R.S. c. C-23 as amended. AND IN THE MATTER OF a Limited Partnership formed to combine the operations of the Reservec and Pegasus computer reservation systems.
AND IN THE MATTER OF The Gemini Group Automated Distribution Systems Inc. B E T W E E N: THE DIRECTOR OF INVESTIGATION AND RESEARCH, Applicant - and -AIR CANADA Place Air Canada 500 Dorchester Blvd. West Montreal, Quebec H2Z 1X5 - and -AIR CANADA SERVICES INC. c/o Air Canada Place Air Canada 500 Dorchester Blvd. West Montreal, Quebec H2Z 1X5 - and -PWA CORPORATION Suite 2800, 700-2nd Street, S.W. Calgary, Alberta T2P 2W2 - and -CANADIAN AIRLINES INTERNATIONAL LTD. Suite 2800, 700-2nd Street, S.W. Calgary, Alberta T2P 2W2 - and -PACIFIC WESTERN AIRLINES LTD. c/o Canadian Airlines International Ltd. Suite 2800, 700-2nd Street, S.W. Calgary, Alberta T2P 2W2
- and -CANADIAN PACIFIC AIR LINES, LIMITED c/o Canadian Airlines International Ltd. Suite 2800, 700-2nd Street, S.W. Calgary, Alberta TZP 2W2 - and -154793 CANADA LTD. c/o PWA Corporation Suite 2800, 700-Znd Street, S.W. Calgary, Alberta T2P 2W2 - and -153333 CANADA LIMITED PARTNERSHIP c/o The Gemini Group Automated Distribution Systems Inc. 1 First Canadian Place P.O. Box 84 5th Floor Toronto, Ontario M5X 1K4 - and -THE GEMINI GROUP AUTOMATED DISTRIBUTION SYSTEMS INC. 1 First Canadian Place P.O. Box 84 5th Floor Toronto, Ontario M5X 1K4 Respondents
TABLE OF CONTENTS I. INTRODUCTION ... I I. THE CRS INDUSTRY . A. Description B. The Canadian CRS Market c. Joint Government-Industry Task Force on Computer Reservation Systems D. North American CRS Vendors ..... E. Other CRS Vendors F. Why Thers Are Only 9-10 CRS Vendors or Potential Vendors in the World •.• 10 I I I. REASONS FOR MERGER .. IV. THE RELEVANT MARKET A. Geographic and Product Dimensions B. Market Share • . . . • • . . . . . . . • V. THE EXTENT OF FOREIGN COMPETITION AND THE EXTENT TO WHICH EFFECTIVE COMPETITION REMA INS . . . • . • . . . . . . . . . . vI . FAILURE OF PEGASUS . VI I. BARRIERS TO ENTRY VI I I. CHANGE AND INNOVATION IN THE RELEVANT MARKET . . . . . IX. OTHER FACTORS . . . • . . . . . . . A. Pro-Competitive Effect on Inter-Airline Competition B. Limitation on Travel Agency Subscriber Fees • • • • • . C. Limitation on Airline Booking Fees . D. Constraints on Abuse . • . • . . . . . . E. Benefits to Travel Agents F. Service in the French Language . G. Other Canadian Competitive Considerations • • • . . . • • . • . • . H. Efficiency Gains . . . . . . . . • • • . I. Dissolution Alternative x . EFFICIENCY GAINS DEFENCE . x I. ADDRESS FOR SERVICE XI I. RELIEF SOUGHT BY AIR CANADA
1 4 . . . . • . . . . 4 . . . . . . 5 7 8 • • • • . . . . • 10 12 . . . . . . • • • . 14 14 15 l 8 21 23 . ... 25 26 . • • . . . . • • . 26 • • • • • • • • • • 26 27 27 • • . . • • . 30 30 31 31 . • • . . . • • 32 33 35 36
RESPONSE OF THE RESPONDENT AIR CANADA
I. INTRODUCTION 1. The Respondent, Air Canada ("Air Canada"), opposes the Notice of Application by the Director of Investigation and Research (the "Director") dated March 3, 1988 (such Notice of Application and the attached Statement of Grounds and Material Facts hereinafter called the "Application") before the Competition Tribunal (the "Tribunal") on the grounds:
(a) that the merger which is the subject of the said Application (the "Merger") neither prevents nor lessens nor is likely to prevent nor lessen competition substantially in the provision of computer reservation systems ("CRS") services to airlines, travel agents and consumers in Canada within the meaning of Section 64 of the Competition Act; and
(b) that even if it prevented or lessened competition substantially, the Merger will bring about gains in efficiency that will be greater than and will offset the effects of any prevention or lessening of competition that could result from the Merger and that the gains in efficiency would not be attained if an order of dissolution were made.
2. In his Application, the Director has set forth a section e n t i t 1 e d " Ba c kg r o u n d " i n wi i i c h h e de s c r i b e s th e p a r t i e s , the transaction, the airline industry and the CRS industry. This section of the Director's Application highlights certain facts and ignores others, thereby providing a distorted overview of the context of the Merger. For example:
- 2 -(a) The Director's description of the CRS industry ignores the small number of participants in this industry in the context of North America and the world. The Application implies that the Canadian CRS market can support a number of CRS vendors on a full cost recovery basis. In fact, industry trends indicate that there will not be more than nine or ten distinct CRS's on a worldwide basis (paragraph 24 hereunder). The Canadian market can support, at most, only one CRS on a full cost-recovery basis.
(b) In paragraphs 5 and 8, the Director, in citing the national CRS market shares held by Reservec and Pegasus prior to the Merger, measures market share by travel agency locations. This produces a combined national market share for Gemini of ninety percent (90%). When stating the national market shares of the airline passenger market held by Air Canada, Canadian Airlines International Limited ("CAIL") and Wardair Canada Inc. ("Wardair") in paragraphs 11, 12 and 13, the Director uses domestic revenue passenger kilometres as his measure. Using this measure, he concludes that Air Canada and CAIL have over ninety percent (90%) of the national airline passenger market. Air Canada disagrees with the way in which each of these market shares has been measured. The Diractor should have used segments booked in the CRS market (as described in paragraphs 6 and 32 hereunder) and domestic, transborder and international revenue passenger kilometres in the air transportation market. By using less appropriate measures, the Director is creating an impression of overall market concentration in the airline and CRS markets that overstates that concentration and erroneously describes the effect of the Merger.
- 3 -(c) In paragraph 11 of the Application, the Director states that Air Canada and CAIL and their affiliates are the only carriers in eighteen of the top twenty-five city pairs in Canada. This allegation ignores trans-border and international services which are also relevant to the provision of CRS services in Canada. Of the top ten city pair markets (domestic, transborder and international) served by Air Canada and/or CAIL, all are served by another competitive carrier. Of the top twenty-five, all but three are served by another competitive carrier. By ignoring trans-border and international services, the Director is again overstating concentration in the airline passenger market and erroneously describing the effect of the Merger.
- 4 -II. THE CRS INDUSTRY A. Description 3. The majority of tickets for passenger airline travel are sold by travel agents and not directly by airlines. A CRS allows travel agents instantly to obtain current schedules, fares and availability of seats, and make reservations and issue tickets electronically for their airline passenger clients. Travel agents purchase or lease necessary equipment such as cathode ray tubes (CRT's) and printers from CRS vendors. There is a contractual relationship between travel agents as buyers and CRS vendors as sellers of CRS services (subscriber relationship).
4. There is also a contractual relationship which may exist between individual CRS vendors and airlines whose services can be sold through a particular CRS (a participating carrier relationship). Airlines must make it convenient for travel agents to sell their flights, and thus are listed on all CRS's that substantial numbers of travel agents are using. At the same time, a CRS vendor desires to make its CRS attractive to agents, and to do this must secure the ability to make reservations on the carriers that agents' customers request most.
5. An airline can be represented in a CRS either as a hosted or participating carrier. If it is hosted, it stores its complete airline seat inventory information in the CRS. In this case, the CRS vendor provides that carrier with both an internal reservation function to manage its inventory for use by its own ticket offices as well as an external agency subscriber system to distribute its product to subscribing travel agents and, ultimately, consumers. An airline can be hosted in only one reservation system - its own, another airline's, or a reservations bureau. Hosted carriers pay
- 5 that reservations system a hosting fee. These are distinct from booking fees paid to a CRS vendor as described in paragraph 6 hereunder.
6. If an airline is a CRS participating carrier, the CRS vendor does not supply an internal reservation and management function, but only lists the information on fares, schedules and seat availability (which the participating carrier supplies) and also accepts reservations and issues tickets on behalf of that participating carrier. The fees paid by a participating carrier to the CRS vendor are called "booking fees" and are based on each flight segment booked through a CRS. A flight segment represents travel on one direct flight which may have intermediate stops but involves no connections.
7. CRS vendors also earn booking fees from hotels, car rental agencies, railway companies and other travel service providers whose products are displayed in and reserved through the CRS.
B. The Canadian CRS Market 8. Since 1962, Air Canada has operated an automated internal reservation system ("Reservec"). Until 1973, Reservec was used solely by Air Canada's own ticket offices and reservation centres. In 1973, Air Canada entered into the business of providing internal reservation (hosting) services to other airlines. Reservec maintained the inventory of not only Air Canada, but also the other carriers hosted in it. At various times, Reservec hosted: Eastern Provincial Airways Nordair Quebecair Pacific Western Airlines Transair Air Jamaica Other smaller Canadian carriers.
- 6 -Reservec also contained information on schedules, fares, rules and seat availability on a wide variety of foreign carriers.
9. Canadian Pacific Air Lines, Limited ("CP Air") also operated its own internal reservation system and was never hosted in Reservec, although CP Air's schedules, fares and seat availability were available through Reservec.
10. Until 1975, Reservec's services were only available to the reservations centres and ticket offices of Air Canada and other hosted carriers. In 1975, access to this service was expanded to include travel agencies and it became a "CRS" in the sense that it provided automated reservation services to travel agents.
11. During the period 1973 to 1982, Reservec automated almost two-thirds of the travel agency locations in Canada. Its market share during that period was thus one hundred percent (100%) of the segments booked through these Canadian automated travel agents.
12. In 1983, a second CRS vendor, American Airlines, through its SABRE system ("SABRE"), entered the Canadian CRS market after it had become established as the largest CRS in the U.S. In fact, SABRE is the largest CRS in the world. SABRE concentrated on marketing its CRS to larger agencies in urban areas of Canada, and by 1986 had obtained a twenty-two percent (22%) share of the Canadian CRS market measured by segments booked and an even higher share in those areas where it particularly focussed its attention. (1) (1) In paragraph 28 of the Application, the Director states that in 1986, Reservec accounted for approximately seventy percent (70%) of segments booked and Pegasus, approximately seven percent (7%). By implication, U.S.
- 7 -13. CP Air entered the Canadian CRS market in 1984 under the name "Pegasus 2000" ("Pegasus"). By 1986 Pegasus had obtained only a seven percent (7%) share of the Canadian CRS market as measured by segments booked.
14. In 1986, Reservec's share of the Canadian CRS market had declined to seventy percent (70%), down from one hundred percent (100%) in 1982.
15. There are other U.S. CRS services in the Canadian market, but their penetration to date has been small.
16. In June of 1987, Reservec and Pegasus were merged to form Gemini (as more fully described in paragraph 25 hereunder).
C. Joint Government-Industry Task Force on Computer Reservation Systems
17. In the U.S. in 1982, complaints of bias and other discriminatory practices in the provision of CRS services (favouring the airlines that owned the CRS) gave rise to Congressional and Department of Justice investigations. The problem warranted a full-scale Civil Aeronautics Board ("CAB") investigation. In 1984, the CAB found that the U.S. airlines providing CRS services to travel agents were using their systems in ways that favoured the selection of their airline services.
CRS's would have accounted for approximately twenty-three percent (23%) of segments booked, with SABRE accounting for approximately twenty-two percent (22%). While a precise count of segments booked through the various CRS's in Canada
is difficult to obtain, the figures used by the Director appear to be reasonably representative and are relied upon by Air Canada for the purposes of this Response.
- 8 -The built-in bias was evident in many ways, most notably in the practice of programming the CRS to give preferential display to its own services and faras. As a result, in November 1984, the CAB issued rules governing access, pricing and bias in the supply of CRS services.
18. In Canada in 1984, a Joint Government-Industry Task Force ("Task Force") was established to look into the question of possible bias (favouring Air Canada) in Reservec. Ultimately, the terms of reference of the Task Force were broadened to look into not only Reservec, but also Pegasus.
19. In contrast to the CAB's findings in the U.S., the Task Force found: ... that neither Air Canada nor CP Air operates its system with a view toward creating a competitive advantage for itself and thereby discriminating against other carriers reliant on their system. The Task Force believes that these CRS owners are committed to supplying a fair, unbiased display of information on airline schedules, fares, seat availability, etc., and to meet carrier automation requirements as effi
ciently and as effectively as possible, consistent with the capabilities of their systems.
(Task Force Report, June 21, 1985, page 2.)
D. North American CRS Vendors 20. The North American CRS Market consists of six vendors. Five are owned by U.S. airlines, either individually or jointly. Table 1 shows the six North American CRS vendors:
- 9 -TABLE 1 NORTH AMERICAN CRS VENDORS 1988 Airline CRS Provider CRS American Airlines AMR Information Services SABRE Inc.
United Airlines Covia Corp. Apollo Texas Air SystemOne Corp. SystemOne nrn/N or th west Pars Travel Information Pars Airlines Systems Delta Airlines Delta Airlines Inc. DATAS II Air Canada/CAIL The Gemini Group Limited Gemini Partnership
21. Table 2 lists, as of 1986, all CRS vendors in North America ranked by estimated segments booked. TABLE 2 SIZE OF NORTH AMERICAN CRS VENDORS 1986 CRS Vendor Estimated Segments Booked (Millions)
SABRE 132 Apollo 98 SystemOne 31 Pars 30 DATAS II 15 Reservec 14 Pegasus 1 22. It is apparent that in 1986, in terms of estimated segments booked, Reservec, Canada's largest CRS vendor, was eleven percent (11%) of the size of SABRE, and Pegasus, one percent (1%) of the size of SABRE. Even combined, the estimated segments booked by Reservec and Pegasus were only twelve percent (12%) of those of SABRE, and were barely equivalent to those of DATAS II, the smallest of U.S. CRS vendors.
- 10 -E. Other CRS Vendors 23. In addition to the North American CRS vendors, there are two other CRS vendor consortia in Europe with the possibility of a third independent CRS being formed in the Asia/Pacific Rim. With the exception of SABRE, Apollo and DATAS II, all CRS's involve ownership consortia of a number of airlines. The European CRS vendors and their major airline owners are listed in Table 3.
TABLE 3 EUROPEAN CRS VENDORS AND OWNERS CRS Vendor Major Owning Airlines Amadeus Air France, Lufthansa, SAS, Iberia Galileo British Airways, KLM, Swissair, Alitalia F. Why There Are Only 9-10 CRS Vendors or Potential Vendors in the World
24. The CRS industry will likely consist of only 9-10 independent firms for the reasons set out hereunder:
(a) CRS technology is extremely large-scale, requiring a large number of subscribing travel agencies to generate the transaction volumes necessary to achieve efficient scale.
(b) Since a large part of the cost of establishing and maintaining a CRS is shared with an airline's own internal reservation system, airline-sponsored CRS's have a distinct cost advantage over non-airline CRS's. Further, because of the desire to avoid paying booking fees to third parties, the airlines have an additional incentive to be more competitive in attracting travel agencies than non-airlines.
- 11 -(c) Outside the U.S., most national markets cannot support even a single CRS at full cost recovery because of the large number of transactions required to achieve the most efficient scale. All of Europe will be served by only two CRS's (Amadeus and Galileo).
- 12 -III REASONS FOR MERGER 25. On June 1, 1987, Air Canada, PWA Corporation ( "PWAC") and 153333 Canada Inc. formed 153333 Canada Limited Partnership for the purposes of operating a CRS. 153333 Canada Inc. was appointed general partner to manage the limited partnership. That limited partnership is now called The Gemini Group Limited Partnership ("Gemini") and the general partner is called The Gemini Group Automated Distribution Systems Inc.
26. Air Canada's reasons for entering into this Merger were:
(a) Air Canada realized that the Merger provided the only opportunity for a Canadian CRS to attempt to achieve the critical mass necessary for full cost recovery and provide to Canadians a sixth and Canadianoriented alternative in addition to the five U.S. CRS's. Given the substantial economies of scale in CRS technology, a "critical mass" in terms of subscriber travel agencies is necessary to enable any CRS to be viable and competitive with full cost recovery. u . s . competitors (such as SABRE, which is many times the size of Reservec) have already reached that scale and can enter the Canadian market on an "incremental" cost basis because the vast majority of the required investment is already recovered from their existing U.S. base. The Canadian travel agency population is approximately one-tenth that of the U.S. agency population. Entry into Canada by U.S. CRS vendors involves the simple leasing of communication lines from the U.S. to Canada's major population centres.
- 13 -( b ) There is free trade in CRS services. Because efficient U.S. CRS vendors were aggressively competing for business in Canada, Air Canada considered it necessary to have a sufficient volume to enable Canadians to build a technologically competitive CRS to survive in the Canadian CRS marketplace.
(c) There are many Canadian centres which have relatively small populations but which should have the benefits of travel agency automation. U.S. CRS vendors have not found it attractive to serve these markets. This is because of the relatively high cost involved in providing service to these non-urban and generally lower volume locations. Economies of scale for a Canadian CRS are thus vitally important to lower unit costs and ensure continued service to such centres. The Merger provided an opportunity to achieve such economies of scale.
(d) As in the case of most other of the world's major airlines, Air Canada viewed a consortium approach with other Canadian carriers as appropriate. In fact, should the Merger be allowed, equity participation will be offered to other Canadian air carriers and investors.
(e) Air Canada recognized the Merger would enhance the ability of Air Canada and CAIL to negotiate raciprocal international agreements with foreign national airlines for unbiased display of their respective schedules and fares in each other's CRS's.
(f) The Alliance of Canadian Travel Associations had promoted the Merger of Air Canada's and CP Air's CRS's. A Canadian CRS designed to meet the particular requirements of Canadian travel agents is most likely to be achieved through the Merger.
- 14 -IV. THE RELEVANT MARKET A. Geographic and Product Dimensions 27. Merger analysis requires definition of the relevant market in terms of geographic and product dimensions. The Director's Application fails to define the geographic or product dimensions of the market.
28. For many CRS issues, the relevant geographic market is national. However, for some purposes it may be more appropriate to consider smaller regions or urban centres, and for others (e.g. the potential suppliers of CRS services in Canada), the relevant geographic market will be North America.
29. CRS services provided to travel agents are principally a travel information service including listings of sche-dules, fares and availability, and a facility to enable travel agents to make reservations or bookings and issue tickets on airlines and with other travel service suppliers whose services are also displayed in the system.
30. In the case of services provided to airlines, hotels and other travel service suppliers, these include listing their services, schedules and fares and other relevant information in the CRS data base, making that data accessible to travel agent subscribers and providing the system through which travel agents may make bookings on them.
31. From the point of view of both the travel agencies and the airlines, the product is the reservation and sale of air and other travel services through the CRS.
- 15 -B. Market Share 32. Market share is one empirical measure of the relative competitive significance of the firms engaged in the relevant market. This is not easily measured in the dynamic Canadian CRS marketplacs. Normally, the measure used to quantify market share should be selected on a basis which reflects each firm's success in competing for business. That measure would be total CRS distribution revenues in Canada (booking fees from airlines and other vendors and subscriber fees from travel agents). However, information as to such revenues (particularly with regard to the other CRS vendors) is not available to Air Canada at this time. Airline "segments booked" (which generate booking fees from airlines when reservations are made through travel agencies) are more readily available and may reasonably be used as a proxy for total CRS revenues.
33. In paragraphs 5, 8 and 25 and elsewhere in the Application, the Director incorrectly uses the share of travel agency locations as the basis for calculating market share. This is inappropriate since it does not take into account variations in volume or revenue generation among travel agency locations. The same is also true of attempts to measure market share on the basis of CRT sets (Appli-cation, paragraph 25(ii)). Either all CRS distribution revenues, or segments booked as a proxy for such revenues, should be used.
34. In the period from 1983 to 1986, SABRE's share of market (measured on a segments booked basis) had increased from zero percent (0%) to twenty-two percent (22%). In the same period, Reservec's share declined from one hundred percent (100%) to seventy percent (70%). From 1984 to 1986, Pegasus had gained only seven percent (7%) of the segments booked in Canada. (Since Pegasus was a failing business,
- 16 even this small market share was not sustainable by Pegasus.)
35. Thus, considering market share alone, it is apparent that the Merger neither prevents, wholly nor substantially, potential competition from SABRE or others coming into the Canadian market and the degree to which actual competition may be said to have been lessened by reason of the Merger is, at most, de minimis. However, the words of subsection 64(2) of the Act indicate that Parliament did not intend that the only issue in a merger case should be market share or industry concentration: (2) For the purpose of this section, the Tribunal shall not find that a merger or proposed merger prevents or lessens, or is likely to prevent or lessen, competition substantially solely on the basis of evidence of concentra~ion or market share.
36. This is particularly important in this case where (as more particularly described below):
(a) Pegasus was a failing business in imminent prospect of being forced to discontinue operations. Thus its modest market share, however measured, overstated its relative competitive significance within the relevant market;
(b) CRS markets are still in the formative stage. Because of the rapid technological change and ease of entry by U.S. CRS's, historic market share may be unreliable as a predictor of inherent or future competitiveness; and
(c) Reservec alone had a substantial position in the relevant market prior to the Merger. Most of the speculation as to the conduct and market control concerning the merged entity (Gemini) suggested by the Director, might equally be attributed to Reservec. The
Merger should not be taken as an opportunity to challenge the pre-merger relative size of one of the parties to it.
- 17 -
- 18 -V. THE EXTENT OF FOREIGN COMPETITION AND THE EXTENT TO WHICH EFFECTIVE COMPETITION REMAINS 37. Paragraph 65(a) and (e) of the Competition Act provide that the Tribunal may have regard to: (a) the extent to which foreign products or foreign competitors provide or are likely to
provide effective competition to the businesses of the parties to the merger or
proposed merger;
(e) the extent to which effective competition remains or would remain in a market that is or would be effected by the merger or proposed merger;
38. Gemini's major competitor currently is SABRE (with potential competition from the other four U.S. CRS's). SABRE has gone from not having any presence in Canada prior to 1983 to the point where by 1986 it is estimated to have a twenty-two percent (22%) share of the Canadian CRS market measured by segments booked. SABRE is the world's leading CRS vendor. It is today aggressively and successfully competing for CRS market share in Canada. For example, in March, 1988 Jarvis Travel of Calgary, Alberta, Gemini's largest single travel agency customer, switched to SABRE and has committed to some thirty SABRE CRT's.
39. Because of the size of SABRE's U.S. subscriber base, its unit costs are lower than those of Gemini. Moreover, SABRE's investment in CRS software and hardware has been made for its operations in the U.S. SABRE's Canadian operations are incremental to its U.S. business. To date, SABRE has tended to market its services primarily to larger travel agencies in Canada's larger urban centres such as Toronto. The incremental cost of access to major Canadian centres is relatively low. These cost advantages are among the reasons why SABRE is and continues to be a vigorous and effective foreign competitor to Gemini.
- 19 -40. SABRE's travel agency contracts usually contain five-year terms with liquidated damage provisions and minimum booking requirements which create a situation of de facto exclusivity. These liquidated damage provisions (which preestimate the loss to SABRE of booking fee revenue in the event of early termination) have been upheld in U.S. courts. The amounts involved can be very large. Thus, SABRE has an assured existing Canadian subscriber base, and can concentrate its attentions on capturing further business from Gemini. As a result, SABRE will continue to be a vigorous and effective competitor in Canada.
41. The Director alleges that SABRE is currently functionally superior to Gemini. As such, it is an attractive option for Canadian travel agents. Any competitor which is perceived by customers as having "state-of-the-art" technology and equipment is in a strong position to provide effective competition.
42. Other U.S. CRS vendors have so far sought only modest Canadian presence. Apollo, for example, has thirty subscribing travel agents. Texas Air's SystemOne also has a presence in Vancouver. Wardair has recently been hosted in Texas Air's SHARES hosting system. This liaison suggests that Texas Air's SystemOne CRS may expand in the Canadian market. As in the case of SABRE, expansion in Canada by Apollo and SystemOne will be on the basis of incremental costs. As a result, they will constitute competition on an ongoing basis for Gemini.
43. In paragraph 51 of the Application, the Director has alleged that the Merger will reduce or eliminate technological innovation. On the contrary, Gemini is planning to spend some $60 million for a new software system to make it better able to compete with SABRE and other U.S. CRS
- 20 -vendors. This is further evidence of the competitive pressure in the Canadian marketplace. (However, it is noteworthy that the Director's Application is now preventing Gemini from making the commitments necessary to undertake these innovations.)
- 21 -VI. FAILURE OF PEGASUS 44. Paragraph 65(b) and (f) of the Competition Act provides that the Tribunal may have regard to: (b) whether the business, or part of the business, of a party to the merger or proposed merger has failed or is likely to
fail;
(f) any likelihood that the merger or proposed merger will or would result in the removal of a vigorous and effective
competitor;
45. Pegasus entered the Canadian CRS market in 1984. By 1986, it had captured only seven percent (7%) of the market measured on a segments booked basis. This compares to SABRE, which entered the market only one year earlier, and achieved a twenty-two percent (22%) market share by 1986. Once it entered the market, Pegasus never even began to approach the necessary critical mass to sustain business operations on an independent basis over the long term.
46. Even if Pegasus, under a new and expanded PWAC, were to capture a larger share of the Canadian CRS market approximating CAIL's share of airline business originating in Canada, this would not have given it the necessary critical mass and would have represented much less than the subscriber base of even the smallest U.S. CRS. However, expansion of Pegasus to this degree would have rendered Reservec unprofitable. Simply redistributing market share between Pegasus and Reservec would have meant that there would be two failing Canadian businesses, not merely one, and would have inevitably resulted in the demise of any independent Canadian CRS competitor. Neither Pegasus nor Reservec on its own could match the lower unit costs of the larger U.S. CRS vendors.
- 22 -47. Because Pegasus was in imminent prospect of having to go out of business, its long run market share was effectively zero.
- 23 -VI I. BARRIERS TO ENTRY 48. Paragraph 65(d) of the Competition Act provides that the Tribunal may have regard to: (d) any barriers to entry into a market, including
(i) tariff and non-tariff barriers to international trade,
(ii) interprovincial barriers to trade, and
(iii) regulatory control over entry, and any effect of the merger or proposed merger on such barriers; 49. In this case, there are no barriers to entry of the type listed in paragraph (d) of Section 65 of the Competition Act. In fact, there is free trade in the CRS market. For this reason, a Canadian CRS must be competitive on a North American scale, since U.S. CRS's are free to, and in fact do, compete in the Canadian market.
50. The Director appears to have placed significant reliance on the fact that there is no direct access link between Gemini and SABRE. He alleges that this precludes Canadian travel agents using SABRE from having "last-seat availability" ("LSA") on Air Canada's or CAIL's flights, thereby constituting a barrier to entry. This is not the case and is at variance with SABRE's rapid and significant growth in Canada.
51. LSA, on Air Canada and CAIL, currently serves to differentiate Gemini from SABRE just as SABRE's alleged superior functionality currently differentiates it and makes it, in that respect, more commercially attractive than Gemini.
52. LSA is a resource over which Gemini has a proprietary right. As such, it has value. Access to LSA on Air Canada and CAIL is a matter for commercial negotiation between
- 24 -SABRE (and other U.S. CRS vendors) and Gemini. SABRE has never directly approached Gemini to commercially negotiate LSA. The fact that there has been no commercial negotiation or resolution does not constitute a barrier to entry.
53. U.S. CRS's, and especially SABRE, have many advantages in competing for business in Canada (see paragraphs 39 to 42 above). The share of the Canadian market held by U.S. CRS vendors was in 1986 some twenty-three percent (23%), up from zero percent (0%) in only three years. The achievement of such a position in such a short space of time demonstrates the lack of any significant barriers to entry.
54. In paragraphs 30 and 31, the Director suggests that Air Canada and CAIL might refusa to provide information on fare schedules and seat availability and refuse to pay booking fees to U.S. CRS operators. Apart from being speculation, such an allegation is illogical. An airline that would deliberately refuse to participate fully in the major U.S. or other world CRS services would limit the distribution of its seat availability and thereby its business. Moreover, Air Canada has only a fifty percent (50%) interest in Gemini, but a one hundred percent (100%) interest in its own airline business. Neither Air Canada nor any other profit motivated airline would follow a course of conduct which would impair its principal business.
- 25 -VI I I. CHANGE AND INNOVATION IN THE RELEVANT MARKET 55. Paragraph 65(g) provides that the Tribunal may have regard to: (g) the nature and extent of change and innovation in a relevant market;
56. The CRS industry is characterized by a high degree of technological change and innovation. American Airlines claims to have invested hundreds of millions of dollars in the development of its SABRE system and has a continuing commitment to keep its system at the "state-of-the-art" level. The escalating competitive performance requirements of continued participation in the CRS industry has forced Air Canada and PWAC, through Gemini, to invest in a new and more costly CRS technology. This competitive response is required in an industry which is continuously under pressure from competitively driven large-scale technological developments by vastly larger U.S. competitors.
57. Paragraph 39 of the Director's Application states that the SABRE system has superior functionality as compared with either Reservec or Pegasus. It is clear that the maintenance by Gemini of a Canadian competitive presence in the Canadian CRS marketplace will very much depend upon how suecessful it is in responding to the technological superiority of U.S. CRS's.
58. Gemini is the smallest of the six North American CRS systems and is operating in a market a fraction the size of the primary market of its major North American competitors. Having regard to the substantially greater financial resources of Gemini's competitors and the commitment to maintaining maximum technological currency in a much larger market, there is a very real question whether Gemini (let alone Reservec or Pegasus individually) can continue to compete at this level.
- 26 -IX. OTHER FACTORS 59. Paragraph 65(h) provides that the Tribunal may have regard to: (h) any other factor that is relevant to competition in a market that is or would be affected by the merger or proposed merger.
A. Pro-Competitive Effect on Inter-Airline Competition 60. Although the airline passenger market is not directly within the CRS marketplace, it is "a market that is or would be affected by the merger". In the case of the airline passenger market, the Merger should stimulate competition as Air Canada and CAIL will, in respect to CRS, operate on an equal basis. It will have a pro-competitive effect on inter-airline competition to the extent that Air Canada possessed an advantage ("halo effect") in its separate ownership of the Reservec system over the less widely accepted and used Pegasus system. Given the large size differential as between the airline and CRS markets in Canada, even a relatively small benefit in the airline sector would tend to offset a proportionately more significant negative impact on the CRS market.
B. Limitation on Travel Agency Subscriber Fees 61. Most Canadian travel agency bookings through CRS's are on Air Canada and CAIL. This represents a lucrative source of revenue for CRS services such as SABRE (and costs for Air Canada and CAIL). Since Air Canada and CAIL would prefer that bookings on their airlines be made through Gemini rather than SABRE (booking fee avoidance), they have a great incentive to cause Gemini to seek to capture as much of the travel agency bookings as possible through competitive travel agency subscriber fees.
- 27 -62. In Canada, CRS vendors are discounting subscriber fees in order to capture a greater share of the travel agency business and a correspondingly larger portion of the booking fees to be derived from the airlines. This demonstrates the importance attached by CRS vendors to booking fees and to the need to maximize the potential generation of such fees through an enlarged travel agency CRS subscriber base.
C. Limitation on Airline Booking Fees 63. In June 1987, the then Minister of Transport sought assurances from Air Canada and PWAC about certain aspects of the conduct of Gemini. Air Canada and PWAC provided undertakings to the Minister stating that access to Gemini would be extended on a fair and equitable basis and that displays would be unbiased. Therefore, all participating carriers will have to be treated equitably in terms of booking fees.
64. In actual practice, it is highly unlikely that Gemini would ever initiate a booking fee increase. Any increase in booking fees initiated by Gemini would undoubtedly be matched by other CRS vendors. Since Air Canada and CAIL pay more in booking fees to other CRS's than other airlines pay to Gemini, Air Canada and PWAC would suffer net losses when booking fees are increased.
D. Constraints on Abuse 65. It would appear that the principal concern of the Director is the possible abuse by Gemini of its CRS market position (paragraphs 30, 31, 33, 43 and 46-52 of the Application).
66. In responding to this concern, the following points are relevant:
- 28 -(a) Prior Hist or}:'.. Until 1983, Reservec controlled one hundred percent (100%) of the Canadian CRS market and even at the time of the Merger it accounted for some seventy percent (70~0. During the whole of such time no significant abuses of the type described in the Application have been documented against Air Canada in regard to its CRS operations.
(b) Task Force Report. The June 21, 1985 Task Force Report on the state of the CRS business in Canada concluded that Reservec and Pegasus were "virtually free of bias".
(c) Commitments of the Parties. In June, 1987 the respective chairmen of Air Canada and CAIL, in response to specific requests from the then Minister of Trans-port, each provided separate written undertakings to the Minister not to engage in the sorts of practices about which the Director has expressed concern.
In his letter to Claude I. Taylor of June 3, 1987, the then Minister of Transport stated, after setting forth the requirements that he had of Air Canada and PWAC: Failure to meet these needs would invite consideration of appropriate measures by the Government.
The Minister's letter implies that the Government would not hesitate to take action, should it determine that there is non-compliance with these undertakings.
(d) Autonomy of Gemini. The Task Force recommended that the airlines provide greater autonomy to their respective CRS's to avoid any possible temptation on their part to engage in anti-competitive practices. Consistent with this recommendation, Reservec and Pegasus salespeople were compensated solely for selling CRS services and not for selling seats on Air Canada or
- 29 -CP Air. The same practice continues under Gemini. Moreover, the Merger results in even a further measure of autonomy through each airline owning not more than fifty percent (50%) of Gemini and neither being in a position to control it. Accordingly, the form of ownership of Gemini itself provides protection from anti-competitive conduct.
(e) Control of Abusive Conduct under the Competition Act. The Act contains provisions directly addressing many of the potential concerns mentioned by the Director. For example, Section 49 of the Act deals with tied selling of the sort speculated about in paragraph 33 of the Director's Application. In addition, there are provisions in the Act expressly dealing with discriminatory pricing (Section 34(l)(a)), predatory pricing (Section 34(l)(c)), refusal to deal (Section 47), exclusive dealing (Section 49) and abuse of dominant position (Sections 50 and 51). If the airlines or Gemini were to engage in such practices, these provisions could be relied upon by the Director to control them.
Given that there is no past history of such practices occurring, that the parties have given serious undertakings not to engage in such conduct and that there are provisions in the Act specifically designed to deal with such practices and which provide substantial sanctions, including in certain cases criminal penalties, it is inappropriate to speculate about the possibility of their occurrence or of possible criminal conduct by the airlines or their employees.
- 30 -E. Benefits to Travel Agents 67. If Gemini is permitted to continue, at least one CRS service will be more responsive to the needs of Canadian travel agencies than if U.S. CRS's dominate the Canadian market.
68. A unification of the Reservec and Pegasus system was an objective that had been sought by the Alliance of Canadian Travel Associations ("ACTA"). In September, 1986 the Executive Committee of ACTA passed a resolution calling for such unification. Again, following the announcement of the Merger, ACTA published an unsolicited advertisement in the Toronto Globe and Mail (Saturday, May 16, 1987) applauding this development.
69. The Merger provides a better and more financially assured basis for undertaking the technological improvements being demanded by travel agents and, as a consequence, affords a greater degree of assurance concerning the ultimate survival of a Canadian-owned system which is sensitive to their needs and the needs of the Canadian travelling public.
F. Service in the French Language. 70. A Canadian CRS will ensure that services can continue to be provided in both of Canada's official languages. It is unlikely that a U.S. CRS vendor would agree to incur the expense of providing French language capability when it views Canada as an incremental market in any event, and the French speaking population of Canada an even smaller subset of that incremental market.
- 31 -G. Other Canadian Competitive Considerations 71. The Merger provides a more assured basis for the continued existence of an all-Canadian system which reflects a sensitivity to Canadian preferences and requirements. This is because it provides to the merged entity, Gemini, what at least approaches the necessary critical mass or scale, in terms of volume of business, to better carry the heavy fixed costs of operating a national system. It is extremely costly to provide a CRS service to travel agencies located throughout the more remote and less populated areas of the country. Gemini's objective, in this respect, is to serve the nation as a whole.
72. The loss or disappearance of Gemini, a separate Canadian CRS system, could involve the loss from Canada of important technical skills which may be expected to migrate outside the country in search of comparable job oppor-tunities.
H. Efficiency Gains 73. The Merger will undoubtedly give rise to efficiency gains. These are described under the heading "Efficiency Gains Defance". Even if (contrary to the submissions made under that heading) the Tribunal were to conclude that the efficiency gains defence provided by Section 68 did not apply, it would nevertheless be proper to consider such gains as part of the Section 64 merger analysis. Section 1.1 of the Act, in expressing the purpose of the Act, specifically states, in part, as follows: 1.1 The purpose of this Act is to maintain and encourage competition in Canada in order to promote the efficiency and adaptability of the Canadian economy, The efficiency gains for which recognition is claimed represent real operating scale economies which are unique to
- 32 the merger and do not include any purely pecuniary economies such as tax advantages.
I. Dissolution Alternative 74. If the Director's Application is successful, Pegasus will have to be liquidated in view of its large operating losses. Quite apart from the undesirable financial and social costs which would be associated with such a business failure, there is no possibility of reverting to the premerger status quo ante.
- 33 -X. EFFICIENCY GAINS DEFENCE 75. Under subsection 68(1), the Tribunal is precluded from making an order against a merger under Section 64: ... if it finds that the merger has brought about or is likely to bring about
gains in efficiency that will be greater than, and will offset, the effects of any
prevention or lessening of competition that will result or is likely to result from the merger ... and that the gains in efficiency would not likely be attained if the order were made.
76. A conservative estimate of the efficiency gains which will be generated as a consequence of the Merger is in excess of $15 million annually, commencing approximately in the fourth year of Gemini's operation (i.e. 1990).
77. The consolidation in Gemini of the internal reservations functions and the subscriber CRS services of Pegasus and Reservec will create the potential for realizing significant efficiency gains (and cost savings). The volume and quality of services currently being provided by Pegasus and Reservec require substantial duplication of capabilities and costs which will be eliminated by a consolidated system.
78. The proposed reorganization contemplates the unification of all the main elements of Reservec and Pegasus (such as computer facilities, software, communication lines and other capital facilities). The reorganization involves the consolidation of data processing facilities at Winnipeg and the centralization of most other functions in Toronto.
79. Were Pegasus and Reservec to continue independent CRS operations in the Canadian market, they would each have required major and parallel investments in enhancing the quality of their CRS service to remain competitive with SABRE and other U.S. CRS's. The Merger avoids duplication of these new investments.
- 34 -80. The anticipated efficiencies are real cost savings. All of such gains are unique to, and result solely from, the Merger and would not likely have occurred but for the Merger. None of these gains is of the nature which are excluded for this purpose by subsection 68(3).
81. As mentioned, such gains have been conservatively estimated at $15 million per annum. Because of the limitation on subscriber fees and booking fees outlined in paragraphs 61 to 64 above, the Merger could not result in price rises that could come close to exceeding these efficiency gains and cost savings.
82. Accordingly, it is submitted that the minimum anticipated efficiency gains will exceed by a wide margin any possible adverse effects of the Merger and that the defence provided by Section 68 is applicable.
- 35 -XI. ADDRESS FOR SERVICE 83. Service on Air Canada of any documents in connection with this proceeding may be effected on M.E. Rothstein, Q.C., Aikins, MacAulay & Thorvaldson, c/o A.E. Lorenz, at Air Canada, 500 Dorchester Blvd. West, Montreal, Quebec H2Z 1X5.
- 36 -XII. RELIEF SOUGHT BY AIR CANADA 84. Air Canada submits that the Application be dismissed in its entirety.
All of which is respectfully submitted this 22nd day of April, 1988.
AIKINS, MacAULAY & THORVALDSON Pceounrsel::Ai: ~ ~a