UNITED STATES of America, Plaintiff,
v.
UNITED TOTE, INC., Defendant.
Civ. A. No. 90-130 LON.
United States District Court, D. Delaware.
May 10, 1991.*1065 Kent A. Jordan, U.S. Attorney's Office, Wilmington, Del. (Richard L. Rosen, Jonathan M. Rich and Linda SooHoo, U.S. Dept. of Justice, of counsel), Washington, D.C., for plaintiff.
Henry N. Herndon, Jr., P. Clarkson Collins, Jr., Barbara MacDonald, of Morris, James, Hitchens & Williams, Wilmington, Del. (Donald I. Baker, Esquire, David R. Boyd, Kenneth G. Starling and Bruce M. Bettigole, of Sutherland, Asbill & Brennen, of counsel), Washington, D.C. for defendant.
OPINION
LONGOBARDI, Chief Judge.
On March 14, 1990, the Government brought the instant antitrust action to prevent and/or restrain an alleged violation of Section 7 of the Clayton Act, as amended, 15 U.S.C. § 18. The violation purportedly arose when Defendant United Tote, Inc. ("United Tote" or "United") acquired all of the outstanding shares of Autotote Systems, Inc. ("Autotote"). On April 11, 1990, the Government and United Tote entered into a hold separate agreement by which United Tote and Autotote would be operated and maintained separately and independently. After a six day trial, the following constitutes the Court's findings of fact and conclusions of law in the matter.
I. BACKGROUND
Pari-mutuel wagering is the most common form of wagering at horse races, grey-hound races and jai alai contests in the United States. The essential characteristic of pari-mutuel wagering is that the wagers of the betting patrons are combined into pools from which the successful bettors recover. As a result, wagering odds are determined by the size of the pools established by the patrons' wagers. The odds, conveyed to the bettors at frequent intervals, change continually until betting closes at the start of a race or event. Docket Item (D.I.) 105, ¶ 9.
A totalisator system supports pari-mutuel wagering by controlling the acceptance of wagers, calculating odds and payout, cashing winning tickets, and performing assorted management, accounting and reporting functions. Id., ¶ 12. Totalisator systems can be used to serve on-track, off-track and inter-track wagering. On-track wagering is conducted on the premises of a racetrack while that event is taking place.[1]Id.
Off-track wagering takes place somewhere other than the racetrack. Most commonly, it takes place at off-track betting "parlors", some of which televise the event taking place. An off-track totalisator system generally combines or "commingles" wagers from the remote locations into a common pari-mutuel pool maintained at the host track. In some instances, the combination of pools and the calculation of payout amounts are done by the off-track betting *1066 system itself. See generally Id.; Transcript ("Tr.") at 337-38, 340.
Inter-track wagering is similar to off-track wagering in that it entails the combination of wagering pools from different locations into a common pool at the host track. The only significant difference is the location where the wagering occurs. Inter-track wagering occurs at one race-track when it accepts bets on a live racing event at another track. In many cases, the racing event is televised from the host track where the contest is conducted to one or more "guest" or "satellite" tracks. Inter-track wagering then occurs when the wagers from the guest tracks are transmitted to the host track and combined with the host track's pari-mutuel pool. The host track's central computer calculates the resulting odds and returns that information to the guest tracks. As a result, betting patrons at the guest tracks share in the pari-mutuel pool of the host track and they receive the same payout for winning wagers as they would have received had they attended the live racing event.[2] D.I. 105, ¶ 28; Tr. at 244-45, 340-41.
By far, the most common form of totalisator system used in North America is the computerized "cash/sell" system that allows patrons to place bets and to cash winning tickets at the same window. A typical cash/sell system consists of a central computer, ticket issuing terminals, display equipment and associated peripheral equipment. The system also includes proprietary software to carry out the complex, high speed wagering functions required by modern pari-mutuel betting. D.I. 105, ¶¶ 15-17.
The interface between the totalisator system and pari-mutuel customer is the totalisator terminal. Totalisator terminals are most commonly operated by pari-mutuel tellers or clerks although a number of companies supply terminals that are operated manually by the patrons as well as some that accept pre-marked betting slips the patrons have prepared. After the terminals accept the wagering information, the terminal transmits it to a central computer system. Id., ¶ 12.
Terminals at a pari-mutuel facility are linked to one or more central processing units ("CPUs") which perform the bulk of the computer calculations for the system. The CPUs verify each bet and authorize the issuance of a printed ticket, aggregate the wagers into pools and calculate the resulting odds. The CPUs also commonly communicate the calculated odds for display on video devices or totalisator boards and calculate the final odds and payout amounts when the racing event is complete and the results are official. At that point, the winning bettors can present their tickets to the pari-mutuel tellers or clerks who enter them into the terminals. The terminals read and confirm the ticket as a winning ticket. The terminals then communicate information from the winning tickets to the CPUs, which calculate and display the payout amount to the clerk, who pays the patron. See generally Id., ¶¶ 12-14; Tr. at 919-21, 1149-50.
Totalisator systems are typically supplied pursuant to lease agreements. Under the agreement, a totalisator supplier provides and operates the totalisator system and guarantees that the system will perform accurately and reliably. Employees of the totalisator company operate, service and maintain the totalisator system. Because of the extreme importance of the totalisator function, a totalisator company must be prepared to prevent breakdowns and quickly repair any problems. For example, computer operators and maintenance personnel are typically located on-site at the race-track to ensure that the system functions smoothly and reliably. In addition, all major totalisator systems include backup computers that take control in the event that the primary computer malfunctions. Finally, the totalisator company often provides *1067 an uninterrupted power supply to protect against power failures. Totalisator suppliers usually assume liability for wagering revenues that are lost by the racetracks as a result of a totalisator system malfunction. Id., ¶ 22; Plaintiff's Exhibit ("PX") 179 at 39.
Prior to 1976, all totalisator systems in the United States market were "sell-only" systems that required the patron to purchase wagering tickets from one location and return to a different window to cash winning tickets. In 1976, American Totalisator ("AmTote") introduced the first cash/sell North American totalisator system at the Ontario Jockey Club in Canada. Tr. at 821-22. The primary innovation of the cash/sell totalisator system was that it allowed wagering patrons to return to the same terminal from which they placed their bets to cash their tickets. This development not only resulted in greater convenience to the wagering customer but also reduced labor costs to the track. The cash/sell totalisator system was immediately recognized as offering a number of advantages over the sell-only system and, at the present time, all but the smallest tracks and state fair events have converted to cash/sell systems. D.I. 105, ¶ 17.
AmTote, which enjoyed a dominant position in the North American totalisator industry prior to 1976, was able to consolidate and maintain its position after its introduction of the first North American cash/sell system, especially at the larger tracks. Tr. at 822. Although AmTote has witnessed a subsequent erosion of its market position, it still captures approximately 55% of the total revenues produced in the North American market. D.I. 105, ¶ 8. In addition, AmTote has announced that it will offer its first new totalisator system since its initial cash/sell system in 1976. AmTote calls its new system "Spectrum" and it has entered into a contract with the New York City Off-Track Betting Corporation to install Spectrum as early as April of 1991. Tr. at 1113-14. It is anticipated that this new system will provide AmTote with additional competitive strength in the coming years.
Autotote is a privately owned Delaware Corporation located in Newark, Delaware. Autotote's primary business is the provision of totalisator systems and services in the North American and foreign markets. At the time AmTote introduced its first cash/sell system in 1976, Autotote was an American subsidiary of ATL, Ltd., an Australian totalisator company. Autotote served approximately forty racetracks in the United States and Canada, including some larger tracks, with its sell-only system. PX 1, ¶ 11. Autotote introduced its first cash/sell system in North America in 1978 or 1979. At the time, AmTote and Autotote were the only totalisator companies offering cash/sell systems in the North American market. Tr. at 826.
United Tote is a Delaware corporation with its principal place of business in Shepard, Montana. United Tote got its start approximately thirty-five years ago when the Shelhamer family purchased a small track in Belgrade, Montana. After discovering that there was no one in that area who was experienced in pari-mutuel wagering, United Tote was formed to provide a variety of totalisator and other racing services to small racing tracks and racing fairs in Montana and adjacent northwestern states. The totalisator services provided at that time used older sell-only equipment that had been purchased from other totalisator companies, including AmTote. Tr. at 543-45, 587.
During the late 1970s, Amtote and Autotote focused their cash/sell systems on the larger tracks while offering only the sell-only systems to the smaller tracks. Recognizing the desire of smaller tracks to obtain cash/sell systems, United attempted unsuccessfully to purchase cash/sell equipment from AmTote. When that effort failed, United decided to develop its own cash/sell totalisator system which it accomplished by contracting with a number of small companies in Southern California that specialized in computer software and related technologies. In April of 1980, United Tote introduced its cash/sell totalisator system known as the United System 1000 at a racetrack where the Shelhamer family had a substantial ownership interest. During *1068 the remainder of the 1980 racing season, United Tote secured contracts to operate its System 1000 at a number of other tracks in which it did not have ownership interest and its system was licensed for use in Canada as well. See generally, Tr. at 545-47, 587-90, 608, 611-12.
From 1980 to 1985, United took approximately twenty-five contracts away from AmTote, all from racetracks which were being served by sell-only AmTote equipment. Tr. at 619. These contracts tended to be with smaller racing facilities with average racing handles in the range of $150,000 or less.[3]Id. at 615. These initial contracts were generally with tracks having short racing seasons ranging from a few days to three to four months in duration. Id. at 620. United Tote also took three contracts away from Autotote during this period. As with AmTote, in each instance, United Tote offered cash/sell equipment and Autotote countered with sell-only equipment rather than a cash/sell system. Tr. at 619, 830-31.
By the mid-1980s, United Tote discovered that AmTote and Autotote were now offering all segments of the market the same opportunity to upgrade to a cash/sell system that was being offered by United. Finding continued market penetration considerably more difficult, United turned its attentions to tracks in new racing jurisdictions that had just begun to authorize pari-mutuel wagering. Tr. at 618; Defendant's Exhibit ("DX") 63s. United Tote has enjoyed considerable success in this area and, over the past five or six years, United has signed forty-two of a potential sixty-two contracts in these new racing jurisdictions. Tr. at 618-19. The remainder went to AmTote with the exception of two tracks which were won by Autotote. Id. at 629.
Over the years, United Tote has expanded its presence throughout the United States and has gradually begun to serve medium-sized tracks as well as some of the smaller statewide wagering systems in such states as Montana, Wyoming, New Mexico and Idaho. DX 62g. United has not, however, significantly penetrated the large tracks and wagering systems whose average daily handles exceed 1 million dollars. According to United Tote, it was partly for this reason that on December 11, 1989, United Tote acquired all of the outstanding shares of Autotote. PX 14 at 5. On March 14, 1990, the United States filed the instant action, alleging that United Tote's acquisition of Autotote violated Section 7 of the Clayton Act, 15 U.S.C. § 18, and seeks an order requiring United Tote to divest its interest in Autotote. This Court has jurisdiction pursuant to Section 15 of the Clayton Act and venue is proper under Section 12 of the Clayton Act and 28 U.S.C. § 1391(c).
II. LEGAL STANDARD
In order to prevail on a Section 7 claim, the Government must establish a reasonable probability that the merger between United Tote and Autotote will substantially lessen competition in the relevant product and geographic market.[4] If it demonstrates that the merger further consolidates an already highly concentrated market for a given product, the Government establishes a rebuttable presumption that the merger is illegal under Section 7. United States v. Citizens & Southern National Bank, 422 U.S. 86, 120-22, 95 S. Ct. 2099, 2118-2120, 45 L. Ed. 2d 41 (1975); United States v. Philadelphia National Bank, 374 U.S. 321, 363, 83 S. Ct. 1715, 1741, 10 L. Ed. 2d 915 (1963).
Once the Government has established a prima facie case, the burden of producing evidence then shifts to the defendant *1069 "to show that the market share statistics [give] an inaccurate account of the acquisition's probable effect on competition." U.S. v. Ivaco, Inc., 704 F. Supp. 1409, 1420 (W.D.Mich.1989) (citations omitted). If the defendant successfully rebuts the Government's prima facie case, the burden of proving the anti-competitive effect of the merger lies with the Government which bears the burden of persuasion at all times. United States v. Baker Hughes, Inc., 908 F.2d 981, 983 (D.C.Cir. 1990) (citing Kaiser Aluminum & Chem. Corp. v. FTC, 652 F.2d 1324, 1340 & n. 12 (7th Cir.1981)).
III. DISCUSSION
Reduced to its essentials, a prima facie Section 7 case requires a showing of undue concentration in a particular product market in a particular geographic area. In the instant case, the parties have stipulated that the relevant product market in which to analyze United Tote's acquisition of Autotote is the manufacture and provision of on-track, off-track and inter-track totalisator systems and services. D.I. 105, ¶ 5. The parties have also agreed that the relevant geographic market is North America, consisting of the United States, Canada and Mexico. Id., ¶ 4.
With regard to concentration, the parties have stipulated that prior to the acquisition, three firms United Tote, Autotote and AmTote designed, manufactured, leased, operated and serviced virtually all of the totalisator systems used in North America. Of the remainder, Control Data Corporation ("CDC") provides a relatively small amount of totalisator services to two regional off-track betting corporations in New York. In addition, there are several small companies which supply totalisator services on a seasonal basis to very small tracks and racing fairs.[5]Id., ¶ 8.
The parties also stipulated that approximate market shares in the totalisator market in 1987, 1988 and 1989 were:
1987 1988 1989
AmTote 60% 58% 55%
Autotote 25% 26% 27%
United Tote 11% 12% 13%
CDC 5% 4% 4%
Others .5% .5% .5%
The Government's position is that United Tote's acquisition of Autotote increased United Tote's share of the North American totalisator market from 13% to 40% and reduced the number of meaningful competitors in that market from three to two. Moreover, the Government contends that prior to the acquisition, the two leading firms in the market accounted for 82% of the revenues while after the acquisition that share rose to approximately 95%. Finally, after applying the Herfindahl-Hirschman Index ("HHI") to this market, the Government asserts that the HHI prior to acquisition was 3940 and the HHI after the acquisition was 4640.[6] Based on the *1070 foregoing, the Court concludes that the Government has established a prima facie case. See, e.g., Hospital Corp. of America v. FTC, 807 F.2d 1381, 1384, 1392 (7th Cir.1986), cert. denied, 481 U.S. 1038, 107 S. Ct. 1975, 95 L. Ed. 2d 815 (1987) (violation of Section 7 where acquisition increased market share of second largest firm from 14% to 26% and increased the market share of the four largest firms from 79% to 91%); FTC v. Coca-Cola Co., 641 F. Supp. 1128, 1134, 1139 (D.D.C.1986) (two-firm concentration ratio of 66.3% and post-acquisition HHI of 2650 violated statute).
As stated previously, statistical evidence of market share and concentration data is not determinative in a Section 7 case. See Baker Hughes, 908 F.2d at 984 ("[e]vidence of market concentration simply provides a convenient starting point for a broader inquiry into future competitiveness"); FTC v. Warner Communications, Inc., 742 F.2d 1156, 1163 & n. 1 (D.C.Cir. 1984) ("statistics concerning market share and concentration are not conclusive indicators of anti-competitive effects, but they provide a meaningful context within which to address the question of the merger's competitive effects."). Accordingly, United Tote may successfully rebut the Government's statistical case by affirmatively showing that the merger between itself and Autotote is unlikely to substantially lessen competition or by discrediting the data underlying the Government's prima facie case. See Baker Hughes, 908 F.2d at 991. United Tote has raised several contentions in this regard.
A. Statistical Evidence
United Tote's first argument is that the Government's statistical analysis is not competent to satisfy its burden of proof or establish a prima facie case. In particular, United Tote contends that the Government's HHI analysis is overstated and misleading because it fails to recognize the segmented nature of the North American market for totalisator services and the fact that United Tote and Autotote have not competed significantly with one another in substantial segments of the market.[7] As a result, United Tote asserts that the Government's HHI overstates the extent of United and Autotote's effective competition with each other in the past and the extent to which effective competition would occur in the future.
If the totalisator market is divided into tracks grouped by average daily handle, the market shares of United Tote, Autotote and AmTote are as follows:
Segment United Autotote AmTote CDC Total Greater than $1 Million 1.13% 31.74% 67.13% 0.0% 47.9% $200,000 to $1 Million[8] 16.47% 41.01% 37.11% 5.42% 44.0% Less than $200,000 34.16% 39.43% 26.41% 0.0% 8.2%
*1071 As a preliminary matter, the Court notes that even if the North American totalisator market is divided into the suggested market segments, the combined market shares of United Tote and Autotote would establish a prima facie case in each and every segment. Moreover, United Tote's assertion that it does not significantly compete with Autotote in these segments directly contradicts prior statements regarding the state of competition between the two suppliers. For example, in a 1988 10-K, United Tote stated that it faced "two major competitors in the domestic [totalisator] market." PX 16 at 10. In its 1988 Annual Report, United stated that its market share in the totalisator market climbed to 15% "in the face of sharp competition among the three major equipment suppliers." PX 17 at 6. Finally, in connection with acquisition, United Tote referred to Autotote as "[a]nother primary competitor of United...." PX 15 at 50. These statements are significant because the extent to which merging firms regard each other as their primary competition has been a consideration in recent Section 7 decisions. See Ivaco, 704 F.Supp. at 1419-20.
But even if one were to discount these statements, evidence presented by the Government at trial clearly shows that from 1988 to 1990, United Tote and Autotote submitted bids against each other on forty-nine of the one hundred sixteen totalisator contracts for which competitive bids were sought. DX 4; PX 148 at 12. In fourteen of these instances, AmTote did not submit a bid so that United's only competitor during that period on the service contract was Autotote. DX 4. While it is true that Autotote may have bid less aggressively in some cases than others, the Government produced ample evidence of situations where Autotote lowered bids either in response to or in anticipation of United Tote bids. Tr. at 252-53, 327-29, 889-90; PX 157i, PX 155g, PX 154t. In addition, when several medium to large contracts recently became available for bidding, United Tote priced quite aggressively against Autotote in an effort to secure those contracts. Tr. at 776-78.
Despite the forgoing, United points to Autotote's focus on the medium and large track segment and United's lack of success in bidding against it as evidence that United does not significantly compete with Autotote. The Court, however, draws the opposite conclusion. The fact that Autotote concentrates on medium and large tracks does not negate the fact that Autotote is the undisputed revenue leader in the small track segment. Autotote derives approximately 22% of its domestic totalisator revenues from the small track segment which is more than it does from the large track segment. Tr. at 830. In fact, Autotote realizes approximately 85% of its revenues from tracks with average daily handles under 1 million dollars. Id. In addition, the medium track segment constitutes approximately 44% of the market and, as noted by Defendant's expert, "in that segment, and in that segment only, you have roughly equal market presences by United Tote, AmTote and Autotote." Tr. at 1101.
Finally, the fact that United Tote has never replaced Autotote at a track where Autotote was the incumbent and offered a cash/sell system, does not necessarily demonstrate a lack of competition between the two suppliers. Quite to the contrary, this evidence strongly indicates that Autotote is quite willing to aggressively lower prices in order to retain market share. Indeed, the record demonstrates that, at least in one instance, Autotote reduced its totalisator system price by as much as 50% to beat a very aggressive United Tote bid to replace it. See Tr. at 327-29. Accordingly, the Court concludes that the Government's statistical case, whether based upon the entire market or segmented market shares, accurately reflects the state of competition in the North American totalisator market.
B. Ease of Entry
United Tote's second argument is that it is so easy to enter the totalisator market that high market share does not accurately reflect an ability to exercise market power. Market power is defined as the ability and incentive for existing competitors within a market to raise prices for any significant period of time. Merger Guidelines § 3.3, reprinted in 4 Trade Reg.Rep. (CCH) ¶ 13,103 *1072 at 20,562 (1988). A finding that there are no significant additional sources of supply and no substitutes from the consumer's perspective makes it more likely that market share accurately reflects power over price. Ball Memorial Hosp., Inc. v. Mutual Hosp. Ins., 784 F.2d 1325, 1336 (7th Cir.1986). On the other hand, if alternative sources of supply could enter the market with relative ease, then no hypothetical monopolist or cartel could achieve or maintain supra-competitive pricing without attracting new entrants. See Statements of the Federal Trade Commission Concerning Horizontal Mergers, 4 Trade Reg.Rep. (CCH) ¶ 13,200 at 20,902, § III(A)(1) (if entry is easy "it is unlikely that market power, whether individually or collectively exercised, will persist for long.").
In the instant action, the Government has established an imposing statistical case for the presumption that a merged United Tote/Autotote will be able to exercise market power either individually or in combination with AmTote. Thus, the focus of the analysis now shifts to whether market realities and, in particular, the relative ability of a new supplier or "entrant" to enter the totalisator market, denigrates the Government's statistical case of purported market power. United Tote asserts that entry into the totalisator market is quite easy and that there are a host of potential entrants ready and willing to enter the market. Potential entrants identified by United Tote include suppliers of foreign totalisator systems, lottery systems and general transaction-processing systems.
In response, the Government contends that entry is not easy because of the existence of numerous barriers to entry in the North American totalisator market. In particular, the Government asserts that entry into the market is difficult because of the time and effort required to develop or modify a totalisator system and then gain customer acceptance. The Government argues that these two factors, when combined, will either prevent entry altogether or delay it such that any entry that does occur will not be meaningful. This latter effect is commonly known as an "impediment to entry" and consists of "any condition that necessarily delays entry into a market for a significant period of time and thus allows market power to be exercised in the interim." In re Echlin Mfg. Co., Inc., 105 F.T.C. 410 (1985).
1. Technical Barriers
The Government's first asserted barrier or impediment to entry is the time and effort required to develop and produce a competent totalisator system for the North American market. As stated previously, the basic components of a totalisator system are the central computer, the totalisator terminals and the associated software. The time and expense connected with the provision of the central computer system is not really an issue in this case because all major totalisator system suppliers use commercially purchased computers for their systems. For example, United Tote uses Hewlett-Packard machines, Autotote uses Digital Equipment Corporation products and AmTote currently uses a CPU formerly produced by Varian Computers. AmTote's newly announced Spectrum system will rely on an IBM CPU.
On the other hand, totalisator terminals are somewhat more unique and design and performance differences can be noted from one company's terminal to the next. Similarly, the design and development of the proprietary applications software package that runs the computer and terminals also are key considerations. The proprietary software in a cash/sell totalisator system is designed to support a number of different types of wagers, each with a separate pari-mutuel pool, and simultaneously to incorporate the rules and regulations pertaining to payout calculations.[9] These rules and regulations, which vary according to the state in which the track is located and the types of races and wagers involved, cover a plethora *1073 of circumstances such as scratches, dead heats and assorted combinations. The software must be written to cover not only normal situations but also very unusual situations so that the system will be able to calculate and display the payout information shortly after the outcome of an event is official. D.I. 105, ¶¶ 12, 14.
In addition, the totalisator system must have a high processing speed because most wagers are placed within the last few minutes before a race begins. During that last minute rush, the system must process a large concentration of transactions, issue a large volume of tickets and continually recalculate and display odds. Id., ¶ 15. Moreover, the turnaround time for each race is very short, typically 13-30 minutes apart depending on the nature of the races involved. During this time the terminals must sell bets, calculate odds and display them every 30-60 seconds as well as display the amounts wagered on each different pool and each horse or greyhound. Id., ¶ 16.
These technical performance requirements, when combined with customer demand for 100% system reliability, complicate a potential entrant's ability to design or modify totalisator terminals or applications software in a timely manner. It may be assumed, however, that some potential entrants will experience less difficulty than others in meeting these requirements. In particular, the group of potential entrants most likely to experience the least amount of technical difficulty in entering the North American market would be foreign totalisator system suppliers because these suppliers would only have to modify their systems rather than start from scratch. United Tote's expert witness testified that entry by suppliers of foreign totalisator system would be easy because their systems are technologically capable of functioning in North America with only "minor modifications" to the applications software. Tr. at 1159.
Yet, when the Court heard testimony on Autotote's modification of portions of an Australian totalisator system for use in North America, a quite different picture was presented. Autotote's Director of Research and Development at the time testified that, from the initial research phase to when the first system was ready for installation, the entire effort took approximately eighteen to twenty-four months to complete. PX 175c; Tr. at 929-30. See also DX 34 at 17 (testimony of Bill DiStephano before the Nevada Racing Commission) ("we spent two years taking that system, designing it, using the concept that had already been in existence, and just growing that into what we thought the United States market required."). The software effort alone took ten man-years to write and cost approximately 1 million dollars. Tr. at 917; DX 34 at 18. It is also significant to note that Autotote was aided in its effort by assistance by ATL, its former parent, which had already developed a similar system based on the same computer for use in the Australian market. Tr. at 896-97, 916-17, 928-29; PX 175c; DX 34 at 17.[10]
Evidence was also presented in the form of deposition testimony by James Walters, the President of International Totalisator Services ("ITS"), a company which uses foreign software in its totalisator system. In his deposition, Mr. Walters testified that if ITS was to modify its system for use in the North American market it would require five to ten man-years of effort and cost approximately 1 to 2 million dollars. PX 172r, PX 172s; DX 65hh. In this respect, Mr. Walter's estimate is consistent with the level of effort expended by Autotote's software engineers and programmers when they modified ATL's system for use in the North American market.
*1074 The second group of potential entrants suggested by United Tote were suppliers of computerized lottery systems. Like totalisator systems, lottery systems are also composed of CPUs, peripherals, communications lines, remote terminals and displays. DX 6 at 9. While software for the two systems is not identical, United Tote asserted at trial that the two systems shared many features. For example, lottery system software is capable of transmitting wagers, assigning ticket numbers, maintaining winners files and performing accounting and reporting functions. Id. at 9-10. United Tote's expert witness was of the opinion that lottery companies could develop a competent totalisator system with a minimal two to four man-years of effort. Tr. at 1171; DX 6 at 19. At a cost of approximately $60,000 per programmer-year, a simple calculation results in an estimated base expenditure of $120,000 to $240,000.
CDC, a major supplier of lottery systems in the United States, had a markedly different estimate. At trial, a representative of CDC testified that it would take approximately eighteen months and cost several million dollars to develop the necessary software. The representative also estimated that three years and 6 million dollars would be required to develop a competitive terminal. Tr. at 200-01. This estimate was based on CDC's prior experience in developing a now obsolete totalisator system for a South African customer in the mid-80s. Tr. at 198. In connection with this effort, CDC spent approximately eighteen months developing the software, another six months modifying and enhancing the software prior to actual operational use and eighteen months redesigning an existing lottery terminal for use in the totalisator system, all at a cost of approximately 4.5 million dollars. DX 71 at 3.
Autotote's experience in developing its first cash/sell system is also instructive. In 1977, Autotote began converting an on-line lottery system previously developed by Autotote for the Massachusetts lottery into a cash/sell system capable of serving the North American market. Tr. at 914-15. Autotote's manager of hardware engineering at the time testified that the actual conversion process took approximately nine to twelve months at a cost in excess of 1 million dollars. Id. at 915. The hardware conversion required four man-years of effort and the software conversion required ten. Id. While the lottery conversion estimate is slightly less than CDC's estimate in terms of linear time, the Court notes that the testimony referred only to the time spent actually making the software and hardware modifications. Although it is difficult to draw any conclusions, it would appear that any time spent doing feasibility studies, initial research or even debugging was not included in the estimate.
It is also significant to note that when Autotote converted the lottery system it did so at a cost four to five times greater than the estimates offered by United Tote's expert witness. Similarly, Autotote's fourteen man-year effort virtually dwarfed United Tote's two to four man-year estimate. In addition, evidence was presented at trial that suggested that today's complex racing environment might make the utility of converting lottery software into totalisator software somewhat questionable. CDC's representative, for instance, stated that the practical difference between the two types of software was "[a]lmost a hundred percent." Tr. at 117. Furthermore, the President of GTECH, a leading lottery system supplier, stated in a deposition that a lottery system would need "[b]rand new software" to function in a totalisator environment. PX 170w.
Finally, Linda Shelhamer, the daughter of Lloyd Shelhamer and the Chief Financial Officer of United Tote, stated in a 1987 letter to the Justice Department that "The requirements of a lottery system and an on-track system are so diverse, that a lottery system could not function as an on-track system. It would also take a complete rewrite of lottery software and modification of the terminal for it to be semi-functional." PX 2 at 2. See also PX 9, Admission 10.
United Tote's third group of potential competitors who would be technologically capable of developing operating totalisator *1075 systems are suppliers of transaction-processing software systems. Tr. at 1172-75. United Tote's expert did not believe that these types of companies were technologically as close as foreign totalisator suppliers or lottery companies but was of the opinion that they were capable of doing the job. Id. at 1175. The expert did not offer any time or effort estimates with regard to these competitors so the Court will assume that these competitors would develop a system from scratch.
In an affidavit given in 1987, Lloyd Shelhamer estimated that, based on United Tote's experience in developing its own system, it would take a prospective new entrant at least eighteen to twenty-four months at a cost of approximately 3 million dollars to produce an operating system. Tr. at 560. Prior to this litigation, Linda Shelhamer also estimated that it would take a new entrant a year and a half to develop a totalisator system as well as an additional six to twelve months to debug the system. PX 2 at 7.[11] In addition, James Pierce, the President of United Tote, stated shortly before the litigation began that it would not take "a heck of a lot less" than two years and 5 million dollars to develop a totalisator system. PX 166b.
Moreover, when Datatrol, the predecessor company of GTECH, developed a now outmoded totalisator system it took twenty to thirty people two to three years at a cost of 5 million dollars. PX 170w. Finally, in passing, the Court notes that by April of 1991, AmTote will have been developing its new Spectrum system for at least eighteen months and possibly longer. Tr. at 577, 719. Based on the foregoing, the Court finds that the evidence is overwhelming that it takes at least eighteen to twenty-four months to study, develop and then adequately debug a truly competitive totalisator system.
2. Reputational Barriers
The significance of reputational barriers to entry in antitrust analysis is a somewhat unsettled question at the present moment. A number of courts give the existence of such barriers at least some weight in their analysis. See, e.g., Henry v. Chloride, Inc., 809 F.2d 1334, 1342 (8th Cir.1987) (significant customer loyalty and reliance upon expertise of existing salesman); Southern Pacific Communications v. A.T. & T., 740 F.2d 980, 1002 (D.C.Cir.1984) (brand preferences created by longstanding existence in market); United States v. Crowell, Collier and MacMillan, Inc., 361 F. Supp. 983, 1003 (S.D.N.Y.1973) (buyers looked to reputation for quality and personal experience with company's product). See also Bain, Barriers to New Competition 210 (1962) ("the advantage to established sellers accruing from buyer preferences for their products as opposed to potential entrant products is on the average larger and more frequent ... than any other barrier to entry."). Other courts, however, give reputational barriers relatively short shrift. See, e.g., Baker Hughes, 908 F.2d at 989 n. 10 (relegating discussion of reputational concerns to footnote). One court has even gone so far as to state that "[w]e fail to see how the existence of good will achieved through effective service is an impediment to, rather than the natural result of, competition." United States v. Waste Management, Inc., 743 F.2d 976, 984 (2nd Cir. 1984).
In reality, the weight accorded to reputational barriers to entry will vary with the circumstances of a given case. Typically, evidence that consumers were reluctant to consider new entrants in the *1076 past is an unreliable indicator of the incumbents ability to exercise market power in the future. The explanation for this observation is quite simple. In a truly competitive market, product sellers compete vigorously on both quality and price. A consumer satisfied with the state of affairs might thus be reluctant to try new sellers, if only out of a sense of inertia. On the other hand, if the nature of the market drastically changed such that price and/or quality suffered, then the consumer's perception of new entrants might change.
It is primarily for this reason that courts have rejected evidence of a given market's vigorous pre-merger price competition as indicative of whether new competitors are likely to enter or not. See, e.g., United States v. Syufy Enterprises, 903 F.2d 659, 667 (9th Cir.1990) ("Confronted with this record and the district court's clear findings, the government trots out a shopworn argument that we had thought long abandoned: that efficient, aggressive competition is itself a structural barrier to entry."). Because quality is just the flip side of the competition coin, it would appear that pre-merger reputational barriers should also be given short shrift in most cases. In some markets, however, switching suppliers entails real costs such as retraining workers or downtime due to installation. In other markets, the need for reliability is so great and the consequences of new product failure so dire that, even if the competitive nature of the market deteriorated, consumers would still be reluctant to switch to new entrants. The totalisator market combines elements of both of these scenarios.
For example, because pari-mutuel wagering is the primary business of a racetrack and is its principal source of revenue, the totalisator system that supports its wagering activities is a critical element in the track operation. It is crucial to the track's financial success that its totalisator system performs with consistent reliability. Malfunctions in the system or its components can cause slowdowns or even suspensions of wagering and racing resulting in losses of revenue and of the goodwill of the track's patrons and the horsemen participating in its races. D.I. 105, ¶ 20.
Some racetracks run on-track races relatively few days a year and the failure of a racetrack's totalisator system on a big day would cause the racetrack to lose pari-mutuel handle, concession and admission revenues and possibly require it to pay employees and horsemen's purses. Tracks would regard the loss of a day's revenue due to a totalisator system's failure to be unacceptable. Thus, the proven ability to provide reliable systems and service generally is an important factor in a racetrack's selection of a totalisator supplier. Id., ¶ 21.[12]
The historical record of entry in the totalisator market has only heightened racetrack owner's fears in this regard. In the early 1980s, perhaps the golden age of new entry into the totalisator market, a number of new firms attempted to penetrate the then two-firm totalisator market. This spate of entry was brought on by the occurrence of two events. First, the incumbent suppliers, AmTote and Autotote, raised their prices by 50%, ostensibly to recoup the research and development costs associated with the introduction of the cash/sell system. Tr. at 827. Second, demand far outstripped supply. Small track owners, in particular, were told that they might never see the new technology in their lifetime. Tr. at 545.
Five new firms attempted to enter the market during this period; only one succeeded, United Tote. When ITS attempted to enter, its system was so unreliable that its customers terminated their contracts early and ITS was forced to withdraw from the market in disgrace. Its performance at a Sacramento harness track had so many system failures and other problems that the customer canceled the contract. PX 9 *1077 at 22-23, PX 136 at 14; Tr. at 67, 998. As a result of that failure and another unsatisfactory experience with the ITS system at Audobon Raceway, Rockingham Park canceled its contract with ITS shortly before opening the track because it did not believe that the ITS system could satisfactorily perform the job. Tr. at 66-68; PX 172h. ITS also had a contract to provide totalisator services to one or more greyhound tracks in Colorado which also was terminated by the customer due to performance problems and the failure of ITS to meet certain schedules and requirements of the contract. PX 135 at 8, PX 9 at 22.
GTECH also experienced difficulties when it tried to enter the totalisator market. GTECH entered the market in 1982 when it acquired a company that provided totalisator systems and services to small racetracks in North America. PX 1 at 18. GTECH's largest customer was Pocono Downs which had an average daily handle of approximately $184,000. GTECH's system suffered from continual problems and Pocono Downs negotiated the termination of the contract. Tr. at 350, 767; PX 1 at 18. Thereafter, GTECH sold off its remaining contracts and exited the market. PX 1 at 18.
Dynatote, a totalisator company formed by a group of racetrack owners, used a totalisator system developed by a Spanish totalisator supplier. Tr. at 616-17, 823-24. The Dynatote system, however, was a "total disaster" at one of the racetracks it supplied and its poor performance allegedly contributed to the demise of the track. Tr. at 264-65. The system also performed quite poorly at the Meadows, a racetrack in Pennsylvania, where one observer noted that Dynatote "almost sunk that track because of the same operational problems" experienced at the other track. Tr. at 265. The system also had "very serious" problems at the Freestate Raceway in 1983. The system occasionally shutdown completely but often operated slowly, leading to long lines at the wagering windows and "very unhappy" patrons. These problems hurt the track financially and Freestate canceled the Dynatote contract after one season. Tr. at 856-57.
CDC's attempt to enter the market is also instructive. CDC was already a participant in the totalisator market by virtue of its longstanding New York OTB contracts and its sale of at least one totalisator system to a racetrack in Natal, South Africa. Tr. at 120. Unlike the other new entrants, CDC apparently targeted the medium to large track segment of the totalisator market to obtain a toehold in the market. CDC solicited various racetrack contracts in the early 1980s but was uniformly unsuccessful. Tr. at 120-21; PX 151x; DX 71 at 3. The reluctance of racetracks to try a new system and CDC's inability to match the prices offered by the other existing totalisator suppliers were factors in its lack of success. Tr. at 120-22, 134-35, 1021. It is also significant to note that in the mid-70s, when CDC offered to supply a track with totalisator service for free, the track rejected the offer because it was uncomfortable with a totalisator system that it had not seen perform dependably and reliably. Tr. at 351-52.
Competition among the three major suppliers to replace each other at established racetracks is also informative. In the last five years, on every totalisator contract available for competitive bidding, United Tote has been uniformly unsuccessful in attempting to replace Autotote as an incumbent supplier and has replaced AmTote on only five occasions. Tr. at 626-27. AmTote has also been unable to replace a single incumbent Autotote cash/sell system and replaced only one United Tote system. Tr. at 627-28. Autotote has been the most successful, replacing two United Tote systems and an unspecified number of AmTote systems. Id.
The Court has gone into considerable detail concerning the historical entry record because it demonstrates several important characteristics about the North American totalisator market. First and foremost, the record of entry failure underscores the technical difficulty in developing a competitive totalisator system. With the exception of United Tote, each and every new competitor that entered the market was forced out, not because of price competition but *1078 rather because they could not develop an adequate system. Second, as dramatically illustrated by CDC's experience, unless the racetrack owner was an investor in the new competitor, no new entrant during this period was able to persuade the medium to large track segment to accept their system. Finally, of the racetrack owners who were willing to "go first", mainly small track owners faced with higher prices and limited supply, many suffered tremendous financial hardships as a result of the system's failures.
Another reason that track owners are reluctant to switch totalisator suppliers is the fact that, while significant, totalisator costs constitute a very small percentage of overall operating expenses. PX 9, Admission 75; Tr. at 568. As stated in Lloyd Shelhamer's 1987 affidavit, a 10 to 20% increase in a track's totalisator expenses would probably increase its total operating costs by 1% or less. PX 1 at 15-16. One racetrack operator, whose track's average daily handle was $800,000, testified that a 10% increase in totalisator costs would amount to approximately $50,000 a year, a cost he would swallow, "if it meant that my customers are all going to be happy." Tr. at 53. On the other hand, if a new system were to fail for even one day at a large track, this could mean a short-term loss of 1 million dollars or more in gross operating revenue.
Long-term losses are also substantial. One racetrack operator testified that within one hundred twenty miles of his facility, there were fifteen to sixteen pari-mutuel programs consisting of thoroughbred racing, harness racing, greyhound racing, jai alai and OTB. Tr. at 39. If his facility's totalisator system were to fail on a consistent basis, patrons would blame the track and go elsewhere. Id. This position is well supported in the record. Tr. at 229, 321, 856, 1067. If attendance declined, non-wagering revenue from such sources as admissions, parking and concessions would also fall. PX 1 at 13-14.
The above factors only emphasize the well-supported proposition that most racetrack owners are reluctant to contract with a totalisator supplier that does not have a record of performance at tracks of similar or greater size. PX 9, Admissions 28, 72; Tr. at 757. Tracks want to see a system perform at a comparable track to know that it has adequate processing speed for their needs and that it will not become overloaded and stop or slowdown. Tr. at 231. One possible reason is that a totalisator system that functions adequately at a small track might run into difficulty serving a track with greater wagering volume. PX 1 at 12.
The unwillingness of medium and large tracks to use a totalisator system unproven at a comparable track is also supported by United Tote's own experiences in expanding its initial customer base. According to the Shelhamers, larger racetracks were reluctant to use United when they could not see its system in operation at a track of similar wagering volume. Tr. at 569, 721. Large tracks refused to contract with United Tote even though United offered prices it believed to be lower than those of its competitors. Tr. at 569-70. After ten years of providing cash/sell services and over thirty years in the totalisator supply business, United Tote has only just begun to persuade large tracks of the capabilities of its system. PX 16 at 8, PX 1 at 11. As of August 1987, United Tote essentially served no tracks or wagering systems with average daily handles exceeding 1 million dollars. PX 1 at 11. Today, United Tote serves only two contracts in this market segment and both are in California. Tr. at 1090; DX 5, Attachment 2a.
In response, United Tote presented testimony that indicated that some racetrack owners and organizations might be willing to "go first" if they could otherwise be persuaded of system reliability. Tr. at 747-48, 962-64, 999; DX 55o. United Tote then elicited testimony from the Government's racetrack witnesses that they would at least consider a new system if it reliably performed at a comparable racetrack for one to two years. Tr. at 87, 279-80, 313, 349, 367. Based on this evidence, United Tote concluded that the Government had failed to demonstrate the existence of a "reputational barrier" that would preclude *1079 entry by companies that lack a "proven track record" of performance in the totalisator market.
Even accepting United Tote's optimistic, if somewhat unlikely scenario, the Court cannot conclude that entry into the totalisator market is easy. As discussed previously, the evidence overwhelmingly demonstrates that it would cost 1 to 2 million dollars and take eighteen to twenty-four months to research, develop and then adequately debug a competitive totalisator system. Then, once the system was technically ready for the North American market, the new system would still have to operate flawlessly at a North American track for another twelve to twenty-four months before the market as a whole would be ready to accept it. Thus, a total of two and one-half to four years would pass before a new entrant would be able to exert any competitive influence on the totalisator market. In addition, the Court notes that even if a theoretical firm were to get a toehold at a medium track, a year's worth of trouble-free operation under this scenario would only satisfy tracks of comparable size or smaller. Larger tracks would remain unaffected.
Therefore, the Court concludes that because entry into the totalisator market with a competitive product, even by firms in related industries, would be costly and time consuming, the threat of actual or potential entry by new firms would not pose a significant constraint on unjustified price increases in the market. See Syufy, 903 F.2d at 666 n. 11 (entry easy where new competitor in less than two years not only successfully entered market but also was operating more first-run screens); Ivaco, 704 F.Supp. at 1420 (entry difficult where it would take approximately three years and cost between 2.5 and 3 million dollars); United States v. Calmar, Inc., 612 F. Supp. 1298, 1306 (D.N.J.1985) (entry easy where it would take a year and a half and cost approximately half a million dollars); FTC v. Bass Brothers Enterprises, Inc., 1984-1 Trade Cas. (CCH) ¶ 66,041 at 68,613-14, 1984 WL 355 (N.D.Ohio 1984) (entry difficult where it would take three to five years and cost approximately 50 million dollars); In re Echlin, 105 F.T.C. 410 (entry easy where it would take as little as 500 dollars and less than a year to successfully enter the market). See also Merger Guidelines § 3.3, reprinted in 4 Trade Reg.Rep. (CCH) ¶ 13,103 at 20,562 (1988) (stating that entry will be considered "easy" if can be successfully accomplished within a two year time period).
At trial, United Tote's economic expert suggested that the Merger Guideline's suggested a two year period for entry was "highly arbitrary" and that perhaps a longer period "might be a much better framework for analyzing this kind of activity." Tr. at 1253-54. While this may be true in some circumstances, certain aspects of the racing industry exacerbate the anti-competitive effects of an unjustified totalisator cost increase.
First, racetrack owners and operators must contend with significant restrictions on the amount of profit that they can make as well as their ability to pass along increased costs to consumers. For example, state law and related regulations strictly control the allocation of the wagering dollar in the pari-mutuel industry. Generally speaking, the substantial majority of the wagering dollar is required to be allocated to the winning pool and thus is returned to the betting public that places winning bets. The percentage of the wagering dollar that is returned to the winning bettors varies by bet type and from one jurisdiction to the next. As a general rule, however, more than 80% of each dollar that is wagered is returned to the winning bettors. D.I. 105 at 5.
State laws and related regulations also provide for the allocation of a percentage of the wagering dollar to state taxes and provide that another specified percentage of the wagered dollar be distributed to the horsemen or the providers of the greyhounds. Finally, state law and regulation control the percentage of the wagering dollar that is to be retained by the racing facility itself. Id. For example, one racetrack operator testified that in his jurisdiction, 81% of the wagering dollar is required by law to be returned to the betting public *1080 while another 8.25% is required by the state to be allocated for the horsemen and taxes. Tr. at 28. The remainder goes to the operator of the facility to cover expenses and hopefully generate a profit. Id.
Second, totalisator service contracts tend to be, on the average, three to five years in length. PX 150g, PX 169z, PX 15 at 45, 63. If, after the merger, the remaining incumbents were able to unjustifiably inflate totalisator service costs, the racetracks that entered into these contracts would be locked-in for the duration of the contract. Over two years, a majority of the three year and a significant portion of the five year contracts will have come up for renewal. Accordingly, the Court concludes that a two year time frame is an appropriate measure of the period in which significant anti-competitive harm can occur in the absence of entry.
3. Actual Entry
Actual entry may also be used to rebut a statistical case of purported market power. A dramatic example of this actual entry may be found in Syufy Enterprises where the government brought an antitrust action against an operator of a chain of movie theaters. Syufy, 903 F.2d at 662. The antitrust violations purportedly arose when the defendant, through a series of mergers and acquisitions, purchased all of the first run movie theaters in the Las Vegas, Nevada, market.[13]Id.
The strength of the government's statistical case was impressive. In a scant three years, the defendant turned what was already a concentrated four firm market into a virtual monopoly. Nonetheless, the Ninth Circuit agreed with the lower court that there was nothing structurally about the Las Vegas market which gave the incumbent a permanent pricing advantage over new entrants. Indeed, the Syufy court characterized the industry as "rough-and-tumble, marked by easy market access, fluid relationships with distributors, an ample and continuous supply of product, and a healthy and growing demand." Id. at 667.
The most striking evidence concerned actual entry into the market. Approximately seven days after Syufy established market dominance, he attempted to exercise market power against a film distributor by refusing to pay contractual guarantees in connection with the showing of the film "The Cotton Club." The distributor, Orion Releasing Group, promptly sued Syufy for breach of contract, licensed the film to a second run movie theater owner and to this day refuses to play its films in any Syufy theater. Id. at 670. The second run movie theater owner, emboldened by the Orion licensing agreement, opened up three new cinema multiplexes within a thirteen month period. Within two years, the competitor was operating more screens than Syufy and within two and one-half years, Syufy's percentage of exclusive exhibition rights dropped 52% and its percentage of box office receipts dropped 18%. Id. at 666.
Similarly, in Baker Hughes, the D.C. Circuit found it significant that two new companies had recently entered the United States drilling rig market by winning one contract apiece, "and were poised for future expansion." 908 F.2d at 988-89. This finding was bolstered by the fact that when a previous competitor entered the market, it was unable sell a single drilling rig for two years. Then, by lowering its price and improving its service, it became the market leader within five years. Id. at 989.
At trial, United Tote introduced evidence that ITS submitted a number of bids to tracks in North America in 1990 and that it also contacted United Tote and Autotote customers to seek the opportunity to bid for their business in the future. DX 65cc, DX 65ff; Tr. at 240, 279, 660-61, 665, 878-79. United Tote also presented evidence that the President of ITS, Mr. Walters, believed that he could have modified his software, obtained regulatory approval, built terminals, have his system installed *1081 and ready for use in training track personnel within seven months. DX 65ee, DX 65k.[14] Mr. Walter's deposition testimony was also used to demonstrate that ITS had won two contracts in the North American market in recent months, although both are contingent on the receipt of regulatory approval to construct the track. DX 65cc, DX 65ff. In May of 1990, ITS announced its acquisition of McKinnie & Associates, a domestic computer company specializing in information management systems for racing organizations. DX 14. ITS also has taken out advertisements in racing publications and attended trade conventions in an effort to increase its market presence. DX 27; Tr. at 654-55, 666. United Tote concludes, citing to Baker Hughes, that these activities make ITS a viable and committed entrant capable of rebutting the Government's statistical case of purported market power.
As a preliminary matter, United Tote's reliance upon Baker Hughes is misplaced. The crucial aspect of Baker Hughes was not that actual competitors had entered the market and established a toehold but rather that the leading firm's "growth suggests that competitors not only can, but probably will, enter or expand if this acquisition leads to higher prices." 908 F.2d at 989. Moreover, market shares were characterized as "volatile and shifting." Id. at 986. Thus, much like Syufy, where the market was characterized as "rough and tumble", new entrants could be expected to expand market share rapidly. This is clearly not the case in the totalisator industry. Technical requirements and the reliability issue, when combined with long-term contracts and a racetrack's inclination to stay with its current supplier, hinder both new entrants and incumbents in their efforts to gain market share or effect prices. See also Merger Guidelines, § 3.3 (emphasizing "magnitude" of entry); FTC Merger Statement, § III A(1) ("Evidence of substantial expansion by firms already in an industry, especially nondominant firms, may persuasively indicate that barriers to larger scale are not high. Conversely, evidence of frequent entry, but on a small scale without significant expansion by fringe firms, may also suggest the existence of barriers to larger scale.").
Second, ITS is burdened by the terrible performance record it incurred when it previously attempted to enter the market. Tr. at 69-70, 998; PX 9 at 22-23, PX 172h, PX 135 at 8. As observed by Linda Shelhamer in 1987, if ITS were to reenter the totalisator market, it is "unlikely racetracks would take a chance on them." PX 2 at 5. In her opinion, "it would take many years at very small racetracks before they had an opportunity to gain any significant marketshare." Id. at 5-6.[15]
Third, ITS generally does not lease, operate or service its totalisator systems, preferring instead to sell them outright. PX 172b. In stark contrast, established racetracks in the North American market typically prefer leasing rather than purchasing totalisator equipment. Tr. at 790; PX 154i. One reason for a racetrack's reluctance to *1082 purchase equipment is that they do not want to discover shortly thereafter that their equipment has become obsolete. PX 168r. Lease agreements are also preferable because software updates may be obtained free of charge. Tr. at 974-75. In addition, tracks lease systems to get "a guarantee of performance" and require the totalisator company to supply operating personnel and be responsible for repairs and maintenance. PX 10 at 7. Finally, one racetrack owner who had experience in both leasing and owning stated that he would be unwilling to purchase another totalisator system because they are very expensive to own and operate in the long run and require the payout of a large amount of cash up front. Tr. at 1324-26.
Finally, despite the fact that ITS is actively bidding in the marketplace, United Tote was unable to offer even a single example of a competitor adjusting its prices in response to an ITS bid. Quite to the contrary, on one recent bid, ITS's price was almost twice that of AmTote's and 50% higher than Autotote's once the cost of buying was converted to the cost of leasing. Tr. at 1324. In addition, despite the fact that the buyer specifically requested a leasing arrangement, ITS did not comply. Id. at 1325-26. Finally, the buyer was troubled by ITS's lack of "any history of their operations in the United States. We didn't have anybody that we could go to and look at the operation and feel comfortable with it." Id. at 1325. The buyer ultimately rejected the bid. These considerations inescapably lead the Court to only one conclusion, actual entry by ITS would not constrain anti-competitive price increases by incumbents.
C. Other Considerations
United Tote also contends that a number of other considerations mandate the conclusion that its acquisition of Autotote will not tend to lessen competition in the market for totalisator system manufacture and service. First, United Tote argues that the racing industry is experiencing dramatic changes, including consolidation of wagering networks into statewide or regional systems, interstate commingling of betting pools, blending of lottery and racing wagers, growth of off-track betting and the advent of sports wagering. United Tote asserts that, without the merger, United will be unable to sustain the research, development, production and service efforts that an increasingly complex racing industry will demand.
Second, United Tote argues that the merger will be pro-competitive in that it will generate synergistic benefits and efficiency gains. In particular, United Tote asserts that the merger will finally make United a full-spectrum supplier to the large wagering networks which will include both large and small tracks. Finally, United Tote suggests that racing track owners are strong and sophisticated and possess the leverage to thwart any anti-competitive conduct that might result from an increase in seller concentration. Moreover, United Tote contends that many of these racetrack owners and operators support the merger and that this factor should be given great weight. The Court will address each of these contentions in turn.
1. Market Changes
The essence of United Tote's first argument is that dynamic technological and structural changes in the totalisator market render any analysis based on present market shares an inaccurate measure of future market power. In particular, United Tote contends that without the merger it will lack access to the large track segment and thus will gradually be squeezed out as the small and medium track segments are consolidated into large racing networks. See, e.g., DX 5 at 23; Tr. at 876-78, 1083, 1102-03. This assertion, in turn, is based on the premise that as racetracks are consolidated into large interdependent networks, the larger tracks will dictate to the smaller tracks who the system's totalisator vendor will be. See, e.g., Tr. at 1014-15. This is a scenario that would allegedly favor AmTote and Autotote over United Tote. Tr. at 549, 551, *1083 645.[16] United Tote argues that, standing alone, United lacks the resources and strategic market position to produce sufficient equipment to keep up with the changing market. Tr. at 639.
As a preliminary matter, the Court finds United Tote's somewhat speculative position severely underestimates its own future competitive abilities. For example, United Tote's current system can handle on-track, off-track and telephone account wagering; handle betting on up to five simultaneous events; and accept up to seven or eight interfaces from other totalisator suppliers for inter-track and off-track wagering. See PX 16 at 6. Since 1988, United Tote has provided totalisator services for on-track and inter-track wagering to the California Authority of Racing Fairs ("CARF") which operates seven live racing meets and satellite wagering at fifteen facilities. PX 45 at 10. The largest of these meets has an average daily wagering volume of 2.4 million dollars and a high of 3.5 million dollars, another has an average daily handle of 2.1 million dollars. Tr. at 986, 1002. United Tote also provides a totalisator system to Los Alamitos, a California racetrack whose average daily handle exceeds 1 million dollars per day. Tr. at 942. In addition, United Tote also serves off-track and inter-track wagering networks in California, New Mexico, Wyoming, North and South Dakota and Montana. Tr. 577-78.
Second, the evidence produced at trial shows that the dynamic changes in the totalisator industry forecast by United Tote are still in their infancy. Interstate commingling of racing pools has only been in existence since the spring of 1990. See supra note 2. As to statewide networks, with the possible exception of New Jersey, every other state in North America is currently served by multiple totalisator vendors who communicate through established protocols. Furthermore, the evidence over-whelmingly demonstrates that, to date, it is common ownership and not inter-system dependence that thus far has been the main driving force behind single vendor inter-track systems. See, e.g., Tr. at 626-28, 632. It is also important to note that inter-track wagering, off-track betting and "commingling" of pools across state lines are all regulated activities and that typically legislation must be passed before a given state will allow racetracks to engage in such activities. It was estimated at trial that it would take approximately three to five years before many of the major states passed such legislation. Tr. at 290-91.[17]
Turning to United Tote's contention that, without the merger, it will be financially and strategically unable to keep up with the technological changes, the Court finds it unsupported in the facts and in the cases on which United relies. In Lektro-Vend Corp. v. Vendo Co., 660 F.2d 255, 274-76 (7th Cir.1981), cert. denied, 455 U.S. 921, *1084 102 S. Ct. 1277, 71 L. Ed. 2d 461 (1982), the Seventh Circuit rejected a Section 7 claim based partially on evidence that the market share of the acquired company steadily declined for four years prior to the merger. Similarly, in United States v. International Harvester Co., 564 F.2d 769, 776-779 (7th Cir.1977), the Seventh Circuit upheld a partial acquisition where under the stock purchase agreement the two companies continued to compete after the merger and, prior to the acquisition, the acquired company was unable to secure any additional lines of credit to meet its capital needs. The Eighth Circuit in FTC v. National Tea Co., 603 F.2d 694, 699-701 (8th Cir.1979), upheld a merger where the acquired company had suffered major losses in a very competitive market for the previous five years and was anticipating an imminent departure from the market. In FTC v. Great Lakes Chemical Corp., 528 F. Supp. 84, 96-97 (N.D.Ill.1981), the court approved an acquisition where the acquired company suffered increasing losses over a number of years and severely underutilized production capacity because it was not economically feasible to repair equipment once it was lost to attrition.
In direct contrast to above cases, prior to the merger, both United Tote and Autotote were not only profitable but also saw their market share increase year after year. Furthermore, as discussed above, United Tote's technological capability has not only kept up rapid technological change but in many ways is technically superior to the market leader's current system. With regard to financing, unlike International Harvester, United Tote has failed to show that the merger is necessary to acquire the financial and service capabilities it needs to compete effectively in the future. See also Merger Guidelines at § 3.5 ("the Department will reject claims of efficiencies if equivalent or comparable savings can reasonably be achieved by the parties through other means.").
Accordingly, the Court concludes that United Tote, with its thirty years of industry experience, its ten years of providing a technologically proven product, aggressive marketing and thriving toehold in the large track segment will be able to successfully meet the above market challenges. This also happens to be the view apparently endorsed by Lloyd Shelhamer in February of 1989 when he expressed his belief that United Tote was "well-positioned to take advantage of the new opportunities in the expanding racing industry." PX 17 at 4. The Court also finds it is somewhat ironic that United Tote claims that it will be unable to meet future market conditions without a merger but that it expects a hypothetical new entrant, without any of United Tote's attributes, will nonetheless successfully compete against AmTote and a combined United Tote/Autotote.
2. Pro-Competitive Effects of the Merger
United Tote also presented the testimony of numerous buyers who believed that the merger would be pro-competitive in that the combined companies would be able to produce a more sustained and effective research and development program and improved product innovation efforts. Tr. at 74, 846, 996, 1029, 1045-46. United Tote would allegedly benefit from a larger revenue base as well as an enhanced ability to meet large network service demands. Tr. at 996, 1027-28. Autotote would purportedly benefit from United Tote's managerial expertise, enhanced research and development capabilities and a better range of products. Tr. at 1028, 1031-32. In addition, many racing industry representatives expressed their belief that two strong competitors will better serve their needs and interests than to continue with a dominant AmTote facing individualized competition from United Tote and Autotote. Tr. at 743-44, 995-96, 1027, 1031-32, 1063.
Assuming, without deciding, that a United Tote/Autotote merger would result in valid efficiency gains for the combined entity, the Court nevertheless finds that the pro-competitive aspects of the merger are insufficient to offset its anti-competitive aspects. The Court's finding is guided, in part, by the reality that even if the merger resulted in efficiency gains, there are no guarantees that these savings would be *1085 passed on to the consuming public. In this respect, the instant case is readily distinguishable from United States v. Country Lake Foods, Inc., 754 F. Supp. 669, 1990-2 Trade Cas. (CCH) ¶ 69,113 (D.Minn.1990), a case relied upon by the Defendant.
In Country Lake Foods, the court was influenced by evidence that a merger between the second and third largest sellers in the Minneapolis-St. Paul fluid milk market would result in "significant efficiencies" that would enable the combined firm to "compete head-to-head" with the market leader. Id. at 674, 1990-2 Trade Cas. (CCH) at 64,117. The court, however, characterized the fluid milk market as having an "absence of entry barriers" and that brand loyalty was "an insignificant competitive factor." Id. Moreover, the court also found that future price competition was ensured by the power of the milk buyers, the three largest of whom accounted for more than 90% of industry sales. In particular, the court relied on the size and volume of buyer's purchases, the ready availability of alternate suppliers outside the geographic area and buyer's ability to vertically integrate should those sources of supply become unavailable. Id. at 674, 676, 1990-2 Trade Cas. (CCH) at 64,117, 64,121. See also, FTC v. R.R. Donnelly & Sons Co., 1990-2 Trade Cas. (CCH) ¶ 69,239 at 64,855 (D.D.C.1990) (evidence that market consisted of a few very large customers, along with other characteristics of the market, made anti-competitive consequences unlikely).
As discussed previously, the totalisator system product market is not characterized by easy entry or lack of brand loyalties. Similarly, the North American totalisator market does not consist of a few, very large consumers. In stark contrast, the totalisator market consists of over two hundred fifty-five pari-mutuel facilities, only thirty-nine of which have average daily handles in excess of 1 million dollars. DX 5, Attachment 2a. Even if the Court were to accept United Tote's argument that the owners of these large, sophisticated facilities would be able to protect themselves from any anti-competitive price increase, this would still leave at least one hundred nine facilities unprotected in the small market segment alone. Id. at 8. This problem is aggravated by the fact that large tracks and multi-track owners are already able to achieve lower totalisator prices as a result of their greater volume. See, e.g., Tr. at 46-47. Given the existence of significant barriers to entry, the United Tote/Autotote merger raises the distinct possibility that the existing price differential could widen if totalisator suppliers continued to provide low prices to their large customers but imposed unjustified price increases on the small to medium buyers. See Bass Brothers, 1984-1 Trade Cas. at 68,615 (noting the consequences of a bifurcated exercise of market power). See also Ivaco, 704 F.Supp. at 1427-29 (rejecting sophisticated buyer defense, noting that customers were dependent upon their suppliers, entry was difficult and that the two firms left in the market could collude).
Therefore, based on the foregoing discussion, the Court finds that the presence of some large sophisticated customers in the North American totalisator market would not be sufficient to offset the anti-competitive effects of the merger. In addition, while it appears that the merger may have some pro-competitive aspects, those aspects are simply insufficient to counter the Government's strong case of anti-competitive effect particularly since there is no guarantee that these benefits will be passed along to the small to medium track segments of the market.
3. Divestiture
United Tote's final argument is that divestiture is inappropriate because the financial hardships associated with divestiture would cripple United as an effective competitor and substantially reduce the competitive viability of Autotote as well. In support of this argument, United Tote presented evidence of costly legal fees, increased interest expenditures associated with the uncertainties of the instant litigation, Autotote's failure to meet projections, the illness of Jim Pierce, the Chief Executive Officer of Autotote and of the combined companies, and estimated lost opportunities *1086 in anticipated corporate efficiencies and marketing opportunities. Tr. at 683, 807-11.
As a preliminary matter, the Court finds that the evidence offered by United Tote, consisting solely of the unsubstantiated and somewhat speculative testimony of John Shelhamer, does not demonstrate that Autotote's competitive effectiveness will be reduced as a result of a court-ordered divestiture. The illness of Mr. Pierce and the failure of Autotote to meet United Tote's projections are difficulties that would have occurred in the absence of the merger. More significantly though, given Autotote's excellent reputation in the North American market, its technologically superior product line and its proven profitability, United Tote has failed to show that either one of the above events would render Autotote unable to compete effectively in the totalisator market in the event of a divestiture.
The Court also finds United Tote's predictions of its own imminent demise and financial hardship to be greatly overstated. Of the various grounds advanced by United Tote as contributing to its purported financial difficulties, the only significant one concerns the large amount of debt and attendant cost that United accumulated when it acquired Autotote. This large amount of costly debt, when combined with the specter that a forced sale may not be sufficient to repay all of United Tote's obligations, apparently forms the basis of United's belief that it will be financially "crippled" if forced to divest Autotote.
The Court cannot agree. Of the approximately 60 million dollars in debt that United Tote incurred as a result of the merger, 15 million dollars of that amount is owed to one of United's largest shareholders, Wembly, Inc. Tr. at 810. Another 16 million dollars is owed to former Autotote shareholders who received United Tote stock as part of merger price. Id. at 807-08. Together, these two groups hold a majority of United Tote's outstanding stock. Id. As pointed out by the Government, these shareholders would run the risk of wiping out all of their equity if they chose to force United into bankruptcy rather than reschedule the debt. This argument is particularly compelling because apparently, if not for the acquisition related debt, United Tote would be quite profitable at the present moment.
The Court also questions the appropriateness of a ruling that would allow participants in unlawful mergers to, in effect, "borrow" their way into legitimacy. If all corporations had to do to evade the structure of the antitrust laws was take out potentially ruinous loans prior to engaging in a merger, the Clayton Act would lose its effectiveness. This same idea was succinctly observed by the Supreme Court in United States v. Du Pont & Co., 366 U.S. 316, 327, 81 S. Ct. 1243, 1250, 6 L. Ed. 2d 318 (1960), when it stated that "[t]hose who violate the Act may not reap the benefits of their violations and avoid an undoing of their unlawful project on the plea of hardship or inconvenience."
More importantly though, United Tote offers no reasonable alternative to the Court other than to allow an unlawful merger to proceed. This is clearly not an effective remedy to the anti-competitive effects in the North American totalisator market attendant to United Tote's merger with Autotote. As stated by the Supreme Court in Du Pont & Co.:
If the Court concludes that other measures will not be effective to redress a violation, and that complete divestiture is a necessary element of effective relief, the Government cannot be denied the latter remedy because economic hardship, however severe, may result. Economic hardship can influence choice only as among two or more effective remedies. If the remedy chosen is not effective, it will not be saved because an effective remedy would entail harsh consequences. This proposition is not novel; it is deeply rooted in antitrust law and has never been successfully challenged.
Id. at 327, 81 S. Ct. at 1250-1251 (citations omitted).
Accordingly, the Court concludes that having declared the merger unlawful, divestiture is the appropriate remedy. In *1087 addition, the Court's retention of jurisdiction until divestiture is accomplished can help assure an orderly transition that meets both the requirements of the anti-trust laws and at the same time considers any obvious hardships on the Defendant.
IV. CONCLUSION
Based on the foregoing, the Court finds that the Government's statistical evidence accurately reflects the future of the North American totalisator market and indicates that the merger between United Tote and Autotote would have anti-competitive effects. The Court also finds that potential entry into the totalisator market is not so easy that it could pose a significant restraint to price increases and that neither actual entry nor any of the other considerations advanced by United Tote are sufficient to offset the merger's anti-competitive effects. Accordingly, the Court concludes that United Tote has failed to rebut the Government's prima facie case of a violation of Section 7 of the Clayton Act. Moreover, the Court concludes that, even if United Tote has successfully rebutted the Government's prima facie case, the Government has demonstrated by a preponderance of the evidence that the merger would have anti-competitive effects.
NOTES
[1] While the Court will refer to racing events only, in some jurisdictions pari-mutuel wagering also includes other types of contests, including betting on jai alai events.
[2] Until 1990, wagering pools were only combined on a intrastate basis because it was commonly thought that federal law precluded the commingling of wagering pools across state lines. In the spring of 1990, however, the racing industry reached an understanding with the Department of Justice as to the legality of interstate commingling. Since then, interstate commingling of wagering pools has increased dramatically. Tr. at 245-47, 638-39.
[3] The "handle" of a racetrack or jai alai fronton is the total amount bet on a pari-mutuel event or events. At a racetrack, the "average daily handle" is the mean amount bet daily during a racing season. D.I. 105, ¶ 10.
[4] Section 7 of the Clayton Act provides in pertinent part:
No person engaged in commerce or in any activity affecting commerce shall acquire, directly or indirectly, the whole or any part of the stock ... of another person engaged also in commerce or in any activity affecting commerce ... the effect of such acquisition may be substantially to lessen competition, or tend to create a monopoly.
15 U.S.C. § 18.
[5] These companies consist of Trac-Tote, Ace Totalisator, Colorado Data Systems, Burt Blackhurst and Microtote of Ohio.
[6] HHI is a measure of market concentration calculated by squaring the market share of each firm in the market and then summing the squares. The HHI takes into account the relative size and distribution of the firms in a market. The HHI approaches zero when the market consists of a large number of firms of relatively equal size and reaches a maximum when a market is a monopoly. The Department of Justice's Merger Guidelines characterize a market as unconcentrated if its HHI is between 1000 and 1800. A market is highly concentrated if its HHI is above 1800. United States Department of Justice Merger Guidelines § 3.1 (1984), reprinted in 4 Trade Reg.Rep. (CCH) ¶ 13,103 at 20,561.
[7] The Court notes that the primary reason for the Government's reliance upon total market revenues rather than segmented market revenues is that both parties stipulated to it. While the Court will address United Tote's contentions, it does appear somewhat disingenuous for United to stipulate to a set of facts, unilaterally change its position during trial and then in post-trial briefs argue that the previous stipulation is misleading.
[8] The middle segment of the market was the only part of the totalisator market about which the two parties disagreed and then only to a small extent. United Tote's proposed market shares would be 17.4% for United, 43.4% for Autotote and 39.2% for AmTote. See D.I. 138 at 10, D.I. 135 at 30-31.
[9] There are forty different types of wagers, five different ways of calculating exacta and quintella payoffs, fourteen different ways of calculating double daily payoffs, sixteen different ways of calculating trifecta payoffs, five different ways of calculating simple win pools and eleven different ways to calculate the state's share of the pari-mutuel handle. Tr. at 708.
[10] This testimony was buttressed by the sworn statements of James Baker, a former ATL employee. In his deposition, Mr. Baker testified that:
Autotote developed their own system in the early '80s late '70s early '80s, which is known as the J-25 system. They had a lot of difficulties with that. And they came to Australia at the time and purchased [a cash/sell] system design from ATL. Of course they have taken [it] since and refined it quite a bit to suit American conditions. But that's the genesis of that system.
PX 151c.
[11] At trial, Lloyd Shelhamer testified that his 1987 estimate referred only to the effort level required for a potential entrant with no racetrack or computer experience. Tr. at 560. John Shelhamer testified that he believed that a totalisator system equivalent to the original United System 1000 could be developed today in the same time frame and at the same cost as the original approximately twelve to thirteen months and 1.5 million dollars. Tr. at 598. While John Shelhamer's estimate may indeed be possible, the record demonstrates that the United System 1000 was developed by a company with twenty years of racing experience for use primarily in the small track segment, the system lacked any multi-track capabilities and that to date, no other totalisator supplier has been able to develop a system as quickly or as cheaply as United Tote originally did. Tr. at 599, 709, 712.
[12] Similarly, racetracks face certain "switching costs" whenever they decide to go with a new vendor of totalisator services. For example, when Rockingham Park switched vendors just prior to the start of the racing season, it was forced to accept a higher price, had to rewire the track and build a new computer room. Tr. at 66-71. In addition, all of the pari-mutuel clerks would have had to have been retrained on the new system. This retraining period likely resulted in a net cost to the track in terms of downtime and errors. PX 1 at 16; Tr. at 331.
[13] When a film is first released by a film distributor for public exhibition, it is in its first run. Once the initial demand for the film has somewhat subsided, the film is shipped to a second run theater in the same area which shows it for a lower ticket price. Syufy, 903 F.2d at 665 n. 7.
[14] The Court notes that Mr. Walter's belief that he could deliver a fully operational system within seven months is somewhat at odds with his earlier estimate that it would take five to ten man-years and cost 1 to 2 million dollars to modify his system for use in the North American market. In addition, even if ITS was able to rush a system into the market, it would face potentially insurmountable reputational problems if that system were to experience operational difficulties.
[15] In 1987, Lloyd Shelhamer declared that "ITS withdrew from the North American market with a very bad reputation. Were ITS to decide to reenter, I believe it would take many years of successfully operating at very small tracks before its negative reputation could be overcome." PX 1 at 17-18. At trial, Lloyd Shelhamer revised his earlier opinion on grounds that in the intervening three to four years ITS had licensed software from the Hong Kong Jockey Club for use with its equipment, "matured" as a company, had a good reputation where they were operating and while it was possible that they might go into small racetracks, it was most likely that they would go into statewide networks, which Mr. Shelhamer described as their "forte." TR. at 558. The record demonstrates, however, that despite having a license to sell the Hong Kong software since 1985, the only installation where ITS equipment is being used in conjunction with the licensed software, other than the Hong Kong Jockey Club itself, is at the Macau Jockey Club in Taiwan. DX 23.
[16] United Tote also claims that the introduction of AmTote's new Spectrum system will also erode and ultimately jeopardize United's competitive position. In particular, United Tote asserts that Spectrum's enhanced networking capabilities will substantially increase dominance at the large tracks and also result in a flood of fully depreciated equipment into the small track market once the older systems at the large tracks are replaced. Tr. 623-24, 648-49, 1113. The Court finds this argument unpersuasive given its highly speculative nature as to whether Spectrum will be a success or failure and as to what quantities the system will be available.
[17] The Court also finds that United Tote's argument that, without the merger, it could not effectively compete in the ever increasingly complicated totalisator system market due to research and development requirements, is somewhat undercut by its own 1988 10K form, where it stated that:
Since incurring the costs to develop the United Tote System 1000 in 1979 and 1980, the Company has not spent significant amounts on research and development. Expenditures for research and development are anticipated to increase in the future as the Company continues the development of self-service and portable terminals and begins to adapt the United Tote System 1000 for other uses. The Company believes that the funds necessary for such development will be generated from operations.
PX 16 at 8-9.
Even if United Tote was unable to fund increased research and development costs from operations, United has failed to demonstrate that it would be unable to easily acquire such funds from alternate sources, such as merging with companies outside the market that would not have the same anti-competitive effects as the United/Autotote merger.