DISSENTING STATEMENT OF COMMISSIONER MARY L. AZCUENAGA

in Time Warner Inc., Docket C-3709

The Commission today issues a consent order to settle allegations that the acquisition by Time Warner Inc. (Time Warner) of Turner Broadcasting System, Inc. (Turner), and related agreements with Tele-Communications, Inc. (TCI),⁽¹⁾ would be unlawful. Alleging that this transaction violates the law is possible only by abandoning the rigor of the Commission's usual analysis under Section 7 of the Clayton Act. To reach this result, the majority adopts a highly questionable market definition, ignores any consideration of efficiencies and blindly assumes difficulty of entry in the antitrust sense in the face of overwhelming evidence to the contrary. The decision of the majority also departs from more general principles of antitrust law by favoring competitors over competition and contrived theory over facts.

The usual analysis of competitive effects under the law, unlike the apparent analysis of the majority, would take full account of the swirling forces of innovation and technological advances in this dynamic industry. Unfortunately, the complaint and the underlying theories on which the order is based do not begin to satisfy the rigorous standard for merger analysis that this agency has applied for years. Instead, the majority employs a looser standard for liability and a regulatory order that threatens the likely efficiencies from the transaction. Having found no reason to relax our standards of analysis for this case, I cannot agree that the order is warranted.

Product Market

We focus in merger analysis on the likelihood that the transaction will create or enhance the ability to exercise market power, i.e., raise prices. The first step usually is to examine whether the merging firms sell products that are substitutes for one another to see if there is a horizontal competitive overlap. This is important in a case based on a theory of unilateral anticompetitive effects, as this one is, because the theory requires a showing that the products of the merging firms are the first and second choices for consumers.⁽²⁾

In this case, it could be argued from the perspective of cable system operators and other multichannel video program distributors (MVPDs), who are purchasers of programming services, that all video programming networks⁽³⁾ are substitutes. This is the horizontal competitive overlap that is alleged in the complaint.⁽⁴⁾

One problem with the alleged all-programming market is that basic cable programming services (such as Turner's CNN) and premium cable programming services (such as Time Warner's HBO) are not substitutes along the usual dimensions of competition. Most significantly, they do not compete on price. CNN is sold to MVPDs for a fee per subscriber that is on average less than one-tenth of the average price for HBO, and it is resold as part of a package of basic services for an inclusive fee. HBO is sold at wholesale for more than ten times as much; it is resold to consumers on an a la carte basis or in a package with other premium services, and a subscription to basic service usually is a prerequisite. It is highly unlikely that a cable operator, to avoid a price increase, would drop a basic channel and replace it with a significantly more expensive premium channel. Furthermore, cable system operators tell us that when the price for basic cable services increases, consumers drop pay services, suggesting that at least at the retail level these goods are complementary rather than substitutes for one another.

Another possible argument is that CNN and HBO should be in the same product market because from the cable operator's perspective, each is "necessary to attract and retain a significant percentage of their subscribers."⁽⁵⁾ If CNN and HBO were substitutes in this sense, we would expect to see cable system operators playing them against one another to win price concessions in negotiations with programming sellers. But there is no evidence that they have been used in this way, and cable system operators have told us that basic and premium channels do not compete on price.⁽⁶⁾ There are closer substitutes, in terms of price and content, for CNN (in basic cable services) and for HBO (in premium cable services).

I am not persuaded that the product market alleged in the complaint could be sustained. CNN and HBO are not substitutes, and they are not the first and second choices for consumers (or for cable system operators or other MVPDs). There are no other horizontal overlaps warranting enforcement action in any other cable programming market.⁽⁷⁾ Under these circumstances, it would seem appropriate to withdraw the complaint.

Entry

The complaint alleges that entry is difficult and unlikely.⁽⁸⁾ This is an astonishing allegation, given the amount of entry in the cable programming market. The number of cable programming services or networks increased from 106 to 129 in 1995, according to the FCC.⁽⁹⁾ One source reported thirty national 24-hour networks expected to launch in 1996,⁽¹⁰⁾ and another source identified seventy-three networks "on the launch pad."⁽¹¹⁾ That adds up to between fifty-three and ninety-six new and announced video programming networks in two years. According to an industry trade association, thirty-three new basic networks and thirteen new premium networks were launched between 1992 and 1995.⁽¹²⁾ Another source listed 141 national 24-hour cable networks launched or announced between January 1993 and March 1996.⁽¹³⁾

This does not mean that entry is easy or inexpensive. Not all the channels that have announced will launch a service, and not all those that launch will succeed.⁽¹⁴⁾ But some of them will. Some recent entrants include CNNfn (December 1995), Nick at Nite's TV Land (April 1996), MSNBC (July 1996), and the History Channel (January 1995).⁽¹⁵⁾ The Fox News Channel, offering twenty-four hour news, began service in October 1996, and Westinghouse and CBS Entertainment have announced that they will launch a new entertainment and information cable channel, Eye on People, in March 1997.⁽¹⁶⁾ The fact of so much ongoing entry indicates that at any given moment, entry from somewhere is imminent, and this, translated for purposes of antitrust analysis, means that entry should be regarded as virtually immediate.

Recent entrants have achieved some measure of success. TV Land reports 15 million subscribers (almost 24% of cable households) less than one year after its launch.⁽¹⁷⁾ The History Channel has obtained carriage to more than 40% of cable households in less than two years. Home & Garden Television, launched in December 1994, reports 18 million subscribers (more than 28% of cable households).⁽¹⁸⁾ The SciFi Channel, launched in September 1992, has 36 million subscribers (57% of cable households).⁽¹⁹⁾ The TV Food Network, launched in November 1993, reportedly has 21 million subscribers (about one-third of cable households).⁽²⁰⁾

New networks need not be successful or even launched before they can exert significant competitive pressure. Announced launches can affect pricing immediately. The launch of MSNBC and the announcement of Fox's cable news channel, for example, enabled cable system operators to mount credible threats to switch to one of the new news networks in negotiations with CNN, the incumbent all-news channel.⁽²¹⁾

Any constraint on cable channel capacity does not appear to be deterring entry of new networks. Indeed, the amount of entry that is occurring apparently reflects confidence that channel capacity will expand, for example, by digital technology. In addition, alternative MVPDs, such as direct broadcast satellite (DBS), may provide a launching platform for new networks.⁽²²⁾ For example, CNNfn was launched in 1995 with 4 to 5 million households, divided between DBS and cable.⁽²³⁾

Nor should we ignore significant technological changes in video distribution that are affecting cable programming. One such change is the development and commercialization of new distribution methods that can provide alternatives for both cable programmers and subscribers. DBS is one example. With digital capacity, DBS can provide hundreds of channels to subscribers. By September 1995, DBS was available in all forty-eight contiguous states and Alaska.⁽²⁴⁾ In April 1996, DBS had 2.6 million customers; in August 1996, DBS had 3.34 million subscribers;⁽²⁵⁾ by the end of January 1997, DBS had more than 4.7 million subscribers⁽²⁶⁾ (compared to 62 million cable customers in the U.S.). AT&T last year invested $137.5 millon in DirecTV, a DBS provider, began to sell satellite dishes and programming to its long distance customers in four markets, and planned to expand to the rest of the country in September 1996.⁽²⁷⁾ By the end of 1996, DirecTV had 2.3 million subscribers (up from 1.2 million in 1995⁽²⁸⁾), giving DirecTV more subscribers than all but the six largest cable system operators.⁽²⁹⁾ Echostar and AlphaStar both have launched DBS services, and MCI Communication and News Corp. last year announced a partnership to enter DBS.⁽³⁰⁾ Some industry analysts predict that DBS will serve 15 million subscribers by 2000.⁽³¹⁾ Direct broadcast satellite already is offering important competition for cable systems.⁽³²⁾

Digital technology, which would expand cable capacity to as many as 500 channels, is another important development. DBS already uses digital technology, and some cable operators were planning to begin providing digital service in 1996. Last fall, Discovery Communications (The Discovery Channel) announced four new programming services designed for digital boxes for TCI's "digital box rollout."⁽³³⁾ (Even without digital service, cable systems have continued to upgrade their capacity; in 1994, about 64% of cable systems offered thirty to fifty-three channels, and more than 14% offered fifty-four or more channels.⁽³⁴⁾) Local telephone companies have entered as distributors via video dialtone, MMDS⁽³⁵⁾ and cable systems, and the telcos are exploring additional ways to enter video distribution markets.⁽³⁶⁾ Digital compression and advanced television technologies could make it possible for multiple programs to be broadcast over a single over-the-air broadcast channel.⁽³⁷⁾ When these developments will be fully realized is open to debate, but it is clear that they are on their way and affecting competition. According to one trade association official, cable operators are responding to competition by "upgrading their infrastructures with fiber optics and digital compression technologies to boost channel capacity . . . . What's more, cable operators are busily trying to polish their images with a public that has long registered gripes over pricing, customer service and programming choice."⁽³⁸⁾

Ongoing entry in programming suggests that no program seller could maintain an anticompetitive price increase and, therefore, there is no basis for liability under Section 7 of the Clayton Act. Changes in the video distribution market will put additional pressure on both cable systems and programming providers to be competitive by providing quality programming at reasonable prices. The quality and quantity of entry in the industry warrants dismissal of the complaint.

Horizontal Theory of Liability

The complaint alleges that Time Warner will be able to exploit its ownership of HBO and the Turner basic channels by "bundling" Turner networks with HBO, that is, by selling them as a package.⁽³⁹⁾ As a basis for liability in a merger case, this appears to be without precedent.⁽⁴⁰⁾ Bundling is not always anticompetitive, and we cannot predict when bundling will be anticompetitive.⁽⁴¹⁾ Bundling can be used to transfer market power from the "tying" product to the "tied" product, but it also is used in many industries as a means of discounting. Popular cable networks, for example, have been sold in a package at a discount from the single product price. This can be a way for a programmer to encourage cable system operators to carry multiple networks and achieve cross-promotion among the networks in the package. Even if it seemed more likely than not that Time Warner would package HBO with Turner networks after the merger, we could not a priori identify this as an anticompetitive effect.

The alleged violation rests on a theory that the acquisition raises the potential for unlawful tying. To the best of my knowledge, Section 7 of the Clayton Act has never been extended to such a situation. There are two reasons not to adopt the theory here. First, challenging the mere potential to engage in such conduct appears to fall short of the "reasonable probability" standard under Section 7 of the Clayton Act. We do not seek to enjoin mergers on the mere possibility that firms in the industry may later choose to engage in unlawful conduct. It is difficult to imagine a merger that could not be enjoined if "mere possibility" of unlawful conduct were the standard. Here, the likelihood of anticompetitive effects is even more removed, because tying, the conduct that might possibly occur, in turn might or might not prove to be unlawful. Second, anticompetitive tying is unlawful, and Time Warner would risk private law suits and public law enforcement action for such conduct.

The remedy for the alleged bundling is to prohibit it,⁽⁴²⁾ with no attempt to distinguish efficient bundling from anticompetitive bundling.⁽⁴³⁾ Assuming liability on the basis of an anticompetitive horizontal overlap, the obvious remedy would be to enjoin the transaction or to require the divestiture of HBO. Divestiture is a simple, easily reviewable and complete remedy for an anticompetitive horizontal overlap. The weakness of the Commission's case seems to be the only impediment to imposing that remedy here.

Vertical Theories

The complaint also alleges two vertical theories of competitive harm. The first is foreclosure of unaffiliated programming from Time Warner and TCI cable systems.⁽⁴⁴⁾ The second is anticompetitive price discrimination against competing MVPDs in the sale of cable programming.⁽⁴⁵⁾ Neither of these alleged outcomes appears particularly likely.

Foreclosure

Time Warner cannot foreclose the programming market by refusing carriage on its cable system, because Time Warner has less than 20% of cable television subscribers in the United States. Even if TCI were willing to join in an attempt to barricade programming produced by others from distribution, TCI and Time Warner together control less than 50% of the cable television subscribers in the country. In that case, entry of programming via cable might be more expensive (because of the costs of obtaining carriage on a number of smaller systems), but it need not be foreclosed.⁽⁴⁶⁾ And even if Time Warner and TCI together controlled a greater share of cable systems, the availability of alternative distributors of video programming and the technological advances that are expanding cable channel capacity make foreclosure as a result of this transaction improbable.

The foreclosure theory also is inconsistent with the incentives of the market. Cable systems operators want more and better programming, to woo and win subscribers. To support their cable systems, Time Warner and TCI must satisfy their subscribers by providing programming that subscribers want at reasonable prices. Given competing distributors and expanding channel capacity, neither of them likely would find it profitable to attempt to exclude new programming.

TCI as a shareholder of Time Warner, as the transaction was proposed to us (with a minority share of less than 10%), would have no greater incentive than it had as a 23% shareholder of Turner to protect Turner programming from competitive entry. Indeed, TCI's incentive to protect Turner programming would appear to be diminished.⁽⁴⁷⁾ If TCI's interest in Time Warner increased, it stands to reason that TCI's interest in the well-being of the Turner networks also would increase. But it is important to remember that TCI's principal source of income is its cable operations, and its share of Time Warner profits from Turner programming would appear to be insufficient incentive for TCI to jeopardize its cable business.⁽⁴⁸⁾ It may be that TCI could acquire an interest in Time Warner that could have anticompetitive consequences, but the Commission should analyze that transaction when and if TCI increases its holdings.

Another aspect of the foreclosure theory alleged in the complaint is a carriage agreement (programming service agreement or PSA) between TCI and Turner. Under the PSA, TCI would carry certain Turner networks for twenty years, at a discount from the average price at which Time Warner sells the Turner networks to other cable operators. The complaint alleges that TCI's obligations under the PSA would diminish TCI's incentives and ability to carry programming that competes with Turner programming,⁽⁴⁹⁾ which in turn would raise barriers to entry for unaffiliated programming. The increased difficulty of entry, so the theory goes, would in turn enable Time Warner to raise the price of Turner programming sold to cable operators and other MVPDs.

It is hard to see that the PSA would have anticompetitive effects. TCI already has contracts with Turner that provide for mandatory carriage of CNN and TNT, and TCI is likely to continue to carry these programming networks for the foreseeable future.⁽⁵⁰⁾ The current agreements do not raise antitrust issues, and the PSA raises no new ones. Any theoretical bottleneck on existing systems would be even further removed by the time the carriage requirements under the PSA would have become effective (when existing carriage agreements expire), because technological changes will have expanded cable channel capacity and alternative MVPDs will have expanded their subscribership. The PSA could even give TCI incentives to compete with Time Warner's programming and keep TCI's costs down.⁽⁵¹⁾ The PSA would have afforded Time Warner long term carriage for the Turner networks, provided TCI with long term programming commitments with some price protection, and eliminated the costs of renegotiating a number of existing Turner/TCI carriage agreements as they expire. These are efficiencies. No compelling reason has been advanced for requiring that the carriage agreement be cancelled.⁽⁵²⁾

In addition to divestiture by TCI of its Time Warner shares and cancellation of the TCI/Turner carriage agreement, the proposed remedies for the alleged foreclosure include: (1) antidiscrimination provisions by which Time Warner must abide in dealing with program providers;⁽⁵³⁾ (2) recordkeeping requirements to police compliance with the antidiscrimination provision;⁽⁵⁴⁾ and (3) a requirement that Time Warner carry "at least one Independent Advertising-Supported News and Information National Video Programming Service."⁽⁵⁵⁾ These remedial provisions are unnecessary, and they may be harmful.

Paragraph VII of the order, the antidiscrimination provision, seeks to protect unaffiliated programming vendors from exploitation and discrimination by Time Warner. The order provision is taken almost verbatim from a regulation of the Federal Communications Commission.⁽⁵⁶⁾ It is highly unusual, to say the least, for an order of the FTC to require compliance with a law enforced by another federal agency, and it is unclear what expertise we might bring to the process of assuring such compliance. Although a requirement to obey existing law and FCC regulations may not appear to burden Time Warner unduly, the additional burden of complying with the FTC order may be costly for both Time Warner and the FTC. In addition to imposing extensive recordkeeping requirements,⁽⁵⁷⁾ the order apparently would create another forum for unhappy programmers, who could seek to instigate an FTC investigation of Time Warner's compliance with the order, instead of or in addition to citing the same conduct in a complaint filed with and adjudicated by the FCC.⁽⁵⁸⁾ The burden of attempting to enforce compliance with FCC regulations is one that this agency need not and should not assume.

The order also requires Time Warner to carry an independent all-news channel.⁽⁵⁹⁾ This requirement is entirely unwarranted. A duty to deal might be appropriate on a sufficient showing if Time Warner were a monopolist. But with less than 20% of cable subscribers in the United States, Time Warner is neither a monopolist nor an "essential facility" in cable distribution.⁽⁶⁰⁾ CNN, the apparent target of the FTC-sponsored entry, also is not a monopolist but is one of many cable programming services in the all-programming market alleged in the complaint. Clearly, CNN also is one of many sources of news and information readily available to the public, although neither televised news programming nor ad-supported cable TV news programming is a market alleged in the complaint.

Antitrust law, properly applied, provides no justification whatsoever for the government to help establish a competitor for CNN on the Time Warner cable systems. Nor is there any apparent reason, other than the circular reason that it would be helpful to them, why Microsoft, NBC or Fox needs a helping hand from the FTC in their new programming endeavors. CNN and other programming networks did not obtain carriage mandated by the FTC when they launched; why should the Commission now tilt the playing field in favor of other entrants?

Price Discrimination

The complaint alleges that Time Warner could discriminatorily raise the prices of programming services to its MVPD rivals,⁽⁶¹⁾ presumably to protect its cable operations from competition. This theory assumes that Time Warner has market power in the all-cable programming market. As discussed above, however, there are reasons to think that the alleged all-cable programming market would not be sustained, and entry into cable programming is widespread and, because of the volume of entry, immediate. Under the circumstances, it appears not only not likely but virtually inconceivable that Time Warner could sustain any attempt to exercise market power in the alleged all-cable programming market.

Whatever the merits of the theory in this case, however, discrimination against competing MVPDs in price or other terms of sale of programming is prohibited by federal statute⁽⁶²⁾ and by FCC regulations,⁽⁶³⁾ and the FCC provides a forum to adjudicate complaints of this nature. Unfortunately, the majority is not content to leave policing of telecommunications to the FCC.

The order addresses the alleged violation in the following way: (1) it requires Time Warner to provide Turner programming to competing MVPDs on request; and (2) it establishes a formula for determining the prices that Time Warner can charge MVPDs for Turner programming in areas in which Time Warner cable systems and the MVPDs compete.⁽⁶⁴⁾ The provision is inconsistent with two antitrust principles: Antitrust traditionally does not impose a duty to deal absent monopoly, which does not exist here, and antitrust traditionally has not viewed price regulation as an appropriate remedy for market power. Indeed, price regulation usually is seen as antithetical to antitrust.

Although the provision ostensibly has the same nondiscrimination goal as federal telecommunications law and FCC regulations, the bright line standard in the proposed order for determining a nondiscriminatory price fails to take account of the circumstances Congress has identified in telecommunications statutes in which price differences could be justified, such as, for example, cost differences, economies of scale or "other direct and legitimate economic benefits reasonably attributable to the number of subscribers serviced by the distributor."⁽⁶⁵⁾ These are significant omissions, particularly for an agency that has taken pride in its mission to prevent unfair methods of competition and, in so doing, to identify and take account of efficiencies. There is no apparent reason or authority for creating this exception to a congressional mandate. To the extent that the proposed order creates a regulatory scheme different from that afforded by Congress and the FCC, disgruntled MVPDs may find it to their advantage to seek sanctions against Time Warner at the FTC.⁽⁶⁶⁾ This is likely to be costly for the FTC and for Time Warner, and the differential scheme of regulation also could impose other, unforeseen costs on the industry.

Efficiencies

As far as I can tell, the consent order entirely ignores the likely efficiencies of the proposed transaction. The potential vertical efficiencies include more and better programming options for consumers and reduced transaction costs for the merging firms. The potential horizontal efficiencies include savings from the integration of overlapping operations and of film and animation libraries. For many years, the Commission has devoted considerable time and effort to identifying and evaluating efficiencies that may result from proposed mergers and acquisitions. Although cognizable efficiencies occur less frequently than one might expect, the Commission has not stinted in its efforts to give every possible consideration to efficiencies. That makes the apparent disinterest in the potential efficiencies of this transaction decidedly odd.

Industry Complaints

We have heard many expressions of concern about the transaction. Cable system operators and alternative MVPDs have been concerned about the price and availability of programming from Time Warner after the acquisition. Program providers have been concerned about access to Time Warner's cable system. These are understandable concerns, and I am sympathetic to them. To the extent that these industry members want assured supply or access and protected prices, however, this is (or should be) the wrong agency to help them. Because Time Warner cannot foreclose either level of service and is neither a monopolist nor an "essential facility" in the programming market or in cable services, there would appear to be no basis in antitrust for the access requirements imposed in the order.

The Federal Communications Commission is the agency charged by Congress with regulating the telecommunications industry, and the FCC already has rules in place prohibiting discriminatory prices and practices. While there may be little harm in requiring Time Warner to comply with communications law, there also is little justification for this agency to undertake the task. To the extent that the consent order offers a standard different from that promulgated by Congress and the FCC, it arguably is inconsistent with the will of Congress. To the extent that the consent order would offer a more attractive remedy for complaints from disfavored competitors and customers of Time Warner, they are more likely to turn to us than to the FCC. There is much to be said for having the FTC confine itself to FTC matters, leaving FCC matters to the FCC.

I dissent.

1. Liberty Media Corporation, a wholly-owned subsidiary of TCI, also is named in the complaint and order. For simplicity, references in this statement to TCI include Liberty.

2. See 1992 Horizontal Merger Guidelines 2.2. The theory is that when the post-merger firm raises the price on product A or on products A and B, sales lost due to the price increase on the first-choice product (A) will be diverted to the second-choice product (B). The price increase is unlikely to be profitable unless a significant share of consumers regard the products of the merged firm as their first and second choices.

3. The terms "programming services," "networks," and "channels" are used interchangeably in this statement. For example, The History Channel is a video programming service or network that is sold to MVPDs for distribution to consumers.

4. Complaint 24. Note that this market excludes broadcast programming, which "is a primary source of programming for most viewers regardless of distribution media." Federal Communications Commission, Third Annual Report on the Status of Competition in the Market for the Delivery of Video Programming at 7 (Dec. 26, 1996) (hereafter "1996 FCC Report").

5. Complaint 4 & 9. To the extent that each network (CNN and HBO) is viewed as "necessary" to attract subscribers, as alleged in the complaint, each would appear to have market power quite independent of the proposed transaction and of each other.

6. If the market includes premium cable programming services, it probably ought also to take account of video cassette rentals, which constrain the pricing of premium channels. See Federal Communications Commission, Second Annual Report on the Status of Competition in the Market for Delivery of Video Programming 121 (Dec. 7, 1995) (hereafter "1995 FCC Report"). If the theory is that HBO and CNN (and other networks) compete for channel space (i.e., for carriage on cable systems), the market probably should include over-the-air broadcast networks, at least to the extent that they compete for cable channel space as the price for retransmission rights. See Complaint 34 (alleging "shortage of available channel capacity").

7. In the two product markets most likely to be sustained under the law, basic cable services and premium cable services, the transaction falls within safe harbors described in the 1992 Horizontal Merger Guidelines, which strongly suggests that no enforcement action is warranted.

8. Complaint 33-35.

9. 1995 FCC Report 10.

10. National Cable Television Association, Cable Television Developments 103-17 (Fall 1995) (hereafter "1995 NCTA").

11. "On the Launch Pad," Cable World, April 29, 1996, at 143; see also Cablevision, Jan. 22, 1996, at 54 (98 services announced plans to launch in 1996).

12. National Cable Television Association, Cable Television Developments 6 (Fall 1996) (hereafter "1996 NCTA").

13. "A Who's Who of New Nets," Cablevision, April 15, 1996 (Special Supp.), at 27A-44A (as of March 28, 1996, 163 new networks when regional, pay-per-view and interactive services are included).

14. "The stamina and pocket-depth of backers of new players [networks] still remain key factors for survival. However, distribution [i.e., obtaining carriage on cable systems] is still the name of the game." Cablevision, April 15, 1996 (Special Supp.), at 3A.

15. The History Channel reportedly had one million subscribers at its launch in January 1995, reached 8 million subscribers by the end of the year and was seen in 18 million homes by May 1996. Carter, "For History on Cable, the Time Has Arrived," N.Y. Times, May 20, 1996, at D1. The History Channel now reports more than 26 million subscribers (which accounts for more than 41% of basic cable television households). See 1996 NCTA at 57.

16. Carmody, "The TV Channel," The Washington Post, Aug. 21, 1996, at D12.

17. 1996 NCTA at 70. The percentage figure is derived from the number of subscribers for the network, divided by the number of basic cable households (62.8 million, as estimated by Paul Kagan Associates, Inc.), reported in 1996 NCTA. As a comparison, CNN has 69.9 million subscribers. 1996 NCTA at 39. HBO has 20.8 million subscribers (about one-third of basic cable households). Id. at 56.

18. 1996 NCTA at 58.

19. 1996 NCTA at 77.

20. 1996 NCTA at 86. Cf. the reply of the majority, at 3 ("None of the channels that has entered since 1991 has acquired more than a 1% market share.") (Separate Statement of Chairman Pitofsky, and Commissioners Steiger and Varney, Time Warner Inc., Docket C-3709).

21. This is the kind of competition we would expect to see between cable networks that are substitutes for one another and the kind of competition that does not exist between CNN and HBO.

22. The entry of alternative MVPD technologies may put competitive pressure on cable system operators to expand capacity more quickly. See "The Birth of Networks," Cablevision, April 15, 1996 (Special Supp.), at 8A (cable system operators "don't want DBS and the telcos to pick up the services of tomorrow while they are being overly arrogant about their capacity").

23. CNNfn has 5.7 million subscribers, with 2.4 million on cable and 3.3 million on satellite. 1996 NCTA at 39.

24. 1995 FCC Report 49.

25. DBS Digest, Aug. 22, 1996 (http://www.dbsdish.com/dbsdata.html (Sept. 5, 1996)).

26. DBS Digest, Jan. 20, 1997 (http://www.dbsdish.com/dbsdata.html (Jan. 27, 1997)).

27. See Breznick, "Crowded Skies," Cable World, April 29, 1996 (http://www.mediacentral,com/magazines/CableWorld/News96 1996042913.htm/539128 (Sept. 3, 1996)). National and regional advertising campaigns have helped popularize DBS. E.g., Newsweek, Dec. 2, 1996, at 23 (DISH Network full page ad for digital satellite system and programming); USA Today, Aug. 20, 1996, at 5D (DISH Network full page ad for digital satellite system and programming); N.Y. Times, July 14, 1996, at 23 (AT&T full page ad for digital satellite system, DirecTV and USSB). For a cable system response to DBS competition, see, e.g., The Georgetown Current (Washington, D.C.), Dec. 18, 1996, at 25 (District Cablevision full page ad: "The DISH Network's real charge to hook up your home is out of this world.").

28. Paikert, "Strong Christmas Revives DBS Sales," Multichannel News Digest, Jan. 13, 1997
(http://www. multichannel.com/digest.htm (Jan. 13, 1997)); see also Breznick, "DBS Celebrates the Holidays: Brisk Year End Sales a Boon for DirecTV, EchoStar," Jan. 6, 1997
(http://www.mediacentral.com/Magazines/CableWorld/News96/1997010601.htm (Jan. 6, 1997)).

29. See 1996 NCTA at 14 (ranking the 50 largest MSOs by number of subscribers).

30. Breznick, "Crowded Skies," Cable World, April 29, 1996 (http://www.mediacentral.com/magazines/CableWorld/News96/1996042913.htm/539128 (Sept. 3, 1996)).

31. Id.

32. See Robichaux, "Time Warner Inc. Is Expected To Buy New Set-Top Boxes," Wall Street Journal, Dec. 10, 1996, at B10 (reporting that Time Warner is "look[ing] for new bells and whistles to protect its base of 12 million subscribers against an escalating raid by direct-broadcast-satellite companies"); Robichaux, "Once a Laughingstock, Direct Broadcast TV Gives Cable a Scare," Wall Street Journal, Nov. 7, 1996, at A1. See also Cable World, Dec. 3, 1996 (reporting that "analysts and industry observers agree that cable operators are losing customers to DBS").

33. Katz, "Discovery Goes Digital," Multichannel News Digest, Sept. 3, 1996 ("The new networks . . . will launch Oct. 22 in order to be included in Tele-Communication Inc.'s digital box rollout in Hartford, Conn.") (http://www.multichannel.com/digest.htm (Sept. 5, 1996)).

34. 1995 FCC Report at B-2 (Table 3).

35. MMDS stands for multichannel multipoint distribution service, a type of wireless cable. See 1995 FCC Report 68-85. Industry observers project that MMDS will serve more than 2 million subscribers in 1997 and grow more than 280% between 1995 and 1998. 1995 FCC Report 71.

36. See 1996 FCC Report 67-79.

37. See 1995 FCC Report 116; 1996 FCC Report 93.

38. Pendleton, "Keeping Up With Cable Competition," Cable World, April 29, 1996, at 158.

39. Complaint 38a.

40. Cf. Heublein, Inc., 96 F.T.C. 385, 596-99 (1980) (rejecting a claim of violation based on leveraging).

41. See Whinston, "Tying, Foreclosure, and Exclusion," 80 Am. Econ. Rev. 837, 855-56 (1990) (tying can be exclusionary, but "even in the simple models considered [in the article], which ignore a number of other possible motivations for the practice, the impact of this exclusion on welfare is uncertain. This fact, combined with the difficulty of sorting out the leverage-based instances of tying from other cases, makes the specification of a practical legal standard extremely difficult.").

42. Order V.

43. Although the proposed order would permit any bundling that Time Warner or Turner could have implemented independently before the merger, the reason for this distinction appears unrelated to distinguishing between pro- and anti-competitive bundling.

44. Complaint 38b.

45. Complaint 38c.

46. According to the FCC, "[t]he available evidence suggests that a successful launch of a new mass market national programming network -- that is, the initial subscriber requirement for long-term success -- requires that the new channel be available to at least ten to twenty million households," which amounts to about 16% to 32% of cable households. 1996 FCC Report 135 (footnote omitted). Cf. the reply of the majority, at 7 ("the successful launch of any significant new channel usually requires distribution on MVPDs that cover 40-60% of subscribers") (Separate Statement of Chairman Pitofsky, and Commissioners Steiger and Varney, Time Warner Inc., Docket C-3709).

47. Turner programming would account for only part of TCI's interest in Time Warner.

48. Looking only at cash flow, even if its share of Time Warner were increased to 18%, TCI's interest in the combined Time Warner/Turner would be only slightly greater than TCI's pre-transaction interest in Turner, and it still would amount to only an insignificant fraction of the cash flow generated by TCI's cable operations.

49. Complaint 38b(2).

50. Cable system operators like to keep their subscribers happy, and subscribers do not like to have popular programming cancelled. For example, TCI recently "decided to yield to subscriber cries of 'I Want My MTV and VH1' and restore the channels on cable systems . . . ." Media Central, Jan. 23, 1997 (http://www.mediacentral.com/Magazines/MediaDaily/#08).

51. TCI would have incentives to encourage new programming entry, to the extent that such entry would reduce the "industry average price" referred to in the PSA and thereby reduce the price that TCI would pay under the PSA.

52. See Order IV. There would appear to be even less justification for cancelling the PSA in light of the requirements (Order II & III) that TCI spin off or cap its shareholdings in Time Warner.

53. Order VII.

54. Order VIII.

55. Order IX.

56. See 47 C.F.R. 76.1301(a)-(c).

57. To the extent that the recordkeeping requirements may replicate what is required by the FCC, no additional costs would appear to be imposed by the Order on Time Warner.

58. See 47 C.F.R. 76.1302. The FCC may mandate carriage and impose prices, terms and other conditions of carriage.

59. Order IX.

60. Even in New York City, undoubtedly an important media market, available data indicate that Time Warner apparently serves only about one-quarter of cable households. See Cablevision, May 13, 1996, at 57; April 29, 1996, at 131 (Time Warner has about 1.1 million subscribers in New York, which has about 4.5 million cable households). We do not have data about alternative MVPD subscribers in the New York area.

61. Complaint 38c.

62. 47 U.S.C.A. 548.

63. 47 C.F.R. 76.1000 - 76.1002.

64. Order VI.

65. 47 U.S.C. 548(c)(2)(B)(i)-(iii).

66. Most people outside the FTC and the FCC already confuse the two agencies. Surely we do not want to contribute to this confusion.