Mr. C. Michael Armstrong

April 23, 2002

Thank you, Mr. Chairman and Members of the Subcommittee, for inviting me here today to testify about the proposed merger between AT&T Broadband and Comcast Corp.

The merger creates a unique opportunity to accelerate the development and widespread deployment of facilities-based local telephony and broadband services. By uniting two companies with remarkably complementary assets, this merger will bring more digital video, data, and voice services and features, to more Americans, more quickly than would be possible without the merger. In short, the merger will benefit American consumers and enhance competition, without violating any FCC or antitrust rule or policy.

I will focus today on two of the principal public interest benefits that will be made possible by the merger. Specifically, I will discuss how the merger will: (1) promote facilities-based local telephone competition; and (2) accelerate the deployment of facilities-based high-speed Internet service (as well as ISP choice), digital video, and other broadband services. I then will explain that the merger will not violate any law or regulation (including any horizontal ownership limit) and will cause no competitive harm in any relevant market.

I. THE MERGER WILL PROMOTE FACILITIES-BASED LOCAL TELEPHONE COMPETITION, PARTICULARLY TO RESIDENTIAL CONSUMERS.

The proposed merger will create substantial benefits in the form of long-awaited local telephone competition, particularly for residential customers. Six years after passage of the Telecommunications Act of 1996 ("1996 Act"), virtually all local exchange traffic -- and particularly residential traffic -- continues to be carried by the incumbent local exchange companies ("ILECs"). While we are making tremendous strides, the ILECs still provide local exchange service to 95% of the customers in their territories. Although our merger obviously cannot be a full solution to producing local competition, it will accelerate the availability of local telephone choice to millions of additional consumers.

The deployment of cable telephony in new markets continues to involve considerable business risks. Cable systems entering the telephony business must underwrite large, upfront investments in new plant and develop and implement order processing, customer care, and other complex support systems, in order to overcome the substantial advantages of incumbent providers. An ILEC's installed infrastructure allows it to serve customers at a lower incremental cost than a new facilities-based entrant and to realize scale efficiencies provided by heavily concentrated customer bases. The magnitude of the risks facing new entrants is underscored by the numerous telecommunications companies that have filed for bankruptcy in recent years.

AT&T Comcast will be on a stronger footing in dealing with these substantial business risks because of the complementary assets and expertise of AT&T Broadband and Comcast and the scale economies created by the merged entity. AT&T Broadband brings to the merged entity extensive experience and expertise in the design, roll-out, provisioning, operations, and marketing of cable telephony in a customer-friendly manner. AT&T Broadband currently offers cable telephony in 16 markets to more than seven million households and has approximately 1.1 million customers. We offer special "Block of Time" plans that allow customers to pay one charge for local, intraLATA toll, and long distance telephone services, with savings in some markets of over 39% when compared to incumbent LEC calling plans. Savings for customers buying more than one line can be even higher.

Our experience has been quite positive. For example, in the Salt Lake City market, we have had consumer take rates of 25% or higher in Ogden, Provo, and Salt Lake. We have had a similar consumer response in several Pittsburgh-area communities, including McKeesport, Aliquippa, East Hills, South Hills, Beaver Falls, Carnegie, McKees Rocks, Ross, and Midland, as well as Bellingham in the Seattle market.

In the past year alone, AT&T Broadband added almost one-half million new cable telephony customers, increasing its customer base by over 100%. As illustrated below, AT&T Broadband is by far the leading provider of cable telephony in the U.S. today:

By contrast, Comcast's cable systems currently provide cable-delivered telephone services on only a very small scale, mostly in cable systems Comcast acquired from third parties which had already launched telephone service.

Fortunately, AT&T Broadband's cable telephony expertise is highly scaleable and can be applied to Comcast's existing cable systems. As a result, AT&T Comcast will be better able to expand the availability of telephony over the Comcast systems more quickly, at less expense, and in a more customer-friendly manner. In light of these synergies, Comcast has announced that, after closing, the merged company intends to begin to deploy telephone service in the Philadelphia and Detroit markets currently served by Comcast, bringing facilities-based local telephone choice to about one million additional homes.

AT&T Broadband's cable telephony expertise will enhance the ability of Comcast's cable systems to offer telephony services in three important respects: technical and operational expertise, back office systems, and marketing.

A. Technical and Operational Expertise.

Comcast will acquire AT&T Broadband's technical and operational expertise in launching and providing cable telephony. AT&T Broadband has already deployed centralized systems to support the design, installation, maintenance, and operation of the complex, two-way hybrid fiber-coaxial systems that support digital voice and data applications and that interconnect with both copper twisted-pair and fiber optic technologies used by the ILECs.

AT&T Broadband has several business units that have developed -- at significant cost -- the technical and operational know-how to provide cable telephony in an efficient and consumer-friendly manner. For example, the AT&T Broadband National Operations team provides support on a wide range of planning, engineering, technical, and operational issues that are faced when deploying complex cable telephony service. AT&T Broadband's Technical Operations Organization has already developed operational performance metrics to ensure quality cable telephony services, effective training of technicians and field fulfillment personnel, and cost-effective investigation and resolution of field performance issues. AT&T Broadband's National Service Assurance Center provides the means for our cable systems to ensure not only that calls are completed successfully and billed correctly, but also that all of the necessary number portability, emergency service, and other databases are managed correctly. And, cable telephony providers must be interconnected to, and coordinate with, ILECs (and other competitive LECs) and interact effectively with a variety of third parties to rate, record, and bill traffic for purposes of reciprocal compensation -- all functions that AT&T Broadband already performs for its systems. Upon closing of the merger, the same organizations at AT&T Broadband that now act as the points of interface for these issues will be available to support cable telephony operations over the Comcast systems. Comcast will also be able to take advantage of certain interconnection agreements that AT&T Broadband has with the incumbent LECs serving Comcast's territories.

B. Back Office Systems.

Comcast will also gain access to AT&T Broadband's existing back office systems that support cable telephony. These systems allow AT&T Broadband to take customer orders and to serve as the point of contact for customer care inquiries. Having in place these "nuts and bolts" back office capabilities and employees is essential to offering local telephone service in competition with incumbent LECs. Not only are AT&T Broadband's back office systems highly robust and efficient, but they employ technologies and processes that will allow AT&T Comcast to use them to support offerings in Comcast territories without incurring substantial additional cost.

The combination with AT&T Broadband will also enhance Comcast's telephone billing capabilities. AT&T Broadband has in place specialized billing software processes, developed over several years, that are sufficiently flexible to handle a service area's unique billing parameters and sufficiently robust to handle substantial increases in volume. These back office billing systems can be used to support telephone entry in Comcast territories at a mass market level.

C. Marketing.

AT&T Broadband's substantial marketing expertise will also help Comcast face the considerable challenge of competing for local telephony customers against formidable incumbents in Comcast's service areas. AT&T Broadband has already conducted primary market research on topics such as pricing and offer design -- benchmarked against the competition -- to assist it in developing successful product offers, programs, and marketing campaigns. And AT&T Broadband has learned a tremendous amount about customer preferences (including the types of marketing that customers like and dislike) as a result of its market experience over the past several years.

Finally, the technical, operational, back office, marketing, and customer care experience AT&T Broadband has gained from its cable-based, circuit-switched telephony operations should be applicable in an IP telephony environment. Comcast and AT&T Broadband have taken leadership roles in developing cable-delivered IP telephony. IP telephony may result in significantly lower roll-out costs and increased flexibility and may also provide a common infrastructure that supports multiple advanced services. AT&T Broadband is committed to the continued development of IP telephony.

I want to emphasize that the synergies detailed above are not merely theoretical. AT&T Broadband's experience in deploying cable telephony after the MediaOne merger has proven that combining new cable assets will result in just such consumer benefits. Indeed, as illustrated below, the number of telephony customers served by AT&T Broadband today is five times greater than the number served by the two separate companies before their merger:

We are confident that AT&T Comcast can build on this successful record, and that the combination of our complementary assets and expertise will further accelerate the pace, broad deployment, and effectiveness of facilities-based local telephone competition, creating substantial benefits for consumers. It is also worth emphasizing, however, that while the promise of facilities-based local telephone competition is a major benefit of this merger, realizing this promise will require a substantial investment of time and money by AT&T Comcast, as well as other cable operators, to deploy the necessary technology and gain the necessary market presence. AT&T Broadband and Comcast are strongly committed to making these investments, but nothing about cable telephony or this merger diminishes the independent need to facilitate the other means of creating local telephone competition that Congress specified in the 1996 Act.

II. THE MERGER ALSO WILL ENHANCE THE DEPLOYMENT OF FACILITIES-BASED HIGH-SPEED INTERNET SERVICE, DIGITAL VIDEO, AND OTHER BROADBAND SERVICES, PARTICULARLY TO RESIDENTIAL CUSTOMERS.

Comcast and AT&T Broadband both offer high-speed Internet services, serving a combined 2.5 million customers. By combining complementary assets and experience and creating economies of scale and scope, the merger will allow us to more efficiently develop and deploy new, innovative broadband applications over the AT&T and Comcast cable facilities, providing substantial benefits to consumers and stimulating productivity gains and growth in the U.S. economy. Moreover, AT&T Comcast's efforts will provide a competitive spur to other entities, including incumbent telephone companies, nationwide direct broadcast satellite ("DBS") providers, and others. The existence of a strong and credible broadband alternative on cable has already generated competitive responses in the form of accelerated DSL deployment by incumbent telephone companies, and this proposed merger will further advance this trend.

A. Capital Improvements And Other Merger Benefits.

AT&T Broadband's merger with Comcast will enhance significantly its access to the capital required to underwrite an aggressive plan for deploying new broadband services such as HDTV, video-on-demand, and expanded Internet offerings to residential consumers over existing AT&T Broadband systems. It is estimated that AT&T Broadband and Comcast collectively will spend approximately $5.5 billion in 2002 on capital expenditure items and, following the merger, AT&T Comcast will continue to make substantial capital expenditures. AT&T Comcast should be able to obtain lower prices for many of these capital items as a result of the increased scale of its purchases.

More generally, the scale economies created by the merger will foster more efficient use of infrastructure (e.g., by allowing for more efficient use of call centers), and provisioning, repair, and maintenance (e.g., by providing local/regional scale to support efficient, centralized truck rolls). The merger will also provide national scale that will allow the merged firm to defray more efficiently the enormous research, development, and testing costs associated with new services and features. This increased scale is particularly important to accelerating the development and testing of new interactive TV services, voice-enhanced data services, home networking and security, and other new, and as yet untested, broadband services.

B. ISP Choice.

I want to address in particular the issue of ISP choice, which I know is of interest to members of the Subcommittee. AT&T Broadband and Comcast share a strong commitment to providing multiple ISP access on their broadband networks. Indeed, both companies have ample market incentives to make commercially reasonable, customer-friendly arrangements with unaffiliated ISPs in order to maximize the attractiveness of their Internet offerings to customers and potential customers. Given the need to compete with DSL and other comparable high-speed data providers, AT&T Comcast will continue to have such incentives to offer its customers a choice of ISPs post-merger.

AT&T Broadband has made real progress in its efforts to provide ISP choice. In particular, in 2000 and 2001, AT&T Broadband conducted a $20 million six-month trial in Boulder with four ISPs (Excite@Home, EarthLink, Juno, and WorldNet) which provided significant experience on the technical and operational requirements needed to support a multiple ISP environment. The Boulder trial enabled us to test our technical infrastructure and assess our key business assumptions. For example, we learned a great deal in Boulder about routing architecture, consumer self-help and diagnostic tools, business-to-business interfaces, and how consumers value ease of ISP selection.

The lessons learned in Boulder will be valuable as we roll out ISP choice in Boston and Seattle this year. The first step in commencing implementation of ISP choice is the agreement we recently entered into with an unaffiliated ISP, EarthLink. Under the agreement, EarthLink will offer high-speed cable Internet service via AT&T Broadband's network. Initially, EarthLink will launch service in Seattle followed by greater Boston. The companies anticipate launching EarthLink's service in additional AT&T Broadband markets in 2003. The planning discussions are underway with EarthLink regarding, for example, the interconnection of our two networks, the deployment of efficient operational interfaces between the companies, and the schedule for rolling out the service in particular communities. In addition, we are actively reaching out to a number of regional ISPs in an effort to provide our customers with even greater ISP choice. We are also migrating to a more robust high-speed data provisioning system across all of our markets to provide more effective support for our high-speed data customers, both those we serve directly and those receiving ISP services from unaffiliated ISPs.

Comcast also has conducted trials of ISP choice, which have provided it with valuable experience and insight into how best to roll out this new offering. The merger will enable our two companies to share the unique experiences we have had and the important knowledge we have gained in our respective ISP choice trials. This sharing of "best practices" will enable AT&T Comcast to overcome the substantial technical and operational complexities involved in implementing ISP choice, so that this important new service offering can be rolled out on a more efficient and widespread basis than the two companies could hope to achieve independently.

III. THE MERGER WILL NOT RESULT IN ANY VIOLATION OF THE COMMUNICATIONS ACT OR THE FCC'S RULES.

The proposed merger will not result in the violation of any provisions of the Communications Act, other applicable statutes, or the FCC's rules. In particular, I will address today the reasons why the proposed merger will not violate the FCC's cable horizontal ownership limit.

In October 1999, the FCC adopted a rule prohibiting a cable operator from having an attributable interest in cable systems that account for more than 30% of all MVPD subscribers nationwide. However, as the Subcommittee knows, last year the D.C. Circuit in Time Warner II reversed the 30% limit and remanded the rule to the FCC for further consideration. The FCC has initiated a proceeding to consider the cable horizontal issue in light of Time Warner II. The FCC has not yet reached a decision in that proceeding.

Of course, AT&T Comcast will take all steps necessary to comply with any new cable horizontal ownership limit that the FCC adopts. But, it is especially noteworthy that the merger would not violate even the 30% limit that was set aside in Time Warner II. AT&T Comcast will serve approximately 27.3 million subscribers, or about 29.7% of the nation's MVPD subscribers. Because this percentage is below the horizontal limit in effect before the ruling in Time Warner II, there can be no reasonable basis for concern that the proposed merger would violate any horizontal ownership rule.

This calculation does not include the subscribers served by the cable systems owned by Time Warner Entertainment Co., L.P. ("TWE"). When AT&T merged with MediaOne, AT&T Broadband acquired a minority, limited partnership interest representing about 25% of TWE. Subsidiaries of AOL Time Warner hold the remaining majority interest in TWE. Under the terms of the TWE Limited Partnership Agreement ("LPA"), AT&T Broadband has no role in or ability to influence the management or operations of TWE, nor does it have the right to communicate with TWE, or AOL Time Warner, the general partner of TWE, on matters pertaining to the day-to-day operations of TWE. The TWE Cable Management Committee (all members of which are appointed by and from AOL Time Warner) has full discretion and final authority over TWE's cable operations. All of MediaOne's rights with regard to the TWE Cable Management Committee were terminated before AT&T merged with MediaOne and acquired the TWE interest. Thus, we believe that the interest in TWE would qualify for insulation from attribution to AT&T Broadband under the FCC's attribution rules today, and to AT&T Comcast post-merger.

In any event, AT&T Broadband and Comcast do not view the TWE interest as a long-term investment and are firmly committed to divesting the interest for a fair price as quickly and efficiently as possible. In fact, the process of attempting to sell the TWE interest is already underway. AT&T Broadband has pursued with AOL Time Warner various options for the sale of its TWE interest to AOL Time Warner in an efficient and expeditious manner. AT&T Broadband also is pursuing the sale of its TWE interest via a public offering pursuant to the registration rights provisions of the TWE LPA. Although AT&T Broadband is pursuing diligently all possible avenues to dispose of its TWE interest, the simple fact is that our ability to sell promptly the interest at a fair price depends almost entirely on the cooperation of AOL Time Warner and its subsidiaries, who do not have the same interests or incentives as AT&T Broadband in this regard.

IV. THE MERGER WILL HAVE NO ANTICOMPETITIVE EFFECTS IN ANY RELEVANT MARKET.

AT&T Broadband and Comcast provide services to consumers in different local markets and therefore the proposed merger will have no measurable impact on horizontal concentration in any relevant market. Additionally, the combined entity will not have either the ability or incentive to exercise buyer market power in any relevant market.

A. Multichannel Video Programming Distribution

The merger will not have any adverse effect on competition in the business of multichannel video programming distribution. AT&T Broadband and Comcast cable systems reach different residences and businesses and compete in different local markets, so the proposed merger will not reduce actual competition in any relevant local distribution market.

Further, the merged company will face intense competition from DBS providers. DirecTV and EchoStar, two DBS providers, distribute video programming throughout the United States and compete directly in all local markets served by AT&T Broadband, Comcast, and other cable operators. In less than ten years, DBS has grown from serving no multichannel video subscribers to serving nearly 18 million subscribers, over 19% of all MVPD subscribers nationwide. Last year alone, DBS grew twelve times faster than cable, with both DirecTV and EchoStar experiencing tremendous subscriber growth. Indeed, four out of five new customers now are choosing DBS over cable, and almost one-half of existing DBS subscribers are former cable customers. In addition, AT&T Comcast will face retail competition in many localities from cable "overbuilders" (including RCN and Knology), electric utilities (including Starpower and Seren), and MMDS and SMATV providers.

B. Video Programming Production And Packaging

The merger will not adversely affect competition in the production and packaging of video programming for sale to MVPDs. As explained below, AT&T Comcast will have neither: (1) "seller power" that would allow it to raise prices for, or discriminate in the distribution of, video programming; nor (2) "buyer power" that would allow it to insist on anticompetitive terms and conditions for programming that it purchases from others.

As the members of the Subcommittee know, a critical element of any competition analysis is the definition of the relevant geographic market. The relevant geographic market for the purchase and sale of video programming is quite broad and, for many types of programming, international in scope. There are no significant limitations on transporting programming and, as a result, video programming can be sent to virtually any distribution outlet in the world for roughly equivalent costs. Moreover, the only limiting factor on the international distribution of U.S.-produced content is whether there is foreign demand for that content. Foreign demand is quite strong; international sales now account for a very substantial portion of video programmers' businesses. By way of example, MTV reaches more than 340 million households in 140 countries.

1. Seller Power.

The merger will not reduce competition or create market power in the sale of video programming by AT&T Comcast. The combined company will have only very modest programming interests and no enhanced ability to control the pricing of video programming to MVPDs. AT&T Comcast will have ownership interests in a total of 24 video programming networks, or 6.4% of the current 374 programming networks. This very limited set of post-merger interests (many of which are minority interests) presents no concentration problem or threat of competitive harm, particularly when viewed against the backdrop of the highly competitive video programming marketplace, and the far more significant program holdings of other media entities.

2. Buyer Power.
AT&T Broadband and Comcast are, of course, buyers of video programming. There are two theories of competitive harm that could be raised by an assertion that the merger creates buyer "market power": first, that the merger would reduce horizontal competition in the purchasing of programming and thereby create buyer monopsony power; and second, that the merger would increase the incentive and ability of the merged firm to engage in distribution foreclosure in the purchase of video programming from video programming producers. As explained below, the merger will not create any anticompetitive consequences under either of these theories.

Monopsony Power. Traditional monopsony theory holds that a firm buying a sufficiently high percentage of the output of a group of sellers may have the ability to set unilaterally the price it pays for goods or services produced by the sellers. This theory has no applicability in the present case for several reasons.

First, companies can only exercise monopsony power over goods that, when sold to one buyer, cannot be sold to another buyer. However, video programming can be consumed by an unlimited number of buyers. This negates the normal intuition that a very large purchaser may be able to exercise monopsony power over sellers.

Second, a programmer's distribution alternatives will largely determine whether the programmer is vulnerable to an attempt to exercise monopsony power. As noted, AT&T Comcast will account for less than 30% of total purchases, and that is not remotely enough to give it buyer market power, since the alternative distribution channels and revenue sources available to video programmers are significant (i.e., over 70% of the U.S. distribution market, as well as substantial international distribution markets).

Third, a cable operator's appetite for quality programming is driven by consumer demand and retail competition. As a cable operator gets bigger, there is no change in its incentives to buy the programming that is likely to produce the greatest number of viewers relative to the cost of the programming. For these reasons, and given the intense competition from DBS and others at the retail level, even if a cable operator was large enough to exercise monopsony power -- and AT&T Comcast clearly will not be large enough -- it would choose the same quantity and quality of programming as a competitive, "non-monopsonist" purchaser.

Distribution Foreclosure. Nor could the combination of AT&T Broadband and Comcast trigger any foreclosure concerns. Such concerns could arise if the merged entity would have sufficient market power in the distribution of programming such that it would have the incentive and ability to foreclose access to its cable systems by refusing to buy programming that viewers desire from unaffiliated program packagers or producers.

As an initial matter, AT&T Comcast will not have the incentive to foreclose unaffiliated video program packagers or producers because AT&T Comcast will have only modest video programming interests, and the damage caused by distribution foreclosure to its core cable distribution business could be substantial. It is clear that consumers view DBS and cable as substitutes and have demonstrated that they would readily switch from cable to DBS if they viewed AT&T Comcast's offering as inferior. As a result, any action by AT&T Comcast that degraded the quality of its programming -- by foreclosing competitively priced unaffiliated programming that consumers want -- would cause AT&T Comcast to lose customers to DBS or other distributors. Moreover, given the modest programming interests of AT&T Comcast, the potential benefits of such a strategy would be essentially non-existent.

AT&T Comcast will also have no ability to foreclose. In order to engage in foreclosure successfully, AT&T Comcast would have to control such a substantial percentage of all distribution channels to which rival video programmers could turn as to be able to drive them out of business or substantially raise their costs. However, even focusing solely on U.S. MVPD distribution channels, AT&T Comcast will purchase programming for systems that serve less than 30% of subscribers. Video programmers, of course, understand marketplace dynamics and would recognize that, even without AT&T Comcast, they still have access to more than 70% of U.S. MVPD subscribers.

C. Set-Top Boxes, Cable Modems, And Other MVPD Consumer Equipment

Likewise, the merger will have no adverse effect on any equipment market. The relevant geographic market for MVPD customer equipment is global. Set-top boxes, modems, and other navigation devices are purchased by MVPDs and MVPD customers in the U.S., as well as by MVPDs, consumers, and other buyers worldwide. With fewer than 30 million subscribers, AT&T Comcast will represent less than 10% of the 317 million worldwide cable and DBS subscribers. Accordingly, AT&T Comcast cannot be considered to have the power to do anything to harm the production or supply of such equipment.

Even if one were to focus on the domestic equipment market, AT&T Comcast would purchase equipment for less than 30% of U.S. multichannel video customers -- a level too low to raise any concerns about anticompetitive harm.

Moreover, given the ubiquitous availability of DBS and DSL alternatives, AT&T Comcast will have no incentive to exercise market power against set-top box or modem manufacturers. Any action by a cable operator that has the effect of restricting the supply of high-quality equipment that enables consumers to access operator-provided services would cause the operator to lose cable customers to the DBS competitors and Internet customers to DSL or other competing providers. Thus, AT&T Comcast will be compelled by market forces to deal fairly with equipment manufacturers and to ensure that it and its customers have access to the best quality state-of-the-art equipment at the best possible price.

D. Interactive TV Services

The merger will not harm consumers or competition with respect to the provision of interactive TV services. As with MVPD and other services discussed above, AT&T Broadband and Comcast do not compete with each other in the provision of interactive TV services, so the merger will have no adverse effect on competition in this business. Moreover, the interactive TV business is in the very early stages of its development and many questions remain about the technology and consumers' demand for it. Indeed, there has not yet emerged a clear definition of what interactive TV is or how the market should be defined. So, it is entirely premature to even speculate on how the merger might affect this business.

Moreover, notwithstanding the nascent stage of the interactive TV business, numerous companies (including DBS providers) are investing substantial resources in developing, deploying, and distributing interactive TV content, equipment, and services. In this highly dynamic environment, AT&T Comcast will have no market power in the provision of interactive TV services.

V. CONCLUSION

Thank you for this opportunity to testify. I would be pleased to answer any questions you might have about the proposed merger.