Brian graduated from the Wharton School of Business in 1981 with a B.S., played a lot of high-quality squash, and became a low-handicap golfer.38 He could have had a comfortable life sitting on boards, dabbling in business, and playing even more squash and golf. But he had other aspirations. Things moved quickly for him; he joined the Comcast board in 1987 in the wake of the Group W acquisition, and he took part in the Storer Communications negotiations. Meanwhile, the cable industry continued to grow; cable in 1986 was in about 37 million U.S. homes, or 43 percent of all households with a television.39
Brian, like Ralph, had no interest in giving up control of Comcast, even for an enormous amount of money. He could see far greater profits ahead in the digital world. Given some well-timed deal making, a favorable regulatory context, and good financing, more riches were bound to come Comcast's way.
The only other cable guy whose ambition has compared with that of Brian Roberts was John Malone, former CEO of TCI, who brought down Al Gore's ire on the cable industry by his arrogance in the early 1990s. But where Malone was rough, curt, and dismissive, Brian Roberts was smooth and polite. Roberts now owns Malone's cable systems; Comcast bought them from AT&T in 2001. Malone, for his part, is proud of what Roberts has done: “Brian has really matured as a business man, as a financial expert,” he told Bloomberg News in 2010. “I take enormous pride that he's come out of our industry.”40
The company grew even faster after Brian's ascendance in 1990. So had the cable industry generally, following CNN's dramatic coverage of Tiananmen Square and the fall of the Berlin Wall; by 1991, cable was in 60 percent of U.S. homes with televisions.41 Comcast aggregated its Philadelphia cluster of systems through nine separate transactions that pulled together more than 1.4 million additional subscribers;42 it bought MacLean Hunter's systems and the Scripps systems, becoming the third-largest multiple-systems operator by 1995.43
Television viewers weren't the only ones taking notice. Julian Brodsky was proud of Brian Roberts for convincing Bill Gates at a dinner in 1997 that “cable is clearly going to be the way to go,” the best high-speed data route into Americans’ homes, far more promising than the phone companies’ copper lines. Following Brian's direct pitch, Microsoft gave Comcast a major shot in the arm by buying 11 percent of the company for one billion dollars.44 The deal was a typical sledgehammer Brodsky arrangement: “They asked about the Microsoft discount. We explained to them the Comcast premium.” The Microsoft billion went right in the bank as general funds supporting Comcast, and Microsoft received no power in return.45
This was a turning point for Brian Roberts and for the cable industry. Gates saw that with television and the Internet becoming one thing, conduits capable of shipping massive amounts of information were going to be dominant. The cable companies could do this more cheaply than the phone companies because they did not have to dig up the streets and install a second network.
At about this time, “clustering” (“You take Philadelphia, I'll take San Antonio”) became the rage for cable-systems operators. The country had been wired; there was no more room for new cabling in metropolitan areas. As a former cable mogul told me in 2010, “I thought that if cable was going to be on the technology cutting edge; if we were going to compete with the likes of an RBOC [local phone company] or a public utility, we had to own whole markets, not parts of markets.” So the operators, primarily TCI, Time Warner, Comcast, and Cablevision, swapped and clustered systems during the summer of 1997—Leo Hindery, the former president of TCI under Malone, has called it the Summer of Love—so that each company could act within clusters of subscribers, a proceeding that helped cut costs.46
The big companies’ acquisitions of smaller cable operators were also proceeding quickly. In 1996, the top five cable distributors controlled 66 percent of all subscribers: John Malone's TCI alone held a 20 percent market share. By 1999, the cable industry was dominated by just six companies: AT&T (which had bought TCI's systems for $48 billion in 1998), Comcast, Time Warner, Cablevision Systems, Charter, and Adelphia. Then, in December 2001, Brian Roberts scored a major coup by buying AT&T Broadband's cable and Internet divisions (including TCI's former cable systems) in a $72 billion quasi-hostile takeover, propelling Comcast into the top spot as the nation's largest cable company.47 FCC chairman Reed Hundt told the Wall Street Journal in connection with Comcast's bid that “the Roberts don't take ‘no’ for an answer. They repeatedly don't take ‘no’ for an answer.” More than a hundred million households were connected to cable wires by then, and Comcast now served twenty-two million of them, in forty-one states.48
RCN, a small cable provider based in Princeton, New Jersey, that has tried to compete with Comcast over the years, sharply opposed Comcast's acquisition of AT&T's cable systems in 2002, accusing Comcast of using “bullying tactics” in the form of non-compete clauses to prevent about fifteen Philadelphia-area cable-installation contractors from doing business with RCN. According to RCN, contractors were followed and photographed when they were thought to be in contact with or working with RCN, and those photographs were used as a basis to cut off the contractors from doing work with Comcast. Without access to construction and installation contractors, RCN could not offer services. The Philadelphia Business Journal noted in 2002 that Comcast responded by saying that it had taken four years for Comcast to obtain a Philadelphia franchise. “RCN chose to abandon its effort … after a significantly shorter period of time (of about two-and-a-half years).”49
By 2005, Comcast was more than twice the size of Time Warner, its closest national rival—but not its competitor in any major geographic market. The Summer of Love and the swaps and deals since then had ensured that no major cable-systems operator competed with any other. After family-run Adelphia, the nation's sixth-largest cable operator, went into bankruptcy in 2002, its assets were divided in 2006 between Time Warner Cable and Comcast. Comcast gave 500,000 customers to Time Warner in Los Angeles, another 500,000 in Dallas, and 100,000 in Cleveland, while Time Warner gave Comcast 750,000 customers in Houston, 50,000 in Philadelphia, and 200,000 in Minneapolis.50 Smaller cable providers did not share in the pie—the diminishing number of huge companies ran these trades for themselves.
As a result of this unofficial non-compete agreement, although Comcast as of 2010 had only about a 30 percent share of the nationwide market for video services (far ahead of Time Warner's 17 percent share), in the local markets where it operated it had almost no video competition from a cable operator; more important, it was just about the only choice in these markets for video-quality high-speed wired data services.51
Comcast historically has stopped at almost nothing to get strategically located exclusive franchises and subscribers that allow it to further cluster its operations. In 2011, the Third Circuit Court of Appeals allowed a class action to proceed that charged that between 1998 and 2002 Comcast increased its share of Philadelphia subscribers from about 24 percent to about 78 percent through a series of nine swaps of systems with AT&T, Adelphia, and Time Warner; acquisitions of competing cable service providers; denial to RCN of key sports programming owned by Comcast; requiring cable-installation contractors to enter non-compete contracts with Comcast; and persuading potential customers to sign up for long contracts with special discounts and penalty provisions in areas where RCN planned to compete—all with the result that consumers in Philadelphia ended up paying a lot more for pay TV than they would have in a competitive market.52
The family story continued. Ralph Roberts transferred much of his voting stock to Brian in 1998.53 And whenever additional shares are issued, the ratio of votes controlled by the supervoting shares to those controlled by ordinary shareholders is adjusted to maintain Brian Roberts's 33 percent voting power over the company.
Thus, through a well-timed series of acquisitions and swaps, as well as the helping hand of his father, by the February 2010 hearing Brian Roberts found himself at the controls of the nation's largest media company in a thoroughly consolidated marketplace.
Rockefeller would have felt a twinge of jealousy.
But if other cable companies no longer were a threat, what about other technologies? Digital technology now provides the key differentiator on the high-speed Internet access side of Comcast's business, where its future growth and dominance lie: only Verizon's FiOS service, which uses fiber-optic lines (the “one competitor” Brian Roberts referred to when talking to analysts in mid-2011), represented competition with Comcast's DOCSIS 3.0 data services. But in March 2010, Verizon indicated that it was suspending FiOS franchise expansion around the country.54 Cities like Boston and Alexandria, Virginia, that had hoped to get FiOS would be left out in the cold; in the end about 15 percent of Americans (only those in zip codes whose characteristics satisfied Verizon's fairly high target rate of return) would have access to FiOS services.
Verizon stopped expanding for a simple reason. Its existing phone lines are made of twisted copper wire. To build FiOS, it has to install a complete second network—roll in the trucks, rip up the streets, and put in fiber—essentially cannibalizing the existing network on which it still sells DSL service. That's an extraordinarily expensive procedure, and Wall Street hates steep, long-term, up-front capital expenditures. Wall Street wants to see high free cash flow, ample dividends, and frequent buybacks. Comcast, meanwhile, only has to swap out some electronics to shift its existing cable network to DOCSIS 3.0 services. Much, much cheaper. And a death knell to potential competition, even though FiOS services are objectively better because uploads and downloads across its fiber optics are evenly fast. (Comcast faces competition from Verizon's FiOS in less than a fifth of its territory; Cablevision, by contrast, is competing with Comcast in almost two-thirds of its territory.55 Some cable companies are bigger and more important to one another than others; Comcast and Time Warner are strategically aligned in a way that sometimes leaves out Cablevision.)
Another possible competing technology, wireless access, cannot match the speeds cable lines provide. It cannot offer the same capacity unless there are towers connected to fiber lines everywhere—and that's another major up-front expense that the telephone companies don't want to incur. John Malone, among many others, has scoffed at the idea that wireless access could make a dent in cable's dominance: “The threat of wireless broadband taking away high-speed connectivity [market share] is way overblown,” he said in May 2011. “There just is not enough bandwidth on the wireless side to substantially damage cable's unique ability to deliver very high-speed connectivity.”56
Comcast has always been quick to adopt new technology. With Brian Roberts's assistance, in 1988 the cable industry created and funded a technology research center—Cable Television Laboratories, usually called CableLabs—that has played a key role in developing shared technologies and technical advances for the industry. CableLabs is an unsung hero of the cable industry; its founder, Richard Leghorn, predicted back in 1987 (before the birth of the commercial Internet) that the cable industry could become “a multi-channel, multi-format video programmer and publisher utilizing its own interactive, point-to-multipoint optical cable plant,” and this is exactly what happened.57
In 1997 CableLabs came up with standards that could be used to deliver packet-switched, Internet Protocol–based voice services over the cable lines (nicknamed VoIP, for Voice over Internet Protocol), and Comcast quickly adopted the technology, making itself the nation's third-largest telephone company.58 The company embraced the “DOCSIS” (Data over Cable Service Interface Specification) standard developed by CableLabs as soon as it was available, and moved its system to all digital communications in 2008–9. That freed up bandwidth inside its pipe (digital signals can be compressed more efficiently than the old analog signals) while enabling new revenue streams for convertor-box rentals (so that analog sets could continue to be attached to cable wires) and high-definition video. More recently, Comcast was the first to offer CableLabs’ DOCSIS 3.0 protocol, a digital channel-bonding technique that makes possible two-way capacity of Internet Protocol traffic of at least 100 Mbps. By 2011, Comcast had covered some 80 percent of its territory with DOCSIS 3.0, on a substantially faster sche-dule than any other cable distributor, and was selling this high-speed access at high prices.59 Again, only Verizon's FiOS service could hope to compete with the speeds possible with DOCSIS 3.0—and Verizon was backing off.
So Comcast was aiming to stand alone in offering truly high-speed Internet access in each of its markets. This was a sensible move: data access is vastly more profitable than video services—it takes two dollars of video revenues to deliver the same profit as one dollar of Internet access revenues—and Internet access uses only about one-sixtieth of a cable system's total bandwidth.
But technology was only part of Comcast's success. Content was also important.
Brian Roberts knew that Comcast needed to maintain, as long as possible, its power to sell subscribers large bundles of programming that included “must-have” content—particularly live sports. To do that, he needed to make sure that live sports would not be available over the Internet on demand, at attractive prices, without a subscription. The programmers and networks had to be assured that they would make more money selling to cable distributors than directly to online consumers. The Comcast-NBCU deal would stave off the day when programmers revolted; Comcast would become itself a major player in the programming market.
Content was always part of the Comcast story. In the early days, there often was not much programming available to cable operators. In an early system in Sarasota, the community could pull in stations from Tampa through rabbit ears, and local televisions could even get the ABC network from Largo, Florida, about half the time. Roberts, Aaron, and Brodsky were offering Sarasota residents just half a channel of ABC in exchange for a monthly subscription fee—not a very attractive deal.60
As Comcast expanded, it looked for ways to build up content. Dan Aaron thought cable would eventually be bigger than just a reception business—that, in Brodsky's words, “there should be things we can do to bring people other than broadcast television.” Aaron's early attempts to offer content provide fodder for Comcast's autobiography. In Tupelo, one of the first Comcast locations, the trio was operating a three-channel system. Doing an electronic upgrade to five channels by moving amplifiers around within the system would have taken a large investment, and Brodsky worried about wasting money. Aaron said he had a feeling they would be able to use five channels. As Brodsky tells the story,
So what does Dan do with this fourth channel, pioneer that he was, a visionary. … He talks to Telemation out in Salt Lake City and they built him a diorama, and he mounts a videocam, a very cheap video camera on a post that rotated 180 degrees and in the diorama he had a clock, a thermometer, a wind gauge, a rain gauge, a barometer and at the end of it was a place to put in a placard. … The first one was Eat at Joe's Diner, which cost Joe's Diner ten dollars a month, could have been the first local advertising that I knew of, and he played background music behind this thing, and he had [the first] time-weather channel.61
Telecommunications, Inc., later known as TCI, made a similar attempt at local programming in the 1960s. As Mark Robichaux puts it in Cable Cowboy, it was “a TV camera aimed at a news ticker service, another fixed on a thermometer and, occasionally, a camera trained on a goldfish bowl.”62
Aaron's programming was hardly a hit, but Comcast continued to explore the content business. Its logic from the beginning has been that if you don't know whether content is king or distribution is king it is best to spread your bets. You want to be selling something that people can get only from you. When a key partner, the McLean newspaper family of Philadelphia, dropped out, Comcast had to sell its Florida cable franchises (a decision Ralph Roberts and Brodsky regretted for years), but it continued to acquire lucrative Muzak franchises across the country. In the end, Comcast became the largest Muzak franchisee in the nation, selling off its interest to Muzak managers only in 1993.63
Comcast's $20 million 1986 investmen
t in the QVC (Quality, Value, Convenience) home-shopping channel, a hedge-your-bets deal made just after its acquisition of the Group W cable systems, was one of the best moves the company ever made: QVC eventually brought in a third to a half of Comcast's revenue. For little to no cost, through QVC, Comcast was paid by its subscribers to watch content that was presented by advertisers—the sellers—and then paid again when the subscribers phoned in their orders to QVC. In 1992, Barry Diller, former second-in-command at Paramount, took over QVC; when Diller made a $7.2 billion bid in 1994 to merge CBS with QVC, Comcast blocked the sale with a $2.2 billion offer to take over QVC entirely. Comcast and Malone's TCI divided ownership of QVC (with Comcast in control), and Diller promptly left. Comcast's $250 million investment paid off handsomely; to help pay for the AT&T systems in 2001, it sold its QVC shares to Malone for almost $8 billion. Comcast continued its diversification into content by buying a majority interest in E! in 1997, as well as the Golf Channel and Versus, its main sports channel.64
Comcast's more important moves by far have been in sports: in the late 1990s, it leveraged its majority interest in the NHL's Philadelphia Flyers, the NBA's Philadelphia 76ers, and Philadelphia's two major sports arenas into a twenty-four-hour regional channel called SportsNet Philadelphia. Within a few years, Comcast owned exclusive rights in broadcasts by teams and regional sports networks from coast to coast, with dominion over games played in the Bay Area, central California, Chicago, the mid-Atlantic, New England, New York, the Northwest, Houston, and the Washington, D.C.–Baltimore area—ten owned-and-operated Regional Sports Networks in seven of the ten largest television markets, which became the Comcast SportsNet. Because no competing video provider can hope to survive without access to local sports programming, Comcast's refusal to license Comcast SportsNet to RCN in Philadelphia helped keep that potential competitor at bay; it did the same thing to DirecTV and Dish Network.65 Comcast has used SportsNet as a sledgehammer in many contexts.
Captive Audience: The Telecom Industry and Monopoly Power in the New Gilded Age Page 10
