The first edition of this book was published in 1991. On page 3 we wrote:

Five years later, Congress passed the Telecommunications Act of 1996. What was "beginning to emerge" in 1991 is now well under way; this second edition marks the beginning of the middle. Regulatory means and ends have shifted fundamentally. Now, the unqualified regulatory objective for all levels of this industry is competition. If anything, however, the sheer volume of regulation has increased dramatically, as regulators have sought to manage all the particulars of the transition. Our third edition will describe, we hope, the beginning of the end, when competition will flourish in a truly deregulated environment.(1) The road, however, has been a long and winding one.

For most of this century, telephony was viewed as a natural monopoly.(2) The high cost of fixed plant, the steadily declining average cost of service, and the need for all customers to interconnect with one another, made it seem both sensible and inevitable to have a single, monopoly provider. The Bell System was born on that premise, and with it an elaborate body of regulation designed to reconcile private monopoly with the public good.

The old regulatory paradigm had three basic pillars. First, the protected franchise: would-be competitors were barred from competing or even interconnecting with the enfranchised carrier; natural monopoly thus became a self-fulfilling prophecy. Second, the quarantine: the monopolist was restricted to its regulated sphere and barred from exporting its expertise (and the corrosive influence of its monopoly) into adjacent competitive markets. Third, cradle-to-grave regulation: prices, terms, and conditions of the monopolist's services had to be sold to regulators before they could be sold to customers.

The old regulatory paradigm served the country adequately for over half a century, permitting the Bell System to deploy the best, most technologically sophisticated telephone system in the world. Bell also deployed Bell Labs and funded it with the enormous profits of the Bell System. And Bell Labs, apparently unfamiliar with regulatory paradigms, developed new technologies that made competition in telephony inevitable.

Today, the two overarching technological trends in the industry are fragmentation and convergence. There are, on the one hand, more switches, more lines, more networks, many more levels of interconnection – the old integrated, centralized media are being fragmented into many smaller, more autonomous, parts. On the other hand, interconnections are proliferating as never before: direct interconnections between competing, wireline phone companies, and also interconnections between different media that never used to compete at all.(3) Television is leaving the air in favor of wires; the telephone is leaving the wires in favor of the air. Copper and coax, wired and wireless, terrestrial and satellite: digital data networks are rapidly emerging as the new, universal, universally-interconnected standard for the transmission of everything – voice, data, video, the lot.

These industry-transforming developments are not, of course, unconditionally welcome on all sides of all the industries involved; "every invention," as Joel Mokyr has noted, "is born into an uncongenial society, has few friends and many enemies."(4) Precisely the same technological developments support new competition by the phone company and against it, and incumbents on opposite sides of traditional regulatory fences have quite different views about which kind of competition should come first. Thus, television broadcasters want to contain phone and cable companies, just as radio broadcasters once hoped to contain television, and just as newspapers once hoped to contain radio. Cable companies want to contain phone companies and vice versa. Providers of Internet services would prefer to contain both. Landline phone companies see a real threat in wireless services, which provide far more convenience; wireless carriers see a real threat in the landline system, which still provides the dominant hub for ubiquitous interconnection. Much of the time, for many of the protagonists, the paramount regulatory objective is to preserve the status quo, and so to fend off the instability of untrammeled new competition.

The dynamics of the new technology have forced regulatory change regardless. The protected franchise has disappeared. The quarantine was eliminated in its entirety by the 1996 Act. Pervasive regulation of prices and terms of service is now rapidly giving way to prices and terms set by competition. These changes have in turn wrought fundamental changes in the nature, purpose and complexity of regulation.

Under the old regulatory model, for example, there was no need for interconnection or equal access standards – the only interconnection was between customers and their monopoly carrier.(5) We may likewise anticipate a day when interconnection regulations will once again become unnecessary: competitive forces will balance out, and interface standards will be set through market forces, voluntary agreements, and evolutionary consensus as they are in other industries, most notably the computer industry. Today, however, while dominant carriers still mingle with fledgling competitors, interfaces are regulated more pervasively than ever before. And day by day the interfaces multiply, as equipment providers, long-distance carriers, Internet providers, wireless providers, and – more recently and importantly – competitive providers of local exchange service – all clamor for new forms of access, steadily blurring the lines between customers and providers.

Price regulation too has become vastly more complicated.(6) With a single monopoly provider, the regulator's task was relatively simple: ensure that the monopolist enjoyed no more than a "fair" rate of return.(7) Within that overall stricture, regulators were free to shuffle costs and prices – which they did with enthusiasm, mostly in pursuit of universal service.(8) Through the "separations" process – the division of costs between interstate and intrastate jurisdictions – the FCC and state regulators heavily subsidized local connections by loading costs on to interstate services.(9) This helped attract long distance competition. And with competition came today's dismaying spectacle of regulators vowing to embrace competition, maintain subsidy, and protect fledgling competitors all at the same time.

Most complicated of all have been relations among regulators themselves.(10) Telecommunications began as a local industry heavily regulated by local authorities. For most of this century, local exchange revenues dwarfed those from long distance service; regulation was accordingly centered in local commissions, and the FCC’s role was correspondingly modest. Hewing to tradition, the Bell divestiture decree divided markets along strictly geographic lines. But the whole point of telephony is to erase geography – so the better service gets, the more difficult it is to regulate through a hodge podge of independent, uncoordinated, geographically-defined authorities. In the final analysis, local, introverted regulation is irreconcilable with a network that must function outward, nationally and beyond. Thus, the major trend today – pushed by the FCC, though resisted by many States – is toward a steady contraction of local power.(11)

For now, we are at the high water mark of regulation, with major components of both the old and the new regulatory regime still in place. The shift from managing monopoly to managing competition was accelerating rapidly in 1991, when our first edition appeared, though we described it then as "one of the great if unheralded regulatory initiatives of our day." The heralds finally made it to Congress, and what emerged was the 1996 Act. Our ability to compete in the new global markets will turn, in significant part, on how – and how quickly – critical disputes are resolved and its objectives are attained.

The history of telephony begins, as it ends, with the working ends, the parts on customer premises that are actually seen and used. Most of the modern telephone network – a gigantic spider web of wires and switches, satellites, radio towers, and trans-oceanic cables circling the globe – is completely invisible to end users. But it was the telephone itself that spawned all the rest. And, as we shall see, a later generation of customer premises equipment impelled the revolutionary changes in marketplace and law that we see all around us today.

The idea of "telephony" (from the Greek, "far speaking") tantalized inventors since time immemorial, but nineteenth century inventors were the first seriously to explore the possibility. In 1831, Michael Faraday demonstrated that a piece of iron vibrating in a magnetic field produces electrical pulses. Soon thereafter, in 1835, Samuel Morse invented the essential elements of a telegraph.(12) A quarter century later, a German researcher, Philipp Reis, developed a device that used a membrane, electrodes, and an electric current to transmit pitch.(13)

Alexander Graham Bell produced his first rough plans for a "harmonic telegraph" in 1872. Within a few years he had developed the "phonautograph," which translated sounds into visible markings.(14) By 1875, he was in a neck-and-neck race with Elisha Gray,(15) co-founder of the Western Electric Company, to create a workable telephone. In June of that year, Bell discovered the basic principles of the electromagnetic microphone and speaker, the key elements of a telephone. And on February 14, 1876, Bell filed his historic patent application at the U.S. Patent Office. Just a few hours later, in the same office, Gray filed a notice of a pending application for a patent on a similar device. The U.S. Supreme Court ultimately ruled, however, that the invention belonged to Bell.(16) Bell did undoubtedly win the race to build an actual, working telephone. The famous words – "Mr. Watson, come here, I want you" – were transmitted from Bell to his assistant in a nearby room on March 10, 1876.

First there was the box telephone – an iron diaphragm encased in a wooden block. Bell’s assistant Thomas Watson then developed the "Thumper," the first telephone signaling device, subsequently superseded by electric "call bells" and, still later, by "magneto call bells." By 1878, Bell was installing two "telephones" at each station – one for talking, the other for listening, so that the user would not constantly have to shift the unit between mouth and ear.(17) "French" handset telephones (in which the handset contains both transmitter and receiver) arrived in the United States in 1927.

From the beginning, Bell confidently predicted that some day "a telephone in every house would be considered indispensable . . . ."(18) Others were much more skeptical. Western Union, the (then) giant telegraph company, flatly rejected an early offer to purchase Bell's patents.(19) Smaller competitors knew better; imitations of Bell's invention soon inundated the market. The Bell Company brought over 600 patent infringement suits in the two decades after the telephone’s invention.(20)

Telephones created an immediate need for telephone wires to connect them. Before long, telephone poles and wires proliferated in city streets, so much so that the New York State Legislature required all telephone wires in major cities to be routed underground.(21)

Initially, telephones were linked one-to-one.(22) It took until 1878, two years after the invention of the telephone itself, for the budding new telephone companies to grasp the necessity of a telephone "exchange." The telephone exchange – a simple switchboard, at first – radically increased a telephone's utility, by enabling each phone to reach any other phone connected to the same exchange. Demand for service increased dramatically. By the mid 1880s, multiple unit switching systems allowed operators to work in banks, each serving the same array of telephones. The first automatic switching system was patented in 1889; a series of pulses was used to raise and then rotate a shaft to make the appropriate contact.(23) A patent for dialing devices was granted in 1898. Newark, New Jersey boasted the first semi-automatic switching system in 1914. Omaha, in 1921, housed the first fully automated system.

The advent of the telephone exchange gave rise to the first rumblings about the need for – or inevitability of – monopoly provision of telephone service.(24) A central switchboard and the wires leading to it represented a large, fixed capital investment; the costs of an exchange seemed to decline rapidly as more subscribers were added. By fragmenting the market, competition appeared to drive up costs and nullify the key advantage of a central exchange, which was to connect everyone to everyone else.

Local exchanges also raised the possibility of "interexchange" connections over longer distances. While few individual owners of telephones could afford to run lines over great distances, the owners of exchanges clearly could.

Initially, however, there was no technology to support such service. Even short telephone connections were of very poor quality. Thunderstorms filled lines with static; switchboards caused all manner of cross-talk. Signal quality deteriorated rapidly over "long" distances – distances that, in Watson's view, spanned more than about two miles.(25) Until the turn of the century, most long distance service was expensive and almost unusable.

In the late 1890s, Bell engineer George Campbell discovered that "loading the lines" – concentrating inductance in coils along the lines – could greatly improve transmission quality.(26) The first long distance cable using Campbell's invention was demonstrated successfully in September 1899.(27) The subsequent struggle to acquire and control the loading patent would be "the most hotly contested and economically significant . . . since the original Bell patents of 1876."(28)

Then in 1906, Lee DeForest developed the audion – the first vacuum tube electronic amplifier.(29) Audions could be used to make a "repeater," with the capacity to amplify an electric signal. Placed at suitable intervals along a telephone wire, repeaters could sustain the strength of a signal over almost any distance. For the next four decades, the vacuum tube amplifier would be the key component of all telecommunications, broadcast, and (eventually) the first-generation electronic computers. The Bell System purchased DeForest's patent in 1913.(30)

Thus, the key technologies needed for telephone service of the kind we know were developed in the four decades between 1876 and 1913. The invention of the telephone led to the invention of the local exchange; the invention of the local exchange led to the development of the technology needed for interexchange service. Each major restructuring of the network in the eight decades since has been triggered by breakthrough developments in one or another of these three principal tiers of the industry.

By 1895, when Bell’s major patents on the telephone had just expired, the new wonder of the technological world was not communication with wire, but communication without it. That year, Guglielmo ("G.M.") Marconi produced the first practical radio. The following year Britain granted Marconi his first patent on a radio – still known, at the time, for precisely what it did not require – wires. It was the "wireless telegraph," or just the "wireless."(31)

Radio looked, at first, like a major new competitive threat to telephone service. Marconi himself had conceived of it as a messaging system, principally for ships.(32) In short order, the wireless telegraph became a wireless telephone – radio communication of human speech first occurred on Christmas Eve in 1906, between Brant Rock, Mass., and ships in the Atlantic Ocean. And Lee DeForest's vacuum tube amplifier, the "audion," improved radio transmission as dramatically as it improved telephony. By all logic, the development of radio and the electronic amplifier(33) should have ushered in an era of fierce competition between the two infant industries of radio and telephone. Bell itself recognized the threat: in 1909, the company's chief engineer sought research funds to put the company "in a position of control with regard to the art of wireless telephony, should it turn out to be a factor of importance."(34)

Competition did indeed start out on a reasonably promising track. Fearing that Britain would become the world hub for radio, just as it had already become for the undersea cable network, the U.S. government backed the formation of a strong radio manufacturing company, the Radio Corporation of America (RCA).

With the early technologies, however, radio’s great advantage was not in point-to-point messaging, but in point-to-everywhere broadcast. David Sarnoff, who started as a telegraph operator for the Marconi Wireless Telegraph Company of America, was perhaps the first to recognize this. In 1916 he submitted his idea for a "radio music box" to the management of Marconi. Westinghouse inaugurated the nation's first true radio station, KDKA in Pittsburgh, in 1920. Bell put its own station, WEAF, on the air in New York City in 1922.

RCA and Westinghouse, the two early leaders in radio, planned to operate radio stations at a loss to promote the profitable sale of radios. The idea of giving away the broadcast to sell the radios worked for a short while, but was doomed to fail in the long run: once patents expired, anyone could build a radio. Bell understood this from the beginning and adopted quite a different approach. The only sustainable revenues in broadcast, the company recognized, would come from the sale of air time. So Bell developed a plan to build transmitters and sell airtime to all comers – entire chunks of airtime, which buyers could fill with a mix of program and advertisement as they saw fit. Bell decided, in short, to run radio just like its phone business, as a common carrier. The company earned solid profits in the first few years of its operations, while RCA was losing money.

When commercial telephone service began in the United States in 1877, Bell held all the essential patents. The company thus enjoyed a complete and legal monopoly. By 1894, however, the essential patents had either expired or been narrowly construed by the courts. Thousands of independent telephone companies took advantage of the opportunities offered by the newly available technology. By 1902, 451 out of the 1002 cities with phone service had two or more companies providing it. By 1907, when a phone census was taken, the "independents" owned nearly as many phone stations as Bell.(35) Average rates were cut in half, and Bell's average return on investment had fallen over 80 percent.

At about this time, Theodore Vail, a brilliant administrator, took charge of the declining Bell empire. The Bell System had lost its original patents on the telephone, but it had acquired new ones, through which it now enjoyed a critical edge at the opposite end of the network, in providing quality long distance connections. Vail believed passionately in universal service, to be supplied by one company – his own. Bell therefore offered its superior long distance service to its own local affiliates, but not to others. The company likewise refused to sell equipment or to provide interconnection even to those independents that did not directly compete with it. In rapid succession, phone companies not affiliated with the Bell System either folded or were acquired.(36) In 1876, Western Union – then the largest corporation in the world – had declined to purchase Bell's patent. In 1910, it was Vail's turn to buy Western Union – for the then astronomical sum of $30 million.

By 1915, when Bell had established transcontinental service, Vail could describe the Bell System as "an ever-living organism" that possessed "one of the largest laboratories of the application of science to industrial development in the world." Vail's slogan, announced in 1908, had an almost Orwellian ring to it: One Policy, One System, Universal Service.

It is at this point in the history of telecommunications that regulation could have forcefully intervened to prevent the rise of monopoly service by forbidding boycotts and requiring carriers to interconnect with one another. Multiple carriers might well have coexisted and competed under such a regime without a balkanized network in which telephone owners could only talk with other customers of the same carrier. The fundamental question arose early on: Must carriers carry other carriers? At first, the answer was no; the direct result was the creation of the largest, most powerful telecommunications monopoly the world has ever known.(37) From the 1960s until 1996, the answer was a tentative yes, and the result was the break up of that same monopoly and a fundamental, irreversible shift toward competition.(38) With the passage of the 1996 Telecom Act, the answer is no longer tentative at all. The Act=s core provisions direct incumbent local phone companies to Aunbundle@ their networks and offer competitors interconnection at low costs.(39)

The earliest common carriers were created when the Crown awarded an exclusive monopoly to a company operating such things as a ferry boat, a wharf – or, for a time, a printing press. The English common law gradually developed rules that both contained monopolists’ excesses and defended their monopolies.(40) Crown monopolies were required to charge only "reasonable and nondiscriminatory" rates, provide adequate service, and accept all customers on the same terms, without discrimination. In time, these principles came to extend to any firm "affected with a public interest" that held itself open to the general public and purported to serve all comers. In return, common carriers enjoyed important legal privileges, most particularly limits on their liabilities – limits appropriate to a business that could not legally discriminate among those it chose to serve. America inherited these core principles from England.

Telegraph and telephone companies plainly fit the "common carrier" model.(41) Almost from the beginning, they were expected to serve all comers and charge similar rates for similar services. They could not easily be sued for libelous utterances transmitted over their systems, or for many other expensive or hurtful consequences that might ensue when transmissions were garbled or lost. They soon came to be viewed as paradigm "common carriers," so common, so ubiquitous, so routinized that one could scarcely imagine them operating in any other way. Except, as it turned out, when a would-be "customer" happened to be another carrier.

The problem had been faced – and resolved correctly – half-a-century before the birth of telephony, in legislation for telegraphy. The Post Roads Act of 1866, which allowed telegraph companies to run their lines freely along post roads and to fell publicly-owned trees for telegraph poles, had required the telegraph companies, in turn, to interconnect and accept each other's traffic.(42) If telephone companies had been held to similar obligations early on, local exchanges might have remained competitive by differentiating their price and services. All might have offered complete service through interconnection with other companies, including – during the lifetime of the Bell patents on superior long distance technology – the best available long distance connections. But somehow, legislators, regulators, and the courts drifted toward a narrow understanding of a common carrier's obligations to carry its competitors' traffic.

The problems first appeared in the Express Packages cases, decided by the Supreme Court in 1885. Four major express companies, the grandparents of today's overnight courier services, had worked out a cozy territorial division of the business. They didn't do much hauling themselves. Instead, they reserved an entire railway car, or perhaps half a car, paid wholesale rates for the space, and used their own agents to load, unload, and supervise packages consigned to their care. In 1880 one of the major railroads abruptly announced it would no longer sell space on those terms; express companies would be treated like ordinary retail customers. Faced with immediate financial ruin, the express companies rushed to court. Their arrangements with the railroads (they argued), though originally simple matters of contract, had become matters of common "law and usage." The U.S. Supreme Court disagreed.(43)

Perhaps the express companies got what they deserved. For 30 years they had systematically monopolized their end of the business by agreeing not to infiltrate each other's territories. For that reason, no very strong common carriage tradition had evolved between express companies and railways. And yet the principle announced by the Supreme Court would have decades of unfortunate consequences. Common carrier obligations, it seemed, extended from the aristocracy of the business world to the serfs and peasants of "the public." But a railroad baron owed nothing to an express package baron. A common carrier's duty not to discriminate ceased when another carrier came asking for service.

The same principle would come to be applied in telephony. In 1903, for example, what would become the Interstate Telephone Company won permission to install a telephone system in Newport, Washington.(44) An early owner of the system agreed he would not extend the system outward to compete against Pacific Telephone, and would connect all long distance calls to Pacific's network. When Interstate bought the system in 1911, it immediately cut off Pacific and connected the Newport exchange to its own long distance lines. Pacific demanded reconnection; Interstate answered that the original contract was illegal in that it froze out other carriers. Not at all, ruled a federal judge. "[E]ach telephone company is independent of all other telephone companies"; common carrier law did not require telcos "to accord to any such outside organization or its patrons connection with its switchboard on an equality with its own patrons."(45) Other cases of the era were to the same effect, even in the teeth of state statutes requiring universal carriage. "Telephone and telegraph companies are common carriers," the Supreme Court of Tennessee conceded, for example, "[b]ut this does not mean that a telephone company is bound to permit another telephone company to make a physical connection with its lines."(46)

Congress had its first important chance to correct this balkanized environment when it passed the Mann-Elkins Act of 1910.(47) The Act brought interstate telecommunications within the regulatory jurisdiction of the Interstate Commerce Commission (ICC) and included telecommunications companies within the definition of "common carrier," thereby obligating them "to provide service on request at just and reasonable rates, without unjust discrimination or undue preference."(48) But the Act failed to specify a common carrier’s obligations to other carriers. The ICC might have taken steps to correct this failure, but was preoccupied by railroads and uncertain of its powers. The Bell System’s march toward monopoly thus continued, unchecked.

In 1912, fresh on the heels of its victory against Rockefeller,(49) the U.S. Justice Department threatened to take on Vail. These followed a great deal of sound and fury, no doubt reflecting America's traditional populist mistrust of monopoly. In the end, however, government officials would conclude that monopoly in communications was much more tolerable than monopoly in oil.

In a 1913 agreement between the U.S. Attorney General and a Bell vice-president, N. C. Kingsbury, the company agreed to stop acquiring independent phone companies and to connect the remaining independents to Bell's long distance network.(50) Bell also agreed to divest itself of Western Union. It appeared to be at least a partial victory for the trust-busters at the Justice Department – an honorable armistice signed by vigilant public servants to contain all-too-acquisitive private interests. But the Kingsbury Commitment was really nothing of the sort.

It did indeed stop the growth of Bell's financial empire – for a few short years, in any event – but it did nothing to promote competition in either telephony or telegraphy. Local exchange monopolies were left intact, utterly free to continue to refuse interconnection to other local exchange companies. Bell's monopoly long distance service was reinforced: Bell would be required to interconnect with all local exchanges, but there was no provision for any competition – or interconnection – among long distance carriers. Western Union was indeed spun off – but only to provide telegraphy, not telephony.

The government solution, in short, was not the steamy, unsettling cohabitation that marks competition, but rather a sort of competitive apartheid, characterized by segregation and quarantine. Markets were carefully carved up: one for the monopoly telegraph company; one for each of the established monopoly local telephone exchanges; one for Bell's monopoly long distance operations. Bell might not own everything, but some monopolist or other would dominate each discrete market. The Kingsbury Commitment could be viewed as a solution only by a government bookkeeper, who counted several separate monopolies as an advance over a single monopoly, even absent any trace of competition among them.

To judge by actions, then, rather than words, government officials had no strong objection to monopoly telephone service. It was just that the Bell empire had been getting too big and wealthy. For reasons of bureaucratic convenience, government officials preferred single-serving monopolies to the large economy size monopolies. This was especially true for state regulators. For them, a local telephone monopoly was both welcome and convenient. The problem with the burgeoning Bell System was that it had been growing larger than local politics.

Kingsbury was the first – though as we shall see, by no means the last – antitrust quarantine placed on the Bell System.(51) The Bell System remained a colossal corporate empire; although it had agreed not to grow by acquisition, nothing prevented it from building up businesses it already owned. It continued to enjoy almost blanket state and federal government protection against competition within its own businesses. The outward expansion of the monopoly had perhaps been stopped, but there was certainly no promise in the Kingsbury Commitment for any new competition within the confines of the Bell Empire.

Radio, originally conceived as an alternative to the telephone, was born within a decade of the signing of the Kingsbury Commitment. Bell took an immediate interest in the new technology. But competition was the last thing that the new radio companies wanted, and there quickly developed a united opposition to Bell's presence in the new market. The code word was "monopoly," and monopoly was resoundingly condemned. In the 1920s, however, that did not mean government-licensed broadcasters or dominant national networks. It meant Bell, and its common carrier notions for marketing the new medium. In Congress and elsewhere the alarm was sounded: No monopoly in radio. Which really meant: No competition – at least not from Bell.

Bell got the message. Ever since the Kingsbury Commitment, the company had adopted a philosophy of accommodating politics rather than fighting it. Bell managed its exit from the broadcast business with characteristic brilliance.

By 1926 it was already clear that much of the future of radio broadcast lay in networking. Indeed, the first radio networks were established that year. And the links between radio stations needed by the national radio networks were not going to be over-the-air – they were going to be by wire. Wire could provide much higher transmission fidelity over long distances. And, in any event, it was uneconomic to use powerful transmitters and to crowd the airwaves with signals that were being transmitted point-to-point, from an originating station out to its network affiliates. This meant that much of the flow of information ultimately delivered over-the-air would in fact occur by wire. The wire networks would move the programs out to the cities and towns. The radio stations would provide local distribution. Or, viewed another way, the phone system would be used for the long distance transmission of radio signals; radio transmitters primarily for the last mile connections.

Here, then, were the perfect elements of a market-dividing deal. Bell already had a wire network. The new radio companies wanted to dominate local, over-the-air broadcast. The markets would be simply and cleanly carved up, to reduce competition in both sectors. That was exactly what happened.

In 1926, Bell sold its flagship radio station, WEAF, to RCA, for $1 million, to serve as the nucleus of RCA's broadcasting network. RCA committed to use Bell's lines, rather than Western Union's, for its networking. Patents on the single critical piece of technology common to both telephony and broadcast – the audion, the electronic amplifier – had previously been pooled by Bell and the leading radio companies.(52) The patent-pool contract was renegotiated to exclude Bell entirely from the over-the-air end of radio broadcasting, while Bell's exclusive hold over the wire network, and all point-to-point messaging systems, was strengthened. It was really the Kingsbury Commitment all over again, this time negotiated privately. Bell stopped expanding its local radio operations and indeed agreed to get out of the local distribution end of the business altogether. In return, the remaining local radio stations agreed to interconnect only through Bell's long distance network. A gentlemanly agreement, reached under political pressure, had once again replaced competition with complementary monopolies. RCA then used WEAF as the centerpiece of its wholly-owned broadcasting subsidiary, the National Broadcasting Corporation. NBC quickly acquired a position of complete dominance in radio,(53) almost as overwhelming (for a time) as Bell's in telephony.(54)

The Radio Act of 1927 sealed the deal.(55) It reaffirmed the general prohibition on "monopoly" of the airwaves – meaning that competition over the airwaves was prohibited, at least if it came from Bell. The Act forbade cross-ownership of telephone companies and broadcasting stations,(56) and flatly rejected the operation of radio stations as "common carriers."(57) None of this could have concerned top officials at RCA or Bell very much. Congress merely cemented and strengthened a division of markets and territories that the parties had already voluntarily embraced.(58)

The Kingsbury Commitment and the 1927 Radio Act defined the regulatory landscape until the Federal Communications Act of 1934 was signed into law on June 18.(59) By this point, monopoly provision of telephone service had become so familiar it seemed inevitable. Several ponderous studies officially confirmed that it was, a conclusion perfectly consonant with the New Deal political winds then blowing. Monopoly was, in any event, the clear premise of the 1934 Act. The monopoly telephone company was to provide service to all customers at "just and reasonable" prices.(60) New firms would be permitted to compete only if they demonstrated that the "public convenience and necessity" so required.(61) And the monopoly was required to provide connection to newcomers only if a commission found such connection "necessary or desirable in the public interest."(62)

For some economists, these arrangements seemed to make perfect sense.(63) If telephony was indeed a natural monopoly, competition would be futile and inefficient and so should be forbidden. Without competition, however, prices would have to be regulated, or they would be set too high. The regulator – advised by a large staff of economists, of course – would intervene as a visible surrogate for the invisible hand of the market, the invisible hand being altogether too invisible when it came to guiding the business practices of a monopoly.

But regulators, it turned out, much preferred to emulate governments than to mimic markets – they preferred transferring wealth to promoting efficiency.(64) For many years, the paramount regulatory objective in all of telephony, at both state and federal levels, was to promote universal service: a telephone within arm's reach of the chicken in every pot.(65) The Communications Act itself established this as one of the FCC's several missions.(66) So the dominant objective was to push local telephone rates down. As for universal service to other telephone companies, equipment providers, sellers of long distance service and such, well, the Communications Act hadn't meant that kind of universality. Universal service was something for consumers, not providers.(67)

The pricing arrangements that developed in this regulatory climate violated the most elementary principle of economic efficiency.(68) The cost of connecting subscribers to the interstate network is largely "non-traffic-sensitive," but was recovered through per-minute markups on long distance calls. Common costs were not merely distributed between local and long distance service, but triply over-distributed to the long distance side.(69) Interconnections between the public phone network and all other devices and services were the most expensive of all – they were illegal. Thus, in the fullness of time, the Bell System would seek (and receive) from the FCC a solemn order forbidding the attachment of an alien device to its network – to whit, a plastic cup called a Hush-a-Phone, that slipped over the mouthpiece to provide some small measure of quiet and privacy in otherwise crowded offices.(70) Telephone service remained largely local, and therefore regulated primarily by state public utility commissions. And the PUCs were the most zealous of the zealous when it came to subsidizing local operations and promoting universal service.(71)

In retrospect, at least, it is remarkable to see how many opportunities there were to set the industry on a competitive path, and how persistently these opportunities were neglected or deliberately turned aside. What would finally become the core interconnection and unbundling provisions of federal law in 1996 could have been put in place on any number of occasions, decades earlier. The courts might have done the job in the beginning, by simply building on ancient principles of common carriage. A one sentence decision in the Express Package cases might have made all the difference – a sentence to the effect that common carriers really were common carriers, even for business brought to them by other carriers. State legislatures and public utility commissions had the next opportunity to set things on the right track; faced with unduly narrow interpretations from the courts, they too could have insisted that carriers really had to be carriers, for each other as well as for the general public. The federal government, first through the ICC and later the FCC, could have demanded the same, at least for interstate traffic. The market itself would have taken care of things, given time, for there was more profit to be made in interconnection and growth than in boycotts and stasis. Every opportunity was missed, however, and when government intervened it did so not to promote market forces but to outlaw them once and for all. Indeed, at almost every stage, the government systematically approved and upheld the boycotts, the exclusive dealing, and the inevitable attrition and fratricide that rapidly transformed telephony from an industry of flourishing competition to one of entrenched monopoly.

Guessing at motives, especially a century later, is not a very useful exercise. But, by all appearances, government authorities – in the courts, in the state agencies and legislatures, and in the federal government – were quite content to let the telephone industry slip from the unruliness of competition to the quiet order of monopoly. Perhaps government officials saw and welcomed what lay ahead. They would arrive at the scene of the great competitive battles late in the day. They would shoot the wounded, congratulate the victors, and then claim the new empire of monopoly as their own. It was an empire worth controlling. At its apogee, the Bell System had annual revenues of $58 billion, total assets of $138 billion, and employed over 1 million people. No other company in the world was as large. Indeed, most other nations owned and produced little by comparison.

The government did not just maintain a Bell System monopoly within telephony. It also carefully maintained divisions between telephony and other media and among those other media themselves.(72) It was tidy arrangement, in which each medium of communication performed a particular function, pursuant to its own particularized set of regulations.

Certainly, considerations of engineering efficiency historically favored different architectures for different types of communications. The telephone network was optimized for point-to-point voice communications between individuals. Broadcasting infrastructure was designed for point-to-all transmissions. Cable plant began life as a Acommunity antenna,@ distributing video from a local headend without two-way or switching capabilities. All of these architectures and capabilities originated in the day of analog technology and narrowband transmission. Only data networks, which arrived much later, were designed from the outset around a digital, broadband model that supported both point-to-point and point-to-multipoint transmission.

These basic engineering differences made it easy for regulators to maintain different regulatory regimes for different media and the communications companies that operated them. Drawing legal distinctions between a Acommon-carrier@ Aphone company@ and a Abroadcaster@ was straightforward. Until the advent of digital computers and modems, there was too little in the way of Adata communications@ to merit much regulatory attention at all.

Thus, a company or network zoned as a carrier is required to convey the content of others; one zoned as a broadcaster must carry content of its own. The 1934 Communications Act defines telephony, the common carrier paradigm, as two-way intercommunicating voice service.(73) It defines broadcasting, by contrast, as one-way dissemination of radio communications.(74) With varying degrees of conviction, regulations promulgated under the 1934 Act have curtailed video on phone lines,(75) voice on cable,(76) and point-to-point messaging on broadcast media.(77) Phone companies have also been required to separate any enhanced services they may provide from basic carriage.(78) Similarly, carrier services offered on the subcarrier and VBI bands of broadcaster spectrum are regulated separately from the broadcasting.(79)

Much zoning has traditionally occurred during licensing.(80) Section 214 of the 1934 Act requires phone companies to obtain advance FCC permission before constructing new interstate facilities.(81) Section 201 empowers the Commission to regulate devices connected to the telephone network and thus, in principle, to regulate every service offered over a telephone line by a phone company or anyone else.(82) The FCC has used this authority to restrict phone companies to providing basic voice services, and little else.(83) The FCC is likewise granted plenary authority over use of the radio waves, and has used it over the years to prescribe in great detail how spectrum may be used.(84) Radio licensees are generally issued for a specific use, such as public broadcasting.(85) A licensee may not convert its spectrum to any other use without first obtaining FCC approval.(86)

For the most part, existing regulatory codes for telephone, cable, and broadcast treat "data" services as largely beneath or outside official attention.(87) AEnhanced services@ have been excluded from the ambit of Abasic,@ regulated telephony. They are exempt from access charges, and almost completely free of most other forms of common-carrier regulation.(88) Data services provided over mobile radio, cable, terrestrial broadcast and DBS have been largely excused from most forms of rate, content, and carriage regulation. Data has long been the Aincidental@ service tagged onto something else older and more important. As such, data has been the fortunate beneficiary of regulatory accident, inattention, neglect, and indifference. Wires and radios alike will all soon be digital, and bandwidth is increasing rapidly in every medium. AData@ traffic is growing far faster than analog voice or video. And on broadband digital channels Adata@ encompasses everything. But as we discuss below, a digital world transforms everything into Adata.@ The data inmates are taking over the regulatory asylum.

The genius of capitalism, as Lenin might have pointed out, is that it develops its own rope, for hanging as much as for other purposes. Even in the shelter of its government-embraced monopoly, the monstrous Bell System remained a for-profit enterprise with some residual instinct to provide better, faster, and cheaper service. The company had been founded on technology, and technology would remain its strength. In time, Bell would establish the greatest peacetime research program the world has ever known, at Bell Laboratories in Murray Hill, New Jersey. The company set its scientists in pursuit of better telephones, better networks, and better switches. The pursuit would span radio, then broadband communications, then telephone exchanges, and would end finally where the industry had begun – with the telephone box, now called a computer.

Since 1927, Bell had been barred from broadcast radio. But not all radio waves spread out like ripples on a pond; at sufficiently short wavelengths radio waves can be made to move in straight lines, like beams of light and at the same speed. And these electromagnetic waves, especially at higher frequencies, have enormous capacity to carry information accurately over long distances. Realizing this, the broadcasters would push their way up the radio spectrum and by the 1950s would be transmitting the considerable volumes of information required for television. Telephone companies, especially Bell's long distance arm, needed equally high-capacity – "broadband" – transmission systems, but systems that could operate over much larger distances, more securely, and more reliably.

Bell Labs researchers began to investigate wave-guide transmission in 1931. The idea was to use the high transmission capacities of high-frequency airwaves, but with signals confined in ducts to maintain quality and assure secure, point-to-point transmissions. By 1934, other researchers had developed a variation on the same idea – the coaxial cable, a copper sheath with a wire running down its center. "Coax," it turned out, could carry huge amounts of information with low attenuation. Within two years, Bell had installed its first coaxial cables in New York. Coax would eventually supply many of the high-capacity "trunks" of the telephone network until the end of World War II.

Meanwhile, research at Bell Labs was laying the foundation for another broadband alternative, microwave communications. Harold T. Friis and his colleagues developed the horn-reflector antenna, now a standard fixture on microwave towers. Unlike their longer wavelength cousins listed on the dial of any car radio, microwaves – short-wavelength radio waves – travel in straight lines and can be accurately focused. Because of their comparatively high frequency, microwaves can also carry far more information; even early systems were designed to carry up to 1,000 voice circuits. Their final advantage for long distance transmission, especially in rural areas, was also the most obvious: microwave towers could be placed twenty or thirty miles apart, and a single license from the FCC could substitute for the cumbersome process of obtaining rights of way over the entire span. By 1959, microwave systems comprised 25 percent of Bell's long distance network.(89)

Coax and microwaves would transform more than the telephone industry.(90) John Walson, Sr. of Mahanoy City, Pennsylvania, recognized that coaxial cable was the perfect medium for connecting homes in rural areas with poor television reception to a large, community, "master antenna." He began work on his first "community antenna television" system in 1948.(91) Others took up the idea. Antennas were placed on hilltops, on tall buildings, on masts. Distant signals were picked and piped to viewers over coaxial cable. Before long, antenna operators began using microwave systems to beam in television signals to the master antennas from still farther afield. Hence the origins of cable television. By 1955 there were 400 such systems in operation, serving 150,000 subscribers.

Not long after, in 1963, a small startup firm, Microwave Communications Inc., underfunded and understaffed, would apply to the FCC to construct a microwave line between St. Louis and Chicago.(92) MCI told the FCC it would offer business customers "interplant and interoffice communications with unique and special characteristics;"(93) in fact what MCI had in mind was direct competition with Bell's long distance operations.(94)

MCI could consider offering private, dedicated, high-capacity connections to large business customers, only because by the 1960s those customers had begun to acquire on their own premises the equipment that would make such connections useful. It had taken the telephone to create the original demand for a telephone network almost a century before. Now, a new generation of sophisticated electronic equipment in private hands – a new generation of "local exchanges," so to speak – was beginning to create demand for the kinds of competing long distance services that MCI proposed to offer.

Appropriately enough, Bell Labs had once again been at the forefront in creating the new technology. The technology of telephone exchanges had languished for some years after the first electromechanical switches began to be used. It had not been until 1938 that the first "crossbar" central switching office was installed in Brooklyn. As late as 1951, operators were still being used to dial almost 40 percent of domestic long distance calls.(95)

In 1936, Bell Labs' Director of Research had first discussed with physicist William Shockley the possibility of creating electronic telephone exchanges. Electronic switching, however, required a better amplifier than the vacuum tube technology pioneered by Lee DeForest. While each triode vacuum tube was capable of operating as a switch in a telephone exchange, an exchange needed thousands of such switches; tubes used too much power, and generated too much heat, to be packed together in the numbers required. Shockley and his Bell Labs colleagues Walter Brattain and John Bardeen set off in search of something better. They found it in 1947: the transistor. A Nobel prize followed in 1956; in late 1963 Bell installed its first electronic switching system.(96)

The transistor, like the vacuum tube it displaced, was a switch, only much more compact and energy-efficient. Switches are the heart of a telephone exchange, for it is by opening and closing an appropriate set of switches that a single continuous line is created between Romeo in San Francisco and his Juliet in New York. Switches are also the heart of a computer: by shifting on and off like beads moving on an abacus, switches can keep track of numbers, and numbers can keep track of everything.

Before the advent of the transistor, both computers and telephone exchanges had required large, cumbersome, costly, custom-configured, labor-intensive centers. With the new electronics, much more powerful telephone switches and computers could be built into much more compact and reliable units – minicomputers and private branch exchanges (PBXs). Larger institutions – hospitals, universities, corporate headquarters, and so on – had once relied on a few, centralized mainframes to do their computing, and on "Centrex" services to handle their telephone requirements, even for internal telephone calls within a single office building. Now these same functions could be – and rapidly were – located in stand-alone units on private premises. Competing manufacturers of PBXs and minicomputers proliferated. By the late 1970s, even Bell was systematically downgrading Centrex service and migrating its larger customers to PBXs.

This dispersion of electronic intelligence created a host of new centers, held in private hands, capable of communicating by wire and in need of the connections to support the same. As had happened almost a century earlier with the rise of the telephone itself, the new talking boxes created new demand. What was critically different about the new-generation local exchanges, whether PBXs or communicating computers, was that they were owned and controlled not by a small number of quasi-governmental, monopoly phone companies, but by a larger number of private, competitive institutions. For the most part, these private owners welcomed competitive bidding for their telecommunications needs. MCI, as we have seen, was one of the first to target the new demand.

At the same time, the transistor was also fulfilling its original mission, which was to transform the public telephone exchange: a new generation of electronic switches was deployed in the 1960s and 70s. These switches were far more efficient, powerful, and flexible than the old switches they replaced. They could support levels of interconnection – and thus offer customers a variety of choices – that would have been prohibitively slow, complex, and unreliable in the days when switching was accomplished by human operators or electro-mechanical devices. As MCI built up its business in the 1970s, the company resolved to carry competition back up the network – to compete not just in connecting private computers and switches, but also in connecting the public exchanges operated by the Bell and other public telephone companies. The capabilities of the new electronic switches made that aspiration quite realistic; as every telephone user knows today, such switches can be programmed with databases to route traffic automatically, Hatfield's to Bell, McCoy's to MCI, effortlessly and invisibly whenever either places a long distance call.

Shockley and his colleagues discovered the transistor in 1947, but neglected to inform lawyers in the antitrust division of the Department of Justice. Two years later those lawyers filed suit to break up the Bell System.(97) In 1956, Shockley, Brattain, and Bardeen traveled to Stockholm to collect their Nobel prizes. The government's antitrust lawyers traveled less far that same year: to a district court in these scientists' home State of New Jersey, where on January 4 the government lawyers threw in the towel on their case. It mattered very little. The important antitrust work was being done by the monopolist itself, at Bell Labs. The first scent of competition was already in the air, discernible even in the torpid air of the nation's capital.

Although the basic premise of the Communications Act was that telephone service, both local and long distance, was a natural monopoly, the FCC's statutory mandate was broad enough to permit competition where the Commission, in its wisdom, thought competition might serve the public interest. When microwaves, satellites, and developments in CPE, computers, and radio communications began to make competition feasible in the 1950s and thereafter, the Commission came under intense pressure from would-be competitors to do just that. The FCC slowly warmed to the idea that competition might not be so bad after all.

For a period it even moved out ahead of the courts. In 1951, for example, the Commission authorized a radiotelegraph company to open new circuits to Portugal and the Netherlands, in direct competition with another company that already provided similar service.(98) Suddenly discovering a "national policy in favor of competition," the FCC authorized duplicate facilities; competition is in the public interest, the Commission declared, wherever it is "reasonably feasible." In a 1953 ruling, however, the Supreme Court could not discern the "national policy" on which the FCC had relied. The "encouragement of competition as such has not been considered the single or controlling reliance for safeguarding the public interest," the Court stated.(99) "To say that national policy without more suffices for authorization of a competing carrier wherever competition is reasonably feasible," the Court continued, "would authorize the Commission to abdicate what would seem to us one of the primary duties imposed on it by Congress."(100) The Commission would have to do a better job at articulating just how competition would serve the public interest.(101)

If authorizing competition had been the only thing required, a new national policy might have evolved with little protest. But the old national policy came with an old set of franchises, quarantines and cross-subsidies that were fundamentally irreconcilable with competition. The government, as in an M.C. Escher drawing, soon found itself fighting an antimarket monstrosity that was largely of its own creation.

Under the paradigm of regulated monopoly, the FCC's primary goal had been to provide universal service at a reasonable cost. Regulators focused primarily on rates and their overall relation to carrier costs. To encourage universal service neither the FCC nor state regulators sought to align prices of individual services with their attendant costs. Long distance rates were used to subsidize local rates. Business rates subsidized residential rates. Urban rates subsidized rural rates. So long as telephone service remained an end-to-end monopoly, schisms between price and cost did not really matter. A large, New York brokerage house making heavy use of long distance service had no choice but to subsidize homeowners in Fishkill.

When competition arrived in some segments of the industry, so too did the significant problem of "cream skimming." New competitors would naturally target low-cost, high-price services, such as high-density routes between major cities. If the Bell System lowered its long distance prices in response, other subsidies would have to be eliminated, over the vehement protest of state regulators and those receiving the subsidies. If the Bell System did not lower its prices, it would lose the business, and with it the subsidy, too. At the same time, fledgling competitors had to be protected from the potentially overwhelming might of the Bell System; the company that could arbitrarily set price above cost could also arbitrarily set it below, and so ruin the competition.

Schooled in the ethic of end-to-end service, the Bell System made it quite plain from the outset that it viewed competition as a direct threat to the higher objective of seamless, universal public service. And Bell had at its disposal a potentially decisive weapon to use against that competition: its control over the local exchange. The local exchange would, in fact, prove to be the most durable component of the old monopoly – either because (as some maintain) local exchange service is a "natural" monopoly, or (as others have suggested) because monopoly is most fiercely protected here by reactionary state regulators. In any event, competitors in all segments of the telephone industry depended on access to (or interconnection with) the local exchange to reach their customers.

In managing a transition to competition, a large part of the FCC's responsibility was thus to ensure that the Bell System did not refuse competitors fair access to the local exchange or shift costs from competitive ventures to local exchange operations.(102) The twin dangers of discrimination and cross-subsidization occupied the attention of the FCC for decades before the Justice Department and the courts – dissatisfied with the FCC's efforts – engineered the Bell System's dismemberment. The Commission's alleged failure to deal adequately with the coexistence of monopoly and competitive markets served as a major impetus in the break-up.

Nonetheless, the Commission's record had been (and would be) better than generally acknowledged. Indeed, most of the transition to a competitive marketplace – complete with safeguards to protect that competition from any bottleneck abuses – were in place prior to divestiture. The Commission's efforts were sometimes clumsy, and often too slow. But, by and large, things were moving in the right direction when the Department, spurred on by MCI, filed the suit that altered the face of the industry forever.

The FCC's reforms were directed at four different markets. With CPE and long distance service, previously monopolistic markets were opened to competition. Protected franchises were eroded, quarantines were lifted, and rate-of-return regulation was replaced by price caps and, even, by prices set by competitive forces. With enhanced and wireless services, brand new markets were opened to competition from the beginning. Rules governing interconnection were changed in parallel: telephone companies were required to provide service to all comers, including competing carriers. Across the board, exclusive franchise gradually gave way to open competition.

The Bell System had long sought to provide "end-to-end" service to its customers. This included not only the wires and switches that made up the phone network, but also the CPE that customers wanted to connect to the network. Bell tariffs forbade all "foreign attachments" – meaning equipment not provided by Bell itself.

Regulators endorsed this practice for almost 70 years.(103) It was not until the 1970s that the foreign attachment provisions were finally and decisively eliminated and standard CPE interfaces with the network were established. The landmark decision came in 1968, in Carterfone.(104) At issue was a device permitting direct communication between a wireless radio and the landline network. The Bell System objected to the device, but failed to demonstrate that it would harm the network in any way. The Commission held that any form of CPE could be attached to the network "so long as the interconnection does not adversely affect the telephone company's operations or the telephone system's utility for others."(105)

With this ruling, the Commission seemingly opened the way for a complete line of competitive products that would interconnect with the network on customer premises. Yet almost immediately, the Commission began to waver.(106) And the Bell System, in filing replacement tariffs, narrowly construed the Commission's directive and tried to interpose expensive "protective connecting arrangements" (PCAs) between non-Bell CPE and the network. By 1975, however, the Commission had screwed up its courage to throw out PCAs and replace them with its own much simpler standards.(107) Finally, in 1980, CPE sales by Bell and other carriers were "detariffed," so that carriers themselves would be free to compete on equal terms in the open market.(108) The provision of CPE, the Commission held, was simply no longer a part of "communications common carriage." Even now, the Commission hesitated again: Bell (and later the divested Regional Companies) would be permitted to provide CPE only through separate subsidiaries – the quarantine still held sway. That last wall would fall, however, in 1985.(109)

From a technological perspective, the issues raised by enhanced services were identical to those in the CPE proceedings.(110) On-line information services, after all, are provided by telephone-compatible equipment located on some telephone customer's premises. But from a market perspective, CPE and enhanced services started worlds apart. On-line electronic information services evolved only after the winds of competition had begun to blow. The FCC accordingly chose not to regulate the new services at all – except as they might be provided by telephone companies. In the spirit of the old quarantine, the Commission resolved that "enhanced services" would not under any circumstances become entangled with plain old telephone service.(111)

The definitional problems soon became overwhelming. A first set of lines were drawn between pure "data processing," "hybrid data processing," "hybrid communications" and pure "communications." A few years later, the Commission tried a simpler pair definitions – "basic" and "enhanced" services – the former to be regulated strictly, the latter to be deregulated and quarantined from "basic" telephony.(112) Then for a while the Commission flirted with three categories: "basic" services, "services ancillary to communications," and "non-communications services." All of this definitional hot air was needed, it was thought, to contain a single company – Bell. The divisions survived divestiture, but only by a few years. It became increasingly apparent to the Commission that, in a post-divestiture world, the costs of structural separation outweighed any conceivable benefits. In 1986, the FCC finally decided to end structural separation for the enhanced services of AT&T and the divested Bell companies.

1.5.3 Long Distance

In the 1940s, long distance service was provided exclusively over wires, and the same basic economics that seemed to preclude competition in local service applied equally to long distance service.(113) The development of microwave and satellite technologies radically changed that picture, making competition both practical and inevitable. Competition was spurred, in addition, by the Bell System's policy of nationwide price-averaging, and the heavy subsidies from long distance to local service.

Initially, the pressure for competition came from large businesses, which sought to build microwave links solely to satisfy their own, private communications needs. As time wore on, however, various "specialized" common carriers sought to provide service on traffic intensive routes to smaller businesses that could not afford their own private facilities. Eventually, these specialized common carriers sought and gained permission to serve the public at large, though this still meant, largely, the urban public. Only the Bell System had the obligation of universal service.

The first few, faltering steps in the direction of a competitive marketplace, were taken by the Commission in 1959, when the Commission allowed private users to construct point-to-point microwave facilities.(114) A decade later, a deeply divided Commission granted MCI's long-pending application to provide private microwave service for businesses as a "specialized common carrier," and then extended its decision to others by establishing general conditions for entry. The Commission also forbade any tariff restrictions on the resale and sharing of long distance private line services, and sought to regulate the terms, conditions, and charges for the interconnection of the private lines of the specialized common carriers with the local distribution facilities of existing common carriers. With a strong shove from the D.C. Circuit court of appeals, the Commission eventually opened a new rulemaking proceeding, and in 1980 formally adopted an open entry policy for all interstate services.(115)

Thwarted in its efforts to prevent competition through regulation, however, the Bell System tried to make competition unprofitable by dramatic price reductions on competitive routes. The system of nationwide price averaging – and the subsidies embedded in that system in an effort to promote universal service – was the first casualty of competition.(116) Few disputed the Bell System's right to bring prices in line with costs to meet competition. But many disputed whether the new tariffs reflected costs or a below-cost attempt to drive out competition. Charges of cross-subsidization and cream-skimming were freely traded. The specialized common carriers (using a fully distributed costing methodology) accused the Bell System of below-cost, predatory pricing and sought to have its tariffs declared unlawful. The Bell System (using a marginal cost method) claimed its tariffs were cost-based and justified by competition. The Commission was unable to tell the difference and sank into a state of almost complete paralysis. These disputes lasted for decades and contributed to the sense that the Bell System was simply too large and too unwieldy for the Commission adequately to regulate.(117) That, in turn, helped spur the Justice Department's calls for breaking up the Bell System.(118)

To create interconnection rules for CPE and long distance competitors, the FCC had to dismantle decades-old practices based on Bell's hegemony. With enhanced services the Commission began at the other extreme, with a policy of complete telco exclusion. Only with wireless services did it get things right from the start.(119)

What it did, from the start, was to embrace competition – but reject quarantine. Wires are dedicated facilities that (in the last mile, at least) may lie idle most of the time; radio spectrum, by contrast, is a shared resource. Thus, the economics of providing wireless services – and the costs of embedded plant – are completely different from those for landline local exchange service.(120) Whether or not the local landline exchange is characterized by the steadily declining average cost of a natural monopoly, the Commission recognized that the market for wireless services plainly is not.

Thus, when it first allocated frequencies for Aland mobile services@ in 1949, the FCC granted separate blocks to telephone companies and to "miscellaneous" or "limited" common carriers.(121) The Commission consistently maintained that procompetitive policy thereafter. In the paging market, the FCC likewise licensed sufficient frequency to permit multiple providers in every service area. And it continued to encourage competition by allocating additional channels as needed and by allowing radio broadcasters to use FM subcarrier frequencies for paging. When it began to issue cellular telephone licenses in the early 1980s, the FCC allocated two licenses for every service area, prohibited any licensee from owning a significant interest in both licenses, and thereafter encouraged the development of other radio technologies capable of providing directly competitive services. The Commission complemented its procompetitive licensing policies with active industry oversight. Most importantly, it required all landline telephone companies to provide unaffiliated wireless concerns with interconnection equal in type, quality, and price to that enjoyed by affiliates.(122) Both competition and service have flourished under these arrangements.

By the early 1980s, with some prodding from the courts, the FCC was thus moving firmly toward competition in CPE, enhanced services, long distance, and wireless. The Commission could not easily dismantle Bell's corporate structure, but it could unpack the network itself. It could require Bell to interconnect with other long distance carriers, wireless carriers, paging companies, information service providers, and in time other local exchange carriers too. Equal access principles were evolving incrementally, product by product, service by service. The pace of change was too slow to satisfy newcomers, who were often still thwarted in their attempt to interconnect. But it also seemed too quick for Bell and state regulators, who saw traditional franchises and subsidies evaporating under the radiant heat of competition. Matters went to court. And the courts . . . the courts at one time or another, in one place or another, sided with every side on every issue.(123)

The first private antitrust case involving CPE was decided summarily in favor of Bell in 1940.(124) An inventor had obtained a patent on an automatic dialer, which allowed "certain pre-selected call stations [to] be signaled by a single manipulation, pressing a button."(125) The inventor was out of luck. Subsequent CPE cases ended in much the same way.(126) Carterfone itself began as an antitrust suit; a federal district court ruled that the FCC had primary jurisdiction, and the Fifth Circuit affirmed in 1966, in an opinion replete with language strikingly deferential to the FCC.(127) For the next decade or so, Bell and other telephone companies continued to win easy dismissals of foreign attachment antitrust cases.

The FCC's decisive shift in favor of CPE competition in 1975, however, changed the atmospherics completely. An avalanche of antitrust suits followed, attacking Bell PCA tariffs as a sham designed to delay the advent of competition.(128) Cases that only a few years earlier would have been decided summarily by reference to extant tariffs were now litigated. A 1979 ruling of the Third Circuit reversed a trial judge's dismissal of a suit by a distributor of an answering device called the Code-a-Phone.(129) A small Connecticut terminal equipment vendor won $16.5 million on various charges of discriminatory interconnection and Bell's abuse of its "utility function."(130) Another case in the same circuit involved a five-month trial directly attacking Bell's PCA tariffs; a jury awarded Litton $276.8 million.(131)

Predictably, that verdict triggered a raft of me-too suits, with many plaintiffs hoping to invoke offensive collateral estoppel. But the easy victories did not materialize. The D.C. Circuit declined to apply offensive collateral estoppel in a 1984 ruling. Most other courts reached similar decisions. Ironically, the last PCA case reembraced the logic of the early cases, and vindicated Bell completely.(132)

None of the antitrust suits against the Bell System in the 1960s and '70s focused on Bell provision of information services. Indeed, most of the antitrust litigation relating to information services has occurred outside the telephone industry and has focused on the content – software, news wire services, movies, airline reservation systems, etc. – rather than the conduit.(133)

The private antitrust suits attacking Bell's dominance of the long distance market matured in the 1970s, smack in the middle of the turn-around in policy under the joint (if often uncoordinated) impetus of the FCC and the D.C. Circuit court of appeals.(134) Three of these cases supply a record of a judiciary at war with itself – mirroring, in other words, the schizophrenic policies emanating at that time from the FCC.

In March, 1974, MCI sued Bell for monopolizing long distance communications.(135) Six years later, a Chicago jury awarded MCI $600 million, trebled to $1.8 billion. On appeal, the Seventh Circuit accepted some aspects of the verdict but rejected others. Bell ultimately paid some $113 million. A second private suit was filed in 1980 by an independent telephone company that had sought to provide local telephone service in a new real estate development.(136) Bell had refused interconnection with its long distance facilities and refused to assign the company a block of telephone numbers. A federal district court awarded $55 million in treble damages; however, the Fifth reversed on an error in jury instructions.(137)

Sprint (at that time Southern Pacific Communications) was the really unlucky one. It filed suit in 1978. After a 33 day bench trial, federal Judge Richey accepted Bell's defense chapter and verse. The D.C. Circuit court of appeals affirmed.(138) The most interesting part of this last case, however, is the one least remembered. The case had originally been assigned not to Judge Richey, but to Judge Joseph C. Waddy. Judge Waddy had also been assigned the federal government's 1974 case against Bell. Waddy, however, died in 1978. Sprint's case was reassigned to Judge Richey; the federal government's was reassigned to Judge Harold Greene. Had the docket assignments been reversed, the Bell System would still be intact today.

Judge Greene quickly took control of the case he had inherited.(139) The action was ultimately settled, so he never had a direct opportunity to rule on whether Bell had violated the antitrust laws. In ruling on Bell's motion to dismiss at the close of the government's case in chief, however, he telegraphed where he was headed. The gargantuan Bell System was too large; something had to be done. Bell, a company with some understanding of telegraphs, took the hint and soon thereafter acceded to the government's terms.

The essence of the settlement was that telephony could be split into two pieces. The larger of the two – the local exchange – was deemed to be a necessary and desirable monopoly, that might quite properly be granted an exclusive franchise, but that should be regulated and quarantined in return. The smaller, comprising long distance service and equipment markets, was thought to be competitive, at least potentially though not yet in actuality. This piece should not be regulated, should not be quarantined, and should not escape competition by way of any exclusive franchise or any equivalently cozy relationship with the divested local franchisees. Competitors dependent on the monopoly services should therefore enjoy "equal access" to Bell local exchanges.(140)

The proposal was thus, at one and the same time, enormously liberating for the one-third of Bell that would become AT&T and more confining than anything that had come before for the two-thirds of Bell that would become the seven Regional Bell Operating Companies (or "RBOCs" or "BOCs," as they came to be known). The seven RBOCs would be forbidden, with minor exceptions, to provide any service other than local exchange service. In particular, they would be forbidden to provide long distance service, "information services," or to manufacture telecommunications equipment and CPE.(141) It was all based on a tidy economic model – tidier on paper than in reality. And tidier before Judge Greene addressed it than it would become thereafter.

The proposed settlement was duly presented to Judge Greene. He immediately took charge of the settlement process and revised the terms of the decree. Almost single-handedly, Judge Greene crafted an intricate common law of decree jurisprudence. His courtroom became a shadow FCC, an independent authority that scrutinized, cajoled, modified, hectored, and enforced indefatigably, in pursuit of Judge Greene's own understanding of where antitrust law intersected with the public interest, convenience, and necessity.(142)

Our first edition dealt with these battles at considerable length. This edition still cites many of them, for two reasons. First, it is not possible fully to understand the changes wrought by the 1996 Act without understanding the twelve years of decree jurisprudence that preceded it. Second, the Act itself adopts, expressly and wholesale, very significant segments of decree jurisprudence.

We must also take note of the divestiture that wasn=t. Certainly, no account of the antitrust epic that reshaped telephony in the 1980s would be quite complete without mention of the other great battle between the federal government and private enterprise, fought in parallel during the same years. In 1969, government lawyers begin contemplating the future of the computer business. The future, to them, is plain enough: An IBM monopoly as far as the eye can see. What is to be done?

Release Bell to provide competition? Certainly not. To be sure, half or so of all federal and state regulators embrace the Bell monopoly, and work vigorously to protect it. The other half, however, wish to punish Bell for the sin of monopoly, and short of that, to make sure that it does not fall prey to the sin of gluttony. Bell, in other words, is to be utterly excluded from any role in the computer industry. Almost no one in government recalls that it was Bell's research labs – not IBM's – that discovered the transistor, the key to all modern computers.

So instead of letting another established firm compete against IBM, the government resolves to have IBM compete against itself. The objective: break up IBM. The means: a mammoth antitrust suit. In the following 13 years of litigation this will prove to be one of the slowest, most expensive, paper clogged, and useless antitrust lawsuit ever undertaken; it comes to be known as the Antitrust Division's "Vietnam." The suit is ultimately abandoned on the same day that the Bell divestiture is announced.

In the interim, three engineers will found a new company that will do the job the litigators never complete. The engineers plan to manufacture an altogether novel device called a microprocessor. They will take a familiar device, the transistor, and make it smaller. A hundred million times smaller, in fact, and still growing smaller year by year. The power of the microprocessor grows apace. It is a power destined to shatter not only the computer monopoly but the telephone monopoly too, and thus to make nonsense of much of the FCC and decree jurisprudence that has come before. The new microprocessor company is called Intel.

The Intels of the world were underestimated by regulators and antitrust authorities time and time again from the 1960s until the 1980s. This is somewhat surprising. After all, the regulatory convulsions of the 1960s and 1970s, which culminated in the Bell divestiture of 1984, all flowed directly from discoveries at Bell Labs, most notably the discovery of the transistor in 1947.

The first-generation computers in 1956 were monstrous devices built around huge racks of vacuum tubes. But the new transistor soon came to the notice of Jack Kilby, an engineer who had been designing compact systems for hearing-aid companies. In 1958, Kilby moved to the Dallas headquarters of Texas Instruments and had a brainstorm. By then, transistors were already being made on silicon. Why not make resistors and capacitors too on the same medium, and thus manufacture entire circuits all at once, in one process, on one substrate?(143) Why not, in other words, manufacture an "integrated circuit"? Robert Noyce, another alumnus of Bell Labs then working at Fairchild Semiconductor, soon radically improved on Kilby's design. In 1968, Noyce and a colleague set up their own new company – Intel. And Intel would eventually become master of the microprocessor, the computer on chip, that would transform all of telephony, computing, and broadcast.

The integrated circuit continued the transistor's restructuring of telephony, but accelerated the pace of change a thousand-fold. Transistors were shrunk from the size of a fingernail to the size of a hair, to the size of a microbe and smaller. The economics of producing electronic equipment shifted dramatically. Designing a single, advanced microprocessor may require a billion-dollar investment. Thereafter, any number of copies can be stamped out at very little cost. The technology thus triggered an efflorescence of new desktop and office systems, as well as consumer electronics. All depended on the same fundamental component – the transistor. All operated digitally. All could be mass produced at little cost once the electronics for the first unit had been designed.

The result has been a radical technological transformation, characterized by two, seemingly contradictory trends: fragmentation and convergence.

The first major technological trend today continues to be one of fragmentation. The once centralized network is becoming decentralized. "Terminals" – dumb end points to the network – are giving way to "seminals" – nodes that can process, switch, store, and retrieve information with a power that was once lodged exclusively in a few fortified centers massive switches and mainframe computers. This is the fragmentation.

The second transcendent reality is convergence. Wires and wireless media, for example, remained separate for most of the first century of telecommunications. But it is clear today that wire and wireless define the two halves of a boundless telecosm. From a physicist=s perspective, they are not very different. Radio waves come in an infinite range of frequencies. AM radio is low on the dial. Ruby-red laser light is high. Fiber-optic transmission is just radio on speed: the oscillating crystal of an old-fashioned radio is replaced by an oscillating cavity of gas called a laser. The only real difference between wireless and wire is the Awave guide@ B the wire itself. An ordinary radio transmitter shines outward in all directions B it Abroadcasts.@ An electronic or photonic transmitter points its signal into a shielded tunnel of metal or glass B it Adrills.@

The tunnel solves the problem of interference. Any number of Aradios@ can transmit side by side on wires; when one strand is filled, another can be unrolled next to it. The wire also excludes dust and rain, so the signal is more dependable. A wire can turn corners, dodge buildings, or track the curvature of the earth. And a wire can be more private. The National Security Agency much prefers glass to air.

Does wire, and particularly glass, therefore have a decisive edge over wireless? Not always. Wireless is often more efficient for one-to-many communication, and is essential for mobility. It is less private B but the whole point of Abroadcast@ is to reach lots of people. Overall, wired networks offer more capacity and privacy, but are most efficient only for stationary, Anarrowcast@ connections.

One vivid illustration of this convergence is cellular telephony, made possible by the synthesis of radio, telephone, and computers. The key problem with the early radio telephones, which persisted until the 1980s, was that there just didn't seem to be enough spectrum available to allow simultaneous use of very many of them. A few dozen stations pretty much fill up the dial of a radio – and radio telephone requires radio stations in pairs to sustain two-way conversation.

In the 1940s, researchers at Bell Labs proposed an ingenious solution. Radio telephones should be low-power, short-range devices. The same frequencies could then be used again and again (just as they are with cordless home telephones); a radio conversation on East 42d street would not interfere with another one on the same frequency on West 51st. A city would be divided into many separate "cells," each one served by its own, low-power transmitter. The capacity of a cellular system could then be increased almost indefinitely by shrinking cells and increasing their number. But cellular telephony required, in exchange, highly sophisticated transmitters and receivers, and massive coordination among cells to "hand off" calls and coordinate frequencies as the car phone on 42d street moved toward 51st. No one had the technology to perform this – until the advent of microelectronics.

After the FCC finally approved commercial cellular telephone systems in 1982, the market grew explosively. The new exchanges – "mobile telephone switching offices" – secured the right to interconnect with the established landline exchanges.(144) By 1990 entrepreneurs and regulators were considering a second generation of over-the-air telephone systems – "personal communications services@ (PCS) – based on "microcells," with base stations linked to either private or public exchanges.(145) Each new cluster of exchanges that appeared on the scene opened up new possibilities for service from competing networks. Cellular companies have quickly recognized the advantages of "clustered" service, and established dedicated links between their own exchanges and those of the long distance carriers. PCS operators have turned to cable companies to provide transport among the transceivers that will be used to support their service.

A less visible but equally revolutionary merger of radio and telephone technologies has occurred below ground, during almost exactly the same years as cellular systems were being deployed above. This too evolved directly from technological developments set in motion at Bell Labs several decades earlier.

The development of coaxial cable and microwave transmission marked major advances in the continuing quest for ever more capacious, reliable, secure transmission systems. For telephonic purposes, microwaves represented an important advance over ordinary radio because they operated at much higher frequencies, capable of carrying much more information over focussed paths. Push the frequencies higher still, and you get ultra-high frequency radio waves, better known as light. A light beam can be shaped and modulated to carry information just like Marconi's radio waves, but in vastly larger amounts. Light is best transmitted in a wave guide, similar (in principle) to those developed by Bell Labs in the 1930s. Extremely pure, hair-thin, stands of glass serve admirably.

Fiber optic systems represent today's pinnacle of telecommunications technology, the finest merger (so far) of radio, telephone lines, and electronics. Integrated circuits provide the highly sophisticated transmitters and receivers at each end of the line. A single strand of glass can today transmit thousands of simultaneous telephone conversations,(146) or hundreds of color television signals. With such capacities, demand for fiber has come primarily from the higher levels of the network, where traffic from many callers is consolidated into interexchange trunks. Fiber is now rapidly replacing copper, coaxial cable, and microwave everywhere in the telephone network, except (so far) in the short last stretch to the user's home.

Wireline telephony, wireless telephony, broadcast television, and cable television, now share a single, integrated future B a future of one (or more) switched, digital, broadband networks, networks that combine the broadband carrying capacity of cable television, the digital power and flexibility of computers, and the switched addressability of telephones. In digital systems, a bit is a bit, whether it represents a hiccup in a voice conversation, or the price at which AT&T stock is selling at this particular instant, or a strand of hair in a rerun of I Love Lucy. The lines between media formerly segregated by mode of transmission (radio vs. landline) and function (telephone, cable, broadcast, computer) are disappearing. We are moving toward a myriad of mixed media (radio/landline), integrated (digital), broadband networks, all interconnecting seamlessly to one another.(147)

We are not there yet. Despite the torrents of electronic information cascading into our offices and homes, the telecosm is still young and, in critical respects, unformed.

Along one dimension, television is broad and telephone narrow. Broadcast television delivers high bandwidth B it transmits an enormous amount of data very fast. Telephone lines are a thousand times narrower. Along a second dimension, however, telephone is broad, and television narrow. There are about 150 million telephone lines in the country, but only about 1,500 TV broadcast stations. From this perspective, television is a hundred-thousand times narrower than the telephone network. Television is a one-way technology, connecting the few to the many. Telephone gives everyone both a transmitter and a receiver.

The missing parts of the network create a remarkable financial paradox. The narrow connections, which are consumed too sparingly, are far more lucrative than the broadband ones, which are consumed in profligate excess. In round numbers, the Super Bowl attracts 100 million viewers and $100 million in sponsors= cash. The game lasts four hours. That=s about 25 cents per viewer-hour. While sports enthusiast Jane Doe watches the big game, John prefers to chat with his college roommate on the phone. The connection is still very narrow. Yet, for entertainment that good, John is willing to pay about $12 an hour to his long-distance phone company.

Multiply these numbers across 150 million households and offices, and you arrive at a comparatively small TV entertainment industry and a much larger telephone industry. Though it provides a broadband signal B sound and full-motion color video B and occupies the average American many hours every day, broadcast and cable combined generate revenues of $ 80 billion a year. Though each telephone line is used only about 25 minutes a day, local telephony alone generates about $ 120 billion a year. Long-distance calling generates another $100 billion. In total, then, telephony generates nearly three times the revenues of video.

Broadcast delivers a great deal, but the network behind it is steeply hierarchical. Telephone is narrow but egalitarian. That simple difference accounts for the two-way telephone=s two-for-one revenue edge over one-way television. If we have to choose, thin, two-way connectivity is worth far more than a fat but one-way channel. But we don=t have to choose any more. Modern digital transmission technology offers both.

It will surely come; the only question is how quickly. The technology to supply almost limitless bandwidth is now at hand. Broadband networks already occupy the top tiers of the telephone network, operated by regional and national telephone companies, and the top tiers of the broadcast networks, operated by video carriers. Only the last mile remains to be conquered. And Congress itself undertook that task with the Telecommunications Act of 1996.

As noted, the AT&T consent decree assumed that the local exchange was a natural monopoly and therefore left that monopoly intact. Many state regulators, however, concluded otherwise and began to embrace local competition. By 1996, Congress was ready to finish the job. In February of that year, Congress passed, and President Clinton signed into law, a sweeping reform of U.S. telecommunications regulation. For better or worse, the Telecommunications Act of 1996 will likely be remembered as the most important piece of economic legislation of the twentieth century. It runs some 100 pages of impenetrably dense and convoluted prose. Its overarching objective is to transition the entire industry from regulated monopoly to unregulated competition. Completing the process will take a decade or more, but the transition is already well under way.(148)

The 1996 Act broadly preempts most state laws to the extent that those laws explicitly bar competitive entry.(149) It specifically affirms the right of cable operators to compete head-to-head against local telephone companies.(150) The legal path has been cleared to allow electric utilities and gas companies to build communications networks, too.(151)

But the 1996 Act does not stop there, by clearing away the obstacles to new entry. It also requires incumbent carriers affirmatively to assist the new entrants in three distinct ways. First, the 1996 Act allows a competitor to construct its own network and "interconnect" with the incumbent telephone company "for the transmission and routing" of traffic between the two networks.(152) Such interconnection is crucial to competitive entry because it allows customers on the new network to place calls to, and to receive calls from, customers on the incumbent's network.

Second, Congress required incumbent telcos to offer competitors "access to network elements on an unbundled basis at any technically feasible point" and "in a manner that allows requesting carriers to combine such elements."(153) Congress added this provision because it recognized that some key parts of the incumbents' network would be difficult to reproduce immediately. For instance, Congress understood that cable television companies planned to use their existing door-to-door networks to carry telephone calls as well as TV signals, but lacked the switches needed to direct calls from one customer to another. Conversely, some interexchange companies might have the necessary switches but lack the loops to the customers= premises. By requiring access to parts of the incumbents' network, Congress allowed such companies to enter local markets as facilities-based carriers without reproducing the incumbent's complete network.

Third, the 1996 Act requires incumbent LECs "to offer for resale at wholesale rates any telecommunication service that the carrier provides at retail to subscribers who are not telecommunications carriers."(154) Resale permits entry by potential competitors who have no facilities of their own.

The flip-side of these obligations placed on incumbent LECs is supposed to be a new freedom to compete. The two remaining pillars of traditional regulation B the quarantine, and end-to-end price regulation – are not long for this world. The 1996 Act sets in place an express mechanism for eliminating the long-distance restrictions on Bell Companies. A number of those restrictions B on wireless and out-of-region services, have already been removed. Others are due to sunset. The main ones will be eliminated when state and federal regulators conclude that incumbent local phone companies have opened their networks sufficiently to put competition on a robust trajectory.

As for pricing regulation, the new statute calls for the elimination of all implicit subsidies and their replacement by a new explicit universal system funded in an equitable and non-discriminatory way by all providers of telecommunications services.(155) And Congress called upon the FCC to forbear from regulating, consistent with the public interest, wherever such regulation is not necessary to ensure that charges are just and reasonable or to protect consumers.(156) As competition spreads at all levels of the network, traditional price regulation should be its first victim.(157)

In August 1996, the FCC promulgated rules purporting to implement these requirements; many of those rules are currently under review in the courts.

Open entry is now the norm in local exchange markets, as it is elsewhere in the industry. The old and simple regulatory principle of exclusive franchise has given way to a new and complex set of rules governing interconnection among horizontal local competitors. The principle underlying the 1996 Act, however, is still a simple one: carriers sell carriage, and their obligation to do so does not depend on whether the customer is itself a competing carrier. That principle is over a century old, dating back to (though not affirmed in) the Express Package cases.

This shift from the exclusive franchise to unbundled interconnection fundamentally changes both the process and objectives of federal regulation. In the past, when local exchange carriers were legal monopolies, all the interconnection issues were vertical, not horizontal.(158) Customer premises equipment, long distance services, enhanced services, and wireless services were complements of, not substitutes for, local exchange service. The principal terms and conditions of interconnection overseen by the Commission B such things as access charges – were not those between direct horizontal competitors. Without such horizontal competition, the Commission had little information upon which to ground its regulatory rulings. Despite pious pronouncements about benchmarks, all the monopolies looked pretty much alike.

The principal regulatory issues before the Commission today are of a fundamentally different character: they concern horizontal interconnection not vertical, direct competitive substitutes, not complements.(159) Competing local carriers now interconnect with incumbents and compete directly against them. Hundreds of interconnection agreements have been signed, and hundreds more are being negotiated. The most important Aconsumers@ of local exchange service now include such Acustomers@ as AT&T, Sprint, Frontier, and MCI/WorldCom. This new class of customer-competitor knows a great deal about the local exchange business B they are currently are local exchange carriers themselves. And these customers are being joined by incumbent local carriers too, as those incumbents venture to compete outside their traditional service areas.

With horizontal interconnection taking center stage, the FCC=s primary mission is no longer to set terms, conditions or rates in an informational vacuum. It is to play its role, along with the states, as arbitrator of interconnection among actual, horizontal competitors. Most significantly, the regulation of horizontal interconnection can rely on direct, quantitative measures of performance and success. The Commission has access to quantitative counts of interconnection agreements signed, minutes exchanged, local numbers ported, and customers switched. No comparable measures of regulatory efficacy were available with regard to vertical interconnection.

The 1996 Act also institutes a variety of procedural mechanisms that provide CLECs, and the Commission alike with relevant data.(160) Both CLECs and the Commission have access to the terms of every interconnection agreement that has been negotiated and may use those terms as a basis for comparison. Indeed, requesting telecommunications carriers can do more than that: the 1996 Act requires LECs to "make available any interconnection, service, or network element provided under an agreement approved under ['251] to which it is a party to any other requesting telecommunications carrier upon the same terms and conditions as those provided in the agreement.@(161) Thus, the Commission and consumers have at their disposal an abundance of new, publicly filed information on which to base their assessment of what types of interconnection are fair and reasonable.

As horizontal competition increases B and it is increasing rapidly – the importance of many traditional types of Commission regulation will diminish apace. For example, the Commission has previously used benchmarks to help determine access charges and productivity factors.(162) As the number of horizontal competitors multiplies, the importance of access charges will decline, simply because long-distance carriers will increasingly reach their customers through CLECs rather than ILECs. As Internet and enhanced-service traffic continues to riseB and it is rising very fast B the importance of access charge rates will decline further, because those services do not pay access charges at all.(163) And regulation will be entirely irrelevant to a fast-growing number of competitive services, including high-speed access services offered by incumbent local carriers through separate subsidiaries, because they will not be regulated at all.

Since divestiture, there has been a steady, inexorable rise in competition in all levels of the industry. MCI and Sprint developed into full-fledged national carriers to compete with AT&T; other smaller carriers built regional networks, and hundreds of resellers entered the market, too. Unexpected though it was by the framers of the decree, competition began to emerge in local markets too, particularly in the business of providing Aexchange access service@ i.e., the local transport of calls to the nearest Apoint of presence@ of a long-distance carrier. In 1984, Teleport Communications Group (TCG) was formed; the company began as a partnership between Western Union, Merrill Lynch, and the Port Authority of New York and New Jersey, and was later acquired by cable companies Cox, TCI, Continental, and Comcast. (TCG was purchased by AT&T in 1998.) Teleport and other competitive access providers (ACAPs@) began building fiber-optic networks designed to provide exchange access for high-volume customers. These networks were typically built as rings that snake through metropolitan areas, passing the major office buildings where large businesses typically locate.

This process has accelerated rapidly since 1996 B wherever competition makes strategic and economic sense for the new entrants. That competitive sphere includes business services of all kinds: short-haul toll services, wireless services, many data services, and other enhanced services. As of late 1998, nearly 300 companies were providing competitive local exchange carrier service of some description – companies like MCI WorldCom and AT&T/TCG, cable companies, interexchange carriers, providers of personal communications services (PCS), providers of shared tenant services (e.g., service to apartment buildings), and others. Well over 2000 interconnection agreements had been reached. New capital investments by competitors had surpassed new capital investment by the incumbents. Competitive local carriers had installed over 1000 switches, and were deploying new switches much faster than the incumbents.

Since passage of the 1996 Act, cable operators have begun offering data services to a rapidly growing number of customers in this high-growth segment of the market. Once customers get used to relying on cable for high speed data services, they may migrate more readily to cable voice services.

Wireless markets are likewise growing very rapidly. By late 1998, providers of wireless PCS had concluded negotiations and signed close to 200 interconnection agreements with incumbent wireline carriers. PCS providers have launched commercial service in markets that serve half of the U.S. population. Wireless prices are falling and the time is fast approaching when PCS will be seen as a genuine substitute for wireline service.

Today, all networks are going digital, and they are getting there fast. Not just wireline telephone networks B wireless networks, cable, and broadcast too. Some carriers are upgrading or replacing existing facilities with new, digital components. Others are entering markets with digital technology from the ground up. All are adding bandwidth (to support voice and video), adding two-way capabilities (to support real-time voice) and adding error correction (to support data).

Telephony. The digitization of the telephone network has been underway for over 20 years. Almost all long-distance transmissions are already digital. The local network is now catching up. Local digital switching has become ubiquitous in the last decade. Even existing copper plant is being upgraded to provide digital circuits. ISDN is already widely available, and prices have dropped significantly.(164) More recently, phone companies have begun deploying Digital Subscriber Line (xDSL) technologies, which supports simultaneous, digital transmission of voice and video over existing copper plant.(165)

Cable. Fiber optics, signal compression technology, and other digital technologies are rapidly transforming cable television into cable telecommunications. Many cable operators are past the halfway point in upgrading their networks to a digital HFC architecture that is capable of delivering telephone and other advanced information and two-way services. Cable operators have already begun to offer cable modems to exploit their bandwidth and new interactive capabilities. According to some estimates, one third of all Internet users will be accessing the Internet over cable networks by 2002. Providers of Awireless cable@ – like MMDS(166) and LMDS(167) – will use recently-obtained regulatory freedoms to upgrade to two-way digital services.

Wireless. Wireless media are going digital from the top down and the bottom up. All-digital PCS networks are now being built in virtually every major metropolitan area in the U.S.(168) Existing cellular systems, which have traditionally used analog technology, are simultaneously being converted to digital radio. Nextel has also converted much of its SMR spectrum into an all-digital, nationwide wireless network.

Broadcast. Over the next decade, terrestrial analog television will give way to digital Advanced Television. The affiliates of the big four networks went digital in the top 10 markets on November 1, 1998.(169) Broadcasters have already announced aggressive budgets for digital conversion. Digital satellite television has been available since 1994. Direct broadcast satellite is now making the transition from wholesale to retail markets, not only for broadcast services but for Internet/data services, too. Hughes Electronics offers high-speed Internet access via DBS satellite nationwide.

Satellites. Beyond DBS, several companies are now making enormous investments to deploy global digital satellite networks. Iridium has launched a 66 satellite-network and began providing worldwide commercial service in late 1998. Teledesic, the $9 billion joint venture between Craig McCaw and Bill Gates, expects to be up and running by 2002. Numerous other satellite ventures have also been formed. This is the AOrbiting Internet@ or AFiber in the Sky.@

Digital networks are very much more flexible and powerful than the analog networks they are replacing. They offer more bandwidth. They more readily support two-way capabilities. And with encryption, they support addressing and narrowcasting, which can erase traditional distinctions between Abroadcast@ and Atelecom@ services.

Bandwidth. Digitization alone will sharply increase the effective bandwidth of existing networks. Digital signals can be much more efficiently multiplexed than analog signals. ISDN boosts the capacity of copper wire by a factor of 4. Digital signals can also be compressed. Advanced compression, multiplexing, and modulation can condition analog copper wire to carry multiple megabits per second of data, or full-motion video, or multiple voice calls. ADSL, for example, increases the effective capacity of copper wire by a factor of 350. DBS operators already use 6 MHZ bands (what analog broadcasters use to transmit one channel) to deliver six or more video channels. Terrestrial broadcasters will deliver high-definition digital television within the same 6 MHZ bands, with space left over for data services. Digital cable plant can support the equivalent of 750 MHZ, (or even 1 GHZ) of bandwidth, transmitting over 100 video channels and high-speed two-way cable modem services. Digital video compression can increase the bandwidth of existing coaxial or copper networks by a factor of between 10 and 24. Digital cellular technologies increase the capacity of existing cells and spectrum between three and ten times current levels.

Two-way Capabilities. As the bandwidth increases, addressing and filtering become essential. Addressability alone is often a close substitute for two-way capability, when the return path is used only to lodge a request for information. A credit card can be verified in response to a request initiated over a two-way phone line; alternatively, it can be verified by continuously transmitting an encoded, suitably addressed list of invalid cards to a storage device at the point of sale.

Digital technology also makes it easy to plug together two-way services by interconnecting hybrid networks. That is often the best way to proceed when upstream and downstream bandwidth requirements are highly asymmetric. Cable or satellite, for example, can provide a downlink, with telephone lines or narrowband terrestrial wireless closing the loop. The abundance of cheap digital processing makes it is easy to combine different media in this way.

Two-way capabilities, or at least addressability, can also be added with encryption and scrambling devices. Sophisticated coding algorithms that scramble what is sent can be readily combined with terminal equipment smart enough to decode it. The descramblers installed at the end of the cable television lines perform such functions today. So do numerous types of communications software located in personal computers. With enough encoding of this type it is possible to add two-way capabilities to any type of digital bandwidth, and convert it for any use. For example, television and cellular telephony both use broadcast technology: a pocket cellular phone, like a TV station, transmits outward in all directions. A television station uses 6 MHZ; a cellular telephone network uses 25 MHZ. Combine four TV licenses and encode the signals, and you have the makings of a cellular network.

In sum, digital technology and high bandwidth obliterate engineering and economic distinctions between different type of electronic communication. Not simply because Aa bit is a bit.@ The key point is that high bandwidth digital networks, both wired and wireless, are extremely flexible. They can readily be configured and interconnected to mimic any of the capabilities of any of the old, analog systems.

And it is on digital, data networks that all traffic, of every description, is going to move. If fax is included as Adata,@ as it should be, data traffic probably already equals or exceeds voice traffic on phone networks. And the volume of data traffic is growing much faster than voice, even on wireless phone networks. Much of this increase can be explained by the increase in the number of people using the Internet, and the related rise in second residential telephone lines and computer modems. Digital video traveling over phone lines will tilt the balance toward data further still. With this new universe of data telecom, services will compete in ways that are unexpected and largely unfamiliar today.

The Internet (together with the World Wide Web, which operates as a set of software protocols on it) is probably the best model at hand today for the broadband architecture of the future. It is already the world=s largest communications system, the data archipelago of the planet. It spans the national, regional, and local communications networks used by commercial, government, and educational organizations worldwide. It is also, at the same time, an array of independently owned and managed networks, separately funded, separately developed, and separately maintained, with no single entity or agency in charge of managing everything.

The Internet has no defined structure B it is best described as a network of networks. At the physical level, it is a network of millions of computers, joined together by wires and radios of varying bandwidth.(170) At the virtual level, it is a common set of protocols: the ATransmission Control Program/Internetworking Protocol@ (TCP/IP), and HTTP B the code of the World Wide Web. Computers, modems, and set-top boxes surround the Internet at its periphery. At the far end: the Ahost@ B another computer that processes requests from its clients, whenever they come a-calling. Providing the links in between, are several thousand AInternet Service Providers,@ and a handful of major backbone providers. For a variety of technical reasons, Internet-based voice telephony is not yet convenient, or of very good quality, but that is changing, and rapidly. A number of carriers are already offering phone to phone service domestically including AT&T, MCI, Sprint and a number of startups including Qwest and Level 3. AT&T has announced plans to deploy Internet Protocol telephony for local calling on TCI and TCI-affiliated cable systems.

A broadband Internet will offer decentralized and bandwidth and non-hierarchical connectivity, but not simplicity or uniformity. It will be an inelegant mix of glass, coax, and copper, of terrestrial transmitters and satellites. No one technology will prevail. Bandwidth and two-way capabilities will constantly improve. The technological mix will constantly change. Bandwidth and connectivity will proliferate, but so will complexity and unpredictability. It has been suggested that the airwaves and the landline network are trading places B that everything that now goes by wire will go by air, and vice versa. But what is really going on is a hybridization of technology, not a simple exchange of chairs. Both wireless and landline technologies can provide broadband links; the choice will therefore depend on such factors as the need for mobility, population densities (which sharply affect the relative costs of wire and wireless), and overall network economics. The market will use broadband media in unexpected ways, if regulators will let it find them. Unforeseen synergies between different broadband media B satellite and cable, for example B will create entirely new markets and wholly new categories of services. They will also utterly destroy traditional regulatory boundaries between the various media.(171)

Under the traditional regulatory paradigm, all customers were treated equally. Universal service required that all be served and non-discrimination principles precluded favoritism. In the new marketplace, however, customers are sharply segmented. They fall into four basic groups: Multinationals, Small and Medium-Sized Businesses, Boomers, and Aunt Tillie.

Multinationals. Approximately one third of incumbent phone company revenues comes from 1 percent of the biggest customers B multinational business customers. They demand one-stop suppliers of state-of-the-art service. They buy more data traffic than voice. Within 10 years, 90 percent of their traffic will be data, only 10 percent voice. Local competitors like WorldCom are going after this market almost exclusively. And they are winning a big share of it. In the first quarter of 1998, for the first time, local competitors added more new business lines than were added by Regional Bells.

Small and Mid-Size Businesses. These customers are not anchor tenants. No one will build a network out to them. But they will be picked up in increasing numbers by competitors who have built their networks to serve large businesses and who have capacity to spare. The fiber rings through downtown areas can serve the flower shop on the ground floor along with the multinational in the penthouse and the various law and accounting firms in between.

Boomers. One residential customer in six spends over $100 a month on telecom services. These are affluent customers. They travel a lot. They use the Internet a lot. They want to buy bundled service. They want one-stop shopping. For this group of customers, data traffic will soon exceed voice. Ten years from now, these customers will be using five to ten times as much service B mostly data service B as they do today. These customers are heavy users of wireless, too. Wireless carriers are adding new subscribers twice as fast as wireline carriers are adding new lines. (Wireless carriers added 13 million new subscribers in 1997.)

Aunt Tillie. About five out of every six residential subscribers. They live in suburbs, or smaller towns, or rural areas. They are still using phone service pretty much the same way they did 20 years ago. They spend $10 to $30 a month on phone service. They want Wal-Mart and McDonald=s phone service. Most still look to AT&T for their long-distance service. And to a Bell Company for their local service. They will buy data and wireless services too B but only when prices drop, services get easier to use, and familiar, brand-name companies offer plug-and-play packages. WorldCom won=t touch them. MCI is transitioning away from serving them. Sprint won=t target them. AT&T doesn=t want them, not their local business. They are still served almost exclusively by traditional phone companies. Only by the lowest cost provider, the one that successfully captures economies of scope and scale, can afford to serve them. They will buy only from a very trusted, brand name; they view telecommunications as too important to entrust to unknown upstarts. They want reliable, familiar service. That=s why most of them never switched from AT&T in their long-distance service.

To serve the most desirable of these customers, companies will increasingly offer packages of bundled services. The breakup of Bell in 1984 separated telephone equipment from local service, and local from long-distance. The post-1996 competitors will bring them all back together.

Equipment is already routinely re-combined with service in many markets. Cellular companies give away phones; cable companies supply converters; modem manufacturers bundle in start-up subscriptions to on-line services. Phone companies will be fully back in the same game soon enough.

The companies are reuniting geographically as well. Long-distance carriers like AT&T and MCI are already free to build their networks down to the consumer. They are doing so as fast as they can. Soon, local phone companies will win permission to build long-distance facilities, or affiliate with existing long-distance carriers, or resell their services. They will do so at once.

The divisions between local and long-distance service have already all but disappeared in data communications. Internet connections cost the same, whether they are used to move data across the street or around the globe. Indeed, on the Internet, it is almost impossible even to track where calls originate, what communications paths are traveled, or where information is ultimately delivered or consumed. Content is readily hybridized; a display on a single computer can be synthesized out of data residing on, and delivered from, any number of others, located anywhere on the globe. Divisions between local and long- distance wireless service are fading, too. GTE and Sprint have been bundling local cellular service with long distance service for some years. By the turn of the century, all wireless carriers will sell one-brand, one-bill, one-stop, integrated wireless service.

With voice and data, wire and wireless, the new, re-integrated phone companies won=t offer their customers Aequal access@ to other vendors. No principle of common carriage requires that they should, so long as they sell what they sell to anyone who wants to buy it, other vendors included. The whole marketing pitch will center on just the opposite: the simplicity and reliability of shopping from a single, name-brand provider, with nothing to pick but the name.

AT&T will exploit its strong brand identity as aggressively as anyone, and will prosper by doing so. A significant fraction of consumers will buy their phone service B all of it, end to end B from AT&T. Just as they did in 1980. But many others won=t. That will be the difference.

Year by year, we can expect to see further erosion of the boundaries that traditionally segmented the industry. It is no longer useful to think of broadband pipes as ordinary wires. They are bundles of virtual channels, carrying clusters of constantly evolving services. With the triumph of digital technology, the old divisions between voice, video and data services disappear once and for all. A full-bore digital, broadband network will empower everyone to cast broadly or narrowly at will. The on-line world already does so: anyone can create a bulletin board, and anyone can post musings on it, whether wise, foolish, tasteful, crass, or crude, to be read by the world. With digital broadband technology, the text-based bulletin boards will support voice and video. Broadband carriers will have room to carry anything anywhere: on their networks, anyone who chooses will be an instant broadcaster. Tomorrow=s broadcasters will have the power to narrow and address their signals at will: on their networks, anyone who chooses will be an instant carrier.

Pure carriage can certainly survive as a business in a digital, broadband world, and undoubtedly will. But who will provide pure carriage, on which assemblies of wires and radios, is much less clear. The owner of a UHF station licensed to broadcast sitcom re-runs twenty-four hours a day might do far better using the same spectrum to transmit digital wireless e-mail, a common-carrier service.(172) In rural areas, wireless telephony may be both cheaper and much more valuable than wired; the best use of existing phone wires may be for a next-generation of digital video-on-demand.

A few die-hard providers may decide to keep their businesses strictly on one side of the old definitional line or the other, or be forced to do so by reactionary regulators. For the most part, however, the clean, familiar legal divisions between carriers and broadcasters will be impossible to maintain. In the broadband telecosm, Acarriers@ and Acasters@ will compete head-to-head in a single unbounded arena.

With bundling and an increasing focus on the needs of large customers, mergers are inevitable. After two decades of fragmentation, the center of the industry is now entering a period of rapid consolidation. In both U.S. and international markets, the telecom industry is consolidating, both horizontally and vertically, through mergers and alliances. Measured by revenues earned, approximately half of the U.S. telecom industry was involved in a merger or acquisition in 1997 and 1998, including Bell Atlantic/NYNEX/GTE, WorldCom/MFS/Brooks Fiber/MCI, AT&T/TCG, and SBC/Ameritech. Almost half (8 million) of the 20 million phone lines served by independent phone companies have been transferred by merger or acquisition in the past decade.

These trends reflect a fundamental economic fact B so fundamental, indeed, that it used to considered immutable, and was often written codified as the law of exclusive franchise. Network industries are characterized by powerful economies of scale and scope. Scale and scope are critical factors in purchasing and deploying new technologies and services. Large buyers of equipment are able to negotiate large discounts with hardware and software vendors, such as Nortel and Lucent. Purchases of bulk services, like wholesale interexchange traffic or Internet backbone access, also become much cheaper with scale. Large providers of service can distribute the costs of funding or soliciting bids for the development of new technology over an extended base of operations. Size likewise diminishes the costs and risks of developing new services. Scale eliminates many duplicative general and administrative costs, providing selling and maintenance efficiencies. Geographic scale and scope are equally important to national and multinational customers. Large companies like AT&T and MCI/WorldCom can plausibly bid to serve a large customer=s telecom needs around the world.

Small wonder, then, that the major players in the industry are very large. Collectively, Bell Companies represent 44 percent of the nation=s telecom revenues, and a still larger percentage of the nation=s telecom workers. The larger regional Bells (like Bell Atlantic and SBC) own about $60 billion in capital plant, and invest $6 billion a year in new capital plant. AT&T already serves 100 million phone lines. Only MCI/WorldCom (28 million lines) and Sprint (12 million lines) can currently lay similar claim to national identities, with established customer bases in every major market and state.

This is not to say that niche players will not still thrive. Indeed, niche players were the first to prosper in the emerging competitive environment. They offered differentiated, specialty services to only a select, high-profit segment of the market. The 1996 Act=s resale and unbundling requirements have significantly reduced entry barriers, and newcomers have no responsibility (at least none comparable to that of incumbents) to offer universal service. These competitors are quite rationally focusing all their competitive energies on the very largest business customers, while ignoring less profitable residential customers. Their competitive strategy is defined by how selectively they choose their customers, and how few customers they actually serve. They leave the mass market, particularly the low-margin residential market, to others.

But, while many companies can and will be effective regional or niche competitors, only a small number of companies can realistically hope to be full service companies, offering a local, long-distance, and international service bundle nationwide and beyond. The financial challenge is daunting. The capital costs for facilities-based local competition in the United States generally range from several hundred to several thousand dollars per customer, depending on demographics, services provided, and facilities deployed. Customer acquisition costs run several hundred dollars more. Incumbent local phone companies serve 150 million customers and have already sunk capital investments in excess of $200 billion in the United States.

Thus, even as the niche players multiply and prosper, the center of the industry is restructuring, and the big are indeed getting bigger. But the competitive implications of mergers, acquisitions, and sheer size must be assessed in the context of the equally profound transformation of geographic and product markets. In the old order, each provider was confined to its own geographic territory, and shielded from competition by an exclusive franchise. There are no exclusive geographic territories in the new: Every player may compete everywhere. In the old order, each provider was confined to a particular category of service: local, long-distance, wireless, or data. Those lines are all collapsing in the new: Every player may provide complete bundles of service. AT&T, MCI/WorldCom, Sprint, and a host of foreign carriers are all, quite clearly, on a similar competitive trajectory, transforming themselves into national and global providers of integrated telecom services.

Consolidation trends notwithstanding, competition is still increasing. The new industry order now emerging is one of four to six well-financed, full-service, national (eventually global) companies, competing head to head to provide a full bundle of services in every major market. Numerous smaller regional players and providers of specialized services will continue to compete against the big nameplates by offering specialized and differentiated services. The upshot will be fewer national or global nameplates, more niche competitors, but B at the same time B far more competition all around.

The global market for all telecommunications services was approximately $700 billion in 1996, and has been growing an estimated 20 percent per year. Traditional local and long-distance voice services account for 60 percent of the total. International services accounted for 10 percent of the total. North America accounts for 28 percent of all worldwide access lines and 33 percent of worldwide revenue. Europe accounts for 38 percent of access lines and 33 percent of revenue. Asia accounts for 26 percent of access lines and 26 percent of revenue.

Global markets are characterized by three major trends: Privatization, Open Entry, and Interconnection. But most of the world (with a few exceptions, e.g., Great Britain, New Zealand) is one to two decades behind the United States in implementing these initiatives. Most countries have separated wireline from wireless services and permit wireless competition. But no other country has followed the United States model of divesting the local from the domestic long-distance operations of their incumbent wireline carrier. And in many countries, the incumbent telephone service provider also provides cable service.

Most major foreign carriers have expanded their operations globally, through alliances with, investments in, or acquisitions of foreign carriers. All major U.S. long-distance carriers are pursuing comparable global strategies: forging multinational alliances with major foreign incumbents and investing in wireless and data providers in foreign countries. Virtually all major wireline incumbents have acquired foreign interests. As of this writing, there are three major global alliances, too.(173)

The competitive struggles discussed in this book are not abstract lessons in economics. They have a very real effect on American competitiveness and the quality of American life. The "information" sector of the economy makes up a large percentage of GNP and of the workforce. Even companies outside the information arena are allocating a steadily increasing fraction of their budgets to telecommunications. Since most unit costs of telecommunications are dropping rapidly, even a slowly growing telecommunications budget represents a much more rapid growth in telecommunications usage. By all indications, these number will continue to grow rapidly.

And by all indications, the supply of bandwidth will grow to meet this demand. As it does, free markets will become increasingly difficult to repress. The genius of a market is that it elicits information about what people have and what they want. That information, however, becomes powerful only when it is communicated to others. The invisible hand has no power unless guided by visible eyes and ears. The expanding universe of high-bandwidth telecommunications will allow communication as never before.

What, after all, are the essential ingredients of a market? Communication, to connect together the willing buyer and the willing seller. Promises, so that trades begun today can be consummated tomorrow. Memory, so that promises will be kept. Memory, to record what belongs to whom. Promises again, to create all other rights beyond property rights, because all social norms depend on a shared commitment to enforce them. Promises again, by which honest traders agree to ostracize cheats, deadbeats, and thieves. Promises and memories, trust and loyalty: These are the essentials upon which all else in the marketplace is constructed. And broadband is communication at its most capacious and powerful, with the wires and the ether terminating at electronic scribes, records, and memories, a laser-light weaver of trust and loyalty.

The giant trusts and corporations that evolved in the early days of capitalism will not reappear. The old corporation operated in the image of Big Brother, as a homogeneous, collectivist autocracy, dominated by a single, all-powerful leader. Such structures cannot survive. Independent yet tightly interconnected business groups, linked by network yet disciplined in all their relationships by market forces, will replace them. In this world, people will be paid because they work, not because they show up daily on the factory floor or in some glass-walled office. Services delivered over the new network will be metered with absolute precision. Specialists will specialize as never before.

Items supplied over the network – most of all labor – will be valued with scrupulous accuracy. People who really produce will be in high demand, and no one will care at all about their race or religion, their sex, their social graces, their physical appearance, or how they smell. Quality services will be purchased wherever they can be found, across the street or across the ocean. Inferior services will be priced accordingly, and incompetence will not be purchased at all, however much employment commissions may protest. No government agencies will even be able to keep track of what is being supplied where, still less to dictate who should be employed or on what terms.

The factory will be irrevocably changed. A car manufacturer will become a truly efficient assembler of parts provided by hundreds of independent suppliers. Secretaries, accountants, designers – most of the enterprise's support services – will be replaced by independent outsiders, knitted together into an efficient whole by the network. Suppliers, assemblers, distributors, and customers will be coordinated with meticulous precision. Even the largest factory will operate with no inventory, no warehouses, no fitful starts and stops caused by shortage of supply or excess of output. A customer's order will be conveyed instantly up the chain of production to the assembly line, and back further still to the factory's suppliers of paint, tires, and radios, and hence back to their suppliers of rubber and steel. Industrialism no longer requires collectivism. Cooperation will be by consent.

Marketing will be transformed beyond recognition. In primitive societies, the market operates with tiny stalls and without reliable currency; trading extends only as far as goods can be carried. In a barter economy, the seller exchanges the pig directly for a dozen chickens. Payment is assured, but the process is terribly cumbersome. In societies that are stable enough to issue currency and imbue it with value, money marks a great advance. It records value in a standardized form that is widely understood; it conveys value without any need to transport pigs or chickens. The paper itself is worthless, except for the information it conveys. Money, then, is just another network – a system of communication, a record of past effort and a promise of future return. With money, the record is on paper, a primitive, inefficient, and vulnerable medium of communication. And a single, master record keeper, the government treasury, with a single, centralized printing press, has absolute power to determine value.

But there is ultimately only one valuable currency: the currency of reputation, of stable, honest, reliable loyalty. Once it is established of a man, or a leader, or a nation's central banker that his word is his bond, he can issue currency at will. Once it is established that he is a chiseler, a deadbeat, or a thief, no amount of currency will do him much good, for his paper will be shunned wherever he tries to peddle it. The value of money thus depends on trust and promises among the people who control the records. If the network is powerful enough, nobody controls the records, or at least no central authority does. Trust begins between individuals; then it coalesces among larger groups; then it coalesces in larger groups still. Today, we put our trust in a thousand different private currencies in varying degrees: personal checks, stock certificates, bonds, credit card slips, futures contracts, green stamps, and patronage accounts of every kind. The network enables us to confirm that an account has assets, that a business is functioning, or that an individual faithfully pays his bills. Private enterprises issue countless private currencies by verifying credit, clearing checks, evaluating investments, dealing in futures, and insuring risks.

As the power of the network increases, so too will the power and reliability of private currency. With perfect communication, the government bank is no longer needed – currencies of every imaginable description can be created by the market itself, like all other goods. Virtually every kind of good can become the equivalent of a banknote, available for inspection, conveyance, and storage at any distance.

For better or for worse – mostly for better – telecommunications will transform the face of democracy. Opinion polls, call-in shows, and electronic town meetings usurp the "mediating function" of people like Sam Donaldson and George Will. The middleman in the dialogue of democratic government is cut out of the process, in much the same way as Sears is cut out by mail-order catalogues and 800 numbers. Governments, it has been noted, derive their just powers from the consent of the governed. Now we have at hand technologies for securing consent in ways never before imagined. Consent can now be sought directly, quickly, and efficiently, on issues both large and small. The man on the couch can have precisely the same chance as Dan Rather to raise his hand and say something rude when the President steps into the room. The machines behind the network can tally national opinion not only instantaneously but also with a level of delicate precision that would have left Jefferson breathless. Power will move into constantly shifting communities of shopkeepers, housewives, Yale bulldogs, fruit-juice thinkers, nudists, sandal-wearers, sex-maniacs, Quakers, Nature-Cure quacks, pacifists, and phesbian leminists. The networked society will be shaped by the accumulation of individual decisions to meet or stay apart, to buy or sell, to speak or remain silent.

Until recently, freedom of the press still belonged only to the few who owned one. But the Internet and other broadband technologies give a voice to the average man, the man who has never before owned a printing press or a broadcast station. By giving people the power to speak freely, such technologies also give them the right. No law will be able to repeal the technologies of freedom.

Ignorance – or at least rigid class ignorance, which endures generation after generation – cannot survive this technological transformation. People will be able to educate themselves and their children, if they wish to do so. The state may continue to own the public schools – the buildings themselves – but broadband can create a school out of any pair of desks. The great universities, the great libraries, will no longer be in places; they will reside in cyberspace, securely out of reach of torches and bonfires. Heresy can no longer be eradicated by fire. Heresy is now fire itself, pulses of light in a network of glass.

In today's world, the entertainment industries still pitch their programs at what the great mass of people share in common – their prurience, their neurotic fears – and still rely on crude polling to estimate how many people were sufficiently attracted by sex and violence to endure advertisements for sugared cereal and laundry soap. Much of this will disappear. Every hobby and pastime – cage-birds, fret-work, carpentry, bees, carrier-pigeons, home conjuring, philately, chess – will have at least one channel devoted to it, and often several. Gardening and livestock-keeping will have at least a score between them. Then there will be the sporting channels, the radio channels, the children's cartoons, the large range of channels devoted to the movies and all more or less exploiting women's legs, the various trade channels, the soap-opera channels, the needlework channels, and countless others. Ours will become, once again, a nation of flower-lovers and stamp-collectors, pigeon-fanciers, amateur carpenters, coupon-snippers, darts-players, and crossword-puzzle fans. The days of mindless broadcast to the mindless masses are at an end.

Amid all this freedom there will still be prurience, violence, necrophilic reveries, and the repulsive art of Salvador Dali. Yet in the broadband telecosm, speech in public spaces no longer needs to be regulated. Public spaces – or at least the ones of any importance – are no longer surrounded by walls or gates. Bulletin boards, auditoriums, theaters, schools, stadia, squares, subway walls – electronic replacements for all the traditional public fora – can be created upon demand. The network gives the pamphleteer and soap-box orator not just a place in Speaker's Corner but the whole of Hyde Park. There is room for the pacifist, the Communist, the anarchist, the Jehovah's Witness, temperance reformers, Communists, Trotskyists, Freethinkers, vegetarians, and any number of plain lunatics: All will be able to speak out over the network, all will receive a good-humored hearing from anyone who chooses to listen. The network will be Alsatia, where no opinions are outlawed. There has never been any place like it before in the physical world.

The network will supply room enough for every sight and sound, every thought and expression that any human mind will ever wish to communicate. It will make possible a wildness of spirit, where young minds can wander in adventurous, irresponsible, ungenteel ways. It will contain not innocence but a sort of naive gaiety, a buoyant, carefree feeling, filled with confidence in the future and an unquenchable sense of freedom and opportunity. It will be capitalist civilization at its best. It will be the new frontier, and also the last frontier, for it extends as far as any human mind may wish to range.

The regulatory debates will not disappear in this new world, but they can be expected to shift to a higher plane. Close government regulation of the electronic media has historically been justified on the theory that over-the-air channels were scarce,(174) and that landline media were monopolies. The differences between the two media were also thought to justify fundamentally different approaches to the twin problems of privacy and free speech. But the media are rapidly converging, and scarcity has been abolished. The old regulatory preoccupation with segregation and scarcity will soon be history; the debate will have turned to more important matters. For beyond the quarantine, beyond the exclusive franchise, beyond pricing regulation, beyond even the endless quarrels about state and federal regulatory jurisdiction lie the two great issues of the information age: free speech, and its mirror image, privacy – the right to telecommunicate, one might say, and the right not to.

Endnotes:

40. "If a ferry is erected on a river, so near another ancient ferry as to draw away its custom, it is a nuisance to the owner of the old one. For where there is a ferry by prescription, the owner is bound to keep it always in repair and readiness for the ease of all the king's subjects; otherwise he may be grievously amerced: it would be therefore extremely hard if a new ferry were suffered to share his profits, which does not also share his burden." 3 W. Blackstone, Commentaries 219 (1768).
41. See § 3.11; see also Thorne, Huber & Kellogg, Federal Broadband Law § 5.3 (1995) (FBL).
42. "A telegraph company represents the public when applying to another telegraph company for service," a New York court would later declare, "and no discrimination can be made by either against the other, but each must render to the other the same services it renders to the rest of the community under the same conditions." People ex rel Western Union Tel. Co. v. Public Serv. Comm’n of New York, 230 N.Y. 95, 129 N.E. 220, 12 A.L.R. 960 (1920).
43. Express Packages Cases, 117 U.S. 1 (1885).
44. Pacific Tel. & Tel. Co. v. Anderson, 196 F. 699 (E.D. Washington 1912).
45. Id. at 703 (citing State v. Cadwallader, 172 Ind. 619, 87 N.E. 644, 89 N.E. 319 (1909)).
46. Home Tel. Co. v. People's Tel. & Tel. Co., 141 S.W. 845, 848 (1911). See also Home Tel. Co. v. Sarcoxie Light & Tel. Co., 236 Mo. 114, 139 S.W. 108, 113 (1911).
47. Commerce Court (Mann-Elkins) Act, ch. 309, §7, 36 Stat. 539 (1910). The Mann-Elkins Act is discussed in § 3.2.2.
48. Essential Communications v. AT&T, 610 F.2d 1114, 1118 (3d Cir. 1979) (citing Mann-Elkins Act §§ 7, 12).
49. Standard Oil Co. of N.J. v. United States, 221 U.S. 1 (1911).
50. See § 4.1.
51. See §§ 4.2-4.4.3.
52. See Radio Corp. of Am. v. Lord, 28 F.2d 257 (3d Cir. 1928).
53. Fifteen years later NBC was finally forced to divest one of its two national networks into what became the American Broadcasting Company. See National Broad. Co. v. United States, 319 U.S. 190 (1943).
54. The American Newspaper Publishers Association then launched a vehement attack against radio in Congress and before the Federal Radio Commission. Radio and press representatives negotiated a cease-fire in 1933. The agreement called for dissolution of network radio newsgathering operations, restrictions limiting newscasters to covering "background matter," and restrictions on news broadcasting. The radio would carry no more than two five-minute news broadcasts (after 9:30 am and 9:00 pm so they wouldn't interfere with reading the newspaper). Each news item was limited to thirty words, and the broadcast was to conclude with the language, " See your daily newspaper for further details." See D. Doeden, The Press-Radio War: A Historical Analysis of Press-Radio Competition 1920-1940, at 130-31 (Ph.D. dissertation, Northwestern U., 1975); The Week, The New Republic, Jan. 3, 1934, at 209.
55. See Radio Act of 1927, 44 Stat. 1162 (1927).
56. See id. § 17. See also FBL § 4.6.2.
57. See de Sola Pool, Technologies of Freedom at 136-38.
58. The hostility against any taint of common carrier operations of radio has survived to this day. The FCC's rules against "time brokering" of radio air time were invoked not long ago against the Cosmopolitan Broadcasting Corporation, which operated an FM radio station, WHBI, licensed to serve Newark, New Jersey, but effectively in competition with the many radio stations in New York City. WHBI failed to make a profit with "black-oriented" broadcast, so in 1969 it began selling most of its air time in large units to "time brokers." These brokers resold the time in smaller units, mostly to foreign language and ethnic specialty programmers, but also to individuals in five minute segments between 3:00 and 6:00 a.m. As a result, WHBI ended up broadcasting material in Spanish, Italian, Greek, Hungarian, Arabic, Polish, Brazilian, Portuguese, Lithuanian, Slovakian, Croatian, Albanian, Ukranian, Rumanian, Armenian, Yugoslavian, Bulgarian, and Norwegian; by 1972, Korean, Macedonian, Urdu, Hindi, Bengali, Japanese, and Russian had been added to the list. Everyone seemed happy about the arrangement except the FCC, which promptly terminated Cosmopolitan's broadcasting license. See Cosmopolitan Broad. Corp. v. FCC, 581 F.2d 917 (D.C. Cir. 1978), cert. denied, 454 U.S. 1143 (1982).
59. See § 3.2.5.
60. 47 U.S.C. §§ 201(a), 201(b).
61. Id. § 214(a).
62. Id. § 201(a).
63. See § 2.1.2.
64. See § 2.2.1.
65. See § 6.2.
66. See 47 U.S.C. § 151.
67. See §§ 8.4, 9.4.
68. See § 2.3.
69. See § 2.3.3.3.
70. See § 8.4.1.
71. The rule of thumb among many state regulators, it has been said, is the Pizza Test: The monthly fee for basic residential phone service must not under any circumstances be permitted to exceed the price of a pizza – medium size, with two ingredients.
72. See FBL Chapter 7.
73. See 47 U.S.C. § 153(47). See also Domestic Public Radio Service, 76 FCC2d 273, 281 ¶ 13 (1980).
74. See 47 U.S.C. § 153(6). See also, 47 U.S.C. § 525(5)(A) (defining cable services as Aone-way transmission@ of video or other like services).
75. See § 13.7.1; FBL §§ 7.4.4., 8.4.6.
76. Local telephone companies have long held exclusive franchises for the provision of local telephone service. See §§ 3.2.1; 6.1.2.
77. See Subscription Video, Report and Order, 2 FCC Rcd 1001, 1003-04 ¶¶ 23-28 (1986).
78. See § 12.7.
79. See FM Licensees: Amendment of the Commission=s Rules Concerning Use of Subsidiary Communications Authorizations, 48 Fed. Reg. 28,445, 28,446-48 (1983); § 11.8.2.
80. See § 3.13.5.
81. See 47 U.S.C. § 214(a).
82. See id. § 201.
83. See § 3.11.1.
84. Under section 307(b), the Commission is simply required to oversee the distribution of broadcast "licenses, frequencies, hours of operation, and power among the several states and communities as to provide a fair, efficient and equitable distribution of radio service to each of the same." 47 U.S.C § 307(b).
85. See FBL §§ 4.6.2-4.6.3.
86. The Commission may revoke a station license for any reason that would have warranted refusing a license in the first place. 47 U.S.C. § 312 (a). Licenses may not be transferred without Commission approval. Id. § 310(d). See § 10.3.1.
87. We discuss the regulation of data in detail in Chapter 11.
88. See § 12.6.2.
89. See Vietor, AT&T and the Public Good: Regulation and Competition in Telecommunications, 1910-1987, in Future Competition in Telecommunications 27, 52 (S. Bradley and J. Hausman eds. 1989).
90. See §§ 11.8; 13.1.
91. See M. Hamburg, All About Cable § 1.02, at 1-6 (rev. ed. 1985).
92. See § 9.3.2; P. Temin, The Fall of the Bell System: A Study in Pricing and Politics 47-54 (1987). See generally L. Kahaner, On The Line: The Men of MCI – Who Took on AT&T, Risked Everything, and Won! (1986).
93. Quoted in Temin, The Fall of the Bell System at 50.
94. For the FCC's (and the Bell System's) reaction to this competitive challenge, see §§ 9.3.5-.9.4.3.1.
95. Brooks, Telephone: The First Hundred Years at 244.
96. Id. at 279.
97. See § 4.4.3. This suit is discussed in greater detail in the first edition of this treatise, Kellogg, Thorne & Huber, Federal Telecommunications Law § 4.3 (1992) (FTL1).
98. MacKay Radio and Tel. Co., 15 FCC 690 (1951) (Three Circuits), rev'd, RCA Communications, Inc. v. FCC, 201 F.2d 694 (D.C. Cir. 1952), vacated and remanded FCC v. RCA, 346 U.S. 86 (1953).
99. FCC v. RCA, 346 U.S. at 93.
100. Id. at 95.
101. Id. at 93.
102. See §§ 2.3.
103. For a discussion of the FCC's evolving CPE policies, see § 8.4.
104. See § 8.4.1.2.
105. Use of the Carterfone Device in Message Toll Telephone Service, Decision, 13 FCC2d 420, 424, recon. denied, 14 FCC2d 571 (1968) (Carterfone).
106. On reconsideration, for example, it insisted that it was not "open[ing] the door to customer ownership of telephone handsets." Carterfone, 14 FCC2d at 572.
107. See § 8.4.1.2.
108. See § 8.6.1.
109. See § 8.3.
110. For a discussion of computer-based enhanced services, see Chapter 12.
111. See § 12.3.
112. See § 12.4.
113. For a discussion of the FCC's evolving long distance policies, see §§ 9.2 – 9.5.
114. See § 9.3.2.
115. See § 9.4.2.
116. See §§ 2.5; § 6.1.1.2.
117. See §§ 2.2.
118. See FTL1 § 4.4.
119. For a discussion of the FCC's wireless policies, see §§ 10.8.1.
120. See G. Calhoun, Digital Cellular Radio 68-86 (1988)
121. General Mobile Radio Serv. Allocation of Frequencies Between 25 & 30 Megacycles, Report and Order, 13 FCC 1190, 1228 (1949).
122. See §§5.3.1, 10.5.1.
123. See Chapter 4.
124. Pastor v. AT&T, 76 F. Supp. 781 (S.D.N.Y. 1940).
125. Id. at 782. An amusing irony is that had Bell permitted that technology to advance, both MCI and the federal government would have had much less to complain about three decades later, when both made the absence of "1+ dialing" the centerpiece of their attacks on the Bell System.
126. See §8.4.1.2.
127. See Carter v. AT&T, 365 F.2d 486, 492 (5th Cir. 1966).
128. See § 8.4.2.
129. See Essential Communications v. AT&T, 610 F.2d 1114 (3d Cir. 1979).
130. Northeastern Tel. Co. v. AT&T, 651 F.2d 76 (2d Cir. 1981), cert. denied, 455 U.S. 943 (1982). The Second Circuit reversed all but one aspect of the verdict, and the case was subsequently settled.
131. See Litton Sys., Inc. v. AT&T, 700 F.2d 785, (2d Cir. 1983), cert. denied, 464 U.S. 1073 (1984). This verdict was affirmed.
132. See Phonetele, Inc. v. AT&T, 889 F.2d 224 (9th Cir. 1989), cert. denied, 60 U.S.L.W. 3599 (Mar. 3, 1992).
133. See, FTL1 §3.8
134. See §4.3.
135. See MCI v. AT&T, 708 F.2d 1081 (7th Cir.), cert. denied, 464 U.S. 891 (1983).
136. See Mid-Texas Communications Sys., Inc. v. AT&T, 615 F.2d 1372 (5th Cir.), cert. denied, 449 U.S. 912 (1980).
137. See id. at 1391.
138. See Southern Pacific Communications Co. v. AT&T, 556 F. Supp. 825 (D.D.C. 1983), aff'd, 740 F.2d 980 (D.C. Cir. 1984), cert. denied, 470 U.S. 1005 (1985).
139. The circumstances leading up to the entry of the decree, and the decree itself, are discussed in greater detail in FTL1 § 4.5.
140. See FTL1 Chapter 5.
141. See FTL1 Chapter 6
142. See FTL1 Chapter 7.
143. See T.R. Reid, The Chip: How Two Americans Invented the Microchip and Launched a Revolution 133 (1984).
144. See §§ 5.3.1; 10.5.
145. See § 10.4.3.
146. Although the present capacity is 24,000 simulataneous conversations, "[e]xperts suggest a theoretical carrying capacity of about 600 million [simultaneous] conversations." Yates, The Promise of Fiber Optics, Pub. Util. Fort., Aug. 15, 1990, at 14.
147. Ithiel de Sola Pool anticipated this trend as early as 1983, in his landmark Technologies of Freedom. "For the first three-quarters of the twentieth century the major means of communications were neatly partitioned from each other, both by technology and by use," Pool wrote. "Now the picture is changing. Many of the neat separations between different media no longer hold. . . . The explanation for the current convergence between historically separated modes of communication lies in the habitability of digital electronics." de Sola Poole, Technologies of Freedom at 26-27.
148. See generally J. Kearney and T. Merrill, The Great Transformation of Regulated Industries Law, 98 Colum. L. Rev. 1323 (Oct. 1998).
149. See 47 U.S.C. § 253(d); see § 3.8.
150. Telecommunications Act of 1996 § 302(b)(1) (repealing 47 U.S.C. § 533(b)). See § 13.7.2.6; see also FTL1 § 14.8 (and 1995 Supp.); FBL § 7.4.4.
151. The 1996 Act authorizes the Commission to permit registered holding companies to offer telecommunications services through a separate, single-purpose telecommunications subsidiary. See Telecommunications Act of 1996 § 402(b)(2)(A) (to be codified at 15 U.S.C. § 79 et seq., amending PUHCA Section 34(a)(1)).
152. 47 U.S.C. § 251(c)(2); see § 5.5.3.4( i).
153. 47 U.S.C. § 251(c)(3) ; see § 5.5.3.6(ii).
154. 47 U.S.C. § 251(c)(4)(A); see § 5.5.3.6.
155. 47 U.S.C. § 254; see Chapter 6.
156. 47 U.S.C. § 160.
157. See Chapter 2.
158. See § 5.2.
159. See § 5.3.
160. For example, each interconnection agreement negotiated under § 252 must include Aa detailed schedule of itemized charges for interconnection and each service or network element included in the agreement.@ 47 U.S.C. § 252(a)(1).
161. 47 U.S.C. § 252(i).
162. See, e.g., Price Cap Performance Review for Local Exchange Carriers, Fourth Report and Order in CC Docket No. 94-1 and Second Report and Order in CC Docket No. 96-262, 12 FCC Rcd 16,642 (1997).
163. See Regulatory and Policy Problems Presented by the Interdependence of Computer and Communications Services and Facilities, Final Decision and Order, 28 FCC2d 267 (1971), aff=d in pertinent part sub nom., GTE Serv. Corp. v. FCC, 474 F.2d 724 (2d Cir. 1973); Amendment of § 64.702 of the Commission=s Rules & Regulations, (Second Computer Inquiry), Final Decision, 77 FCC2d 384 (1980), modified on recon., 84 FCC2d 50 (1980), further modification on recon., 88 FCC2d 512 (1981), aff=d sub nom., Computer and Communications Indus. Ass’n v. FCC, 693 F.2d 198 (D.C. Cir. 1982), cert. denied, 461 U.S. 938 (1983), aff=d on second further recon., 56 Rad. Reg. 2d (P&F) 301 (1984).
164. See FBL § 2.7.3.
165. DSL lines have advanced electronics – DSL modems – installed on either end of the line that provide advanced signal processing, digital multiplexing, and compression algorithms to transmit a tremendous amount of information. See FBL § 2.7.4.
166. See FBL § 2.5.1(vi).
167. See FBL § 2.5.1(vii).
168. Unlike conventional cellular, PCS networks were designed from the start as digital networks, transmitting information in bits instead of analog waves.
169. These top 10 markets serve 30 percent of U.S. television households. By Nov. 1999, the top 30 markets – covering 53 percent of U.S. TV households – will convert to digital. See Advanced Television Systems and Their Impact upon the Existing Television Broadcast Service, Fifth Report and Order, 12 FCC Rcd 12,809, 12,841 ¶ 76 (1997).
170. See § 11.2.2.
171. See generally E. Leo & P. Huber, The Incidental, Accidental Deregulation of Data . . . and Everything Else, 6 Industrial and Corporate Change 807 (Dec. 1997).
172. See generally, H. Shelanski and P. Huber, Administrative Creation of Property Rights to Radio Spectrum, 41 J.L. & Econ. 581 (1998).
173. Global One (Sprint, France Telecom, Deutsche Telekom), WorldPartners (AT&T, KDD, Unisource, Singapore Telecom, Telstra, and several non-equity partners), and Unisource (Dutch, Swedish, Swiss and Italian PTTs). AT&T and British Telecom have announced plans to merge their international operations into a jointly owned company that would have $10 billion in revenues.
174. Red Lion Broad. Co., Inc. v. FCC, 395 U.S. 367, 376 (1969).