Back to main page

Mergers, Divestitures and the Internet: Is Ownership of the Media Industry Becoming too Concentrated?⁽¹⁾

Benjamin M. Compaine

Telecommunications Policy Research Conference, September 26, 1999
(Based on Chapter 10 of Who Owns the Media?, 3rd ed., 2000)

With mergers and acquisitions in broadcasting, publishing, cable and the Internet making news almost weekly in the late 1990s, a question that regularly arises with each announcement is whether the media are becoming more or unduly concentrated in fewer hands. The answer to this question has cultural and political implications as well as ramifications for antitrust policy. To address this question, this paper asks whether the old rules, the classic verities about the media and its control, have changed at the end of the 20^th century.

At least as early as 1982 the blurring of the boundaries among the traditional media hinted that change might be in the offing:

...[T]he information dissemination process is rapidly changing. Computers and connected terminals in homes and offices increasingly allow users to select the information they wish to receive, at precisely the time they wish to use it. Computers have made it economically feasible to mail identical, "personalized" messages to millions of recipients using the postal system that at one time was reserved for point-to-point communication. The telephone can give countless users almost simultaneous access to the same computer data base. The telephone and computer are also being combined to provide "electronic mail," perhaps doing for mail what the Xerox machine did for memos. Video and audio cassette recording devices allow individuals to record broadcast programs for replay at a time of their own choosing.⁽²⁾

One need not look much further than the case of the breaking of the Monica Lewinsky story in 1998 for validation:

Matt Drudge heard rumors about Lewinsky and President Clinton. He heard that Michael Isikoff, a Newsweek reporter, was working on this story. Drudge was a freelance writer with no formal journalism training -- not even a college degree. Drudge's father bought Matt a computer during a visit with his son in the mid 1990s. Drudge was soon exchanging gossip on Internet news groups. He started collecting e-mail addresses from these exchanges and began an e-mail list with what he dubbed "The Drudge Report." First he had 100 names, soon with 5,000 and then 100,000 names. His Web site by the same name was born. It specialized in political gossip -- not reporting.⁽³⁾

In 1998 Drudge was running his Web site from his $600 a month apartment, using a modest computer based on an outdate Intel 486 microprocessor. In January he did not have the corporate nor ethical burdens on him that Isikoff and his colleagues had. The Drudge Report, on the Web, published a story about what Newsweek was holding back: the story of Bill Clinton and Monica Lewinsky. Now in the public domain, the mainstream media picked up the story. Two pieces of history were made that day. The most obvious are the forces it unleashed that lead to the impeachment of a President of the United States for only the second time. But it also made media history. Though not the first time that the mass media were goaded into running a story after its break in the small media or even the Internet, it was by far the biggest and most consequential.

As the 1990s ended The Drudge Report received more visitors each day than the weekly circulation of Time magazine.⁽⁴⁾

Matt Drudge had some insight on the connection between these two: the relationship between the big established media companies and the very small:

What's going on here? Well, clearly there is a hunger for unedited information, absent corporate considerations. As the first guy who has made a name for himself on the Internet, I've been invited to more and more high-toned gatherings such as this...

Exalted minds--the panelists' and the audience's average IQ exceeds the Dow Jones--didn't appear to have a clue what this Internet's going to do; what we're going to make of it, what we're going to--what this is all going to turn into. But I have glimpses...

We have entered an era vibrating with the din of small voices. Every citizen can be a reporter, can take on the powers that be. The difference between the Internet, television and radio, magazines, newspapers is the two-way communication. The Net gives as much voice to a 13-year-old computer geek like me as to a CEO or Speaker of the House. We all become equal.⁽⁵⁾

WHAT IS GOING ON HERE?

This paper looks at the implications of the common digital nature of what had been discrete media industries for determining the basis for analyzing the degree of media ownership and concentration in the U.S. Besides looking at the empirical trends of concentration, it addresses some of the legal, economic social, and cultural concerns that are part of the public policy issues raised when judging media competition.

When U.S. media pundit A.J. Liebling wrote that freedom of the press belongs to those who own one he summed up the emotion that separates the media business from virtually any other enterprise. The press -- or today more generically the mass media -- stands not simply for the power to convey information, but more crucially for the assumed ability to shape attitudes, opinions and beliefs. The media are the vehicles for education--and propaganda. Who controls these outlets and what the players' intentions are for their use has been a contentious issue at least since the 15th century, when both Church and State recognized the potential of the printing press and immediately sought to control it.

From time to time in recent history public policy has become concerned with apparent trends towards concentration in one branch or another of the media industry. Since the 1930s various federal bodies have legislated, adjudicated, or regulated such areas as ending newspaper-television cross ownership, breaking up theatrical film distribution-exhibition combinations, limiting broadcast station ownership, prohibiting most television network program ownership, and preventing telephone ownership of cable.⁽⁶⁾

At the start of the 21^st century, the issue remains salient. The stakes have risen higher than ever, as ownership has broken out of national boundaries. The United States has long had a major media presence in much of the world through the preeminence of Hollywood in film and television production. British, German, Dutch, French, Japanese and Australian players have become prominent in the U.S., initially in publishing, then in all aspects of media.

The substance conveyed by the media can no longer be stopped at national boundaries by customs services. A dial-up telephone connection between computers can transfer information in seconds. A videocassette or disc can be easily smuggled and inexpensively duplicated, and a television program can be transmitted by satellite across thousands of miles for reception by an antenna that can be purchased by anyone at a local electronic store. In 1999, $600 worth of computing equipment and $20 monthly access to an Internet Service Provider could produce a globally accessible presence that only a decade before required tens or hundreds of thousands of dollars for equipment and distribution.

There appear to be two trends pulling in opposite directions. One trend, suggested by a reading of the headlines, is that there is a new round of consolidation within the media industry. This would imply a lessening of separately owned outlets for information. At the same time, the trend of smaller, faster, better and cheaper information-related technologies appears to be generating the ability to create, store and transmit many types of information faster and less expensively, with greater "production values," than ever before. This trend would appear to imply that there is opportunity for a greater number of information outlets.

SIZE AND SCOPE OF THE MEDIA BUSINESS

Figure 1 is a map of the information business (omitted in online version]. On it are placed some of the products and services that are part of that industry.⁽⁷⁾ The businesses that make up the media industry occupy roughly the right quadrant along the process-substance axis and vertically span the range of both products and services. The media include the virtually pure service function of the news wire services used by publishers as well as the pure products called books or magazines. But they also stretch two-thirds of the way towards service -- the left side of the map along the top -- reflecting the broad range of transmission vehicles that are available as distribution conduits. To the extent that information services increasingly use the telecommunications networks to transmit computer-based content, the line could be extended even further to the left. Indeed, given the substantial reliance of magazine and book publishers on the Postal Service and private delivery services, one could argue that the media extend completely along the horizontal axis as well. The demarcation criteria in 1999, however, may be based on the extent to which the conduit operator has responsibility for content. Cable and broadcast operators do make content decisions, whereas in 1999 the telephone companies and Postal Service remained common carriers and thus exercise no substantial content decisions.

The entire media and entertainment business accounted for almost one third of the total information industries revenue of an estimated $947 billion in 1996.⁽⁸⁾ Between 1987 and 1996 the information industry grew at nearly twice the rate of the overall economy as measured by GDP.

Although the media and entertainment sector grew nearly 50% faster than the economy as a whole, it was less than the information industry as a whole. Within this sector rates of growth varied dramatically. The mature print business underperformed the economy, with growth rates ranging from 2.6% (newspapers) to 4.9% (periodicals). The motion picture sector, which included increasing revenue from videocassettes and cable networks was the fastest growing media segment.⁽⁹⁾

BLURRING BOUNDARIES OF MEDIA INDUSTRIES

Mass communication historically has had certain characteristics that differentiate it from other forms of communication. First, it was directed to relatively large, heterogeneous and mostly anonymous audiences. Second, the messages were transmitted publicly, usually intended to reach most members of the audience at about the same time. Finally, the content providers operated within or through a complex, often capital-intensive industry structure.⁽¹⁰⁾ Point-to-point forms of communication, such as telephone or letter mail, traditionally have had only the third of these characteristics.

Digital technologies and the Internet infrastructure undermines those long-held characteristics. Digital means that text, audio and visual information are in the identical and interchangeable format of bits. The Internet, while relying on a complex and overall expensive structure, is more like the highway system it is often compared to than a printing press or broadcast network. It is owned by hundreds of thousands of entities but works as a whole to connect hundreds of millions of users, much like the switched telephone network. Thus, the media arena, which in an earlier era could be described as encompassing industries known as newspaper, film, books, television, etc., today must recognize less precise boundaries for the term "mass."

More crucially, the traditional media industries are finding a blurring of the boundaries among themselves. For example, a television set may gets its picture and audio at any given moment from a broadcast signal, a coaxial cable signal, a video cassette, an optical disk or a telephone line. What then is the relevant medium? A person viewing the screen may not even know what the conduit is at any moment.

The changing media environment that makes a precise definition of the media arena difficult also means that competition may be coming from new, less traditional players, such as telephone companies, computer firms, financial institutions and others involved in the information business. This suggests not only a broadened arena for conflict in the marketplace, but in the regulatory environment overall, as government agencies seek to identify their territory.

DETERMINING MEDIA CONCENTRATION

The implied policy issue is whether the mass media industries or any segment has a degree of concentration that would be in some way detrimental to some vaguely defined political, social, economic and cultural interest of society. This is by far a more difficult question to answer definitively than the relatively specific tabulation of who owns what entity. Moreover, it should be equally valid to ask whether the media were in 1999 or were likely in the foreseeable future to be too diffuse, decentralized and competitive.

The implications of concentration are the more typically voiced. Bagdikian has articulated this concern in writing that "our view of the social-political world is deficient" if there is a regular omission or insufficient inclusion or certain elements of reality. And that is happening, he believes when "the most important institutions in the production of our view of the real social world," the mass media, are becoming "the property of the most persistent beneficiaries" of the mass media's biases.⁽¹¹⁾ Although corporations "claim to permit great freedom" for their editors and producers, he asserts that these businesses "seldom refrain from using their power over public information."⁽¹²⁾ Moreover, as Bagdikian concludes, it is not the total number of outlets that matters, but the number of owners.

Thus, it would seem that when it comes to the media industry, there are at least two thresholds in the continuum from monopoly to unfettered competition. First, there is the conventional antitrust standard. This is primarily the realm of concentration ratios, Lorenz Curves, and Herfindahl-Hirschmann indexes. Second, there is the socio-political standard, the one that says we need to ensure diversity of sources, accessability by consumers, and uncontrolled distribution. There are no known indexes, curves or standards for this measurement of competition. It tends to be on an "I'll know it when I see it." basis. Presumably, however, it is the socio-political standard that the antitrust standard is intended to promote: as the media approaches concentration that gets closer to the antitrust trip-wire, the more likely there is to be the threat of narrowness and missions that Bagdikian and others worry about.

To that end, this paper next examines the economic and antitrust version of the media world. Then it returns to address some of the socio-political concerns.

THE ECONOMIC NATURE OF INFORMATION

The study of the media is frequently an attempt at understanding the status of the flow of content -- of information. It is the communication process itself that ultimately has meaning for society. It has been challenging for economists to get a handle on the nature of information: What is it, how is it used, what is its value, how can-- or should -- it be allocated?

In the manufacturing environment, economists have pointed to the "law of diminishing returns" to suppport practical, limits on enterprise size. In the new digital world of information, there is growing recognition for a "law of increasing returns." If accurate, this is an inflection point in our understanding of bigness and its economic consequences.

The term "marketplace," as in "marketplace of ideas" is frequently applied to describe the ideal environment for information. But describing the information marketplace is a different order of problem than characterizing the marketplace for toothpaste, or even for newspapers. Conventional measurements do not suffice when dealing with the amorphous and inexact concept of information. For example, how does one place a monetary value on the information an airline pilot uses to guide a jet not in sight of land to a precise destination -- the weather reports, the navigational aids, the on-board computer read-outs, etc., not to mention the knowledge and intuition gained by years of accumulated experience?

Marketplace seems to presuppose information is indeed a commodity, like cotton, paper, or hamburger. This may be a reasonable assumption, but it must be tested in light of other economic approaches to its nature. For example, at the other extreme, information may be viewed as a theoretical construct, having features unlike other commodities and therefore requiring unique treatment. In between there is an alternative that grants some commodity-like characteristics to information, but recognizes other distinctive features as well. For example, typical commodities are tangible, but information may not be. Most commodities lend themselves to exclusivity of possession, but information can be possessed by many individuals at the same time without any other being deprived of it. In addition, there is frequently little or no marginal cost to the provider of information in reaching a wider audience.⁽¹³⁾

This last viewpoint is the one that is accepted for this paper. It considers information a "public good." One key characteristic of public goods is that of essentially no marginal cost associated with adding distribution. An example is a television broadcast. Once the fixed costs of production have been incurred and the show is sent out over the air, there is no difference in expense to the broadcaster whether 10 households or 10 million households tune in to the show. Thus, broadcast television (and radio) can be given away. However, advertising is not sold at its marginal cost, since that would be zero.

The "product" of the media differs from most commodities, which are private goods. Every orange, for instance, has a cost, and each one adds weight in shipment. Selling more oranges means adding more orange trees, etc. There can be a real marginal cost -- the expense of growing and shipping one more orange.

In print media, the informational content is really the public good, while the physical product -- paper and ink -- is a private good. In many cases, the cost of producing the first copy constitutes the bulk of total cost, just as in broadcasting the production is virtually the total cost. Costs of editorial staff, typesetting and plate making are all necessary whether the print run will be 100 or 100,000. The incentive, therefore, for broadcasters and publishers is to increase circulation or audience for a product, since that adds little or nothing to marginal costs while justifying higher marginal revenue from advertisers in the form of higher advertising rates. The public good aspect of information is what encourages television networks and syndicated shows, as well as the desire for a firm to trade up from stations in smaller markets to larger ones. News services and print syndicates are encouraged by the same economic facts. Information provided over the Internet is a public good. The critical characteristics here is that many users can have the same data or information without depriving another else from having it. The same cannot be said for a newspaper. If I have the physical paper, you can't have it. But I can read an online newspaper at the same time as dozens or thousands of others.

COMPETITION IN THE MEDIA ARENA

One of the key needs for determining competition is first determining the relevant market. This can be applied to the product market and the geographic market. The distinctions are particulary critical for many media, which are geographically very local, while being part of a broader product market.

Section 7 of the Clayton Antitrust Act, along with years of court interpretations, makes determination of "line of commerce," or product market and "section of the country" or geographic market, the first step in any determination of concentration.⁽¹⁴⁾ The more that products are reasonably interchangeable, the more likely it is that they should be considered as the same product market. This applies both from the perspective of consumers (demand) and potential market entrants (supply). After the product market is determined, the geographic market is addressed. Traditionally the geographic market may be a city, a region or the entire country. Increasingly, there may be a global component as well.⁽¹⁵⁾

Most daily newspapers, cable systems and television and radio stations are distinctly local for the geographic markets. Periodicals are for the most part geographically national or regional. Film, recordings, books, television programming are, for the most part national in their geographic market.

Thus, in analyzing competition for narrow media industry segments, it is critical to distinguish what the standard for the relevant market might be. There remain dozens of cable operators, but the household in any given locality has from the start generally had only one choice (enforced in part by the economics of building a cable system, but as well by the monopoly franchising power of local governments). Similarly, the residents and merchants of most cities and towns have only a single daily local newspaper to buy or advertise in. Though there are 9000 radio stations, a given locality may reliably have access to 10 or 40 that have transmitters in the area. Thus, while it may be useful to aggregate the overall market "power" of a newspaper or cable chain, the degree of competition at the local level needs to be considered separately than do the less geographically-based media. One reason why the Internet is such a break with old media is that it has virtually no geographical market limitations.

Quite relevant to the issue of local competition for specific media, however, is the degree of substitutability among media. If an advertiser is not satisfied with the pricing or service of the local newspaper, what options does it have? If a household is displeased with the local cable operator, what reasonable alternatives, if any, exist?

This was the central point of Theodore Levitt's enduring concept of "marketing myopia." To prevent marketing myopia demands that a firm carefully determine its field of operations. For example, decades ago the railroads conceived themselves (as did the regulators) as being in the railroad business. The relevant antitrust question was the market share of competing railroads. But with the expansion of the Interstate Highway System in the 1950s and 1960s, all railroads started losing tons of freight to trucks, as well as passengers to cars. The relevant market turned out to be "transportation." So it is with beverage containers: steel, aluminum, glass, plastic, cardboard. Could a single manufacturer in the aluminum can and the glass and plastic businesses be considered monopolists? Would a buyer -- say Coke or Pepsi -- in fact have a choice and be able to bid one against the other?

And so might well it be with the media. Granted, each medium is not perfectly interchangeable with another. Classified advertising does not work well on television, music cannot be played in a magazine. However, there is probably more fungibility than not. Daily newspaper circulation has declined steadily (as measured by household penetration), while the percentage of adults who claim to get most of their news from television has increased. Are those trends related? DBS service is a close alternative to cable. Video cassette and disk sales and rentals compete with movie theaters as well as premium and pay-per-view service. Direct mail competes with newspapers.

Again, these do not need to be perfect substitutes, at all times, for all types of content. Cardboard containers compete with glass or plastic containers for juice, but not for carbonated beverages. The overlap is substantial, not perfect. Similarly, consumers and advertisers may find some bundle of options that erodes the notion of a local cable, newspaper, or broadcaster bottleneck. And, here again, the Internet is changing everything. Bits -- video, text, audio, and graphics -- have become increasingly fungible, so that audio and video can be part of the Web site of a newspaper, while text is now part of the Web site of a television producer's's site.⁽¹⁶⁾

Criteria for Ascertaining Antitrust

Starting with the Sherman Antitrust Act in 1890, Congress has taken legislative steps targeted at the breakup and prevention of industry concentration. Authority for implementing antitrust policy is shared by the Justice Department and, since the Clayton Act in 1914, the Federal Trade Commission.

There is a rich history of antitrust activity. For the most part, antitrust cases are seldomly clear cut and frequently involve years, if not decades, of fact finding, negotiation, trial and appeals. In the media industry one of the first antitrust cases was Associated Press v. United States. In this case the U.S. government sued the newspaper cooperative on antitrust grounds for its restrictive policies for membership. The AP argued both that it was protected by the First Amendment and immune from the Sherman Act, as newspapers were not engaged in interstate commerce. The Supreme Court clearly placed newspapers within the jurisdiction of antitrust legislation, holding that "Freedom to publish is guaranteed by the Constitution, but freedom to combine to keep others from publishing is not."⁽¹⁷⁾

In a 1948 ruling, the Court ruled against the vertical integration of Paramount Pictures as a motion picture distributor and theater chain. In a consent degree, Paramount had to divest its theater chain.⁽¹⁸⁾ For decades the Paramount decision was the basis for restricting vertical integration between producer/distributors of motion pictures and exhibition.

Historically there have been two critical newspaper antitrust cases. In the case of the joint operating agreement between the two daily newspapers in Tucson, Arizona, the Supreme Court upheld a judgement that charged the two papers with price fixing, profit pooling and market allocation.⁽¹⁹⁾ This was the catalyst for the Newspaper Preservation Act that Congress passed, essentially giving newspapers a dispensation from this form of otherwise anticompetitive behavior, in the interest, it was viewed, of the greater benefit of preserving limited competition. The other case involved, the Times Mirror Co, owner of the Los Angeles Times, which was ordered to divest the neighboring newspaper in San Bernardino, though another chain, Gannett faced no obstacle in being the new purchaser.

New Technologies and Consumer Behavior and Media Markets

The courts and ultimately the antitrust litigators in government have recognized changing technologies and consumer media user patterns. As far back as 1975, the Justice Department argued that, in certain circumstances, newspapers, television stations, and radio stations compete and, therefore, should be included in the same product market.⁽²⁰⁾ In Satellite Television v. Continental Cablevision, the Court of Appeals held that "cinema, broadcast television, video disks and cassettes, and other types of leisure and entertainment-related businesses for customers who live in single-family dwellings and apartment houses" were reasonably interchangeable and constituted a single product market.⁽²¹⁾ Another Appeals Court decision, Cable Holdings of Georgia v. Home Video, Inc., found that consumers perceive cable television, satellite television, video cassette recordings, and free broadcast television to be reasonable substitutes.⁽²²⁾ Thus, the relevant market definition, was that all "passive visual entertainment" are reasonably interchangeable by consumers and constituted a single product market.⁽²³⁾ The result was that the court upheld a merger between two cable companies.

Similar reasoning by the courts ultimately eroded and broadened the definition of the product market for first run films in the Syufy case in 1989.⁽²⁴⁾ At issue was the alleged concentration of ownership of motion picture theaters in Las Vegas. The defendant, Syufy Enterprises, argued that its competition was not just movie theaters but videocassette rentals, cable and pay-TV. With evidence that owners of VCRs and subscribers to cable did attend first run theaters less often than nonVCR and cable consumers, the court agreed that the relevant competitive market was greater than just the market share of movie house attendance. This determination has lead to a lessening of federal oversight of vertical integration in the motion picture industry.⁽²⁵⁾

The reality of the new mix of the media is suggested in a case study of a film, The Shawshank Redemption. It was produced and released in 1994 by Castle Rock Entertainment, a small studio that subsequently became part of Time Warner. Despite some excellent reviews, it's initial theatrical release produced a rather poor $18 million in box office receipts. Here's how other media affected Shawshank:

• Broadcast TV: The attention from several Academy Award nominations (the ceremony viewed by millions on television) enabled it to bring in another $10 million in 1995.
• Cable TV: Cable network TNT heavily promoted the movie as part of its "New Classics" campaign.⁽²⁶⁾
• Internet: About a dozen Web sites devoted to "Shawshankmania" were created by individuals. In the evaluation of film critic Roger Ebert, "The Web has become an 'important element in any film reaching cult status, because people who like it can find a lot of others who agree with them. Movie lovers with specialized tastes no longer feel isolated.'"⁽²⁷⁾

With these other media in play The Shawshank Redemption became the top video rental in 1995.⁽²⁸⁾ It was the top movie on the Internet Movie Database list, ahead of Godfather, Star Wars and Schindler's List in 1999.⁽²⁹⁾

Cable television, videos and the Internet have all given movies more avenues to reach viewers by -- and, in turn, they have given audiences more say in a movie's long-term appeal.... "There are just so many more ways to discover a movie than there used to be," says Martin Shafer, a principal at Castle Rock Entertainment."⁽³⁰⁾

Flowing from examples such as this, there is growing recognition by the courts that in determining economic concentration there is the need for broadened product market definitions for the media industry, transcending the traditional boundaries of standard industry codes.

MEASURING MEDIA COMPETITION

Among the various economic measures of concentration, the Herfindahl-Hirschmann Index is one of the more robust because "it reflects...the number and size distribution of firms in a market, as well as concentration of output."⁽³¹⁾ It is calculated by squaring the market share of each player in the industry. Generally an HHI score of greater than 1800 indicates a highly concentrated industry. Under 1000 is considered unconcentrated, with scores in between degrees of moderate concentration.

In an example of an industry with 10 providers, it can differentiate between a playing field where the market is relatively equally divided and one where a few players hold most of the revenue. In example A, the three largest players account for 30%, 25% and 20%, respectively, of industry sales. The remaining seven divided up 25% about equally. It has an H-H Index of 2014, highly concentrated. In example B, the largest firm has a 15% market share, the second firm 12%, the third 10%, and the seven remaining firms roughly divide the rest. The HHI in this industry is 1036, low concentration.

Greco applied the HHI to the book publishing industry over the years 1989 to 1994.⁽³²⁾ This followed several decades of apparent consolidation from mergers, including the 1970s period that encouraged the Federal Trade Commission to investigate media concentration in 1978. Seen in Table 1, during the period studies the 14 largest book publishers accounted for 75% to 80% of total book industry revenue, but with downs and ups over the years. The fragmented nature of the industry is seen in the HHI. Even at the its highest, in 1994, the HHI indicated a very competitive industry, well below even the low boundary of oligopoly. Greco went on to calculate the effect of a single company controlling the 25% to 20% of the industry revenues not account for by the 14 firms covered in his study. This would have shown an HHI of 931 in 1994 -- a drop from 1101 in 1989, below the minimum for low concentration. Greco further documented that over the decades of mergers the volume of new titles published grew dramatically, a sign of great competition, adding credibility to the HHI data of a highly competitive market.

A similar type of analysis may be applied to the television broadcasting segment. In 1997 the 20 largest broadcast companies had an aggregate of $23.9 billion in broadcast revenue. CBS and NBC each accounted for about 20% of the total, ABC 19% and Fox 11%. From Table 2, these four accounted for 60% of the revenue of the top 20 broadcasters, which in turn was the dominant share of the total broadcast market. But, contrary to what might have been assumed, this share, and the HHI was actually lower in 1997 than in 1994, before the wave of mergers in response to the liberalized ownerships standards of the 1996 Telecommunications Act.

Table 2 suggests that television broadcasting is a moderately concentrated industry -- no surprise. But it further shows that over this period the concentration, as measured by HHI, also decreased by nearly 12 percent among the 20 largest companies. The revenue share of the four and 10 largest players was lower This is largely due to the mergers at the bottom of the industry, creating stiffer competition for an industry that, until 1986, was dominated by only three networks and limited to many small groups that could have no more than seven stations each. In 1980 the three largest players (the networks) held an industry share about equal to that of the four major networks in 1997.⁽³³⁾ These are small changes, but at the very least suggest a different outlook than the intuitive one created by merger announcements.

TRENDS IN MEDIA CONCENTRATION, 1986 TO 1997

Nonetheless, focusing on trends in a specific market segment is a distraction from the prevailing trends. With the continued blurring of the boundaries of the old media as all become essentially digital in nature, the product market distinctions have become all but meaningless. Broadcasters compete with programming that is available only over cable: few viewers with cable (the majority) care or even know the difference. Cable operators that provide Internet access and switched telephone service blend with telephone companies offering their own high speed data services and even video. Thousands of newspapers have World Wide Web sites that are accessible not only by the hometown residents but by anyone, anywhere. Radio broadcasters are also available via Internet, while other programmers, without any government license, are available via the Internet, including users with portable wireless connection. Record manufacturers are facing Internet delivered music. It is quite difficult to sustain a fiction of old boundaries: that newspapers compete only with newspapers, the local tv stations only with the few others in the market. It is through the merger of digital technologies, more than mergers of companies that has brought "all modes of communications into one grand system."⁽³⁴⁾

How does the overall media industry look based on concentration percentages as well as the HHI? Table 3, identifies the 50 largest media companies in 1986 and 1997 by the revenue from their media activities. In most cases, this is 100% of their revenue. For a few companies, the parent company has much greater revenue. For example NBC's revenue in Table 3 was about 6% of parent General Electric's revenue.

Using 1986 as the base year for comparison is appropriate as it was the first year after the Federal Communications Commission eased the number of television stations under the ownership of a single firm from seven to 12. It was in that year that News Corporation launched the first successful challenge to the long dominance of the older three commercial networks, opening the gates to new competition in broadcasting. The timing of the Fox network a year later was not coincidental. The ability of News Corp. to gain ownership of local stations in 12 major markets gave it a needed core of network affiliates. In the early 1990s the FCC's restrictions on broadcast networks owning a financial interest in prime time programming were phased out. In early 1996 the Telecommunications Act substantially eliminated the size of broadcast radio groups and further loosened restriction on television station group ownership.

Findings

Tables 3 and 4 evaluate media ownership as a single industry. Among the observations are:

• As measured by revenue, there was little change in media concentration between 1986 and 1997. In the former period the top 50 accounted for about 79% of revenue. By the end of the period it edged up to under 82%. The change in concentration among the top 20 and top eight was similarly small. Only at the top four level has there been substantially greater concentration (see next bullet).
• At the very top, the two largest companies in 1986 (CBS and Capital Cities/ABC) accounted for 10.5% of industry revenue. The top duo in 1997 (Time Warner and Disney, with most of Capital Cities/ABC) had 16.4% of industry revenue. This is the only economic measure by which the notion of increased concentration of ownership of the media had substantive backing.
  
  
  

Table 4-- Concentration of Media Industry Revenue by Number of Companies, 1986 and 1997

• The HHI increased from an extremely low 206 in 1986 to a still very low 268 in 1997. Thus, this measure did show some increased concentration, but with HHI levels well under 1000 indicating low concentration in the media industry and one of the most competitive major industries in United States commerce.
• There has been a substantial turnover in the companies in the top 50 and even the top dozen. CBS, the largest in 1986, was eleventh in 1997. Dun & Bradstreet, Gannett, Times Mirror, Newhouse, Knight Ridder and Tribune Co. are firms that were still around but had dropped from the top tier. Gulf + Western become Paramount and was acquired by Viacom. New to the top tier in 1997 were Bertelsmann, Viacom (with Paramount), Sony, News Corp., TCI, Thomson, Seagram (with MCA) and Polygram.
• Indeed, fully half the names in the 1997 list were not in the top 50 in 1986. In some cases they were too small in 1986 but grew rapidly (e.g., Cox Enterprises, Cablevision). In other cases, they were new to the U.S. market (e.g., Bertelsmann, News Corp.). Others reflect new owners and new names for old players (e.g., Sony, which renamed Columbia Pictures; Seagrams, which renamed MCA). Yet others were companies that are totally new to the media industry or did not even exist (e.g., AOL, SBC Communications, Primestar, Hughes Electronics/ DirecTV).
• Of the 25 names from 1986 that were no longer in the top 50 in 1997, 15 disappeared as the result of mergers and acquisitions. The other 10 simply did not grow fast enough to stay at the top. They are identified Table 5.
• The total media industry's revenue nearly tripled from 1986 to1997, while the economy as a whole did not quite double.⁽³⁵⁾ Thus bigger media companies did not necessarily grow in relative size to the industry.
• The role of synergy in mergers may play themselves out differently depending on management as well as product factors. For example, in 1986 Capital Cities/ABC and Disney added together accounted for 5.5% of media revenue. Time plus Warner plus Lorimar Telepictures plus Turner Broadcasting were 9.0%. After its merger with Capital Cities/ABC, in 1997 Disney was at 7.2% of revenue, while the combined Time/Warner/Turner was 9.2%. In relative terms, therefore, Disney showed much greater true growth, perhaps due to synergy among the pieces. That is, above and beyond its growth from mergers it generated growth greater than the overall media industry. Time Warner increased its relative size only marginally beyond what the combined companies would have been.
• There has been a pronounced shift in the nature of the players in Table 3. In 1986 five of the top 12 companies ( Gannett, Times Mirror, Newhouse, Knight Ridder, Tribune) were best known as newspaper publishers, though with substantial other print and electronic media interests. A sixth (Capital Cities/ABC) also had a group of large city newspapers. By 1997 there were no newspaper publishers in the top tier. Thomson, which still did have a large division composed of very small dailies, received most of its revenue from electronic information services, magazines and books. It was in the process of divesting itself of its newspapers. News Corp., which owned the New York Post and Boston Herald, among others was out of that business as well. Thus, electronic media owners were displacing the old guard print media at the top of the media industry.

Policy Implications of the Trends in Media Ownership

Policy, of course, is determined by more than lofty ideals of what is right or wrong, what is best for society, or what is technologically feasible. In the case of media concentration and ownership issues, policy combines at least four separate factors: the legal, economic, socio-political and technological.

Legal Factors. Those who have followed the attempts of successive Congresses in trying to rewrite the Communications Act of 1934 are well aware of the political booby traps in policymaking. Any time a part of a bill deregulated one piece of the pie, some new player appeared to either claim injury or a piece of the pie itself. The outcome, the Telecommunications Act of 1996, was therefore written in general terms, leaving most of the decision-making to the Federal Communications Commission.

The process of changing FCC policy can be torturously slow, especially when the guiding document is as vague as the 1996 Telecommunications Act. A simple content analysis of the wording of the Telecommunications Act tells a tale. "Deregulation" was mentioned twice, More ominous, however: Regulation (and its derivatives) -- 202 times. Fair or unfair -- 35 times. Reasonable or unreasonable -- 37 times. The "FCC shall" -- 94 times and The "FCC may" -- 30 instances. The Act called for 80 proceedings to be initiated by the FCC.⁽³⁶⁾ This wording lead to continued reliance on the courts, administrative proceedings and appeals by the losing parties.

Developments such as the pace for introduction of digital television or availability of access to broadband data services could also be hampered by lack of regulatory direction. In the case of digital TV, local station license holders were required to implement digital broadcasting in exchange for being given free digital spectrum. However, cable operators had no obligation to transmit digital signals and, in 1999 seemed to be in no rush to make the investment to do so.

Economic Factors. While it may be a pleasant fantasy (except to the incumbents) to wish there could be two or three independent newspapers in every city or 15 radio stations in every town and village, the reality is that the economic infrastructure does not support such dreams. Indeed, the limitation on the number of radio stations in most parts of the country is not due to spectrum scarcity any more than the number of newspapers in a town is related to lack of printing presses. There is just not a large enough economic base to support more broadcasters or newspapers. The implications of this reality for public policy-makers was recognized in a congressional staff report in 1981. It noted:

Since scarcity due to economic limitations does not provide a rationale for regulating other media, a strong argument can be maintained that such a rationale should not be a basis for broadcast regulation either.⁽³⁷⁾

Similarly, it may be argued that the tendency toward mergers and acquisitions in cable is in large measure the result of the economic demands being made of cable systems. Initially there were the costs of wiring entire towns and cities, the poorer areas along with the middle-class neighborhoods. Typically local franchise authorities required the cable operator to provide neighborhood studios and programming funds for public access channels, link the city's educational facilities as well as the government offices together and remit a franchise fee to the city in addition. In the late 1990s there were massive new investments to provide expanded channel capacity for all the new program services, for digital services, and most recently for switched telephony. Small firms could not handle these demands. So the older cable systems combined with larger, better financed systems.

The role of increased competition is an economic force as well. Among the competitive factors that are changing the economic models of the media are the doubling in the number of broadcast television networks since 1985, the availability of cable to over 90% of households, the vast number of channels available to the three fourths of households that have cable or satellite services, the news, entertainment and information available on the Internet. Meanwhile, none of the old industries have faded away. But new industries and players have been added: the online aggregators lead by America Online, Yahoo, Excite and Lycos; the financial services players, such as Bloomberg and Intuit; the Internet Service Providers, including the regional Bell telephone companies; and others, such as Microsoft with both content (financing the Slate online magazine) and aggregation, with the Microsoft Network. E-commerce has a profound economic impact on many of the older media: books or toys sold by online merchants erode the sales of retailers who might have to cut down their newspaper or magazine advertising schedule. On the other hand, the online merchants have been paying some of those same publishers and broadcasters for banners and links from their Web sites and advertising on radio and television.

Socio-Political Factors. Social factors are related to political factors. In this case, the real question is, "How much diversity is enough?" And a corollary question is, "How is that determined?"

If it is generally agreed that the antitrust standard for concentration as applied to the media would be insufficient to fulfill the objective of having many unaffiliated "voices," there is no acceptable guideline for what constitutes too few voices. It cannot be seriously proposed that the mass communications business must be so structured that any person or group can have unlimited access to whatever medium for whatever purpose for whatever period of time they so desire. Short of that impractical standard, what is acceptable and how can that be determined?

The issue of media control is particularly important to many critics and analysts because of the presumption of the media content's great influence on mass society. Those who control the media, goes the argument, establish the political agenda, dictate tastes and culture, sell the material goods and in general manipulate the masses. While there is certainly great power in the media, for two related reasons its strength may also be overemphasized.

First, so long as there are reasonably competing media sources as there are today, these can cancel each other out. Why is it we do not all eat Wheaties or use Exxon gasoline? Second, there are media other than the "big" media that can be very effective, especially for reaching easily identified groups. Indeed, replacing the fear that society is the victim of a few mass media moguls is a new specter of such a fragmented media landscape that society becomes captive to narrows interests, following the news groups on the Internet and the myriad of Web sites from which individuals assemble their own, almost unique stew of content.

The use of media in the 1978 Iranian revolution was an early prototype case study.⁽³⁸⁾ ln the typical coup d'etat, the rebel forces are supposed to take over the television and radio stations. The government meanwhile imposes censorship in the press. The Iranian revolution succeeded without the Ayatollah Khomeini overrunning a single broadcast facility. The Shah had control of all the media to the day he left. The revolutionary forces relied quite effectively on the "small" media. Khomeini used audio tapes to get his message to the mullahs, who in turn spread the word in the mosques. The Xerox machine, Everyman's printing press, was used to distribute his instructions. And the telephone was used to coordinate efforts between Teheran and exile headquarters in Paris.

Now, the Internet has taken its place among the applications of information technologies in a political setting. Here is how one report started:

As rebellions broke out across Indonesia... protesters did not have tanks or guns. But they had a powerful tool that wasn't available during the country's previous uprisings: the Internet.

Bypassing the government-controlled television and radio stations, dissidents shared information about protests by e-mail, inundated news groups with stories of President Suharto's corruption, and used chat groups to exchange tips about resisting troops. In a country made up of thousands of islands, where phone calls are expensive, the electronic messages reached key organizers.

''This was the first revolution using the Internet,'' said W. Scott Thompson, an associate professor of international politics at the Fletcher School of Law and Diplomacy at Tufts University.⁽³⁹⁾

Still, the perception no doubt persists that the mass media are all powerful in the industrialized world, so this factor will be a dominant force in determining policy.

Technological Factors. Technological factors are addressed last to emphasize that they are only one of many interacting factors. With the rapid advancement in developments of microprocessors, telecommunications processes and software, it sometimes seems that the communications world is technology driven. The preceding sections indicate that technology interacts with other forces. History seems to provide several lessons about the role of technology in change.

First, technology is rarely adapted for its own sake. It must fulfill some need. Off and on from the 1960s, the Bell System and its successors tried to introduce PicturePhone' service. It did not catch on. Also during the 1960s and 1970s the educational establishment tried to implement computer-aided instruction. It too failed miserably. In 1978, the government-owned telephone system in Great Britain, looking at its underutilized network, initiated an electronic data base service for the home market, dubbed Prestel. They expected it to have 100,000 households subscribing by the end of 1980. It had fewer than 10,000.

Second, whereas is was not unusual for some technology to take five or more years to get from discovery to commercial availability, the rate of time to the marketplace seems to have contracted. That gives existing industry participants less time to adjust. The ubiquitous telephone was not in place in 50% of U.S. households until 1946, 70 years after its invention. But the graphical browser for hypertext was invented in late 1990 and introduced in 1991. The working version of Mosaic, the first browser for PCs and Macintosh computers, was made available mid-1990s and its commercial version, Netscape, was available in 1994. By 1998, 83 million U.S. adults -- 40% of the over 16 year old population, had access to the Internet.⁽⁴⁰⁾

Finally, there is an important difference between that which is technologically feasible and what is economically viable. Indeed, the technological graveyard is littered with better mousetraps that failed because they cost too much. What will the technology do, at what price and what will it replace are questions that must be resolved as part of the policymaking process. One example was the uncertainty about digital television and high definition television in 1999. The original motivation of moving to digital TV was to make it feasible to provide a wide, high definition picture while using the same bandwidth (6 MHZ) as older analog transmission. However, using bandwidth compression it was also feasible to use the 6 MHZ of digital space to offer four or five digital channels at a similar resolution and size as the NTSC standard. Should high resolution be employed to show the "talking heads" of a newscast just because it is available? Or should it be used at the discretion of video distributors -- based on audience considerations -- for events such as action movies or sporting events? The answer may well help determine how many channels of video are available to consumers and who may have the ownership of them.

Implications for Public Policy

One critical question that has faced government agencies over the years was to what degree they could and should be more concerned about concentration in the media than other industries. Should a stricter standard apply to the media than to other industries because of the media's position in American society and the importance of having many channels available for speech? Can free speech be separated from the economic structure that controls the media? Should the government promote diversity and independence to avoid having to regulate?

In 1979, the antitrust division of the Justice Department investigated the merger between newspaper giant Gannett and Combined Communications, with its extensive broadcast holdings. However, a top Justice Department official admitted:

The antitrust laws do not flatly prohibit media conglomerates any more than they prohibit other kinds of conglomerates. Under present law, some measurable impact on competition in some market must be proven before a merger or acquisition will be held to violate the antitrust laws. Indeed, the courts have been generally reluctant to condemn conglomerate mergers where such an impact has not been shown, regardless of the social or other objections that have been asserted.⁽⁴¹⁾

Government policymakers are faced with challenges to long standing practices. At the top of the list are decisions on defining the product and geographical boundaries for the old and new media industries. It is perhaps nonproductive in the longer run to focus on the concentration of media ownership using conventional concepts of newspapers, television, magazines, etc. Rather, the criteria that government policymakers must be concerned with instead may be a more generalized goal: encouraging diversity of conduits for information and knowledge. Do the major news weekly magazines have direct competition from newspapers, televised news programs, and all news cable and radio programmers? Do motion picture distributors compete with book publishers and certain periodicals? Do special interest magazines, compete for advertiser dollars, consumer dollars and time with niche video programming?

Questions and More Questions

There remain questions that need to be considered in the discussion of policy formation. A selection of such questions includes:

• Does increased diversity and access imply greater quality? What happened when the FCC took 30 minutes of prime-time programming from the three networks (via the Prime Time Access Rule) and forced this time on the individual stations? The prohibitive costs of single market productions resulted in few quality shows and opened up the market to syndicators of low-cost game shows of little substance -- and great popularity. On the other hand, cable television has spawned the Discovery and History Channels, among other quality and small audience niches.

• Who should be the arbiter of what type of programming or content is most desirable for society? Much of the criticism of the old broadcast networks centered on the supposedly mindless grade of the programming. However, when given a choice, the viewing public has "voted" by the way it clicks the remote. Many of the top-rated shows have outperformed presentations of supposedly higher intellectual content. But the sheer volume and variety of books, magazines and video seems to be pushing publishers and programmers in directions that fill ever smaller niches of all variety and quality.

• How much control by any firm or group of firms must be manifest before we are threatened with perceivable restraints on true access to a broad spectrum of opinion and information? Most crucially, how can this be measured? On the one hand, there is a point at which some combinations may have to be limited. On the other hand, there can be no credence given to the argument advanced by some that every opinion or creative idea has a right to be heard through the mass media. However, anyone with a few dollars can make up a picket sign or hand out leaflets at City Hall or create a Web page or post a message to dozens or even hundreds of Internet News Groups. Can concentration of ownership be measured by the total number of media properties? By the number of households reached by the media owned by a given firm? By the geographical concentration of the firm's properties?

Conclusion: More Concentration or Greater Competition?

Are the mass media now unduly concentrated, heading toward dangerous concentration, or are they and will likely remain sufficiently competitive? The answers depend on what is to be measured, whether it can be measured, what judgement policy-makers want to apply to the findings.

Looked at in small, industry-specific pieces, there are trends toward both consolidation and greater competition. In this study we saw that by one criteria book publishing is not concentrated, but marginally more than in the recent past. Broadcasting is moderately concentrated, but less than in the past.

Looked at as a single industry, there can be little disagreement that there is, overall, more competition than ever among media players. The issue could be stopped with a single word, Internet. But it goes beyond this development.

The combination of Twentieth Century Fox with News Corporation's television stations helped create a fourth television network. The wiring of the cities with coaxial cable has created an infrastructure of scores of programs and hundreds of channels. The introduction of satellite receivers for the cost of what a terrestrial home TV antenna used to cost has provided a measure of competition for cable operators. Computerized data based management has provided direct mail with every greater accuracy as an alternative to advertising in newspapers and magazines. As seen in Table 5, the owners of these outlets remain many, diversified and in constant flux.

But it is and will be the Internet that ultimately appears to erode many of the old notions of bottlenecks. Users can easily and cheaply access essentially any newspaper from almost anywhere. Musicians that are not offered any or decent recording contracts can distribute via the Internet. (And listeners who can't find the type of music they like can probably find it on the Internet). Publishers who have titles that are not bought by a bookstore chain can get ready distribution via online booksellers or sell economically direct to customers. Home sellers and car dealers and anyone else dependent on local newspaper classified ads rates can use online options instead. Government agencies, public service organizations, indeed, any organization or individual with a message that it cannot get covered in the traditional media can get it out, often with startling speed and coverage, using the Internet. Conversely, consumers of all stripes who want some type of information can, sometimes with little effort, sometimes with the need for search skills, find most of what they may want. This includes specific needs -- how to find out more about Lyme disease, for example -- to pure browsing. And there is every indication that the capability to disseminate as well as the ability to aggregate will get more accurate and require lower skill levels.

The difference between the Internet and newspapers, books, records or television is that it can be all those things. There may be large players who continue to provide content, packaging, and promotion that make them popular providers via the Internet. Unlike the older media, there are not the high regulatory and/or capital barriers to entry using the Internet. If it is diversity, accessability, and affordability that are society's goal for the media, then the Internet appears to have laid the foundation for its success. For better or worse.

Notes

1. Copyright © 1999, Benjamin M. Compaine. All rights reserved, hard copy and electronic.

2. Benjamin M. Compaine, et al, Who Owns the Media? Concentration of Ownership in the Mass Communications Industry, 2^nd ed. (White Plains, NY: Knowledge Industry Publications, Inc., 1982), p. 3.

3. Matt Drudge, "Anyone with a Modem Can Report to the World," transcript of speech to The National Press Club, Washington, DC, June 2, 1998. Accessed May 5, 1999 at http://www.frontpagemag.com/ archives/drudge/drudge.htm.

4. Drudge transcript.

5. Ibid.

6. For a historical review of such concerns and efforts, see Benjamin M. Compaine, et al, Who Owns the Media?, 2nd ed. (White Plains, NY: Knowledge Industry Publications, 1982).

7. Anthony G. Oettinger and John F. McLaighlin, "Charting Change: The Harvard Information Business Map," Chapter 10 in Benjamin M. Compaine and William H. Read, eds., The Information Resources Policy Handbook: Research for the Information Age (Cambridge, Mass.: MIT Press, 1999).

8. Derrick C. Huang, "Size, Growth and Trends of the Information Industries, 1987-1996," Chapter 11 in B. Compaine and W. Read, eds. The Information Resources Policy Handbook, (Cambridge, Mass: The M.I.T. Press, 1999), pp. 347-361.

9. Ibid.

10. Reed H. Blake and Edwin D. Haroldson, A Taxonomy of Concepts in Communication (New York: Hastings House, 1975), p. 34.

11. Ben H. Bagdikian, The Media Monopoly, 4^th ed.(Boston: Beacon Press, 1992), p. xxiv.

12. Ibid, p.xxxi.

13. Benjamin M. Compaine, "Shifting Boundaries in the Information Marketplace," Journal of Communication, Vol. 31, No. 1 (Winter 1981), pp., 132-133.

14. Clayton Act of 1914, ch. 323, § 7, 38 Stat. 730, 731-32 (1914). There are two other statutes that comprise the core of federal antitrust law: The Sherman Anti-Trust Act of 1890. Section 1 of the Sherman Act forbids contracts, combinations, and conspiracies that are in restraint of trade. Section 2 of the Sherman Act prohibits monopolization, attempts to monopolize, and conspiracies to monopolize. The other principal antitrust statute is the Federal Trade Commission Act of 1914, which, as amended, prohibits "unfair methods of competition" and "unfair or deceptive acts or practices."

15. The detailed legal and economic analysis of antitrust are beyond the scope of this section. Highly recommended for the basic of economic concepts such as cross elasticities, especially as applied to the media, is Alan B. Albarran, Media Economics (Ames, Iowa: Iowa State University Press, 1996).The legal analysis drawn on here is from H. Peter Nesvold, "Communication Breakdown: Developing an Antitrust Model for Multimedia Mergers and Acquisitions," at http://www.vii.org/papers/peter.htm.

16. ' ""

17. Associated Press v. United States 326 U.S. 1 (1945) at 20.

18. United States v. Paramount Pictures, 334 U.S. 131.

19. Citizen Publishing Co. v United States, 394 U.S. 131 (1969).

20. According to Nesvold, note 365 (citation omitted): "See In re Multiple Ownership, 50 F.C.C.2d at 1056 n.11. According to the DOJ, newspapers, television stations, and radio stations are all engaged in the same business of attracting audiences and selling them to advertisers. While the DOJ does acknowledge that the three are not interchangeable for all advertisers, it asserts that the three are far more alike than they are different. Also, at least one commentator has suggested that there is some substitutability between news on the radio and on other media. In New York City, for example, consumers may obtain local news, social calendars and sports information from: (1) local newspapers, including the New York Post and New York Newsday ; (2) news radio stations, including WINS Radio, 1010 AM; and (3) New York 1, a 24 hour all news cable channel that focuses on events within and concerning New York City."

21. 714 F.2d 351 (4th Cir. 1983), cert. denied, 465 U.S. 1027 (1984).

22. 825 F.2d 1559 (11th Cir. 1987).

23. Ibid, at 1563.

24. 712 F. Supp. 1386 (N.D. Cal. 1989), aff'd , 903 F.2d 659 (9th Cir. 1990).

25. Nesvold, at note 353.

26. TNT, part of Turner Broadcasting, became part of Time Warner in late 1996, well after the decision to promote the movie.

27. Stephen Schurr, "Shawshank's Redemption," The Wall Street Journal, April 30, 1999, p. B4. Subsequently, the low budget film The Blair Witch Project owed much of its break-out success to a carefully crafted Web site. See Newsweek, August 17, 1999.

28. Ibid.

29. Accessed May 3, 1999 at http://www.imdb.com.

30. Shurr, "Shawshank's Redemption," p. B1.

31. S.A. Rhodes, "Market Share Inequality, the HHI and Other Measures of the Firm Competition of a Market," Review of Industrial Organization, 10, pp. 657-674. See also A. Golan, G. Judge, and J. Perloff, "Estimating the Size Distribution of Firms Using Government Summary Statistics," Journal of Industrial Economics, 44, pp. 69-80. This study demonstrated the efficacy of the HHI in a study of 20 industries.

32. Albert N. Greco, "The Impact of Horizontal Mergers and Acquisitions on Corporate Concentration in the U.S. Book Publishing Industry, 1989-1994," Journal of Media Economics, 12(3), 1999, pp. 165-180.

33. This is drawn from Who Owns the Media?, 2^nd ed., Table 6.6, subtracting estimated nontelevision revenue and adjusting for inclusion of only 16 rather than 20 firms in the 1980 data.

34. Pool, p. 28.

35. Media industry revenue was up 188% in current dollars, while GDP was up 83% in current dollars (from Table 715, 1998 Statistical Abstract).

36. Benjamin M. Compaine, "Regulatory Gridlock and the Telecommunications Act of 1996," National Cable Television Association Academic Seminar Keynote Address, Atlanta, GA

May 2, 1998. Accessed April 28, 1999 at http://www.cablecenter.org/Main/INSTITUTE/ speech.cfm?SelectedSpeech=8

37. U.S. Congress, "Telecommunications in Transition: The Status of Telecommunications in the Telecommunications Industry," report by the Majority Staff of the Subcommittee on Telecommunications, Consumer Protection and Finance, U.S. House of Representatives, Nov. 3, 1981, pp. 310-325.

38. Majid Tehranian, "Iran: Communication, Alienation, Evolution," Intermedia, March 1979, p. xiii.

39. 38.David L. Marcus, "Indonesia Revolt was Net Driven," The Boston Globe, May 23, 1998, P. A1.

40. Estimates of Internet use (which may include e-mail but not necessarily World Wide Web) vary, depending on definitions of "use" and what age groups are included. This survey was done by Intelliquest and was accessed April 29, 1999 at http://www.nua.ie/surveys/?f=VS&art_id=905354866&rel=true

41. I. William Hill, "Justice Department Probes Gannett-Combined Merger," Editor & Publisher, March 24, 1979, p. 11. Quotes John H. Shenefield, then Assistant Attorney General for antitrust.