Cooperative Integration: A Structural Geography of Primetime Program Production and Broadcast

Tonya Lindsey

[WJMCR 8:4 September 2005]

Sections:

Introduction|Research Questions|Hypotheses|Method|Results|Conclusion

Abstract

This article asks: what is the influence of policy change on primetime TV programming, from production to broadcast? To understand the structure of primetime programming from production to broadcast, social network analysis is used to study the pattern of ties between parent companies of production and broadcast outlets using primetime programs as the ties. A hierarchical cluster analysis, structural equivalence profile, multidimensional scaling, and an indegree/outdegree procedure are each employed to measure the similarity of ties between parent companies of production and broadcast outlets from 1989, 1994, and 1999. The years studied reflect before, during, and after the time The Financial Interest and Syndication Rules were rescinded and The Telecommunications Act of 1996 was enacted. The results illustrate that, overtime, the ties between and among parent companies grow more and more similar, with independent producers being utilized less frequently and large parent companies both producing and broadcasting more primetime programs. The arbitrary thresholds measuring monopoly and oligopoly are bypassed in favor of a more dynamic analysis of how companies can be cooperatively integrated.

Introduction

In 1993 the Federal Communications Commission rescinded the Financial Interest and Syndication Rules (Fin-Syn rules) allowing owners of television and cable networks (broadcast networks) to own the programs they broadcast. Later, in 1996 Congress passed, and President Bill Clinton signed, the Telecommunications Act of 1996 changing the 1954 Telecommunications Act to allow media companies to own both more networks and for those networks to reach a higher percentage of the population than previously allotted. The goal of this study is to explain the structural shifts, over an 11-year period, in ownership of primetime programs and the networks they are broadcast on.

Over the 11-year period there have been a decreasing number of parent companies producing or owning the programs broadcasted during primetime. The companies are increasingly more alike in their production to broadcast patterns of primetime programs; their business practices are increasingly similar. Overtime, with mergers and acquisitions, fewer companies have limited choices in who owns programs to broadcast and which networks the programs can be broadcast on. To understand the shifting structure overtime, data from 1989, 1994, and 1999 have been analyzed using UCINET 5.0�s1 multidimensional scaling (MDS), structural equivalence profile, in/outdegree procedure, and hierarchical cluster analysis procedures. I employed these procedures to describe the pattern of ties representing primetime program production to broadcast among parent companies and to develop a visual representation of the structure of the industry.

Concentration and Ownership

Measuring the degree of concentration in the industries in question is one way to discuss the effects of policy and legislation shifts. Some authors believe, and have shown statistically, that there has been a decrease in the concentration of television programming. Specifically, one of Noam and Freeman�s conclusions in their article, �The Media Monopoly Myth,� is that television programming is actually less concentrated than in previous years. The authors� state, �Concentration in TV programming dropped with the launch of new broadcast and cable networks.�2 This statement is supported with the percentages of the top four firms�3 share in the three major TV networks, which do indeed decline.

The claim that concentration has decreased is misleading. Measuring concentration of programming by the number of broadcast networks does not measure concentration in the sense of ownership; it explains how programming is dispersed over new networks. As new networks appear there must be more programming to fill up broadcast time. For example, if there are two, new networks created there will have to be two, new programming schedules requiring many programs to fill otherwise empty airtime. While additional programming does support a decrease in concentration of content as the content is spread out, decreasing concentration of programming does not indicate concentration of ownership in the shares of networks, a better measure of ownership. Without measuring ownership of the programming how is a media monopoly tested as a myth or not? To test for concentration in programming ownership of the programs can be used instead. Further, in regards to the top four firms� share of the three major networks, why did their share (holdings) decrease? In an analysis of the consolidation trend in television stations Howard found,

�there [had] been evidence of significant ownership consolidation since 1995 when passage of the Telecommunications Act was anticipated. This recent trend is shown by the declining number of group owners from 210 in 1995 to 184 in 1997�In the consolidation process, the decline in the number of group owners was accompanied by an increase in the number of TV stations per owner.4

With a decline of the number of group owners in mind, it may be reasonable to conclude if there are more broadcast networks, then the top four firms divested themselves of some of their share in the top three networks in order to create the new networks.5 Shares in the top three networks may have declined, but overall with the addition of new networks, shares and/or ownership in/of all of the networks may have increased. For example, if the top four firms A, B, C, and D have decreased their shares in networks 1, 2 and 3 and simultaneously created networks 4, 5, 6, and 7, does it necessarily mean concentration in ownership of networks has decreased overall? To test for concentration in ownership of networks a broader measure of ownership including all of the participating companies and networks is more complete.

This study challenges Noam and Freeman�s conclusion of decreasing concentration by first, locating the parent companies producing primetime programs, second, locating the parent companies that own the networks on which programs are broadcast, and third, aggregating the ties among the parent companies responsible for primetime production to broadcast, program by program. This will offer a more accurate representation of concentration in the two industries�, motion picture6 production and broadcast, by including all of the participating companies and the relationships among them. In this case concentration is more than percent ownership, shares, or based on content; it is the dynamic relationship among the companies involved.

Another way to discuss the effects of policy and legislation is to make claims that ownership is not only concentrated but also a monopoly or oligopoly. The question of whether or not there is a media monopoly is a question that has been asked over and over again. Industry sources such as Television Quarterly have published studies claiming a media monopoly is a myth7, while progressive publications such as The Nation have focused on the deleterious effects of consolidated media power both domestically and abroad.Books have been written examining the media as a monopoly9, and about the power of concentrated ownership in news media to frame political issues.10 Further, more has been written concerning cultural and psychological issues: the �determined� effects of media technology on collective, cognitive functioning of the masses11, the effects of the aesthetic quality, or lack of aesthetic quality, of mass culture on high culture and society12, and the framing of mass culture as a means for hegemonic control.13

While each of the above approaches is interesting and rich in information, my focus is to discuss ownership in terms of the relationship between policy and structure. I want to measure changing relationships among companies in terms of social network analysis. The idea behind employing social network analysis is to understand the structure through a geography of ownership. The geography is mapped according to ties among companies rather than simply which companies won how much of a market. Using structural equivalence measures will show how alike the companies are in their patterns, and the MDS will illustrate how those patterns look in a spatial model of the structure. Overtime, these measures will show if and which companies are becoming more similar in their production and broadcast patterns. Increasing similarity could mean that, overtime; certain companies are producing and broadcasting more of each other�s programs.

Although the issue of the extent of concentration (monopoly or oligopoly) is prevalent in the literature on media ownership, I am not attempting to report conclusive results affirming or denying claims of monopoly or oligopoly. To affirm monopoly would be to find only one company owns all of the media industry in question, a claim I do not expect to have supported in my research. Alternatively, though concentrated media ownership by a few corporations indicates oligopoly, something I do expect to find, I don�t want to define an arbitrary threshold beyond which oligopoly exists and within which it does not.

Ramstad14 reports there are three levels to analyze structure. Company level analysis seeks to understand a single company. Industry level analysis studies the interactions between companies within the same industry. Finally, researching at the market level focuses on either the interindustry relations or the entire market. In Ramstad�s terms I am analyzing at a market level the interactions between the television production and broadcasting industries overtime.

In this article I do the following: First, the interaction between technology, policy, economics and structural change is examined to provide the background for my discussion of policies that have created shifts in the structure. Second, academic and industry studies and opinions of entertainment and news media concentration are examined for posited effects of structural changes. Third, issues concerning democracy, capitalism, and media are presented. The fourth and final part is the empirical study of the relationship between companies involved in the production and broadcast of primetime programming before and after significant policy shifts.

Research Questions

Two research questions arise concerning the organization of the broadcast and program industries. First, what has happened structurally in the primetime production and broadcast industries before and after the Fin-Syn rules were rescinded and the Telecommunications Act (1996) was enacted?

It is my assertion a social network analysis will show the increasing similarity in ties among a small number of corporations that dominate production and broadcasting. Theoretically, the increasing similarity creates more bridges between companies that already dominate the industries. The repercussions of the move towards more bridges are on the one hand increasing profits for large, parent companies and on the other hand, decreasing profits for smaller, independent companies and individuals.

A second and related question is: overtime, has the increased similarity led to cooperative integration? Cooperative integration is loosely defined as the practice of companies across industries relying primarily on one another or themselves for product and/or distribution (in this case broadcast) of product. Unlike vertical integration, where one company produces and distributes its own product, cooperative integration occurs between companies. The cooperation is not necessarily akin to a monopoly or an oligopoly in the sense explicit price-fixing takes place; however, the effect may be similar in that independent and smaller companies may find it more difficult to participate in the market as large, parent companies produce programs that it, or another large media company, will also broadcast.

One effect of cooperative integration may be companies airing programs they have produced or co-produced could inhibit independent writers and producers who do not have previous successes. Bielby and Bielby15 explain that programmers frame primetime programming before it is scheduled:

�to justify their actions in highly institutionalized contexts. In developing new primetime series, network programmers are compelled to provide legitimate accounts for their decisions (Meyer and Rowan 197716). We argue that they do so by organizing discourse around widely accepted frames. Specifically, in describing series in development, network executives invoke the framing devices of genrereputation, and imitation.

The framing devices most often used by networks are the reputation of the producer or a celebrity. Thus, if a program is written or produced by an unknown author, gatekeepers with established frameworks for marketing new programming more often than not bypass the unknown work in favor of that which they can frame in the standard rhetorical discourse. Further, concentration and consolidation shrinks the number of resources an independent writer producer may have to sell their wares to, and may, overall, not be conducive to the untested and innovative.

Background: The Industries, Policy, Technology, Ecomomics and Change

Two pieces of legislation are important in discussing how the television programming industry operated in 1999 as opposed to how it did in 1989. Rescinding the Fin-Syn Rules and enacting the Telecommunications Act of 1996 loosened regulations on the television industries, and in response the structure of the industries began to shift as mergers and acquisitions that previously could not have happened took place.

In 1970 the Federal Communications Commission (FCC) implemented what are known as the Financial Interest and Syndication (Fin-Syn) rules. These rules were developed to constrain the then major networks� (ABC, CBS, and NBC) ability to control development and syndication of television programs through financial interest or syndication rights in programs that were broadcast on each network.17The original intent of the rules was to ensure there was a competitive market for myriad of producers. Many producers were thought to develop diverse programming that could then be sold into syndication for the profit of the producer. Diversity was at stake.

By 1993, cable and new television networks had expanded the range of choices to which many television viewers had access. With the expansion of cable outlets came the idea that more outlets equaled more diversity and a decrease in the ability of big corporations to monopolize production and/ or syndication rights of programming. When a U.S. District Court ruled networks were no longer subject to many of the Fin-Syn regulations, the market for creative and diverse programming was understood to open up as numerous, new cable channels which could �narrow-cast� to smaller audiences with diverse tastes.18 Diversity was believed to increase because there were so many more networks to choose from whether or not production studios or networks were producing the programming.

Even before the rescinding of the Fin-Syn rules, the preliminary discussions opened up what has been called the fight between Hollywood and the networks. The argument over which side would have creative control and access to the profits from programming developed as both producers and network executives developed �rhetorical strategies that associated their own interests with the values of creativity, diversity, and quality programming.�19 The rhetorical strategies are perhaps what prompted Schrage20 to write,

Get real. The networks don�t care about programming diversity, they care about return on investment. (Indeed, the diversity argument is particularly moronic given the rumors that Disney or another studio would buy CBS, or that General Electric/NBC might buy a studio if the restrictions were lifted). Similarly, the studios and independents care far less about �diversity� than protection from network customers who would subsequently become their fiercest competition. Please, I love stories about entrepreneurial independents as much as anyone, but it�s crazy to think that they would shrivel or vanish in the face of network competition. Yes life would get tougher. Too bad. Remember, the role of the FCC is not to maximize shareholder value for CBS or enable a sitcom producer to build an addition to her Malibu beach house: It is to protect the public interest. (italics added)

Similar to arguments concerning diversity are ideas of pluralism in media content. Cavallin21 has explained media concentration using an early definition from The Council of Europe.

In relation to media concentrations, the notion of pluralism is understood to mean the scope for a wide range of social, political and cultural values, opinions, information and interests to find expression through the media.

Pluralism may be internal in nature, with a wide range of social, political and cultural values, opinions, information and interests finding expression within one media organisation, or external in nature through a number of such organisations, each expressing a particular point of view.

Content is emphasized in this definition of concentration. With good reason, notes Cavallin, because the assumption that a lack of concentration in media ownership indicates pluralism of content has been proven wrong. Just because there are many companies producing does not mean content is diverse. The proof of this assertion is cited in the example of the Soviet Union. The former Soviet Union supported many media producers from the Communist Party, none of whom used their programs to advocate the principles of the European Convention of Human Rights. Many media producers do not necessitate diverse information if the sources advocate one particular worldview. Alternatively, the idea a small number of strong media companies indicate homogeneity of content is also not necessarily true.

Collette and Litman�s22 study of broadcast network entry offers different effects from additional networks.

�there are expected consumer welfare gains in the form of increased program diversity that may be realized through the addition of new networks. As the number of networks rises, the number of viewers a new entrant can expect to attract by duplicating programming of incumbent networks will decline, thus making the most profitable strategy one in which a new program type is offered.

As competition for audiences rises with additional networks there will be more different types of programs produced and broadcasted.

The structure of U.S. media was headed for further change when three years after the 1993 rescinding of the Fin-Syn rules, the Telecommunications Act of 1996 was signed by President Clinton.23 Surpassing the effects of the Fin-Syn rules that were narrowly targeted at the television industry, the Telecommunications Act of 1996 took on prior ownership restraints on radio, television, cable systems, newspapers, phone companies and long distance service.24 Hickey25 specifies that the 1996 Act changed regulation that had been in place since 1934, and

�removed all limitations on the number of radio stations one company can own nationally, and allowed up to eight per company locally (instead of only four); relaxed the rules about how many TV stations one company can operate; ordered the FCC to consider easing the rule limiting ownership to one TV station per market, as well as the bar to ownership of a newspaper and a broadcast outlet in the same city; permitted common ownership of cable systems and broadcast networks; ended all rate regulation of smaller cable TV systems and promised the same for large ones later on; extended the license term of TV and radio stations to eight years from four; allowed TV networks to start and own another broadcast network if they choose; required that all TV sets come equipped with a V-chip to help screen out violent and sexually explicit shows; imposed prison terms and fines on anybody who transmits pornography over the Internet.

Economists and communications scholars writing about media economics focus on The Telecommunications Act of 1996. For these two groups of scholars, The Telecommunications Act of 1996 ushered in economic transactions in the forms of mergers, acquisitions, and consolidations of media companies. Compelling these transactions is strategic reasoning that assumes the reduction of risk in the television industry.26 In particular, technology has precluded the strategic reasoning behind mergers and acquisitions that were to come.

Before President Clinton signed the Telecommunications Act of 1996 into law, digital spectrum technology was poised to bring sharper images to viewers through high-definition television (HDTV) powered by digital versus analog systems of transmission. In the period between 1987 and 1993 U.S. electronics companies scrambled to create the technology for HDTV before the Japanese could. During the same time period the FCC decided to allow every television station27 in the U.S. to have one transitional channel to simultaneously broadcast their analog programming on by digital transmissions. This would allow for consumers to transition their old TV sets for new ones that supported the newly developed technology.28

Along with the discovery that the additional channel could and would transmit the digital signal the discovery was made that the single channel could be split into six channels of traditional, non-HDTV, channels. The free channel per station suddenly became six. The profits that could be generated from not one, but six channels, multiplied exponentially, and broadcasters successfully lobbied for their free channel(s) to remain free.29 In the same year companies owning television stations could own more stations reaching a larger number of people, those same companies received a free channel that was in fact six wrapped into one. Technology and policy effectively worked together to increase potential profits for companies who had just been allowed to have access to more broadcast outlets than before.

As for economic strategy, Howard30 writes,

With the enactment of The Telecommunications Act of 1996 and its more liberal approach to multiple-station ownership, a new era of consolidation of broadcast stations has begun. It is unlikely that this trend will abate in the near future. Indeed, this recent wave of consolidation may result in a stronger television industry that must exist and compete with other video delivery systems in a complex multimedia environment.

As media companies bought up more stations31 equipped with the free channel from the FCC, the parent companies then had access to the number of free channels owned by the acquired stations times six. The scope of delivery for programming parent companies previously could not own widened. The rescinding of the Fin-Syn rules worked well with the ability the parent companies had to broadcast programs they owned after the revised Telecommunications Act allowed ownership of more broadcast networks; more programs and more broadcast outlets might mean more profit.

Technology affected more than one industry deregulated in the Telecommunications Act. Fiber optics and digitization allowed for information to be transformed from medium to medium and delivered quickly.32 As telephone networks could provide video and data communications services, cable modems could carry high-speed data and voice switched messages, and computer networks could transfer voice and video data, the boundaries of the traditionally separated industries of telephone, cable, and computer were blurred. In a discussion of the blurring of boundaries in telephone, computer, and cable that was incited by new digital technology Tseng & Litman33 explain that,

The interaction between technology and conditions of entry creates a new dynamic of intensified competition and increased flow of new products. These factors fuel each other and intensify the process, creating an ever-accelerating spiral of new competition and barrier disintegration.34

Technology has led media economists to view the media industries as needing fewer restrictions on ownership of multiple forms of media to avoid competitive disadvantages. For already large telephone and cable companies, cable systems powered with digital technology congruent with telephone transmission could and did necessitate the process of mergers and acquisitions (M & As) so companies could keep up with one another. Deregulation in the television industries may have helped giant companies remain competitive with one another.

One area of policy that is sometimes neglected in the literature is any mention of the Fin-Syn rules. Howard 35attributes the merger frenzy of the mid ’90s to The Telecommunications Act.

Although the growth of multiple station ownership was a long-term trend in the television industry, 1995 saw an acceleration of merger activity fueled largely by anticipation of the passage of a new and liberalized telecommunications act. (italics added)

And, in a discussion of the merger between US West and Continental Cablevision Tseng & Litman36 discuss the vertical integration of system and programming without mention of the Fin-Syn rules. �Not only is it cheaper to acquire control vertically over programming than to buy from competitors, but it also minimizes risk.� In a discussion that does mention the impact of the Fin-Syn rules Collette and Litman37 state that

[The impact of the changes in regulation of ownership of programs] is important because programming and the associated sale of time represents the core business of the television networks. The expiration of the Financial Interest and Syndication Rules in 1995 permitted greater network ownership of programming, thereby increasing network incentives for vertical integration into production. The networks could attempt to reign in program costs and utilize their own network as a �launch pad� for the sale of content in syndication aftermarkets.

Not only could these companies �reign in� program costs for future sales they could also produce original programs for newly acquired networks.

Overall, in discussions of the strategic planning of media companies it is important to analyze the relaxing of rules concerning vertical integration of systems and content. Some of the anticipation of the Telecommunications Act of 1996 may actually be attributed to the consolidated libraries of content network owners bought after the rescinding of the Fin-Syn rules. Once these companies bought film libraries they pushed for a legal means to own more broadcast networks. Owning more broadcast networks meant more places to air acquired program libraries and higher profits from ad revenues. Profits stood to be even higher as companies owning previous hit programs like The Cosby Show virtually guaranteed audiences for advertisers.38

Simply lending or renting out the right to broadcast programs like The Cosby Show to another company is not worth as much as the revenues generated from the ability to market the show�s previous success to advertisers on a parent company�s own network. Caves39 describes what happened when Time Warner was going to sell film exhibition licenses to Sony,

A decision by Warner Communications to make a profitable sale of film exhibition licenses to CBS was stopped by Ted Turner, Time Warner�s largest share holder, who demanded that they go instead to Time Warner�s own cable networks that were part of his former Turner Broadcasting�

Further,

In the 1990s Disney, Warner and Paramount all integrated into broadcast network operations�Disney by acquiring ABC, Warner and Paramount by starting their own networks. The three studios accounted for one-third of the leading four networks� primetime programming, and the abolition of the Federal Communication Commission�s financial interest and syndication rules restored a strong incentive for the networks to supply themselves with programming. (italics added)

Film and television program libraries were thus kept in-house as M & As that could not have happened prior to changes in policy that allowed for companies to own both programs and the networks on which they were aired. Previous programming is just one of the benefits of owning both programs and broadcast networks.

In a discussion of strategic alliances through M & As related to the Telecommunications Act of 1996 Chan-Olmstead40 observes that,

Strategic alliances through M & A present an especially attractive avenue for telecommunications and media industries as these firms will be able to integrate different communication segments quickly, capture a developed customer base, consolidate smaller niches, and accelerate the implementation of new technologies with combined resources.

Further, she describes, there are

�the needs for corporations to identify and exploit interrelationships among related businesses to reduce costs or enhance differentiation.

Strategic alliances, where companies work to achieve an advantage in the market by combining operations and sharing risks41, work well in industries propelled by the �nobody knows� property, a concept that originally referred to the inability to predict the success of a motion picture, but is used here to describe the inability to predict the success of a television program.42 Strategic M & As vertically align production to broadcast of previously aired and �nobody knows� programs creating the interrelationships between related businesses that can reduce costs and risk.43

A vertically integrated company owns all parts of the production to distribution process. An example might be if a company owns the motion picture studio that produces films and owns movie theaters that display those films. Vertical integration differs from horizontal integration in that the same company might also own a cable movie channel that will air the film, a video manufacturer that will create rental tapes/DVDs, video stores to rent the movie from, a music production company that produces a soundtrack, radio stations that play the soundtrack, and finally television stations that will air the commercials. Vertical integration refers to cooperation or synergy within a company to produce and distribute one form of media, and horizontal integration refers to cooperation or synergy across different forms of media.

M & As incorporate shifts in technology and propel capitalist markets forward, allow companies to profit from previous hits and help to reduce the risk in nobody knows properties. Similarly, contracts align companies and individuals in the quest for successful programs from which they will profit. Contract theory, �addresses why self-interested parties structure their deals as they do.�44 It is the specifics of contracts that are a feature of the study of industrial organization ns and its question of, �why transactions occur within continuing firms or between independent parties and why those firms are few or many, operating in one market or several.�45

Contracts are like terminal M & As operating within a larger structure of ownership. In this case, contracts create cooperation between the parent companies and within the parent companies� holdings that are in the business of production and broadcast of primetime programs. There are two types of deals, or contracts, influenced by ownership structure.46 Externalized and internalized deals describe two types of agreements companies can make in their pursuit for successful content and profits. First, external deals may require complicated contracts and increased financial risk. When one company solicits programs from another company�s program library they must negotiate a �rent�. Further, if production studios from multiple parent companies collaborate profits must be shared. Alternatively, internalized deals describe parent companies� ability to use programs from within their holdings. Program libraries owned by subsidiaries can be broadcast on another subsidiary�s network. Further, when parent companies own production studios new programs can be produced in-house without the added costs and risks of hiring independent studios by contract. Contracting internally is easier and more cost effective. Theoretically, as companies utilize cost-effective internal deals more often, competition between fewer companies for audiences may actually require creativity. Content that previously may not have been aired may be broadcast in the hopes of finding a new audience.47 However, it is also possible some content from independent producers may not ever be seen.

Effects of Structural Change

Policy has worked to change the structure in which economic transactions take place. The rescinding of the Fin-Syn rules allowed companies that owned networks to also own content. The Telecommunications Act allowed for broadcasters to own more networks and for other telecommunications industries to invest in networks where they once could not. Combined, the deregulation and regulation have helped to keep the market competitive with the advent of new technologies.

Now, years after the Fin-Syn rules were rescinded and the Telecommunications Act of 1996 was enacted, we have watched M & As between mass media giants that have already swallowed smaller independent competitors.48 While diversity in content can be difficult to measure, diversity in ownership is not. Although there continues to be many different production companies these companies tend to, first, be owned by one of a decreasing number of parent companies49, second, have their programs distributed on one of the parent companies� network(s), and/or third, jointly produce programs or produce programs in-house.50 Indeed, due to the shrinking number of parent companies, the parent companies production holdings �swap� programming appealing to one another�s network audiences. Contracts work to align parent companies in a market structure that has, overtime, come to require some cooperation between a dwindling number of major players. Thus, the parent companies are both producing and broadcasting their own programs (vertical integration) and producing and/or broadcasting programs for one another (cooperative integration51).

Disney�s touted �synergy�, which refers to vertical integration within a company, is similar to cooperative integration. �Synergy�52 implies the ability of a single parent company to produce and broadcast programs, movies, etc., and then market associated products both vertically and horizontally. Cooperative integration is the process or business of a small number of parent companies relying on one another as well as themselves to produce and broadcast programs, movies etc.

Bielby and Bielby53 have done studies about the specific effects of the changes in the television industry. They found that consolidation within the primetime television industry has affected which ideas are made into pilots for primetime slots. While the networks are as likely to schedule externally produced pilots in primetime slots as their own, they are actually looking at fewer and fewer externally produced pilots. When they do see an outside pilot has high potential for audience shares, they use their market power to extract an ownership stake after the pilot has been produced and/or produce the program in-house. Thus, many programs are produced in-house or are co-ventures that could not have happened prior to 1993. Networks are playing favorites in deciding which ideas are made into pilots.

What happens when primetime programs without the Fin-Syn rules are sold into syndication? One example of what can happen when programming goes into syndication is the �other� production company (usually the independent, or smaller production company) can lose out on revenues that would have been theirs under the Fin-Syn rules. For instance, if a network sells a co-produced program to its parent company�s network for a moderate price without other networks being allowed to compete, the smaller production company can lose potential profits. Smaller companies who have an ownership stake in a program can lose out on additional revenues that bidding in a competitive market place once garnered.

With so much money at stake media corporations spend millions of dollars to lobby for beneficial legislation to pass. Lewis54 claims that since the 1996 Telecommunications Act was signed that not only are there more registered, media-related lobbyists, but the amount of money spent on those lobbyists has increased by 26.4 percent to over $31,000,000. In addition to media related lobbyists, media companies donate millions to political campaigns and pay for members of Congress� travel expenses. Specifically, when Viacom (National Amusements Inc.) bought out CBS it was over the cap of only reaching 35 percent of the national audience through the stations it directly owned.

Within a week of the merger�s announcement, Senate Commerce Chairman McCain introduced a bill that would help Viacom skirt that requirement, raising the audience cap to 50 percent of the national audience�Viacom was McCain�s fourth most generous �career patron.�55

Advantageous changes in regulation of ownership and financing those who pass the regulations is the linchpin to the structural shifts then and for the future.

Overall, two primary events in the last 14 years have changed the structure of the television production and broadcast industries: rescinding the Fin-Syn rules and enacting the Telecommunications Act of 1996, while hotly debated, have allowed parent companies to merge with both smaller and larger companies; creating vertical and cooperative integration of resources. The integrative effects also influence which ideas are made into pilots and can decrease revenues smaller companies and individuals may have garnered in competitive bidding.

Democracy and Capitalism

In Noam and Freeman�s56 quantitative analysis of whether or not there is a media monopoly, they broadly define media to encompass entertainment and telecommunications media resulting in large entertainment companies appearing to own a very small portion of the entire industry. Situating entertainment companies within a broadly defined media doesn�t capture the nuances that can be found in a market-level analysis. Ramstad57 recognizes that,

Due to the inherent differences in media products content type and distribution mode, Gomery58 (1993) recommended that one analyze the different industries separately, and then combine the analyses to create a picture of the market. Analyses that aim to be thorough must be divided into industries to enable one to see the fine details in the market. On the other hand, there is little reason to divide the analysis into industries when they do not aim to provide a high level of nuance (e.g. pure ownership analysis).

While Noam and Freeman59 are thorough in their inclusion of each industry in the media market, the level of nuance suffers. Offering only the percentage of the entire market and the percentage of the industry a company owns (e.g. pure ownership analysis) does not give much information about the industry or the market.

For example, when Noam and Freeman60 analyze companies like Disney, they evaluate in terms of the percentage share of the entire information industry. Disney, evaluated as owning only 2% of the entire information industry, might be better and more accurately evaluated in terms of the percentage of the entertainment industry in which it owns or has subsidiaries. Thus, Disney and other companies could be evaluated in terms of the industries each is a part. Although the authors do put Disney and other companies into official industry classifications and sub-classifications61 they do not attempt to quantify the network of interindustry relationships from production to distribution for any of the forms of the media. There is no market level analysis that is dynamic; rather there is only a static analysis of ownership.

In part, the absence of a dynamic analysis can be attributed to Noam and Freeman�s62 use of a government index that doesn�t require interrelationships within a larger context. To evaluate concentration in ownership in what they call the �classic�63 media sector the Justice Department�s Herfindahl-Hirschman Index (HHI Index) is used. According to the U.S. Department of Justice the HHI index is an accepted measure of market concentration found by squaring and then summing the market shares of each firm competing in the market. Markets with an HHI between 1000 and 1800 are considered to be moderately concentrated, and anything in excess of 1800 points is considered to be concentrated (Department of Justice Website).

Using this index to evaluate the unspecified top four firms Noam and Freeman64admit that, �If one looks at classic media industries alone (excluding telecommunications, computers, software, and equipment) [the top 4 firms] did increase in concentration� but remained unconcentrated by justice department standards.� They attribute the increased concentration to cable television systems and home video. While they claim the increased concentration is due to cable systems, they also claim the top four firms� shares in pay cable has had declining concentration from 1986 to 1995. Assuming pay cable and cable systems are the same thing, how is this possible? An answer may be that companies previously not investing in cable systems might have increased their share of the cable systems market. As �new� companies entered this industry the original top four firms� shares might appear to be declining.65

In contrast to Noam and Freeman, McChesney, and others (for instance Bagdikian)66 assume domestic entertainment and news media concentration as a given. In their view concentration is ongoing and increasing. McChesney67addresses the continuing consolidation of entertainment media in his 1999 publication in The Nation by focusing on the attempts of media companies to consolidate international entertainment media. In addition to domestic corporate concentration, he observes there are two tiers of global media corporations. The first, made up of companies like General Electric, AT&T/Liberty Media, Disney, AOL/Time Warner, Sony, News Corp, Viacom, and Seagram, is understood to be competing for domination of the global media markets and accompanying revenues. The second tier includes lesser corporations from North America, Europe, Japan and Australia.

Authors like McChesney68 have focused on the global expansion of a small number of media giants across all sub-classifications of the entertainment industries. Others, like Noam and Freeman69, focus on and show that media concentration has for the most part, overtime, decreased. While these articles address essentially different entertainment markets (domestic versus international), McChesney�s stance assumes Noam and Freeman�s claims about the domestic market are false. Even though both focus on a broadly defined media of multiple industries and or/ genres of entertainment, their fundamental assumptions are polarized. Further, McChesney assumes the small number of corporations that control domestic markets are expanding to consolidate international markets. The two articles present contrasting viewpoints of the extent and direction of media concentration.

Bagdikian70 understands media ownership is more and more monopoly-like with a few companies controlling a large portion of the information disseminated. When he discusses media ownership he predicts that, �To give citizens a choice in ideas and information is to give them a choice in politics: if a nation has narrowly controlled information, it will soon have narrowly controlled politics.� Underlying Bagdikian�s analysis and predictions is the idea that narrowly controlled information by a small, powerful group of companies will limit the ability of opposing views to be aired and offered as alternatives to the general public. Concentrated ownership of information, in effect, stifles the activity of free speech.

A free market may not be conducive to democratic ideals like free speech,

�because it is the free market system that has produced local newspaper monopolies. Entmen71 (1989) is skeptical about the claim that we should look to competition for a solution. He makes an important distinction between the economic market and the marketplace of ideas, arguing that ensuring diversity in the latter should be the principal focus for those concerned about democracy.72

In fact, for those �concerned with democracy� and ideas from many sources, it sounds as if capitalism and democratic ideals are incompatible. Perhaps as capitalism requires deregulation for established media companies to diversify their businesses and continue to grow, our socio-political ideals must necessarily suffer.

Michael Parenti�s book, Inventing Reality: The Politics of News Media73, supports the idea that concentrated ownership of the news media across industries (print and television/cable) ensures only certain views are allowed to be disseminated. Parenti states:

To think that information and viewpoints circulate in �a free market of ideas� is to conjure up a misleading metaphor. A �market� suggests a place of plenitude, with the consumer moving from stall to stall as at any bazaar, sampling and picking from an array of wares. But the existing media market of ideas is more like the larger economic market of which it is a part: oligopolistic and accessible mostly to those who possess vast amounts of capital, or who hold views that are pleasing to the possessors of capital.

To Parenti not only does the concentration of media inhibit views displeasing to owners of the media outlets, but the media also deliberately and systematically misrepresent labor, foreign powers, and what has come to be known as a �liberal� bias.

Like McChesney, Bagdikian�s claim of a media monopoly and Parenti�s74assertions about the functioning of consolidated media power are much different than the conclusion made by Noam and Freeman that media, overall, has actually become less concentrated. McChesney, Bagdikian, and Parenti75 assume there is an oligopoly or monopoly of the media industry as a whole, with the latter two claiming democratic ideals necessarily suffer. Further, Bagdikian�s claim of a media monopoly and Parenti�s definition of media both seem to be as narrowly based in the news media, as Noam and Freeman are overbroad by including too many industries in their definition. None of these authors has tested the full gamut of the media industry, nor have some of them focused on the dynamic relations constituting a market in order to make the generalizing claim of a media monopoly or a media monopoly myth.

The Study

This is a study of the structure of a system in terms of dis/similarity of ties among parent companies involved in production or distribution of primetime programs. I have chosen parent companies that own productions as one indicator of the overall structure, rather than the programming itself, an indicator of content. In addition, I have also chosen parent companies of the broadcast networks as indicators of structure. The number of programs each company produces and/or broadcasts is used to quantify the relationships among the companies. Each company is tied to another company if it produces or owns a program the other broadcasts. Companies can and are tied to themselves. The outcome is a network of ties the companies share. Overtime, the network will include companies that either become more alike (similar) or less alike (dissimilar) in their relationships with one another.

To understand structure, data from 1989, 1994, and 199976 have been analyzed using UCINET 5.0�s77 hierarchical cluster analysis, structural equivalence profile, multidimensional scaling (MDS) and indegree/outdegree procedures. The rational for using these procedures considers the pattern of ties representing program production to broadcast among parent companies in order to develop a visual representation of clusters forming overtime. Using a distance measure based on dissimilarity of ties will display which nodes, or parent companies, are more or less alike in their production to broadcast patterns.

Companies that were both broadcasters and program owners are listed once to account for self loops. For instance if Companies A, B, C, D, etc�are described as nodes 1, 2, 3, 4, etc�respectively, whether they are broadcasting at first and then later owning programs, like Disney, they remain the same node to account for self loops and to accurately track the company overtime. For a complete list of which nodes represent which parent companies please contact me.

Mapping the ownership of a program using the path it takes from production to broadcast ties the companies together into a larger system. Instead of bifurcating production and broadcast into two separate industries, mapping the entire process accurately represents two industries actually functioning together. Studying the paths programs take according to ownership offers a system rather than only the units of that system.

To analyze concentration overtime the nodes are defined as the parent companies that own primetime programs or studios that produce them, and/or broadcast networks in the years 1989, 1994, and 1999. A tie connects parent production company (ies) (i) to parent broadcast company (ies) (j) through the primetime program (p1, p2, etc.�). The weighting (value) of the tie is the number of programs linking two nodes. The choice of years allows an analysis before and after the Fin-Syn rules were rescinded in 1993 and the Telecommunications Act was enacted in 1996.

Data Collection

Information regarding networks that distribute programming is found in The Complete Directory to Primetime Network and Cable TV Shows 1946 � Present.78 Each primetime program in this study was aired in 1989 (272 programs), 1994 (362 programs), or 1999 (291 programs) and has been broadcast on one or more networks over time. The above data source offers the specific network for each program aired (ABC, NBC, Fox, A&E, etc.), except it does not specify syndication channels that broadcast syndicated programs. For instance, The Cosby Show might be listed as aired on NBC but once this program moved into syndication only the word �syndication� is found when referring to where the program was aired. Thus, in this study syndication is its own �parent company� and node.

A brief explanation of the syndicated market and associated methodological issues will clarify further my choice to have a single node represent so many programs. Syndication is a broadcast outlet that is comprised of local markets. For example if program A is sold into syndication it is offered to areas A, B, C, D, etc� separately and each of the areas may or may not buy the rights to broadcast the syndicated program. Thus, some areas may have aired the programs Brooks and Marsh79 listed as syndicated, but other areas may not have aired the programs at all. I had considered looking up which areas did broadcast the syndicated programs, but each local markets can be owned by multiple companies and are usually affiliated with other companies that do not necessarily own the local broadcast outlets. This posed two problems�one, obtaining a data source that would list all, or even some, local markets where each syndicated program was broadcast, and two, deciding how to treat affiliates. After looking for data sources, such as TV Guide, to locate particular programs in syndication I discovered that it would be too expensive to order issues from different local areas from past years. In essence, syndicated programs appear to be a project in and of themselves. Instead, I opted to include syndication as its own node because I wanted to control for its existence; I thought that excluding syndication would skew the resulting network more than if I included it. Therefore, syndication only has indegrees as it is not a programmer.

Information concerning which companies or entities produced the 1989, 1994, or 1999 programs is found from an online industry database offered through BibNet.80 Broadcasters in the entertainment industry utilize the database so they can immediately find who owns distribution rights to programs they may want to purchase or broadcast.

Parent companies of both program producers and broadcasters have been derived using both The Million Dollar Directory81 and the Directory of Corporate Affiliations.82 To be included in the directories the thresholds are as follows: The Million Dollar Directory, 180+ employees if the headquarters is in a single local, 900+ employees if a branch or $9+ million in sales volume, and, for the Directory of Corporate Affiliates, if a domestic company has $10+ million in sales volume or 300+ employees, or, for non-U.S. companies $50+ million in sales volume. Only the Directory of Corporate Affiliates offers non-U.S. companies in their annual reference volumes.83

Whenever possible84 for the years 1989, 1994, and 1999 I identified parent companies by using the Master Indexes (volumes I and II) of the Directory of Corporate Affiliates. I used both SIC (The Standard Industrial Classification) codes and the alphabetical listing of every company included in the public, private and international volumes. Any production companies not listed in The Directory of Corporate Affiliates were then cross-referenced in The Million Dollar Directory.85Any company or individual still not included was classified as its own parent company. For example, if production company A was not listed in the first directory then I looked for it in the second. If neither directory listed production company A as a parent company or as a part of a parent company, then I categorized production company A as a parent company.

Often production companies are a conglomeration of last names I have left as one entity. If people�s names were attached by �name/company/etc.� to small company names I also left it as one entity. Finally, I used the Directory of Corporate Affiliates volumes for public, private, and international companies to look up all the holdings of parent companies to uncover overlooked subsidiaries or affiliates. The same process was used to identify parent companies of the networks on which the programs were aired.

Programs BibNet classified as broadcast on �syndicated� stations have been coded under �syndication� as a node. This will affect distance measures of structural equivalence because ties that may have gone to larger parent companies, increasing weighted ties and changing parent companies� positions in the MDS, will be absorbed and analyzed as �syndicated.�

Individuals or groups of individuals that produce programs are aggregated into a single node called �individuals�. Although this will affect distance measures of structural equivalence, listing each individual together will show how the broadcast of programs produced by individuals has changed overtime. Listing each individual as their own parent company may render them invisible in the center of the MDS.

Nineteen percent of the programs from 1989, 28 percent from 1994, and 26 percent from 1999 did not have a producer listed, or were not in the BibNet database at all. These programs were omitted from the study.86

Finally, the total number of parent companies broadcasting programs for each year is considerably less than the number of production companies. For example, for the aggregated years there are approximately 25 companies that own networks and a total of over 400 production companies, including those that broadcast as well. This means that because each of the 400+ companies must utilize one of the 25 parent companies that own broadcast networks, many of the 25 parent companies will be represented as dissimilar from production companies that are not broadcasters. The ties between companies producing and broadcasting are representative of each primetime program from 1989, 1994, and 1999.

Hypotheses

To test for dis/similarity, two hypotheses have been formulated. My initial hypothesis is that each year, measured individually, will show that two clusters of companies will emerge that are each more alike in their patterns of production to broadcast. Thus, in 1989 before policy and regulatory changes were in place, there will be a cluster of companies primarily producing programs and there will be a second cluster consisting of companies that broadcast programming. In 1994 and 1999 after policy and regulatory changes were in place there will continue to be two clusters, however, while one cluster continues to be those companies producing programs, the second cluster will be made up of those companies both producing and broadcasting programs.

Next, I hypothesize that, overtime, any clustering in 1989 will become more distinct after the Fin-Syn rules are rescinded and The Telecommunications Act of 1996 is enacted. Where in 1989 there were some companies producing and other companies broadcasting, in 1994 and 1999 companies both producing and broadcasting will be more dissimilar from those only producing. For either hypothesis, any clustering of companies exemplifies that the companies in any cluster are more alike in their business practices.

Method

UCInet converted data from the three years into adjacency matrices. Then UCInet calculated Euclidean distances and created structural equivalence matrices based on comparisons of both the rows and columns. The resulting coordinates from each matrix were then used to produce a multidimensional model of dissimilarity (MDS). Companies closer together (similar) in social space are more structurally equivalent than those farther apart. The two-dimensional solution offered low stress values with a large amount of improvement for fit and is used for each of the three years. Those parent companies differing most in their pattern of production and broadcasting occupy social space farther from any clusters of parent companies having similar patterns. Alternatively, those parent companies most alike will occupy social space closer to other parent companies with similar patterns. Hierarchical cluster analysis was used to partition parent companies into clusters according to the proximity of their ties between parent companies.

Indegree and outdegree were computed and are presented to illustrate the number of programs each company broadcasts (indegree) and produces (outdegree). Reporting the indegree and outdegree elaborates whether or not the patterns of production and broadcasting are changing overtime. For instance, if the nodes appear to be shifting from year to year in the MDS, reporting the indegree and outdegree specifies whether or not the shift was due to changes in the activities of companies producing or broadcasting a larger or smaller number of programs.

Results

Figure 1 represents the MDS for 1989. Nodes representing CBS Corporation (2), ABC/ Capital Cities (20), Syndication (13), GE (7), Individual producers (95), Discovery Communications (5), News American Publishing (28), TNN (21), The Hearst Corporation (15), Viacom (32), Warner Communications (29), MCA (23), Stephen J. Cannell Productions (155) and Gulf and Western (30) all lie distinctly outside of a central cluster that includes all other companies producing and/or distributing in 1989. A hierarchical cluster analysis elaborates the MDS.

The hierarchical cluster analysis87 for 1989 provides an image illustrating clusters developing at 1.01. At 1.01 it appears there is one central cluster represented in the MDS by the central cluster and then there are the remaining companies that occupy distinct positions outside of the center. The central cluster contains hundreds of structurally equivalent companies that occupy almost the exact same space, whereas, the remaining companies are few and not structurally equivalent.

The MDS positions are supported by the cluster analysis. The distinct companies from the hierarchical cluster analysis appear on the MDS in positions lying on what could be considered concentric circles around the center. The outermost ring includes nodes 95, 13, 2, 7, and 20 and represents the most dissimilar nodes from the center. These 1989 parent companies, with the exception of 95 and 13, are the broadcasting parent companies with the following indegrees (i) and outdegrees (o) (i,o)88: (13) syndication (61,0), (95) individuals (0, 33), (2) CBS International (60,4), (7) GE (58,5), and (20) ABC/Capital Cities (60,6)89. The second ring contains (5) Discovery Communications (13,0) and (28) News American Publishing (17,3). The third ring consists of (30) Gulf and Western (6,10), (15) The Hearst Corporation (6,6) and (32) Viacom (8,7), (23) MCA (0,8) and (29) Warner Communications (0,7). As the rings get closer to the center the indegrees are diminishing as the outdegrees are increasing. In the center are the primary production companies used before the innovations in technology and subsequent policy changes.

Moving toward the center the next ring begins. (1) CBN (1,3), (21) TNN (7,2), (26) Turner Broadcasting (3,0), and (155) Stephen J. Cannell Productions (0,6), (341) Lorimar Telepictures (0,4), (354) Coca-Cola Company (0,5), (39) Family Channel (2,0), (190) Yorkshire TV (0,2), (228) CBC Canada (0,1), (342) ABC Australia Production (0,1), (348) Touchstone TV (0,1), (390) National Geographic TV (0,2), and (393) Universal Pay TV (0,1) all appear to be almost a part of the central cluster. The remaining companies are in the one, central cluster90 identified from the cluster analysis.

The emergence of pattern is clear from the MDS, the hierarchical cluster analysis, and the indegrees and outdegrees for companies from each ring and cluster that there is a pattern emerging. Those companies furthest from the center are those that have a high outdegree, indicating that broadcasting companies are dissimilar from the center production companies. As we approach the center, companies broadcast fewer and fewer programs (lower indegree) but begin to show higher outdegrees or production levels. Finally, at the center, a larger number of independents offer production only. The companies with largest in and/or out degree are the thirteen companies (including the aggregated syndication node) responsible for all of the broadcasting.

The first hypothesis is supported. There is a very similar, central cluster surrounded by ten broadcast companies (except for the nodes representing individual producers and syndication) and one production company that occupy distinct, dissimilar, positions in the social space. The broadcast companies are somewhat similar to one another, though not as similar as the companies located in the center.

Figure 2 represents the MDS for 1994 after the Fin-Syn rules were rescinded and parent companies could begin to own productions they broadcast on their channels/networks. The Nodes representing companies/entities outside of the central cluster are as follows: Individual producers (95), CBS Corporation (2), GE (7), ABC/ Capital Cities (20), National Amusements (9), News Corporation (10), Discovery Communications Inc. (5), The Hearst Corporation (15), Syndication (13), Time Warner (16), Disney (6), Sony International (151), Paramount (22), Matshushita (25) and TNN (21) are.

While some companies have disappeared as distinct nodes, others have appeared. News American Publishing disappears as Newscorp appears.91 Viacom disappears as National Amusements increases its control of the company by 199392 and appears. National Amusements as Viacom�s parent company now moves further away from the central cluster as it broadcasts more programs and relies more on its increasing number of productions. Time Warner appears after Warner Communications merges with Time Corporation in 1989. The newly merged Time Warner remains in a similar position in 1994 as Warner Communications did in 1989. MCA disappears93 and Matshushita appears after it acquires MCA in 1991.94 Sony International appears after it buys Columbia pictures in 1989 (www.caslon.com). Gulf and Western is renamed Paramount Communications Inc. in mid 1989 and Paramount appears.95 Stephen J. Cannell productions, the only independently owned production company that was distinctly dissimilar from the center in 1989, disappears as a distinct node and is absorbed into the center. Node 95, for individuals moves closer to the center as those individuals producing programs become more like the production-only companies. Disney appears as it too becomes involved in broadcasting.

Again the MDS positions are supported by the cluster analysis. At 1.7 it appears from the cluster analysis that there is one giant cluster. The cluster analysis helps refine the MDS representation. At level 15.9 the outermost ring includes (2) CBS Corporation (73,10), (7) GE (74,14), and (20) ABC/ Capital Cities (73,26). At level 10.7 the second ring contains (95) Individuals (0,42) and Syndication (72,0), (9) National Amusements (39,14), (5) Discovery Communications (27,5), (10) News Corporation (35,10), and (15) The Hearst Corporation (22,13). At 6.8 a third ring appears with (6) Disney (0,13), (16) Time Warner (1,19), (22) Paramount (7,12), and (151) Sony International (0,15). At 3.0 the ring includes (21) TNN (7,1), (25) Matshushita (7,9), (190) Yorkshire TV (0,3), (214) The Distribution Co. (0,6), and (215) BBC (0,5). Finally at1.5, (1) CBN (2,3), (19) Western International Communications (5,3), (26) Turner Broadcasting (2,1), and (155) Stephen J. Cannell Productions (0,4) are beginning to move into the center.

The rest of the companies, even though they fall outside of the giant cluster in the cluster analysis, are indiscernible in the central cluster. Now, the remaining companies are in two clusters at 1.4 with approximately 24 in the first cluster and all others in the second. All of these companies are indiscernible in the central area of the MDS.

The first hypothesis is again supported with a central cluster of production companies and another cluster of 12 (again the individuals and syndication aren�t counted as companies) production/broadcast companies occupying distinct space from the central cluster. The number of broadcasting companies actually increases, as additional companies that could not previously have broadcasted productions, like Disney, begin to broadcast. This information could support Noam and Freeman in that there are more companies broadcasting;; however, as the number of broadcast companies increases the pattern of ties amongst broadcast companies becomes more alike and in/outdegree increases. Though the number of distinct companies is larger, each is increasingly responsible for more and more of the production and/or broadcast.

The second hypothesis is also supported as the distinct companies surrounding the center move closer together, an indication that they are more similar in their ties to themselves and each other. Overall, even though there are more broadcast companies, the parent companies that own each are becoming more similar in their pattern of production and broadcast. Companies in the center are also becoming more similar in that outdegree is decreasing overall. The clusters are tightening.

Figure 3 represents the MDS for 1999 after the Telecommunications Act of 1996 was enacted. Nodes representing National Amusements (9), GE (7), Disney (6), Time Warner (16), The Hearst Corporation (15), News Corporation (10), Chris Craft Industries (3), CBS Corporation (2), Sony International (151), and Individuals (95) lie distinctly outside the center again indicating a high degree of production and/or distribution. Once again the central most nodes appear more clustered as they produce fewer primetime shows and have more similar choices in where the program is broadcasted.

The cluster analysis for 1999 supports the MDS representation. At 1.5 there is one large, central cluster and then the remaining companies that again appear to lie on concentric rings around the center. The first ring forms at 12.4 and includes (6) Disney (62,31), (7) GE (51,22), and (9) National Amusements (54,36), and the second loosely includes (10) News Corporation (26,12), (15) The Hearst Corporation (24,18), and (16) Time Warner (33,21).96 At level 6.9 (13) Syndication (53,0) and (151) Sony International (0,22), (2) CBS International (33,3) and (3) Chris Craft Industries (4,20) appear.

National Amusements, Disney, and GE are now the most dissimilar companies from the center, having the highest indegree and outdegree. Acquiring ABC/Capital Cities in 199597 increases Disney�s indegree by 1999 and pushed it further from the center production-only companies as Discovery Communications disappears into the center. Chris Craft Industries appears after creating an affiliation with News Corporation in 1998.98 Sony International moves further away from the center as it is produces a number of programs that are broadcast on one of the other distinct company’s channels/networks. Sony is the one production-only company that has enough ties to the other outer companies to remain outside of the central cluster.99 The node for Individual producers, even with joint ventures, begins to be absorbed into the center as the distinct broadcast companies utilize fewer productions from independent producers (the outdegree decreases by 32, a 76 percent decrease from 1994 to 1999).

There is a central core of companies with nine production and/or broadcast companies occupying distinct space around the center, which again supports hypothesis number one. Whereas in 1994 there was a brief moment of increase in number of owners of broadcast networks, by 1999 with mergers and acquisitions there are fewer companies that own more networks developed after the implementation of digital technology and cable systems. Still, those few companies cluster towards one another as they become more alike in the production to broadcast patterns.

The second hypothesis is also supported. The distinct companies from 1994 are now even more similar and structurally equivalent in their pattern of ties. These companies are also more dissimilar from the central cluster. This is because first, these companies have more ties to one another (more similar in/outdegree), and second, have fewer ties from the center. The central cluster becomes even more similar as fewer productions are broadcast from those companies or individuals occupying the center cluster.

Summary

In summary, both of the hypotheses were supported with the hierarchical cluster analysis, MDS and in/outdegree procedures. Overtime, any clustering became more and more pronounced as companies became more and more alike in their business practices. In 1989 there was a large, central cluster of similar companies which were production-only. Surrounding this cluster were 10 broadcast companies and one production company that occupied distinct, dissimilar positions from the center. While each of these 10 companies were somewhat similar and had high indegree, their source companies for production differed; thus, the distance between them is greater than between those companies in the central cluster. The end effect was that the outermost companies occupied distinct space outside of the center. The node for individual producers had a high outdegree and also occupied space well outside of the center, because the outdegree is so much higher than the other production companies.

After the Fin-Syn rules were rescinded in 1994 there was again a central cluster containing production-only companies. The number of companies distinctly different than the center increased to twelve as companies that previously could not combine production and broadcast began to do so. Each of the distinct companies also moved closer together as their pattern of production and broadcast to and from one another became more similar. The node representing individual producers moves closer to the center cluster as the outdegree decreases and individual producers� programs are utilized less frequently.

By 1999 after The Telecommunications Act allowed for mergers and acquisitions that could not have previously happened, the central core of companies appears to be a tight knot of companies. Whereas in 1994 there was a brief moment when the number of distinct companies increased after mergers and acquisitions, now there is a decrease in the number of companies occupying the space around the central cluster. Further the distinct companies from 1994 moved closer together to occupy more similar social space. They became more structurally equivalent than in the previous years. The node for individuals appears almost absorbed into the center as individual producers were utilized less frequently than in previous years.

The pattern that emerged is that overtime a small number of parent companies of program production and broadcast have become more similar in their ties to and from one other. Even when the number of different companies broadcasting increases those companies continue to become more alike in their ties to one another. There is every reason to believe that the pattern that emerged in this geography of the television program production and broadcast industries will continue in the direction of distinct, parent companies having more and more similar ties to one another if regulation and legislation remains as it is. Eventually, the market may begin to stagnate as just a few companies own all that they can. Ostensibly, this may lead to further deregulation and larger and larger parent companies.

Conclusion

The ongoing academic and industry debate has been concerned with whether or not the media industry is monopolized after shifts in policy and regulation. As the literature has shown the opposing sides in the debate support their arguments with studies that can be misleading and/or too narrowly or broadly define what media is. Defining a part of the media industries in terms of the similarity of ties between production and broadcast companies bypasses the need to decide if the industries are monopoly- owned. Overtime, more similar ties between certain companies indicate that shifts in policy and legislation have created interindustry relations that include fewer and fewer companies in the system of production and broadcast. These companies may or may not have concentrated ownership by varying standards and definitions, but they do have very similar patterns of production and broadcast of programs that are becoming more similar overtime.

Using social network analysis as the method to study the structure of the television program production and broadcast industries overtime offers a geography of a system. Describing the terrain of ownership in terms of similarity of ties instead of how monopolized ownership is bypasses arbitrary thresholds that determine if an industry is over-concentrated or not. Instead of understanding the increase in the similarity of ties between distinct companies as concentrated, monopolistic, or oligopolistic, the system is seen as cooperatively integrated.

Cooperative integration is a better way to describe the relationship among the companies in this study. Recalling that cooperative integration refers to the practice of companies across industries relying primarily on one another or themselves for production and/or distribution of product, the companies in this study are cooperatively integrated. Understanding companies as embedded in a system of cooperative relationships or ties also allows for us to go outside of the argument about whether or not an industry is controlled by a monopoly. Instead we can discuss a cooperatively integrated industry as excluding companies that could potentially compete in the market, concentrated or not.

Further studies could map the relations between media companies that own other forms of media. Each of the distinct companies furthest from the center in the MDS also own other forms of media such as music production studios, VHS distribution companies, movie theaters, etc. Studying the entire market landscape, from production to distribution, by including each of the various forms of media may show that the trend in the television industries studied here is a more general pattern. Cooperative integration across the entire media market landscape may be a reality and can be measured and described using the social network analysis methods presented in this article.

Tonya Lindsey is a Candidate for the Ph.D. attending UC Santa Barbara�s Department of Sociology. Her contact information is: Department of Sociology, University California Santa Barbara, Santa Barbara CA, 93106. tdl@umail.ucsb.edu. This paper is based on her thesis and was presented at the Pacific Sociological Association�s annual meeting in Portland, Oregon, 2005.