Article 2

2009 43

Albinger J.. - Economic Model for Monopoly Analysis in Telecommunication: Proposal to Demonstrate Uniqueness

Jürgen Albinger

ECONOMIC MODEL FOR MONOPOLY ANALYSIS IN TELECOMMUNICATION: PROPOSAL TO

DEMONSTRATE UNIQUENESS

Jürgen Albinger (PhD)

Abstract

The Telecommunications Act of 1996 sought to end the monopoly that once existed in the telecommunications industry. Since its adoption, the telecommunications industry has been undergoing a period of rapid change and development. The entry of new players into the market encouraged them to seek new ways to attract and keep customers. These changes have led to a rapid influx of new technology and services. Many times what defines a monopoly is not clear in every circumstance and there are many pending lawsuits for violations of antitrust laws in the courts today. Economic models are useful in resolving issues of whether a monopoly truly exists, or whether claims are unsubstantiated. Previous models were applicable only in certain situations. These models are unreliable in predicting monopolies outside the parameters for which they were designed. This research will develop and test an economic model that accurately predicts the existence of a monopoly in the telecommunications sector, based on the United States of America’s marketplace case, as the best representative example to deal with.

Business Intelligence Journal - January, 2009 Vol.2 No.1

44 Business Intelligence Journal January

Introduction

In some ways, it would appear that a monopoly represents a good deal for consumers. By having all of the resources needed to provide a good or service managed by one organization, there are some enormous advantages to be derived including efficiency and availability. Given the negative connotations that have been assigned to the term in recent years, it becomes necessary to understand why these business forms are not being used more frequently today. Indeed, monopolies have been the subject of large amounts of media attention in recent years. Recently, there has been the question of whether Bill Gates constituted a monopoly, or whether his vast empire was a just and fair reward for good business savvy. There have been many questions regarding mergers and acquisitions of corporate giants would create a monopolistic environment, this creating an unfair advantage and virtually destroying any smaller companies that were unable to compete on their scale. Courts are flooded with cases such as these and surrounding all of these cases is one key question, “What exactly is a monopoly?”

Everyone is familiar with the textbook definition of a monopoly, that is, a marketplace where a lack of competition exists and one company and all or a significant portion of the market place. Notwithstanding the outward appearances of representing a good deal for consumers, in reality, a monopoly is only good for one entity: the monopoly itself. A lack of sufficient competition allows them to set their own places and in some cases, no incentive to produce the best quality product available. If the product or service that the monopoly produces is necessary for life, then the customer has only one choice. They must take the product that is offered, and pay the price that is offered, especially if there is a lack of a sufficient substitute product.

Monopolies tend to drive up prices and have many negative effects on the establishment of a fair marketplace. One of the key issues surrounding the court cases regarding monopolies is that there are many different types of monopolies that exist for many different reasons. Some monopolies exist because of scale. In this case, the monopoly simply out-produces anyone else in the industry. They are price makers and everyone else must follow their lead. Because of their size, they can often produce goods cheaper than their competition and there fore can offer them at lower prices. Eventually, the smaller companies may be forced out of business and a true monopoly will then emerge.

Some monopolies exist due to sheer geographic location. This is particularly common in the telecommunications sector and other utility sectors. In these cases, there may only be one service provider established in an area and the product may be necessary to life. Therefore a monopoly exists due to location. There are other more subtle forms and reasons for a monopoly existing, however, such as the effect that at a provider is the first to offer a service. They establish brand equity and name recognition. This may make entry into the market more difficult for the competition.

Monopolies may be complete, where the monopoly is the only provider and has complete control over the marketplace. The monopoly may be partial, where there is competition, but the monopoly retains a significantly large portion of the marketplace. By all accounts, a monopoly makes it difficult to enter into the marketplace. The more complete, the monopoly, the more difficult, entry into the market will be. However, in an incomplete monopoly, it is possible for the monopoly to drive prices up and a lower priced competitor steal a portion of the marketshare from the monopoly.

2009 45

As can be readily seen, there is not one hard and true definition of what constitutes a monopoly. The term “monopoly” has many shades and subtleties that make a black-and-white definition a difficult task. There have been many mathematical economic models that have attempted to provide a legal definition by which the courts, businesses, and general public could use to classify whether a market is a monopoly or not. If this clear-cut definition existed, the courts would not be bogged with Anti-trust suits; there would be no argument. This is the purpose for the following research. To this end, this research will focus on the telecommunications industry and develop a mathematical economic model that will be useful in determining if a monopoly exists in the telecommunications industry. This model will be tested through empirical research.

Rationale for Study

Recent focus in the telecommunications industry has been on the cell phone sector and other new technologies such as broadband communication and other wireless services. However, the predecessor to this expansion in the telecommunications industry began with the Telecommunications Act of 1996. Prior to this act, one primary company owned a large portion of the communications industry, which at the time consisted of telephones. This company was Bell Telephone, the pioneers of modern communication. In his essay, “TELRIC vs. Universal Service: A Takings Violation?,” Buck (2003) reports that:

The Telecommunications Act of 1996 (“Telecom Act” or “1996 Act”), however, introduced a new wrinkle into the realm of telecommunications pricing. By law, local telephone companies, known as incumbent local exchange

companies (“ILECs”), would be forced to lease virtually all of their equipment and facilities to their competitors, the idea being that the competitors could then offer a competitive product to the consumers. The pricing terms of such leases are to be set by state public utility commissions, who must calculate the theoretical cost of constructing a new network. This system of cost-based pricing is known as TELRIC, which stands for ‘total element long-run incremental cost.’ (p. 2)

As others began to wish to enter the marketplace, they found it difficult or nearly impossible to compete with the giant. Clearly, this scenario and comparable ones that resulted were not conducive to the ideals of free enterprise and true capitalism, so the government decided to act. They passed the Telecommunications Act of 1996, which de-regulated the industry and paved the way for open competition. The act included provisions for helping new companies enter the marketplace including providing extra licenses in many areas and funding sources for new business startups. This created a wealth of opportunity for new businesses. In an attempt to out-compete their competition, companies experimented with new services, pricing schemes and eventually, new technology. This created a boom in new technology and led to the invention of cell phones, pagers, and other wireless products. It seemed that everyone has benefited from this move by enjoying lower prices and a wealth of new products and services. Every one has benefited, that is, except “Ma Bell.”

As in other sectors, the privatization of the telecommunications industry has led to a plethora of business startups. Some of these new enterprises became shining stars and rose to the top of their industries quickly, while others crashed and burned. As in any other industry that experiences this type of boom,


Jürgen Albinger



there were many allegations of mergers and shining stars becoming monopolistic in nature. This led to a backlog of court cases that is likely to be there for some time in the future.

An economic model that tests for the existence of a monopoly in the telecommunications industry would be of great use in determining if a monopoly truly exists, or of the claims are unsubstantiated. This research will examine several existing models for the existence of monopolies. There are many instances in which these existing models reliably predict the existence or nonexistence of a monopoly; however, they were developed for prediction under certain types of markets and market conditions. As a result they are reliable in predicting monopolies in certain sectors of the market, however, prove themselves to be inadequate in predicting monopolies in other sectors. The existing models do work in certain circumstances’ however, their inability to work in other circumstances has resulted in some cases, particularly that of the telecommunications field, in a predicament where no clear decisions can be reached. Furthermore, there are still some who argue that the Telecommunications Act of 1996 was unnecessary because no true monopoly existed in the first place.

For the purposes of this analysis, it is important to keep in mind that every sector of the market has its own set of characteristics. For instance, some are cyclical, such as retail, and some are steady growth, as in the case of many manufacturing sectors. Some sectors have many small players, while others are made of a few larger players. Sometimes one company is the clear leader. Sometimes there is a group who emerges ahead of the pack and sometimes all are on a level field. These varying characteristics make it necessary to develop a separate economic

model for each sector or sub-sector in the market.

The telecommunications sector has experienced a fundamental restructuring in recent years and is still in a period of rapid change. The most drastic change came after the adoption of the Telecommunications act of 1996; however, this is a rapidly changing industry and by all accounts, there are many more changes to come. In this environment, some observers have suggested that even though there are more companies in competition, that there is still a monopoly, or an environment closely resembling a monopoly. This is particularly true in the case of companies such as Lucent Technologies and AT & T. Mergers Such as Viacom and WorldCom draw even more suspicion as they form conglomerates large enough to force smaller companies out of business, by their shear scale.

This is the rationale behind the need for the proposed research. It is the objective of this research to develop an economic model that accurately predicts the existence of a monopoly in the telecommunications industry. The telecommunications industry consists of many sub-sectors and the model will be tested in each of the sub-sectors to assure that it can accurately predict conditions in a variety of conditions within the telecommunications sector.

The proposed model will be tested both before the Telecommunications Act of 1995 and after its inception. It will be tested in the rapidly changing environment that exists today. It is the goal of this research to develop a real-world working model that will be helpful in solving the many issues that exist today regarding monopolies and the telecommunications industry.

2009 47

Scope of Problem

The telecommunications sector has been a leader in the technology sector since the influx of new companies and new technologies that resulted in the invention and proliferation of technology after the adoption of the Telecommunications Act of 1996. While different components of the industry, the regulators, and the economists expressed different concepts concerning what changes should be made in the telecommunications industry, virtually all stakeholders agree that competition in all segments of telecommunications should be encouraged (Mcmaster, 2002). Anti-trust lawsuits have caused many disruptions in the development of the sector and as a result caused massive overheads for companies that could barely afford it. Many of these lawsuits have centered on the existence of local or large monopolies. These lawsuits lowered the profitability of many companies and forced many of them to go bankrupt pr make drastic cuts in their workforce. The costs of these lawsuits have been devastating and have hurt the profitability of many companies. As a result, the sector, as a whole, has suffered a downturn that will likely take some time to recover.

Statement of Hypothesis and Research Questions

The hypothesis of this research is that a mathematical economic model can accurately predict the existence of a monopoly in the telecommunications industry as follows:

Hypothesis 1 (H1): As market concentration shifts from low to high, the effect on the top four players out of ten in an industry shifts becomes more pronounced than for the remaining top six in a manner discernible through one of the economic models to be analyzed.

The successful model will be useful in predicting the existence of a monopoly in the telecommunications industry, and will thereby become a useful tool in resolving the issue of whether a monopoly exists or not. The guiding research questions of this research will demonstrate that the economic model developed as a part of this research will accurately predict the existence or nonexistence of a monopoly in the telecommunications sector. The primary research question will be, “Can the model developed for this model accurately predict the existence or nonexistence of a monopoly in the telecommunications sector?”

Literature Review

Given the importance of the industry, it is not surprising that much literature exists on the telecommunications industry and the history of the Telecommunications Act of 1996. There is a wealth of scholarly debates regarding whether it has been a help or whether it has been of harm to the industry. It can be debated as to which side holds the most truth at the present time. The following will review the current literature available, both on the telecommunications industry and the models that have been used to describe other monopoly models. The review will be critical in nature and will attempt to summarize as much of the information as possible on the topics at hand.

The Telecommunications Industry

The telecommunications industry is important and considered a vital part of our everyday lives. The telecommunications industry represents only a small portion of the country’s Gross Domestic Product, only 1-2 percent (Stigiltz, 1998). While this amount may seem insignificant, the services that it provides are vital to every other sector


Jürgen Albinger



in the economy. Telecommunications is the backbone of many other sectors.

The Telecommunications Act of 1996 is one of the most highly debated topics in economics. There are some that say that it has been ineffective and that we now have a monopoly again, as a result of mergers and acquisitions. Despite its constraints, the Telecommunications Act of 1996 was an absolute necessity according to Henricks (1999). This author points out that, “Despite the amendments and revisions that have been added throughout the years, Congress saw no choice but to completely replace the obsolete Communications Act of 1934. Because communication technology has grown by leaps and bounds over the last few years, the Communications Act of 1934 was considered an antiquated document. Edmund L. Andrews states that advances “in technology and rapid change in the marketplace have made [the 1934 act] increasingly outdated and, many experts believe, [the 1934 act] has been harmful to competition and consumers’” (p. 39). As a result, Congress replaced the 1934 act with a new, more contemporary law that was intended to regulate the electronic media of the 21st century (Hendricks, 1999). There are others who say that it has had the intended result, but that the movement towards a competitive marketplace does not happen overnight. Poulson (1997) believes that achieving a fair market in Colorado will not be immediate and will take some time. There are others who believe that it is working in some cases and not working in others. Alaska is moving towards a more competitive marketplace on a local level. Rural communities often have a localized monopoly as there are not enough customers to attract competition (APUC, 1997).

Many have made predictions, similar to the one that will be made by this research project. Many of these feel that the

telecommunications market in 2010 will be a pure competition, with no single company emerging as the dominant force. Pehal (2000) believes that the market in 2010 will not be dominated by one type of technology and that the result will be a mixture of the new and the familiar. Powers and associates (2000) predict a similar scenario. Others see the market moving towards a monopolistic setting in the future (Farrell, 1995, Goldstein, 1993, Kanell, 1998, Kitmari, 1998, and others). They are basing these claims on the fact that some companies are now beginning to emerge as monopolies at the current time. Sun is currently being accused of attempting to gain a monopoly of scale (Malik, 1999).

According to Black’s Law Dictionary (1990), a monopoly is “A privilege or peculiar advantage vested in one or more persons or companies, consisting in the exclusive right (or power) or carry on a particular business or trade, manufacture a particular article or control the sale of the whole supply of a particular commodity” (p. 1007). Not surprisingly, monopolies can have many negative effects on the marketplace. Monopolies prevent competition and raise barriers of entry into a market. Senator Leahy points out that the cable television companies are a prime example of this (Kenyon, et. al., 1995). A free and open competition ensures that customers will have many benefits, such as lower prices, a variety of new and developing products and services.

A competitive market forces companies to strive to maintain a customer base, whereas a monopoly promotes routine and is resistant to change. In a free competition, businesses must strive to provide customers with new and innovative products at reasonable prices or face the loss of customers and declining profits (FCC, 1995). It is the customer and society as a whole that benefits from free competitions.

2009 49

A natural monopoly exists when the people in an area need a service or product to sustain their daily lives and there is no product substitution (Stoffels, 1999). It also exists where there is no other close competition, due to the fat that there is a lack of resources for expansion into the area. A prime example of this is a rural electric cooperative. Usually the one who ran the lines first can have a monopoly. The people will use whoever is there and have no other choice. It is unlikely that another company will run lines to the same locations due to the cost involved and the uncertainty that customers will switch. In this regard, Gasmi, et. al., (1999) developed a test for a natural monopoly that he feels would be applicable to the telecommunications industry. It was developed to determine natural monopolies in the rural electricity industry. Upon examination of the test, it was felt that it would be useful in certain sectors of the telecommunications industry, especially where rural hardwire telephones were involved.

However, it did not appear that it would be useful in other segments of the market, such as cell phones and other wireless products, or in urban areas. In their chapter, “A Technico-Economic Methodology for the Analysis of Local Telephone Markets,” Gasmi, Laffont and Sharkey (1998) provide an empirical methodology for analyzing the regulation of local telephone markets that combines an engineering process model of costs with models from the new regulatory economics. According to Gasmi and his colleagues (1998), this technico-economic methodology is comprised of two separate analyses.

The first step involves studying the properties of optimal regulation under asymmetric information in detail by examining three issues:

The extent of natural monopoly when 1. informational rents associated with regulation are taken into account;

The extent of the divergence of pricing 2. under the optimal regulatory mechanism from optimal pricing under complete information (incentive correction); and,

The implementation of optimal 3. regulation through a menu of linear contracts (Gasmi et al., 1998).

The authors determined that for fixed territory, strong economies of scale allow local exchange telecommunications to retain monopoly characteristics even when the (informational) costs of regulation are properly accounted for. In addition, they report that the incentive correction term is small in magnitude, and optimal regulation can be well approximated through relatively simple linear contracts (Gasmi et al., 1998).

The second phase of the technico-economic methodology involves an evaluation of the relative performance of various regulatory mechanisms, from both traditional and modern (incentive) points of view. This analysis allows us to quantitatively assess the social value of regulatory transfers and of good cost auditing procedures, the redistributive consequences of the various forms of regulation, and the sensitivity of the relative performance of the various methods of regulation to the cost of public funds.

The importance of costing methods in regulation of network industries is by now well established. This is particularly the case in telecommunications. Historically, various approaches have been used to evaluate costs in telecommunications, including accounting and econometric methods. However, with the rapidly changing technology, forward-looking engineering models of costs have increasingly proven useful. Meanwhile,


Jürgen Albinger



new forms of regulation have made their way into telecommunications, triggered, on the one hand, by widespread dissatisfaction with the performance of traditional forms of regulation, and on the other hand by the broad political trust in private incentives for enhancing the performance of operators. The technico-economic methodology described by Gasmi and his associates uses the engineering approach in a modern regulatory economics framework for analyzing some important issues facing regulation of telecommunications today.

The usefulness of this technico-economic methodology is demonstrated through two investigations:

This approach was used to empirically 1. investigate in detail the properties of the optimal regulation of the local telephone service and its implementation; specifically, Gasmi and his colleagues researched ways to generalize the natural monopoly test in order to take account of the (informational) costs of regulation.

The technico-economic methodology 2. was used to compare the performance of various regulatory schemes from both traditional and modern incentive regulation.

In order to successfully assess the various forms of regulation that impact on the technico-economic methodology, a detailed specification, in the context of a representative local telecommunications network, of the firm’s observable cost function, the market demand function, the disutility of effort function, and the distribution of the technological uncertainty parameter (Gasmi et al., 1998).

The regulatory schemes involved in this analysis are subject to constraints of three types:

Mechanisms differ in the (ex post) 1. observability of costs;

In the feasibility of lump sum transfers 2. to the firm, and,

In the degree of bounded rationality of 3. the regulator (in addressing incentive issues) (Gasmi et al., 1998).

A duopoly exists where there are two companies that compete in a marketplace. In this case, there is limited competition, but no clear competitors exist, other than the two duopolists there are some that many consider this situation to be a small oligopoly. In this case, the two competitors must match each other in price and quality. One cannot afford to fail to match the other’s price. This market can be good for the monopolists, if cooperation exists between them. However, if they are fierce competitors, the competition will serve to harm both, as they strive to lower prices and out-perform one another; however, Pociask and Rutner (2000) believe that a separate test must be applied when analyzing a monopoly versus a duopoly.

In a similar vein, Huisman (2000) explores the specific benefits and effects that a monopoly will have on the marketplace and draws the conclusion that the long-distance market is currently an extremely concentrated market. The proposed MCI WorldCom/Sprint merger will make it even more concentrated. There are those who feel that the MCI WorldCom/Sprint merger will make it even harder to compete for smaller companies.

Furthermore, Huisman argues that entry into the marketplace is currently high. The established companies have a considerable amount of brand equity attached to their names. It would take a massive and very expensive media campaign to compete with MCI WorldCom and Sprint. There is

2009 51

a certain amount of trust associated with a name. People will not be likely to switch to a small, barely known carrier, even if their prices are substantially lower. There are also considerable regulatory and capital expense barriers that make it nearly impossible to enter into the market. These factors tend to weaken the opinions of those who claim that MCI WorldCom and Sprint do not resemble a monopoly, or a strong duopoly. These carriers are able to meet future demands of a growing client base, therefore it seem unlikely that the current situation will change (Huisman, 2000).

To this end, Huisman argues that the MCI WorldCom/Sprint merger should not be allowed, as this will create a true monopoly in the telecommunications industry. As long as the two remain separate, there is at least a limited amount of competition. He feels that a limited amount of competition is better than allowing this market to become a pure monopoly. These two companies have cornered the market on an economy of scale and it would not be possible to retain efficiency in any other way. This is the case in the long-distance area of the market; however, Huisman feels that the merger would be beneficial in other markets such as in the wireless and local exchange services. The merger would provide better interconnectivity but there remains a concern that this would create a smaller competitive market in these areas as well (Huisman, 2000).

Clearly, Huisman is against a further concentration in the telecommunications marketplace and argues that re-creates the same situation that the Telecommunications Act of 1996 sought to prevent and suggests that a strong campaign be launched to convince the government to disallow the merger of MCI WorldCom and Sprint; nevertheless, this author maintains that this is not enough and calls for a divestiture of

Sprint’s equipment and backbone to allow others to enter into the marketplace (Huisman, 2000). In addition, this author suggests that the Telecommunications Act of 2996 was necessary and that the intentions were good; yet, since its adoption, the market has been allowed to develop with no guidance or intervention and that this has created a market similar to that, which existed prior to the Telecommunications Act of 1996. The merger will effectively eliminate the two companies who currently have a chance to pose competition to the impending “supergiant,” Bell Atlantic and GTE. Due to a lack of maintenance, it would seem that we have taken a step backward and that it is “Ma Bell” all over again (Huisman, 2000).

Economic Models of a Monopoly

Even a casual review of its circumstances today makes it quickly apparent that the telecommunications industry is a complex entity and there are multiple sub-industries within the primary industry. The telecommunications industry has gone from a relatively pure monopoly to an attempted competition, and now it is questionable as to whether it is gravitating towards a monopoly again. In addition, there are now more products and services available. The market is no longer comprised of one market. There is a long-distance market, a local service market, and a cell phone and wireless market. All of these markets have different characteristics and the previously existing models fail to useful in all areas of the telecommunications industry. For instance, the Gasmi Model (Gasmi, et. al., 1999) is useful in the rural local phone service industry in rural areas. However, is very ineffective in prediction in an urban market, where there are many competitors, such as the cell phone market. In this market, there may still be a monopoly but the monopoly


Jürgen Albinger



does not have as much power as in a natural monopoly setting.

Studies have been conducted on monopolistic markets in other areas. It has been found that monopolies retain certain characteristics and that they are subject to many factors that influence their behavior. Consumers have a set of guidelines, by which they make purchases. Even in a pure monopoly, the company must provide a certain level of perceived quality and cannot engage in overly high pricing schemes. It would seem as if the consumer would have no recourse, especially if they need the product or service. However, disgruntled customers will complain to their government and could pressure the government to break up their monopoly.

Many studies by Shimonura and Shimonura and associates (1197-1198) demonstrated that market equilibrium could be manipulated by placing a set of export constraints on the market (Kemp, M., Okawa, M., Shimonura, K., 1996). Other researchers such as Graham, et. al. (1983) studied the airline industry after deregulation. They found that fares are independent of market concentration. Likewise, Rogerson (1982) found that monopolies seek resources directed at obtaining profit and that there are social costs to the monopoly because of this rent-seeking activity.

According to Miller (1996), “A major rationale of the assault on regulation that has taken place in the work of orthodox economists over the past several decades, and in the courts and legislatures since the late 1970s, is the view that regulation retards technological advance and that an ‘optimal’ rate of investment in new technology will be a happy by-product of deregulation” (p. 719). A monopoly firm’s pricing strategy in a market in which consumers have varying perceptions of the quality of the firm’s product. The effect is reflected in the

price that the firm will be able to secure for the product (Kehoe, 1996). For example, Caminal (1996) found that a monopoly can choose between no advertising and large-scale advertising. The monopoly will adopt the most efficient advertising strategy in conjunction with lower fixed prices for their services. In a duopoly, when consumers have a fixed time-horizon, such as signing a one-year contract, the two firms may alternately dominate the customer base, alternately charging high and low prices (To, 1996).

Using a maximum entropy technique, it is possible to approximate the market shares of each firm in an industry using the available government summary statistics such as the four-firm concentration ratio (C4) and the Herfindahl-Hirschmann Index (HHI). The HHI technique is reported to be effective in estimating the distribution of market shares in 20 different industries. Golan et al. (1996) support practice of using HHI rather than C4 as the key explanatory variable in many market power studies (Golan et. al. 1996). According to Jorde and Teece (1992), with regard to the issue of the appropriate Herfindahl-Hirschmann Index thresholds, the Merger Guidelines identifies critical HHIs at 1000 and 1800; however, it is difficult to hypothesize and propose alternative HHIs for technologically dynamic markets, but the inclusion of performance competition and the extension of the time frame of competitive response may mean that it is not necessary to change these critical HHIs (Jorde & Teece, 1992).

In addition, Jorde and Teece suggest that with technologically dynamic markets, the dynamics of market structure in the past should provide some guidance to assessing market definition and predicting likely changes in market concentration. “Key factors are the change in concentration and the trend in the number of competitors,” they advise. “Failure to recognize that competition

2009 53

is often on the basis of performance attributes and not price will lead courts and the antitrust agencies to underestimate the breadth of product markets in industries characterized by rapid technological change. This process, in turn, will lead courts and agencies to exaggerate antitrust dangers. As a consequence, technological development may be retarded” (Jorde & Teece, 1992, p. 11).

Golan et al. (1996) presented their HHI technique as a one-size-fits all factor that would be useful and applicable in a variety of situation, citing its usefulness in 20 industries. Upon examination of the technique used, however, it was found that many variables were not taken into account, such as changes in industry structure and the existence of markets within markets, Golan and associates tout their method as being useful in almost any situation. However, it was not found to have the wide applicability that they claim. Before applying their method, one should be careful to examine the circumstances to which it was originally applied and to be certain that the situation being tested has similar characteristics to those used by Golan and associates.

Golan and associates did apply their method to 20 industries; however, they did not state the conditions in the industry. Generalizations such as these can be dangerous and should be carefully examined when choosing to use the method for one’s own analysis. This is not to say that the method is not useful, it certainly has some merit. It is the widespread applicability that is in question in this case, not the method itself. In this regard, Young (2000) notes that the predominant assumption in media economics to date has been that market structure is a central feature of media markets:

The problem that has emerged is that once market outcomes (prices and

outputs) under conditions of oligopoly are sensitive to the precise specification of firms’ conjectures with regard to other incumbents and to potential rivals, there is a need to theorize in terms of firm specific behavior (conduct) and less in terms of any structural characteristics of a given industry or market. It is less likely that a summary measure of concentration, such as the widely adopted Herfindahl-Hirschmann index, can convey as much regarding the likely performance of a particular industry. That is, we cannot easily deduce that an increase in the level of concentration is likely to lead to a higher price-cost margin (or high supernormal profits) and a corresponding misallocation of resources. As noted previously, it might, on the contrary, be the case that a particular set of interactions between rival oligopolists, along with specific conjectures regarding likely potential entrants, leads to an outcome that is relatively competitive (i.e., one that is much closer to a perfectly competitive or Pareto efficient state than a pure monopoly outcome). (Young, 2000, p. 31)

Lieberskand et al. (1996) examined the impact of corporate restructuring measured at the industry level on industry concentration in US industries. These industries were engaged in the basic, manufacturing, and services sectors between 1981 and 1989. The results demonstrated a modest increase in median industrial concentration in sample industries between 1981 and 1989. There were few sell-offs of assets at the industry level through horizontal mergers, acquisitions, and inter-firm asset sales increased US industrial concentration during the 1980s (Lieberskand et al., 1996).

The Lieberskand study is interesting when one compares the manufacturing industry


Jürgen Albinger



to the telecommunications industry. The telecommunications industry has undergone a complete restructuring, as compared to the manufacturing industry. The manufacturing industry has been relatively steady, yet still shows a trend toward a higher level of concentration. The manufacturing industry has high barriers to entry, and in many cases, only one of two producers of a particular product. Lieberskand was baffled by the trend towards a more concentrated marketplace in what he considers to be a free competition.

Lieberskand considered the entire manufacturing sector to be a free competition. However, in reality it is a bunch of small natural monopolies. For instance, Colgate/Palmolive makes hundreds of consumer products. There are only two or three competitors that manufacture similar products. One of their primary competitors is Proctor and Gamble. They could be considered to be a duopoly. However, a closer examination of the market shares would have to be undertaken. This industry does not have hundreds of competitors. This represents only one product line.

Another example is in the airline industry. There are only two or three major manufacturers of airplanes. Lieberskand made the error in his analysis of considering manufacturing to be one entity. However, it is really a number of smaller entities and this is the factor that confounded Lieberskand’s analysis. Lieberskand was confused by his findings and this is explanation that eluded him. Some of these monopolies are natural monopolies. However, most are monopolies of scale.

Economic textbooks have called the telecommunications industry an ideal model of a natural monopoly (Thierer, 1994). However, Theirer points out that this monopoly came into existence with the aid of the US government and that many would-be competitors were excluded and not allowed

to obtain licenses in the beginning. As stated before, monopolies are detrimental to the market place by creating higher prices and sometimes-lower quality. Let us examine one possible reason why the government would wish to promote a monopoly.

A natural monopoly can serve consumers at lower costs than two or more firms. This would seem contrary to popular convention until one considers the high barriers to entry into a natural monopoly. Once a single firm overcomes the initial costs, such as laying the cable, or building the sub-stations, their average cost of doing business drops rapidly, relative to newcomers in the industry.

Kellogg, (2000) found that if a monopolist sets their price before they have determined the level of demand for their product, then they have essentially limited the availability of output at lower prices. It cannot lower its prices and this can lead the way for competition and lead to its eventual displacement as a monopoly. Kellogg cautions companies to wait until demand can be determined before setting prices.

Kellogg (2001) examined the homogeneous good Cournot model with two existing companies and one potential entrant into the market. This researcher demonstrated that entry can occur even if the entrant has no cost advantage and must rely on existing companies to distribute their product. This is exactly what happened in the wireless portion of the telecommunications industry. In the beginning there were only a few companies manufacturing cell phones. The costs to enter this market are high. However, in spite of this, other manufacturers were able to enter the market and there are now a large number of manufacturers in the industry. Some manufacture their phones and sell them under a larger conglomerate name. This gives the larger conglomerate another product line, for which it does not have all of the manufacturing costs and

2009 55

gives the smaller company the advantage of the larger companies brand equity. The Cournot model could prove itself to be highly useful in the development of a model for the telecommunications industry.

The practice described above is called “bundling.” Martin (1999) found that bundling be a firm with a monopoly over one product is strategic and that it changes the substitution relationships between the goods. Now the monopoly does not have to worry about the competition as a substitution product. It can now use the product produced by the other company to its advantage. This is a highly profitable proposition for the monopoly. This strategy is usually used by the monopoly to reduce or eliminate competition. In many cases, the larger company will force the smaller company into this arrangement as they can easily override them otherwise.

Labor Unions are not usually considered to be an monopoly. However, there have been many models that prove them to have all of the characteristics of a monopoly. Many models have been developed describing the characteristics of labor unions as being monopolistic in nature. Some of these models may be useful in determining the existence of monopolies in the telecommunications industry because the structure of the labor union is similar to the telecommunications industry.

Labor unions are national in scope, just as with the national telephone carriers. However, they are local in nature as well, small branches serve limited sections of the population. Labor unions engage in a variety of trades, such as plumbing, painting, electricians, etc. They have many products to offer and each of these products represents a different market. This is very similar to the telecommunications industry where there are long-distance markets, local service markets, cell phones, PDA’s and

other products, all representing a different market. For these reasons, models that are used to describe the labor unions may prove to be the most useful in describing the telecommunications industry.

In order to develop an adequate model for describing the telecommunications industry, it is obvious that an examination of other industries and comparisons will be required. Similarities and differences will have to be carefully considered. Currently, there are many models that claim to be the answer in all cases involving the existence of a monopoly. In the models that make these claims, one thing is obvious: They did not compare the market characteristics of the industries for which they have proposed the model. In the case of the telecommunications industry, it was surprising to find that the labor unions were the most similar in structure and market scope. These may be the most useful in constructing a model for conduct of this research.

Many of the models were sound in theory. However, sufficient testing was not performed to substantiate the claims being made. It is not expected that any one of these models will adequately apply to the telecommunications industry. However, in the case of the labor union models, a few modifications may make the most meaningful model. These statements are only speculative at this point and no clear conclusion regarding the methodology can be proposed at this time.

The literature on the telecommunications industry has to this point, focused on the effects of the Telecommunications Act of 1996. There are as many reports that state that the act created a competitive marketplace. Research has shown that this may have been true, at least for a while. However, it is possible that the marketplace may be changing to a monopolistic model in


Jürgen Albinger



the near future, especially with the upcoming proposed merger between MCI WorldCom and Sprint and AT&T and Bell South.

Michael Porter and Monopolies and Clusters

By all accounts, segment differentiation and geographic scope are central concepts of international strategy at the business level (Porter, 1986). According to Cooke (2001), Michael Porter is one of the leading proponents of clustering for competitive advantage. From a competitive advantage perspective, the goal of segment differentiation is to customize products and services to better fit the needs of different groups of customers; however, Porter points out that such segment differentiation is expensive because differences in product designs, advertising programs, distribution channels, and so forth work against the development of economies in research and development, production and marketing. “In contrast,” Carpano, Chrisman, and Roth (1994) advise, “if a firm fails to segment its market, it risks sacrificing effectiveness for efficiency; few customers will be attracted by a product that fails to meet their needs, no matter how economical the purchase may be” (p. 640). In an industry characterized by multidomestic enterprises, the needs of customers are regarded as being heterogeneous; companies that do not adequately respond to the diversity of customer needs and buying motives may find themselves at a competitive disadvantage compared to companies that do (Carpana et al., 1994). “On the other hand,” they advise, “customer needs in a global industry are predominantly homogeneous. In this instance, a firm that attempts to differentiate its offerings to different national markets may see its attempt go unrewarded. It has,

in other words, sacrificed efficiency with no corresponding gain in effectiveness” (Carpana et al, 1994, p. 640).

Knowledge Engineering in Relation to Monopolies and Business Intelligence Applied to Monopolies

According to Davidson (1999), “Today’s knowledge-based organisations have a problem. Their most important assets tend to be in the heads of their workers. As a result they need tools to capture, store and redistribute this knowledge, so that the corporation doesn’t grind to a halt every time an employee walks out of the door” (p. 37). To date, computer scientists have been able to integrate various individuals’ knowledge into “expert systems” that autonomously perform complex tasks such as configuring large-scale computer systems and assessing credit applications. It was not that long ago that such knowledge engineering systems, though, were considered too restrictive in their scope to be of any real significant use to all but the most esoteric applications.

For example, Davidson notes that although experts in artificial intelligence had created a number of “knowledge engineering tools” to help build expert systems, at the time, “Few companies have the time or resources to use them to capture knowledge in this depth. Nor would they necessarily want to. Expert systems are limited inevitably to a single, specific field of interest, whereas what companies often want from their knowledge workers is flexibility and creativity” (Davidson, 1999, p. 37). This situation is no longer the case today, however, and a wide range of expert systems, algorithms, knowledge bases and various business intelligence applications can help model real-world settings and help researchers better understand the various forces at play. According to Hakken (2003),

2009 57

knowledge engineering requires a solid theoretical basis: “[A] computer, no matter how sophisticated, is useless if you have…no rules to drive the inference machinery designed to process…data. Knowledge Engineering attempts to codify and distil information so that it can fuel the increasingly powerful logic machines processing the data stored in their memories. Intelligent knowledge-based systems (IKBSs) and expert systems are computer-based systems which take information and process it in an ‘intelligent’ manner” (Hakken, 2003, p. 154).

From another perspective, then, clusters are monopolies in the market for economic space (Cooke, 2001). Some early uses of knowledge engineering for estimating forward-looking costs of a local telephone exchange were made by Mitchell (1990, cited in Mackie-Mason & Waterman, 1998) who used a heuristic model of the local exchange network, including the local loop, end-office switching, tandem switching, and interoffice transport. According to these authors:

Simple linear functional forms were assumed for individual cost components of the network, and these were calibrated either by econometric techniques where data were available, or by engineering estimates provided by Pacific Bell and GTE of California, in whose territory the study was conducted. Both capital costs (network investments multiplied by a 15 percent annualization factor) and operating expenses (primarily maintenance and billing costs) were considered. The end result of the study was an estimate of both the fixed costs and average incremental costs associated with each of the network components. Mitchell’s study focused on estimating the incremental cost of an additional access line or an additional minute of usage of the network. These results are of

potential interest in guiding investment decisions of telecommunications providers and, in a regulatory context, for setting price floors or for detecting predatory behavior.

Using such knowledge engineering systems to improve returns on investment tend to accrue to first-mover firms in a knowledge-based industry or those that otherwise gain a monopoly position in a given market has become increasingly common today. In this regard, more recently, Cooke (2001) reports that:

Some Knowledge Economy firms with an arm-lock on key Internet technologies, like Cisco Systems, virtual monopolists in Internet ‘routers’ which control Internet communication channels, have this characteristic, as did telecommunications monopolies before privatization and the introduction of competition for telecom services market-share. Increasing returns arise from the widespread and repetitive gains that accrue from the continuously improved codified knowledge responsible for the technological change in question. In Cisco’s case this is achieved by acquisition of specialist knowledge-intensive smaller businesses in appropriate technologies. In other cases it arises from equity investment by corporate venturing arms of such (quasi)-monopolists in equivalent small businesses, as in the case of Intel Technology, which is assumed to possess this increasing returns characteristic. This, in turn, requires the replacement of the neoclassical cornerstone assumption of perfect by one of imperfect competition, since increasing returns to scale clearly imply monopolistic rather than fair competition. [M]ost of these models suffer from some of the same unrealistic assumptions and the same


Jürgen Albinger



measurement problems as the ‘old’ growth models. In particular, they take little or no account of organizational innovations and of the interplay between institutional change, technical change and investment. (Cooke, 2001, p. 35).

Finally, other knowledge engineering techniques that have been applied to the telecommunications industry in recent years include the Local Exchange Cost Optimization Model (LECOM), used to calibrate the cost function for local telecommunications costs. According to Mackie-Mason and Waterman (1998), “LECOM has been developed to determine the degree of economies of scale and scope in local telephone markets and the cost-minimizing deployment of technology (e.g., analog vs. digital or fiber vs. copper). The software for LECOM combines an engineering process model, which computes the cost function for a local exchange network with a given configuration of switch locations, using an optimization algorithm which solves for the optimal number and location of switches” (p. 21).

The LECOM knowledge engineering applications allows the user to specify the size of the local exchange territory, which is comprised of three concentric rectangles representing a central business district, an area of mixed residential and commercial demand, and a residential district. The size and the population density of each region are both specified by the user, as well as specific data concerning calling patterns. Furthermore, the user is able to specify a set of technological inputs and a detailed set of input prices in order to calibrate the model to the specific characteristics of an exchange area (Mackie-Mason and Waterman, 1998). By all accounts, more sophisticated applications of these techniques are expected in the near future.

Methodology

This research will be conducted in two phases, to be discussed in detail in the remainder of this section. The first phase of research involved a detailed review of the telecommunications industry, both historically and in present times. Existing methodologies for determining the existence of monopolies were explored in detail and compared to the needs of the telecommunication industry. Finally the developed model was tested using examples from the telecommunications industry both before and after the Telecommunications Act of 1996.

Description of the Study Approach

The first section of this research involves development of the model. This involved a detailed study of existing models and how they apply to the circumstances for which they were originally developed. These models were then studied to determine if they would be useful in the telecommunications industry. Unique factors in the telecommunication industry will be identified and worked into the equation. The model in this research served as the research instrument.

After the development of the model, the model must be tested to ensure that it works as planned. It will be tested in several environments. It will be tested before the adoption of the Telecommunications Act of 1996. If the model reacts as planned it would be expected that the model will show strong evidence for the existence of a monopoly. The model will be tested after the adoption of the Telecommunications Act of 1996, where it should not show strong evidence for the existence of a monopoly. A variety of tests will be conducted in more recent times where the existence of a monopoly is questionable. If the results obtained fail to

2009 59

make the expected results, then the model will be reexamined and adjusted. The tests will be repeated until the desired results are obtained and found to be statistically significant.

Sample Population

The sample population for this study will be various sub-sectors of the telecommunications sector over the period of time from 1986 through 2002; this period will cover the period both before and after the Telecommunications Act of 1996. It is not believed from historical literature examined that going farther back than 1986 will produce significant results, as the market remained relatively unchanged until this time. According to Compaine and Gomery, using the year 1986 as the base year for comparison is appropriate for this purpose because it was the first year after the Federal Communications Commission relaxed the number of television stations under the ownership of a single firm from 7 to 12; in addition, 1986 was the year that News Corp. first successfully challenged the longstanding dominance of the major three commercial networks, thereby creating an opportunity for additional competition in the broadcasting industry (Compaine & Gomery, 2000). Furthermore, in the early 1990s, the FCC’s restrictions on broadcast networks owning a financial interest in prime time programming were eased and by early 1996, the Telecommunications Act essentially eliminated the size of broadcast radio groups and further loosened restriction on television station group ownership (Compaine & Gomery, 2000).

As the research progresses, the exact time periods of interest are expected to emerge. One example may be to study sector structure at the beginning of the advent of cell phones, and then one year after their

inception to determine if any leaders have emerged and if they did, if they began to resemble a monopoly, either by scale or by location. Then the same sector would be analyzed three years after its beginning to study how it has changed in that time. This is only one example of the types of questions that may be examined in this research. There are many such issues that may be examined upon closer investigation of the sectors over time.

Data Analysis

The first portion of data analysis will be to determine if, when the model is applied, a monopoly is found to exist. Each time period examined will be considered to comprise one data set. Each data set will be either positive or negative for the existence of a monopoly. Some may be considered to be partial monopolies. These will be treated as a monopoly for purposes of this study.

This data will be analyzed and scatter-plotted according to time to determine if more or less monopolies come into existence over a long period of time. Patterns in this data are expected to reveal trends in the formations of monopolies over a period of time. Trends in the movement towards a monopoly or towards a more even competition are expected to be revealed by this graphical method. Regression analysis will be used to attempt to predict future trends in the formation of monopolies in the telecommunications sector in the future.

Findings

It is expected that the data will reveal definite trends in the telecommunication industry. For many years the telecommunications industry was dictated by a few large-scale companies. There were virtually no smaller players and even if they


Jürgen Albinger



did exist, they could not compete with the Bell conglomerate. The Telecommunications Act of 1996 paved the way for a free market in the Telecommunications sector.

Immediately following the passage of the 1996 Act, a plethora of small companies emerged. The government supported the growth of the industry by providing more licensing opportunities and providing grants and low interest loans to those who wished to enter into the lucrative enterprise. When an industry booms such as this one, there are many who enter that are not prepared to compete. Many lack the knowledge and skills necessary to make a sustainable proposition. As a result, many of these fledgling businesses fail in a short amount of time.

This left gaps in the market to be filled, either by another new company, or by an existing company that is positioned to take over that area of the market. More often than not, the latter is the case. This makes the existing company stronger and more able to control a larger section of the market. During this phase of market development, this same process occurs many times over. In the end, many of the small companies that were unable to compete will disappear and the market will emerge with fewer players, each controlling a larger section of the market than in the beginning of this development.

Sometimes, as many believe is the case currently in the telecommunications sector, one of these strong contenders will capture a large enough marketshare that they themselves become another monopoly by scale. Mergers and acquisitions can help play a role in this process. At this stage, prices tend to level, production levels and the market will eventually reach equilibrium. This has been the stage that has occurred in the telecommunications industry. Currently the market is in the stage where many of the smaller companies are beginning to fade

into the sunset, and the few stable success stories are beginning to emerge.

The main question that must be posed at this juncture in time is whether there is currently a monopoly either existing now, or arising in the telecommunications sector? The other question is to attempt to determine whether this trend will continue in the future or whether there will again be another restructuring in the telecommunications industry. It is expected that the data will demonstrate the trends described earlier in an emerging market and that the market will be found to be reaching a period of more stability and steadier, less explosive growth. No predictions regarding findings concerning the existence of a monopoly can be made at this time. Nor can predictions regarding future trends be revealed until the data for the final report is collected and analyzed.

Conclusion

The telecommunications industry is unique and previous models to product the existence of a monopoly have to this point proven inadequate. Many of them only apply in certain sectors of the market, while others are only applicable on a limited basis, such as in the rural local telephone industry. It is expected that the final model will draw from these previous models, but that it will have to be modified to fit the circumstances unique to the industry. The possibility exists that one model may not be adequate in assessing the entire telecommunications industry. A series of models may eventually have to be developed which are applicable in various markets, or situations in the market. This cannot be determined until the research is underway.

Many of the models discussed in the literature review purport to have a wide applicability. However, upon closer

2009 61

examination, it can be found that they were not adequately tested to make such claims. In addition, the samples for these tests were not selected according to a certain criteria. It cannot be determined if the sample sets used to make draw these conclusions contained markets that were similar in characteristic, or if the models were tested on a variety of markets under widely varying conditions. In order to make the types of claims made by these models, one would have to have more information on the markets and test conditions than was provided in the reports.

The primary concern in developing the model or models that will be used to determine that existence of a monopoly will be certain to learn from past mistakes and not make the same mistakes as our predecessors. In order to avoid this conundrum, the research and test methods must include protocol for testing in various time periods under varying market conditions. In addition enough tests will have to be performed to make an accurate determination as to the applicability. The actual test results will have to be critically view to be certain that they are logical given what is qualitatively know about the market place. The entire industry will have to be tested as a whole, as well as the individual sectors separately. Only then can one make a determination as to the usefulness of the model in varying market conditions.

Though the previous models have been criticized, they cannot be discounted altogether. They will still be useful as examples in developing the final model. Surprisingly, the labor union models for determining monopolies were initially revealed to hold the greatest promise for providing a basis for the model. The market is nationwide, yet local in nature. It involves many trades, all representing different markets. These markets are similar to the various sub-sectors in the telecommunications industry. The labor union models may provide an

interesting ground for cross- testing our own model to determine if it applies in other industries as well.

Today, the telecommunications industry is changing rapidly and there is a growing need to define what constitutes a monopoly in this sector. This research hopes to resolves the existing dilemma by developing and testing a model for predicting the existence, either now or in the future of monopolies. The telecommunications industry is unique and models developed in the past fail to meet the needs of this industry. It is hoped that the model developed as a result of this research will prove useful in settling the many questions surrounding anti-trust in the telecommunications market.

Data Analysis

Herfindahl-Hirschmann Index.

According to Jorde and Teece (1992), the term “Herfindahl-Hirschmann Index” is a term used to refer to the total of the squares of the market shares of all actual competitors in any relevant market (Compaine & Gomery, 2000). This measure of concentration is increasingly being used in empirical work and is typically abbreviated to “HHI” (Sawyer, 1991, p. 29). This index is defined according to the number of firms in the industry (T), with the share of each firm (si) weighted by itself; while the concentration ratio places a weight of 1 on the share of the largest n firms and zero on the share of the other firms, this index places a weight equal to the share on the share of the firm. As a result, the index is able to incorporate information on all firms rather than just the largest n firms. In addition, the HHI can vary between a value of zero (where there are a large number of roughly equally sized firms) and unity (where there is just one firm) (Sawyer, 1991).


Jürgen Albinger



Traditionally, economic criteria have been the only or at least the key criteria that have been used to determine the concentration in all industries; while the congressional debate over the Celler-Kefauver Amendment to Section 7 of the Clayton Act suggested that social and political criteria should be considered in antitrust proceedings, the amendment itself did not include such language (Compaine & Gomery, 2000). Consequently, the assessments of efficiency and entry to an industry are measured by such criteria as the percentage of industry revenue accounted for by the largest players, as well as by the more complex Herfindahl-Hirschmann Index. In the most conservative determination of oligopoly, proposed by Carl Kaysen and Donald Turner (1959, cited in Compaine & Gomery, 2000), the eight largest firms would have at least 33-1/3 percent of sales and the 20 largest at least 75 percent; this is called a type II oligopoly.

In this calculation, the index number increases as the number of companies declines and the inequality in the largest and smallest increases; HHIs of 1,800 or greater represent industries of great concentration and index numbers at or below 1,000 are signs of no concentration; according to Compaine and Gomery (2000), “From an antitrust viewpoint, an industry must reach a type I oligopoly, at which time the eight largest firms have 50 percent of receipts and the 20 largest at least 75 percent, before the concentration allows firms to charge prices and make profits above competitive levels and to misallocate resources” (pp. 555-6).

Among the various economic measures of concentration, the Herfindahl-Hirschmann Index (HHI) 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” (Compaine & Gomery, 2000, p. 558). As noted above, the HHI is calculated by squaring the market share of each player

in the industry. Generally speaking, an HHI score of greater than 1,800 indicates a highly concentrated industry; a rate that is under 1,000 is regarded to be unconcentrated, with scores in between representing degrees of moderate concentration (Compaine & Gomery, 2000).

In an example of an industry with 10 providers, the HHI has the ability to differentiate between a marketplace 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 percent, 25 percent and 20 percent, respectively, of industry sales. The remaining seven divided up 25 percent about equally. This example has an HHI of 2,014, which is highly concentrated. By contrast, in example B, the largest firm has a 15 percent market share, the second firm 12 percent, the third 10 percent and the seven remaining firms roughly divide the remainder; the HHI in this industry is 1,036, which is a low concentration (Compaine & Gomery, 2000).

Yet another study applied the HHI model to the book publishing industry from 1989 to 1994 (Greco, 1999). According to Greco’s analysis of a “media monopoly” in the book publishing industry using the Herfindahl-Hirschman Index to ascertain whether Department of Justice antitrust guidelines were violated because of these mergers and acquisitions. “Between 1960 and 1994, there were approximately 72,000 mergers and acquisitions in the United States. This metamorphosis captured the attention of scholars eager to understand the impact this transformation had on the business landscape” (Greco, 1999, p. 165). This period followed several decades of apparent consolidation from mergers, including the 1970s period that resulted in the decision by the Federal Trade Commission to investigate media concentration in 1978 (Greco, 1999).

2009 63


Jürgen Albinger

According to Greco, the impact horizontal mergers and acquisitions had on concentration in the U.S. book publishing industry between 1989 through 1994 were appropriate for investigation because these years represented peak-to-peak periods during a period of relative economic prosperity; furthermore, previous studies on book publishing ended with 1989, so this research, which was based on more recently released statistical data, was an important study of mergers and acquisitions into the mid-1990s (Greco, 1999). The author analyzed empirically data sets on book publisher’s market shares and revenues between 1989 through 1994 (due to data collection lags, the most recent ones available) using the Herfindahl-Hirschman Index to determine whether antitrust guidelines were violated because of the mergers and acquisitions in the book industry; additional issues related to market entry barriers, title output, and pricing were also take into account (Greco, 1999).

The Book Industry Study Group divides the U.S. book industry into three separate market categories: (a) consumer books, (b) professional and reference books, and (c) the education or textbook market (including workbooks); each of these markets has specific considerations that Greco took into account in his analysis. The author first examined a variety of analytical methodologies in an effort to accurately measuring market concentration levels, including “Lerner’s Index of Monopoly Power” (Lerner, 1934, cited in Greco, 1999), in which p is price and MC is marginal cost:

L = (p-MC)/p (1)

Likewise, Greco reviewed the “Bain Index of Monopoly” (Bain, 1941, cited in Greco, 1999) was also investigated; this methodology focuses on excess profits,

which would indicate the presence of a monopoly. To accomplish this analysis, all relevant costs must be deducted from total revenue; thereafter, the index is expressed in terms of where PQ is price times quantity or total revenue, C(Q) represents the current cost of generating the income, D represents the depreciation on fixed capital investment, and iV is the opportunity cost on the owned assets of the firm or the implicit costs of the firm.

p = PQ - C(Q) - D - iV (2)

The author reports that both of these methodologies proved unusable because the information needed to assess demand elasticity and marginal costs was not available in annual reports, industry data, or U.S. government statistics, a point that was also identified by the author of the instant report as well. Based on this review, Greco reports that the HHI emerged as the best tool to measure market concentration levels because:

It was the analytical formula employed 1. by the U.S. government (by both the Department of Justice and the FTC) and,

The necessary post-1989 data were 2. available.

The Department of Justice Manual (1997, cited in Greco) stated that “market concentration is a function of the number of firms in a market and their respective market shares”; as an aid to the in terpretation of market data, the Agency will use the [HHI] of market concentration” (p. 7-58). Once an index is determined, the Department of Justice (p. 7-58) divided a market into three distinct regions: (a) unconcentrated (HHI below 1,000), (b) moderately concentrated



(HHI between 1,000 and 1,800), and (c) and highly concentrated (HHI above 1,800). The author concludes that the HHI is “widely considered to be a useful summary measure of market concentration because it reflects, at least to some extent, the number and size distribution of firms in a market, as well as concentration of output”; researchers have also consistently demonstrated the efficacy of the HHI in their study of various industries. For example, Viscusi, Vernon, and Harrington (1995, cited by Greco, 1999) described the HHI as the “weighted average slope of the concentration curve. . . . The weight for the slope of each segment of the curve is the corresponding MS i for that segment” (p. 150). According to Greco, “Where MS i represents the market share of firm i and where there are n firms in the market. If a market consisted hypothetically of a single seller, the HHI would be 10,000, a perfect example of a monopoly” (p. 171).

As can be seen in Table 1 below, during the period studied, the 14 largest book publishers accounted for 75 percent to 80 percent of total book industry revenue; however, the industry experienced some significant swings over the years as well.

Table 1. HHI for Book Publishing, 1989-1994.% of Industry Revenue by 14 Largest Publishers

HHI Index

1989 74.6 4541990 78.6 4881991 73.7 4431992 74.6 4501993 76.0 4641994 80.0 511Source: Compaine & Gomery, 2000, p. 559

Figure 1. HHI for Book Publishing, 1989-1994.Source: Based on data in Compaine & Gomery,

2000

The highly fragmented nature of the industry can be seen in the HHL; even when it was reported at its highest (in 1994), the HHI reflected a highly competitive industry, well below even the low boundary of oligopoly (Compaine & Gomery, 2000). According to Greco (1999), the U.S. data collected for the top 14 book publishers showed that they held approximately a 75% market share between 1989 through 1994; an analysis of the HHI totals found that in 1989 these top 14 firms accounted for 74.57% of all book sales ($10.523 billion) in the United States and had an HHI of 454.06 and by 1990, these corporations had sales of $11.68 billion (78.62% share), with a HHI of 488.46 (Greco, 1999)

The industry enjoyed another increase in revenues in 1991; as the result of fairly intense competition in the market place, though, Greco notes that their total market share declined to 73.65%, as did the HHI (443.49), lower than the 1989 mark. Nevertheless, by 1992, revenues had increased once again and reached the $12.173 billion mark; market shares eased upward (74.55%) with an HHI of 449.77. In the following year, the top firms’ posted revenue increases (75.96%), topping the $13.2 billion plateau. The HHI

0

100

200

300

400

500

600

% Ind Rev - 14Largest

HH1 Index

1989 1990 1991 1992 1993 1994

2009 65


Jürgen Albinger

increased to 463.82; in 1994, the last year in this study, the major firms accounted for $14.591 billion in sales, a 79.95% market share, and a 510.88 HHI, the highest totals in this 6-year cycle (Greco, 1999).

The study also calculated the effect of a single company controlling the 25 percent to 20 percent of the industry revenues not accounted for by the 14 in the study, a figure that would have resulted in an HHI of 931 in 1994, representing a decrease from 1,101

in 1989, below the minimum required for designation as low concentration according to the Herfindahl-Hirschman Index (Compaine & Gomery, 2000). In addition, this research further confirmed 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 (Compaine & Gomery, 2000).

Table 2. Top 14 Book Publishers’ Market Share: 1989-1994.

PublisherMarket Share

1989 1990 1991 1992 1993 1994

Simon & Schuster 9.35 9.57 9.31 9.33 9.77 9.79

Time Warner Book Group 8.02 6.58 7.07 6.31 6.20 6.25

HarperCollins 8.14 8.19 6.67 5.88 5.70 6.01

Readers Digest 6.69 7.82 8.36 8.62 7.54 7.94

Random House 6.02 6.66 6.49 6.74 6.61 7.21

Harcourt General 5.85 5.72 5.14 5.30 5.43 5.04

Bantam Doubleday Dell 5.49 5.52 5.30 5.51 5.17 5.21

Encyclopaedia Britannica 4.42 4.38 3.90 3.47 2.63 --

Maxwell Macmillan 3.76 3.75 3.42 -- -- --

McGraw-Hill 3.54 3.71 3.42 3.47 3.83 6.37

Times Mirror 3.54 3.70 3.97 4.95 4.85 4.82

Thomson 3.52 4.68 4.82 5.31 7.03 7.46

Viking-Penguin/Addison Wesley 3.38 3.73 4.12 4.45 5.43 5.27

Houghton Mifflin 2.86 2.83 2.68 -- 2.67 2.65

Scholastic -- -- -- 2.65 3.09 3.36

Harlequin -- -- -- 2.56 -- 2.60

14 firm totals 74.57 78.62 73.65 74.55 75.96 79.95

All other firm totals 25.43 21.38 26.35 25.45 24.04 20.50

Source: Greco, 1999, p. 172.



Figure 2. Top 14 Book Publishers’ Market Share: 1989-1994. Source: Based on data in Greco, 1999, p. 172.

Table 3. Top 14 Book Publishers’ Herfindahl-Hirshman Indices: 1989-1994.

PublisherHHI

1989 1990 1991 1992 1993 1994

Simon - Schuster 87.42 91.61 86.68 87.05 95.45 95.84

HarperCollins 66.23 67.10 44.49 34.57 32.49 36.12

Random House 36.24 44.36 42.12 45.43 43.69 51.98

Readers Digest 44.76 61.15 69.89 74.30 56.85 63.04

Time Warner 64.32 43.30 49.98 39.82 38.44 39.06

Harcourt General 34.22 32.72 26.42 28.09 29.48 25.40

Bantam Doubleday Dell 30.14 30.47 28.09 30.36 26.73 27.14

The Thomson Corporation 12.39 21.90 23.23 28.20 49.42 55.65

Encyclopedia Britannica 19.54 19.18 15.21 12.04 6.92 --

Macmillian/McGraw-Hill

School Publishing -- 14.06 -- -- -- --

Viking-Penguin;

Addison-Wesley 11.42 13.91 16.97 19.80 29.48 27.77

McGraw-Hill 12.53 13.76 11.70 12.04 14.67 40.58

Times Mirror 12.53 13.69 15.76 24.50 23.52 23.23

Houghton Mifflin 8.18 -- 7.18 -- 7.13 7.02

Scholastic -- -- 5.77 7.02 9.55 11.29

Harlequin -- -- -- 6.55 -- 6.76

Maxwell Macmillan 14.14 21.25 -- -- -- --

Total industry HHI 454.06 488.46 443.49 449.77 463.82 510.88

Source: Greco, 1999, p. 173.

0

10

20

30

40

50

60

70

80

1989 1990 1991 1992 1993 1994Simon & Schuster Time Warner Book Group HarperCollins Readers Digest Random House Harcourt General Bantam Doubleday Dell Encyclopaedia Britannica Maxwell Macmillan McGraw-Hill Times Mirror Thomson Viking-Penguin/Addison Wesley Houghton Mifflin Scholastic Harlequin 14 firm totals All other firm totals

2009 67

Figure 3. Top 14 Book Publishers’ Herfindahl-Hirshman Indices: 1989-1994.Source: Based on data in Greco, 1999, p. 173.


Jürgen Albinger

Although the specific HHI for the top 14 firms was calculated, this was not possible for each one of the remaining 2,630 firms because data sets are not available for this portion of the industry; assuming a hypothetically perfect sharing of the remaining highly fragmented market, though, Greco notes that the “average” firm in 1994 would have revenues of $1.39 million, a 0.01% market share, and an HHI of 0.0001. The total HHI for this entire industry subset in 1994 would fall below the 0.30 rate (Greco, 1999).

In addition, if it is assumed that the entire remaining market were held by one company instead of 2,630 firms, then the adjusted total HHI totals for the entire industry would be as follows: (a) 1989: 1,100.74; (b) 1990: 945.56; (c) 1991: 1,137.81; (d) 1992: 1,097.47; (e) 1993: 1,041.74; and (f) 1994: 931.13. The HHI dropped 169.61

points between 1989 through 1994, falling below the U.S. Department of Justice and FTC unconcentrated 1,000 threshold. Entry barriers were quite low to support a total increase of 1,708 firms (+182.48%) since 1963 and marginal changes in the HHI between 1989 through 1994. Based on his review of the 1989 through 1994 empirical data on market concentration, as well as statistical information concerning new entrants into the book marketplace, new title output, and book prices, Greco concluded that the creation of a “media monopoly” or illegal levels of concentration in the U.S. book publishing industry could not be substantiated.

A similar type of analysis may be applied to the television broadcasting segment as shown in Table 4 below.

0

100

200

300

400

500

600

1989 1990 1991 1992 1993 1994Simon - Schuster HarperCollins Random House Readers Digest Time Warner Harcourt General Bantam Doubleday Dell The Thomson Corporation Encyclopedia Britannica Macmillian/McGraw-Hill School Publishing Viking-Penguin; Addison-Wesley McGraw-Hill Times Mirror Houghton Mifflin Scholastic Harlequin Maxwell Macmillan Total industry HHI



Company 1997 (in millions)

1996(in millions)

1995(in millions)

1997 Media % of Total Revenue

1996Media % of Total Revenue

1995Media % of Total Revenue

1. CBS $4,839 3,390.0 3,483.0 88.6 78.2 79.6

2. NBC (GE) 4,803 4,940.0 3,659.0 93.2 94.5 93.4

3. ABC (Walt Disney) 4,572 4,005.0 4,177.0 66.2 61.1 67.2

4. News Corporation 2,730 2,500.0 1,580.0 49.2 62.4 53.7

5. Tribune 927 681.0 630.0 39.2 32.3 31.2

6. Gannett 704 641.5 523.8 15. 7 15.2 13.1

7. Cox 650 391.0 354.0 19.6 12.7 13.0

8. A.H. Belo 537 333.4 322.6 43.1 40.4 43.9

9. Sinclair 517 326.0 273.5 100.0 86.1 100.0

10. Univision 460 370.3 321.3 100.0 100.0 100.0

11. BHC Commun* 444 446.3 454.7 100.0 100.0 100.0

12. Viacom 422 390.3 385.5 15.7 16.2 19.0

13. Hearst 422 368.0 285.0 14.9 14.3 11.3

14. Washington Post 338 335.2 306.0 18.8 19.6 19.3

15. E.W. Scripps 331 349.6 320.3 30.1 29.9 29.2

16. Lin Television 292 273.4 217,2 100.0 100.0 100.0

17. Raycom Media 286 200.0 187.0 100.0 97.3 97.2

18. Young Broadcasting 264 261.5 245.8 100.0 100.0 100.0

19. Pulitzer Publishing 227 208.5 188.0 38.8 39.0 39.8

20. Chronicle Publishing 148 147.0 131.0 32.8 31.5 24.5 Source: Compaine & Gomery, 2000, p. 196.

Figure 4. Broadcast Revenues of Media Largest Companies, 1995-1997.Source: Based on data in Compaine & Gomery, 2000.

0

10

20

30

40

50

60

70

80

90

100

1997 1996CBS NBC (GE) ABC (Walt Disney) News Corporation Tribune Gannett Cox A. H. Belo Sinclair Univision BHC Commun* Viacom Hearst Washington Post E. W. Scripps Lin Television Raycom Media Young Broadcasting Pulitzer Publishing Chronicle Publishing

Table 4. Broadcast Revenues of Media Largest Companies, 1995-1997

2009 69


Jürgen Albinger

Figure 5. Average Weekly Television Viewing, Selected Years, 1950-1997. Source: Based on data

in Compaine & Gomery, 2000.

During 1997, these companies had an aggregate of $23.9 billion in broadcast revenue; CBS and NBC each accounted for approximately 20 percent of the total; i.e., ABC 19 percent and Fox 11 percent (Compaine & Gomery, 2000). Table 5 below shows these four top firms accounted for fully 60 percent of the revenue of the top 20 broadcasters, which in turn had the dominant share of the total broadcast market; however, 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 (Compaine & Gomery, 2000).

Table 5. Average Weekly Television Viewing, Selected Years, 1950-1997.

Weekly Set Usage Per Home

Channels Available Per Home

Channels Viewed Weekly

Time Spent Per Channel Weekly

1950 32.5 Hours 2.9 2.8 11.6 Hours

1960 36.5 5.9 4.2 8.7

1970 42.0 7.4 4.5 9.3

1980 46.5 10.2 5.6 8.3

1990 483.5 27.2 8.8 5.5

1997 50.0 43.0 10.3 4.9Source: Compaine & Gomery, 2000, p. 206.

05

101520253035404550

No. of Hours

1950 1960 1970 1980 1990 1997

The data in Table 6 below indicates that television broadcasting is a moderately concentrated industry which represents no major surprise; however, the table also shows that over this period the concentration, as measured by HHI, also decreased by almost 12 percent among the 20 largest companies. The revenue share of the top four and top 10 players was found to be lower, due mostly to the mergers at the bottom of the industry which resulted in more intensive 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. An approximate estimate by the authors is that, in 1980, the three largest players (the networks) held an industry share similar to that of the four major networks in 1997 (Compaine & Gomery, 2000).

The authors report that these are small changes; however, at a minimum, they suggest a different outlook than the intuitive one created by merger announcements (Compaine & Gomery, 2000).

Table 6. Revenue Share and HHI of Largest Broadcasters, 1994-1997.

Total Revenue Top 20 (billion)

Share Top 4 Share Top 10 HHI

1994 18.9 72.6 87.8 1553

1995 18.0 71.5 86.3 1455

1996 20.6 72.2 86.4 1432

1997 23.9 70.9 86.7 1372Source: Compaine & Gomery, 2000, p. 559.



Figure 6. Revenue Share and HHI of Largest Broadcasters, 1994-1997. Source: Based on data in

Compaine & Gomery, 2000.

70

70.5

71

71.5

72

72.5

73

Share Top 4

1994 1995 1996 1997

A comparison of the overall media industry look based on concentration percentages as well as the HHI is shown in Table 7 below.

Table 7. Media Revenue of the Largest Media Companies, 1986 and 1997.

No. Parent Company

1997 Media Revenue % (mil)

Total Parent Company1986 Media Revenue %

Total HHI1997

HHI1986

1 Time Warner 22,283 9.22 CBS 4,714 5.61 85.03 31.52

2 Disney 17,459 7.22 Capital-Cities/ABC 4,124 4.91 52.20 24.13

3 Bertelsmann 9,525 3.94 Time 3,828 4.56 15.54 20.79

4 Viacom 9,051 3.75 Dun & Bradstreet 3,114 3.71 14.03 13.76

5 Sony 8,253 3.42 GE (NBC) 3,049 3.63 11.66 13.19

6 News Corp 7,695 3.18 Warner Comm 2,849 3.39 10.14 11.51

7 TCI 6,803 2.82 Gannett 2,802 3.34 7.93 11.14

8 Thomson 5,849 2.42 Times Mirror 2,684 3.20 5.86 10.22

9 Seagram 5,593 2.31 Newhouse 2,371 2.82 5.36 7.97

10 PolygramN.V. 5,535 2.29 Gulf + Western 2,094 2.49 5.25 6.22

11 CBS 5,363 2.22 Knight Ridder 1,880 2.24 4.93 5.01

12 GE(NBC) 5,153 2.13 Tribune 1,830 2.18 4.55 4.75

13 Reed Elsevier 4,902 2.03 MCA 1,829 2.18 4.12 4.75

14 Gannett 4,730 1.96 Hearst 1,688 2.01 3.83 4.04

15 Reuters 4,729 1.96 McGraw Hill 1,577 1.88 3.83 3.53

16 Cox 4,591 1.90 New York Times 1,565 1.86 3.61 3.47

17 Newhouse 4,250 1.76 Cox 1,544 1.84 3.09 3.38

18 EMI Group 4,088 1.69 News Corp 1,510 1.80 2.86 3.23

19 MediaOne 3,586 1.48 Coca Cola (Columbia) 1,374 1.64 2.20 2.68

20 McGraw Hill 3,534 1.46 Readers Digest Assoc 1,255 1.49 2.14 2.23

21 Times Mirror 3,298 1.36 Washington Post Co 1,162 1.38 1.86 1.92

22 Pearson 3,066 1.27 Dow Jones 1,135 1.35 1.61 1.83

23 Knight Ridder 2,879 1.19 Thomson 1,000 1.19 1.42 1.42

2009 71


Jürgen Albinger

No. Parent Company

1997 Media Revenue % (mil)

Total Parent Company1986 Media Revenue %

Total HHI1997

HHI1986

24 New York Times 2,866 1.19 Thom EMI 959 1.14 1.41 1.30

25 Hearst 2,800 1.16 Viacom 932 1.11 1.34 1.23

26 Tribune 2,720 1.13 Westinghouse 839 1.00 1.27 1.00

27 Readers Digest 2,662 1.10 Harcourt Brace Jovanovich 800 0.95 1.21 0.91

28 Dow Jones 2,573 1.06 Thomson 756 0.90 1.13 0.81

29 Hollinger 2,538 1.05 Storer Communications 649 0.77 1,10 0.60

30 Dun & Bradstreet 2,154 0.89 Tele Communications 646 0.77 0.79 0.59

31 SBC Comm 2,110 0.87 Maclean Hunter 638 0.76 0.76 0.58

32 Cablevision Sys 1,949 0.81 Macmillan 611 0.73 0.65 0.53

33 BellSouth 1,934 0.80 Harte Hanks Comm 576 0.69 0.64 0.47

34 Washington Post 1,799 0.74 Disney 512 0.61 0.55 0.37

35 AOL 1,685 0.70 Affiliated Publications 401 0.48 0.49 0.23

36 Primedia 1,488 0.62 Amer Television & Comm 569 0.68 0.38 0.46

37 Sprint 1,454 0.62 A.H.Belo 399 0.48 0.38 0.23

38 Grupo Televisa 1,446 0.60 Houghton Mifflin 321 0.38 0.36 0.15

39 Harcourt General 1,376 0.60 Lorimar Telepictures 757 0.90 0.36 0.81

40 A.H. Belo 1,284 0.57 Media General 431 0.51 0.32 0.26

41 Hughes Electronics 1,277 0.53 Meredith Corporation 507 0.60 0.28 0.36

42 E.W. Scripps 1,246 0.53 MGM/UA 355 0.42 0.28 0.18

43 Ziff Davis 1,154 0.52 Multimedia 372 0.44 0.27 0.20

44 PrimeStar 1,097 0.48 Orion Pictures 328 0.39 0.23 0.15

45 Rogers Comm 958 0.45 Pulitzer Publishing 329 0.39 0.21 0.15

46 Media General 910 0.40 Southam 530 0.63 0.16 0.40

47 Torstar 894 0.38 Taft Broadcasting Co. 490 0.58 0.14 0.34

48 Meredith 830 0.37 Turner Broadcasting 507 0.60 0.14 0.36

49 Houghton Mifflin 797 0.34 Advo Systems 460 0.55 0.12 0.30

50 USA Networks 796 0.33 Berkshire Hathaway 400 0.48 0.11 0.23 Source: Compaine & Gomery, 2000, p. 561.



0

10

20

30

40

50

60

70

80

90

1997 1986Time Warner Disney Bertelsmann Viacom Sony News Corp TCI Thomson Seagram PolygramN.V. CBS GE(NBC) Reed Elsevier Gannett Reuters Cox Newhouse EMI Group MediaOne McGraw Hill Times Mirror Pearson Knight Ridder New York Times Hearst Tribune Readers Digest Dow Jones Hollinger Dun & Bradstreet SBC Comm Cablevision Sys BellSouth Washington Post AOL Primedia Sprint Grupo Televisa Harcourt General A.H. Belo Hughes Electronics E.W. Scripps Ziff Davis PrimeStar Rogers Comm Media General Torstar Meredith Houghton Mifflin USA Networks

Figure 7. Media Revenue of the Largest Media Companies, 1986 and 1997.Source: Based on data in Compaine & Gomery, 2000.

Table 7 above identifies the 50 largest media companies in 1986 and 1997 by the revenue from their media activities. In most cases, this is 100 percent of their revenue; in the case of a few companies, though, the parent company enjoyed much greater revenue. For example, NBC’s revenue in Table 8 below is shown to be about 6 percent of parent General Electric’s revenue (Compaine & Gomery, 2000).

Table 8. Concentration of Media Industry Revenue by Number of Companies, 1986 and 1997 % of

Industry Revenue 1997 % of Industry Revenue 1986.

% of Industry Revenue 1997

% of Industry Revenue 1986

Top 50 81.81 78.67Top 20 59.16 56.79Top 8 35.97 32.35

Top 4 24.13 18.79Source: Compaine & Gomery, 2000, p. 562.

0

10

20

30

40

50

60

70

80

90

1997 1986

Top 50 Top 20 Top 8 Top 4

Figure 8. Concentration of Media Industry Revenue by Number of Companies, 1986 and 1997 % of

Industry Revenue 1997 % of Industry Revenue 1986.Source: Based on data in Compaine & Gomery,

2000.

According to Compaine and Gomery, the use of 1986 as the base year for comparison is appropriate because it was the first year after the Federal Communications Commission

2009 73


Jürgen Albinger

relaxed the restrictions on the number of television stations under the ownership of a single firm from 7 to 12. “It was in that year that News Corp. 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 the core of network affiliates” (Compaine & Gomery, 2000, p. 562). During the closing years of the 20th century, the FCC’s restrictions on broadcast networks owning a financial interest in prime time programming were gradually removed and in early 1996, the Telecommunications Act in essence eliminated the size of broadcast radio groups and further eased restriction on television station group ownership (Compaine & Gomery, 2000).

An evaluation of media ownership as a single industry indicates that if the mergers that were pending in 1999 between Viacom and CBS and between Seagram and Polygram had been in effect in 1997 (assuming no divestitures of any lines of business of the participants), the top 50 would have been 82.13 percent, top 20, 61.80 percent, top 8, 40.37 percent and top 4, 27.02 percent and the HHI would be 295 instead of 268 (Compaine & Gomery, 2000).

The highlights of the HHI analysis of the media industry by Compaine and Gomery were as follows:

As measured by revenue, there was very 1. little change in media concentration between 1986 and 1997. In the former period, the top 50 accounted for about 79 percent of revenue. By the end of the period, it edged up to under 82 percent. The change in concentration among the top 20 and top 8 was similarly small.

Only at the top four level has there been a sign of greater concentration.

At the very top, the two largest companies 2. (CBS and Capital Cities/ABC) accounted for 10.5 percent of industry revenue in 1986. The top duo in 1997 (Time Warner and Disney, with most of Capital Cities/ABC) had 16.4 percent of industry revenue. This is the only economic measure by which the notion of increased concentration of ownership of the media had substantive backing. But it was prior to Disney selling off substantial parts of the newspaper and magazine properties that were part of the acquisition.

The HHI increased from an extremely 3. low 206 in 1986 to a still very low 268 in 1997. Therefore, while this measure did show some increased concentration, with HHI levels of under 1,000 indicating low concentration, the media industry remains one of the most competitive major industries in U.S. commerce.

There has been a substantial turnover 4. in the companies in the top 50 and even the top 12. CBS, the largest in 1986, was 11th 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.

Fully half the names in the 1997 list 5. 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; Seagram, which renamed MCA). Yet others were companies that are totally new to the media industry or did not exist (e.g., AOL, SBC Communications, Hughes Electronics/DirecTV).

Finally, the authors report that out of the 25 names from 1986 that were no longer in the top 50 in 1997, 15 were completely eliminated as the result of mergers and acquisitions. The other 10 simply did not grow fast enough to stay at the top as shown in Table 9 below.

Cournot Model.

According to Fellner (1949), “It was Augustin Cournot’s great achievement to have discovered the distinctive feature of the oligopoly problem. He also showed in 1838 that on certain assumptions a determinate equilibrium solution is obtained for the duopoly problem and that this solution can be extended from duopoly to oligopoly” (p. 56). In their book, A History of Game Theory, Dimant and Dimant (1996) report that Cournot pioneered the most widely used equilibrium concept in the course of analysing duopolies. According to these authors, “Recognition of Cournot’s eminence has become almost axiomatic to economists. One of the most familiar treatments of duopoly and the most-frequently employed equilibrium concept in game theory have been named after him: Cournot duopoly and Cournot—Nash equilibrium. Before Cournot (1838), writers on strategic interdependence such as Sun Tzu, Pascal and Waldegrave had formulated problems in the fields of war, ethics or card games. It was Cournot who first gave a rigorous

analysis of market structure, and he gave it from a game-theoretic perspective” (Dimant & Dimant, 1996, p. 18).

The Cournot solution is founded on the tenet that (in undifferentiated duopoly) each duopolist will believe that his competitor will go on producing a definite quantity irrespective of the quantity he produces. In this regard, Fellner (1949) suggests, “Obviously Table 9. Change in Firms on Largest 50 List, 1986

and 1997.

Top 50 Companies 1986 Merged/Acquired by 1997 .Capital Cities/ABC with Walt Disney Co.

Top 50 Companies 1997 Not in 1986 List Bertelsmann

Warner Communications with Time Inc.

Sony Pictures (formerly Columbia)

Gulf + Western with Viacom News Corporation

MCA with Seagrams Seagram

Westinghouse Broadcasting with CBS

Reed-Elsevier

Storer Broadcasting Reuters

MacLean Hunter Cox Enterprises

Macmillan, pieces sold to various

EMI

Affiliated Publications with New York Times Co

MediaOne

American Television & Comm

Pearson

Lorimar Telepictures with Time Inc.

Hollinger

Multimedia SBC Communications

Orion Pictures Cablevision Systems

Taft Broadcasting Bell South

Turner Broadcasting with Time Warner

America Online

Primedia

Sprint

Grupo Televisa

E.W. Scripps

Hughes Electronics (DirecTV)

Rogers Communications

USANetworks

Ziff-Davis

Torstar

Source: Compaine & Gomery, 2000, p. 563.

2009 75


Jürgen Albinger

in these circumstances each duopolist believes that he can calculate the quantity he should produce in order to maximize his profits. For, by deducting the fixed quantity of the rival’s output from the total quantity indicated by the market demand function for each price, he can obtain his own assumed individual demand function; and he may then proceed to equate his individual marginal revenue to his marginal cost. Yet this is not the reason why the solution is regarded as determinate. Neither duopolist is likely to make a correct guess of the quantity his rival will produce, because the rival produces the quantity which maximizes his profits on the same assumption with respect to his rival’s behavior, and hence does not generally keep his output constant. Consequently, the “position” (pair of outputs) obtained at first does not last; each firm will change its output” (p. 57).

“The characteristic feature of the Cournot model is that if each duopolist continues to assume that the other will not change his rate of output, then ultimately they will prove to be correct although during the approach to the equilibrium, for a limited period of time, they will be wrong. A produces a quantity which maximizes his profits on the assumption that B will go on producing his present output, whereupon B adjusts his output so as to maximize his profits on the assumption that A will go on producing his present output, which induces A to adjust his output, etc. What Cournot proved is that these adjustments ultimately result in an output for A which he can actually go on producing on the assumption that B will continue to produce his “present” output because B will ultimately produce an output which truly justifies A’s output on the assumption he makes. This statement must clearly be symmetrical; e.g., B must in this equilibrium position also be producing an output which is truly justified (that is, need not be further

adjusted) on the assumption he makes about A’s behavior (Fellner, 1949).

Clearly, this means that ultimately they are proven to be “right,” but for the wrong reasons. In this regard, Fellner points out that each competitor will assume that his rival follows a policy of fixed output while in reality each follows a policy of adjusting his own output to the requirement of profit maximization, on the assumption that the other follows a policy of fixed output; however, if, on this incorrect assumption, they both have actually adjusted their output to the simultaneous output of the other, then (from there on) the assumption they make with respect to one another is “quasi-correct” (Fellner, 1949, p. 57). Furthermore, it is reasonable to assume that the other producer will continue producing a fixed output, although the reason is not (as is mutually assumed) that he follows a policy of producing a fixed output disregarding his rival’s behavior. “This is what we meant by saying that they are right for the wrong reason” (Fellner, 1949, p. 57). The geometrical proof provided by Fellner is as follows:

Measure along the abscissa the output of A, and along the ordinate that of B. Label the curve indicating A’s output as a function of B’s output F 1, and the curve expressing B’s output as a function of A’s output F 2. The shape of these reaction functions is given by the postulate that each duopolist maximizes his individual profits on the assumption that the other will go on producing his “present” output. Let us first assume that the market demand function is linear (which Cournot did not assume) and that the producers have constant costs at the identical level (which Cournot did assume in the first exposition of his theory) (Fellner, 1949, p. 57).

The main proposition is established by



the fact that the two curves intersect and that the intersection marks stable equilibrium. In other words, to the right of the intersection, A’s output tends to fall and B’s to rise, while to the left of the intersection A’s output tends to rise and B’s to fall, so that the equilibrium becomes restored in the event of disturbances. This must be so because F 1 must intersect with the ordinate above F2, while it must intersect with the abscissa to the left of F 2, For point 1 shows that output of B which would induce A to produce nothing, while point 2 shows the output which B produces if A produces nothing. The first of these two quantities is the competitive output for the industry in question. 4 The second quantity is B’s monopoly output. The competitive output exceeds the monopoly output. For the analogous reason A’s output in point 3 must be smaller than in point 4. Point 4 marks the competitive output and point 3 marks A’s monopoly output (Fellner, 1949).

The reaction functions (F 1 and F 2) are linear if it is assumed linear demand functions and constant costs. Consequently, the stable intersection point just described is the only intersection point. Furthermore, given identical cost functions in undifferentiated duopoly, the individual outputs of the duopolists will be the same. It also is straightforward to demonstrate that, based on these assumptions, the aggregate duopoly output is two-thirds of the competitive output, while the monopoly output would be one-half of output under pure competition (Fellner, 1949).

In his assessment of the economics of monopoly and equilibrium analysis, Schumpeter (1954, cited in Dimant & Dimant, p. 18) suggested that:

The chief performance was Cournot’s and the period’s work may be described as a series of successful attempts to develop his statics of straight monopoly and as another series of much less successful

attempts to develop and to correct his theories of oligopoly and bilateral monopoly. Second honors are divided between Marshall and Edgeworth. (p. 976).

According to Sarkar, Gupta, and Pal (1998), “The first and still one of the most widely cited models of noncooperative oligopoly behavior is the Cournot model, developed by the French mathematician Augustin Cournot in 1838. The Cournot model is the fundamental model used to study strategic interactions among quantity-setting firms in an imperfectly competitive market” (p. 118). In recent years, there has been a renewed interest in Cournot-based models of strategic behavior that can be used to analyze various real-world phenomena ranging from horizontal mergers to intra-industry trade. In this regard, Sarker and his colleagues advise, “A proper understanding of the Cournot model of imperfect competition and strategic interactions among firms in various contexts is thus essential” (p. 118).

In his analysis, “Structural Screens in Stochastic Markets,” Shaffer (1996) reports that if companies manage to negotiate multi-period contracts with their suppliers, then differences in the timing of contracts may lead to variations in input prices across firms, even within a single input market; these effects are not explicitly represented in the Cournot model, but the author emphasizes that these effects may underlie some of the assumed interfirm asymmetries in real-world settings (Shaffer, 1996). According to Boumans (2005), a number of economists have maintained that models function as instruments of investigation that can be used to learn about the world as well as the theories involved because they both involve some type of representation. “They either represent an aspect of the world or an aspect of theories about the world, or both,” he advises (Boumans, 2005, p. 74).

2009 77


Jürgen Albinger

In order to gauge an instrument’s accuracy, it is important to test it to determine the association, if any, between the representation itself and the aspect of the world that is to be represented. In this regard, Boumans notes that, “The accuracy of this correspondence depends on the complexity of the system under investigation and our ability to construct representations of it. This is, in principle, a technical problem, termed by Kevin Hoover as the “Cournot problem” (after the man who formulated it explicitly, Antoine Augustin Cournot, 1801-77) (2005).

According to Cournot:The economic system is a whole of which all the parts are connected and react on each other. … It seems, therefore, as if, for a complete and rigorous solution of the problems relative to some parts of the economic system, it were indispensable to take the entire system into consideration. But this would surpass the powers of mathematical analysis and of our practical methods of calculations, even if the values of all the constants could be assigned to them numerically. (Cournot, 1971, p. 127, cited in Boumans, 2005)

In their essay, “Profitable Mergers in a Cournot Model of Spatial Competition,” Norman and Pepall (2000) report that whenever two companies merge, they act as Cournot competitors against other merged firms but as Stackelberg leaders with respect to the remaining nonmerged firms. With linear demand and costs, any two-firm merger is profitable. We adopt a less extreme approach to commitment by introducing the assumption that our Cournot firms compete across a set of spatially differentiated markets. In this setting, firms choose not just the quantity of o utput to supply to the spatially differentiated markets but also where to locate their production

plants to serve these markets (Norman & Pepall, 2000).

Location is a key factor underlying why a merger can lead to a bigger and better firm. Firms that have different locations have different locational advantages in serving the set of spatially separated markets. As a result, in contrast to the standard nonspatial Cournot model, a merger between two firms need not result in one of them effectively being shut down. Rather, a merger between two firms allows them to coordinate their location decisions with the result that the merged firm becomes potentially bigger than its rivals, better adjusted to consumer locations (Norman & Pepall, 2000).

The results of the investigation by Norman and Pepall (2000) show that a two-firm merger results in the merged firm relocating its plants away from the market center, whereas the nonmerged firms remain at the center. “Being a leader or Stackelberg first mover in location choice and coordinating the two plants’ locations to serve certain segments of the market more efficiently lead to a merged firm with a larger overall market share than its nonmerged rivals,” they advise. “This explains why, in sharp contrast to the standard Cournot model, a two-firm merger can be profitable even in relatively unconcentrated markets” (p. 667).

Not only does a two-firm merger improve locational efficiency, it also increases market concentration and reduces competitive pressure across the set of markets (Norman & Pepall, 2000). The bottom-line for companies is that it generally leads to higher prices across the set of consumer markets, in sharp contrast to nonspatial analysis, though, the increased profit of both the merged firm and the nonmerged firms, together with improved locational efficiency, is such that the merger increases total surplus. “In other words,” the authors conclude, “the introduction of commitment



through location leadership means that merger can indeed lead to a bigger firm, which so far as total welfare is concerned serves the market ‘better’” (Norman & Pepall, 2000, p. 667). In this regard, Herk (1993) reports that allowing firms to set prices individually once capacities are fixed provides a capacity-constrained model that offers more complete and plausible description of duopoly competition than the widely used Cournot model, which in effect assumes that firms always select the market-clearing price corresponding to their aggregate capacity.

An example of the oligopoly Cournot model is shown below:

Ex: The demand function is given by: p = 105 - 5 (q1 + q2) where Q= q1 + q2 and C(q) = 5q (both firms face the same cost function.)

max p = (105 - 5 (q1 + q2)) q1 - 5q1

d p/dq1 = 105 - 10q1 - 5q2 - 5q1 = 0

Simplifying = q1 = 10 - ½ q2

The first company’s decision to produce is a function of the second firm’s decision. This formula is known as a reaction function. Once firm one has an idea of what firm 2 will produce, it can decide on its own production. So the next question is, are the firms likely to be able to second guess each other’s decisions. If so, will the solution be a Nash equilibrium?

This equation can be plotted with q2 and q1 on the X and Y axes respectively; it can also be solved for the second firm’s reaction function, which turns out to be identical to the first (except q1 and q2 are reversed), since the cost functions for both firms are identical.

Therefore, firm 2’s reaction function will be q2 = 10 - ½ q1

This can also be graphed and an equilibrium point identified; this equilibrium point is the solution to the Cournot model.

If the second equation is substituted into the first, it is shown that:

10 - ½ (10 - ½ q1) = q1

10-5 + 1/4 q1 = q1

5 = 3/4 q1

q1 = 20/3

It will also be the case that q2 = 20/3

p = 105 - 5(40/3) = 105 - 200/3 = 115/3

“So how do the monopoly, perfect competition and Cournot duopoly fare in terms of p and q outcomes?” (Olmsted, 2002, p. 3). The answer provided by this author is shown in Table 8 below:

Table 10. Sample Cournot Model Analysis.

Monopoly: p = 105 - 5q(105 -5q) q - 5q105 -10q -5 =0100 = 10qq=10p = 105 -50p=55PCp = 55 = 105 - 5q100 = 5qq=20

Source: Olmstead, 2002.

2009 79


Jürgen Albinger

It can be seen from the above that the Cournot duopoly is between PC and the monopoly in terms of pricing and quantity supplied to the market (Olmsted, 2002).

There are some fundamental constraints to the Cournot model, particularly as it applies to the telecommunications industry, however. For example, in his book, Competition, Commitment and Welfare, Kotaro Suzumura (1995) reports that, “It goes without saying that Cournot conjecture, whereby an oligopolistic competitor expects that rival firms will not adjust their strategies in response to a change in its own strategies, is a great simplifier of the analysis of oligopolistic competition and economic welfare. It is clear that our analyses in this monograph have extensively utilized the services rendered by this simplification. Unfortunately, however, Cournot conjecture is self-defeating out of equilibrium, as it just implies that firms keep on expecting that other firms will not respond when they are in fact changing their strategic variables throughout the process of adjustment towards equilibrium. We should also add that the assumption of Cournot conjecture prevents us from examining the welfare effects of collusion among oligopolistic firms” (p. 160).

Four-Firm Concentration Ratio (C4).

This model seeks to identify and understand the combined share in released length of the four largest companies in a given industry (Pokorny & Sedgwick, 2004). In order to capture the interdependence of firms, measures of industries’ competitiveness as represented by each industry’s four-firm concentration ratio (C4) and industry-level estimates of the price elasticity of demand (PELAS) can be made (Cohen & Levinthal, 1990).

By conventional economic measures, the newspaper industry would seem quite competitive. For example, Table 11 below shows that in 1947 the four and eight firm concentration ratio for newspapers was 21 percent and 26 percent, respectively. The four largest newspaper publishing companies in 1992 accounted for 25 percent of industry shipments. The eight largest accounted for 37 percent of shipment value. These ratios are similar to the book and magazine publishing sectors of the publishing industry. This compares, just to choose several unrelated industries for context, with 63 percent and percent, respectively, for the soap and other detergent industry (SIC 2841); 70 percent and 77 percent, respectively, for commercial printing, gravure; and 34% and 49%, respectively, for bread, cake and related product manufacturing. Compared to the median concentration ratios for all manufacturing industries, newspaper publishing, in 1992, was considerably less concentrated at all size levels (Compaine & Gomery, 2000).

Table 11. Share of Total Dollar Shipments by Largest Firms in Publishing Industries, Selected

Years, 1947-1992 Newspapers (SIC 2711)

Periodicals (SIC 2721)

Book Publishing (SIC 2731)

Median All Manufac-turers

19474 largest 21% 34% 18% NA

8 largest 26 43 28

50 largest NA NA NA

19674 largest 16 24 20 NA

8 largest 25 37 32

50 largest 56 72 77

19924 largest 25 20 23 37

8 largest 37 31 38 52

50 largest 70 62 77 87 NA: Not availableSource: Compaine & Gomery, 2000, p. 6.



Industry consolidation and a trend toward market oligopoly have been the evolutionary paths of a various industries in the U.S.; some examples from various sectors of the economy include the airline industry, the wine and beer industries, the petroleum production industry, the motion-picture distribution industry, the microcomputer industry, and the steel industry, among numerous others (Caroll, Dobrev & Kim, 2002).

According to Carroll and his colleagues (2002), “Concentration propels the formation of a strong and visible market structure whose effects inevitably reverberate through all levels of the social system, influencing the behavior of both economic and non-economic actors” (p. 233). For example, the authors cite the instance of studies by industrial organization economists that found that a broad spectrum of a firm’s activities and processes are directly affected by the rising industrial concentration: incentives to innovate, mechanisms for price setting, the expectation for investment returns and stability, budgeting for advertising expenses, and the distribution of wages all seem to hinge on the rising market power of a few dominant producers and the relationships that develops among them; however, to date, there has been a dearth of research concerning the social dimension of market structure (Carroll et al., 2002).

In many industries in which scale provides an advantage, the gradual rise to dominance of a few large competitors is accompanied by a horizontal expansion of their market positions or niches. This was what happened with Daimler-Chrysler’s merger in the auto industry, WarnerBros.-Lorimar’s merger in the motion-picture distribution industry, LTV-Republic merger’s in the steel industry, and American-TWA’s merger in the airline industry (Carroll et al., 2002). The authors offer somewhat disingenuously

that, “Accordingly, it seems important that analyses of organizational evolution in concentrating industries deal with questions of niche and scale simultaneously and how they interrelate. The answer is far from obvious and might be complex” (emphasis added) (Carroll et al., 2002, p. 234).

Variations in the industry size distribution that may give rise to the emergence of structure and market partitioning are captured by our measure of market concentration; in this regard, Carroll and his colleagues depended on the frequently used concentration-ratio measure, defined as the ratio of the annual production of the four largest firms to the total industry output for that year (C4). For their analysis, this measure has an advantage over other concentration measures because it captures well the process of consolidation in the market center, where only a few firms are able to dominate, which has been the case in the U.S. automobile industry. The measure implies that as the combined market share of the top four industry firms increases, concentration increases proportionately. It also implies that if the market in any given year consists of four or fewer producers, concentration will equal one, even if production is evenly divided among them. This is not a concern in their analysis, however, because they used two other size-based measures to account for positional and scale differences among individual firms: relative size and scale competition (described further below) (Carroll et al., 2002).

Organizational ecologists have cited a pattern of density evolution that is relevant across a wide range of industries, including the U.S. automobile industry; in these cases, the number of organizations initially grows slowly until the form acquires taken-for-grantedness, causing density after that point to increase steeply (Carroll et al., 2002).

2009 81


Jürgen Albinger

According to these authors, as the carrying capacity of the population is reached and competition increases, the number of companies involved will tend to decrease, reflecting the consolidation of the market. The subsequent entry of specialists leads to a modest resurgence of density in scale-dominated industries. One important implication of this density pattern for interpreting variation in concentration is that the slow increase in the number of firms in the early years of the industry confines our measure of concentration to equal 1 until at least five firms are present on the market. Because their argument is based on the concept that rising concentration solidifies industry structure, and because high mark et concentration in the very early years of the industry clearly does not signify this theorized process, we estimated models in which we controlled for the effect of high concentration during that initial period (Carroll et al., 2002).

Niche overlap density (NO) represents the total number of companies involved whose niches overlap with the niche of the focal firm; while there are different ways to compute niche overlap, we chose to use a density measure because it is a better fit with our hypotheses. A more complex alternative cited by the authors would require weighing the overlaps based on the portion of the niche that overlaps with that of competitors (Carroll et al., 2002).

Next the market center addresses the range of the niches of the four largest firms in the industry; this measure of the market center is based on the rationale of the concentration measure described above. Should the four largest firms in the industry provide a revealing example of the level of concentration, then the range of their niches should provide an adequate description of the most resource-abundant segment in the

market, where the dominant players position themselves (Carroll et al., 2002).

The authors report that they used the midpoint of a firm’s technological product range to indicate its niche position in the market. The midpoint could be used to mark position (or location) whenever niche width on a focal dimension can be represented as a continuous variable and the distribution of a firm’s capabilities across the range of the niche is symmetrical. Distance away from the market center is the difference between the midpoint of the focal firm’s niche and the midpoint of the market center. Finally, the authors estimated the effects of the distances of firms both “above” the market center (position: DAMC), meaning a niche width that contains a larger engine capacity than the center, and “below” the market center (position: DBMC) (Carroll et al., 2002).

The assumptions used by Carroll and his colleagues differ significantly from those of the original theory of niche width, in which other authorities assumed that organizations face only one environmental state at a point in time and that the alternation of states over time imposes contradictory demands on the organization; in this case, the various states of the environment may impose complementary demands (Carroll et al., 2002).

Summary, Conclusions and Recommendations

Summary

The purpose of this study was to determine the efficacy of the Herfindahl-Hirschmann Index in identifying a monopoly within the telecommunications industry based on an analysis of the relevant statistical from before and after the enactment of the Telecommunications Act. The research showed that the economic models reviewed



were useful for certain aspects of analyzing the telecommunications industry. In this regard, the Herfindahl-Hirschmann Index (HHI) was shown to be useful in identifying market concentration among firms on the Largest 50 List before the Telecommunications Act for 1986 and subsequently in 1997. Likewise, the trend of fewer owners generally earning a larger percentage of market revenue can be identified by analyzing the revenue share of the top four owners in the market using the four-firm concentration ratio (Compaine & Gomery, 2000). Nevertheless, the research also showed that there are a wide range of confounding variables that must be taken into account in these models. Furthermore, while these economic models are useful in determining whether a high concentration actually exists in a given industry, they remain applicable only in certain situations.

Conclusions

This study set out to identify a reliable economic model that could be used to identify and predict monopolies in the telecommunications industry. This research determined that the Herfindahl-Hirschmann Index economic model can be used to accurately predict the existence of a monopolizing condition in the telecommunications sector, assuming that the data sets employed and findings derived therefrom are viewed in the appropriate context. For example, Jorde and Teece point out that, “It is difficult to hypothesize and propose alternative HHIs for technologically dynamic markets. However, the inclusion of performance competition and the extension of the time frame of competitive response may mean that it is not necessary to change these critical HHIs. Furthermore, we believe that with technologically dynamic markets, the dynamics of market structure in the past should provide some guidance to

assessing market definition and predicting likely changes in market concentration. Key factors are the change in concentration and the trend in the number of competitors” (p. 11). In reality, economists have been using economic criteria as the only or at least the primary criteria for determining concentration in all industries for some time now. While the congressional debate over the Celler-Kefauver Amendment to Section 7 of the Clayton Act indicated that social and political criteria should be taken into account in antitrust proceedings, the amendment itself was devoid of any such language; consequently, assessments of efficiency and entry to an industry are measured by such criteria as the percentage of industry revenue accounted for by the largest players, as well as by the more complex Herfindahl-Hirschmann Index (Compaine & Gomery, 2000). Based on the foregoing considerations, the hypothesis was confirmed that the Herfindahl-Hirschmann Index represents an accurate measure of high market concentration for virtually any industry where the index is 1,800 or greater. In reality, though, the nature of the telecommunications industry today makes such blanket assessments problematic but the HH1 remains the most effective tools for this analysis.

Recommendations

A number of economists and researchers have become increasingly concerned about the influence mergers and acquisitions had on specific industries. Taken together, the critical review of the literature and the results of the data analysis suggest that if a degree of significance is placed on the diversity of program choice, then it is important to consider the number of firms within the market and their relation to each other. These are the basic dimensions of

2009 83


Jürgen Albinger

market structure, which are proxied by the number of firms and their size distribution and these, in turn, are reflected by standard measures of concentration such as the Herfindahl-Hirschmann index. Although some experts have maintained that a high number of companies in a given industry may result in some program duplication, the reality of the marketplace is that if pluralism is to involve more than simply the diversity of programs but something that embodies the nature and quality of what is broadcast, then the production and ownership pattern may be very important (Young, 2000).

By any measure, though, the ownership link between different firms does not necessarily imply a similar approach to program types but clearly suggests that greater scrutiny of the plurality of the types of programs offered is required. For standard economic analysis it may not matter because the concern is whether the market structure produces the socially optimal program variety but, as we have argued, ownership does matter if we are concerned with a diversity of source and a diversity of opinions as embodied in the ways in which programs are presented (Young, 2000).

References

Abbott, A., and Brady, G. (1997) Dezoning Spectrum: Opportunities For Rent Seeking. Journal of Private Enterprise, Fall 1997, 13 (1).

Adams, W.J., and Yellen, J.L., (1976) Commodity Bundling and the Burden of Monopoly. Quarterly Journal of Economics 90 (3): 475 498.

APUC Annual Report (1997) H t t p : / / W w w. S t a t e . A k . U s / A p u c /Annualreports/97annl_Rpt/13.Htm Accessed January, 2002.

Armentano, D.T. (1990) Antitrust and Monopoly: Anatomy Of A Public Policy Failure. New York: Holmes & Meier.

Baumol, W.J., J. Panzar and R.D. Willig, (1986), “On The Theory of Perfectly Contestable Markets,” In New Developments In The Analysis of Market Structure. Edited By J.E. Stiglitz and G.F. Matthewson, Cambridge, Mass.,

Benham, H. Union-Nonunion Wage Differentials Revisited Journal of Labor Research. Fall 1987. 8(4): 369-383.

Black’s Law Dictionary. (1990). St. Paul, MN: West Publishing Co.

Bolter, W. G. (1990) Telecommunications Policy For The 1990s and Beyond, M.E. Sharpe: Armony.

Boumans, M. (2005). How economists model the world into numbers. New York: Routledge.

Brennan, T., (2002) Preventing Monopoly Or Discouraging Competition? The Perils of Price-Cost Tests for Market Power In Electricity. Resources For The Future. Discussion Paper 02 50 October 2002.

Buck, S. (2003). TELRIC vs. universal service: A takings violation? Federal Communications Law Journal, 56(1), 1.

Caminal, T. Price Advertising And Coupons In A Monopoly Mode. Journal of Industrial Economics. March 1996. 44 (1): 33-52.

The Evolution of Organizational Niches: U.S. Automobile Manufacturers, 1885-1981. Contributors: Glenn R. Carroll - author, Stanislav D. Dobrev - author,



Tai-Young Kim - author. Journal Title: Administrative Science Quarterly. Volume: 47. Issue: 2. Publication Year: 2002. Page Number: 233

Carpano, C., Chrisman, J. J., & Roth, K. (1994). International strategy and environment: An assessment of the performance relationship. Journal of International Business Studies, 25(3), 639.

Chen, B., & Shimonura, K. (1998) Self-Fulfilling Expectations And Economic Growth: A Model of Technology Adoption and Industrialization. International Economic Review. February 1998.

Cherry, B., & Wildman, S. (1995) A Framework For Managing Telecommunications Deregulation While Meeting Universal Service Goals. Presented at Universal Service in Context: A Multidisciplinary Perspective. New York, New York, December 6, 1995

Choi, S., Stahl, D., & Whinston, A. (1997) A Commentary On Is Microsoft A Monopolist? Recife, July 29th, 1998. Http: / /Www.Beje.Decon.Ufpe.Br/Taylor/Monopolist.Htm Accessed January, 2003.

Cohen, M (2002). “Morningstar Introduces New Industry Sector Structure, Divides Economy Into Three “Super Sectors.” Morningstar. Jun 25, 2002. Http://Www.Morningstar.Com/Press/Release/Item/0,1045,661,00.Html Accessed January, 2003.

Cohen, W. M., & Levinthal, D. A. (1990). Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 35(1), 128.

Compaine, B. M., & Gomery, D. (2000). Who owns the media? Competition and concentration in the mass media industry. Mahwah, NJ: Lawrence Erlbaum Associates.

“Competition’s Final Hurdle?” (2002) Journal of Competition In Telecommunications. October/November 2002 1(003).

Compete Against Each Other? That Would Be “Fundamentally Wrong... It Might Be A Way To Turn A Quick Dollar,” Says One Of The Monopolists, “But That Doesn’t Make It Right.” (2002, November/December). Journal of Competition in Telecommunications. 1(004).

Cooke, P. (2001). Knowledge economies: Clusters, learning and co-operative advantage. New York: Routledge.

Cournot, A. A. [1838] (1971). Researches into the Mathematical Principles of the Theory of Wealth, trans. N.T. Bacon, New York: Augustus M. Kelley.

Crandall, R.W. (1991) After The Breakup: U.S. Telecommunications In A More Competitive Era. Washington, D.C.: The Brookings Institution.

Davidson, C. (1999, September 27). Corporate brain transplant. New Statesman, 4456(128), 37.

Davidson, C. and R. Deneckere, (1986), Long-Run Competition In Capacity,Short-Run Competition In Price, and the Cournot Model. The Rand Journal of Economics,17, 404-415.

Dei, F. (1997, July). Industrialization And Expectations In A Small Open Economy. Japanese Economic Review.

2009 85


Jürgen Albinger

Demsetz, H. (1968, April). Why Regulate Utilities? Journal of Law And Economics 11, 55-65.

Dimand, M. A., & Dimand, R. W. (1996). A History of Game Theory. New York: Routledge.

Dixit, A., (1979). A Model of Duopoly Suggesting A Theory of Entry Barriers. Bell Journal of Economics 10 (1), 20 32.

Drexler, M. (1999). NTT Head Talks About Life After Monopoly. IDG News Service\Tokyo Bureau. H t t p : / / W w w. E - C o m m e r c e w o r l d .C o m / I d g n s / 1 9 9 9 / 0 9 / 1 6 /Nt theadta lksabout l i feaf te r.Shtml Accessed January, 2003.

Epstein, G. and Nitzan, S. (2002) Political Culture and Monopoly Price Determination. Center For Economic Studies and IFO Institute For Economic Research. Working Paper Series No. 646. January 2002.

Farrell, C. (1995). “Media Control Is Narrowing; Should We Worry? (Competition and Monopolies in Mass Media.” Business Week, August 14, 1995:37.

FCC. (2003) Connecting The Globe. Http://Www.Fcc.Gov/Connectglobe/Sec5.Html Accessed January, 2003.

Fellner, W. (1949). Competition among the few: Oligopoly and similar market structures. New York: Alfred A. Knopf.

Gabszewicz, J.J., & Thisse, J. F. (1986) On The Nature Of Competition With Differentiated Products, The Economic Journal,96: 160-172.

Gabszewicz, J.J. And J.F. Thisse, (1979) Price Competition, Quality and In-Come Disparities. Journal Of Economic Theory, 20:340-359.

Gabriel, S. (2002) Telecommunications Sector, Bruised And Battered, Looks To A Brighter Future. FDIC. Http://Www.Fdic.Gov/Bank/Analytical/Fyi/062002fyi.Html Accessed January, 2003.

Gasmi, F., Laffont., J., Sharkey, W., (1999) No Title. Toulouse, Institute Economie Industrielle. Http://Ideas.Repec.Org/P/Fth/Touein/91.Html.

---. A technico-economic methodology for the analysis of local telephone markets. In MacKie-Mason, J. K., & Waterman, D. (Eds.) (1998). Telephony, the internet, and the media: Selected papers from the 1997 telecommunications policy research conference. Mahwah, NJ: Lawrence Erlbaum Associates.

Gilligan, T. W. (2002). Industrial Concentration. The Concise Encyclopedia of Economics. Http://www.econlib.org/library/Enc/IndustrialConcentration.html. Accessed March, 2006.

Golan, A., Judge, G., Perloff, J. (1996) Estimating The Size Distribution Of Firms Using Government Summary Statistics. Journal of Industrial Economics. March 1996. 44(1): 69-80.

Goldstein, T. (1993) Playing Monopoly With The Media. New York Times Dec. 10, 1993):A19 Npa35(L), Col 3, 18 Col In.

Graham, D., Kaplan, D., And Sihley, D. (1983) Efficiency and Competition In The Airline Industry. Bell Journal of



Economics. Spring 1983 14 (1): 118-138. Http://Ideas.Repec.Org/A/Rje/Bellje/V14y1983ispringp118-138.Html Accessed January, 2003.

Greco, A. N. (1999). 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), 166.

Greenhut, M.L., G. Norman and C.-S. Hung, (1987) The Economics Of Imperfect Competition. A Spatial Approach, Cambridge, Cambridge University Press, Chapters 2, 3, 5.

Gruenwald, J. (1996). Beleaguered ‘96 Overhaul Hits Another Snag. Congressional Quarterly Weekly Report Feb 21, 1998:438.

Gruley, B., Simons, J., Wilke, J. (1998)Is This Really What Congress Had In Mind With The Telecom Act? Wall Street Journal. May 12, 1998):A1(W), A1(E), Col 6, 48 Col In.

Hakken, D. (2003). The knowledge landscapes of cyberspace. New York: Routledge.

Hammond, A. (1997) The Telecommunications Act Of 1996: Codifying The Digital Divide. Federal Communications Law Journal Dec, 1997:179-214.

Harper, J. (1997) Monopoly and Competition in British Telecommunications, Pinter, A Cassell Imprint, London.

Hendricks, J. A. (1999). The Telecommunications Act of 1996: Its impact on the electronic media of the

21st century. Communications and the Law, 21(2), 39.

Herk, L. F. (1993). Consumer choice and Cournot behavior in capacity-constrained duopoly competition. Rand Journal of Economics, 24(3), 399.

Hoover’s Online. Telecommunications Sector. Description. Http://W w w . H o o v e r s . C o m / S e c t o r /Description/0,2176,27,00.Html Accessed January, 2003.

Huber, P., Kellogg, M., Thorne, J. (1999) Special Report: The Telecommunications Act of 1996. Http://Www.Phuber.Com/Huber/Legis/Intro.Html Accessed January, 2003.

Hughes, M. (1996, February). Telecommunications Reform and the death of the local exchange monopoly. Law Review. Http://216.239.33.100/Search?Q=Cache:85-Qfl5kos0c:Www.Law.Fsu.Edu/Journals/Lawreview/Downloads/241/Hughes.Pdf+Telecommunications+Monopoly&Hl=En&Ie=UTF-8 Accessed January, 2003.

Huisman, M. (2002) Let’s Play Monopoly. Ht tp : / /Www.Aivf .Org/Advocacy/Communications_Policy/Monopoly.Html Accessed January, 2003.

IDA. (2002) Full Competition In Telecommunications Sector. Http://Www.Ida.Gov.Sg/Website/Idacontent.Nsf/Vsubcat/Policy+&+Regulationfull+Competition+In+Telecommunications+Sector?Opendocument Accessed January, 2003.

2009 87


Jürgen Albinger

Jorde, T. M., & Teece, D. J. (1992). Antitrust, innovation and competitiveness. New York: Oxford University Press.

Kaysen, C., & Turner, D. F. (1959). Antitrust policy: An economic and legal analysis. Cambridge, MA: Harvard University Press.

Khan, M. (2000) CPP-Interconnect Issue: The Case of Pakistan Paper For ITU-BDT Presentation. Symposium For Regulators November 20-22, 2000 Geneva Switzerland.

Kanell, M.(1998). Telecommunications ‘Convergence’ Returns. The Atlanta Journal and Constitution, June 28, 1998; BUSINESS; Pg. 01R

Kehoe, M. (1996) Quality Uncertainty and Price in Monopoly Markets. Journal of Industrial Economics. March 1996. 44 (1): 25-32.

Kemp, M., Okawa, M., Shimonura, K. (1996) Voluntary Export Restraints And Economic Welfare: A General Equilibrium Analysis. Japanese Economic Review. November 1996.

Kemp, M., And Shimonura, K. (1997) Increasing Returns And International Trade. Review of International Economics. July 1997.

Kitman, M. (1998) Phone and Cable Sharks Merge In A Feeding Frenzy. Newsday, June 29, 1998. PART II; Page B23 June 29, 1998

Kellogg (1992) Monopoly Price Dispersion Under Demand Uncertainty. International Economic Review.

Kellogg (2001) Entry Deterrence In A Duopoly Market. Unpublished Working Paper. April 2001

Kenyon, L., Craig, L., Treuer, P. (1995). Communications Competition and Deregualtion Act. Senate Hearings. January, 9, 1995. Http://Leahy.Senate.G o v / P r e s s / 1 9 9 5 0 6 / 9 5 0 6 0 9 . H t m l Accessed January, 2002.

Kreutzer, D., And Wood, W. (1997) Private Monopoly Vs. Public Shortsightedness: Private The Natural Monopoly Rationale Revisited. Journal of Private Enterprise, Fall 1997, 13 (1).

Kurtin, O. (1998) Everything That Communicates Must Converge; The Merging Of Technology and Services and The Growth of The ‘Net Raise Thorny Legal Issues. National Law Journal V20, N34. April 20, 1998:B12, Col 1, 59 Col In.

Lacy, S. The Financial Commitment Approach To News Media Competition. Journal of Media Economics 5:(2):5-22 (1992).

Lacy, Stephen (2000). Commitment of Financial Resources As A Measure Of Quality. In Robert G. Picard (Ed.), Measuring Media Content, Quality, and Diversity: Approaches and Issues In Content Research. Turku, Finland: Business Research and Development Centre, Turku School of Economics and Business Administration.

Leland, H. E., & Meyer, R. A. (1976). Monopoly Pricing Structures With Imperfect Discrimination, Bell Journal of Economics, 7: 449-462.



Lieberrskand, J., Opler, T., And Hatfield, D. (1996) Corporate Restructuring And The Consolidation of US Industry. Journal of Industrial Economics. March 1996. 44 (1): 53-68.

Lima, T. (2002, November). The Economics Of The Microsoft Antitrust Case. Windowwatch. Http://Www.Windowatch.Com/2002/November/Lima8_10.Html Accessed January, 2003.

Mclaughlin, L. (1998) “Telecom Reform.” PC World 16(2) (Feb, 1998):58 –63.

Malik, O. (1999, June 24) “Is Sun Ringing Up A Telecom Monopoly?” Forbes. Http://Www.Forbes.Com/1999/06/24/Feat.Html Accessed January, 2003.

Martin, S. (1999) Strategic and Welfare Implications of Bundling. Martin Centre For Industrial Economics, Institute of Economics, University of Copenhagen, Economics Letters 62: 371 376.

Mcmaster, S. E. (2002). The telecommunications industry. Westport, CT: Greenwood Press.

Miller, E. S. (1996). Economic regulation and new technology in the telecommunications i n d u s t r y . Journal of Economic Issues, 30(3), 719.

Mills, M. (1998)”Making The Connection; AT&T Revives Debate Over One Wire For All Communications Needs.” The Washington Post, June 27, 1998, FINANCIAL; Pg. E01

Mills, M.(1998a)” The Telecom Big Top; Firms Old And New Jockey For Position On Fiber-Optic High Wire.”

The Washington Post, May 24, 1998, FINANCIAL; Pg. H01

“Monopoly? Not Me. He’s The Monopoly.” (2002) Journal of Competition In Telecommunications. October/November 2002 1(002)

Mulligan, C. (2000) “Can Monopoly Unionism Explain Publicly Induced Retirement?” National Bureau Of Economic Research, Inc. April, 2000.

Musa M. And S. Rosen, (1978) “Monopoly And Product Quality,” Journal of Economic Theory 18: 301-317

Negrin, J. (2002) Access Pricing Determination In A Vertically Integrated Network Industry Using A Simulations Model . Rice University, Houston, Texas. Unpublished Dissertation.

Nelson, J.R. (1966) “The Role Of Competition In The Regulated Industries.” The Antitrust Bulletin XI: 1-36. Jan.-Apr.

Norman, G, (1981) Spatial Competition and Spatial Price Discrimination. Review of Economic Studies. 48, pp. 77-90.

Norman, G., & Pepall, L. (2000). Profitable mergers in a Cournot model of spatial competition. Southern Economic Journal, 66(3), 667.

Novskek, W., (1980) Equilibrium In Simple Spatial (Or Differentiated Products) Models. Journal of Economic Theory. 22, 313-326.

Olmstead, J. (2002). Oligopoly. Http://users.drew.edu/jolmsted/io/iochap7.htm Accessed March 23, 2006.

2009 89


Jürgen Albinger

Oxohy, R. (2001) Monopoly Classroom Experiment. Journal of Economic Education. Spring 2001: 160-169.

Park, V. (1997) The Stability Of Networks:An Evolutionary Approach To Standardization. EURAS Yearbook Of Standardization 1: 347-356,

Pehal, M. (2000, April). Regulation and Strategy for Telecommunications, Information and Media. Journal of Policy, 2(1), 2-9.

Poulson, B.W. (1997) Deregulating Telecommunications In Colorado. Independence Institute. Press Release June 11, 1997. Http://I2i.Org/Publications/Op-Eds/Other/Op061297.Htm Accessed January, 2003.

Pociask, S., And Rutner, J.(2000) MCI Worldcom’s Sprint Toward Monopoly. Published By The Economic Policy Institute.

Pokorny, M., & Sedgwick, J. (2004). An economic history of film. New York: Routledge.

Porter, M. E. (1980). Competitive strategy: Techniques for analyzing industries and competitors. New York: Free Press.

Powers, Smith, Rutkowski, Lukasik, Kozel, and Khan (2000) The Evolution of Telecommunications Infrastructure over The Next Decade. CCIP Workshop. Http://216.239.51.100/Search?Q=Cache:Cw_H494-Zfec:Www.I e e e u s a . O rg / C o m m i t t e e s / C C I P /Workshop/STG4.Pdf+Telecommunications+Monopoly&Hl=En&Ie=UTF-8

President’s Council On Competitiveness (1992) “Telecommunications.” In The Legacy Of Regulatory Reform: Restoring America’s Competitiveness, 32-41. U.S. Government Printing Office. United States Department Of Commerce (October 1988) NTIA Telecom 2000: Charting The Course For A New Century. Washington, D.C.: U.S. Government Printing Office.

Roberts, J. And H. Sonnenschein (1977) On The Foundations Of The Theory of Monopolistic Competition. Econometrica 45, 101-113.

Rogerson, W. (1982, Autumn). The Social Costs Of Monopoly And Regulation: A Game-Theoretic Analysis. Bell Journal Of Economics,13(2), 391-401.

Rowse, A. (1998, May). A Lobby The Media Won’t Touch. The Media Lobby Itself. Cover Story. Industry Overview. Washington Monthly 30(5), 14.

Salop, S., (1979) Monopolistic Competition with Outside Goods. Bell Journal of Economics. 10:141-156.

Sapolsky, Harvey M., Ed., The Telecommunications Revolution: Past, Present, And Future, Routledge, New York, 1992.

Sarket, Y., Gupta, B., And Pal, D. (1998) Geometric Solution Of A Cournotoligopoly With Nonidentical Firms. Journal of Economic Education. Spring 1998: 118-126.

Sawyer, M. C. (1991). Economics of industries and firms: Theories, evidence and policy. London: Routledge.



Schnalensee, R. (1978) A Note On Economies of Scale and Natural Monopoly In The Distribution of Public Utility Services. Spring, 1978.

Silverstein, K. (1998) “His Biggest Takeover: How Murdoch Bought Washington.” (Rupert Murdoch’s Lobbying)(The National Entertainment State) (Cover Story) (Industry Overview) Nation V266, N21 (June 8, 1998):18-25.

Shimonura, K. (1997) A Dynamic Equilibrium Model Of Durable-Goods Monopoly. Journal of Economic Behavior and Organization. September 1997

Spulber, D.F. (1995) “Deregulating Telecommunications.” Yale Journal of Regulation. 12(1): 25-67.

Stigitz, J. (1998) Creating Competition In Telecommunications. Speech. Conference On Managing The Telecommunications Sector Post-Privatization. The George Washington University, Washington, D.C., April 27, 1998. Http://Www.Worldbank.Org/Html/Extdr/Extme/Jssp042798.Htm Accessed January, 2002.

Stoffels, B. (1999) What’s Natural About A Natural Monopoly?

Outside Plant Magazine. February 1999. Http://Www.Ospmag.Com/Columns/Stoffels/Stoffels_0299.Htm Accessed January, 2003.

Suzumura, K. (1995). Competition, commitment and welfare. Oxford: Oxford University.

Szidarovsky, F. And S. Yakowitz (1977) A New Proof Of The Existence and uniqueness of The Cournot Equilibrium,

International Economic Review,18, 787-789.

Thierer, A. (1994) Unnatural Monopoly: Critical Moments in The Development of The Bell System Monopoly. The Cato Journal Fall 1994. 14(2).

To, T. (1996) Multi-Period Competition With Switching Costs: An Overlapping Generations Formulation. Journal of Industrial Economics. March 1996. 44(1): 81-87.

United States. (1995) Congress. Senate. Committee On The Judiciary. Subcommittee On Antitrust, Monopolies, And Business Rights. Examining The Effects Of Megamergers In The Telecommunications Industry: Hearings Before The Subcommittee On Antitrust, Monopolies, And Business Rights Of The Committee On The Judiciary, United States Senate, One Hundred Third Congress, First Session: QVC Network, Inc., And Viacom, Inc., For paramount Communications, Inc., and The Merger telecommunications, Inc., And Liberty Media Corp. Into Bellatlantic Corp., October 27, November 16, And December 16, 1993. Washington : U.S. G.P.O, 1995. S. Hrg.; 103-850

United States Telephone Administration (1990) The History Of The Telephone Industry. Washington, D.C.

Utne Reader (1997)”To Free The World: George Gerbner Wants To Save Us From The Global Media Monopoly.” (Interview) Utne Reader, N79 (Jan-Feb, 1997):80-82.

Van Long, N. And Shimonura, K. (1997) Some Results On The Markov Equilibria

2009 91


Jürgen Albinger

of A Class Of Homogeneous Differential Games. Journal of Economic Behavior and Organization. November 1997.

Victor, K.(1996). Media Monsters.” (Fear Of Giving Corporations Too Much Control Over Media Leads To Some Ownership Restrictions In New Telecommunications Law. National Journal V28, N9 (March 2, 1996):480-485.

Virginia’s Center For Innovative Technology (2002) State of The Telecommunications Industry In Virginia CIT Industry Report. Telecommunications. June 2002.

Vives, X., (1989), Cournot and The Oligopoly Problem. European Economic Review. 33, 503-514.

Vives, X., (1993), Edgeworth and The Modern Oligopoly Theory. European Economic Review. 37, 463-476.

Wiley, R.E. (1981) Competition And Deregulation In Telecommunications: The American Experience. In Lewin, L. (Ed.) Telecommunications In The U.S.: Trends And Policies, Boston: Artech House: 38-42.

Winton, A. (1994) Delegated Monitoring And Bank Structure In A Finite Economy. Journal of Financial Intermediation. 4: 158-187.

Yano, M. (1997) On The Dual Stability Of A Von Neumann Facet and the Inefficacy of Temporary Fiscal Policy. Econometrica. March 1997.

Young, D. P. T. (2000). Modeling media markets: How important is market structure? Journal of Media Economics, 13(1), 31.

Article 2

Documents

q1 q2

gulf western

administrative science quarterly

propose alternative hhis

technologically dynamic markets

business intelligence journal

hl en

blacks law dictionary