Standards for Personal
`Communications in Europe
`and the United States
`
`David J. Goodman
`
`Program on Information Resources Policy
`
`Harvard University
`
`Cambridge, Massachusetts
`
`Center for Information
`Policy Research
`
`

`
`A publication of the Program on Information Resources Policy.
`
`Standards for Personal Communications in Europe and the United States
`David J. Goodman
`April 1998, P-98-1
`
`Project Director
`Anthony G. Oettinger
`
`The Program on Information Resources Policy is jointly sponsorod by Harvard University and the
`Center for Information Policy Research.
`
`Chairman
`Anthony G. Oettinger
`
`MQIUlging Director
`John C. B. LeGates
`
`David Goodman is Professor of Electrical and Computer Engineering at Rutgers University and
`Director of the Wireless Information Network Laboratory (WINLAB), a National Science Foundation
`Center at Rutgers. He performed the research for this report while he was a Research Associate at the
`Program in 1995, on sabbatical from Rutgers.
`
`Copyright© 1998 by the President and Fellows of Harvard College. Not to be reproduced in any
`form without written consent from the Program on Information Resources Policy, Harvard University,
`65 Rear Mt. Auburn Street, Cambridge MA 02138. (617) 495-4114. E-mail: pirp@deas.harvard.edu
`http://www .pirp.harvard.edu Printed in the United States of America. ISBN 1-879716-46-1
`
`Printing 5 4 3 2 1
`
`

`
`PROGRAM ON INFORMATION RESOURCES POLICY
`
`Harvard University
`
`Center for lnfonnation Policy Research
`
`Affiliates
`
`April1998
`
`AT&T Corp.
`Australian Telecommunications Users Group
`Bell Atlantic
`Bell Canada
`BellSouth Corp.
`The Boeing Company
`Cable & Wireless (U.K.)
`Carvajal S.A. (Colombia)
`Center for Excellence in Education
`Centro Studi San Salvador, Telecom ltalia
`(Italy)
`CIRCIT (Australia)
`Commission of the European Communities
`Computer & Communications Industry
`Assoc.
`CSC Index (U.K.)
`CyberMedia Group
`DACOM (Korea)
`Deloitte & Touche Consulting Group
`Dialog Corp.
`ETRI (Korea)
`European Parliament
`FaxNet Corp.
`First Data Corp.
`France Telecom
`Fujitsu Research Institute (Japan)
`GNB Technologies
`Grupo Clarin (Argentina)
`GTE Corp.
`Hearst Newspapers
`Hitachi Research Institute (Japan)
`IBM Corp.
`Intel Corporation
`Investment Company Institute
`Korea Telecom
`Lee Enterprises, Inc.
`Lexis-Nexis
`Lincoln Laboratory, MIT
`Litton Industries, Inc.
`Lucent Technologies
`John and Mary R. Markle Foundation
`Microsoft Corp.
`
`MicroUnity Systems Engineering, Inc.
`MITRE Corp.
`National Telephone Cooperative Assoc.
`NEC Corp. (Japan)
`The New York Times Co.
`Nippon Telegraph & Telephone Corp.
`(Japan)
`NMC/Northwestern University
`Pacific Bell
`Pacific Bell Directory
`Pacific Telesis Group
`The Post Office (U.K.)
`Raytheon Company
`Research Institute of Telecommunications
`and Economics (Japan)
`Revista Nacional de Telematica (Brazil)
`Samara Associates
`Scaife Family Charitable Trusts
`Siemens Corp.
`SK Telecom Co. Ltd. (Korea)
`Strategy Assistance Services
`TRW, Inc.
`UNIEMP (Brazil)
`United States Government:
`Department of Commerce
`National Telecommunications and
`Information Administration
`Department of Defense
`Defense Intelligence Agency
`National Defense University
`Department of Health and Human Services
`National Library of Medicine
`Department of the Treasury
`Office of the Comptroller of the Currency
`Federal Communications Commission
`National Security Agency
`United States Postal Service
`Viacom Broadcasting
`VideoSoft Solutions, Inc.
`Weyerhaeuser
`
`

`
`Acknowledgements
`
`The author gratefully acknowledges the following people who reviewed and commented
`critically on the draft version of this report. Without their consideration, input, and
`encouragement, this study could not have been completed:
`
`John A. A reate
`Gustave Barth
`Robert P. Bigelow
`Judson C. French
`Richard Frenkiel
`James Mikulski
`
`Michel Mouly
`Milton Mueller
`Marie-Bernadette Pnutet
`Alain Servantie
`Seo Jung Uck
`
`Four reviewers, in particular, helped me to correct some errors and otherwise improve
`the report. I thank Gustave Barth, James Mikulski , Michael Mooly, and Marie Pautet for their
`helpful suggestions.
`
`These reviewers and the Program's affiliates, however, are not responsible for or
`necessarily in agreement with the views expressed here, nor should they be blamed for any
`errors of fact or interpretation.
`
`The research for this report was conducted during 1995, while I was a research associate
`at the Program, on sabbatical leave from Rutgers University.
`
`

`
`Executive Summary
`
`In the 1990s, cellular and personal communications have been among the most dynamic
`areas of the economy, both in terms of markets and technology. As in other branches of
`information technology, the technology of personal communications is embodied in published
`standards. Standards for personal communications systems, however, are exceptional in
`several ways. For example:
`
`• They are directly influenced by government regulatory policy.
`
`• They are created before the underlying technology is mature.
`
`• Markets have accepted a proliferating number of standards rather
`than consolidating around one single standard.
`
`This report chronicles the history of cellular and personal communications standards in
`Europe and the United States. It then analyzes the standards in the context of economic
`theories developed since around 1980, with an emphasis on changes that have occurred since
`the early 1980s, when the first cellular systems reached the market, and on the differences
`between government policies in Europe and the United States. European governments have
`increasingly worked together to guide the creation and deployment of standards, while the
`United States government has taken a less active role than it did in the early 1980s. It can be
`argued that in the mid-1990s, European industry and consumers have been better served by
`government policies than their American counterparts have been.
`
`

`
`cisco Systems, Inc., Exhibit 1137
`Page 6
`
`
`

`
`Contents
`
`Acknowledgements
`
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`
`iv
`
`Executive Summary
`
`Chapter One
`
`Introduction . ... .. ............................ .
`
`v
`
`1
`
`. . . . . . . . . . . . . . . . . . . . . . . . 5
`Compatibility Specifications
`Chapter Two
`2.1 Standards in General
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
`2.1.1 Types of Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
`2.1.2 Why Have Standards? Benefits and Some Costs . . . . . . . . . . . . . . . . 6
`2.2 Compatibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
`2.2. 1 Methods of Achieving Compatibility . . . . . . . . . . . . . . . . . . . . . . . 7
`2.2.2 The Growing Importance of Compatibility Specifications . . . . . . . . . . 8
`2.2.3 The Economics of Compatibility
`. . . . . . . . . . . . . . . . . . . . . . . . . 9
`2.2.4 Examples of Compatibility . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`11
`2.3 Issues Addressed Here
`12
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`2.3 .1 Creation and Adoption of Standards . . . . . . . . . . . . . . . . . . . . . .
`13
`2.3 .2 Status of Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`14
`2.3.3 Figures of Merit for Standards
`15
`. . . . . . . . . . . . . . . . . . . . . . . . .
`2.3.4 Adoption of Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`16
`
`17
`First-Generation Cellular Systems . . . . . . . . . . . . . . . . . . . .
`Chapter Three
`17
`3.1 The Cellular Idea
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`3.2 Introduction of Cellular Service into the United States
`19
`. . . . . . . . . . . . . . .
`3.2.1 Technology
`19
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`3.2.2 Radio Spectrum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
`3.2.3 Operating Companies
`21
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`3. 3 Introduction of Cellular Service into Europe . . . . . . . . . . . . . . . . . . . . . .
`23
`3.4 Stakeholders in Europe and the United States . . . . . . . . . . . . . . . . . . . . .
`24
`26
`3. 4. 1 Consumers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`3.4.2 Network Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`26
`3.4.3 Infrastructure Vendors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`27
`3.4.4 Subscriber Equipment Vendors
`. . . . . . . . . . . . . . . . . . . . . . . . . 27
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
`3.4.5 Governments
`
`31
`. . . . . . . . . . . . . . . . .
`Second-Generation Cellular and PCS
`Chapter Four
`31
`4.1 Evolution of Technology in Europe
`. . . . . . . . . . . . . . . . . . . . . . . . . . .
`31
`4 .1.1 Motivation and Goals of GSM . . . . . . . . . . . . . . . . . . . . . . . . . .
`32
`4.1.2 Technology Development by GSM . . . . . . . . . . . . . . . . . . . . . . .
`34
`4.1.3 GSM's Decisions and Deployment . . . . . . . . . . . . . . . . . . . . . . .
`4.1.4 Other European Wireless Telecommunications Standards . . . . . . . . . 35
`4.2 The Evolution of Technology in the United States . . . . . . . . . . . . . . . . . .
`36
`4.2. 1 NA-TDMA Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
`4.2.2 The Development of Code Division Multiple Access (COMA)
`. . . . .
`39
`4.2.3 Intersystem Operations
`41
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`4.2.3 Narrowband AMPS (NAMPS)
`41
`. . . . . . . . . . . . . . . . . . . . . . . . .
`4.2.4 PCS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
`
`

`
`-viii -
`
`4 .3 Stakeholders in Europe and the United States . . . . . . . . . . . . . . . . . . . . .
`42
`45
`4.3. 1 Consumers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`4. 3. 2 Network Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
`4. 3. 3 Infrastructure Vendors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
`4.3.4 Subscriber Equipment Vendors . . . . . . . . . . . . . . . . . . . . . . . . . 47
`4.3.5 Government . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
`
`. . . . . . . . . . . . . . . . . . . . . . . .
`Analysis of PCS Standards
`Chapter Five
`51
`51
`5.1 Creation of Standards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`52
`5.2 Status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`53
`5.3 Figures of Merit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`5. 3.1 Extrinsic Figures of Merit
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
`5.3.2 Intrinsic Figures of Merit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
`5.4 Adoption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
`56
`5.4. 1 Equipment Vendors
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`5.4.2 Operating Companies
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`57
`5 .4. 3 Subscribers
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`59
`5.5 Relation of This Study to Theoretical Studies of Standards . . . . . . . . . . . . .
`59
`5.6 Implications for Public Policy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`61
`
`Acronyms
`
`. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`
`63
`
`

`
`Illustrations
`
`Figures
`1-1
`
`3-1
`
`3-2
`
`4-1
`
`Tables
`3-1
`
`4-1
`
`4-2
`
`5-l
`
`Four Factors in Cellular Communications . . . . . . . . . . . . . . . . . . . . . . . . 1
`
`Elements of a Cellular System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`
`18
`
`Cellular Stakeholders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`
`26
`
`Global System for Mobile Communications (GSM) Architecture . . . . . . . . .
`
`33
`
`First-Generation Cellular Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . .
`
`25
`
`Digital Cellular Systems in 1997
`
`. . . . . . . . . . . . . . . . . . . . . . . . . . . .
`
`43
`
`Personal Communications Systems in 1997 . . . . . . . . . . . . . . . . . . . . . .
`
`44
`
`Standards Widely Adopted in the United States and Europe . . . . . . . . . . . .
`
`52
`
`

`
`cisco Systems, Inc., Exhibit 1137
`Page 10
`
`
`

`
`Chapter One
`
`Introduction
`
`Since their birth in the early 1980s, cellular communications have proved a major
`success throughout the world. The public appetite for cellular telephones has consistently
`exceeded expectations of market analysts and fueled the rapid growth of service and
`equipment industries. Although the interplay of suppliers and consumers is common
`throughout the economy, two other factors have had an unusually strong influence on cellular
`communications: uncertainties about new technologies and government policy. As a
`consequence, the nature of cellular communications has been determined by the four factors
`illustrated in Figure 1-1, each of which influences all the others. This situation differs from
`other areas of information technology, for example video recording and personal computing,
`where the interaction of industry and consumers determines market growth in an environment
`of uniform government policy and predictable technology evolution. The situation portrayed
`in Figure 1-1 is complex. Moreover, the details of the interactions among the four factors are
`in a state of flux and differ considerably in different parts of the world.
`
`C 1998 President and Fellows of Harvard College. Program on Information Resources Policy.
`
`Figure 1-1
`
`Four Factors in Cellular Communications
`
`This report analyzes the relationship between government policy and the evolution in
`technology in Europe and the United States. The focus is on the standards that govern cellular
`communications and cellular's 1990s' counterpart, "personal communications." Usually,
`
`

`
`-2-
`
`personal communications are referred to as PCS, in which "S" stands for services or systems.
`PCS is presented to the public as cellular telephone service with frills, such as paging, voice
`mail, and calling number identification. To government regulators, PCS refers to frequency
`bands in which telephones operate. The cellular bands are around 450 megahertz (MHz), 850
`MHz, or 900 MHz, depending on national regulations. PCS bands are around 1900 MHz in
`the United States and Canada and around 1800 MHz in many other countries. For the most
`part, in this report the terms "cellular" and "PCS" are used as synonyms, because they share
`common technologies and in many respects are treated similarly by government regulators. In
`a few places (especially in Chapter Four), it is helpful to distinguish between services offered
`in the cellular and PCS frequency bands.
`
`As in other branches of information technology, PCS equipment must conform to
`established standards, but the creation and adoption of standards for PCS differ substantially
`from procedures in other application areas. There are two major, interrelated differences:
`
`• PCS standards were created before the conforming technologies were
`developed; and
`
`• there are many competing PCS standards.
`
`The striking effect of these differences is that the PCS industry, while mature commercially,
`remains a standards battleground, with an increasing, rather than decreasing, number of
`warring factions. This situation appears to be unique in information technology.
`
`This report documents the status of PCS standards in the United States and Europe
`during the past fifteen years since roughly 1980. Many of these issues addressed here are
`identified in a body of literature that has grown steadily since the mid-1980s in response to a
`recognition of the increasing economic importance of information technology standards. Most
`of the standards literature falls into two categories: case studies, such as this one, and analyses
`based on abstract models. This report is a case study of PCS that refers to the abstract models
`in order to explain some of the economic forces guiding the evolution of PCS.
`
`The report is aimed at two audiences. It offers members of the PCS community a
`context for analysis and decisonmaking relevant to the creation and adoption of standards.
`And to scholars with a general interest in standards, it offers another case study, distinguished
`in two ways: first, by the diversity of competing standards existing in a single industry, and,
`second, contrary to many other case studies, by describing issues still far from resolved. In
`this sense, PCS standards are a laboratory for examining hypotheses elsewhere deduced from
`abstract models or induced from retrospective case studies.
`
`Chapter Two presents a framework for examining standards in general, with a focus on
`compatibility specifications. This framework, largely derived from the standards literature,
`
`

`
`-3 -
`
`emphasizes criteria for judging the merits of specific standards. Chapters Three and Four
`describe the history of first- and second-generation PCS systems in Europe and North
`America. The striking differences in the status of standards in these two continents is in itself
`surprising. Moreover, since 1988 the situation has flip-flopped so that the status of second(cid:173)
`generation European standards resembles that of first-generation North American standards,
`and vice versa. Chapter Five uses the economics framework developed in Chapter Two to
`interpret the case studies presented in Chapters Three and Four. A principal conclusion is
`that, to date, the policies adopted in Europe in the 1980s have served consumers and industry
`better than policies in the United States.
`
`

`
`cisco Systems, Inc., Exhibit 1137
`Page 14
`
`

`
`Chapter Two
`
`Compatibility Specifications
`
`To establish a framework for examining PCS compatibility, compatibility specifications
`must first be placed in the general context of standards for products and services. There are
`many types of standards, each serving its own purposes. Compatibility specifications serve the
`purpose of interoperability. They insure that two or more products will function properly
`when part of the same system. Although the details of interoperability remain the fundamental
`purpose of compatibility specifications, specifications also address other goals that influence
`the value of standard products to their producers and consumers.
`
`2.1 Standards in General
`
`2.1.1 Types of Standards
`
`Another author, introducing standards to his readers, explained that when he began his
`investigation, "Standards of every kind started to appear everywhere. " 1 To prevent
`proliferation from obscuring the principal issues addressed here, this investigation begins with
`a definition that encompasses the scope of this report:
`
`Standards are formal rules that guide the manufacture of products and
`the delivery of services.
`
`Although this definition excludes, in the words of that author, "the way my wife and I write
`phone messages, "2 it still incorporates a wide variety of practices.
`
`To appreciate the wide range of issues addressed by standards, it is helpful to identify
`categories that refer to the principal purpose of a standard:
`
`1. Appearance, labeling
`2. Dimensions, such as size and weight
`3. Safety
`4. Quality
`5. Compatibility specifications
`
`1Yesha Y. Sivan, Knowledge Age Standards: Present Scope and Potential Use in Education (Cambridge, Mass:
`Harvard University Program on Information Resources Policy, P-96-1, March 1996), 7.
`
`2Ibid.
`
`

`
`-6-
`
`All of them could be observed while I was recently examining a compact disc (CD) player in
`a store. Like other components of an audio system, the CD player was in a black cabinet (1).
`The controls were labeled with the same icons that appear on other CD players ( 1). The disc
`player has the same width and depth as other stereo components (2). It carries an endorsement
`from Underwriters Laboratories, certifying that it meets electrical safety standards (3). "Eight(cid:173)
`times oversampling" in combination with an "18-bit digital-to-analog converter" indicate the
`quality of the audio signals (4). And, finally, the CD player is compatible with all my
`compact discs (5).
`
`2.1.2 Why Have Standards? Benefits and Some Costs
`
`Standards deliver a variety of benefits to industry and to the public at large. Appearance
`and labeling standards promote a company's image and help consumers understand the nature
`and function of a product. The benefits of safety and quality are obvious and inherent in the
`products and services they cover. Other benefits fall into the category of network economies
`and network externalities. 3 These are economic benefits that derive from wide adherence to
`standards, beyond the inherent benefits of the products and services to their users.
`Externalities can be positive or negative, depending on whether the total added benefit of one
`more unit is greater or less than the benefit to the person obtaining it.
`
`Dimensional standards are useful principally because they produce network efficiencies.
`There is nothing inherently optimum in stereo components that are 44 centimeters (em) wide
`and 29 em deep. However, wide adherence to this convention produces efficiencies in storage
`and display for retailers. It also serves consumers by stimulating the availability of
`inexpensive, space-efficient audio equipment furniture. Dimensional conventions promote
`efficiency of service and repair facilities. Car maintenance would be even more expensive
`than it is if each model were to come with its own size of spark-plug threads.
`
`Externalities can also be negative. For example, the batteries in portable electronic
`equipment increase the value of the equipment to its users but create disposal problems for the
`community at large. In some instances, the nature of the externality depends on how many
`similar products already exist. Citizens band (CB) radios at first had positive externalities.
`When there were few CB radios, the purchase of each new one gave existing owners new
`opportunities to communicate. But when the airwaves became congested, each new radio,
`although beneficial to its purchaser, added to the likelihood that other people would encounter
`congestion when they tried to use theirs.
`
`3In this context, the word "network" is used in the general sense of a collection of related goods and services.
`Elsewhere in this report the discussion refers extensively to communications networks consisting of electronic
`equipment that stores and transports information.
`
`

`
`-7-
`
`Even when they produce externalities that are positive in aggregate, standards can have
`negative side effects. By limiting variety, they can circumscribe efficiency and quality. Many
`stereo components could fit comfortably in boxes that are less than 44 em wide and 29 em
`deep. Smaller boxes would consume less metal for construction, and smaller boxes would be
`more economical to store and transport than standard ones. When standards become
`entrenched they can inhibit innovation. Someone may invent a better technique than Dolby for
`suppressing noise in audio tapes, but it would be hard to fmd a market for the new method in
`the presence of the "excess inertia"4 associated with the millions of tapes and tape players
`that conform to the Dolby standard.
`
`Some standards, such as the dimensions of audio equipment, embody a mixture of
`advantages and disadvantages. They are adopted when the pluses outweigh the minus. By
`contrast, compatibility specifications are often compelling. Whatever its color and size, a CD
`player that is incompatible with CD recordings would be useless. Similarly, compatibility of
`PCS system components is essential.
`
`2.2 Compatibility
`
`2.2.1 Methods of Achieving Compatibility
`
`The word "compatibility" implies a relationship between two or more different things.
`Gabel's book5 begins with an excellent discourse on the economic implications of
`compatibility. He points out that a standard is one way, but not the only way, of achieving
`compatibility. In some cases, compatibility is inherent in the relationship between separate
`properties of two things. He offers the example of tea and lemons. Other times, compatibility
`is achieved by means of an adaptor, which is an additional piece of equipment that allows two
`incompatible items to function together. To travelers, an electrical adaptor is a familiar means
`of making an appliance compatible with a local power source. Compatibility can also be
`achieved by gateways, which are products capable of conforming to more than one standard.
`An example of a gateway is a hair dryer that can be used with both European and North
`American electricity supplies. Gateway technologies, in the form of dual-mode cellular
`telephones, figure prominently among emerging cellular telephone products in North America.
`
`The focus of this report is on products that conform to formal compatibility
`specifications. Sometimes a specification describes only one part of a system, such as a
`television transmitter. In doing so, it implicitly establishes rules for manufacturing compatible
`
`4Joseph Farrell and Garth Sa loner, "Standardization, Compatibility, and Innovation, Rand Journal of Economics
`16, 1 (Spring 1985), 70-83.
`
`5H. Landis Gabel, Competitive Strategies for Product Standards: The Strategic Use of Compatibility Standards for
`Competitive Advantage (London: McGraw-Hill, 1991), Chapter 1.
`
`

`
`-8 -
`
`receivers. By contrast, a cellular phone standard typically contains rules for manufacturing
`groups of products that have to be mutually compatible.
`
`2.2.2 The Growing Importance of Compatibility Specifications
`
`Prior to roughly 1980, the creation of standards in teleconununications was, for the most
`part, a technical activity confmed to experts within one company or to small teams of
`specialists nominated by their employers. In the telecommunications industry, standards were
`typically produced by representatives of national monopolies. When a single company
`produced a specification, the task of the experts was to formulate recipes for producing
`products that conformed to existing technology. Committees spanning several companies,
`usually from different countries, had the task of ironing out differences between similar, but
`not identical, approaches. In the late 1990s, this situation is rare. Standards groups today are
`battlegrounds where company representatives assert the interests of their employers and work
`to block competitors from gaining an advantage. Several circumstances explain this
`transformation.
`
`Long-standing telecommunications monopolies are rapidly disappearing. Traditionally,
`companies like AT&T and government-owned post, telegraph, and telephone authorities
`(PTTs) set national standards. With their monopoly status gone, these companies no longer
`have the authority to dictate standards. Their competitors have a stake in the details of
`emerging standards, and often their interests are at variance with those of the former
`monopolies. This situation has opened the way for many more players in the standards arena,
`and for more contention, which together demand substantially more company resources.
`Twenty years ago, telecommunications standards were set by groups of experts working
`together to achieve a good technical solution. Today, standards are set by representatives of
`companies, each trying to obtain maximum advantage for their organizations.
`
`In the 1970s, IBM derived enormous wealth and power from its ownership of mainframe
`computing standards. Today, Microsoft and Intel together hold a similar advantage in the
`realm of personal computing. It is an interesting paradox that their monopolies in software
`and microelectronics emerged concurrently with the eclipse of the world's telecommunications
`monopolies. This situation attracts the attention of everyone in the information technology
`industries who recognize the critical importance of standards to their businesses. It engages
`the U.S. government. The executive branch works to restrain Microsoft by exercising antitrust
`oversight, while the courts define the limits of the power Intel derives from its intellectual
`property rights.
`
`Technologies have become increasingly complex and interdependent as the number and
`diversity of components that have to interact to produce an effective information system grow.
`In a system, the merits of each component depend not only on the quality of the particular
`
`

`
`-9-
`
`component but also on how it functions in the presence of all the other components.
`Compatibility standards govern the interfaces between system components and critically
`influence the quality of a system. With the increasing complexity of a system, the task of
`producing a standard that promotes the quality of the system and advances the interests of a
`specific company requires more and more company resources. In contrast to issues of trade
`and industry structure, which have attracted increased government attention, technical
`complexity has in many instances had the opposite effect. Where the government once played
`a role in determining standards, officials and politicians have decided that the issues are too
`complex to be decided by administrative means. Instead, they leave standards setting to
`"market forces," which can assert themselves in various ways, depending on the technology
`involved.
`
`The expansion of product markets to cover the entire world forces companies to be
`acquainted with standards beyond their home bases. They need to acquire expertise on
`existing standards and influence the creation of new standards that will govern their products.
`This situation also engages governments. Diplomats examine national standards ranging from
`the characteristics of rice to automobile pollution controls and argue about whether they really
`promote quality and safety or whether they serve as restraints to free trade.
`
`The globalization of communications networks also commands increased attention to
`standards. For a long time, international communications accounted for a small fraction of
`information traffic. Under the auspices of the International Telegraphic Union (ITU),
`monopoly telephone companies collaborated to establish technical interfaces (adaptors) that
`reconciled differences between the national networks. Relative to compatible products
`throughout an international network, adaptors incur penalties in efficiency and quality, but
`these were of little consequence when there was little international traffic. Anticipating the
`growth of international communications, in the 1960s telephone industries moved toward
`product standardization for digital telephony. In the 1990s, with international calling a
`conspicuous growth area for telephony and the worldwide Internet expanding rapidly, global
`standards no longer are a side issue. All the other factors that lead to increased standards
`activity come together in the international scene, where their complexity is compounded by
`the need to reconcile differing national and regional interests.
`
`2.2.3 The Economics of Compatibility
`
`The academic community, recognizing the growing importance of compatibility
`specifications in an information society, has since the mid-1980s taken increasing interest in
`compatibility. Studies of compatibility fall into two categories, case studies and economic
`models. Many of the economic models focus on the effects of positive externalities. They
`address the question of whether markets with network externalities are "efficient," in the
`
`

`
`- 10-
`
`classical economists' sense of producing socially optimal characteristics.6 Typically, a model
`describes a market containing products that conform to two or more competing standards,
`each with its own price. Owing to the presence of externalities, the value of a product to a
`consumer depends on the number of compatible products. Consumers choose among
`competing products on the basis of price and assumptions about the decisions of other
`consumers. 7 Other elements of mathematical models describe pricing policies of suppliers.
`Although the models differ with respect to details such as the nature of the information
`available to consumers,8 they generally imply that in the presence of positive extern