Citation
Internet diffusion and communications policy

Material Information

Title:
Internet diffusion and communications policy a compasrison of policy styles and patterns
Creator:
Kwon, Byung W
Publication Date:
Language:
English
Physical Description:
335 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Internet -- Government policy -- United States ( lcsh )
Internet -- Government policy -- Finland ( lcsh )
Internet -- Government policy -- Korea ( lcsh )
Communication policy -- United States ( lcsh )
Communication policy -- Finland ( lcsh )
Communication policy -- Korea ( lcsh )
Communication policy ( fast )
Internet -- Government policy ( fast )
Finland ( fast )
Korea ( fast )
United States ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 303-335).
General Note:
School of Public Affairs
Statement of Responsibility:
by Byung W. Kwon.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
51805756 ( OCLC )
ocm51805756
Classification:
LD1190.P86 2002d .K86 ( lcc )

Downloads

This item has the following downloads:


Full Text
INTERNET DIFFUSION AND COMMUNICATIONS POLICY:
A COMPARISON OF
POLICY STYLES AND PATTERNS
in'
THE UNITED STATES, FINLAND, AND KOREA
by
Byung W. Kwon
B.A., Sogang University, 1989
M.A., Seoul National University, 1993
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Public Affairs
2002


2002 by Byung W. Kwon
All rights reserved.


This thesis for the Doctor of Philosophy
Degree by
Byung W. Kwon
has been approved
by
I c 'Yku^ Wi


Kwon, Byung W. (Ph.D., Public Affairs)
Internet Diffusion and Communications Policy: A Comparison of Policy Styles and
Patterns in the United States, Finland, and Korea
Thesis directed by Professor Peter deLeon
ABSTRACT
This thesis investigates policy styles and patterns for the advanced use of the
Internet in the context of global development toward an information society.
Theoretically, this dissertation seeks to fill a gap in the literature by combining the
synchronic analysis with the diachronic analysis of Internet diffusion. Practically, this
thesis suggests the appropriate governmental role for the dissemination of the Internet.
Based on synthesizing the state of knowledge on both communications policy
and Internet diffusion, this thesis develops an analytic framework composed of both
synchronic and diachronic models. Five propositions regarding the relationship of
communications policy to Internet diffusion are proposed utilizing their political,
economic, and social dimensions. They are tested via a pattern-matching comparison,
which is conducted using the three stages of Internet diffusion initiation, proliferation,
and control of proliferation. For this purpose, the United States, Finland, and Korea are
selected and compared as the most different systems.
The several findings and subsequent arguments of this dissertation can be
summarized: First, the Internet service can be theoretically identified as displaying
characteristic known as the commons, which can serve as a rationale for
governmental involvement in Internet diffusion. Second, a key success factor of
IV


Internet diffusion in the three nations as high IT intensity users is an explicit
government policy in each developmental stage of Internet diffusion, but not non-
regulation of government. Third, the economic logic of liberalization that competition
is supposed to facilitate the investment in the alternative networks and to enhance
social welfare of users cannot adequately function in the local communications market
because of the distorted pricing mechanism by non-market considerations. Fourth, the
most significant factor for Internet diffusion is not the movement toward the
deregulation in communications sector but the contingent policy mix of (1) political
national planning for digitizing the network in initial stage, (2) economic liberalization
for enhancing efficiency and entrepreneurial activities of providers in proliferation
stage, and (3) social control for enhancing equity of users in control of proliferation
stage. In conclusion, government involvement as a fine tuner of three dimensions in
each stage of Internet diffusion can lead to the end stage of Internet diffusion.
This abstract accurately represents the content of the candidates thesis. I recommend
its publication.
Signed
Peter deLeon


DEDICATION
Dedicated to my parents for their unfaltering understanding and great sacrifice.


ACKNOWLEDGMENT
I wish to express my gratitude to the chairman of my committee, Dr. Peter
deLeon, for his heartfelt guidance and never-ending patience. I also wish to thank the
other members of my committee, Dr. Robert Gage, Dr. M. J. Moon, Dr. Ralph
Longobardi, and Dr. Jennifer Oetzel, for their wise advice and helpful comments.
I especially want to thank my wife, Sun-Ju (Sunny) Chung, and my son, Tae-
Hun (Tenny), for their understanding and support during my studies.
The opinions expressed in this thesis are the authors personal views and do not
necessarily reflect those of the organization to which the author is affiliated.


CONTENTS
Figures..........................................................xiii
Tables............................................................xiv
CHAPTER
1. INTRODUCTION: INTERNET AND GOVERNMENT...........................1
Research Problems and Justification...........................1
Research Goals and Questions..................................6
Chapter Outline...............................................7
2. LITERATURE REVIEW..............................................10
Internet as Commons........................................ 11
Internet and Characteristics..............................11
Nature of Internet Service.............................. 14
Allocation Mechanisms and Internet Policy.................17
Three Theoretical Approaches.................................23
Economic Approach: Market Model...........................23
Political Approach: Government Model......................26
Social Approach: Community Model..........................28
Summary...................................................32
Synchronic Causal Studies....................................34
Internet Diffusion and Indicators.........................34
Diffusion Determinants in General.........................37
Diffusion Determinants and Communications Policy..........39
Summary...................................................47
Diachronic Process Studies...................................48
Technological Development of Internet.....................48
Logistic Model and Limits.................................52
Interactive Model and Communications Policy...............54
Summary...................................................57
Synthetic Comparative Studies................................59
Comparative Studies in General............................59
Comparative Studies and Communications Policy.............66
viii


Summary
,70
3. ANALYTIC FRAMEWORK AND METHODOLOGY................................71
Analytic Framework and Propositions.............................71
Three Dimensional (3-D) Synchronic Model.....................71
Three Stage (3-S) Diachronic Model...........................73
Research Propositions........................................75
Summary......................................................81
Research Methods................................................83
Choice of Method: Comparative Dynamics.......................84
Choice of Nations: Most Different System Design..............85
Choice of Focus and Scope: 3-D & 3-S.........................86
Choice of Mode of Analysis: Pattern Matching Comparison......88
Choice of Criteria: System Analysis and Choice as Framework.89
Data Collection and Manipulation................................91
Data Collection..............................................91
Data Manipulation............................................93
Limitations.....................................................96
4. NATIONAL POLICY STYLES AND INTERNET DIFFUSION.....................98
Three Nations and Internet Diffusion............................98
Global Diffusion of Internet.................................99
Always-on National Group and United States..................100
Scandinavian National Group and Finland.....................104
East Asian National Group and Korea....................... 108
Internet Diffusion in Three Nations.........................112
Policy Styles and Internet Diffusion...........................112
De-regulative Model: United States Case.....................113
Re-distributive Model: Finland Case.........................119
Distributive Model: Korea Case..............................122
Summary.....................................................126
High IT Intensity and Internet Diffusion.......................127
High IT Intensity Group.....................................128
Industrial Distribution of IT Employment....................128
Research and Development (R&D)..............................129
Innovativeness of IT Sector.................................129
IT Networking...............................................130
Summary and Implications....................................132
5. POLITICAL INITIATION AND INTERNET DIFFUSION......................134
IX


Political Imperative: Imperative to Distribute..................135
Theoretical Policy Rationale: Political Distribution........136
Individual Policy Rationale.................................137
Summary.....................................................143
Policy Programs and Agencies....................................144
Theoretical Policy Programs and Agencies....................145
Individual Policy Programs and Agencies.....................145
Summary.....................................................156
Policy Impact: Political Initiation and Internet Diffusion......157
Nil and Internet: Digitizing Networks.......................158
Policy Relevance: Nil and Telecommunications Networks.......T59
Nil and Initiation..........................................161
Summary and Implications........................................163
6. ECONOMIC DE-REGULATION AND INTERNET DIFFUSION.....................166
Economic Imperative: Imperative to De-regulate..................167
Theoretical Policy Rationale: Economic De-regulation........167
Individual Policy Rationale.................................168
Summary.....................................................174
Policy Programs and Agencies....................................175
Theoretical Policy Programs and Agencies....................176
Individual Policy Programs and Agencies.....................178
Summary.....................................................195
Policy Impact: Economic De-regulation and Internet Diffusion....197
De-regulation and Internet: Two Routes to Internet Taking Off..... 197
Policy Relevance.......................................... 198
De-regulation and Proliferation.............................202
Summary and Implications........................................204
7. SOCIAL CONTROL AND INTERNET DIFFUSION.............................207
Social Imperative: Imperative to Re-distribute..................208
Theoretical Policy Rationale: Social Re-distribution........208
Individual Policy Rationale.................................209
Summary.....................................................216
Policy Programs and Agencies....................................217
Theoretical Policy Programs and Agencies....................218
Individual Policy Programs and Agencies.....................218
Summary.....................................................226
Policy Impact: Social Control and Internet Diffusion............227
'I
X


Universal Service and Internet: Solving the Digital Divide.228
Policy Trends toward Social Control:
New Universal Service and Digital Divide...................232
Universal Service and Control of Proliferation.............237
Summary and Implications......................................239
8. CONCLUSION: REVIEW OF POLICY OBSERVATION.......................243
Summary of Findings...........................................244
Technological Factors......................................244
Policy Styles..............................................246
Policy Patterns............................................248
Pattern Matching Comparison................................253
Policy Observations........................................258
Policy Recommendations and Government Role....................261
Policy Recommendations.....................................261
Government Role in Each Stage..............................266
Discussion of Framework and Findings..........................269
Theories and Analytic Framework (3-D & 3-S)................269
Validity of Findings.......................................272
Future Academic Study.........................................274
Theoretical Policy Systems in Information Society..........274
Third Way in Communications Sector.........................276
Unbundling and High Speed Internet Access..................277
Post and Press.............................................278
Utilizing 3-D & 3-S...................................... 279
Closing Remarks...............................................279
APPENDIX
A: INTERNET HOSTS PER 1000 INHABITANTS..............281
B: WEB SITES IN OECD................................282
C: SECURE SERVERS PER 1 MILLION INHABITANTS.........283
D: INTERNET SUBSCRIBERS (JANUARY 2000)..............284
E: TELECOMMUNICATION FACILITIES COMPETITION IN OECD.285
F: MAJOR TELECOMMUNICATIONS OPERATORS IN OECD.......286
G: CHARGING STRUCTURE FOR RESIDENTIAL USERS IN OECD.288
H: REGULATION ON UNIVERSAL SERVICE (US).............289
xi


I: INTERNET HOSTS AND INTERNATIONAL DIGITAL DIVIDE.....291
J: NON-OECD NATIONS WITH LARGEST INTERNET HOSTS........292
K: BROADBAND ACCESS TO INTERNET IN OECD............293
L: KOREAN PLAN FOR NKN-G...........................294
M: KOREAN PLAN FOR NKN-P...........................295
N: COMMUNICATIONS SERVICE IN THREE NATIONS (1991-2000).296
0: INTERNET DIFFUSION IN THREE NATIONS (1991-2000).297
GLOSSARY.............................................298
REFERENCES...........................................303
Xll


FIGURES
Figure
2-1 Communications Policy Associated with Traditional Industries..............14
2-2 Classification of Nature of Goods and Service.............................16
2-3 Triad of Internet Policy..................................................23
2-4 Three Dimensional Approaches to Internet..................................33
2-5 Growth of Internet Hosts: 1989-1997.......................................51
2-6 Bell-Shaped Frequency Curve and S-Shaped Cumulative Curve for
Adopter Distribution.....................................................53
2-7 Noams Diffusion Model of Telephone Network...............................56
2-8 Methods of Comparison across Time and Space...............................60
2- 9 Causes and Consequences of Public Policy..................................65
3- 1 System Model of Internet Policy...........................................74
3-2 Combined Interactive Process Model........................................73
3- 3 Theoretical Policy Styles and Patterns....................................82
4- 1 Synchronic Triad of Internet Policy in Three Nations.....................133
8-1 Policy Patterns and Internet Users in Three Nations......................257
8-2 Policy Styles and Patterns in Three Nations..............................260
xiii


TABLES
Table
2- 1 Comparison of Telecommunications and Internet......................12
3- 1 Government Reports on Nil..........................................94
4- 1 Internet Hosts per 1000 Inhabitants in Scandinavian Nations.......107
4-2 Telecommunications Access (Lines per 100) in Scandinavian Nations.108
6-1 Liberalization of U.S. Telecommunications Market..................169
6-2 Liberalization of Finnish Telecommunications Market...............171
6-3 Liberalization of Korean Telecommunications Market................173
6-4 Pros and Cons of Unbundling.......................................181
6-5 Korean Classification of Telecommunications Service Providers.....192
XIV


CHAPTER 1
INTRODUCTION:
INTERNET AND GOVERNMENT
Writing about the Internet is a dangerous business: one runs the risk of adding
to the incomplete description that has spawned misperceptions and myths about this
global network.
Brian Kahin (1997:47)
Research Problems and Justification
The end of the Twentieth Century, especially the 1990s, can be characterized by
a worldwide transition from an industrial society towards an information society
triggered by the rapid development of computer-centered information technology (IT).
The latest development to transform radically the IT landscape has been the
widespread adoption of Internet technology and its integration into the National
Information Infrastructure (Nil).
This new information technology, including an interactive communication
medium such as the Internet, is fundamentally different from traditional technologies.
Most traditional technologies have extended a persons muscular and manipulative
power and have led the development of society from an agricultural to an industrial
society (Riggs, 1957, 1964), whereas the information technology essentially extends
the human brain and neural system and leads human society toward a knowledge-based,
1


or information society (Bell, 1973; Katz, 1988; Zuboff, 1988; Kurxweil, 1999; Mitchell,
1999). Thus, the Internet is a major technological innovation of the 20 Century that
portends key political, social, and economic consequences in the context of global
development toward an information society (Castells, 1996; Albarran et al., 2000).
Technologically, the relationship between the Internet and the traditional media-
telecommunications, computing, and broadcasting is growing intimate and more
complex with the emergence of a mass market for the Internet. Moreover, the Internet
blurs traditional demarcations based on types of media in the regulation of
communications services, combining features of telephony, broadcast, and data
communications.
Politically, the new wave of large-scale infrastructure projects, such as the Nil
Initiative triggered by the U.S. government in 1993, will transform the old social order
of the industrial era. The new information society is a development goal or ideology
politically supported by many governments through a variety of policy instruments
(Gamham, 2000). For instance, many scholars propose that the Internet might be
expected to revive participatory democracy (Naisbitt, 1982; Deaken, 1981; Rheingold,
1993) and has even been used as an indicator of a countrys level of democracy
(Anderson et al., 1995).
Socially, there is the threat that technology can contribute to increasing
inequality, given that it is unequally distributed among the population (NTIA, 1999;
Novak & Hoffman, 1998), the so-called digital divide. Culturally, the Internet could
2


affect human interaction (Kiesler, 1997; Gackenbach, 1998; Wallace, 1999). For
instance, the empowering role of the Internet is illustrated by how online interactions
enrich interpersonal relations through the development of social relationships and by
online community building (Parks and Floyd, 1996). Furthermore, the Internet brings
various cultures into closer proximity by its death-of-distance phenomenon
(Caimcross, 1997).
Economically, in 1991, the Internet had less than 3 million users around the
world, and its application to e-commerce was non-existent. By 1999, an estimated 250
million users accessed the Internet and approximately one-quarter of them made
purchases online from electronic commercial sites, worth approximately $ 110 billion
(OECD, 2000a). The Internet via e-commerce offers incredible potential and will
change the ways in which businesses trade with each other and people shop for goods
and services. In short, public access to the Internet is not a technical luxury but a
critical necessity for full participation in economic, political, and social life in the
information society of the 21st Century (Burgess et al., 1999: 3).
Therefore, it is appropriate to assess the role of government and public policy to
ensure the mature use of the Internet in the context of social transformation to an
information society. Nevertheless, especially among the international organizations, it
is commonly held that the movement toward deregulation of communications markets
is a worldwide phenomenon (OECD, 1997a, 1999a, 2001a; ITU, 1997, 1999;
Wellenius, 1992). The external enforcement of competition in the telecommunications
3


sector is now possible with the World Trade Organizations (WTO) Basic
Telecommunications Agreement. In addition, the faith in market-based solutions to the
telecommunications sector is extended to the studies of Internet diffusion regarding the
Internet-related regulatory issues. The terms such as privatization, deregulation,
commercialization, and liberalization are surrogates for the extension of economic
trends in the telecommunications sector to the development of the Internet.
At the core of the economic studies of Internet diffusion are the arguments that
non-government regulation (un-regulation) is a key success factor of Internet
diffusion (European Commission Report, 2000), and that the movement toward
deregulation in the telecommunications sector is the single and most significant factor
for the development of the Internet (FCC Working Papers, 1997, 1999). Consequently,
keeping abreast with the empirical data from the international organizations regarding
the international comparison of Internet diffusion, the academic studies of Internet
diffusion are skewed toward the synchronic and static analyses based on the economic
and monocausal perspectives (Paltridge, 1996; Hargittai, 1999; Mann, 2000; Guillen
and Suarez, 2001).
Nevertheless, with the emergence of a mass market for the Internet caused by
technological innovation, the nominal explanatory power of monocausal and static
economic perspective in the Internet arena is now being challenged. First of all, these
studies usually pay little attention to the diachronic and dynamic process of Internet
diffusion in the context of the social transformation toward an information society.
4


Second, the economic studies of Internet diffusion set aside as a given the huge
network externalities caused by the multidisciplinary characteristics of the Internet in
nature. In the Internet arena, network externalities can cause huge and significant
consequences to the transformation toward an information society on the outside of the
Internet market where the invisible mechanism of the price is supposed to function. For
instance, social solidarity can be reinforced by an online community such as the
alumnus reunion websites. The political democratization might be reinforced by an
e-govemment movement such as the firstgov.gov website.
Third, technological traits of the Internet e.g., no physical connection between
client and host computer make it impossible to impose a usage-based charge, which
render the typical cost function ill-suited to the Internet arena. For example, the most
striking characteristics of the communications markets shared by Australia, Canada,
New Zealand, and the U.S. (which show a high rate of Internet performance) is that
they have a flat rate charging structure of the dial-up connection to the Internet (OECD,
2000a).
Fourth, practical problems can arise when monocausal economic theories are
applied to the explanation of Internet diffusion phenomena among the respective
nations. For example, both the United Kingdom (hereafter, U.K.) and Japan are fully
liberalized nations with competition in local telephone and privatized Public
Telecommunications Operators (PTOs), but they show low Internet performance. On
5


the other hand, Iceland is a non-liberalized nation with a monopoly in local telephone
and government-owned PTOs, but shows high Internet performance.1
Research Goals and Questions
Broadly speaking, the goal of this dissertation is to consider the appropriate
government role that would result in a nations full participation in economic, political,
and social life of the 21st Century information society via the Internet. More
specifically, the practical goal of this thesis is to provide a better understanding of
emerging issues regarding the Internet and its related regulatory policy. The theoretical
goal is to address a gap in the literature by combining synchronic analysis with
diachronic analysis of the relationship of communications policy to Internet diffusion.2
For the purpose of achieving these research goals, this thesis examines the
national policy styles and policy patterns for the mature use of the Internet across time
and space in the context of the global development toward an information society.3
More specifically, the policy rationale, policy instrument, policy effectiveness, and
1 Among the various criteria, Internet hosts per 1000 inhabitants on July, 1999 (Appendix A) can be used
to compare Internet performance: the U.K. (43.7), Japan (18.1), Iceland (108.2), and OECD average
(471);
2 In this thesis, a synchronic analysis means an analysis of the variables similarly timed but differently
placed, while a diachronic analysis means an analysis of the variables differently timed but similarly
placed. Furthermore, a synthetic analysis means an analysis of the variables differently timed and
placed. They can be matched with Roses (1993) comparative statics, trend analysis, and comparative
dynamics, respectively (see page 60).
3 In this dissertation, the national policy styles of Internet policy mean the comparative and synchronic
characteristics of Internet policy in a nation, while the policy patterns mean the comparative and
synthetic characteristics (comparative dynamics) of Internet policy in a nation. In addition, the mature
use of the Internet means the last stage of Internet diffusion with the subscription of over half of the
people in a nation.
6


policy shift for the dissemination of the Internet are compared and contrasted through
the proposed analytic framework and hypothetical propositions regarding the dynamic
relationship of communications policy to Internet diffusion. In this respect, three
interrelated questions warrant special attention:
What are the comparative characteristics of national policy styles for the
mature use of the Internet (Internet policy) across space?
What are the comparative characteristics of national policy patterns for the
mature use of the Internet across time and space?
What is the appropriate role of government in terms of national policy
styles and patterns across time and space?
Chapter Outline
To answer these questions, a brief look is to be taken in chapter 2 at the findings
from previous research about synchronic causal studies, diachronic process studies, and
synthetic comparative studies of the relationship between communications policy and
Internet diffusion. Especially, reviewing the synchronic causal theories focuses on the
governmental determinants of Internet diffusion, while reviewing the diachronic
process models focuses on the policy implications in each stage of Internet diffusion.
As a synthesis of the relationship of communications policy to Internet diffusion, the
relevant comparative studies are critically reviewed.
After this, based on the critical synthesis of the existing knowledge on the
relationship of communications policy to Internet diffusion, an analytic framework,
composed of a three-dimensional (3-D) model and a three-stage (3-S) model, is
7


proposed in the research design chapter (chapter 3). These models suggest five
theoretical propositions regarding the dynamic relationship of communications policy
to Internet diffusion, which are designed to be theoretically tested. Thus, the analytic
framework and the propositions are intended to produce a theory about the styles and
patterns of Internet policy over time and space.
A theory about an ideal style and pattern of Internet policy is tested using a
pattern-matching comparison, which is conducted in view of the 3-D and 3-S of
Internet policy. For this purpose, this study selects the three relevant cases of the
United States of America (hereafter, U.S.), the Republic of Finland (Finland), and the
Republic of Korea (Korea) as a most different system design to show the similarities
of Internet policy patterns over time. Internet policies of each of the three nations are
compared and evaluated in terms of a triad of Internet policy.
In chapter 4, after identifying the technological similarities, the national policy
styles in the U.S., Finland, and Korea are compared and contrasted to show that each of
the three nations has relatively different national policy styles for the mature use of the
Internet that is, a de-regulative model of the U.S. with the free market orientation, a
re-distributive model of Finland with the welfare state orientation, and a distributive
model of Korea with the high stateness condition.
In chapters 5, 6, and 7, the theoretically drawn patterns of Internet policy of the
three nations in each stage of Internet diffusion are compared and contrasted
respectively with a predicted theoretical pattern of Internet policy over time under the
8


basic assumption of a S-curve process of Internet diffusion. With the pattern-matching
comparison, these three chapters provide both the similarities (a literal replication) and
the differences (a theoretical replication) of the longitudinal patterns of Internet policy
in the three nations. For this purpose, policy rationale, policy instruments, and policy
impacts on Internet diffusion of the three nations in each stage are analyzed with the
ideal pattern variables of Internet policy drawn from the research design chapter.
Chapter 8 concludes this dissertation by summarizing its findings, discussing its
theoretical and practical implications, and suggesting avenues for future research.
Especially, the governmental role for the mature use of the Internet is drawn from the
contingent perspective, that is, the appropriate policy responses need to be taken to
reflect the changes of the policy environments in each stage of Internet diffusion.
9


CHAPTER 2
LITERATURE REVIEW
After briefly reviewing the background literature about the multidisciplinary
nature of the Internet, this dissertation draws on the literature from synchronic causal
studies, diachronic process studies, and synthetic comparative studies on the
relationship of public policy to Internet diffusion over time. These literature reviews
are conducted from the perspective of public policy to provide insight toward the three
research questions.
For this purpose, the focus of reviewing the causal studies is on the influential
theoretical imperatives of communications policy for the mature use of the Internet, i.e.,
the political, the economic, and the social imperatives. In addition, the focus of
reviewing the process studies is on the two influential models on the relationship of
communications policy to the developmental stages of Internet diffusion, i.e., the
logistic model and the interactive model. The purpose of reviewing both causal and
process studies is to identify the two policy controversies in view of the observation
that the literature on Internet diffusion is multidisciplinary in nature (Hindle et al.,
1997), which leads us to review the final and synthetic comparative studies.
10


Internet as Commons
Internet and Characteristics
The Internet4 is a revolutionary and global medium that combines the elements
of telecommunications, computing, and broadcasting services into a new model for
communications. When compared with telecommunications, the Internet uses radically
different technologies and pricing mechanisms (Manishin, 1996; Werbach, 1997;
Gibson, 1999). Unlike circuit-switched telecommunications such as the public
switched telephone network (PSTN), packet-switched Internet communications share
bandwidth, so that there is no physical connection between client and host computers.5
Some top-level domains (such as .uk for Britain) are country-specific; others (such as
.com) are inherently generic and non-geographic. Perhaps most significantly in the
telecommunications domain, transport and applications are bundled, so that most of the
intelligence of the medium resides in the network. On the Internet, however,
intelligence is decentralized, and applications are delivered by client-oriented software.
Each also is classified into different services especially in the U.S.; that is, the basic
4 The Internet can be simply defined as an interconnected global computer network of tens of thousands
of packet-switched networks using the Internet Protocol (IP). IP defines the structure of data, or
packets, transmitted over the Internet. The higher-level transmission control protocol (TCP) controls
the routing and transmission of these packets across the network. Most Internet services use TCP, and
thus the Internet is often referred to as a TCP/IP network (Werbach, 1997:10).
5 A packet-switched network means that data transmitted over the network is split up into small chunks,
or packets. For a packet-switched network, a dedicated end-to-end transmission path (circuit) does not
need to be opened for each transmission. Rather, each router calculates the best routing for a packet at a
particular moment in time, given current traffic patterns, and sends the packet to the next router
(Werbach, 1997:17).
11


service and the enhanced service,6 which causes information service providers to
exempt the access charges to local networks and not to contribute to universal service
funding (USF). With relation to the pricing mechanisms, the packet switching tariffs
are based on bandwidth, content, and the bit, not by the minute as for a traditional
circuit switched network (Table 2-1).
Table 2-1: Comparison of Telecommunications and Internet
Telecommunications (PSTN) vs. Internet

analog, becoming more digital digital
circuit-switched packet-switched
physical connection connectionless
static switching dynamic routing
geographic numbering neo-geographic addressing
centralized intelligence decentralized intelligence
< Economic Aspect-Pricing>
usage-based pricing flat-rate pricing
by the minute by the bandwidth the content and the bit

basic services enhanced services
regulated -* competitive competitive -* ????
contribution to USF non-contribution to USF
access charges no access charges
6 Such a legal distinction between the unregulated Internet and the regulated telecommunications was
drawn from the FCCs three Computer Inquiries (so-called, Computer I, II, and III) in the U.S. For the
purpose of preventing cross subsidization and discrimination between unregulated and regulated service,
two questions for essential determination were crucial to all Computer Inquiries: (1) how to draw a line
between data processing and communications services and (2) whether, and if so, under what conditions
to permit regulated common carriers into unregulated data processing services. For further information,
see First Computer Inquiry, 28 FCC 2d 267 (1971); Second Computer Inquiry, 77 FCC 2d 384, 419
(1980); Third Computer Inquiry, NPRM, FCC 85-358 (1985).
12


Furthermore, the Internet enables structural convergence among the domains of
communications, information, and computing, which brings three different policies and
values into contact on the web (Albarran, 2000; Kahin, 1996).
In the past, communications networks were designed to carry different types of
information separately. Telephone networks were designed for voice, broadcast
networks for video, and so forth. Traditionally, communications regulation has been
concerned with the provision and operation of the physical network and access to that
network, that is, the regulation of the conduit. First of all, the telecommunications
industry has traditionally been controlled in terms of market entry, service pricing, and
technology to ensure interoperability of equipment. Second, in the broadcasting
industry (radio and television, for instance), licensing has provided the basis for
regulation on political and cultural criteria. The objectives of governmental regulation
have typically included ensuring pluralism, impartiality, protection of vulnerable social
groups, and the promotion of cultural heritage. Third, the computing industry
fundamentally was able to develop in a largely unregulated fashion (Pool, 1983).
Now, digital technologies permit the manipulation of all forms of information -
voice, data and video across all types of networks. We can easily see the convergent
trends of traditional services in the Internet in which applications are spreading across
communications media, from the telephone to the wireless, cable, and mass media such
as Digital TV (Clements, 1996; Blackman, 1998) (Figure 2-1).
13


Figure 2-1: Communications Policies Associated with Traditional Industries
freedom of expression
government information
mandatory standards
patents
research funding
tax credits
antitrust
Source: Brian Kahin, The Internet Business and Policy Landscape. 1997. p.66
Consequently, in light of both the dramatic differences between
telecommunications and the Internet and the convergent trends of communications
policy on the web, PSTN-type policy tools cannot simply be imported to the Internet
(Froomkin, 1996; Lenin & Meisel, 1997). Thus, the literature on the inherent nature of
the Internet needs to be critically reviewed.
Nature of Internet Service
Commonly cited criteria for the distinction between the public and the private
sectors are funding and ownership (Wamsley and Zald, 1973). If these are used as the
criteria for the Internet, Internet service can be considered public goods in view of
14


public funding and ownership in its initial development, on the one hand. The Internet
was initially developed by the U.S. Department of Defense Advanced Research
Projects Administration (ARPA) as a publicly financed defense initiative. Furthermore,
later the NSFNET that was evolved directly out of ARPANET was also funded by the
National Science Foundation (NSF) until its privatization on April 30, 1995
(Moschovitis, 1999; Varian, 1997). On the other hand, the Internet also can be regarded
as private goods in terms of both private ownership and termination of public funding
after the privatization of the NSFNET (Hallgren et al., 1999).
The main problem of this dichotomous scheme is that it makes no mention of
government regulation7 (Rainey, 1997: 66). Therefore, to understand the relationship
between Internet diffusion and communications policy, a review of the literature on the
inherent nature of Internet service that provides us with the rationale for governmental
intervention is indispensable (Head, 1974; Ostrom and Ostrom, 1977; Bobrow and
Dryzek, 1987). It can be analyzed in terms of both possibility of exclusion (exclusion,
failure of exclusion) and mode of use (separable use, joint use)8: in a matrix format,
this pairing results in private goods, toll goods, common pool resources (CPRs), and
public goods (Jennings, 1986: 117) (Figure 2-2).
7 As Rainey noted, many corporations, such as IBM, receive funding from government contracts but
operate so autonomously that they clearly belong in the private category.
8 Exclusion is characteristic of a goods when individuals can be denied access to the goods by the
provider. A goods is characterized by joint consumption when the use of that goods by one person does
not detract from or prevent its use by another person.
15


Figure 2-2: Classification of Nature of Goods and Service
Exclusion Failure of Exclusion
Separable Use Private Goods CPRs
(PC) (Natural Resources)
Joint Use Toll Goods Public Goods
(Roadways) (National Defense)
Source: Jennings, 1986:117
When there is no network congestion,9 the Internet service gets its nature of
public goods because it is subject to joint consumption and exclusion is infeasible
(Hallgren and McAdams, 1994). First of all, the cloud10 of the Internet is common to
all Internet users. When a new user sends traffic over the network, current users see no
difference in their performance; service is subject to joint consumption. Anyone at a
network node can make use of the network theoretically; exclusion is not possible.
When network congestion on the Internet occurs, the Internet can be classified
into the CPRs exclusion is not possible, but consumption is not joint. For instance, in
the U.S. out of the situation of degradation, the Internet service manifests itself as
congestion of the ARPANET. So, the main characteristic of Internet service is a
failure of exclusion, causing it to get inherent natures of public goods and CPRs. In
this respect, the Internet can be classified into the commons (Brin, 1995; MacKie-
9 The Internet service can be a congestible resource like the roadways in that the Internet service has a
classic problem of the commons and the free riders. For instance, with a fixed fee, many users can
send lots of data (live video) simultaneously via the Internet, which can be beyond the network capacity.
The congested network will cause delayed or dropped transmission of the given data.
10 The Internet cloud, which can be defined as the Internets network of networks, is a perfect analogy
for the wide area networking capabilities for the Internet since there is no fixed form which keeps it in
place (Mcknight & Bailey, 1999: 156).
16


Mason and Varian, 1996), but cannot be classified into both toll goods (roadways) and
private goods (PC).11 For instance, Internet users pay a fixed fee in exchange for
unlimited access up to the maximum throughput of their particular connection, which
will bring on the classic free riders problem in the Internet network (MacKie-Mason
and Varian, 1996).
Allocation Mechanisms and Internet Policy
The major question arises about the allocation of the Internet resources as the
commons, that is, how one can solve the dilemma of the so-called tragedy of the
commons of Internet networks (Olson, 1965; Hardin, 1968). Central governments
intervention is one traditional solution, and marketization is another. Thus, the state
and the market may be considered dichotomous solutions, but this thesis intends to
include the newly emerging community as the third solution from the practical
perspective of division of labor among the allocation mechanisms and the three major
variables of Internet policy (network, service providers, and users). In this section, after
reviewing the historical overview of the state, the market, and the community, the
theoretical foundations of the triangular nature of both the allocation mechanisms and
the Internet policies are briefly reviewed.
11 Toll goods can be simply defined as those for which exclusion is feasible but there is joint use, and
private goods as those which are not subject to joint use and for which exclusion is feasible.
17


Historical Overview. From the historical point of view, there is an ebb and flow
in a worldwide tendency to place our faith in the private market, public government,
and non-profit community (Mishra, 1984; Kettle, 1993; Adams and Hess, 2001).
Reliance was on the private market before the Great Depression and World War
II. After World War II and for the subsequent three decades, welfare states were
premised on state intervention to steer markets and communities in the direction of
increased growth and equity. Thus, reliance was on the public government.
By the 1970s there was a crisis of legitimacy for the welfare state caused by
government failure (Mishra, 1984), and the 1980s and 1990s saw the influence of
markets and public choice reasoning dominate in the public sector. So, reliance was on
the private market again. While these policies based on market approaches have
succeeded in introducing efficiency instruments to public sector management, they
have failed in areas in which political rather than economic judgments, and qualitative
rather than quantitative instruments, are necessary (Wolf, 1993; Adams and Hess,
1999).
Thus, the failures of government intervention and the negative impacts of the
subsequent capture of public policy by market-based rationality have created the
political space for the recent emergence of community-based ideas (Adams and Hess,
2001).
18


Dichotomy. The relevant examples of the dichotomous distinction between the
government and the market can be drawn from Deborah Stones (1997) public policy
theory and James Careys (1989) communications theory. The two theories together
show the combined traits of Internet policy that are closely connected to both a public
policy and a communications medium.
Stones (1997: 18-34) two competing policy models i.e., the market model and
the polis model is one example of the dichotomous distinction. The polis model is a
model of political society, that is, a model of the simplest version of society that retains
the essential elements of politics. Its unit of analysis is community, and its glue
bridging the vast gap between self-interest and public interest is influence, cooperation,
and loyalty. In the polis model, political resources are governed by the laws of passion
- human resources are renewable and expand with use. In the market model,
individuals act only to maximize their own self-interest, and the nature of collective
activity is competition. In the market model, economic resources are governed by the
laws of matter material resources are finite and diminish with use.
From the perspective of the division of labor among the three allocation
mechanisms of scare resources, Stones market model gives us a clear-cut overview of
the economic approach, but her polis model may be considered mixed and represents
an unclear model of government and community.
James Careys (1987) communication models that is, the transmission model
and the community-cultural-ritual (CCR) model are other examples of dichotomous
19


distinctions. Careys two models can be compared in four areas. First, in terms of a
level of analysis, the transmission model places primary focus on the individual,
whereas the CCR model embraces a larger conceptual territory called community.
In terms of the mode of interaction, the transmission model focuses on the
movement of messages from sender to receiver. The CCR model, on the other hand,
begins with a study of the social practices and rituals associated with a communication
event. In terms of domain, the primary analytical focus of the transmission model is on
the spatial dimension, whereas the CCR model seeks to understand the temporal
dimension. Finally, the transmission model looks to the consequences of
communication in terms of individual ratification, private goods, and rationality. The
CCR model looks, instead, to the public consequences of communication, including
traditions, public goods, and the dialogic interaction necessary to promote and sustain
democratic discourse (Lenert, 1998: 8).
James Carey established a fundamental distinction between the two models of
communication that is important for understanding socio-cultural implications of
communications policy. E. Lenert (1998) posits that current telecommunications policy
discourse, drawing upon the transmission model, aims to achieve liberalization of
telecommunication markets. From the perspective of the division of labor among the
three allocation mechanisms of scare resources, the transmission model can give us a
clear-cut overview of market logic applied to communication, but the CCR model may
be considered a mixed model of government and community.
20


Trichotomy. Considering the limits of the dichotomous distinction between the
public and the private in the Internet arena caused by the inherent nature of Internet
service as the commons, this dissertation utilizes the trichotomous distinction among
state, market, and community. This distinction is analogous to the trichotomous
distinction among the public, the private, and the non-profit.
Burton A. Weisbrod (1988) posits that it is the nature of different economic roles
of the public, the private, and the non-profit sectors that allows clear distinction: the
government exists to ameliorate market failures; the private sector exists to satisfy
market demand; and the private nonprofit sector meets the diversity of demand that is
too small to elicit a response from either market or government.
Furthermore, Adams and Hess (2001: 17) clearly show the division of labor
among the three allocation mechanisms: the government model relies on legal authority
to regulate relation and its preferences are revealed by votes; the market relies on
voluntary monetary contracts, and its preferences are revealed by prices; the
community relies on shared values as the glue explaining social relations, and its
preferences are revealed by its values and engagement with reciprocity.
Triad of Internet Policy. From the perspective of system or input-output analysis
(Easton, 1979; Dye, 1998), Internet policy, which can be briefly defined as the
communications policy for the mature use of the Internet, has the three theoretical
21


dimensions of its causes and consequences from the three major players of government
(or network), service providers, and service users, respectively.
In this regard, the trichotomous distinction among state, market, and community
can be analogous to the three major inputs from the three major players in Internet
policy. First, input from the state for network digitalization, especially from the
political arena, to the implementing administration as a policy system can be termed as
apolitical dimension of Internet policy.12
Second, input from the market, especially from the service providers, to the
policy system can be termed as an economic dimension of Internet policy. Third, input
from the community, especially from the service users, to the policy system can be
termed as a social dimension of Internet policy. These distinctions of Internet policy
based on the trichotomous division of labor among state, market, and community can
be referred to as a triad of Internet policy (Figure 2-3).
12 In this dissertation, from the perspective of the traditional politics-administration dichotomy
(Goodnow, 1893; Willoughby, 1927; Gulick, 1933; Dimock, 1958; Appleby, 1949; Waldo, 1948), the
player of Internet policy as a government can be divided into two; politics (Chief Information Officer)
and administration (Ministry or Independent Regulatory Body).
22


Figure 2-3: Triad of Internet Policy
State or Politics (Network)
Market
(Service Providers)
(Service Users)
Three Theoretical Approaches
Based on the trichotomous distinction among the state, the market, and the
community, the three theoretical approaches to the Internet as the commons can be
reviewed in view of their theoretical backgrounds and their applications to the Internet.
Economic Approach:
Market Model
Overview. The economic approach inherently originated from an extension of
neoclassical, micro-economic theory on the workings of the free market. This approach
applies to the public policy arena as in Stones (1997) market model, to the
communication theory as in Careys (1987) transmission model, and to the legal theory
as in Posners (1974) Economic Analysis of Law (EAL) model.
23


Especially regarding the legal arena, EAL was bom in the early 1960s, but it is
with the 1973 publication of Judge Richard Posners (1992) Economic Analysis of Law
that EAL began to have a dramatic impact on legal education and scholarship. Posners
central thesis was that judicial opinions do and/or should display an economic logic
(Ostas, 1999).
The economic approach is connected to the policy goal of liberalization13 and
policy instruments of competition, privatization, and price deregulation. Especially,
telecommunications policy is predominantly based on this approach.
Liberalization achieves competition by deregulation and laissez faire policy.
Liberalization is also closely connected to its policy emphasis on both the
technologically efficient transport of messages and the private ownership of the
communication networks (Hill, 1998; Lenert, 1998; Cuilenburg et al., 1995). To
achieve the policy goal of liberalization, an economic approach focuses on correction
of market failure based on welfare economics (Friedman, 1953, 1982). The policy
instrument for correcting the market failure is the introduction of competition in the
communications market through privatization. The economic approach also focuses on
deregulation of the communications infrastructure based on information economics,
including public choice theory, principal-agency theory, and transaction cost theory
(Walsh, 1995: 17; Willianson, 1985; Ghertman et al., 1995). The deregulation
movement prefers market-type allocation mechanism to governmental mechanism
13 Liberalization can be simply defined as the unleashing of the magic of the marketplace. This
simple definition is the often-cited formulation of former U.S. President Ronald Reagan.
24


under the basic assumption of Internet service as private goods. Information economics
views market strategies as a means to correct for government failure such as under- or
over-supply of public goods, and X-inefficiency of governmental bureaucracy. In the
Internet arena, Internet Economics attempts to improve the understanding of the
Internet as an economic system based on the quantitative analysis of network
congestion controls and pricing mechanism (McKnight and Bailey, 1999; Mackie-
Mason et al., 1993, 1995).
Assessment. A major strength of this approach is the incentive for its managers
to achieve greater rewards for greater efficiency and responsiveness. The second
strength is its relative independence from government and political constraints. But its
weakness is that the pursuit of profit sometimes results in social inefficiencies and
inequities. These kinds of market failures can be caused by the lack of information and
the existence of the network externality. Lack of information causes failures in the
sense that consumers do not obtain the goods and services that they would demand if
they were well informed. The existence of the network externality causes failures in the
sense that profit-motivated firms do not take into account any effects of their activities
that do not influence their revenues or costs (Weisbrod, 1988). Furthermore, the main
problem of this approach is that traditional price regulation on the Internet brings
distorted pricing structures from the perspectives of the different pricing mechanisms
between the two media (Cawley, 1997)
25


Political Approach:
Government Model
Overview. The political approach, based on a deliberative14 and a
communitarian15 framework, tries to connect theories of democracy with theories of
communication to allow new communication technologies to contribute to the
expansion of democracy and the empowerment of more people (Abramson et al., 1988;
Hacker, 1996). The interactive Internet can revive ancient Greek ideas of direct
participation in speech and debate, by reestablishing the connection between voters and
candidates, improving voter information, and increasing civic participation (Friedland,
1996; Ogden, 1994).
Furthermore, the political approach emphasizes apolitical goal of electronic
democratization, meaning that adding channels to the political communication system
will empower the citizen by popular participation in decision making in the context of
liberal guarantees of equality and individual rights16 (deLeon, 1992, 1997). Recently,
the e-govemment movement has shown us ways to improve the effectiveness of
political decision-making by making citizens aware of the how and the why of political
14 The deliberative model takes representative democracy as a starting point and asks how it might be
strengthened and made more participatory (Friedland, 1996; Dryzek, 2000).
15 A communitarian framework emphasizes deliberation over direct voting and is rooted in identifiable
communities and a concept of the public good that goes beyond both individual and group interest
(Abramson et al., 1988, Etzioni, 1994).
16 Electronic democratization can also be termed increasing freedom of communication (Lenert, 1998;
Cuilenburg et al., 1995). Freedom of communication may be defined as the right to send and not to send,
or to receive and not to receive messages without any interference whatsoever by someone else.
Information that is essential to fully participate in a democratic society should be provided free.
26


decision-making and facilitating their participation in this process17 (Milward and
Snyder, 1996). The promises of this perspective are well reflected in A1 Gores Earth
in the Balance (1992: 359):
. ... the best way for a nation to make political decisions about its future is to
empower all of its citizens to process the political information relevant to their lives
and express their conclusion in free speech designed to persuade others ....
The political approach prefers the government model to the market model as the
allocation mechanism of the resources under the basic assumption that Internet service
can be classified as public goods. The political approach postulates that collective
action through government has the potential of correcting market failures.
Assessment. The strength of political approach lies in its power of compulsion -
to regulate, to tax, and to subsidize and to monitor. It can also encourage the provision
of goods and services that are inadequately provided by the private sector or that are
not provided to particular groups in the population. Its handicaps are its political
entanglements for example, the dependence of senior officials on periodic elections -
and the weak incentives it provides for efficiency, which can be symbolized in
government failure. The major problem for the government model is dealing with
17 The Cyberspace Policy Research Group (CyPRG) studies diffusion and use of the World Wide Web in
government worldwide, particularly in terms of organization openness and internal effectiveness. For
further information, see http://www.cypre.arizona.edu
27


diversity, which is fundamentally a problem of information.18 Inevitably, it causes the
problems of over- and under-utilization.
Considering its limits as an allocation mechanism of scare resources, the key
policy issue is whether the disadvantages of using an indirect form of regulation are
offset by the greater ease of monitoring it rather than regulating production and its
distribution directly. The government regulates what it can monitor easily, and it
monitors what it can gauge usefully and inexpensively. If and when regulation of
nonprofits is easier than direct regulation of output, production processes, or the
distribution of output, the nonprofit form of institution (community) is attractive
(Weisbrod, 1988).
Social Approach:
Community Model
Overview. The social approach emphasizes a policy goal of increasing social and
cultural interaction (Lenert, 1998; Cuilenburg et. al., 1995). By treating communities as
social capital network, rather than strictly as discourse communities (Kubicek, 1997),
this approach, focusing on social capital19 and community assets-based approaches,20
tries to ground the connective elements of the Internet in social life and social culture.
18 Some people who benefit from governmental services are unintended beneficiaries; they are included
only because of the cost of excluding them. At the same time, some who do not obtain the services are
excluded unintentionally, victims of the high costs of targeting the consumer population more accurately.
19 Social capital is the necessary infrastructure of civic and community life that generates norms of
reciprocity and civic engagement (Coleman, 1990; Putnam, 1993,1994a, 1994b, 1995).
20 Assets-based development (Kretzmann and Mcknight, 1993; Mcknight, 1995) stresses locally
generated knowledge that permits communities to mobilize their assets, broadly conceived, to address
problems.
28


The failures of governmental intervention and the negative impacts of the subsequent
capture of public policy by market-based rationality have created the political space for
the (re-) emergence of community-based ideas (Adams & Hess, 2001). Traditional
broadcasting policy used to be based on the social approach. In the telecommunications
arena, social approach can be characterized by customer-centered regulation (Kahin,
1997) that promotes the interests of the customer through symmetrical removal of
restraints on any competitors ability to compete. Under this approach, a government
can limit the freedom of all carriers to compete and rely on regulatory monitoring to
ensure against undue discrimination. The government would impose administrative
burdens symmetrically on all carriers and subject all carriers to strategic abuse of the
regulatory process by rivals (Haring & Weisman, 1993).21
Increasing social interaction means universal access to the information
infrastructure (Hudson, 1994). Fundamental to life, liberty, and the pursuit of happiness
in the information age is access to video, voice, and data networks that provide a broad
range of news, education, health, and government information and services. Such
services should be provided in a user-friendly format, widely available to everyone,
including persons with disabilities. In the Internet arena, universal access
means narrowing the digital divide.
21 Such a symmetric regulation can be achieved by the essential facility doctrine (MCI Communications
Corp. v. AT&T Co.) for securing nondiscriminatory access to the local telephone facilities and
bottleneck facilities in communications on which the Internet communications is heavily dependent.
29


The digital divide can be defined as a gap between those individuals and
communities that have access to computers and the Internet and the ability to
effectively use this technology and those who do not, in other words, a gap between
information haves and have-nots (NTIA, 1999). The report of the U.S. Department of
Commerce (July, 1999) says that the following examples highlight the breadth of the
digital divide today:
(1) Those with a college degree are more than eight times as likely to have a computer
at home, and nearly sixteen times as likely to have home Internet access, as those
with an elementary school education.
(2) A high-income household in an urban area is more than twenty times as likely as a
rural, low income household to have Internet access.
(3) A child in a low-income white family is three times as likely to have Internet access
as a child in a comparable Black family, andfour times as likely to have access as
children in a comparable Hispanic household.
(4) A wealthy household of Asian/Pacific Islander descent is nearly thirteen times as
likely to own a computer as a poor Black household, and nearly thirty-four times as
likely to have Internet access.
(5) Finally, a child in a dual-parent white household is nearly twice as likely to have
Internet access as a child in a white single-parent household, while a child in a
dual-parent Black family is almost four times as likely to have access as a child in
a single-parent Black household.
Furthermore, this approach basically prefers the community model to the market
and the government model as the allocation mechanism of the resources under the
basic assumption that Internet service can be classified as a CPR. Social approach puts
forth that collective action through self-governing community has a potential of
correcting market failures. Based on the social characteristics of the Internet, the self-
30


governing model emphasizes self-regulation of the Internet by the individual user,
social body or culture, and industry and commerce, and the Internet itself.22 In 1996,
John Perry Barlow produced a proclamation he called the Declaration of the
Independence of Cyberspace:23
Governments of the Industrial World... I ask you to leave us alone. You are not
welcome among us. You have no sovereignty where we gather... You have no moral
right to rule us nor do you possess any methods of enforcement we have true reason to
fear.
Assessment. A self-governing community (or, social capital network) has the
hybrid combination of the strengths and weaknesses of government and profit-oriented
organizations. Like for-profits, nonprofits have no power to compel action, but they
can be more flexible in meeting the heterogeneous demands of consumers for
collective action. Like governments, nonprofits are being restricted in the use of any
surplus like the informational superiorities they generate. Yet, nonprofits may be more
trustworthy forms of institution than the private market (Weisbrod, 1988).
The major problem of this model is that the community has no power to compel
action. This is the reason why government involvement is justified in the arena of
appropriate Internet security controls, such as public key cryptography, digital
signature and its authentication, and credentials and certificates in todays electronic
marketplace (Kutais, 1999).
22 McCloskey, 1998. Available online: http://www.ilpf.org/selfreE/bib4-15.htm
23 Available online : http://www.dfF.org/pub/Publications/John-PeiTV-Barlow/barlow-0296.declaration
31


Summary
To obtain a comprehensive solution to the multi-dimensional Internet-related
policy controversies, a comprehensive model is useful because each theoretical
approach to the Internet gets its strengths and weaknesses as an allocation mechanism.
Hence, from the perspective of public policy, the ideal relationship among the nature of
Internet service, the allocation mechanism, and the dimensions of Internet policy can
be summarized in Figure 2-4. The political dimension of Internet policy ideally
matches with both the input from the state, especially from the political arena, and
Internet service as a public goods. The economic dimension of Internet policy ideally
matches with both the input from the market, especially from the service providers, and
Internet service as a private goods.24 The social dimension of Internet policy ideally
matches with both the input from the community, especially from the service users, and
Internet service as a CPR.
24 In Economics, CPRs used to look upon it as private goods. Martyne M. Hallgren and Alan K.
McAdams, The Economic Efficiency of Internet Public Goods, in McKnight, L.and J. Bailey (eds.),
(1999: 455-478).
32


Figure 2-4: Three Dimensional Approaches to Internet
33


Synchronic Causal Studies
Internet Diffusion and Indicators
After the ARPANET was developed by the U.S. DODs ARPA to link together
universities and high-tech defense contractors in the late sixties (Mackie-Mason &
Varian, 1995), the Internet grew rapidly, perhaps at an unprecedented rate. Significant
Internet diffusion can be observed worldwide only in the past few years with the global
number of network connected computers surpassing 35 million in 1998 compared to
less than 1.5 million in 1993 (OECD, 1999a). The current research on the indicators of
Internet diffusion focuses on data on the number of Internet hosts, web sites (web
servers and secure servers), and Internet subscribers.
A commonly cited measure of Internet diffusion is the number of Internet hosts25
(Appendix A). Hosts surveys have been undertaken by the Network Wizards (NW)
since 1981. NW defines an Internet host as a domain name that has an associated
Internet Protocol (IP) address record. This would be any computer system connected to
the Internet (direct or dial-up connections), such as firstgov.gov. NW surveys include
all country code Top Level Domains (ccTLDs such as .kr for Korea) and generic Top
Level Domains (gTLDs such as .com) and is undertaken every six months. By July
1999, the OECD analyzed the NW data and estimated that there were more than 52
million hosts in the OECD area representing 93% of the global total (Paltridge, 1999).
25 Internet hosts can be simply defined as any computer system with an Internet Protocol (IP) address
connected to the network (Coppel, 2000).
34


But, host data do not provide a full count of users because surveys do not capture all
computer systems connected to the Internet (e.g. computers behind firewalls) and thus
provide an indicator of the minimum size of the Internet (OECD, 2000; Coppel, 2000).
Recently, Telcordia Technologies26 has provided daily updates of number of Internet
hosts based on a random sample of IP addresses sampled throughout each day.
The diffusion of the Internet accelerated with the invention of HTTP, the World
Wide Web hypertext transfer protocol, and the number of web sites brings another
important indicator. These data were compiled by Netcraft27 28 and the Online Computer
Library Corporation (OCLC). Especially, the Netcraft Survey is a survey of Web
server software used on computers connected to the Internet. Netcraft collects and
collates as many hostnames providing an http-service as their survey can find, and
systematically polls each one with an HTTP request of the server name. A host name is
the first part (before the first dot) of a hosts domain name (e.g., www). Between July
1998 and July 2000, the number of web sites in the world increased from 2.6 million to
18.2 million (Appendix B). At the same time, the number of web sites under OECD
ccTLDs increased from 643.4 thousand to more than 4.3 million, and web sites under
gTLDs increased from 1.8 million to 12 million.
As one of the best indicators of the growth and diffusion of e-commerce, a count
of secure servers is used (Appendix C). Secure servers allow users to encrypt
26 Refer http://vyww.netsizer.com
27 Refer http ://www.netcraft.com
28 Refer http://www.oclc.org
35


information on, for instance, credit card data that facilitates e-commerce. Netcraft
developed the Secure Socket Layer (SSL) protocol for encrypted transmission over the
TCP/IP network. In August 1999, the OECD analyzed the Netcraft data and estimated
that there were over 46,000 secure servers in member nations. They estimated that this
represented 95 % of the global total and a 109 % increase over the previous year
(OECD, 2000a).
The numbers of Internet subscribers in a nation can be the fourth indicator
(Appendix D). This information can be drawn from the largest telecommunications
carriers reporting both the number of subscribers to their Internet services and their
estimates of market share. At the end of 1999, there were at least 49 million Internet
subscribers in the U.S., approaching 11 million in Japan and in Korea, 9 million in
Germany. A ranking of countries in terms of Internet subscribers per 100 population
shows high levels of take-up in Korea, Sweden, Denmark, Canada, the United States,
Netherlands, Iceland and Norway. In the private sector, NUA29 estimates the number of
users in each nation by averaging journalistic and market research reports from around
the world. In addition, Computer Industry Almanac30 estimates and projects it as a
function of the number of personal computers and other determining factors.
29 http://www.nua.ie
30 http://www.c-i-a.com
36


Diffusion Determinants in General
Among the various studies on the determinants of diffusion of innovation, this
literature review focuses on the determinants of both inter-country diffusion and the
diffusion of the interactive31 medium. The inter-country approach to the diffusion of
innovation attempts to explain international differences in the speed of diffusion of
innovations in terms of the characteristics of the countries and industries concerned
(Swan, 1973, Nabseth and Ray, 1974; Davies, 1979:25) or in terms of the attributes of
technology development and the objectives of technology development (deLeon, 1979).
Furthermore, among the various technological innovations, the literature on the
innovation of the interactive technologies such as telecommunications, broadcasting,
and especially the Internet is to be focused (Allen, 1988; Antonelli, 1989; Mahler and
Rogers, 1999).
In this respect, existing literature about important determinants of Internet
diffusion can be divided into two parts: non-governmental determinants, such as
economic factors, human capital, and existing technologies; and government
determinants, such as institutional and legal environments (Press, 1997: Hargittai,
1999: Guillen & Suarez, 2001).
Previous studies on non-governmental determinants of Internet diffusion have
found that economic wealth strongly predicts a populations adaptation of the Internet
(Lin, 1993). A countrys overall economic strength can affect Internet diffusion in that
31 Interactivity is the degree to which participants in a communications system can exchange roles in,
and have control over, their mutual discourse (William, Rice & Rogers, 1988).
37


the necessary resources are more likely to be present, and capital required for the
expansion of the technology is more available in richer countries.
There are two ways in which the level of human capital may be relevant to
Internet diffusion: the populations level of education and its English language
proficiency. Most studies that have examined the education level of adopters of new
technologies find that people with higher levels of education are quicker to adopt new
innovations than people with comparatively less education (Hargittai, 1999).
Furthermore, it is suggested that academic institutions often play an important role in
spreading the Internet since they are often among the first institutions in a nation to be
wired (Lin, 1993; Kelly and Petrazzini, 1997). So, the education level of a nation may
affect the rates of Internet diffusion.
In addition, Laponce (1987) suggests that some languages have greater status
than others and they dominate certain areas of life, such as the English language having
a prominence in the computer industry. In the same vein, the prominence of the
English language on the content of the Internet was reported that more than 94% of the
links pointing to pages on secure servers were in English (OECD, 2000a). So, the level
of English proficiency may affect the number of people interested in using the medium.
Thomas (1988) found that the spread of technology is contingent upon certain
technological factors being present in the target nation. With respect to the Internet,
existing telecommunication facilities may be crucial for understanding variation in the
spread of the Internet. It is because the Internet in its initial stage was developed as the
38


expansion of computer applications over the telephone lines. For instance, nearly all
the 121 million Internet subscribers at the end of 1999 still used dial-up connections
via the telephone network (PSTN) (OECD, 2000a).
Diffusion Determinants and Communications Policy
The institutional and legal environment (Hargittai, 1999) in a country is also
relevant to the Internets spread because public policies can enhance or hold back
Internet diffusion, depending on their approach to regulating mechanisms. Among
various theories on the policy determinants of Internet diffusion (Press, 1997; Press et
al., 1998; Guillen & Suarez, 2001), there are three distinctive types of Internet policy
that emphasize different influences, and warrant special attention: the de-regulative
policy; the distributive policy; and the re-distributive policy. These also mirror the
distinction among the three dimensions of Internet policy in the previous review.
Before reviewing the three types of policy determinants, Lowis policy typologies need
to be briefly introduced in that three types of policy determinants in this thesis also
mirror Lowis three types of policy.
Policy Typologies. Theodore J. Lowi (1964, 1972), in developing the three types
of policy32 (regulative, distributive, and re-distributive), has argued that policies
determine politics and thus policy processes will differ significantly depending on the
32 Lowi added constituency policy later, but this policy is not utilized in that this thesis focuses on the
external causes and consequences among the three major players of Internet policy.
39


policy type involved. So, each type of policy reveals its several comparative
characteristics.
First, regulative policy is distinguishable in that the regulatory decision involves
a direct choice as to who will be indulged and who deprived. Furthermore, regulative
policy cannot be disaggregated to the level of the individual or the single firm, because
individual decisions must be made by application of a general rule and therefore
become interrelated within the broader standards of law. Regulatory decisions are
cumulative largely along sectoral lines since the most stable lines of perceived
common impact are the basic sectors of the economy. These characteristics of Lowis
regulative policy fall on the later developed competitive regulatory policy (Ripley
and Franklin, 1980) and economic or so-called old regulation (Tatalovich and
Daynes, 1984; Lilley and Miller, 1977; Weaver, 1978; Wilson, 1980).
Regulative policy inherently based on the economic theory well matches with the
economic dimension of Internet policy. Furthermore the term needs to be changed into
de-regulative policy in that deregulation is considered to be one of the main success
factors and is a dominant policy instrument in the Internet arena.
Second, distributive policy is characterized by the ease with which it can be
disaggregated and dispensed unit by small unit, each unit more or less in isolation from
other units and from any general rule. So, distributive policies are policies that are
highly individualized decisions. Patronage or pork barrel programs like R& D can
be taken as synonyms for the main characteristic of the distributive policy.
40


Distributive policy inherently based on the political bargaining matches well with the
political dimension of Internet policy and is utilized as the policy instrument of the
political dimension.
Third, re-distributive policy is closely related to the so-called social equity and
welfare state programs so that it deals with haves and have-nots, bigness and
smallness, bourgeoisie and proletariat. The aim involved is not use of property but
property itself, not equal treatment but equal possession, not behavior but well-being.
These broad or vague characteristics of Lowis re-distributive policy are, more
specifically and clearly, developed later by Riple and Franklins (1980) protective
regulatory policy concept.
Especially, Riple and Franklins division of regulative policy well fits on the
division of the dimensions of Internet policy in that the criteria of division are the same.
For instance, protective regulatory policy focuses on the protection of the consumer,
the deprived, and the public, while competitive regulatory policy focuses on the
regulation of the monopolistic supplier.
On the contrary, the social regulation or new regulation concept focuses on
the regulation of the broad concept of non-market behaviors and consequently mixed
distributive policy with re-distributive policy. So, in this thesis, Lowis re-distributive
policy and Riple and Franklins (1980) protective regulatory policy are utilized as
the policy determinants of the social dimension of Internet policy focusing on
enhancing the equity of Internet usage and on protecting the consumer.
41


De-regulative Policy. The economic dimension of Internet policy emphasizes
the specific policy instruments of competition, privatization, and price regulation as the
policy determinants of Internet diffusion. From the perspective of Lowis (1979) type
of policy, the policy instruments of economic dimension get their nature of (de)
regulative policy. Especially, Ripley and Franklins (1980) competitive regulatory
policy focuses on the (de) regulation of monopolistic suppliers.
Competition is measured by the level of competition (monopoly, duopoly, and
competition) in the provision of public service telephone networks, especially in local
service (OECD, 1999a) (Appendix D). Furthermore, privatization is measured by the
level of privatization (privately owned or state owned) of major public
telecommunications operators (PTOs) (OECD, 1999a) (Appendix E). Pricing
regulation is measured by the PSTN charging structure on local telephony (metered or
unmetered).
Based on the arguable economic logic of liberalization,33 it is suggested that by
introducing a combination of privatization and competition into the
telecommunications sector, policy-makers can reduce or eliminate the local call
charges by competition, thereby encouraging Internet use (Mann, 2000).
Nevertheless, the so-called always-on national group of Australia, Canada,
New Zealand and the United States, with high Internet performance, shows that they
have a flat rate of local phone charges to the Internet caused by social consideration for
33 The arguments on the rationale for de-regulation will be further reviewed in chapter 6.
42


universal access not by economic logic (OECD, 2000a).34 The combination of a fixed
fee with unlimited access can cause the classic free riders problem in Internet service
as the commons. This demonstrates that flat rates of access pricing for social
consideration can independently affect Internet diffusion.
Recently, an empirical study demonstrated that competition and privatization
do not exert systematically significant effects on Internet diffusion (Guillen et al.,
2001). Therefore, considering the invisible function of the price does not work well in
the Internet market by dint of (1) the technological traits of the Internet, (2) the
convergent trends of communications policy on the web, (3) the inherent nature of the
Internet as the commons, and (4) huge socio-political network externalities, policy
instruments in the political and the social dimension needed to complement a policy
instruments based on a purely economic perspective.
Distributive Policy. The political dimension of Internet policy emphasizes the
specific policy instruments of government investment policy on National Information
Infrastructure (Nil) as a policy determinant of Internet diffusion (Drake, 1995; Kahin
& Wilson III, 1997). From the perspective of Lowis (1979) type of policy, the policy
instruments of political dimension get their nature of distributive policy in that the
34 The access price of the Internet is composed of both local call charges and ISP charges. In this study,
focus will be given on the local telephony market and its pricing structure because most people access
the Internet through a dial-up connection. Furthermore, it is because local call pricing structure is closely
connected with government policy (OECD, 2000a).
43


process of government planning and investment on Nil shows the political bargaining
and patronage trends of pork barrel programs.
In particular, several empirical studies have indicated that the governments
credible commitment to a predictable set of rules for investment facilitates Internet
diffusion (Press et al., 1998; Masaic Group, 1998; Henisz, 2000).
For instance, Henisz (2000) tries to demonstrate empirically that a governments
ability to credibly commit not to interfere with private property rights is instrumental in
obtaining the long-term investments required for countries to experience rapid
economic growth. Henisz also posits that predictable conditions for entrepreneurship
are likely to promote investment growth, so investment in Internet activities, in turn,
would contribute to an increase in the numbers of users and hosts. As political
constraints bringing high predictability in policy-making, he emphasizes (1) the
number of independent branches of government with veto point, and (2) the
distribution of preferences across and within those branches.
Furthermore, to show the relationship between branch interaction and Internet
diffusion, he argues that (1) branch interaction working as veto point provides a
positive but diminishing effect on the total level of constraints on policy change; (2)
the rate at which infrastructure penetration grows is positively related to limits on the
feasibility of policy change or constraint on political discretion. Therefore, branch
interaction working as veto point on political discretion is positively related to the
44


infrastructure penetration growth; thus political interactions across branches can be a
direct indicator of Internet diffusion.
For the purpose of the empirical testing, the Henisz index derives a new
measure of political constraints from a simple spatial model of political interaction that
incorporates information on the number of independent branches of government with
veto power and the distribution of preferences across and within those branches.
Re-distributive Policy. The. social dimension of Internet policy emphasizes a
policy goal of increasing social interaction, which can be achieved by a universal
service funding mechanism for universal access. Government sponsorship of public
access in the spirit of social equity can affect the diffusion of the Internet (Press, 1997),
as it did with the national telephone companies.
From the perspective of Lowis (1979) type of policy, the policy instruments of
the social dimension get their nature of re-distributive policy. Ripley and Franklins
(1980) protective regulatory policy, focusing on the protection of the consumer, the
deprived, and the public, well fit with the social dimension of Internet policy.
The specific subsidies on universal service are compared as to whether or not
there is a universal service funding mechanism or direct funding for universal access
(OECD, 2000b) (Appendix G).
In detail, the determination of the coverage of universal service that requires
social consideration (rather than economic rationale) is a policy matter designed to
45


encourage certain classes of user. For example: under the Telecommunications Act of
1996, the U.S. has subsidized certain classes of users, especially through subsidizing
university, school, and library connections, while Singapore provides a free Internet
account and subsidized computers for any school teacher who wants one (and over
90% of the teachers have them) (Press, 1997).
So, the ministry and the legislature determine the coverage of universal service,
while the technical cost finding and allocation of universal service stipulated in the
telecommunications law is understood as an economic regulatory dimension in the
majority of countries (OECD, 2000b).
There are different approaches to ensure universal service provision, i.e.,
Finland and Iceland utilize direct subsidy from government as a social assistance of
welfare state to the lower-income subscribers; in some countries, like the U.K., the
incumbent operator like British Telecom (BT) bears the responsibility for universal
service provision. In other countries, like the U.S., this responsibility is spread across
the industry via the universal service funding mechanism. Where telecommunications
operators share responsibilities, the independent regulator determines the cost of
universal service as well as the cost allocation among telecommunications operators
(Cherry et al, 1999; OECD, 2000b).
46


Summary
Limit of Monocausal Studies. Monocausal theories, especially those based on
economic perspectives, have long been dominant in the causal theories of Internet
diffusion (Levy & Spiller, 1994, Megginson & Netter, 1999; Gutierrez & Berg, 2000).
Yet, considering the malfunction of the price in the Internet market caused by (1) its
technological traits, (2) the nature of the Internet as the commons, and (3) large socio-
political network externalities, the explanatory power of the economic perspective is
being challenged (as reviewed before). Furthermore, the main controversy of
monocausal arguments is that the U.K. with full liberalization shows us a low Internet
host penetration rate, whereas Iceland, with a monopoly and stated-owned network,
especially demonstrates a high Internet host penetration rate. Such a policy controversy
of monocausal theory between liberalization and Internet diffusion can be solved by
multicausal theories of Internet diffusion.
Observations. Through reviewing the causal studies of Internet diffusion, the
specific policy instruments can be added to each dimension of Internet policy: that is,
de-regulative policy can match with the economic dimension of Internet policy;
distributive policy can match with the political dimension of Internet policy; and re-
distributive policy can match with the social dimension of Internet policy.
47


Diachronic Process Studies
In contrast with the synchronic approach to the determinants of Internet diffusion
reviewed above, the process models apply a diachronic approach to Internet diffusion
patterns by focusing on the adoption process and behavioral characteristics of
individual adopters (logistic model) or interactive network externalities (interactive
model). To conduct a comparative policy analysis in view of each stage of Internet
diffusion, we briefly look at both the brief history of technological development of the
Internet and the process models of interactive medium, which will bring us useful
implications on both the trends of Internet diffusion and their relationships with the
communications policy, respectively (Flynn & Preston, 1999: Rai et al., 1998).
Technological Development of Internet
Since its emergence in the late 1960s, the Internet has undergone constant
technological transformation. The technological development of the Internet can be
distinguished into three successive generations (Sim and Rudkin, 1997; Bar et al.,
2000).
First Generation. The first-generation Internet from the late 1960s to the early
1990s was a network and social engineering prototype of interest to military and
research organizations. The central first-generation applications were file transfers and
e-mail. The Internet began in the late 1960s as ARPANET, a project of the Advanced
48


Research Projects Administration of the U.S. Department of Defense, designed to link
together universities and defense contractors using TCP/IP packet-switching
technology (Gromov, 1996). As ARPANET grew during the 1970s and early 1980s,
several similar networks were established, primarily between universities.
In the mid-1980s, the National Science Foundation (NSF) created the first
backbone for the U.S. portion of the Internet, the NSFNET (Moschovitis, 1999).
NSFNET provided connectivity to NSFs supercomputer centers and provided a high-
speed backbone for developing the Internet (MacKie-Mason and Varian, 1997). The
NSF subsequently awarded a contract to a partnership of Merit (the Michigan
Educational Research and Industrial Triad, a nonprofit regional network for the state of
Michigan), IBM, MCI, and the State of Michigan to upgrade NSFNET, and to
interconnect several additional research networks (Werbach, 1997). The new NSFNET
completed in 1988, initially connected thirteen regional networks. Connections to the
federally-subsidized NSFNET were generally free for the regional networks, but the
regional networks generally charged smaller networks a flat monthly fee for their
connections.
Second Generation. From the early 1990s to the late 1990s the second-generation
Internet saw the mass adoption and commercialization of narrowband access, largely
through dial-up connection to the telephone network providing intermittent, low-
49


bandwidth connection. Commercialization and the explosion of the World Wide Web
constituted the main event of the second generation.
In 1992, the NSF announced its intention to phase out federal support for the
Internet backbone, and encouraged commercial entities to set up private backbones.
The NSF paid about $11.5 million annually for the NSFNET operation for several
years, but eventually decided that the technology was mature enough that it could be
more effectively provided by the private market (Varian, 1996; Mackie-Mason et al.,
1995).35 When the NSFNET backbone was shut down on April 30, 1995, the transition
to the new privatized network went relatively smoothly. The NSF continues to fund
some regional nets, but this funding is steadily decreasing to zero over five years.
The World Wide Web was developed by Tim Berners-Lee, a computer
programmer at CERN, a research lab for high-energy physics in Switzerland. Berners-
Lees idea was to use hypertext to link together different documents on the Internet,
which later allowed whole pages of text, pictures, sound, music, voice, animation, and
video to be transmitted over the Internet through a new technology called Hyper Text
Transmission Protocol (HTTP). Due to both commercialization and the Webs many
capabilities, the number of hosts increased dramatically from 1.8 million in 1993 to
over 16 million in 1997 (Kantor and Neubarth, 1996; Rutkowski, 1996) (Figure 2-5).
35 For Merits overview of the New Network Architecture, which provides details of the privatization
plan, see the online: http://www.merit.edu/nsf.architecture/index.html
50


Figure 2-5: Growth of Internet Hosts: 1989-1997
Source: Westphal and Towell, (1998:27)
Figure 2-5 shows the trends of Internet diffusion during 1987-1997, and the
beginning of Internet diffusion is familiar with the s-shaped development, which leads
to review of the logistic model.
Third Generation. From the late 1990s, the third-generation of the Internet can be
characterized by the fact that massive users are about to experience always-on high-
speed access to the Internet from their homes. The technological innovation of
broadband access to the Internet through cable, Digital Subscriber Lines (DSL), and
wireless Internet is the main event of this generation. In 1996, the research and
education community of the U.S., with the support of the White House and several
federal agencies, announced the Internet II or Next Generation Internet (NGI)
initiative to establish a new high-speed Internet backbone dedicated to non-commercial
51


uses. By mid-2000 the number of DSL connections to the Internet still numbered less
than two million in the OECD area (OECD, 2001b). Worthy of note is that more than
half these subscribers were in Korea. During 2000 Korea witnessed a surge in
broadband subscriptions. By mid-year, Korea Telecom was reporting an increase in
DSL subscribers from 11,925 at the end of 1999 to 545,553 by July 2000.
Logistic Model and Limits
When considering the diffusion of the Internet, it is useful to start with one of the
most influential generic models of the diffusion process, Everett M. Rogers The
Diffusion of Innovation (first published in 1962 and now in its fourth edition).
Furthermore, it is also useful in that the beginning of Internet diffusion is similar to the
s-shaped development.
For Rogers, diffusion can be described as the process by which an innovation is
communicated through certain channels over time among the members of a social
system (Rogers, 1983:12), thus suggesting a contagion effect, or snob and
bandwagon effect is the basis for the innovation-diffusion process. The diffusion
process is influenced by four elements: innovation characteristics, communication
channels, time elapsed since the innovation was introduced, and the social system. 36
36 See Clinton Announces Moves for Improving Access to the Internet Wall Street Journal, October 11,
1996, and Internet 2 Project General Information, available on the Web at
http://www.intemet2.edu/about i2.
52


Based on the three assumptions (1) composition of the social system itself is
considered unchanging, (2) homogeneity is assumed over time, and (3) external factors
are considered irrelevant in the diffusion process the rate of diffusion is modeled
solely as a function of interactions between existing and potential adopters in the social
system. Rogers proposes the bell-shaped frequency curve37 and the s-shaped
cumulative curve (logistic curve) that can be applied to any diffusion (Figure 2-6).
Figure 2-6: Bell-shaped Frequency Curve and S-Shaped Cumulative Curve for Adopter
Distribution
Source: Rogers, (1983: 243)
In this regard, several recent studies show us that Internet diffusion can be
characterized to date by the s-shaped growth patterns (Rutkowski, 1996; Rai et al.,
37 The bell-shaped normal frequency distribution of the adopters is divided into five categories:
innovators (2.5%), early adopters (13.5%), early majority (34%), later majority (34%), and laggards
(16%).
53


1998; Press, 2000). The logistic curve especially echoes in virtually all respects the
Product Life Cycle curve used in marketing circles (Flynn and Preston, 1999). Both
curves posit that in the introductory phase, an innovation/new product initially diffuses
slowly as early adopters begin to purchase it. This is followed by a growth take-off
phase as early adopters begin to convince others of the products value. The third phase
is the maturity phase or late adopters phase where the product reaches and plateaus at
its life cycle peak sales. Finally, the product enters a phase of decline as the market is
saturated.
Since its first publication, this model has been the subject of heated debate, not
least within the interdisciplinary fields concerned with communications and socio-
economic development, where it was criticized for failing to take into account specific
institutional contexts and socio-political power structures (Mattelart and Mattelart,
1998). Furthermore, this model provides little idea as to where the inflection point
might be, that is, where a product might be located on the s-curve at any given time.
Interactive Model and Communications Policy
The utility theory of the interactive communication medium (Gurbaxani, 1990)
suggests that users subscribe to a technology only if it provides a net positive utility.
Eli Noam (1994) has proposed a model that attempts to explain telephone network
diffusion in terms of the relationship between the cost of telephone access and the
utility of the telephone network. Noam proposes the four stages of telephone network
54


diffusion: growth by external subsidy (0~ nj), self-sustained growth (ni~n2), directed
growth (entitlement growth) (n2~n4), and growth by external subsidy (n4~).
IQ
In the first stage, or below a critical mass point (nj), a network will not be
feasible unless subsidized by external sources, either by government or by the network
operators willingness to accept losses in the early phases of operation because where
the network is small, average cost is high and externalities small. Beyond that point the
network will grow on its own.
Through the second stage (ni~n2), which Noam called the cost-sharing phase,
the network users can lower their cost by adding members. However, at some point
average costs increase, and utility plateaus. The private optimum point is n2. Left to
themselves, the existing subscribers of the network would not accept members beyond
that private optimum. From a social point of view, however, social welfare still
increases at n2, because the positive utility to additional network users is not considered
by the existing network participants should they stop expanding at n2 under the
assumption of a network with an equal-price system. The insiders do not take the
outsiders into account.
If the benefits are added, the social optimum n3 lies between n2 and n4. This
implies that, in the third stage (n2~n4), the socially optimal size (n3) will therefore not 38
38 Critical mass is defined as a small segment of the population that chooses to make big contributions
to the collective action, while the majority do little or nothing (Oliver et al., 1985: 524; Pearce, 1998).
Oliver et al. (1985) developed the theory of critical mass to integrate theory on phenomena variously
labeled snob and bandwagon effects, the free rider problem, and the tragedy of the commons
(Markus, 1987).
55


be reached by itself, but by some external governmental direction through required
expansion. It may be achieved by a differentiated pricing scheme or through some
internal politics of expansion, but study will focus on the external governmental
direction.
The final point (ru) lies beyond the point at which the net benefits of the network
will be negative. Beyond that point, or the fourth stage (ilt), the network would again
need outside support to exist (Figure 2-7).
Figure 2-7: Noams Diffusion Model of Telephone Network
mass point optimum optimum point
Growth by Self-sustained Entitlement growth Growth by
External growth (directed growth) External
subsidy subsidy
Source: Noam (1994: 687-704).
56


Figure 2-7 shows that initially high average costs decline rapidly as more
subscribers join the system (i.e. as the utility increases). This fall in costs continues
until the system has expanded to a stage whereby new subscribers can only be added
by accepting an increasing average cost. Utility39 increases dramatically at first
followed by a reduction in the rate of increase as the number of new subscribers begins
to tail off when the telephone system enters its mature stage. This figure also shows
that falling average cost40 leads to an increase in subscriber numbers. Yet at the same
time, because the telephone network is an interactive system, the increase in subscriber
numbers also increases the utility of the system as a whole, generating a further
increase in subscriber numbers.
This model is also relevant to the present conceptual framework in that it deals
with the diffusion process of interactive network and policy implications for each stage
of Internet diffusion.
Summary
Process vs. Causality. Process models have applied different mathematical
formulae to various time-series data, and reached a general agreement that most
diffusion processes are s-shaped curves. Although the s-curve model reveals the
socioeconomic characteristics of innovation adopters, it overlooks such crucial factors
39 Utility measured in terms of subscriber numbers: the more people can be reached, the more useful is
the network.
40 AC can be measured in terms of the basic cost of system entry i.e., rental cost.
57


as government policies in Internet diffusion emphasized by causal theories. Causal
theories, on the other hand, tend to use regression methods to analyze cross-sectional
data, but generally pay little attention to the longitudinal and dynamic processes of
Internet diffusion. Considering the distinctive merits of both process models and causal
theories, a dynamic understanding of policy changes in the diffusion process may
deepen our comprehension of the role of different policy determinants in Internet
diffusion process, and vice versa.
Observations. Through reviewing the process studies on Internet diffusion, the
specific policy pattern in each stage of Internet diffusion can be added to each
dimension of Internet policy.
In the initial stage of Internet diffusion or below a critical mass point,
government initiation of Internet network can be justified because huge amounts of the
initial investment cause a network not to be feasible unless subsidized by external
sources. So, the political dimension of Internet policy is well matched with the initial
stage of Internet diffusion.
In the second stage of Internet diffusion or beyond a critical mass point to a
private optimum point, enhancing the entrepreneurship of service providers through
liberalization can be justified in that the Internet network can achieve self-sustained
growth. Thus, the economic dimension of Internet policy is well matched with the
second stage of Internet diffusion.
58


In the third stage of Internet diffusion or beyond a private optimum point,
government guidance of growth toward enhancing social equity of service users can be
justified in that the Internet network can not achieve the socially optimal size. So, the
social dimension of Internet policy is well matched with the third stage of Internet
diffusion.
Synthetic Comparative Studies
Comparative analysis could be difficult because different political, economic,
and social structures and systems render the analytic comparison of various issue
areas problematic at best, and misleading at worst.
Peter deLeon (1979:10)
Comparative Studies in General
The comparative study can be characterized by its extreme diversity (Peters,
1998), which used to be considered its strength and weakness. On the one hand, the
field of comparative study is so diverse that it can scarcely be called a field at all
(Wiarda, 1986: 5). On the other hand, the heterogeneity of this field will continue, and
this diversity is a source of strength rather than of weakness in that the openness of the
field to various theories and methodologies helps to maintain its vitality and its
capacity to cope with a rapidly changing policy world (Verba, 1986: 36). Considering
the diversity of comparative study, a focus of this review is in the literature that is well
matched with the triad of Internet policy: that is, Richard Roses comparative dynamics,
Theodore Lowis policy typologies, Thomas Dyes causes & consequences of public
59


policy, and Heidenheimer et al.s choices as framework. For the purpose of logical
consistency, Giovanni Sartoris (1994:14-34) basic questions are used, that is, (1) why
compare? (2) what is comparable? (3) compare how?
Comparative Dynamics. Among the other diverse literature on comparative
analysis (e.g., Dierkes et al., 1987; Ashford, 1992; Steinmo et al., 1992; Dogan and
Kazangigil, 1994; Baker, 1994; Peters, 1998; Castles, 1998; Maor and Lane, 1999;
Heady, 2001), Richard Roses (1993:71-76) lessons drawing across time and space is
useful in that his synthetic perspective combining time with space is well matched with
the present thesis. He classifies the methods of comparison into four categories in
terms of both across space (yes or no) and across time (yes or no): routine satisfaction,
trend analysis, comparative statics, and comparative dynamics (Figure 2-8).
Figure 2-8: Methods of Comparison across Time and Space
Across Time
No Yes
Routine Trend
Satisfaction Analysis
Comparative Comparative
Statics: Dynamics:
League tables Lesson-drawing
Source: R. Rose (1993:72)
60


Roses focus is on the comparative dynamics where lesson-drawing occurs about
closing the performance gap by catching up with the performance of leaders. For the
purpose of lesson-drawing, he suggests analyzing in sequence: (1) the initial state in a
society prior to public recognition that a policy need exists, (2) how policy issues are
placed on the agenda of public controversy, (3) how demands are advanced, (4) the
importance of the form of government for policy deliberations, (5) available resources
and existing constraints, (6) the move towards a policy decision, (7) the determinants
of governmental choice, (8) the context of the choice, (9) implementation, (10) the
production of outputs, (11) policy evaluation, and (12) feedback (Rose, 1973). His
analytic framework provides an appropriate guidance to the comparative study in
general.
Why Compare. Regarding the objects of the comparison, Przeworski (1987: 35)
asserts that a consensus exists that comparative research consists not of comparing but
of explaining. The general purpose of cross-national research is to understand.... In
similar fashion, Regin (1987: 6) holds that comparative knowledge provides the key to
understanding, explaining, and interpreting. Especially, Mayer (1989: 12) redefines
comparative studies as a field whose goal is the building of empirically falsifiable,
explanatory theory.
Mayers goal of comparative analysis is closely related to the method of ideal
type that has the virtue of providing a standard against which real world systems can be
61


compared. Its typical examples are Webers (1949) ideal type model, Easton (1965,
1979) system model based on the input-output analysis, and Lowis (1979) policy
typology41 of distributive, regulatory, and redistributive policy. In this thesis, the
method of ideal type is to be used for the purpose of drawing a general pattern of
Internet policy across time and space.
What Compare. With regard to the question of the areas and issues to compare,
Dye (1998:7) presents a useful summary of the complex linkages that operate in the
process of policy-making (Figure 2-9).
According to his framework, the primary concern of this dissertation is
represented by arrows B, C, and D. Regarding arrow B, the effect of government as a
policy system on public policies will be compared and contrasted. As the theoretical
types of policy systems for the mature use of the Internet, this dissertation proposes the
three different policy systems i.e., a national Chief Information Officer (CIO), an
independent regulatory body, and a national and functional ministry in view of the
three dimensions of Internet policy.
First, a national CIO can be ideally matched with the political dimension of
Internet policy in that a national CIO is involved in the allocation of the information
technology in general, or the Internet resources in specific.
41 The main point of criticism of these typojogies is that the categories of policy types are not used
mutually exclusively (Bressers and Honigh, 1986). For example, distributive policy is at the level of
policy goals, whereas regulatory policy is at the level of policy instrument. Nevertheless, the basic
concept of policy typology gives us a useful insight into the type of the policy.
62


Considering the multidisciplinary nature and impacts of the Internet as an
interactive social network, the Internet-related policy issues can be interrelated issues
among the functional ministries in the era of the industrial society. In this respect, a
new organizational scheme for coordination in the information society can be termed
as a national CIO. The examples of institutionalizing this new organizational scheme
are the so-called e-Minister and e-Envoy in the U.K.
In the U.K., the e-Minister has overall political responsibility for the
Government's Information Age agenda, and chairs the so-called Information Age
Ministerial Network purpose of which is to ensure that the social, economic and e-
Govemment strands of the Governments information age programs are combined as
an integrated strategy by developing a shared vision (Chatrie and Wraight, 2000). An
Envoy structure is composed of an e-Envoy and an e-Envoy office that is leading the
drive to get the UK online (i.e., the American firstgov.gov).
Second, an independent regulatory agency can be ideally matched with the
economic dimension of Internet policy in that an independent regulatory body is
inherently involved in the (de) regulation of the service providers considering its own
origin of birth with the logic of the separation of regulatory function from the policy
function.
Third, a traditional and functional ministry can be ideally matched with the
social dimension of Internet policy in that the protection of consumer and social equity
of the public is considered a policy function of the traditional ministry.
63


Regarding arrow C, both the effect of the economic input from the market on
Internet policy (the economic dimension of Internet policy) and the effect of the social
input from the community on Internet policy (the social dimension of Internet policy)
will be compared and contrasted across time and space.
Regarding arrow D, the effectiveness of policy determinants of each dimension
of Internet policy on Internet diffusion will be compared and contrasted across time
and space.
64


Figure 2-9: Causes and Consequences of Public Policy
Society
Political System
Public Policy
Including:
Wealth and income
Inflation, recession,
Unemployment
Educational achievement
Environmental quality
Poverty
Racial composition
Religious and ethnic make-up
Health and longevity
Inequality, discrimination
Including:
Federalism
Separation of powers
Checks and balances
Parties
Interest groups
Voting behavior
Bureaucracy
Power structures
Congress, president,
Courts
Including:
Civil rights
Educational policies
Welfare policies
Health care policies
Criminal justice
Taxation
Spending and deficit
Defense policies
Regulations
Linkage A: What are the effects of social economic conditions on political and
governmental institutions, processes, and behavior?
Linkage B: What are the effects of political and governmental institutions, processes,
and
behaviors on public policies?
Linkage C: What are the effects of social economic conditions on public policy?
Linkage D: What are the effects (feedback) of public policies on social economic
conditions?
Linkage E: What are the effects (feedback) of political and governmental institutions,
processes, and behavior on social and economic condition?
Linkage A: What are the effects (feedback) of public policies on political and
governmental institutions, processes, and behavior?
Source: Dye (1998: 7)
65


Compare How. With relation to the choice of the criteria of the cross-time-and-
space comparison, Heidenheimer et al.s (1990) choice as framework gives us useful
criteria of policy comparison (deLeon and Resnick-Terry, 1999). They first define
public policy in terms of social choice, and the effects that those choices have on
individuals. They then marshall a great deal of evidence about the differences among
policy areas in the several countries and are able to demonstrate some consistent
patterns in the choices made (Heidenheimer et al., 1990; Peters, 1998).
They argue that four types of choices are applicable to each policy area: choice
of scope, choice of policy instruments, choice of distribution, and choice of restraints
and innovation. Choice of scope concerns whether and where lines shall be drawn
between public and private responsibilities. In other words, this type of choice is
related to the rationale for governmental intervention. Choice of policy instrument
includes the policy agencies to implement the specific policy program. Choice of
distribution concerns the policy effectiveness from the perspective of policy typologies.
Choice of restraints and innovation can be closely related to the change of policy
environments and the subsequent policy shift. These criteria can provide a useful
perspective for formulating the analytic framework of this dissertation.
Comparative Studies and Communications Policy
In the communication sector, cross-national comparison of communications
policy is a commonly used method in that the diffusion of communication technology
66


is an inherently worldwide phenomenon. In this regard, the four literature sources on
communications policy of which the theoretical and conceptual frameworks as well as
the methodologies can give useful implications for the present dissertation are
reviewed briefly.
The theme of this dissertation is closely related to the classical trade-off
relationship between efficiency and equity, market and government, private and
public, economics and politics, or democratization and liberalization (Dahl and
Lindblom, 1953; Okun, 1974; Stone, 1997; Wolf, 1993; Rainey, 1997; Carey, 1987).
Such a traditional trade-off relationship of the policy goals may be closely linked with
the discrepancy of the policy instruments between universal service policy and
competition policy in the telecommunications and Internet sectors. In this respect, a
brief look is focused on four dissertation studies that deal with similar issues but
different methods (two econometric analyses vs. two case studies).
K. P. Jayakar (1999) studied the effect of telecommunications reform on
universal service, in both its network growth and distributional equity dimensions. The
three main macro-systemic variables affected by telecommunications reforms were
identified for the purpose of the econometric analysis: industry structure (the nature
and extent of competition), ownership (public, private, corporatized), and the
independence of the main regulatory institution. To measure universal service, two
indicators of universal service teledensity and residential teledensity share were
used. Based on the statistical multiple regression model, this dissertation analyzes the
67


causal relationship between reform and universal variables with the data of the selected
24 countries such as rich countries (U.K., Japan, France, Canada, and German) and
poor countries (Sri Lanka, Jamaica, Thailand, Venezuela, etc.). The findings indicate
that less developed countries may benefit more from telecommunications reform than
their richer counterpart.
S. H. Lee (1999) quantitatively investigated how the social and institutional
endowments of the countries in the Asia-Pacific region affected the performance of the
telecommunication sector (such as network expansion and efficiency) by using an
economic model and assumptions based on transaction cost economics. As an
empirical framework, a descriptive overview of general trends of telecommunications
reforms undertaken in the Asia-Pacific region was made in view of organization theory.
This dissertation uses a regression model to analyze statistically the political, economic,
and demographic characteristics for 15 countries in the region over the period 1982-
1995 in terms of level of network penetration, network efficiency, and growth rate of
network penetration. The main findings are: (1) a country with fewer opportunities for
rent seeking behavior is more likely to achieve a rapid growth rate of network
penetration, and (2) governments ability to establish institutional constraints on the
exercise of executive power has a positive impact on the growth of network penetration.
Based on its theoretical framework on political economy, J. Y. Kim (1999)
examined the broadly-defined impacts of telecommunications deregulation on two
significant public interests, universal service and free speech on the Internet. This
68


dissertation selects the relevant experiences of the United States, Germany, and Japan
as main areas of research concern because they represent different paths of
telecommunications evolution. Based on the qualitative case studies, the findings are
then synthesized and discussed in a way that re-frames the role of the state in a
deregulated communication environment. The findings are that all three nations have
attempted to achieve their universal service goals by creating a competitive
marketplace. It also indicates that all three nations have different views on how to
guarantee free expression on the Internet: a hands-off policy of the U.S.; a top-down
approach in Germany; and a bottom-up approach in Japan.
Comparing the United States and France, A. L. Fletcher (1997) analyzed the
effect of divergent telecommunications regimes on public access to information. The
American telecommunications regulatory framework relies on market incentives and
private ownership, while the state-led French model is based on public ownership and
centralized regulatory authority. Indices of telecommunication efficiency support the
argument that each model can produce similar outcomes. This dissertation analyzes
whether these disparate regulatory models also produce comparable levels of access to
information or if this outcome varies. Access to information refers inclusively to both
access to the telecommunications network and to information services carried by the
network. By using a comparative analysis of binary cases, data for this dissertation are
both qualitative and quantitative and this dissertation includes statistical data, primary
source documents (such as telecommunications reforms laws) and archival material.
69


The main findings are that the state has an important function to perform in the
research and development of information technologies, the creation of an initial
demand for advanced information services, and the promotion of nonprofit and civic
applications of information technology.
Summary
Through reviewing the comparative studies of Internet diffusion, the specific
types of policy system can be added to each dimension of Internet policy: that is, a
national CIO can match with apolitical dimension of Internet policy, a independent
regulatory agency can match with a economic dimension of Internet policy, and a
functional ministry can match with the social dimension of Internet policy.
Furthermore, considering the limits of both the pure synchronic causal studies
and the pure diachronic process studies of Internet diffusion, this section suggests that
a synthetic comparative dynamics, combining the synchronic causal studies with the
diachronic process studies of Internet diffusion, can be adopted as a research method.
70


CHAPTER 3
ANALYTIC FRAMEWORK AND METHODOLOGY
Based on the literature review, this chapter develops a triad of Internet policy as
an analytic framework composed of both a three-dimensional (3-D) synchronic model
and a three-stage (3-S) diachronic model, and proposes five research propositions to
explain a dynamic relationship between communications policy and Internet diffusion
over time. These propositions are to produce an explanatory theory about a synchronic
style and a diachronic pattern of Internet policy, which is formulated to provide
guidance to conduct a cross-national and pattern-matching comparison and evaluation
across time and space.
Analytic Framework and Propositions
Three Dimensional f3-D) Synchronic Model
Initially grounded on the observation of the first theoretical controversy
(monocausal vs. multicausal theories), a synchronic model of this dissertation explains
the causes and consequences of Internet policy that can be termed a policy style
(Richardson, 1982). The term of Policy style is used in this dissertation for the purpose
of showing comparative and synchronic characteristics of Internet policy in a nation,
71


and then portrayed as having three dimensions: economic, political, and social. To
resolve the above-mentioned policy controversies of Internet diffusion, the three key
dimensions of Internet policy identified from the existing literature are to be focused.
Furthermore, for the purpose of the across-time-and-space comparison and evaluation,
they need to be adjusted from the perspective of the system model of public policy
based on Eastons input-output model and Dyes policy analysis model of causes and
consequences (Figure 3-1). Figure 3-1 shows that the 3-D synchronic model focuses on
the three dimensional inputs from state (politics), market (service providers), and
community (service users) and the three dimensional outputs (distributive, de-
regulative, and re-distributive).
Figure 3-1: System Model of Internet Policy
72


Three-Stage f3-S) Diachronic Model
To complement the static design of the above synchronic model of Internet
policy, a diachronic model is designed to show the dynamic changes of the policy
pattern. The term of policy pattern is used in this dissertation for the purpose of
showing the comparative and diachronic characteristics of Internet policy in the
Internet diffusion process. The diachronic model is grounded on the observation of the
second policy controversy (process vs. causal models) in the literature review.
To resolve the second controversy, the two process models identified from the
existing literature need to be further refined to formulate empirically testable
propositions under the guidance of a consistent analytic framework of a triad of
Internet policy. For this purpose, the logistic model and Noams model can be
combined (Figure 3-2).
Figure 3-2: Combined Interactive Process Model
73


Figure 3-2 shows the main reason why the two process models are combined.
The policy implications of Noams model are drawn from the interaction between
subscribers (or hosts) and dollars, while the s-shaped diffusion trends of interactive
medium of Rogers model are drawn from the interaction between subscribers (or
hosts) and time. Therefore, for the purpose of getting the policy implications of Internet
diffusion across time, the two models need to be combined to show the interaction
between subscribers (or hosts) and time. The combined diachronic model can thus
show the policy implications in each stage of Internet diffusion.
The combined interactive model can be applied to Internet diffusion in that the
diffusion of the Internet has already shown itself an initial s-shaped pattern since its
creation. In addition, Noams final stage (n4~) will not be examined in this dissertation
in that the policy implications of the final stage simply mirror or replicate that of the
first stage.
Furthermore, a short review of the historical development of Internet technology
shows that the chronological evolution of Internet technologies manifests itself as an
on-going process and thus the adequate governmental response to the technological
evolution is not policy termination but policy succession in an interactive medium
(Hindle et al., 1997). For instance, for the purpose of facilitating the technological
evolution of the Internet from the second generation of Internet (like dial-up or
narrow-band connections) to the third generation of Internet (like Internet II or
74


broad-band Internet), what is needed by a government for successful technological
transformation is not exit but an active initiation (OECD, 2001b).
Therefore, in this dissertation, the three stages of Internet diffusion initiation
(0~ ni), proliferation (ni~ n2), and control of proliferation (n2~ n4) can be
distinguished by the critical mass point (ni), private optimum point (n2), and exit point
(ru) respectively. Furthermore, a diachronic 3-D model of this thesis can provide the
policy implications both in each stage and of policy system (i.e., an external funding
and a national CIO in the first stage, a self-sustained growth and an independent
regulatory agency in the second stage, and a social control and a functional ministry in
the third stage).
Research Propositions
Under the guidance of the 3-D synchronic and 3-S diachronic models, we make
five hypothetical propositions addressing the dynamic relationship of communications
policy to Internet diffusion across time and space for the purpose of deriving both
policy implications and the appropriate role of government in each stage of Internet
diffusion.
Policy Style and Policy Pattern. In view of the fact that the Internet market gets
its problem of the commons (Brin, 1995; MacKie-Mason and Varian, 1996) because of
its inherent nature of public goods and CPRs, the over-arching proposition is that a
75


policy style (Dye, 1998; Lowi, 1979) and a policy pattern (Rogers, 1983; Noam, 1994)
of communications policy affecting Internet diffusion can be generalized across time
and space (Rose, 1993).
Proposition 1: A synchronic policy style of communications policy affecting
Internet diffusion in a nation can be generalized in view of its
policy input, policy system, and policy output.
Proposition 2: A diachronic policy pattern of communications policy affecting
Internet diffusion in a nation can be generalized in view of each
stage of Internet diffusion.
In addition, the second research proposition can be divided into three sub-
propositions in view of the three stages of Internet diffusion.
Political Dimension and Initiation.
Political Imperative. Both the huge amounts of initial investment on Nil and its
characteristics of natural monopoly and externalities caused by the unique nature of
the social network as public goods are hypothesized to bring government intervention
and initiatives to digitize the existing network in the first stage of Internet diffusion.42
The system of information superhighways with government support during the 1990s
used to be analogous to the interstate highway system built during the 1950s with
42 Interestingly enough, Hanaro Telecom, the second private-owned local telephone company in Korea
initiated the Digital Subscriber Lined (DSL) network with the non-economic willingness to accept the
loss in 1999, and now the market share in local service in 2001 is just 1.8 percent of the total (MIC,
2001). In this dissertation, such an initiation by the network operators non-economic willingness to
accept loss is to be set aside as a given.
76


government support in the U.S. (Shew, 1993). So, the prominent before
marketization issues and sequential government initiatives affect Internet diffusion in
the first stage of Internet diffusion.
National Information CIO and Distributive Policy. In this stage of Internet
diffusion, a higher level of a national information CIO can affect Internet diffusion by
envisioning the national goal of building the so-called information society and by
initiating a national backbone plan such as a national information infrastructure (Nil).
This goal can be achieved by policy instruments of investment or subsidies to Nil,
which can be classified into a distributive policy in view of Lowis policy typology.
Research Proposition. In the first stage of Internet diffusion (initiation), before
marketization issues, such as the huge amounts of initial investment and natural
monopoly, and network externalities of Internet service as public goods, are posited to
generate government initiative and intervention. So, political imperative from the state
(political goal-setting of achieving an information society), and its subsequent
distributive policy by a national CIO affect Internet diffusion in the first stage of
Internet diffusion.
Proposition 2a: Political imperative from the state (politics) and its distributive
policy by a national CIO for mature use of the Internet are
the comparative characteristics in the first stage of Internet
diffusion.
77


Economic Dimension and Proliferation.
Economic Imperatives. In this stage, marketization issues, such as the network
congestion caused by the undersupply of Internet service as common pool resources,
are posited to result in privatization and competition. Privatization and competition are
necessary so that the fixed-line and wireless telecommunications infrastructure can
support and encourage the growth of Internet traffic (Mann, 2000). As such, economic
imperatives from the market, especially from the supply-side service providers for the
development of the Internet infrastructure, are expected to affect Internet diffusion in
the second stage of Internet diffusion.
Independent Regulatory Bodies and (de)regulative Policy. In this stage,
policy programs of liberalization (privatization, competition, and price regulation)
which can be classified into a (de)regulative policy in view of Lowis policy typology
can affect Internet diffusion. Furthermore, an independent regulatory body used to be
involved to secure a fair environment for the competition. The economic policy goal
and instruments are traditionally identified as the best policy determinant of Internet
diffusion in many causal analyses. To achieve the economic goal of creating national
wealth by developing an efficient technology, the policy instruments of competition
and privatization are frequently used as the indicators of economic deregulation.
Furthermore, considering the independent influence of local call pricing on Internet
diffusion, the structure of local call charges needs to be added. The indicator of
78


competition is the level of competition in the provision of public service telephone
networks, especially in local service (OECD, 1999a) (Appendix D). The key indicator
of privatization is the level of privatization of major public telecommunications
operators (PTOs) (OECD, 1999a) (Appendix E). The indicator of pricing structure is
the PSTN charging structure on local telephony (Appendix F): that is, whether it is
metered or unmetered (flat rate).
Research Proposition. In the second stage of Internet diffusion (proliferation),
marketization issues, such as the network congestion caused by the undersupply of
Internet service as common pool resources, are proposed to result in privatization and
competition. So, economic imperative from the market and its subsequent de-regulative
policy by an independent regulatory agency affect Internet diffusion in the second
stage of Internet diffusion.
Proposition 2b: Economic imperative from the market (service providers) and
its de-regulative policy by an independent regulatory agency
for the mature use of the Internet are the comparative
characteristics in the second stage of Internet diffusion.
Social Dimension and Control of Proliferation.
Social Imperatives. In this stage, after marketization issues of narrowing
digital divide is supposed to bring the call for more governmental involvement in the
Internet and its application. The social optimum is hard to be achieved by the market
forces via the invisible function of the price due to the technological traits of the
79


interactive Internet, as seen in Noams process model. This is one of the primary
reasons why many arguments call for greater government involvement in the Internet
and its applications in the stage of after marketization (Noam, 1994; Westphal &
Towell, 1998; Shapiro, 1999; Lessig, 1999). Government involvement will focus on
the demand side of the Internet market, that is, achieving the social optimum of the
user of Internet service.
National Ministries and Re-distributive policy. In this stage, the policy program
of enhancing public access to the Internet that can be classified into a redistributive
policy in view of Lowis policy typology can affect Internet diffusion. Furthermore,
national ministry as a policy maker used to be involved. The social goal of increasing
social equity via the Internet can be achieved by the policy instruments of the subsidies
on universal service. The indicator of the subsidies on universal service is whether or
not there is a universal service funding mechanism or direct funding for universal
access (OECD, 2000b) (Appendix G).
Research Proposition. In the third stage of Internet diffusion (control of
proliferation), after marketization issues of narrowing digital divide are supposed to
bring the call for more governmental involvement on the Internet and its application.
Therefore, social imperative from the community and its subsequent re-distributive
80


policy by a functional ministry affect Internet diffusion in the third stage of Internet
diffusion.
Proposition 2c: Social imperative from the community (service users) and its
de-regulative policy by a functional ministry for the mature
use of the Internet are the comparative characteristics in the
third stage of Internet diffusion.
Summary
This dissertation developed the 3-D and 3-S analytic frameworks that is
composed of the several trichotomous variables of Internet policy in view of their
synchronic 3-D causes and consequences and their diachronic 3-S changes of the
times: that is, (1) major variables (network, service providers, and users), (2) allocation
mechanisms (state, market, and community), (3) theoretical dimensions (political,
economic, and social), (4) system styles (national CIO, independent regulatory agency,
and functional ministry), (5) policy styles (distributive, de-regulative, and re-
distributive), and (6) policy stages (initiation, proliferation, and control of proliferation).
The hypothetical relationship among these variables can be summarized as in
Figure 3-3 and identified as in the three dimensions.
81


Figure 3-3: Theoretical Policy Styles and Patterns
Policy Input Policy System Policy Output Internet
Political Imperative Distributive Policy Stage 1
Huge Amount of Initial Investment - National Information - National Planning - Before
Achieving an Information Society CIO Investment Marketization (Network)
Economic Imperative - (De) regulative Policy Stage 2
Government Failure Independent
Positive Network - Regulation - Competition ->
Externalities Agency Privatization Marketization
Network Congestion Price regulation (Providers)
Social Imperative Re-distributive Policy Stage 3
Market Failure
Negative Network Externalities Ministry - Universal Service After Marketization
Digital Divide (Users)
Policy Patterns (Vertical Line, Across Time, Diachronic Characteristics)
^ Policy Styles (Horizontal Line, Across Space, Synchronic Characteristics)
In the political dimension, political imperative from the state (national planning
of Nil) and subsequent policy program of investment in Nil that can be classified into a
distributive policy in view of Lowis policy typology can affect Internet diffusion. In
this dimension, a higher level of national planner such as a national Chief Information
Officer (CIO) used to be involved to project a national route to an information society.
82


In the economic dimension, the imperative from the market (especially, from the
supply-side service providers) and subsequent policy program of liberalization
(privatization, competition, and price deregulation) that can be classified into a
(de)regulative policy in view of Lowis policy typology can affect Internet diffusion. In
this dimension, the independent regulatory body is involved to secure a fair
environment for the competition.
In the social dimension, the imperative from the community (especially, from the
demand-side service users) and subsequent policy program of enhancing public access
to the Internet that can be classified into a re-distributive policy in view of Lowis
policy typology can affect Internet diffusion. In this dimension, a national ministry as a
policy maker is involved to promote social equity.
Research Methods
By choosing a method of comparative dynamics, a preferred pattern of Internet
policy over time is tested across space. For this purpose, the three nations are chosen as
a most different system design: the U.S., Finland, and Korea. The scope and the focus
of this study are on the three synchronic dimensions (3-D) of Internet policy and the
three diachronic stages (3-S) of Internet diffusion. Furthermore, a cross-national
comparison and evaluation is conducted based on a pattern-matching strategy. For this
purpose, the pattern variables of Internet policy are also identified in terms of four
criteria for the comparison; that is, choice of scope (policy rationale), choice of policy
83


instruments (policy agency and program), choice of allocation (policy effectiveness),
and choice of restraints and innovation (policy shift).
Choice of the Method:
Comparative Dynamics
Under the guidance of the 3-D and 3-S frameworks and the theoretical
propositions, a generalized pattern of Internet policy across time is to be empirically
tested across space using a comparative policy analysis strategy. By using the method
of comparative dynamics (Rose, 1993) in a sense that a general pattern of
communications policy for the mature use of the Internet is tested across time and
space, this dissertation intends to propose descriptive theory to understand, explain,
and interpret the Internet-related policy controversies (Mayer, 1989: Regin, 1987).
Furthermore, this thesis utilizes a comparative case method as a distinctive form
of multiple-case studies (Lijphart, 1975; Agranoff & Radin, 1991). The present
dissertation explores a case study method because the boundaries between
phenomenon and context are not clearly evident in that Internet-related phenomena
are multidisciplinary in nature and the researcher deliberately wants to cover
contextual condition articulated in the 3-D and 3-S frameworks (Yin, 1994: 13).
According to Yins types of case study, this thesis is both exploratory and descriptive
in that it tries to build a theory to describe the contemporary Internet-related policy
controversies.
84


Choice of Nations:
Most Different System Design
The present dissertation selects the three relevant cases of the U.S., Finland, and
Korea as the most different system design to test whether or not a theory about a
pattern of Internet policy across time can be generalized across space.
Thus, among the nations with high Internet performance (hosts or subscribers)
showing the different roads to the mature use of the Internet, each case is
systematically selected in view of both driving imperatives (or motives) from the three
major players and policy styles of Internet policy so that it produces contrasting results
but for predictable reasons (a theoretical replication) about the comparative motives
and styles of Internet policy (Yin, 1994).43
From the perspective of the comparative characteristics of the policy
surroundings conditioning communication policy for the mature use of the Internet, an
individualistic and pluralistic policy surroundings of the U.S. can be compared and
contrasted with the communitarian and deliberative policy surroundings of the Finland
(Venturelli, 1995). The driving motive and the policy style for achieving an
information society in the U.S. can be characterized by a market-driven strategy based
on a trust in the free market ethos, while those of Finland can be characterized by a
society-driven strategy based on a trust in the welfare society ethos. In addition, in
Korea, sharing the same policy surroundings of the Confucian cultural traditions with
43 According to Yins (1994: 45-46) the replication logic, analogous to that used in multiple
experiments, is the same as the logic underlying the use of multiple-case studies. For more information
about the replication logic, see Hersen, M & D.H. Barlow (1976).
85


the other far eastern Asian nations like Japan and Singapore, the driving motive and the
policy style can be characterized by a government -driven strategy based on a trust in
the strong state ethos.
Therefore, the different driving motives and the styles of Internet policy for
achieving an information society in each of the three nations can be compared and
contrasted the free market ethos and its subsequent policy style of the U.S., the
welfare society ethos and its subsequent policy style of Finland, and the high state
ethos and its subsequent policy style of Korea. The systematically-chosen three nations
are matched with the triad of Internet policy. The synchronic differences of the
imperatives and policy styles of Internet policy in the three nations are to be further
investigated in chapter 4.
Choice of Focus and Scope:
3-D & 3-S
The field of inquiry must be narrowed to focus on the manageable three sets of
variables of Internet policy, that is, political, economic, and social dimensions. More
specifically, the causes and consequences of Internet policy and the implementing
agencies and the specific programs of Internet policy in the three nations are to be
compared and contrasted in each dimension of Internet policy. According to Dyes
diagram (1998: 7), the primary focus is on (1) the affect of government as a policy
system on Internet policy, (2) the affect of the economic input from the market on
86


Full Text

PAGE 1

INTERNET DIFFUSION AND COMMUNICATIONS POLICY: A COMPARISON OF POLICY STYLES AND PATTERNS ._ ... ; IN THE UNITED STATES, FINLAND, AND KOREA by Byung W. Kwon . Sogang University, 1989 M.A., Seoul National University, 1993 A thesis submitted to the University of Colorado at Denver in partial fulfillment of the requirements for the degree of Doctor of Philosophy Public Affairs 2002

PAGE 2

by Byung W. Kwon AJJ rights reserved.

PAGE 3

This thesis for the Doctor of Philosophy Degree by Byung W. Kwon has been approved by Date

PAGE 4

Kwon, Byung W. (Ph.D., Public Affairs) Internet Diffusion and Communications Policy: A Comparison of Policy Styles and Patterns in the United States, Finland, and Korea Thesis directed by Professor Peter deLeon ABSTRACT This thesis investigates policy styles and patterns for the advanced use of the in the context of global development toward an information society. Theoretically, this dissertation seeks to fill a gap in the literature by combining the synchronic analysis with the diachronic analysis oflnternet diffusion. Practically, this thesis suggests the appropriate governmental role for the dissemination of the Internet. Based on synthesizing_ the state of knowledge on both communications policy arid Internet diffusion, this thesis develops an analytic frameworkcomposed of both synchronic and diachronic models. Five propositions regarding the relationship of communications policy to Internet diffusion are proposed utilizing their political, economic, and social dimensions. They are tested via a pattern-matching comparison, which is conducted using the three stages of Internet diffusion initiation, proliferation, and control of proliferation. For this purpose, the United States, Finland, and Korea are selected and compared as the most different systems. The several findings and subsequent arguments of this dissertation can be summarized: First, the Internet service can be theoretically identified as displaying characteristic known as the "commons," which can serve as a rationale for governmental involvement in Internet diffusion. Second, a key success factor of iv

PAGE 5

Internet diffusion in the three nations as high IT intensity users is an explicit government policy in each developmental stage oflnternet diffusion, but not non regulation of government. Third, the economic logic ofliberalization that competition is supposed to facilitate the investment in the alternative networks and to enhance social welfare of users cannot adequately function in the local communications market because of the distorted pricing mechanism by non-market considerations. Fourth, the most significant factor for Internet diffusion is not the movement toward the deregulation in communications sector but the contingent policy mix of (I) political national planning for digitizing the network in initial stage, (2) economic liberalization for enhancing efficiency and entrepreneurial activities of providers in proliferation stage, and (3) social control for enhancing equity of users in control of proliferation stage. In conclusion, government involvement as a "fine tuner" of three dimensions in each stage of Internet diffusion can lead to the end stage of Internet diffusion. This abstract accurately represents the content ofthe candidate's thesis. I recommend its publication. Signed Peter deLeon v

PAGE 6

DEDICATION Dedicated to my parents for their unfaltering understanding and great sacrifice.

PAGE 7

ACKNOWLEDGMENT I wish to express my gratitude to the chairman of my committee, Dr. Peter deLeon, for his heartfelt guidance and never-ending patience. I also wish to thank the other members of my committee, Dr. Robert Gage, Dr. M. J. Moon, Dr. Ralph Longobardi, and Dr. Jennifer Oetzel, for their wise advice and helpful comments. I especially want to thank my wife, Sun-Ju (Sunny) Chung, and my son, Tae Hun (Tenny), for their understanding and support during my studies. The opinions expressed in this thesis are the author's personal views and do not necessarily reflect those of the organization to which the author is affiliated.

PAGE 8

CONTENTS Figures .............................................................................................................. xiii Tables ............................................................................................................... xiv CHAPTER 1. INTRODUCTION: INTERNET AND GOVERNMENT. ............................. Research Problems and Justification ......................................................... I Research Goals and Questions .................................................................. 6 Chapter Outline ......................................................................................... ? 2. LITERATURE REVIEW ........................................................................... 10 Internet as Commons ......................... ..................................................... 11 Internet and Characteristics ................................................................ 11 Nature ofinternet Service .................................................................. l4 Allocation Mechanisms and Internet Policy ...................................... I? Three Theoretical Approaches ................................................................ 23 Economic Approach: Market Model ................................................. .23 Political Approach: Government Model. ........................................... 26 Social Approach: Community Model ................................................ 28 Summary ............................................................................................ 32 Synchronic Causal Studies ...................................................................... 34 Internet Diffusion and Indicators ...................................................... .34 Diffusion Determinants in General .................................................... 37 Diffusion Determinants and Communications Policy ........................ 39 Summary ............................................................................................ 47 Diachronic Process Studies .................................................................... .48 Technological Development of Internet ............................................ .48 Logistic Model and Limits ................................................................. 52 Interactive Model and Communications Policy ................................. 54 Summary .. .......................................................................................... 57 Synthetic Comparative Studies ............................................................... 59 Comparative Studies in General... ..................................................... 59 Comparative Studies and Communications Policy ............................ 66 viii

PAGE 9

Summary ............................................................................................ 70 3. ANALYTIC FRAMEWORK AND METHODOLOGY .............................. 71 Analytic Framework and Propositions ..................................................... 71 Three Dimensional (3-D) Synchronic Model... .................................. 71 Three Stage (3-S) Diachronic Model... ............................................... 73 Research Propositions ......................................................................... 75 Summary ............................................................................................. 81 Research Methods ..................................................................................... 83 Choice of Method: Comparative Dynamics ....................................... 84 Choice ofNations: Most Different System Design ............................ 85 Choice of Focus and Scope: 3-D & 3-S ............................................. 86 Choice of Mode of Analysis: Pattern Matching Comparison ............ 88 Choice of Criteria: System Analysis and Choice as Framework ....... 89 Data Collection and Manipulation ........................................................... 91 Data Collection ................................................................................... 91 Data Manipulation .............................................................................. 93 Limitations ............................................................................................... 96 4. NATIONAL POLICY STYLES AND INTERNET DIFFUSION ............... 98 Three Nations and Internet Diffusion ...................................................... 98 .Global Diffusion oflnternet. ............................................................... 99 Always-on National Group and United States .................................. I 00 Scandinavian National Group and Finland ....................................... I 04 East Asian National Group and Korea ..................... ; ........................ I 08 Internet Diffusion in Three Nations .................................................. ll2 Policy Styles and Internet Diffusion ...................................................... ll2 De-regulative Model: United States Case ......................................... ll3 Re-distributive Model: Finland Case ................................................ ll9 Distributive Model: Korea Case ........... ; ........................................... l22 Summary ........................................................................................... l26 High IT Intensity and Internet Diffusion ............................................... l27 High IT Intensity Group .................................................................... 128 Industrial Distribution of IT Employment ........................................ l28 Research and Development (R&D) .................................................. l29 Innovativeness of IT Sector .............................................................. l29 IT Networking .................................................................................. l30 Summary and lmplications ............................................................... l32 5. POLITICAL INITIATION AND INTERNET DIFFUSION .................... l34 ix

PAGE 10

Political Imperative: Imperative to Distribute ....................................... l35 Theoretical Policy Rationale: Political Distribution ...................... .l36 Individual Policy Rationale ............................................................ l37 Summary ......................................................................................... l43 Policy Programs and Agencies .............................................................. 144 Theoretical Policy Programs and Agencies ..................................... l45 Individual Policy Programs and Agencies ...................................... l45 Summary .......................................................................................... I 56 Policy Impact: Political Initiation and Internet Diffusion ..................... I 57 Nil and Internet: Digitizing Networks ............................................. I 58 Policy Relevance: Nil and Telecommunications Networks ........... ] 59 Nil and Initiation ............................................................................. 161 Summary and Implications .................................................................... 163 6. ECONOMIC DE-REGULATION AND INTERNET DIFFUSION .......... 166 Economic Imperative: Imperative to De-regulate ................................. 167 Theoretical Policy Rationale: Economic De-regulation ................. .167 Individual Policy Rationale ............................................................. 168 Summary ......................................................................................... l74 Policy Programs and Agencies .............................................................. 175 Theoretical Policy Programs and Agencies ................................... 176 Individual Policy Programs and Agencies ...................................... 178 Summary ......................................................................................... 195 Policy Impact: Economic De-regulation and Internet Diffusion ........... 197 De-regulation and Internet: Two Routes to Internet Taking Off .... .l97 Policy Relevance ............................ ................................................ 198 De-regulation and Proliferation ....................................................... 202 Summary and Implications .................................................................... 204 7. SOCIAL CONTROL AND INTERNET DIFFUSION .............................. 207 Social Imperative: Imperative to Re-distribute ...................................... 208 Theoretical Policy Rationale: Social Re-distribution ...................... 208 Individual Policy Rationale ............................................................ .209 Summary ......................................................................................... 216 Policy Programs and Agencies .............................................................. 217 Theoretical Policy Programs and Agencies ..................................... 2 I 8 Individual Policy Programs and Agencies ....................................... 218 Summary .......................................................................................... 226 Policy Impact: Social Control and Internet Diffusion ............................ 227 X

PAGE 11

Universal Service and Internet: Solving the Digital Divide ............ 228 Policy Trends toward Social Control: New Universal Service and Digital Divide .................................... .232 Universal Service and Control ofProliferation ............................... 237 Summary and Implications ................................................................... .239 8. CONCLUSION: REVIEW OF POLICY OBSERVATION ..................... .243 Summary ofFindings ............................................................................ 244 Technological Factors ...................................................................... 244 Policy Styles .................................................................................... 246 Policy Patterns ................................................................................. 248 Pattern Matching Comparison ......................................................... 253 Policy Observations ......................................................................... 258 Policy Recommendations and Government Role ................................. .261 Policy Recommendations ................................................................ 261 Government Role in Each Stage ...................................................... 266 Discussion of Framework and Findings ............................................... .269 Theories and Analytic Framework (3-D & 3-S) ............................. .269 Validity ofFindings ........................................................................ .272 Future Academic Study ......................................................................... 274 Theoretical Policy Systems in Information Society ......................... 274 Third Way in Communications Sector ............................................ 276. Unbundling and High Speed Internet Access .................................. 277 Post and Press .................................................................................. 278 Utilizing 3-D & 3-S .......................... ............................................... 279 Closing Remarks .................................................................................... 279 APPENDIX A: INTERNET HOSTS PER 1000 INHABITANTS ........................................ .281 B: WEB SITES IN OECD .................................................................................. 282 C: SECURE SERVERS PER I MILLION INHABITANTS ............................ .283 D: INTERNET SUBSCRIBERS (JANUARY 2000) ........................................ .284 E: TELECOMMUNICATION FACILITIES COMPETITION IN OECD ....... .285 F: MAJOR TELECOMMUNICATIONS OPERA TORS IN OECD ............... .286 G: CHARGING STRUCTURE FOR RESIDENTIAL USERS IN OECD ......... 288 H: REGULATION ON UNIVERSAL SERVICE (US) .................................... .289 xi

PAGE 12

I: INTERNET HOSTS AND INTERNATIONAL DIGITAL DIVIDE .......... .291 1: NON-OECD NATIONS WITH LARGEST INTERNET HOSTS ............... 292 K: BROADBAND ACCESS TO INTERNET IN OECD ................................. .293 L: KOREAN PLAN FOR NKN-G .................................................................... 294 M: KOREAN PLAN FOR NKN-P .................................................................... .295 N: COMMUNICATIONS SERVICE IN THREE NATIONS (1991-2000) ..... .296 0: INTERNET DIFFUSION IN THREE NATIONS (1991-2000) ................... 297 GLOSSARY .............................................................................................................. 298 REFERENCES .......................................................................................................... 303 xii

PAGE 13

FIGURES Figure 2-1 Communications Policy Associated with Traditional Industries ...................... ;.14 2-2 Classification of Nature of Goods and Service ................................................... 16 2-3 Triad of Internet Policy ....................................................................................... 23 2-4 Three Dimensional Approaches to Internet.. ....................................................... 33 2-5 Growth of Internet Hosts: 1989-1997 ................................................................. 51 2-6 Bell-Shaped Frequency Curve and S-Shaped Cumulative Curve for Adopter Distribution .......................................................................................... 53 2-7 Noam's Diffusion Model ofTelephone Network ............................................... 56 2-8 Methods ofComparison across Time and Space ................................................ 60 2-9 Causes and Consequences of Public Policy ........................................................ 65 3-1 System Model of Internet Policy ....................................................................... 74 3-2 Combined Interactive Process Model ................................................................. 73 3-3 Theoretical Policy Styles and Patterns ................................................................ 82 4-1 Synchronic Triad of Internet Policy in Three Nations ...................................... 133 8-1 Policy Patterns and Internet Users in Three Nations ......................................... 257 8-2 Policy Styles and Patterns in Three Nations ..................................................... 260 xiii

PAGE 14

TABLES Table 2-1 Comparison of Telecommunications and Internet .............................................. 12 3-1 Government Reports on NII ................................................................................ 94 4-1 Internet Hosts per 1000 Inhabitants in Scandinavian Nations ......................... .l07 4-2 Telecommunications Access (Lines per 1 00) in Scandinavian Nations ........... 1 08 6-1 Liberalization ofU.S. Telecommunications Market ......................................... l69 6-2 Liberalization of Finnish Telecommunications Market. ................................... 171 6-3 Liberalization of Korean Telecommunications Market. ................................... 173 6-4 Pros and Cons of Unbundling ........................................................................... 181 6-5 Korean Classification of Telecommunications Service Providers .................... 192 xiv

PAGE 15

CHAPTER 1 INTRODUCTION: INTERNET AND GOVERNMENT Writing about the Internet is a dangerous business: one runs the risk of adding to the incomplete description that has spawned misperceptions and myths about this global network. Brian Kahin (1997:47) Research Problems and Justification The end ofthe Twentieth Century, especially the 1990s, can be characterized by a worldwide transition from an industrial society towards an information society triggered by the rapid development of computer-cen.tered information technology (IT). The latest development to transform radically the IT landscape has been the widespread adoption oflntemet technology and its integration into the National Information Infrastructure (Nil). This new information technology, including an interactive communication medium such as the Internet, is fundamentally different from traditional technologies. Most traditional technologies have extended a person's muscular and manipulative power and have led the development of society from an agricultural to an industrial society (Riggs, 1957, 1964), whereas the information technology essentially extends the human brain and neural system and leads human society toward a knowledge-based,

PAGE 16

or information society (Bell, 1973; Katz, 1988; Zuboff, 1988; Kurxweil, 1999; Mitchell, 1999). Thus, the Internet is a major technological innovation of the 20th Century that portends key political, social, and economic consequences in the context of global development toward an information society (Castells, 1996; Albarran et al., 2000). Technologically, the relationship between the Internet and the traditional mediatelecommunications, computing, and broadcasting-is growing intimate and more complex with the emergence of a mass market for the Internet. Moreover, the Internet blurs traditional demarcations based on types of media in the regulation of communications services, combining features oftelephony, broadcast, and data communications. Politically, the new wave of large-scale infrastructure projects, such as the Nil Initiative triggered by the U.S. government in 1993, will transform the old social order of the industrial era. The new "information society" is a development goal or ideology politically supported by many governments through a variety of policy instruments (Garnham, 2000). For instance, many scholars propose that the Internet might be expected to revive participatory democracy (Naisbitt, 1982; Deaken, 1981; Rheingold, 1993) and has even been used as an indicator of a country's level of democracy (Anderson et al., 1995). Socially, there is the threat that technology can contribute to increasing inequality, given that it is unequally distributed among the population (NTIA, 1999; Novak & Hoffman, 1998), the so-called "digital divide". Culturally, the Internet could 2

PAGE 17

affect human interaction (Kiesler, 1997; Gackenbach, 1998; Wallace, 1999). For instance, the empowering role of the Internet is illustrated by how online interactions enrich interpersonal relations through the development of social relationships and by online community building (Parks and Floyd, 1996). Furthermore, the Internet brings various cultures into closer proximity by its "death-of-distance" phenomenon (Caimcross, 1997). Economically, in 1991, the Internet had less than 3 million users around the world, and its application to e-comrrierce was non-existent. By 1999, an estimated 250 million users accessed the Internet and approximately one-quarter of them made purchases online from electronic commercial sites, worth approximately $ 11 0 billion (OECD, 2000a). The Internet viae-commerce offers incredible potential and will change the ways in which businesses trade with ea:ch other and people shop for goods and services. In short, public access to the Internet is not a technical luxury but a critical necessity for full participation in economic, political, and social life in the information society ofthe 2151 Century (Burgess et al., 1999: 3). Therefore, it is appropriate to assess the role of government and public policy to ensure the mature use of the Internet in the context of social transformation to an information society. Nevertheless, especially among the international organizations, it is commonly held that the movement toward deregulation of communications markets is a worldwide phenomenon (OECD, 1997a, 1999a, 2001a; ITU, 1997, 1999; Wellenius, 1992). The external enforcement of competition in the telecommunications 3

PAGE 18

sector is now possible with the World Trade Organization's (WTO) Basic Telecommunications Agreement. In addition, the faith in market-based solutions to the telecommunications sector is extended to the studies of Internet diffusion regarding the Internet-related regulatory issues. The terms such as "privatization," "deregulation," "commercialization," and "liberalization" are surrogates for the extension of economic trends in the telecommunications sector to the development of the Internet. At the core of the economic studies oflnternet diffusion are the arguments that "non-government regulation (un-regulation) is a key success factor of Internet diffusion" (European Commission Report, 2000), and that ''the movement toward deregulation in the telecommunications sector is the single and most significant factor for the development ofthe Internet" (FCC Working Papers, 1997, 1999). Consequently, keeping abreast with the empirical data from the international organizations regarding the international comparison of Internet diffusion, the academic studies of Internet diffusion are skewed toward the synchronic and static analyses based on the economic and monocausal perspectives (Paltridge, 1996; Hargittai, 1999; Mann, 2000; Guillen and Suarez, 2001). Nevertheless, with the emergence of a mass market for the Internet caused by technological innovation, the nominal explanatory power of monocausal and static economic perspective in the Internet arena is now being challenged. First of all, these studies usually pay little attention to the diachronic and dynamic process oflntemet diffusion in the context of the social transformation toward an information society. 4

PAGE 19

Second, the economic studies oflnternet diffusion set aside as a given the huge network externalities caused by the multidisciplinary characteristics of the Internet in nature. In the Internet arena, network externalities can cause huge and significant consequences to the transformation toward an information society on the outside of the Internet market where the invisible mechanism of the price is supposed to function. For instance, social solidarity can be reinforced by an online community such as the "alumnus reunion" websites. The political democratization might be reinforced by an e-government movement such as the firstgov.gov website. Third, technological traits of the Internet-e.g., no physical connection between client and host computer make it impossible to impose a usage-based charge, which render the typical cost function ill-suited to the Internet arena. For example, the most striking characteristics ofthe communications markets shared by Australia, Canada, New Zealand, and the U.S. (which show a high rate of Internet performance) is that they have a flat rate charging structure of the dial-up connection to the Internet (OECD, 2000a). Fourth, practical problems can arise when monocausal economic theories are applied to the explanation of Internet diffusion phenomena among the respective nations. For example, both the United Kingdom (hereafter, U.K.) and Japan are fully liberalized nations with competition in local telephone and privatized Public Telecommunications Operators (PTOs), but they show low Internet performance. On 5

PAGE 20

the other hand, Iceland is a non-liberalized nation with a monopoly in local telephone and government-owned PTOs, but shows high Internet performance.1 Research Goals and Questions Broadly speaking, the goal of this dissertation is to consider the appropriate government role that would result in a nation's full participation in economic, political, and social life of the 21st Century information society via the Internet. More specifically, the practical gmil of this thesis is to provide a better understanding of emerging issues regarding the Internet and its related regulatory policy. The theoretical goal is to address a gap in the literature by combining synchronic analysis with diachronic analysis of the relationship of communications policy to Internet diffusion.2 For the purpose of achieving these research goals, this thesis examines the national policy styles and policy patterns for the mature use of the Internet across time and space in the context of the global development toward an information society.3 More specifically, the policy rationale, policy instrument, policy effectiveness, and 1 Among the various criteria, Internet hosts per I 000 inhabitants on July, 1999 (Appendix A) can be used to compare Internet performance: the U.K. (43.7), Japan (18.1), Iceland (108.2), and OECD average (47.1). 2 In this thesis, a "synchronic analysis" means an analysis of the variables similarly timed but differently placed, while a "diachronic analysis" means an analysis of the variables differently timed but similarly placed. Furthermore, a "synthetic analysis" means an analysis of the variables differently timed and placed. They can be matched with Rose's (1993) comparative statics, trend analysis, and comparative dynamics, respectively (see page 60). 3 In this dissertation, the national "policy styles" of Internet policy mean the comparative and synchronic characteristics oflnternet policy in a nation, while the "policy patterns" mean the comparative and synthetic characteristics (comparative dynamics) oflnternet policy in a nation. In addition, the "mature use" of the Internet means the last stage of Internet diffusion with the subscription of over half of the people in a nation. 6

PAGE 21

policy shift for the dissemination of the Internet are compared and contrasted through the proposed analytic framework and hypothetical propositions regarding the dynamic relationship of communications policy to Internet diffusion. In this respect, three interrelated questions warrant special attention: What are the comparative characteristics of national policy styles for the mature use of the Internet (Internet policy) across space? What are the comparative characteristics of national policy patterns for the mature use of the Internet across time and space? What is the appropriate role of government in terms of national policy styles and patterns across time and space? Chapter Outline To answer these questions; a brief look is to be taken in chapter 2 at the findings from previous research about synchronic causal studies, diachronic process studies, and synthetic comparative studies of the relationship between communications policy and Internet diffusion. Especially, reviewing the synchronic causal theories focuses on the governmental determinants of Internet diffusion, while reviewing the diachronic process models focuses on the policy implications in each stage oflnternet diffusion. As a synthesis of the relationship of communications policy to Internet diffusion, the relevant comparative studies are critically reviewed. After this, based on the critical synthesis of the existing knowledge on the relationship of communications policy to Internet diffusion, an analytic framework, composed of a three-dimensional (3-D) model and a three-stage (3-S) model, is 7

PAGE 22

proposed in the research design chapter (chapter 3). These models suggest five theoretical propositions regarding the dynamic relationship of communications policy to Internet diffusion, which are designed to be theoretically tested. Thus, the analytic framework and the propositions are intended to produce a theory about the styles and patterns oflnternet policy over time and space. A theory about an ideal style and pattern oflnternet policy is tested using a pattern-matching comparison, which is conducted in view ofthe 3-D and 3-S of Internet policy. For this purpose, this study selects the three relevant cases of the United States of America (hereafter, U.S.), the Republic of Finland (Finland), and the Republic of Korea (Korea) as a "most different system design" to show the similarities oflnternet policy patterns over time. Internet policies of each of the three nations are compared and evaluated in terms of a triad of Internet policy. In chapter 4, after identifying the technological similarities, the national policy styles in the U.S., Finland, and Korea are compared and contrasted to show that each of the three nations has relatively different national policy styles for the mature use of the Internetthat is, a de-regulative model of the U.S. with the free market orientation, a re-distributive model of Finland with the welfare state orientation, and a distributive model of Korea with the high stateness condition. In chapters 5, 6, and 7, the theoretically drawn patterns of Internet policy of the three nations in each stage oflnternet diffusion are compared and contrasted respectively with a predicted theoretical pattern oflnternet policy over time under the 8

PAGE 23

basic assumption of a S-curve process of Internet diffusion. With the pattern-matching comparison, these three chapters provide both the similarities (a literal replication) and the differences (a theoretical replication) of the longitudinal patterns oflnternet policy in the three nations. For this purpose, policy rationale, policy instruments, and policy impacts on Internet diffusion of the three nations in each stage are analyzed with the ideal pattern variables oflnternet policy drawn from the research design chapter. Chapter 8 concludes this dissertation by summarizing its findings, discussing its theoretical and practical implications, and suggesting avenues for future research. Especially, the governmental role for the mature use of the Internet is drawn from the contingent perspective, that is, the appropriate policy responses need to be taken to reflect the changes of the policy environments in each stage oflnternet diffusion. 9

PAGE 24

CHAPTER2 LITERATURE REVIEW After briefly reviewing the background literature about the multidisciplinary nature of the Internet, this dissertation draws on the literature from synchronic causal studies, diachronic process studies, and synthetic comparative studies on the relationship of public policy to Internet diffusion over time. These literature reviews are conducted from the perspective of public policy to provide insight toward the three research questions. For this purpose, the focus of reviewing the causal studies is on the influential theoretical imperatives of communications policy for the mature use of the Internet, i.e., the political, the economic, and the social imperatives. In addition, the focus of reviewing the process studies is on the two influential models on the relationship of communications policy to the developmental stages of Internet diffusion, i.e., the logistic model and the interactive model. The purpose of reviewing both causal and process studies is to identify the two policy controversies in view ofthe observation that the literature on Internet diffusion is multidisciplinary in nature (Hindle et al., 1997), which leads us to review the final and synthetic comparative studies. 10

PAGE 25

Internet as Commons Internet and Characteristics The Internet4 is a revolutionary and global medium that combines the elements of telecommunications, computing, and broadcasting services into a new model for communications. When compared with telecommunications, the Internet uses radically different technologies and pricing mechanisms (Manishin, 1996; Werbach, 1997; Gibson, 1999). Unlike circuit-switched telecommunications such as the public switched telephone network (PSTN), packet-switched Internet communications share bandwidth, so that there is no physical connection between client and host computers.5 Some top-level domains (such as ".uk" for Britain) are country-specific; others (such as ".com") are inherently generic and non-geographic. Perhaps most significantly in the telecommunications domain, transport and applications are bundled, so that most of the intelligence ofthe medium resides in the network. On the Internet, however, intelligence is decentralized, and applications are delivered by client-oriented software. Each also is classified into different services especially in the U.S.; that is, the basic 4 The Internet can be simply defined as an interconnected global computer network of tens of thousands of packet-switched networks using the Internet Protocol (IP). IP defines the structure of data, or "packets," transmitted over the Internet. The higher-level "transmission control protocol" (TCP) controls the routing and transmission of these packets across the network. Most Internet services use TCP, and thus the Internet is often referred to as a "TCP/IP" network (Werbach, 1997: I 0). 5 A packet-switched network means that data transmitted over the network is split up into small chunks, or "packets". For a packet-switched network, a dedicated end-to-end transmission path (circuit) does not need to be opened for each transmission. Rather, each router calculates the best routing for a packet at a particular moment in time, given current traffic patterns, and sends the packet to the next router (Werbach, 1997:17). II

PAGE 26

service and the enhanced service, 6 which causes information service providers to exempt the access charges to local networks and not to contribute to universal service funding (USF). With relation to the pricing mechanisms, the packet switching tariffs are based on bandwidth, content, and the bit, not by the minute as for a traditional circuit switched network (Table 2-1 ). Table 2-1: Comparison of Telecommunications and Internet Telecommunications (PSTN) analog, becoming more digital circuit-switched physical connection static switching geographic numbering centralized intelligence usage-based pricing by the minute basic services regulated -+ competitive contribution to USF access charges vs. Internet digital packet-switched connectionless dynamic routing neo-geographic addressing decentralized intelligence flat-rate pricing by the bandwidth the content and the bit enhanced services competitive ... ???? non-contribution to USF no access charges 6 Such a legal distinction between the unregulated Internet and the regulated telecommunications was drawn from the FCC's three Computer Inquiries (so-called, Computer I, II, and III) in the U.S. For the purpose of preventing cross subsidization and discrimination between unregulated and regulated service, two questions for essential determination were crucial to all Computer Inquiries: (I) how to draw a line between data processing and communications services and (2) whether, and if so, under what conditions to permit regulated common carriers into unregulated data processing services. For further information, see First Computer Inquiry, 28 FCC 2d 267 (I 971 ); Second Computer Inquiry, 77 FCC 2d 384, 419 (1980); Third Computer Inquiry, NPRM, FCC 85-358 (1985-). 12

PAGE 27

Furthermore, the Internet enables structural convergence among the domains of communications, information, and computing, which brings three different policies and values into contact on the web (Albarran, 2000; Kahin, 1996). In the past, communications networks were designed to carry different types of information separately. Telephone networks were designed for voice, broadcast networks for video, and so forth. Traditionally, communications regulation has been concerned with the provision and operation of the physical network and access to that network, that is, the regulation ofthe conduit. First of all, the telecommunications industry has traditionally been controlled in terms of market entry, service pricing, and technology to ensure interoperability of equipment. Second, in the broadcasting industry (radio and television, for instance), licensing has provided the basis for regulation on political and cultural criteria. The objectives of governmental regulation have typically included ensuring pluralism, impartiality, protection of vulnerable social groups, and the promotion of cultural heritage. Third, the computing industry fundamentally was able to develop in a largely unregulated fashion (Pool, 1983). Now, digital technologies permit the manipulation of all forms of information voice, data and video-across all types of networks. We can easily see the convergent trends of traditional services in the Internet in which applications are spreading across communications media, from the telephone to the wireless, cable, and mass media such as Digital TV (Clements, 1996; Blackman, 1998) (Figure 2-1 ). 13

PAGE 28

Figure 2-1: Communications Policies Associated with Traditional Industries freedom of expression government information public librari s copyright privacy antitrust rate regulation universal service nondiscrimination mandatory standards computing patents research funding tax credits antitrust Source: Brian Kahin, The Internet Business and Policy Landscape. 1997. p.66 Consequently, in light of both the dramatic differences between telecommunications and the Internet and the convergent trends of communications policy on the web, PSTN-type policy tools cannot simply be imported to the Internet (Froomkin, 1996; Lenin & Meisel, 1997). Thus, the literature on the inherent nature of the Internet needs to be critically reviewed. Nature oflnternet Service Commonly cited criteria for the distinction between the public and the private sectors are funding and ownership (Wamsley and Zald, 1973). If these are used as the criteria for the Internet, Internet service can be considered public goods in view of 14

PAGE 29

public funding and ownership in its initial development, on the one hand. The Internet was initially developed by the U.S. Department of Defense' Advanced Research Projects Administration (ARPA) as a publicly financed defense initiative. Furthermore, later the NSFNET that was evolved directly out of ARPANET was also funded by the National Science Foundation (NSF) until its privatization on April 30, 1995 (Moschovitis, 1999; Varian, 1997). On the other hand, the Internet also can be regarded as private goods in terms of both private ownership and termination of public funding after the privatization of the NSFNET (Hallgren et al., 1999). The main problem of this dichotomous scheme is that it makes no mention of government regulation7 (Rainey, 1997: 66). Therefore, to understand the relationship between Internet diffusion and communications policy, a review of the literature on the inherent nature of Internet service that provides us with the rationale for governmental intervention is indispensable (Head, 1974; Ostrom and Ostrom, 1977; Bobrow and Dryzek, 1987). It can be analyzed in terms of both possibility of exclusion (exclusion, failure of exclusion) and mode of use (separable use, joint use)8 : in a matrix format, this pairing results in private goods, toll goods, common pool resources (CPRs), and public goods (Jennings, 1986: 117) (Figure 2-2). 7 As Rainey noted, many corporations, such as IBM, receive funding from government contracts but operate so autonomously that they clearly belong in the private category. 8 Exclusion is characteristic of a goods when individuals can be denied access to the goods by the provider. A goods is characterized by joint consumption when the use of that goods by one person does not detract from or prevent its use by another person. 15

PAGE 30

Figure 2-2: Classification ofNature of Goods and Service Exclusion Failure ofExclusion Separable Use Private Goods CPRs (PC) (Natural Resources) Joint Use Toll Goods Public Goods (Roadways) (National Defense) Source: Jennings, 1986: 117 When there is no "network congestion,"9 the Internet service gets its nature of public goods because it is subject to joint consumption and exclusion is infeasible (Hallgren and McAdams, 1994). First of all, the "cloud"10 of the Internet is common to all Internet users. When a new user sends traffic over the network, current users see no difference in their performance; service is subject to joint consumption. Anyone at a network node can make use of the network theoretically; exclusion is not possible. When network congestion on the Internet occurs, the Internet can be classified into the CPRsexclusion is not possible, but consumption is not joint. For instance, in the U.S. out of the situation of degradation, the Internet service manifests itself as congestion of the ARPANET. So, the main characteristic of Internet service is a "failure of exclusion," causing it to get inherent natures of public goods and CPRs. In this respect, the Internet can be classified into "the commons" (Brin, 1995; MacKie-9 The Internet service can be a congestible resource like the roadways in that the Internet service has a classic problem of"the commons" and the "free riders." For instance, with a fixed fee, many users can send lots of data (live video) simultaneously via the Internet, which can be beyond the network capacity. The congested network will cause delayed or dropped transmission of the given data. 10 The Internet "cloud," which can be defined as the Internet's network of networks, is a perfect analogy for the wide area networking capabilities for the Internet since there is no fixed form which keeps it in place (Mcknight & Bailey, 1999: 156). 16

PAGE 31

Mason and Varian, 1996), but cannot be classified into both toll goods (roadways) and private goods (PC).11 For instance, Internet users pay a fixed fee in exchange for unlimited access up to the maximum throughput of their particular connection, which will bring on the classic "free riders" problem in the Internet network (MacKie-Mason and Varian, 1996). Allocation Mechanisms and Internet Policy The major question arises about the allocation of the Internet resources as "the commons," that is, how one can solve the dilemma ofthe so-called "tragedy of the commons" oflnternet networks (Olson, 1965; Hardin, 1968). Central government's intervention is one traditional solution, and marketization is another. Thus, the "state" and the "market" may be considered dichotomous solutions, but this thesis intends to include the newly emerging "community" as the third solution from the practical perspective of division of labor among the allocation mechanisms and the three major variables of Internet policy (network, service providers, and users). In this section, after reviewing the historical overview of the state, the market, and the community, the theoretical foundations ofthe triangular nature of both the allocation mechanisms and the Internet policies are briefly reviewed. 11 Toll goods can be simply defined as those for which exclusion is feasible but there is joint use, and private goods as those which are not subject to joint use and for which exclusion is feasible. 17

PAGE 32

Historical Overview. From the historical point of view, there is an ebb and flow in a worldwide tendency to place our faith in the private market, public government, and non-profit community (Mishra, 1984; Kettle, 1993; Adams and Hess, 2001). Reliance was on the private market before the Great Depression and World War II. After World War II and for the subsequent three decades, welfare states were premised on state intervention to steer markets and communities in the direction of increased growth and equity. Thus, reliance was on the public government. By the 1970s there was a crisis of legitimacy for the welfare state caused by "government failure" (Mishra, 1984), and the 1980s and 1990s saw the influence of markets and public choice reasoning dominate in the public sector. So, reliance was on the private market again. While these policies based on market approaches have succeeded in introducing efficiency instruments to public sector management, they have failed in areas in which political rather than economic judgments, and qualitative rather than quantitative instruments, are necessary (Wolf, 1993; Adams and Hess, 1999). Thus, the failures of government intervention and the negative impacts ofthe subsequent capture of public policy by market-based rationality have created the political space for the recent emergence of community-based ideas (Adams and Hess, 2001). 18

PAGE 33

Dichotomy. The relevant examples of the dichotomous distinction between the government and the market can be drawn from Deborah Stone's (1997) public policy theory and James Carey's (1989) communications theory. The two theories together show the combined traits of Internet policy that are closely connected to both a public policy and a communications medium. Stone's (1997: 18-34) two competing policy models-i.e., the market model and the polis model-is one example of the dichotomous distinction. The polis model is a model of political society, that is, a model of the simplest version of society that retains the essential elements of politics. Its unit of analysis is community, and its glue bridging the vast gap between self-interest and public interest is influence, cooperation, and loyalty. In the polis model, political resources are governed by the laws of passion -human resources are renewable and expand with use. In the market model, individuals act only to maximize their own self-interest, and the nature of collective activity is competition. In the market model, economic resources are governed by the laws of matter-material resources are finite and diminish with use. From the perspective ofthe division of labor among the three allocation mechanisms of scare resources, Stone's market model gives us a clear-cut overview of the economic approach, but her polis model may be considered mixed and represents an unclear model of government and community. James Carey's (1987) communication models-that is, the transmission model and the community-cultural-ritual (CCR) model -are other examples of dichotomous 19

PAGE 34

distinctions. Carey's two models can be compared in four areas. First, in terms of a level of analysis, the transmission model places primary focus on the individual, whereas the CCR model embraces a larger conceptual territory called community. In terms ofthe mode of interaction, the transmission model focuses on the movement of messages from sender to receiver. The CCR model, on the other hand, begins with a study ofthe social practices and rituals associated with a communication event. In terms of domain, the primary analytical focus of the transmission model is on the spatial dimension, whereas the CCR model seeks to understand the temporal dimension. Finally, the transmission model looks to the consequences of communication in terms of individual ratification, private goods, and rationality. The CCR model looks, instead, to the public consequences of communication, including traditions, public goods, and the dialogic interaction necessary to promote and sustain democratic discourse (Lenert, 1998: 8). James Carey established a fundamental distinction between the two models of communication that is important for understanding socio-cultural implications of communications policy. E. Lenert (1998) posits that current telecommunications policy discourse, drawing upon the transmission model, aims to achieve liberalization of telecommunication markets. From the perspective of the division oflabor among the three allocation mechanisms of scare resources, the transmission model can give us a clear-cut overview of market logic applied to communication, but the CCR model may be considered a mixed model of government and community. 20

PAGE 35

Trichotomy. Considering the limits of the dichotomous distinction between the public and the private in the Internet arena caused by the inherent nature of Internet service as the commons, this dissertation utilizes the trichotomous distinction among state, market, and community. This distinction is analogous to the trichotomous distinction among the public, the private, and the non-profit. Burton A. Weisbrod (1988) posits that it is the nature of different economic roles of the public, the private, and the non-profit sectors that allows clear distinction: the government exists to ameliorate market failures; the private sector exists to satisfy market demand; and the private nonprofit sector meets the diversity of demand that is too small to elicit a response from either market or government. Furthermore, Adams and Hess (200 I: 17) clearly show the division of labor among the three allocation mechanisms: the government model relies on legal authority to regulate relation and its preferences are revealed by votes; the market relies on voluntary monetary contracts, and its preferences are revealed by prices; the community relies on shared values as the glue explaining social relations, and its preferences are revealed by its values and engagement with reciprocity. Triad oflnternet Policy. From the perspective of system or input-output analysis (Easton, 1979; Dye, 1998), Internet policy, which can be briefly defined as the communications policy for the mature use of the Internet, has the three theoretical 21

PAGE 36

dimensions of its causes and consequences from the three major players of government (or network), service providers, and service users, respectively. In this regard, the trichotomous distinction among state, market, and community can be analogous to the three major inputs from the three major players in Internet policy. First, input from the state for network digitalization, especially from the political arena, to the implementing administration as a policy system can be termed as a political dimension of Internet policy. 12 Second, input from the market, especially from the service providers, to the policy system can be termed as an economic dimension oflnternet policy. Third, input from the community, especially from the service users, to the policy system can be termed as a social dimension oflnternet policy. These distinctions oflnternet policy based on the trichotomous division of labor among state, market, and community can be referred to as a triad oflntemet policy (Figure 2-3). 12 In this dissertation, from the perspective of the traditional politics-administration dichotomy (Goodnow, 1893; Willoughby, 1927; Gulick, 1933; Dimock, 1958; Appleby, 1949; Waldo, 1948), the player oflntemet policy as a government can be divided into two; politics (Chieflnforrnation Officer) and administration (Ministry or Independent Regulatory Body). 22

PAGE 37

Figure 2-3: Triad oflnternet Policy Market (Service Providers) State or Politics (Network) Mature Use Of Three Theoretical Approaches Community (Service Users) Based on the trichotomous distinction among the state, the market, and the community, the three theoretical approaches to the Internet as the commons can be reviewed in view of their theoretical backgrounds and their applications to the Internet. Economic Approach: Market Model Overview. The economic approach inherently originated from an extension of neoclassical, micro-economic theory on the workings of the free market. This approach applies to the public policy arena as in Stone's (1997) market model, to the ) communication theory as in Carey's (1987) transmission model, and to the legal theory as in Posner's (1974) Economic Analysis ofLaw (EAL) model. 23

PAGE 38

Especially regarding the legal arena, EAL was born in the early 1960s, but it is with the 1973 publication of Judge Richard Posner's (1992) Economic Analysis of Law that EAL began to have a dramatic impact on legal education and scholarship. Posner's central thesis was that judicial opinions do and/or should display an economic logic (Ostas, 1999). The economic approach is connected to the policy goal ofliberalization13 and policy instruments of competition, privatization, and price deregulation. Especially, telecommunications policy is predominantly based on this approach. Liberalization achieves competition by deregulation and laissez faire policy. Liberalization is also closely connected to its policy emphasis on both the technologically efficient transport of messages and the private ownership ofthe communication networks {Hill, 1998; Lenert, 1998; Cuilenburg et al., 1995). To achieve the policy goal of liberalization, an economic approach focuses on correction ofmarket failure based on welfare economics (Friedman, 1953, 1982). The policy instrument for correcting the market failure is the introduction of competition in the communications market through privatization. The economic approach also focuses on deregulation ofthe communications infrastructure based on information economics, including public choice theory, principal-agency theory, and transaction cost theory (Walsh, 1995: I 7; Willianson, I 985; Ghertman et al., 1995). The deregulation movement prefers market-type allocation mechanism to governmental mechanism 13 "Liberalization" can be simply defined as ''the unleashing of the magic ofthe marketplace". This simple definition is the often-cited formulation of former U.S. President Ronald Reagan. 24

PAGE 39

under the basic assumption of Internet service as private goods. Information economics views market strategies as a means to correct for government failure such as underor over-supply of public goods, and X-inefficiency of governmental bureaucracy. In the Internet arena, Internet Economics attempts to improve the understanding of the Internet as an economic system based on the quantitative analysis of network congestion controls and pricing mechanism (McKnight and Bailey, 1999; Mackie Mason et al., 1993, 1995). Assessment. A major strength of this approach is the incentive for its managers to achieve greater rewards for greater efficiency and responsiveness. The second strength is its relative independence from government and political constraints. But its weakness is that the pursuit of profit sometimes results in social inefficiencies and inequities. These kinds of market failures can be caused by the lack of information and the existence of the network externality. Lack of information causes failures in the sense that consumers do not obtain the goods and services that they would demand if they were well informed. The existence of the network externality causes failures in the sense that profit-motivated firms do not take into account any effects of their activities that do not influence their revenues or costs (Weisbrod, 1988). Furthermore, the main problem of this approach is that traditional price regulation on the Internet brings distorted pricing structures from the perspectives of the different pricing mechanisms between the two media (Cawley, 1997) 25

PAGE 40

Political Approach: Government Model Overview. The political approach, based on a deliberative14 and a communitarian 15 framework, tries to connect theories of democracy with theories of communication to allow new communication technologies to contribute to the expansion of democracy and the empowerment of more people (Abramson et al., 1988; Hacker, 1996). The interactive Internet can revive ancient Greek ideas of direct participation in speech and debate, by reestablishing the connection between voters and candidates, improving voter information, and increasing civic participation (Friedland, 1996; Ogden, 1994). Furthermore, the political approach emphasizes a political goal of"electronic democratization," meaning that adding channels to the political communication system will empower the citizen by popular participation in decision making in the context of liberal guarantees of equality and individual rights16 (de Leon, 1992, 1997). Recently, the e-government movement has shown us ways to improve the effectiveness of political decision-making by making citizens aware ofthe how and the why of political 14 The deliberative model takes representative democracy as a starting point and asks how it might be strengthened and made more participatory (Friedland, 1996; Dryzek, 2000). 15 A communitarian framework emphasizes deliberation over direct voting and is rooted in identifiable communities and a concept of the public good that goes beyond both individual and group interest (Abramson et al., 1988, Etzioni, 1994). 16 Electronic democratization can also be termed "increasing freedom of communication" (Lenert, 1998; Cui len burg et al., 1995). Freedom of communication may be defined as the right to send and not to send, or to receive and not to receive messages without any interference whatsoever by someone else. Information that is essential to fully participate in a democratic society should be provided free. 26

PAGE 41

decision-making and facilitating their participation in this process17 (Milward and Snyder, 1996). The promises ofthis perspective are well reflected in AI Gore's Earth in the Balance (1992: 359): .... the best way for a nation to make political decisions about its future is to empower all of its citizens to process the political information relevant to their lives and express their conclusion in .free speech designed to persuade others .... The political approach prefers the government model to the market model as the allocation mechanism of the resources under the basic assumption that Internet service can be classified as public goods. The political approach postulates that collective action through government has the potential of correcting market failures. Assessment. The strength of political approach lies in its power of compulsion to regulate, to tax, and to subsidize and to monitor. It can also encourage the provision of goods and services that are inadequately provided by the private sector or that are not provided to particular groups in the population. Its handicaps are its political entanglements-for example, the dependence of senior officials on periodic elections-and the weak incentives it provides for efficiency, which can be symbolized in government failure. The major problem for the government model is dealing with 17 The Cyberspace Policy Research Group (CyPRG) studies diffusion and use of the World Wide Web in government worldwide, particularly in terms of organization openness and internal effectiveness. For further information, see http://www.cypre.arizona.edu 27

PAGE 42

diversity, which is fundamentally a problem of information.18 Inevitably, it causes the problems of overand under-utilization. Considering its limits as an allocation mechanism of scare resources, the key policy issue is whether the disadvantages of using an indirect form of regulation are offset by the greater ease of monitoring it rather than regulating production and its distribution directly. The government regulates what it can monitor easily, and it monitors what it can gauge usefully and inexpensively. If and when regulation of nonprofits is easier than direct regulation of output, production processes, or the distribution of output, the nonprofit form of institution (community) is attractive (Weisbrod, 1988). Social Approach: Community Model Overview. The social approach emphasizes a policy goal of increasing social and cultural interaction (Lenert, I 998; Cuilenburg et. a!., 1995). By treating communities as social capital network, rather than strictly as discourse communities (Kubicek, 1997), this approach, focusing on social capital19 and community assets-based approaches,20 tries to ground the connective elements of the Internet in social life and social culture. 18 Some people who benefit from governmental services are unintended beneficiaries; they are included only because ofthe cost of excluding them. At the same time, some who do not obtain the services are excluded unintentionally, victims of the high costs of targeting the consumer population more accurately. 19 Social capital is the necessary infrastructure of civic and community life that generates norms of reciprocity and civic engagement (Coleman, 1990; Putnam, 1993, 1994a, 1994b, 1995). 20 Assets-based development (Kretzmann and Mcknight, 1993; Mcknight, 1995) stresses locally generated knowledge that permits communities to mobilize their assets, broadly conceived, to address problems. 28

PAGE 43

The failures of governmental intervention and the negative impacts of the subsequent capture of public policy by market-based rationality have created the political space for the (re-) emergence of community-based ideas (Adams & Hess, 2001 ). Traditional broadcasting policy used to be based on the social approach. In the telecommunications arena, social approach can be characterized by customer-centered regulation (Kahin, 1997) that promotes the interests of the customer through symmetrical removal of restraints on any competitor's ability to compete. Under this approach, a government can limit the freedom of all carriers to compete and rely on regulatory monitoring to ensure against undue discrimination. The government would impose administrative burdens symmetrically on all carriers and subject all carriers to strategic abuse ofthe regulatory process by rivals (Haring & Weisman, 1993)?1 Increasing social interaction means universal access to the information infrastructure (Hudson, 1994). Fundamental to life, liberty, and the pursuit ofhappiness in the information age is access to video, voice, and data networks that provide a broad range of news, education, health, and government information and services. Such services should be provided in a user-friendly format, widely available to everyone, including persons with disabilities. In the Internet arena, universal access means narrowing the digital divide. 21 Such a symmetric regulation can be achieved by the essential facility doctrine (MCI Communications Corp. v. AT&T Co.) for securing nondiscriminatory access to the local telephone facilities and bottleneck facilities in communications on which the Internet communications is heavily dependent. 29

PAGE 44

The digital divide can be defined as a gap between those individuals and communities that have access to computers and the Internet and the ability to effectively use this technology and those who do not, in other words, a gap between information haves and have-nots (NTIA, 1999). The report ofthe U.S. Department of Commerce (July, 1999) says that the following examples highlight the breadth of the digital divide today: (I) Those with a college degree are more than eight times as likely to have a computer at home, and nearly sixteen times as likely to have home Internet access, as those with an elementary school education. (2) A high-income household in an urban area is more than twenty times as likely as a rural, low income household to have Internet access. (3) A child in a low-income white family is three times as likely to have Internet access as a child in a comparable Black family, and four times as likely to have access as children in a comparable Hispanic household. (4) A wealthy household of Asian/Pacific Islander descent is nearly thirteen times as likely to own a computer as a poor Black household, and nearly thirty-four times as likely to have Internet access. (5) Finally, a child in a dual-parent white household is nearly twice as likely to have Internet access as a child in a white single-parent household, while a child in a dual-parent Blackfamily is almost four times as likely to have access as a child in a single-parent Black household. Furthermore, this approach basically prefers the community model to the market and the government model as the allocation mechanism of the resources under the basic assumption that Internet service can be classified as a CPR. Social approach puts forth that collective action through self-governing community has a potential of correcting market failures. Based on the social characteristics of the Internet, the self30

PAGE 45

governing model emphasizes self-regulation of the Internet by the individual user, social body or culture, and industry and commerce, and the Internet itself.22 In 1996, John Perry Barlow produced a proclamation he called the "Declaration of the Independence of Cyberspace":23 Governments of the Industrial World ... I ask you to leave us alone. You are not welcome among us. You have no sovereignty where we gather ... You have no moral right to rule us nor do you possess any methods of enforcement we have true reason to fear. Assessment. A self-governing community (or, social capital network) has the hybrid combination of the strengths and weaknesses of government and profit-oriented organizations. Like for-profits, nonprofits have no power to compel action, but they can be more flexible in meeting the heterogeneous demands of consumers for collective action. Like governments, nonprofits are being restricted in the use of any surplus like the informational superiorities they generate. Yet, non profits may be more trustworthy forms of institution than the private market (Weisbrod, 1988). The major problem of this model is that the community has no power to compel action. This is the reason why government involvement is justified in the arena of appropriate Internet security controls, such as public key cryptography, digital signature and its authentication, and credentials and certificates in today's electronic marketplace (Kutais, 1999). 22 McCloskey, 1998. Available online: http://www.ilpf.org/selfreg/bib4-15.htm 23 Available online: http://www.dff.org/pub/Publications/John-Perrv-Barlow/barlow-0296.declaration 31

PAGE 46

Summary To obtain a comprehensive solution to the multi-dimensional Internet-related policy controversies, a comprehensive model is useful because each theoretical approach to the Internet gets its strengths and weaknesses as an allocation mechanism. Hence, from the perspective of public policy, the ideal relationship among the nature of Internet service, the allocation mechanism, and the dimensions of Internet policy can be summarized in Figure 2-4. The political dimension oflnternet policy ideally matches with both the input from the state, especially from the political arena, and Internet service as a public goods. The economic dimension oflnternet policy ideally matches with both the input from the market, especially from the service providers, and Internet service as a private goods.24 The social dimension oflnternet policy ideally matches with both the input from the community, especially from the service users, and Internet service as a CPR. 24 In Economics, CPRs used to look upon it as private goods. Martyne M. Hallgren and Alan K. McAdams, 'The Economic Efficiency of Internet Public Goods", in McKnight, L.and J. Bailey (eds.), ( 1999: 455-478). 32

PAGE 47

Figure 2-4: Three Dimensional Approaches to Internet Three Dimensions Political Achieving Network Digitization Allocation Mechanism Government Votes Representation Order Redistribution Nature of Internet Service lnfonnation & Computing (Data) Media Policy Mature Use ofthe Internet Economic Increasing Provider's Entrepreneurship Market Prices Competition Efficiency productivity Private Goods Telecommunications Service (Voice) Telecommunications Policy 33 Social Achieving User's Equity Community Values Reciprocity Equity Cohesion Transport Service (Video) t_ Broadcasting & CATV Policy

PAGE 48

Synchronic Causal Studies Internet Diffusion and Indicators After the ARPANET was developed by the U.S. DOD's ARPA to link together universities and high-tech defense contractors in the late sixties (Mackie-Mason & Varian, 1995), the Internet grew rapidly, perhaps at an unprecedented rate. Significant Internet diffusion can be observed worldwide only in the past few years with the global number of network connected computers surpassing 35 million in 1998 compared to less than 1.5 million in 1993 (OECD, 1999a); The current research on the indicators of Internet diffusion focuses on data on the number of Internet hosts, web sites (web servers and secure servers), and Internet subscribers. A commonly cited measure oflnternet diffusion is the number oflnternet hosts25 (Appendix A). Hosts surveys have been undertaken by the Network Wizards (NW) since 1981. NW defines an Internet host as a domain name that has an associated Internet Protocol (IP) address record. This would be any computer system connected to the Internet (direct or dial-up connections), such as firstgov.gov. NW surveys include all country code Top Level Domains (ccTLDs such as .kr for Korea) and generic Top Level Domains (gTLDs such as .com) and is undertaken every six months. By July 1999, the OECD analyzed the NW data and estimated that there were more than 52 million hosts in the OECD area representing 93% of the global total (Paltridge, 1999). 25 Internet hosts can be simply defined as any computer system with an Internet Protocol (IP) address connected to the network (Coppel, 2000). 34

PAGE 49

But, host data do not provide a full count of users because surveys do not capture all computer systems connected to the Internet (e.g. computers behind firewalls) and thus provide an indicator of the minimum size of the Internet (OECD, 2000; Coppel, 2000). Recently, Telcordia Technologies26 has provided daily updates of number oflnternet hosts based on a random sample addresses sampled throughout each day. The diffusion of the Internet accelerated with the invention of HTTP, the World Wide Web hypertext transfer protocol, and the number of web sites brings another important indicator. These data were compiled by Netcraft27 and the Online Computer Library Corporation (OCLC)?8 Especially, the Netcraft Survey is a survey of Web server software used on computers connected to the Internet. Netcraft collects and collates as many hostnames providing an "http-service" as their survey can find, and systematically polls each one with an HTTP request of the server name. A host name is the first part (before the first dot) of a host's domain name (e.g., www). Between July 1998 and July 2000, the number of web sites in the world increased from 2.6 million to 18.2 million (Appendix B). At the same time, the number ofweb sites under OECD ccTLDs increased from 643.4 thousand to more than 4.3 million, and web sites under gTLDs increased from 1.8 million to 12 million. As one ofthe best indicators ofthe growth and diffusion ofe-commerce, a count of secure servers is used (Appendix C). Secure servers allow users to encrypt 26 Refer http://www.netsizer.com 27 Refer http://www.netcraft.com 28 Refer http://www.oclc.org 35

PAGE 50

information on, for instance, credit card data that facilitates e-commerce. Netcraft developed the Secure Socket Layer (SSL) protocol for encrypted transmission over the TCP/IP network. In August 1999, the OECD analyzed the Netcraft data and estimated that there were over 46,000 secure servers in member nations. They estimated that this represented 95% of the global total and a 109% increase over the previous year (OECD, 2000a). The numbers of Internet subscribers in a nation can be the fourth indicator (Appendix D). This information can be drawn from the largest telecommunications carriers reporting both the number of subscribers to their Internet services and their estimates of market share. At the end of 1999, there were at least 49 million Internet subscribers in the U.S., approaching 11 million in Japan and in Korea, 9 million in Germany. A ranking of countries in terms of Internet subscribers per 1 00 population shows high levels oftake-up in Korea, Sweden, Denmark, Canada, the United States, Netherlands, Iceland and Norway. In the private sector, NUA29 estimates the number of users in each nation by averaging journalistic and market research reports from around the world. In addition, Computer Industry Almanac30 estimates and projects it as a function ofthe number of personal computers and other determining factors. 29 http://www.nua.ie 30 http://www.c-i-a.com 36

PAGE 51

Diffusion Determinants in General Among the various studies on the determinants of diffusion of innovation, this literature review focuses on the determinants of both inter-country diffusion and the diffusion ofthe interactive31 medium. The inter-country approach to the diffusion of innovation attempts to explain international differences in the speed of diffusion of innovations in terms of the characteristics of the countries and industries concerned (Swan, 1973, Nabseth and Ray, 1974; Davies, 1979:25) or in terms ofthe attributes of technology development and the objectives of technology development ( deLeon, 1979). Furthermore, among the various technological innovations, the literature on the innovation of the interactive technologies such as telecommunications, broadcasting, and especially the Internet is to be focused (Allen, 1988; Antonelli, 1989; Mahler and Rogers, 1999). In this respect, existing literature about important determinants of Internet diffusion can be divided into two parts: non-governmental determinants, such as economic factors, human capital, and existing technologies; and government determinants, such as institutional and legal environments (Press, 1997: Hargittai, 1999: Guillen & Suarez, 2001). Previous studies on non-governmental determinants oflnternet diffusion have found that economic wealth strongly predicts a population's adaptation of the Internet (Lin, 1993). A country's overall economic strength can affect Internet diffusion in that 31 Interactivity is the degree to which participants in a communications system can exchange roles in, and have control over, their mutual discourse (William, Rice & Rogers, 1988). 37

PAGE 52

the necessary resources are more likely to be present, and capital required for the expansion of the technology is more available in richer countries. There are two ways in which the level of human capital may be relevant to Internet diffusion: the population's level of education and its English language proficiency. Most studies that have examined the education level of adopters of new technologies find that people with higher levels of education are quicker to adopt new innovations than people with comparatively less education (Hargittai, 1999). Furthermore, it is suggested that academic institutions often play an important role in spreading the Internet since they are often among the first institutions in a nation to be wired (Lin, 1993; Kelly and Petrazzini, 1997). So, the education level of a nation may affect the rates of Internet diffusion. In addition, Laponce (1987) suggests that some languages have greater status than others and they dominate certain areas of life, such as the English language having a prominence in the computer industry. In the same vein, the prominence of the English language on the content of the Internet was reported that more than 94% of the links pointing to pages on secure servers were in English (OECD, 2000a). So, the level of English proficiency may affect the number of people interested in using the medium. Thomas ( 1988) found that the spread of technology is contingent upon certain technological factors being present in the target nation. With respect to the Internet, existing telecommunication facilities may be crucial for understanding variation in the spread of the Internet. It is because the Internet in its initial stage was developed as the 38

PAGE 53

expansion of computer applications over the telephone lines. For instance, nearly all the 121 million Internet subscribers at the end of 1999 still used dial-up connections via the telephone network (PSTN) (OECD, 2000a). Diffusion Determinants and Communications Policy The institutional and legal environment (Hargittai, 1999) in a country is also relevant to the Internet's spread because public policies can enhance or hold back Internet diffusion, depending on their approach to regulating mechanisms. Among various theories on the policy determinants of Internet diffusion (Press, 1997; Press et al., 1998; Guillen & Suarez, 200 I), there are three distinctive types oflntemet policy that emphasize different influences, and warrant special attention: the de-regulative policy; the distributive policy; and the re-distributive policy. These also mirror the distinction among the three dimensions of Internet policy in the previous review. Before reviewing the three types of policy determinants, Lowi's policy typologies need to be briefly introduced in that three types of policy determinants in this thesis also mirror Lowi's three types of policy. Policy Typologies. Theodore J. Lowi ( 1964, 1972), in developing the three types ofpolicy32 (regulative, distributive, and re-distributive), has argued that "policies determine politics" and thus policy processes will differ significantly depending on the 32 Lowi added constituency policy later, but this policy is not utilized in that this thesis focuses on the external causes and consequences among the three major players oflntemet policy. 39

PAGE 54

policy type involved. So, each type of policy reveals its several comparative characteristics. First, regulative policy is distinguishable in that the regulatory decision involves a direct choice as to who will be indulged and who deprived. Furthermore, regulative policy cannot be disaggregated to the level of the individual or the single firm, because individual decisions must be made by application of a general rule and therefore become interrelated within the broader standards of law. Regulatory decisions are cumulative largely along sectoral lines since the most stable lines of perceived common impact are the basic sectors ofthe economy. These characteristics ofLowi's "regulative policy" fall on the later developed "competitive regulatory policy" (Ripley and Franklin, 1980) and "economic" or so-called "old" regulation (Tatalovich and Daynes, 1984; Lilley and Miller, 1977; Weaver, 1978; Wilson, 1980). Regulative policy inherently based on the economic theory well matches with the economic dimension oflntemet policy. Furthermore the term needs to be changed into de-regulative policy in that deregulation is considered to be one of the main success factors and is a dominant policy instrument in the Internet arena. Second, distributive policy is characterized by the ease with which it can be disaggregated and dispensed unit by small unit, each unit more or less in isolation from other units and from any general rule. So, distributive policies are policies that are highly individualized decisions. "Patronage" or "pork barrel" programs like R& D can be taken as synonyms for the main characteristic of the "distributive" policy. 40

PAGE 55

Distributive policy inherently based on the political bargaining matches well with the political dimension of Internet policy and is utilized as the policy instrument of the political dimension. Third, re-distributive policy is closely related to the so-called "social equity" and "welfare state" programs so that it deals with haves and have-nots, bigness and smallness, bourgeoisie and proletariat. The aim involved is not use of property but property itself, not equal treatment but equal possession, not behavior but well-being. These broad or vague characteristics ofLowi's re-distributive policy are, more specifically and clearly, developed later by Riple and Franklin's (1980) "protective regulatory policy" concept. Especially, Riple and Franklin's division of regulative policy well fits on the division of the dimensions of Internet policy in that the criteria of division are the same. For instance, protective regulatory policy focuses on the protection of the consumer, the deprived, and the public, while competitive regulatory policy focuses on the regulation of the monopolistic supplier. On the contrary, the "social regulation" or "new" regulation concept focuses on the regulation ofthe broad concept of"non-market behaviors" and consequently mixed distributive policy with re-distributive policy. So, in this thesis, Lowi's re-distributive policy and Riple and Franklin's (1980) "protective regulatory policy" are utilized as the policy determinants of the social dimension of Internet policy focusing on enhancing the equity oflnternet usage and on protecting the consumer. 41

PAGE 56

De-regulative Policy. The economic dimension oflntemet policy emphasizes the specific policy instruments of competition, privatization, and price regulation as the policy determinants oflntemet diffusion. From the perspective ofLowi's (1979) type of policy, the policy instruments of economic dimension get their nature of (de) regulative policy. Especially, Ripley and Franklin's (1980) "competitive regulatory policy" focuses on the (de) regulation of monopolistic suppliers. Competition is measured by the level of competition (monopoly, duopoly, and competition) in the provision of public service telephone networks, especially in local service (OECD, 1999a) (Appendix D). Furthermore, privatization is measured by the level of privatization (privately owned or state owned) of major public telecommunications operators (PTOs) (OECD, 1999a) (Appendix E). Pricing regulation is measured by the PSTN charging structure on local telephony (metered or unmetered). Based on the arguable economic logic of liberalization,33 it is suggested that by introducing a combination of privatization and competition into the telecommunications sector, policy-makers can reduce or eliminate the local call charges by competition, thereby encouraging Internet use (Mann, 2000). Nevertheless, the so-called "always-on" national group of Australia, Canada, New Zealand and the United States, with high Internet performance, shows that they have a flat rate of local phone charges to the Internet caused by social consideration for 33 The arguments on the rationale for de-regulation will be further reviewed in chapter 6. 42

PAGE 57

universal access not by economic logic (OECD, 2000a).34 The combination of a fixed fee with unlimited access can cause the classic "free riders" problem in Internet service as the commons. This demonstrates that flat rates of access pricing for social consideration can independently affect Internet diffusion. Recently, an empirical study demonstrated that competition and privatization do not exert systematically significant effects on Internet diffusion (Guillen et al., 2001). Therefore, considering the invisible function of the price does not work well in the Internet market by dint of(l) the technological traits ofthe Internet, (2) the convergent trends of communications policy on the web, (3) the inherent nature of the Internet as the commons, and (4) huge socio-political network externalities, policy instruments in the political and the social dimension needed to complement a policy instruments based on a purely economic perspective. Distributive Policy. The political dimension of Internet policy emphasizes the specific policy instruments of government investment policy on National Information Infrastructure (Nil) as a policy determinant of Internet diffusion (Drake, 1995; Kahin & Wilson III, 1997). From the perspective ofLowi's (1979) type of policy, the policy instruments of political dimension get their nature of distributive policy in that the 34 The access price ofthe Internet is composed of both local call charges and ISP charges. In this study, focus will be given on the local telephony market and its pricing structure because most people access the Internet through a dial-up connection. Furthermore, it is because local call pricing structure is closely connected with government policy (OECD, 2000a). 43

PAGE 58

process of government planning and investment on Nil shows the political bargaining and "patronage" trends of"pork barrel" programs. In particular, several empirical studies have indicated that the government's credible commitment to a predictable set of rules for investment facilitates Internet diffusion (Press et al., 1998; Masaic Group, 1998; Henisz, 2000). For instance, Henisz (2000) tries to demonstrate empirically that a government's ability to credibly commit not to interfere with private property rights is instrumental in obtaining the long-term investments required for countries to experience rapid economic growth. Henisz also posits that predictable conditions for entrepreneurship are likely to promote investment growth, so investment in Internet activities, in tum, would contribute to an increase in the numbers of users and hosts. As political constraints bringing high predictability in policy-making, he emphasizes (1) the number of independent branches of government with veto point, and (2) the distribution of preferences across and within those branches. Furthermore, to show the relationship between branch interaction and Internet diffusion, he argues that (I) branch interaction working as veto point provides a positive but diminishing effect on the total level of constraints on policy change; (2) the rate at which infrastructure penetration grows is positively related to limits on the feasibility of policy change or constraint on political discretion. Therefore, branch interaction working as veto point on political discretion is positively related to the 44

PAGE 59

infrastructure penetration growth; thus political interactions across branches can be a direct indicator of Internet diffusion. For the purpose of the empirical testing, the Henisz' index derives a new measure of political constraints from a simple spatial model of political interaction that incorporates information on the number of independent branches of government with veto power and the distribution of preferences across and within those branches. Re-distributive Policy. The. social dimension oflnternet policy emphasizes a policy goal of increasing social interaction, which can be achieved by a universal service funding mechanism for universal access. Government sponsorship of public access in the spirit of social equity can affect the diffusion of the Internet (Press, 1997), as it did with the national telephone companies. From the perspective ofLowi's (1979) type of policy, the policy instruments of the social dimension get their nature of re-distributive policy. Ripley and Franklin's (1980) "protective regulatory policy," focusing on the protection ofthe consumer, the deprived, and the public, well fit with the social dimension oflnternet policy. The specific subsidies on universal service are compared as to whether or not there is a universal service funding mechanism or direct funding for universal access (OECD, 2000b) (Appendix G). In detail, the determination of the coverage of universal service that requires social consideration (rather than economic rationale) is a policy matter designed to 45

PAGE 60

encourage certain classes of user. For example: under the Telecommunications Act of 1996, the U.S. has subsidized certain classes of users, especially through subsidizing university, school, and library connections, while Singapore provides a free Internet account and subsidized computers for any school teacher who wants one (and over 90% ofthe teachers have them) (Press, 1997). So, the ministry and the legislature determine the coverage of universal service, while the technical cost finding and allocation of universal service stipulated in the telecommunications law is understood as an economic regulatory dimension in the majority of countries (OECD, 2000b ). There are different approaches to ensure universal service provision, i.e., Finland and Iceland utilize direct subsidy from government as a social assistance of welfare state to the lower-income subscribers; in some countries, like the U.K., the incumbent operator like British Telecom (BT) bears the responsibility for universal service provision. In other countries, like the U.S., this responsibility is spread across the industry via the universal service funding mechanism. Where telecommunications operators share responsibilities, the independent regulator determines the cost of universal service as well as the cost allocation among telecommunications operators (Cherry et al., 1999; OECD, 2000b ). 46

PAGE 61

Summary Limit ofMonocausal Studies. Monocausal theories, especially those based on economic perspectives, have long been dominant in the causal theories of Internet diffusion (Levy & Spiller, 1994, Megginson & Netter, 1999; Gutierrez & Berg, 2000). Yet, considering the malfunction ofthe price in the Internet market caused by (1) its technological traits, (2) the nature of the Internet as the commons, and (3) large socio political network externalities, the explanatory power of the economic perspective is being challenged (as reviewed before). Furthermore, the main controversy of monocausal arguments is that the U.K. with full liberalization shows us a low Internet host penetration rate, whereas Iceland, with a monopoly and stated-owned network, especially demonstrates a high Internet host penetration rate. Such a policy controversy of monocausal theory between liberalization and Internet diffusion can be solved by multicausal theories of Internet diffusion. Observations. Through reviewing the causal studies oflnternet diffusion, the specific policy instruments can be added to each dimension oflnternet policy: that is, de-regulative policy can match with the economic dimension of Internet policy; distributive policy can match with the political dimension oflnternet policy; andredistributive policy can match with the social dimension oflnternet policy. 47

PAGE 62

Diachronic Process Studies In contrast with the synchronic approach to the determinants of Internet diffusion reviewed above, the process models apply a diachronic approach to Internet diffusion patterns by focusing on the adoption process and behavioral characteristics of individual adopters (logistic model) or interactive network externalities (interactive model). To conduct a comparative policy analysis in view of each stage of Internet diffusion, we briefly look at both the brief history oftechnological development of the Internet and the process models of interactive medium, which will bring us useful implications on both the trends oflnternet diffusion and their relationships with the communications policy, respectively (Flynn & Preston, 1999: Rai et at., 1998). Technological Development of Internet Since its emergence in the late 1960s, the Internet has undergone constant technological transformation. The technological development of the Internet can be distinguished into three successive generations (Sim and Rudkin, 1997; Baret al., 2000). First Generation. The first-generation Internet from the late 1960s to the early 1990s was a network and social engineering prototype of interest to military and research organizations. The central first-generation applications were file transfers and e-mail. The Internet began in the late 1960s as ARPANET, a project of the Advanced 48

PAGE 63

Research Projects Administration of the U.S. Department of Defense, designed to link together universities and defense contractors using TCP/IP packet-switching technology ( Gromov, 1996). As ARPANET grew during the 1970s and early 1980s, several similar networks were established, primarily between universities. In the mid-1980s, the National Science Foundation (NSF) created the first backbone for the U.S. portion of the Internet, the NSFNET (Moschovitis, 1999). NSFNET provided connectivity to NSF's supercomputer centers and provided a high speed backbone for developing the Internet (MacKie-Mason and Varian, 1997). The NSF subsequently awarded a contract to a partnership of Merit (the Michigan Educational Research and Industrial Triad, a nonprofit regional network for the state of Michigan), IBM, MCI, and the State of Michigan to upgrade NSFNET, and to interconnect several additional research networks (Werbach, 1997). The new NSFNET completed in 1988, initially connected thirteen regional networks. Connections to the federally-subsidized NSFNET were generally free for the regional networks, but the regional networks generally charged smaller networks a flat monthly fee for their connections. Second Generation. From the early 1990s to the late 1990s the second-generation Internet saw the mass adoption and commercialization of narrowband access, largely through dial-up connection to the telephone network providing intermittent, low-49

PAGE 64

bandwidth connection. Commercialization and the explosion of the World Wide Web constituted the main event of the second generation. In 1992, the NSF announced its intention to phase out federal support for the Internet backbone, and encouraged commercial entities to set up private backbones. The NSF paid about $11.5 million annually for the NSFNET operation for several years, but eventually decided that the technology was mature enough that it could be more effectively provided by the private market (Varian, 1996; Mackie-Mason et al., 1995). 35 When the NSFNET backbone was shut down on April 30, 1995, the transition to the new privatized network went relatively smoothly. The NSF continues to fund some regional nets, but this funding is steadily decreasing to zero over five years. The World Wide Web was developed by Tim Berners-Lee, a computer programmer at CERN, a research lab for high-energy physics in Switzerland. BernersLee's idea, was to use hypertext to link together different documents on the Internet, which later allowed whole pages of text, pictures, sound, music, voice, animation, and video to be transmitted over the Internet through a new technology called Hyper Text Transmission Protocol (HTTP). Due to both commercialization and the Web's many capabilities, the number of hosts increased dramatically from 1.8 million in 1993 to over 16 million in 1997 (Kantor and Neuharth, 1996; Rutkowski, 1996) (Figure 2-5). 35 For Merit's overview of the New Network Architecture, which provides details of the privatization plan, see the online: http://www.merit.edu/nsf.architecture/index.html 50

PAGE 65

Figure 2-5: Growth of Internet Hosts: 1989-1997 18,000,000 16,000,000 14,000,000 12,000,000 10,000,000 8,000,000 6,000,000 4,000,000 2,000,000 Jan-89 Jan90 Jan-91 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97 Jan-98 Source: Westphal and Towell, (1998:27) Figure 2-5 shows the trends of Internet diffusion during 1987-1997, and the beginning oflnternet diffusion is familiar with the s-shaped development, which leads to review of the logistic model. Third Generation. From the late 1990s, the third-generation of the Internet can be characterized by the fact that massive users are about to experience "always-on" highspeed access to the Internet from their homes. The technological innovation of broadband access to the Internet through cable, Digital Subscriber Lines (DSL), and wireless Internet is the main event of this generation. In 1996, the research and education community of the U.S., with the support ofthe White House and several federal agencies, announced the "Internet II" or "Next Generation Internet" (NGI) initiative to establish a new high-speed Internet backbone dedicated to non-commercial 51

PAGE 66

uses.36 By mid-2000 the number ofDSL connections to the Internet still numbered less than two million in the OECD area (OECD, 2001b). Worthy of note is that more than half these subscribers were in Korea. During 2000 Korea witnessed a surge in broadband subscriptions. By mid-year, Korea Telecom was reporting an increase in DSL subscribers from 11,925 at the end of 1999 to 545,553 by July 2000. Logistic Model and Limits When considering the diffusion of the Internet, it is useful to start with one of the most influential generic models of the diffusion process, Everett M. Rogers' The Diffusion of Innovation (first published in 1962 and now in its fourth edition). Furthermore, it is also useful in that the beginning of Internet diffusion is similar to the s-shaped development. For Rogers, diffusion can be described as "the process by which an innovation is communicated through certain channels over time among the members of a social system" 1983:12), thus suggesting a contagion effect, or "snob and bandwagon effect" is the basis for the innovation-diffusion process. The diffusion process is influenced by four elements: innovation characteristics, communication channels, time elapsed since the innovation was introduced, and the social system. 36 See "Clinton Announces Moves for Improving Access to the Internet" Wall Street Journal, October 11, 1996, and "Internet 2 Project General Information," available on the Web at http://www.internet2.edu/about i2. 52

PAGE 67

Based on the three assumptions(1) composition ofthe social system itself is considered unchanging, (2) homogeneity is assumed over time, and (3) external factors are considered irrelevant in the diffusion process -the rate of diffusion is modeled solely as a function of interactions between existing and potential adopters in the social system. Rogers proposes the bell-shaped frequency curve37 and the s-shaped cumulative curve (logistic curve) that can be applied to any diffusion (Figure 2-6). Figure 2-6: Bell-shaped Frequency Curve and S-Shaped Cumulative Curve for Adopter Distribution 100% 90% 80% 70% 60% Percentage of Adopters 50% 40% 30% 20% 10% Source: Rogers, (I 983: 243) Time Cumulative S-shaped curve Bell-shaped Frequency curve In this regard, several recent studies show us that Internet diffusion can be characterized to date by the s-shaped growth patterns (Rutkowski, 1996; Rai et al., 37 The bell-shaped normal frequency distribution of the adopters is divided into five categories: innovators (2.5%), early adopters (I 3.5%), early (34%), later majority (34%), and laggards (16%). 53

PAGE 68

1998; Press, 2000). The logistic curve especially echoes in virtually all respects the Product Life Cycle curve used in marketing circles (Flynn and Preston, 1999). Both curves posit that in the introductory phase, an innovation/new product initially diffuses slowly as early adopters begin to purchase it. This is followed by a growth take-off phase as early adopters begin to convince others of the product's value. The third phase is the maturity phase or late adopter's phase where the product reaches and plateaus at its life cycle peak sales. Finally, the product enters a phase of decline as the market is saturated. Since its first publication, this model has been the subject of heated debate, not least within the interdisciplinary fields concerned with communications and socio economic development, where it was criticized for failing to take into account specific institutional contexts and socio-political power structures (Mattelart and Mattelart, 1998). Furthermore, this model provides little idea as to where the inflection point might be, that is, where a product might be located on the s-curve at any given time. Interactive Model and Communications Policy The utility theory ofthe interactive communication medium (Gurbaxani, 1990) suggests that users subscribe to a technology only if it provides a net positive utility. Eli Noam (1994) has proposed a model that attempts to explain telephone network diffusion in terms of the relationship between the cost of telephone access and the utility ofthe telephone network. Noam proposes the four stages of telephone network 54

PAGE 69

diffusion: growth by external subsidy (0n1), self-sustained growth (n1-n2), directed growth (entitlement growth) (n2-n4), and growth by external subsidy (nc). In the first stage, or below a critical mass point (n1),38 a network will not be feasible unless subsidized by external sources, either by government or by the network operator's willingness to accept losses in the early phases of operation because where the network is small, average cost is high and externalities small. Beyond that point the network will grow on its own. Through the second stage (n1-n2), which Noam called the "cost-sharing phase," the network users can lower their cost by adding members. However, at some point average costs increase, and utility plateaus. The private optimum point is n2 Left to themselves, the existing subscribers of the network would not accept members beyond that private optimum. From a social point of view, however, social welfare still increases at n2 because the positive utility to additional network users is not considered by the existing network participants should they stop expanding at n2 under the assumption of a network with an equal-price system. The insiders do not take the outsiders into account. lfthe benefits are added, the social optimum n3 lies between n2 and n 4 This implies that, in the third stage (n2-n4 ), the socially optimal size (n3 ) will therefore not 38 Critical mass is defined as "a small segment of the population that chooses to make big contributions to the collective action, while the majority do little or nothing" (Oliver et al., 1985: 524; Pearce, 1998). Oliver et al. (I 985) developed the theory of critical mass to integrate theory on phenomena variously labeled "snob and bandwagon effects," "the free rider problem," and ''the tragedy of the commons" (Markus, 1987). 55

PAGE 70

be reached by itself, but by some external governmental direction through required expansion. It may be achieved by a differentiated pricing scheme or through some internal politics of expansion, but study will focus on the external governmental direction. The final point (14) lies beyond the point at which the net benefits of the network will be negative. Beyond that point, or the fourth stage (14-), the network would again need outside support to exist (Figure 2-7). Figure 2-7: Noam's Diffusion Model of Telephone Network Average cost 0 02 Critical Private Social Exit Network size LT_::rint I optrLI u_m ___ o_p-tim-r--u-m ___ P_,fy Growth by Self-sustained Entitlement growth Growth by External growth (directed growth) External subsidy subsidy Source: Noam (1994: 687-704). 56 Utility n

PAGE 71

Figure 2-7 shows that initially high average costs decline rapidly as more subscribers join the system (i.e. as the utility increases). This fall in costs continues until the system has expanded to a stage whereby new subscribers can only be added by accepting an increasing average cost. Utilitl9 increases dramatically at first followed by a reduction in the rate of increase as the number of new subscribers begins to tail off when the telephone system enters its mature stage. This figure also shows that falling average cost40 leads to an increase in subscriber numbers. Yet at the same time, because the telephone network is an interactive system, the increase in subscriber numbers also increases the utility of the system as a whole, generating a further increase in subscriber numbers. This model is also relevant to the present conceptual framework in that it deals with the diffusion process of interactive network and policy implications for each stage of Internet diffusion. Summary Process vs. Causality. Process models have applied different mathematical formulae to various time-series data, and reached a general agreement that most diffusion processes are s-shaped curves. Although the s-curve model reveals the socioeconomic characteristics of innovation adopters, it overlooks such crucial factors 39 Utility measured in terms of subscriber numbers: "the more people can be reached, the more useful is the network." 40 AC can be measured in terms of the basic cost of system entry-i.e., rental cost. 57

PAGE 72

as government policies in Internet diffusion emphasized by causal theories. Causal theories, on the other hand, tend to use regression methods to analyze cross-sectional data, but generally pay little attention to the longitudinal and dynamic processes of Internet diffusion. Considering the distinctive merits of both process models and causal theories, a dynamic understanding of policy changes in the diffusion process may deepen our comprehension of the role of different policy determinants in Internet diffusion process, and vice versa. Observations. Through reviewing the process studies on Internet diffusion, the specific policy pattern in each stage oflnternet diffusion can be added to each dimension oflnternet policy. In the initial stage oflnternet diffusion or below a critical mass point, government initiation of Internet network can be justified because huge amounts of the initial investment cause a network not to be feasible unless subsidized by external sources. So, the political dimension of Internet policy is well matched with the initial stage of Internet diffusion. In the second stage of Internet diffusion or beyond a critical mass point to a private optimum point, enhancing the entrepreneurship of service providers through liberalization can be justified in that the Internet network can achieve self-sustained growth. Thus, the economic dimension oflnternet policy is well matched with the second stage of Internet diffusion. 58

PAGE 73

In the third stage of Internet diffusion or beyond a private optimum point, government guidance of growth toward enhancing social equity of service users can be justified in that the Internet network can not achieve the socially optimal size. So, the social dimension of Internet policy is well matched with the third stage of Internet diffusion. Synthetic Comparative Studies Comparative analysis could be difficult because different political, economic, and social structures and systems render the analytic comparison of various issue areas problematic at best, and misleading at worst. Peter de Leon (1979: 10) Comparative Studies in General The comparative study can be characterized by its extreme diversity (Peters, 1998), which used to be considered its strength and weakness. On the one hand, the field of comparative study is so diverse that it can scarcely be called a field at all (Wiarda, 1986: 5). On the other hand, the heterogeneity ofthis field will continue, and this diversity is a source of strength rather than of weakness in that the openness of the field to various theories and methodologies helps to maintain its vitality and its capacity to cope with a rapidly changing policy world (Verba, 1986: 36). Considering the diversity of comparative study, a focus of this review is in the literature that is wed matched with the triad of Internet policy: that is, Richard Rose's comparative dynamics, Theodore Lowi's policy typologies, Thomas Dye's causes & consequences of public 59

PAGE 74

policy, and Heidenheimer et al's choices as framework. For the purpose of logical consistency, Giovanni Sartori's (1994: 14-34) basic questions are used, that is, (1) why compare? (2) what is comparable? (3) compare how? Comparative Dynamics. Among the other diverse literature on comparative analysis (e.g., Dierkes et al., 1987; Ashford, 1992; Steinmo et al., 1992; Dogan and Kazangigil, 1994; Baker, 1994; Peters, 1998; Castles, 1998; Maor and Lane, 1999; Heady, 2001), Richard Rose's (1993:71-76) "lessons drawing across time and space" is useful in that his synthetic perspective combining time with space is well matched with the present thesis. He classifies the methods of comparison into four categories in tenns of both. across space (yes or no) and across time (yes or no): routine satisfaction, trend analysis, comparative statics, and comparative dynamics (Figure 2-8). Figure 2-8: Methods of Comparison across Time and Space 0 z rJl (j) >No Routine Satisfaction Comparative Statics: League tables Source: R. Rose (1993 :72) Across Time Yes Trend Analysis Comparative Dynamics: Lesson-drawing 60

PAGE 75

Rose's focus is on the comparative dynamics where lesson-drawing occurs about closing the performance gap by catching up with the performance of leaders. For the purpose oflesson-drawing, he suggests analyzing in sequence: (1) the initial state in a society prior to public recognition that a policy need exists, (2) how policy issues are placed on the "agenda of public controversy," (3) how demands are advanced, (4) the importance of the form of government for policy deliberations, (5) available resources and existing constraints, (6) the move towards a policy decision, (7) the determinants of governmental choice, (8) the context of the choice, (9) implementation, (1 0) the production of outputs, ( 11) policy evaluation, and ( 12) feedback (Rose, 1973). His analytic framework provides an appropriate guidance to the comparative study in general. Why Compare. Regarding the objects ofthe comparison, Przeworski (1987: 35) asserts that a "consensus exists that comparative research consists not of comparing but of explaining. The general purpose of cross-national research is to understand .... In similar fashion, Regin (1987: 6) holds that comparative knowledge provides the key to understanding, explaining, and interpreting. Especially, Mayer (1989: 12) redefines comparative studies as a field whose goal is the building of empirically falsifiable, explanatory theory. Mayer's goal of comparative analysis is closely related to the method of ideal type that has the virtue of providing a standard against which real world systems can be 61

PAGE 76

compared. Its typical examples are Weber's (1949) ideal type model, Easton' (1965, 1979) system model based on the input-output analysis, and Lowi's (1979) policy typology41 of distributive, regulatory, and redistributive policy. In this thesis, the method of ideal type is to be used for the purpose of drawing a general pattern of Internet policy across time and space. What Compare. With regard to the question of the areas and issues to compare, Dye (1998:7) presents a useful summary of the complex linkages that operate in the process of policy-making (Figure 2-9). According to his framework, the primary concern of this dissertation is represented by arrows B, C, and D. Regarding arrow B, the effect of government as a policy system on public policies will be compared and contrasted. As the theoretical types of policy systems for the mature use of the Internet, this dissertation proposes the three different policy systems i.e., a national Chieflnformation Officer (CIO), an independent regulatory body, and a national and functional ministry-in view of the three dimensions oflnternet policy. First, a nationai"CIO can be ideally matched with the political dimension of Internet policy in that a national CIO is involved in the allocation of the information technology in gerieral, or the Internet resources in specific. 41 The main point of criticism of these typo!ogies is that the categories of policy types are not used mutually exclusively (Bressers and Honigh, 1986). For example, distributive policy is at the level of policy goals, whereas regulatory policy is at the level of policy instrument. Nevertheless, the basic concept of policy typology gives us a useful insight into the type of the policy. 62

PAGE 77

Considering the multidisciplinary nature and impacts of the Internet as an interactive social network, the Internet-related policy issues can be interrelated issues among the functional ministries in the era ofthe industrial society. In this respect, a new organizational scheme for coordination in the information society can be termed as a national CIO. The examples of institutionalizing this new organizational scheme are the so-called "e-Minister" and "e-Envoy" in the U.K. In the U.K., thee-Minister has overall political responsibility for the Government's Information Age agenda, and chairs the so-called "Information Age Ministerial Network" purpose of which is to ensure that the social, economic and e Government strands ofthe Government's information age programs are combined as an integrated strategy by developing a shared vision (Chatrie and Wraight, 2000). An Envoy structure is composed of an e-Envoy and an e-Envoy office that is leading the drive to get the "UK online" (i.e., the American firstgov.gov). Second, an independent regulatory agency can be ideally matched with the economic dimension of Internet policy in that an independent regulatory body is inherently involved in the (de) regulation ofthe service providers considering its own origin ofbirth with the logic ofthe separation of regulatory function from the policy function. Third, a traditional and functional ministry can be ideally matched with the social dimension of Internet policy in that the protection of consumer and social equity of the public is considered a policy function of the traditional ministry. 63

PAGE 78

Regarding arrow C, both the effect of the economic input from the market on Internet policy (the economic dimension oflnternet policy) and the effect of the social input from the community on Internet policy (the social dimension of Internet policy) will be compared and contrasted across time and space. Regarding arrow D, the effectiveness of policy determinants of each dimension oflnternet policy on Internet diffusion will be compared and contrasted across time and space. 64

PAGE 79

Figure 2-9: Causes and Consequences of Public Policy Society economic conditions Including: Wealth and income Inflation, recession, Unemployment Educational achievement Environmental quality Poverty Racial composition Religious and ethnic make-up Health and longevity Inequality, discrimination Political System Institutions, Processes, Behaviors D Including: Federalism Separation of powers Checks and balances Parties Interest groups Voting behavior Bureaucracy Power structures Congress, president, Courts Public Policy Policies Including: Civil rights Educational policies Welfare policies Health care policies Criminal justice Taxation Spending and deficit Defense policies Regulations Linkage A: What are the effects of social economic conditions on political and governmental institutions, processes, and behavior? Linkage B: What are the effects of political and governmental institutions, processes, and behaviors on public policies? Linkage C: What are the effects of social economic conditions on public policy? Linkage D: What are the effects (feedback) of public policies on social economic conditions? Linkage E: What are the effects (feedback) of political and governmental institutions, processes, and behavior on social and economic condition? Linkage A: What are the effects (feedback) of public policies on political and governmental institutions, processes, and behavior? Source: Dye (1998: 7) 65

PAGE 80

Compare How. With relation to the choice of the criteria of the cross-time-and space comparison, Heidenheimer et al.'s (1990) "choice as framework" gives us useful criteria of policy comparison (deLeon and ResnickTerry, 1999). They first define public policy in terms of social choice, and the effects that those choices have on individuals. They then marshall a great deal of evidence about the differences among policy areas in the several countries and are able to demonstrate some consistent patterns in the choices made (Heidenheimer et al., 1990; Peters, 1998). They argue that four types of choices are applicable to each policy area: choice of scope, choice of policy instruments, choice of distribution, and choice of restraints and innovation. Choice of scope concerns whether and where lines shall be drawn between public and private responsibilities. In other words, this type of choice is related to the rationale for governmental intervention. Choice of policy instrument includes the policy agencies to implement the specific policy program. Choice of distribution concerns the policy effectiveness from the perspective of policy typologies. Choice of restraints and innovation can be closely related to the change of policy environments and the subsequent policy shift. These criteria can provide a useful perspective for formulating the analytic framework of this dissertation. Comparative Studies and Communications Policy In the communication sector, cross-national comparison of communications policy is a commonly used method in that the diffusion of communication technology 66

PAGE 81

is an inherently worldwide phenomenon. In this regard, the four literature sources on communications policy of which the theoretical and conceptual frameworks as well as the methodologies can give useful implications for the present dissertation are reviewed briefly. The theme of this dissertation is closely related to the classical trade-off relationship between "efficiency and equity," "market and government," "private and public," "economics and politics," or "democratization and liberalization" (Dahl and Lindblom, 1953; Okun, 1974; Stone, 1997; Wolf, 1993; Rainey, 1997; Carey, 1987). Such a traditional trade-off relationship ofthe policy goals may be closely linked with the discrepancy of the policy instruments between "universal service policy and competition policy" in the telecommunications and Internet sectors. In this respect, a brief look is focused on four dissertation studies that deal with similar issues but different methods (two econometric analyses vs. two case studies). K. P. Jayakar (1999) studied the effect of telecommunications reform on universal service, in both its network growth and distributional equity dimensions. The three main macro-systemic variables affected by telecommunications reforms were identified for the purpose of the econometric analysis: industry structure (the nature and extent of competition), ownership (public, private, corporatized), and the independence of the main regulatory institution. To measure universal service, two indicators of universal service teledensity and residential teledensity share were used. Based on the statistical multiple regression model, this dissertation analyzes the 67

PAGE 82

causal relationship between reform and universal variables with the data of the selected 24 countries such as rich countries (U.K., Japan, France, Canada, and German) and poor countries (Sri Lanka, Jamaica, Thailand, Venezuela, etc.). The findings indicate that less developed countries may benefit more from telecommunications reform than their richer counterpart. S. H. Lee (1999) quantitatively investigated how the social and institutional endowments of the countries in the Asia-Pacific region affected the performance of the telecommunication sector (such as network expansion and efficiency) by using an economic model and assumptions based on transaction cost economics. As an empirical framework, a descriptive overview of general trends of telecommunications reforms undertaken in the Asia-Pacific region was made in view of organization theory. This dissertation uses a regression model to analyze statistically the political, economic, and demographic characteristics for 15 countries in the region over the period 19821995 in terms of level of network penetration, network efficiency, and growth rate of network penetration. The main findings are: (1) a country with fewer opportunities for rent seeking behavior is more likely to achieve a rapid growth rate of network penetration, and (2) government's ability to establish institutional constraints on the exercise of executive power has a positive impact on the growth of network penetration. Based on its theoretical framework on political economy, J. Y. Kim (1999) examined the broadly-defined impacts oftelecommunications deregulation on two significant public interests, universal service and free speech on the Internet. This 68

PAGE 83

dissertation selects the relevant experiences of the United States, Germany, and Japan as main areas of research concern because they represent different paths of telecommunications evolution. Based on the qualitative case studies, the findings are then synthesized and discussed in a way that re-frames the role of the state in a deregulated communication environment. The findings are that all three nations have attempted to achieve their universal service goals by creating a competitive marketplace. It also indicates that all three nations have different views on how to guarantee free expression on the Internet: a hands-off policy ofthe U.S.; a top-down approach in Germany; and a bottom-up approach in Japan. Comparing the United States and France, A. L. Fletcher (1997) analyzed the effect of divergent telecommunications regimes on public access to information. The American telecommunications regulatory framework relies on market incentives and private ownership, while the state-led French model is based on public ownership and centralized regulatory authority. Indices of telecommunication efficiency support the argument that each model can produce similar outcomes. This dissertation analyzes whether these disparate regulatory models also produce comparable levels of access to information or if this outcome varies. Access to information refers inclusively to both access to the telecommunications network and to information services carried by the network. By using a comparative analysis of binary cases, data for this dissertation are both qualitative and quantitative and this dissertation includes statistical data, primary source documents (such as telecommunications reforms laws) and archival material. 69

PAGE 84

The main findings are that the state has an important function to perform in the research and development of information technologies, the creation of an initial demand for advanced information services, and the promotion of nonprofit and civic applications of information technology. Summary Through reviewing the comparative studies oflnternet diffusion, the specific types of policy system can be added to each dimension oflnternet policy: that is, a national CIO can match with a political dimension oflntemet policy, a independent regulatory agency can match with a economic dimension oflnternet policy, and a functional ministry can match with the social dimension of Internet policy. Furthermore, considering the limits of both the pure synchronic causal studies and the pure diachronic process studies of Internet diffusion, this section suggests that a synthetic comparative dynamics, combining the synchronic causal studies with the diachronic process studies of Internet diffusion, can be adopted as a research method. 70

PAGE 85

CHAPTER3 ANALYTIC FRAMEWORK AND METHODOLOGY Based on the literature review, this chapter develops a triad oflnternet policy as an analytic framework composed of both a three-dimensional (3-D) synchronic model and a three-stage (3-S) diachronic model, and proposes five research propositions to explain a dynamic relationship between communications policy and Internet diffusion over time. These propositions are to produce an explanatory theory about a synchronic style and a diachronic pattern oflnternet policy, which is formulated to provide guidance to conduct a cross-national and pattern-matching comparison and evaluation across time and space. Analytic Framework and Propositions Three Dimensional (3-D) Synchronic Model Initially grounded on the observation of the first theoretical controversy (monocausal vs. multicausal theories), a synchronic model of this dissertation explains the causes and consequences of Internet policy that can be termed a "policy style" (Richardson, 1982). The term of Policy style is used in this dissertation for the purpose of showing comparative and synchronic characteristics of Internet policy in a nation, 71

PAGE 86

and then portrayed as having three dimensions: economic, political, and social. To resolve the above-mentioned policy controversies oflntemet diffusion, the three key dimensions of Internet policy identified from the existing literature are to be focused. Furthermore, for the purpose of the across-time-and-space comparison and evaluation, they need to be adjusted from the perspective ofthe system model of public policy based on Easton's input-output model and Dye's policy analysis model of causes and consequences (Figure 3-1 ). Figure 3-1 shows that the 3-D synchronic model focuses on the three dimensional inputs from state (politics), market (service providers), and community (service users) and the three dimensional outputs (distributive, deregulative, and re-distributive). Figure 3-1: System Model oflntemet Policy Policy Input (I) Policy Output (I) State (Politics) Distributive Policy Policy Input (II) Policy System Policy Output (II) Policy Impact Market r--'1' Implementing .... De-regulative ---1! Internet (Providers) Institution Policy Diffusion Policy Input (III) Policy Output (III) Community Re-distributive (Users) Policy 72

PAGE 87

Three-Stage (3-S) Diachronic Model To complement the static design of the above synchronic model oflnternet policy, a diachronic model is designed to show the dynamic changes of the policy pattern. The term of policy pattern is used in this dissertation for the purpose of showing the comparative and diachronic characteristics oflnternet policy in the Internet diffusion process. The diachronic model is grounded on the observation of the second policy controversy (process vs. causal models) in the literature review. To resolve the second controversy, the two process models identified from the existing literature need to be further refined to formulate empirically testable propositions under the guidance of a consistent analytic framework of a triad of Internet policy. For this purpose, the logistic model and Noam's model can be combined (Figure 3-2). Figure 3-2: Combined Interactive Process Model Dollars Social Optimum (n3 ) Private Internet Host or (Subscribers) Stage I Stage 2 Stage 3 73 Time

PAGE 88

Figure 3-2 shows the main reason why the two process models are combined. The policy implications ofNoam's model are drawn from the interaction between subscribers (or hosts) and dollars, while the s-shaped diffusion trends of interactive medium ofRogers' model are drawn from the interaction between subscribers (or hosts) and time. Therefore, for the purpose of getting the policy implications of Internet diffusion across time, the two models need to be combined to show the interaction between subscribers (or hosts) and time. The combined diachronic model can thus show the policy implications in each stage of Internet diffusion. The combined interactive model can be applied to Internet diffusion in that the diffusion of the Internet has already shown itself an initial s-shaped pattern since its creation. In addition, Noam's final stage (nr) will not be examined in this dissertation in that the policy implications of the final stage simply mirror or replicate that of the first stage. Furthermore, a short review of the historical development of Internet technology shows that the chronological evolution of Internet technologies manifests itself as an on-going process and thus the adequate governmental response to the technological evolution is not "policy termination" but "policy succession" in an interactive medium (Hindle et al., 1997). For instance, for the purpose of facilitating the technological evolution of the Internet from the second generation of Internet (like "dial-up" or "narrow-band" connections) to the third generation oflnternet (like "Internet II" or 74

PAGE 89

"broad-band" Internet), what is needed by a government for successful technological transformation is not exit but an active initiation (OECD, 2001 b). Therefore, in this dissertation, the three stages of Internet diffusion initiation (0n1), proliferation (n 1 ni), and control of proliferation (n 2 -n4)-can be distinguished by the critical mass point (n1), private optimum point (n2), and exit point (14) respectively. Furthermore, a diachronic 3-D model of this thesis can provide the policy implications both in each stage and of policy system (i.e., an external funding and a national CIO in the first stage, a self-sustained growth and an independent regulatory agency in the second stage, and a social control and a functional ministry in the third stage). Research Propositions Under the guidance of the 3-D synchronic and 3-S diachronic models, we make five hypothetical propositions addressing the dynamic relationship of communications policy to Internet diffusion across time and space for the purpose of deriving both policy implications and the appropriate role of government in each stage of Internet diffusion. Policy Style and Policy Pattern. In view ofthe fact that the Internet market gets its problem ofthe commons (Brin, 1995; MacKie-Mason and Varian, 1996) because of its inherent nature of public goods and CPRs, the over-arching proposition is that a 75

PAGE 90

policy style (Dye, 1998; Lowi, 1979) and a policy pattern (Rogers, 1983; Noam, 1994) of communications policy affecting Internet diffusion can be generalized across time and space (Rose, 1993). Proposition 1: A synchronic policy style of communications policy affecting Internet diffusion in a nation can be generalized in view of its policy input, policy system, and policy output. Proposition 2: A diachronic policy pattern of communications policy affecting Internet diffusion in a nation can be generalized in view of each stage of Internet diffusion. In addition, the second research proposition can be divided into three subpropositions in view ofthe three stages of Internet diffusion. Political Dimension and Initiation. Political Imperative. Both the huge amounts of initial investment on Nil and its characteristics of natural monopoly and externalities caused by the unique nature of the social network as public goods are hypothesized to bring government intervention and initiatives to digitize the existing network in the first stage of Internet diffusion.42 The system of information superhighways with government support during the 1990s used to be analogous to the interstate highway system built during the 1950s with 42 Interestingly enough, Hanaro Telecom, the second private-owned local telephone company in Korea initiated the Digital Subscriber Lined (DSL) network with the "non-economic" willingness to accept the loss in 1999, and now the market share in local service in 2001 is just 1.8 percent of the total (MIC, 2001). In this dissertation, such an initiation by the network operator's non-economic willingness to accept loss is to be set aside as a given. 76

PAGE 91

government support in the U.S. (Shew, 1993). So, the prominent "before marketization" issues and sequential government initiatives affect Internet diffusion in the first stage of Internet diffusion. National Information CIO and Distributive Policy. In this stage oflnternet diffusion, a higher level of a national information CIO can affect Internet diffusion by envisioning the national goal of building the so-called "information society" and by initiating a national backbone plan such as a national information infrastructure (Nil). This goal can be achieved by policy instruments of investment or subsidies to Nil, which can be classified into a distributive policy in view ofLowi's policy typology. Research Proposition. In the first stage of Internet diffusion (initiation), "before marketization" issues, such as the huge amounts of initial investment and natural monopoly, and network externalities of Internet service as public goods, are posited to generate government initiative and intervention. So, political imperative from the state (political goal-setting of achieving an information society), and its subsequent distributive policy by a national CIO affect Internet diffusion in the first stage of Internet diffusion. Proposition 2a: Political imperative from the state (politics) and its distributive policy by a national CIO for mature use of the Internet are the comparative characteristics in the first stage of Internet diffusion. 77

PAGE 92

Economic Dimension and Proliferation. Economic Imperatives. In this stage, "marketization" issues, such as the network congestion caused by the undersupply of Internet service as common pool resources, are posited to result in privatization and competition. Privatization and competition are necessary so that the fixed-line and wireless telecommunications infrastructure can support and encourage the growth of Internet traffic (Mann, 2000). As such, economic imperatives from the market, especially from the supply-side service providers for the development of the Internet infrastructure, are expected to affect Internet diffusion in the second stage of Internet diffusion. Independent Regulatory Bodies and (de)regulative Policy. In this stage, policy programs of liberalization (privatization, competition, and price regulation) which can be classified into a (de)regulative policy in view ofLowi's policy typology can affect Internet diffusion. Furthermore, an independent regulatory body used to be involved to secure a fair environment for the competition. The economic policy goal and instruments are traditionally identified as the best policy determinant of Internet diffusion in many causal analyses. To achieve the economic goal of creating national wealth by developing an efficient technology, the policy instruments of competition and privatization are frequently used as the indicators of economic deregulation. Furthermore, considering the independent influence of local call pricing on Internet diffusion, the structure of local call charges needs to be added. The indicator of 78

PAGE 93

competition is the level of competition in the provision of public service telephone networks, especially in local service (OECD, 1999a) (Appendix D). The key indicator of privatization is the level of privatization of major public telecommunications operators (PTOs) (OECD, 1999a) (Appendix E). The indicator ofpricing structure is the PSTN charging structure on local telephony (Appendix F): that is, whether it is metered or unmetered (flat rate). Research Proposition. In the second stage of Internet diffusion (proliferation), "marketization" issues, such as the network congestion caused by the undersupply of Internet service as common pool resources, are proposed to result in privatization and competition. So, economic imperative from the market and its subsequent de-regulative policy by an independent regulatory agency affect Internet diffusion in the second stage oflnternet diffusion. Proposition 2b: Economic imperative from the market (service providers) and its de-regulative policy by an independent regulatory agency for the mature use of the Internet are the comparative characteristics in the second stage of Internet diffusion. Social Dimension and Control ofProliferation. Social Imperatives. In this stage, "after marketization" issues of narrowing digital divide is supposed to bring the call for more governmental involvement in the Internet and its application. The "social optimum" is hard to be achieved by the market forces via the invisible function ofthe price due to the technological traits of the 79

PAGE 94

interactive Internet, as seen in Noam's process model. This is one of the primary reasons why many arguments call for greater government involvement in the Internet and its applications in the stage of"after marketization" (Noam, 1994; Westphal & Towell, 1998; Shapiro, 1999; Lessig, 1999). Government involvement will focus on the demand side of the Internet market, that is, achieving the social optimum of the user of Internet service. National Ministries and Re-distributive policy. In this stage, the policy program of enhancing public access to the Internet that can be classified into a redistributive policy in view ofLowi's policy typology can affect Internet diffusion. Furthermore, national ministry as a policy maker used to be involved. The social goal of increasing social equity via the Internet can be achieved by the policy instruments of the subsidies on universal service. The indicator of the subsidies on universal service is whether or not there is a universal service funding mechanism or direct funding for universal access (OECD, 2000b) (Appendix G). Research Proposition. In the third stage of Internet diffusion (control of proliferation), "after marketization" issues of narrowing digital divide are supposed to bring the call for more governmental involvement on the Internet and its application. Therefore, social imperative from the community and its subsequent re-distributive 80

PAGE 95

policy by a functional ministry affect Internet diffusion in the third stage of Internet diffusion. Proposition 2c: Social imperative from the community (service users) and its de-regulative policy by a functional ministry for the mature use of the Internet are the c,omparative characteristics in the third stage of Internet diffusion. Summary This dissertation developed the 3-D and 3-S analytic frameworks that is composed of the several trichotomous variables of Internet policy in view of their synchronic 3-D causes and consequences and their diachronic 3-S changes of the times: that is, (1) major variables (network, service providers, and users), (2) allocation mechanisms (state, market, and community), (3) theoretical dimensions (political, economic, and social), (4) system styles (national CIO, independent regulatory agency, and functional ministry), (5) policy styles (distributive, de-regulative, andre-distributive), and (6) policy stages (initiation, proliferation, and control of proliferation). The hypothetical relationship among these variables can be summarized as in Figure 3-3 and identified as in the three dimensions. 81

PAGE 96

Figure 3-3: Theoretical Policy Styles and Patterns L-__ P_o_li_cy_In_p_ut __ _.l I Policy System I ._I __ P_ol_ic_y_O_u_tp_u_t ___JI .__I __ I_n_te_rn_e_t _, Political Imperative Distributive Policy Stag_e 1 Huge Amount of National Initial Investment .... Information National Planning Before Achieving an CIO Investment Marketization Information Society (Network) J Economic Imperative (De) regulative Policy Stage 2 Government Failure Independent Positive Network .... Regulation I+ Competition Externalities Agency Privatization Marketizati on Network Congestion Price regulation (Providers) I Social Imperative Re-distributive Policy Stage 3 Market Failure Negative Network __. Ministry __. 1 ..... After Externalities Universal Service Marketization Digital Divide (Users) ...... Policy Patterns (Vertical Line, Across Time, Diachronic Characteristics) _____..Policy Styles (Horizontal Line, Across Space, Synchronic Characteristics) In the political dimension, political imperative from the state (national planning of Nil) and subsequent policy program of investment in Nil that can be classified into a distributive policy in view ofLowi's policy typology can affect Internet diffusion. In this dimension, a higher level of national planner such as a national Chief Information Officer (CIO) used to be involved to project a national route to an information society. 82

PAGE 97

In the economic dimension, the imperative from the market (especially, from the supply-side service providers) and subsequent policy program of liberalization (privatization, competition, and price deregulation) that can be classified into a (de)regulative policy in view ofLowi's policy typology can affect Internet diffusion. In this dimension, the independent regulatory body is involved to secure a fair environment for the competition. In the social dimension, the imperative from the community (especially, from the demand-side service users) and subsequent policy program of enhancing public access to the Internet that can be classified into a re-distributive policy in view ofLowi's policy typology can affect Internet diffusion. In this dimension, a national ministry as a policy maker is involved to promote social equity. Research Methods By choosing a method of comparative dynamics, a preferred pattern of Internet policy over time is tested across space. For this purpose, the three nations are chosen as a most different system design: the U.S., Finland, and Korea. The scope and the focus of this study are on the three synchronic dimensions (3-D) oflnternet policy and the three diachronic stages (3-S) of Internet diffusion. Furthermore, a cross-national comparison and evaluation is conducted based on a pattern-matching strategy. For this purpose, the pattern variables oflnternet policy are also identified in terms of four criteria for the comparison; that is, choice of scope (policy rationale), choice of policy 83

PAGE 98

instruments (policy agency and program), choice of allocation (policy effectiveness), and choice of restraints and innovation (policy shift). Choice ofthe Method: Comparative Dynamics Under the guidance of the 3-D and 3-S frameworks and the theoretical propositions, a generalized pattern of Internet policy across time is to be empirically tested across space using a comparative policy analysis strategy. By using the method of "comparative dynamics" (Rose, 1993) in a sense that a general pattern of communications policy for the mature use of the Internet is tested across time and space, this dissertation intends to propose descriptive theory to understand, explain, and interpret the Internet-related policy controversies (Mayer, 1989: Regin, 1987). Furthermore, this thesis utilizes a comparative case method as a distinctive form ofmultiple-case studies (Lijphart, 1975; Agranoff & Radin, 1991). The present dissertation explores a case study method because "the boundaries between phenomenon and context are not clearly evident" in that Internet-related phenomena are multidisciplinary in nature and the "researcher deliberately wants to cover contextual condition" articulated in the 3-D and 3-S frameworks (Yin, 1994: 13). According to Yin's types of case study, this thesis is both exploratory and descriptive in that it tries to build a theory to describe the contemporary Internet-related policy controversies. 84

PAGE 99

Choice ofNations: Most Different System Design The present dissertation selects the three relevant cases of the U.S., Finland, and Korea as the "most different system design" to test whether or not a theory about a pattern oflntemet policy across time can be generalized across space. Thus, among the nations with high Internet performance (hosts or subscribers) showing the different roads to the mature use of the Internet, each case is systematically selected in view of both driving imperatives (or motives) from the three major players and policy styles oflnternet policy so that it produces contrasting results but for predictable reasons (a "theoretical replication") about the comparative motives and styles oflntemet policy (Yin, 1994).43 From the perspective of the comparative characteristics of the policy surroundings conditioning communication policy for the mature use of the Internet, an individualistic and pluralistic policy surroundings of the U.S. can be compared and contrasted with the communitarian and deliberative policy surroundings of the Finland (Venturelli, 1995). The driving motive and the policy style for achieving an information society in the U.S. can be characterized by a market-driven strategy based on a trust in the free market ethos, while those of Finland can be characterized by a society-driven strategy based on a trust in the welfare society ethos. In addition, in Korea, sharing the same policy surroundings of the Confucian cultural traditions with 43 According to Yin's (1994: 45-46) "the replication logic," analogous to that used in multiple experiments, is the same as the logic underlying the use of multiple-case studies. For more information about the replication logic, see Hersen, M & D.H. Barlow (1976). 85

PAGE 100

the other far eastern Asian nations like Japan and Singapore, the driving motive and the policy style can be characterized by a government -driven strategy based on a trust in the "strong state" ethos. Therefore, the different driving motives and the styles oflnternet policy for achieving an information society in each of the three nations can be compared and contrasted-the free market ethos and its subsequent policy style of the U.S., the welfare society ethos and its subsequent policy style of Finland, and the high state ethos and its subsequent policy style of Korea. The systematically-chosen three nations are matched with the triad of Internet policy. The synchronic differences of the imperatives and policy styles of Internet policy in the three nations are to be further investigated in chapter 4. Choice of Focus and Scope: 3-D & 3-S The field of inquiry must be narrowed to focus on the manageable three sets of variables of Internet policy, that is, political, economic, and social dimensions. More specifically, the causes and consequences oflnternet policy and the implementing agencies and the specific programs oflnternet policy in the three nations are to be compared and contrasted in each dimension oflnternet policy. According to Dye's diagram (I 998: 7), the primary focus is on (I) the affect of government as a policy system on Internet policy, (2) the affect of the economic input from the market on 86

PAGE 101

Internet policy, (3) the affect of the social input from the community on Internet policy, and (4) the effectiveness of Internet policy on Internet diffusion. In addition, among the other Internet-related policy issues, this thesis focuses on the policy issues regarding the investment in Nil, liberalization, and universal access. The scope of case studies that are to be compared and contrasted is restricted to the initiation, proliferation, and control of proliferation of the Internet among the four stages ofNoam's (1994) model. Furthermore, among the other access technologies to the Internet (such as cable modem, wireless, and satellite), the dial-up connection to the Internet needs to be focused, not because most users access the Internet through a dial up connection, but because communications policy concentrates on the local telephone facility as the essential (bottleneck) one. In addition, the time frames for the national programs for the mature use of the Internet are not synchronous. Not all the three nations initiated the Nil program at the same time.44 So, the relative developmental styles and patterns of Internet policy in the three nations are to be compared and contrasted. Finally, according to Graham Allison's (1999) three models, the unit of analysis ofthis study is the combination of Model I (the National Actor) and Model II (the Organizational Actor). Allison's Model III (the Individual Actor) is not considered to focus on the national and organizational characteristics in a nation. 44 For further information on government planning of Nil in the three nations, see Table 3-1 (page 94). 87

PAGE 102

In comparing and contrasting the driving imperatives for achieving an information society and the Internet performances, the three nations are considered singular or unitary policy-making bodies. In this respect, this dissertation falls into Allison's national actor model. Furthermore, in comparing and contrasting the implementing institutions of Internet policy and the specific policy programs in a nation, the middle-ranged organization will be considered as a policy-making unit to show the complexity of policy-making process for the mature use of the Internet. In this respect, this study falls into both what Allison has termed the Organizational Actor, the Model II and what Ostrom (1999) later has termed the Operational Level (agency-level decision).45 Choice ofthe Mode of Analysis: Pattern-matching Comparison For a mode of analysis, this study intends to use a pattern-matching logic (Trochim, I 989), which compares an empirically based pattern with a predicted one. In this dissertation, a predicted ideal pattern oflnternet policy over time will be compared with the empirically based patterns of Internet policy over time in the three nations. If the patterns coincide, the results can help a case study strengthen its internal validity (Yin, I 994: I 06I 07). Furthermore, pattern-matching comparison of Internet policy across time and space will strengthen its external validity by (1) predicting similar 45 E. Ostrom (1999) classifies the level of institutional analysis into three; that is, collective choice level (the statute governing the legislature), constitutional level (the constitution governing the legislature), and the operational level (agency-level decision). 88

PAGE 103

results (a literal replication) or (2) producing contrasting results but for predictable reasons (a theoretical replication). For this purpose, the more specific pattern variables need to be identified. Choice of Criteria: System Analysis and Choice as Framework For the purpose of cross-time-and-space comparison and evaluation of the relevance of communications policy for the mature use of the Internet in the three nations, the three policy models i.e., (1) Heidenheimer et al.' s (1990) four types of choice, (2) Easton's (1965, 1979) input-output model, and (3) Dye's (1998) policy analysis model of causes and consequences are applicable to this thesis as the criteria of policy comparison and evaluation. Heidenheimer et al.' s choice of scope concerns whether and where lines shall be drawn between public and private responsibilities. In other words, this type of choice is related to the policy rationale for governmental intervention in the Internet arena. This criterion is well matched with Easton's input and Dye's arrow C (see pages 65 and 72). In this dissertation, the political imperative from the politics in the first stage, the economic imperative from the service providers in the second stage, and the social imperative from the service users in the third stage in the three nations can be compared and contrasted as policy input in each stage oflnternet diffusion. Heidenheimer et al.'s choice of policy instrument includes both the policy agencies to implement and the specific policy program. This criterion more broadly 89

PAGE 104

includes Easton's policy system and output and Dye's arrow B. In this dissertation, the investment plan on Nil by a national CIO in the first stage, liberalization (local competition, privatization, local pricing structure) by an independent regulatory agency in the second stage, and universal service by a functional ministry in the three nations can be compared and contrasted as the policy instruments in each stage of Internet diffusion. Heidenheimer et al.'s choice of distribution, concerning the policy relevance from the perspective of policy typologies, and choice of restraints and innovation, concerning the policy shift can be simply represented by Easton's policy impact and Dye's arrow D. In this dissertation, policy relevance and the diachronic relationship of communication policy to Internet diffusion in the three nations are compared and contrasted as policy impacts on each stage oflnternet diffusion. More specifically, in the first stage of Internet diffusion, (1) the relevance of distributive policy in the three nations and (2) the diachronic relationship of the Nil programs to Internet diffusion can be compared and contrasted in view of the focus of subsidization (i.e., network side or service-provider side or service-user side). In the second stage of Internet diffusion, ( 1) the relevance of de-regulative policy in the three nations and (2) the diachronic relationship of liberalization programs to Internet diffusion can be compared and contrasted in view offacility-based competition and the lowering of the local telephone charges through liberalization. 90

PAGE 105

In the third stage of Internet diffusion, (I) the relevance of re-distributive policy in the three nations and (2) the diachronic relationship of the universal service programs can be compared and contrasted in view of narrowing the digital divide. Data Collection and Manipulation Data Collection Primary data for this thesis are collected by the documentation for the purpose of broad coverage of the multidisciplinary Internet-related regulatory issues-i.e., long span oftime, many events, and many settings (Yin, 1994). Furthermore, this dissertation utilizes the so-called methodological triangulation of both quantitative data (empirical data from OECD and ITIJ) and qualitative data (government reports on NII)46 to calibrate the balance between specificity and generalizability (Patton, 1987; Creswell, 1994). The vast majority of the material is in analog form, mostly in print, but has also been deposited in digital form on the World Wide Web. The websites ofthe international organizations that provide a good deal of relevant information include those of the World Bank,47 the International Telecommunications Union (ITU),48 and the Organization for Economic Cooperation and Development (OECD).49 The empirical data will be collected from the official 46 For the purpose of empirical testing, quantitative primary data are attached in Appendix A to 0. 47 see at http://www.worldbank.org 48 see at http://www.itu.int 49 see at http://www.oecd.org 91

PAGE 106

reports ofOECD (Communications Outlook) and ITU (Telecommunication at a glance) in the years 1993-2001. The ITU and the OECD are the main organizations that track cross-national differences in the Internet sector. The ITU, headquartered in Geneva, Switzland, is an international organization within which governments and the private sector coordinate global telecommunications networks and services. At its website, more than 4,000 titles concerning international telecommunications standards, policy analyses, and statistics are available in printed fonn. The OECD was fonned by 20 European and North American countries in 1961. Nine countries have since joined the OECD. Currently, the OECD has a somewhat diverse membership, including some less developed nations such as South Korea, Mexico, and Hungary, but it is still largely known as an organization of the advanced industrial nations. Among many items available at its website, Communications Outlook provides updated infonnation about Internet diffusion and its policy. In addition, the government websites providing useful materials include those of the U.S. Federal Communications Commission (FCC), 50 the U.S. National Telecommunications and Infonnation Administration (NTIA),51 the Finnish Ministry ofTransport and Communications, 5 2 the Finnish Communications Regulatory Authority (FICORA),53 the South Korean Ministry oflnformation and Communication 50 see at http://www.fcc.gov 51 see at http://www.ntia.doc.gov 52 see at http://www.mintc.fi/www/sivut/english/default.html 53 see at http://www.ficora.fi/englanti/index.html 92

PAGE 107

(MIC) and the Korea Communications Commission (KCC).54 The secondary sources include periodicals, and empirical studies obtained from economic, law, and other scholarly publications. Data Manipulation This dissertation investigates the hypothetical relationship of the "three dimensions of Internet policy" to the "three stages of Internet diffusion" given the non governmental variables oflntemet diffusion. Especially, OECD and ITU provide the official and quantitative data on both government policy and Internet diffusion, which can be manipulated into a comparative evaluation of the member-nations including the U.S., Finland, and Korea. Political Dimension. The political dimension of Internet policy in each nation can be compared by the specific policy programs of national planning and investment in Nil (Table 3-1 ). 54 see at http://www.mic.go.kr and http://www.kcc.go.kr respectively. 93

PAGE 108

Table 3-1: Government Reports on Nil U.S.: Finland: Korea: The national information infrastructure: agenda for action (15 September 1993, NTIA/IITF) National information infrastructure: progress report ( 1993-1994, NTIA/IITF) Global information infrastructure: agenda for co-operation (31 January 1995, Vice President ofthe United States, NTIAIIITF) Finland's way to the information society: An action plan (1994, Ministry of Finance): Finland I Developing a Finnish information society: decision in principle ( 1996, Council of State): Finland II Korean information infrastructure: blueprint for implementation (1995, Ministry of Information and Communication) CYBER KOREA 21: an informatization vision for constructing a creative, knowledge-based nation (March 1999, Ministry oflnformation and Communication) Especially, the three nations of this dissertation show the different time frames for their planning: the U.S issued its government plan first in 1993 with several revisions until 1995; Finland issued its plan in 1994; and Korea issued first in 1995 with revision in 1999. These time frames ofNII planning in each nation can be compared with the time frames ofthe first stage of Internet diffusion in each of the three nations. Economic Dimension. The economic dimension of Internet policy can be measured by the specific policy programs of liberalization (i.e., local competition, privatization ofPTOs, and local charging structure). 94

PAGE 109

The competition in each nation can be compared by the level of competition (monopoly, duopoly, and competition) in the provision of public service telephone networks, especially in local service in a sense that this thesis focuses on the dial-up connection to the Internet (OECD, 1999a) (Appendix D). Furthermore, privatization in each nation can be compared by the level of privatization (privately-owned or state owned) of major public telecommunications operators (PTOs) (OECD, 1999a) (Appendix E). Local pricing structure in each nation can be compared by whether it is metered or unmetered (or flat rate of local phone charges to the Internet) (OECD, 2000a). The time frames ofthese specific liberalization programs in the three nations can also be compared with the time frames of the second stage oflnternet diffusion in each nation. Social Dimension. The social dimension oflnternet policy can be measured by the specific policy programs ofthe universal service funding mechanisms for closing or narrowing the so-called "digital divide." The specific subsidies on universal service can be compared and contrasted by whether or not there is a asymmetric universal service funding mechanism focusing on re-distribution among the service providers or a direct funding mechanism for universal access focusing on re-distribution among the service users (OECD, 2000b) (Appendix G). The time frames of these specific universal service programs in the three nations can be compared with the time frames of the third stage oflnternet diffusion in each nation. 95

PAGE 110

Policy System. The data about the implementing agencies in each stage of Internet diffusion can be gathered from the official websites of the three nations by tracking after the specific policy programs of each stage oflnternet diffusion (i.e., the national planning ofNII in the first stage, the program for liberalization in the second stage, and universal service funding mechanism in the third stage). Internet Diffusion. In this dissertation, Internet diffusion is measured by the three indicators: Internet host per 1000 inhabitants, secure servers per 1 million inhabitants, and Internet subscribers (or users) per 1 00 inhabitants. The data on the three indicators can be gathered from both OECD Communication Outlook and ITU Year Book of Statistics (Appendices A-E). These data give a specific time frame about the three stages oflnternet diffusion in the three nations. Limitations The conclusion of this dissertation can be limited by the arguments based on their validity. Three types of validity can be identifiedconstruct validity, internal validity, and external validity (Yin, 1994). First, construct validity is closely related to the following question: Does the operational definition fit with the reference concepts? It can be tainted by the researcher's expectancies. To cope with the construct validity threat, this thesis tries to calibrate the balance between specificity and generalizability 96

PAGE 111

by using a data triangulation of quantitative and qualitative information (Patton, 1987; Creswell, 1994). Second, internal validity can be expressed by the following set of questions: Can we assert causality among variables? If so, which direction? Presence of internal validity threats causes a net bias in our measurement of our ability to establish a relationship between the variables (Cook and Campbell, 1979). To cope with the internal validity threat, this study will utilize the analytic tactic of pattern-matching (Yin, 1994). Finally, external validity can be expressed by the following questions: Is the study population representative of the target population? Can one generalize across types of persons and types of settings? Yin ( 1994) posits that the external validity problem has been a major barrier in doing case study because of using analytic generalization. To cope with the external validity threat, this thesis utilizes both a literal replication and a theoretical replication. Using a replication logic, this analysis may be applicable to the nation-states where the highly developed communications network including wired or wireless is established in that communications facilities can be considered a prerequisite for the development of the Internet network. Furthermore, this research is strongly applicable to the diffusion of interactive communication technology such as broadband access to the Internet (DSL) and digital TV. 97

PAGE 112

CHAPTER4 NATIONAL POLICY STYLES AND INTERNET DIFFUSION This chapter examines what are the comparative and synchronic characteristics of national policy styles for the mature use of the Internet for the purpose of answering the first research question (see p.7). For this purpose, the three nations-the United States, Finland, and Korea-are chosen as a most different system design among the three different national groups -the "always-on" group, the Scandinavian group, and the East Asian group showing the unique cultural, economic, and geographical circumstances respectively. Keeping in line with these differences, the three different models of national policy styles for the mature use of the Internet across space are identified as the policy factors affecting Internet diffusion based on the synchronic 3-D analytic model. Finally, the technological factors affecting Internet diffusion in the three nations are compared to identify the technological similarities of the three nations as a high IT intensity group. Three Nations and Internet Diffusion In terms of both the regional diffusion rate of Internet and the regional traits of national policy styles, the three different roads to the mature use of the Internet can be identified among the nations of achieving high Internet performance: "Always-on" 98

PAGE 113

national group in North America and Oceania, the Scandinavian national group in Europe, and the East Asian national group in Asia. Global Diffusion of Internet The Internet has been developing at a stunning pace throughout the world. Its diffusion, however, has been uneven, or has caused the so-called "international digital divide,"55 not only between developed and developing nations, but also among the developed nations with similar socio-economic development as well as within individual nations (Kelly and Petrazzini, I997; ITU, 2000; OECD, 2000a). By October 2000, there were just 94 million Internet hosts in the world. Some 95.6% of these hosts were in the OECD area (Appendix I), just 4.4% outside the OECD area. The best indication ofthe international digital divide, in relation to the Internet, is in the penetration rate for Internet hosts. By October 2000, there were 8I.5 Internet hosts for every I 000 inhabitants in OECD nations. In contrast, there were only 0.85 Internet hosts per I 000 inhabitants outside the OECD area. All indications are that in 200I the OECD area will surpass IOO Internet hosts per 1000 inhabitants whereas non-OECD nations will have just one Internet host per I 000 inhabitants. The concentration of Internet hosts in the OECD area means that on a regional basis, North America and Europe account for 89% of all Internet hosts. The regions 55 For more information on the "international digital divide," see G8 Kyushu-Okinawa Summit Meeting 2000 (http://www.g8kyushu-okinawa.go.jp/), OneWorld.net Digital Divide Campaign (http://www .oneworld.net/campai gns/di gitaldivide ), and Know Net Initiative (http://www .knownet.org). 99

PAGE 114

with the third and fourth largest shares are Asia and Oceania. In contrast, the regional share of Internet hosts is very low in Central and South America and Africa. Among non-OECD nations, the 15 nations with the largest number of Internet hosts represent 92% of the total number outside the OECD area (Appendix J) (OECD, 2001a). Among them, Taiwan, Singapore, Hong Kong and Israel together account for 52% of all hosts outside the OECD area. Always-on National Group and United States This national group includes the "always-on" national group ofthe United States, Canada (North America), Australia, and New Zealand (Oceania) (OECD, 2000a). This group is typified by both the first rank in Internet diffusion and the leading role in the full-blown liberalization ofthe communications market (based on the trust in the free market ethos). Market Liberalization. Based on the free market ethos, this national group shows their leading role in the liberalization oftelecommunications service. By 1990 the United States and New Zealand had liberalized (i.e., de-regulated) the provision of public switched telecommunication network (PSTN). Between I 990 to 1995, the pioneers of liberalization were joined by Australia and Canada (Green and Teece, 1999). While the different models of market liberalization in both long distance and local service can be identified in this group, the underlying objectives of market 100

PAGE 115

liberalization have been broadly similar, namely the introduction of competitive pressure on incumbent telecommunications operators based on the free market ethos (Cuilenburg and Slaa, I 994; OECD, 1997b; Hills, 1998). With respect to the liberalization of the long distance service market, the modes of liberalization can be distinguished into three different models. The first model was adopted by the United States, the first OECD nation to begin opening its market to competition, where the process of liberalization began as early as I 956 (Kellerman, 1997). In 1982, the U.S. opened up the long distance market to competition, but allowed the regional companies "structurally separated" from AT&T to operate as local monopolies with the FCC implementing detailed regulations for both markets (Philip and Tsoi, 1988; Catinat, 1997; Brock, 1998; OECD, 1998c). The second model of market liberalization initially adopted by the United Kingdom was to convert a monopoly market structure into that of a duopoly (Philip and Tsoi, I 988; Cave, 1997). The reasoning was that completely opening the market to competition would result in the new entrants competing against and weakening each other. With a duopoly situation, the new entrant could concentrate on trying to reduce the incumbent's market share. A modification of this approach has been adopted by Australia with the temporary nature of the duopoly period being well defined so as to give notice to both market participants that the market would be completely open to competition by a certain date (Joseph, 1993; Xavier, 1997). 101

PAGE 116

The third and more radical model was adopted by New Zealand, where liberalization took place in April 1989 without any market entry restrictions. Moreover, no "sector specific regulatory framework"56 was enacted and reliance was placed on existing competition law framework, under the mandate of the Commerce Act (Joseph, 1993). The government, however, explicitly indicated that it would consider more extensive regulation should the need arise. This model of immediate and full liberalization with "light-handed" regulation was a significant contrast to other nations that have tended to phase in competition and manage the competitive process through sector specific regulation (Green and Teece, 1999). In relation to the liberalization of the local service market, there are two fundamentally different views of what a competitive end state should look like (Green and Teece, 1999): that is, a retail model oflocal competition (U.S.) or a model of facilities-based local competition (Australia and New Zealand). U.S. policy makers envision a network of networks where providers compete in the retail distribution of network services. "The local network is to be unbundled at every technically feasible point"57 so that providers of retail services can use these elements of the network in a piecemeal fashion (Katz, I 997; Harris and Kraft, I 997; Noll, I 998; Brock, I 998; Cimatoribus et al., 1998). Competitors can combine the 56 In this regard, a sector-specific approach of the U.S. to telecommunications regulation (telecommunications-specific rules) is contrasted with a general antitrust approach of Australia to telecommunications regulation (general antitrust rules). For more information, see Kerf and Geradin, 1999. 57 This can be considered the real cornerstone of local competition regulation in the U.S. since it forces the incumbent to establish a wholesale business for disaggregated elements of the network. 102

PAGE 117

purchase of incumbent firm's facilities with their own facilities, may rebundle them with other purchased network elements, or they may simply resell retail services. Policy makers in the other nations of this group, including the U.K. in particular, envision the construction of alternative networks that will compete for subscribers (Philip and Tsoi, 1988; Cave, 1997). In this vision, customers would have a choice of networks and not just of retailers. goal of this model is to foster the development of facilities-based competition. Facilities-based competition allows competitors to succeed when their networks and operations are more efficient than that of the existing firm. These types of policies discourage inefficient competition but allow competitors to enter who can offer differentiated services and/or be of equal or greater efficiency. Unmetered Access to the Internet. This group as the pioneer of market liberalization can be characterized by an unmetered local telecommunications charges structure for the access to the Internet (users pay a flat rate per local call), which is important in facilitating Internet diffusion. In fact, this group shows a high penetration rate oflnternet hosts and secure servers (Appendices A and C). In the United States, Canada, and New Zealand, residential users in large part have unmeasured local rates. In Australia, users pay a flat rate per local call irrespective of the duration (OECD, 2000a). Among the members ofthis group, the U.S. shows its leading role in Internet diffusion considering both the American origin of the Internet network and the high 103

PAGE 118

performance in Internet hosts (Appendix A), web sites (Appendix B), secure servers (Appendix C), and subscribers (Appendix D). Scandinavian National Group and Finland This group includes the Scandinavian or Nordic national group of Finland, Sweden, Norway, Iceland, and Denmark. This group is typified by both the information welfare society vision (based on the trust in social solidarity) and the second rank in Internet diffusion. Information Welfare Society Vision. This group is characterized by a powerful commitment to collective social responsibility for the optimal welfare of citizens based on the strong trust in the so-called "Scandinavian model of welfare society" (Erikson et al, 1987; Esping-Andersen, 1990; Kolberg, 1992; Greve, 1996; Cababrese and Burgelman, 1999; Rodger, 2000). Some critics may envision that the Scandinavian nations have been more different than alike in their embrace of the Scandinavian welfare state, and that welfare state principles have not applied only to Scandinavian nations, and thus the term "Scandinavian model" seems a misnomer (Mjoset, 1992). Nevertheless, this group shows several comparative characteristics in pursuing social welfare (Erikson et al., I 987, Kolberg, 1992). 104

PAGE 119

An important characteristic of the Scandinavian model is the extent to which rights on a normal living standard are divorced from market criteria (Erikson et al., 1987). The individualistic self-reliance of market criteria has been replaced by a communitarian commitment to collective social responsibility for the optimal welfare of citizens in this model (Kvist, 1999). Another feature ofthe Scandinavian model is the way in which welfare expenditures are financed (Erikson et al., 1987). The underlying principle has been to spread financial responsibility in a solidaristic way across society, rather than to tie benefits to individual contribution (Esping-Andersen, 1990). General government revenues therefore play a heavy role in total social expenditure finance. The last feature ofthe Scandinavian model lies in broad public participation in various areas of economic and social life, the purpose of which is to improve the ability of society to master its problems, and to enrich and equalize the living conditions of individuals (Erikson et al, 1987; Greve, 1996). Historically, there are heated debates on the validity of the welfare state model in the era of industrial society between market-pros (such as liberals and conservatives) and market-cons (such as social democrats) (Esping-Andersen, 1990; Roger, 2000). Market cons (Erikson et al, 1987; Esping-Andersen, 1990) defend the welfare state and believe that it should distribute income and social services according to universal principles of equity and social justice. Market-pros (Hayek, 1989, 1991; Friedman, 1982, 1993) criticize the welfare state for interfering with the market and undermining family and community responsibilities for social welfare. Furthermore, new public 105

PAGE 120

management or a reengineering movement (Osborne and Gaebler, 1992) offers a new llne of attack on the welfare state by suggesting that social services should be decentralized and privatized to better serve diverse social needs. Recently in the era of information society, there is a theoretical trend transforming or rejuvenating the welfare state model of industrial society into a welfare society model of information society with good work conditions and empowered citizens (Pestoff, 1998; Riis, 1996). Most interestingly, Pestoff(l998) argues, using his terminology of "social enterprises" and "social accounting," that the alternative to the state is not the market, but rather complementary third-sector solutions. He thus posits that "social enterprises," which can be either worker cooperatives, consumer cooperatives or voluntary organizations, can serve as a way to rejuvenate the welfare state. His theoretical argument is that since social enterprises are based on other values (fraternity, self-help, and benevolence) than state and market alternatives, social enterprises demand both multi-stake holder control and social accounting (instead of traditional accounting) as the most relevant methods to evaluate performance. Internet Diffusion and Direct Subsidies. This national group with information welfare society vision shows a high Internet penetration rate among EU member nations (Table 4-1 ). Among the seven nations with the highest Internet penetration, five are the Nordic nations. Furthermore, Finland stands out as the leader. 106

PAGE 121

Table 4-1: Internet Hosts per I 000 Inhabitants in Scandinavian Nations Iceland Norway Sweden Finland Denmark EU OECD January 1998 63.7 65.6 36.1 87.5 30.4 11.2 26.2 July 1998 77.5 71.6 43.0 99.9 36.3 14.4 32.5 January 1999 81.7 73.3 49.0 106.4 53.0 18.2 *Sources: OECD: 'Communications Outlook 1999',0ECD, Paris, 1999, page 86, and Network Wizards, HYPERLINK http://www.nw.com. Among many of the factors considered to be critical to success in the development of the Internet in Nordic nations, some are included below (Henten and Kristensen, 2000; OECO, 2000a): (I) Nordic nations were among the first to connect to NSFNET. (2) Nordic nations have some of the most competitive telecommunication markets and several nations liberalized markets ahead of most OECD nations. (3) Nordic nations have traditionally had the highest penetration rates for PSTNs. (4) Nordic nations have had the highest cellular penetration in Western Europe with the major mobile manufacturers like Nokia in Finland and Ericsson in Sweden (Table 4-2). 107

PAGE 122

Table 4-2: Telecommunication Access (Lines per I 00) in Scandinavian Nations 1990 1995 Fixed Mobile Fixed Mobile Iceland 51.4 3.9 55.3 11.5 Norway 50.3 4.6 56.1 22.5 Sweden 68.3 5.4 68.4 22.7 Finland 53.5 4.5 55.0 19.7 Denmark 56.6 2.9 61.3 15.8 OECD 39.1 0.8 45.6 6.8 1997 Fixed Mobile 56.7 24.0 62.6 38.0 68.0 35.8 55.6 41.7 63.6 27.4 48.9 15.6 *Source: OECD Communications Outlook 1999,0ECD, Paris, 1999, pp.74 and 76, and JTU Yearbook of Statistics, ITU, Geneva, 1998, pp.52, 81, 77, 124 and 153. (5) The prices oflnternet access, and of the underlying telecommunication networks, are among the lowest in the OECD area. (6) Based on the information welfare society vision, this national group uses direct subsidies for securing universal access to the communication network. East Asian National Group and Korea This group includes the East Asian national groups of Japan, Korea, Taiwan, Singapore, and Hong Kong. This group is typified by both a government-led catch-up strategy (based on the trust in the high stateness ethos) and the third rank group in Internet diffusion. Government-led Catch-Up. This group is characterized by a government-led catch-up strategy based on the trust in the high stateness ethos. The high stateness ethos 108

PAGE 123

is closely connected with Asian cultural tradition, that is, so-called "Asian values" or "Confucianism" (Lim, 2000; Hamilton and Biggart, 1988). Asian values are characterized by the following: ( 1) priority of society over the individual, (2) importance of emotional bonds based on reciprocal obligations, as in family, rather than legalistic relationships based on voluntary contract, and (3) power ideally legitimated by virtue and the consequent and comparative weakness of both adequate feedback mechanisms and an institutionalized system of checks and balances. With relation to the second element, Japanese family-like enterprises are theorized into "theory Z" (Ouchi, 1980, 1984), and Chinese family-like management is described as a "nesting box" system (Omohundro, 1981; Redding, 1980). With respect to the third element, men of power were supposed to be men of virtue in traditional Asian politics, and rulers were expected to check themselves against the abuse of power (Lim, 2000). Thus, there was little inclination to establish an equilibrium by depositing power in the hands of the rulers and then constructing a "countervailing force to oppose it" like a western political system of check and balance (Hahm, 1986). This element of Asian values is ironically described by a western observer as" The Politics ofthe Vortex." Gregory Henderson writes (1968:5): The physics of Korean political dynamics of the society upward toward central power ... Vertical presses cannot be countered because local or independent aggregations do not exist to impede their formation or to check the resulting vortex once formed. 109

PAGE 124

These Asian values are embedded in the relationship established between the state and the business sector. While three different models on the relationship can be identified in this group as in Hamilton and Biggart's distinction (1988: S77-S85), the underlying trust in the high stateness ethos is broadly similar, especially compared to the relationship established between the state and the business sector. The Korean government/business relations were described to follow the fonn of what Hamilton and Biggart called the "strong state" model. In Korea, the state was described to participate actively in the public and private spheres ofthe economy and be in fact the leading actor (SaKong, 1980). The state achieved its central position through centralized economic planning and through aggressive implementation procedures by an economic planning agency like the extinct "Economic Planning Board." The Japanese government was portrayed to have developed quite a different relationship with business, what Hamilton and Biggart called the "strong intennediate power" model. The state policy toward business was one of creating and promoting strong intennediate powers, each having considerable autonomy, with the state acting as coordinator of activity and mediator of conflicting interest (Johnson, 1982). The Taiwanese government was described to have developed quite a different relationship with business, what Hamilton and Biggart called the "strong society" model of state/business relations. The government in Taiwan was by no means weak, but promoted what might be identified as "virtually free condition" (Little, 1979: 475), 110

PAGE 125

or "planning within the context of a free economy" (Myers, 1984: 522). Such policies had allowed familial patterns to shape the course of Taiwan's industrialization; this had in turn Jed to decentralized patterns of industrialization, a low level of firm concentration, and a predominance of smalland medium-sized firms. Hong Kong and Singapore followed Taiwan, or the Chinese model of state/business relations to develop highly industrial societies (Nyaw and Chan, 1982; Redding, 1980). In short, the key strand of the East Asian approach to development stressed the need for a bureaucracy, able to conceive and implement the design of a "strong state," and credibly committed to long-run development (World Bank, 1993, 1999; Ohno, 1998, Stiglitz and Yusuf, 2000). A second strand was governments' activist policies to quicken the pace of industrialization and export an increasing proportion of industrial output. Internet Diffusion and Government-led Nil. The above mentioned "high stateness" ethos developed in the era of industrial society is embedded in the East Asian national strategy to develop an information society: that is, government-led catch-up ofthe first and second group of Internet diffusion. This regional.group shows the third rank in the worldwide Internet diffusion (Appendix I). Especially, Singapore and Korea show high Internet penetration among this group. Most interestingly, Korea shows the first rank of both Internet subscribers and DSL subscribers in OECD (OECD, 200lb). Ill

PAGE 126

Internet Diffusion in Three Nations Among the three indicators of Internet diffusion-Internet hosts per I 000 inhabitants (Appendix A), secure servers per I million inhabitants (Appendix B), Internet subscribers per I 00 Inhabitants (Appendix C)-gathered from both OECD Communication Outlook and ITU Year Book of Statistics, the ITU (2001a) gives the more comparable data on Internet users per 100 inhabitants for 10 years (or, 19912000) in the three nations. As ofDecember 2000, the estimated Internet users per 100 inhabitants are 34.65 in the U.S., 37. 22 in Finland, and 40.25 in Korea (Appendix N). Furthermore, the chronological developments of Internet diffusion for I 0 years ( 19912000) in the three nations show the initial s-shaped patterns that will be further reviewed on chapters 5, 6, and 7 (Appendix 0). Policy Styles and Internet Diffusion Based on the investigation on the three different regional roads to the mature use ofthe Internet, three different comparative policy styles can be identified across space: that is, (I) a de-regulative model ofthe U.S. among the "always-on" national group; (2) a re-distributive model of Finland among the Scandinavian national group, and (3) a distributive model of Korea among the East Asian national group. Based on the 3-D synchronic model, the comparative differences of national policy styles for the mature use of the Internet can be identified in terms of both the driving motives (or described in this dissertation as imperatives or policy inputs) from 112

PAGE 127

the three major dimensions (3-D) in a triad oflntemet policy and the characteristics of their policy systems. De-regulative Model: United States Case Among the three major imperatives from the triad oflntemet policy, the U.S. strongly supports the market function that can be called the "free market" ethos, and heavily depends on a independent agency as a regulatory system under the name of the separation of regulatory function with policy function. Those comparative characteristics can be called a "de-regulative model" of national policy style for the mature use of the Internet. Policy Imperative (Free Market Ethos). The United States initiated worldwide liberalization of the communications market and has provided the world with a successful model of the approach (Kellerman, 1997). The specific measures for market liberalization, such as privatization and deregulation, were relatively easy to adopt in a country where "the ingrained free market ethos ... fosters a presumption that all consequences of increased competition ... will be beneficial" (Blumler, 1991 :213). Because of the historical recognition of the state as a threat to personal and commercial freedom, the American government's role in determining the competitive environment of the communications sector has been limited to regulating the "monopolistic 113

PAGE 128

dominance" ofthe service providers. Instead, the market is considered a neutral regulatory mechanism. John Zelezny (1997: 35) put it this way: Experience with suppressive policies of Great Britain during the American colonial period caused many Americans ... to be feaiful of any centralized government. Furthermore, Lowenstein and Merrill (1990: 216) presented the following five historical ideas that have notably advanced and perpetuated the marketplace model in the United States: I. The philosophy of freedom and individualism originating in the Age of Reason in Europe 2. The hard-work-and-competition value of the Protestant Ethic (well explicated by Max Weber in The Protestant Ethic and the Spirit of Capitalism) 3. The influence of Social Darwinism, with its concepts of "natural selection" and "survival of the fittest" as applied to business competition 4. Adam Smith's theory of capitalism, built on a laissez-faire, free market approach wherein the laws of supply and demand determine the flow of goods and services 5. Oliver Wendell Holmes' notion of a "marketplace of ideas" (updating Milton's self-righting process of three centuries earlier), which he saw as based on the "power of thought to get itself accepted in the open competition of the market" Indeed, such a trust in the free market ethos is believed to be wholly compatible with the American commitment to the liberalization of the communications market. As we reviewed before, the U.S. demonstrates its leading role in the liberalization of long distance and local service even among the "always-on" national 1 14

PAGE 129

group. The recent intent of the U.S. to facilitate development of competition in the telecommunications market, especially in the local market, can be represented by the Telecommunication Act of 1996. The Act was implemented by three major FCC orders known as the "Competition Trilogy" -the rules for ( 1) local interconnection or socalled local competition (FCC Order 96-325), (2) universal service (FCC Order 97157), and (3) access charge reform (FCC Order 97-158) (Harris and Kraft, 1997; Noll, 1998). De-regulative Policy. With relation to the comparative characteristic of national policy style for the mature use of the Internet, the strong trust of the U.S. in the free market brings in the more "de-regulative" policy style. Its typical examples are privatization ofNSFNET and subsequent network openness, which can be considered the major two of the main determinants of global proliferation of the Internet (Oxman, 1999, Werbach, 1997). Privatization of the NSFNET. In 1993, the NSF issued a request for proposals outlining the structure of a new, privatized, and commercialized Internet (Varian, 1997). The final privatization plan58 had four main components: First, the NSF would eliminate all financial support for the NSFNET backbone. Second, the NSF would phase out support for the regional networks. Third, the NSF would help fund the 58 For Merit's overview of the New Network Architecture, which provides details of the privatization plan, see the online: http://www.merit.edu/nsf.architecture/index.html 115

PAGE 130

creation ofNAPs (network access points) at which multiple backbone providers could interconnect. Fourth, the NSF would fund development of a very high-speed back-bone network service to serve as a test site for network research and to provide support for scientific applications. The NSF paid about $I I .5 million annually for the NSFNET operation over several years, but eventually decided that the technology was mature enough that it could be more effectively provided by the private market. The movement toward the privatization of the NSFNET is one example of U.S. de-regulative policy and can be justified as a market solution to the problem caused by so-called "network congestion." Network Openness. America's remarkable success in promoting the Internet revolution owes a major debt to determined regulatory action that encouraged all aspects of network openness: that is, the openness of both the Internet and the underlying telecommunication infrastructure (Oxman, I 999). The most important technical feature of the Internet is its openness, which allows any user to develop new applications and to communicate with virtually any other user. This openness is driven by the sharing of that common communications protocol, the Internet Protocol (IP) (Bar et al., 2000). In addition to the technical traits of the Internet, promoting ever-greater openness ofthe U.S. telecommunications infrastructure has been a significant theme of U.S. regulative policy and an important factor in the Internet's success (Oxman, 1999; Bar 116

PAGE 131

et al., 2000). As a "retail model" of local competition, the FCC chose to unbundle "network elements," the functional elements of the network, rather than to regulate end services (Harris and Kraft, 1997; Noll, 1998). This policy allowed a variety of actors to take basic network building blocks and combine them in diverse ways. So, hundreds of Internet service providers offered unlimited dial-up Internet access over that inexpensive phone line for $ 13-29 per month in the U.S. (Waldon, 1998). Policy System (Independent Regulatory Agency). The regulatory structure in the U.S. is a complex web that involves the interface of jurisdiction over sector-specific regulation between the states (Public Utility Commissions, PUCs) and the federal government (Federal Communications Commission, FCC), the relationship between sector-specific regulation (FCC) and antitrust law (Federal Trade Commission), as well as between these agencies and the courts (OECD, 1998c). Among the various actors and agencies influencing public policy, the focus for reviewing the policy system is on the implementing agencies ofNII programs, liberalization programs, and universal service programs. The National Telecommunications and Information Administration (NTIA), an agency within the U.S. Department of Commerce, is the Executive branch's principle voice on domestic and international telecommunications and information technology issues (OECD, 1998c). The NTIA focuses on so-called "policy functions" like 117

PAGE 132

telecommunications infrastructure and the U.S. government's legislative initiative in telecommunications, while the FCC focuses on so-called "regulatory functions." While public utility commissions (PUCs) in each of the fifty states and the District of Columbia have jurisdiction over intra-state matters, such as prices and entry conditions into local markets as well as universal service, the FCC has exclusive jurisdiction over inter-state matters, as well as intra-state matters where legislation pre empts state authority. Such a dual federal-state role in the U.S. telecommunications industry has been the source of numerous jurisdictional battles in the U.S. Courts. For example, the divestiture consent decree (i.e., Modification of Final Judgment) that led to the breakup of the AT&T monopoly was administered by the U.S. District Court for the District of Columbia. Among others, the U.S. Federal Communications Commission (FCC) is the typical example ofthe independence of regulatory agency, which is considered an important policy factor affecting Internet diffusion. The de-regulative model emphasizes that policy-makers can reduce or eliminate the local call charges, thereby encouraging Internet use, by introducing into the telecommunications market a combination of privatization, competition, and independent regulation (Wellenius et al., 1992; Mann, 2000). This model especially posits that independent regulation by an independent regulatory agency is critical to guard against a monopoly where telecommunications providers extend their rent-seeking behavior into the Internet. I I 8

PAGE 133

Re-distributive Model: Finland Case Among the three major imperatives from the triad oflnternet policy, Finland strongly supports the community function that can be called "welfare society" ethos, and can be characterized by the utilization of the existing functional implementation agency in the era of industrial society. Those comparative characteristics can be called a "re-distributive model" of national policy style for the mature use of the Internet. Policy Imperative (Welfare Society Ethos). Finnish society has experienced dramatic and extremely rapid structural changes. In the mid-1960s, Finland still had a very large agricultural sector, whereas by the mid-1980s, it was a modem industrial society characterized by a rapid expansion of the welfare state (Uusitolo, 1989). In the early 1990s, Finland experienced the most critical economic problems among the Nordic nations, with the Finnish economy suffering a severe backlash as a result of the combined effect of international recession, the effect on trade of the fall of the Soviet Union, and of the political decision to link the Finnish Mark to the Deutschmark at a time when the latter was extremely solid (Heikkila and Uusitalo, 1997; Kuhnle, 2000). Finland was no longer viewed as "Europe's Japan," as was the case in the late 1980s when the economy grew rapidly. Gross domestic product declined during each of the three years 1991-1993, in 1991 by close to eight percent. From an average unemployment rate of 4.8 percent during the 1980s, the figure during 1991-1997 average 15 percent (OECD Economic Outlook, 1997; Kuhnle, 2000). 119

PAGE 134

Finnish politics in the 1990s was the politics of cuts and retrenchment in the area of social welfare. Nearly all key social benefits were subject to cuts during 1992-1997, but cuts were made in such a way that broad political compromises secured their legitimacy and in such a way that income distribution was barely affected (Heikkila and Uusitalo, 1997). The economic downturn did not create large-scale new poverty or dramatic new social and economic inequalities in society. The institutional characteristics of the Scandinavian type of welfare state were preserved. The Finnish economy began to pick up again in the latter half of 1993, and has shown steady growth since then. Unemployment is still high, but the rate was brought down to 11-12 percent by 1999, from a peak of 17.7 percent in 1993 (Kuhnle, 2000). The Finnish experience showed that an advanced, universalistic, and re distributive Nordic type of welfare state is not a handicap when a sudden, unexpected economic crisis occurs. The politically and socially controlled cuts and policy adjustments made it easier to recuperate economically and quickly. International show that Finland is still one of the world leaders in welfare expenditure (OECD Economic Surveys on Finland, 1998; OECD Economics at a Glance, Structural Indicators, 1996; OECD Historical Statistics, 1997). Relative equality in income distribution, low poverty rates, world-class standards in public services, and many other facts testify to the robust health of the Finnish welfare state (Ministry of Social Affairs and Health, 1996, 1998). 120

PAGE 135

Re-distributive Policy (Direct Subsidization of Usage). With respect to the comparative characteristic ofthe specific type of Internet policy, the welfare society ethos in Finland brings in the more "re-distributive" policy style, especially in the area of universal access to the network. In Finland, there is a unique funding mechanism for universal service that can be called a "direct subsidization of usage." The method employed in Finland to ensure universal access to communication service is direct "social assistance" (or "income support") to the customers in question from the Ministry of Social Affairs and Health (Nattennann and Murghy, 1998). The direct "income support" is the financial assistance provided by social welfare as the last resort to secure the person's or the family's living and to promote coping independently. The content and amount of income support and other grounds for granting income support are prescribed in a separate "Act on Social Assistance" and Decree. Policy System (Utilizing the Functional Ministries). Among the various actors and agencies influencing public policy for the mature use of the Internet, the comparative characteristic of the policy systems for implementing Nil programs, liberalization programs, and universal service programs, Finland with welfare society ethos utilizes the existing functional Ministries in the era of industrial society. The Ministry of Finance is basically responsible for the distributive function of national 121

PAGE 136

planning ofFinland's way to the information society from the perspective ofthe public management of the IT -related budget allocation. The Ministry ofTransport and Communications is responsible for (de)regulative function of the communication service industry with an independent regulatory agency, the Telecommunications Administration Center (TAC). The Ministry of Social Affairs and Health is responsible for the re-distributive function of the universal access to the communications network as a social support based on the strong trust in the welfare society ethos. Distributive Model: Korea Case Among the three imperatives from the triad oflnternet policy, Korea strongly attests to the government function that can be called the "high stateness" ethos, and is characterized by the creation of a new IT implementation agency with the same logic as the creation of the centralized economic planning agency in the era of industrial society. Those comparative characteristics can be called a "distributive model" of national policy style for the mature use of the Internet. Policy Imperative (High Stateness Ethos). To understand the foundation ofthe high stateness ethos in Korea, a brieflook at the origins of the present form of government must be taken. In Korea, the present form of government arose in a time of crisis, during a brutal Korean war (1950) in which over 1 million Koreans died and 5.5 122

PAGE 137

million more were dislocated (Cole and Lyman, 1971:22). Social disruption on an extraordinary scale, destruction of rural society, and the historical absence of strong intermediary institutions placed great power in the hands of a state structure propped up by U.S. aid and an occupying U.S. military force (Hamilton and Biggart, 1988). The postwar government ofRhee Syngman (1948-1960) shaped the basic institutions that the military government of Park Chung Hee (1961-1979) later gained control of and turned in the direction of economic development. Especially during the presidency of Park, the government actively participated in the public and private spheres of the economy and was in fact the leading actor (SaKong, 1980). The government achieved its central position through centralized economic planning such as the so-called "Five Year Economic Development Plans" and through comprehensive implementation procedures. Furthermore, a strong stateness ethos was legitimized by both the so-called "Asian Values" under the threat ofNorth Korea and the Cold War environment. Such a leading role of government in Korea in the era of military power ofPark is well expressed in the following (Mason et al., 1980: 257): The entire government is geared toward economic policy-making and growth ... Economic decision-making is extremely centralized, and the executive branch dominates. To maintain legitimacy and to manage the complexities of economic development strategies capably, Korea established and increasingly came to rely upon bureaucratic arrangements dealing with such areas as finance, industrialization, and 123

PAGE 138

technology policy (Cheng, 1990; Hahm and Plein, 1995; Wade, 1992; Lim, 1999). Together, these historical conditions and forces helped to facilitate the emergence of strong executive-bureaucratic government arrangements. Recently, several forces, such as social and political democratization in the era of the so-called "civilized power" (1993-present), and economic liberalization and globalization, especially after the economic crisis of 1997-1998, may weaken the high stateness in Korea, but Korea still has comparative characteristics of high stateness (Visco, 1999). Distributive Policy (National Planning of the Network). In relation to the comparative characteristic of the specific style oflnternet policy, the strong stateness in Korea brought in the more "distributive" policy, especially in the area of national planning of the information network. In April 1994, the Korean government released a plan for establishing a high speed information infrastructure called "Korea Information Infrastructure (KII)," which will be completed by the year 2015 with a budget of 45 trillion won (US$ 60 billion) (Ministry oflnformation and Communication, 1994). The KII is composed of high speed and high-capacity government and public information networks, called the ''New Korea Net-Government (NKN-G)" and the "New Korea Net-Public (NKN-P)." The NKN-G was planned to be constructed with the use of government funds, estimated to spend 44.8 trillion won (U.S. $55.29 billion) (MIC, 1994). 124

PAGE 139

Thus, to achieve an information society, the Korean government has played a leading role in planning the KII and allocating its funds in the same manner as in planning economic development and allocating its funds for achieving an industrial society. Through the successful accomplishments ofthe Five Year Economic Development Plans, Korea has achieved rapid economic growth over the past three decades. A strong government intervention on the planning and the allocation of its funds was considered one ofthe key factors of Korea's economic development in the era of industrial society. Policy System (Creating a new IT Ministry). Among the various actors and agencies influencing public policy for the mature use of the Internet, the comparative characteristic of the policy systems for implementing Nil programs, liberalization programs, and universal service programs, Korea, based on the historical foundations in high stateness ethos, utilized the creation of the integrated IT Ministry for achieving an information society with the same logic as the earlier creation of the centralized economic planning agency for achieving an industrial society. In December 1994, the responsibilities ofthe existing Ministry of Communication (MOC) were significantly enlarged under a new Ministry of Information and Communication (MIC) (Larson, 1995). A significant shift came in transferring both responsibility for all software and information service from the Ministry of Science and Technology and responsibility for the Information Technology 125

PAGE 140

(IT) industrial policy from the Ministry of Trade, Industry and Energy to the new Ministry oflnfonnation and Communication. The creation ofthe integrated IT Ministry can be considered a significant step towards achieving an infonnation society in Korea. So, the MIC is responsible for implementing the Nil programs, liberalization programs, and universal service program. Summiuy Considering the preexisting differences of the national structures and the unique cultural, economic, and geographical circumstances in each nation, a homogeneity of policy style in the development of the Internet is not expected. Therefore, the reasoning underlying the approaches taken by each nation to achieve the mature use of the Internet could be quite different and unique. Among the "always-on" national groups, the U.S. strongly supports the market function (or service providers in a triad oflnternet policy), that can be called "free market" ethos, and heavily depends on the independent regulatory agency as an implementing system. Furthennore, the U.S. shows its leading role in the de-regulation of the communications market. These synchronic and comparative characteristics thus can be called a de-regulative model of national policy style for the development of the Internet. Among the Scandinavian national groups, Finland strongly holds community function (or service users in a triad oflnternet policy), that can be called 126

PAGE 141

''welfare society" ethos, and utilizes the existing functional Ministries as an implementing system. Moreover, Finland has demonstrated its comparative characteristics of direct subsidization of usage. These comparative characteristics can be referred to a re-distributive model of national policy style for the development of the Internet. Among the East Asian national groups, Korea historically argues for the government function (or politics in a triad oflnternet policy), that can be called "high state" ethos, and created a new IT Ministry as an implementation system. Furthermore, Korea demonstrated direct involvement on the planning of Nil. So, these comparative characteristics can be termed a distributive model of national policy style for the development ofthe Internet. High IT Intensity and Internet Diffusion In this section, the technological factors affecting Internet diffusion in the three nations are identified as a backbone for the development of Internet technology. The high performance of Internet diffusion in the three nations is technologically based on a high intensity of information technology (IT), which can be compared in view of (I) industrial distribution of IT employment, (2) research and development (R&D), (3) innovativeness of IT sector, and (4) IT networking. This comparison ofthe technological factors intends to reveal a similarity of the technological development in the three nations as a high IT intensity group. 127

PAGE 142

High IT Intensity Group The IT sector can be defined to include industry sectors within the manufacturing industry, the telecommunications industry (wired or wireless), and other IT services, which mainly comprise all computer and related activities and the wholesaling of machinery, equipment, and suppliers (OECD, 1998c; 2000c; 2000f). All the three nations show a high IT intensity measured by the following three variables industrial distribution ofiT employment, R&D (contribution to overall business sector R&D and industrial distribution of IT sector R&D), and innovativeness of the IT sector-among OECD nations. Industrial Distribution of IT Employment In terms of industrial distribution oflT employment among manufacturing, telecommunications, and other IT services in the three nations, Korea has the greatest proportion ofiT sector employment in IT manufacturing, over 70% (telecommunications around 20% and other IT sectors around 10%). The U.S. and Finland have proportions of IT sector employment in manufacturing around 40% (telecommunications around 20% and other IT sectors around 40 %). In this respect, the deployment ofiT in the U.S. and Finland are well balanced among the three components, while that of IT in Korea is centered on the manufacturing industry. Note that the low deployment of the other IT sectors like computers can affect Internet diffusion. 128

PAGE 143

Research and Development (R&D) In terms of the contribution to overall business sector R&D, the share ofiT R&D in total business R&D in Finland was just over 51%, the largest for any OECD nations. The shares ofiT R&D in the U.S. and Korea are well above the OECD average (around 34%)-that is, around 40% in Korea and around 38% in the U.S. In terms of the industrial distribution of IT sector R&D, the proportions of IT R&D undertaken by IT manufacturing business in both Korea and Finland are close to 90%, while that ofiT R&D undertaken by IT manufacturing in the U.S. is around 74%. Especially, the R&D data of Finland (over 51% in IT and around 90% in manufacturing) shows that a well-known wireless maker, Nokia, have played great role in IT development in Finland, which can affect Internet diffusion by providing a technological background. Innovativeness of IT Sector The extent to which IT finns and IT industries undertake innovative activities can be a key detenninant of IT perfonnance, which can provide an economic basis for the development of Internet technology. In terms ofthe ratio ofR&D to value added and Gross Domestic Product (GDP), OECD (1998c) showed that Finland (including Japan) had the largest ratios of R&D to value added in the IT sector, also had high R&D to GDP ratios, while the U.S. and Korea (including France) had high ratios of R&D to value added, both for the IT sector and the total business sector. Considering 129

PAGE 144

that Japan and France with low Internet performance show high innovativeness of the IT sector, this non-policy factor can be a necessary condition or prerequisite for the high performance of Internet diffusion, but not a sufficient requirement. In other words, the policy factors need to be supplemented with the non-policy factors. 5 9 IT Networking Public IT networking, or inter-finn coordination systems, combining research laboratories, academia, and industry affects Internet diffusion. This system makes inter-finn communications possible by providing institutional platforms (OECD, 2000c: 210). Its typical example is Finland's cluster programs (Lyytinen and Goodman, 1999). The Finnish telecommunications cluster consisted of equipment manufacturers (Nokia), electronics manufacturers (Elcoteq Network), software firms (F-Secure), and telephone operators (Son era, Elisa, Tel_ia) (Steinbock, 200 I a, 2001 b). Among the firms in the Finnish telecom cluster, Nokia perceived itself as an infrastructure leader whose spill-over effects benefited the national economy and several industries, as well as various value added services like the Internet, component suppliers, and strategic partners (Steinbock, 2001a).60 In addition, the Finnish telecom cluster is also closely 59 As of January 2000, Internet users per 100 inhabitants (Appendix D) are 8.4 in Japan and 5.1 in France (11.0 in OECD average). 6 For example, by 1999, Nokia alone contributed about one-fourth of Finland's economic growth (Steinbock, 200la). In this respect, it is criticized in that "putting all of its eggs in one baskets puts the whole national economy in a fragile situation." (Wall Street Journal, October 11, 1999). 130

PAGE 145

related with public technology R&D funding by the Finnish National Technology Agency (TEKES), targeting IT and the information society as a special area. In the U.S., early federal investments in networking R&D helped build the technological foundation oftoday's global Internet. Through the combined efforts of federal sponsors (ARPA and NSF), academia, and industry, prototype networking capabilities were deployed on a national scale and popular applications were introduced (OECD, 2000f). Federally funded IT R&D is coordinated with the Computing, Information, and Communication (CIC) programs, which are successors to the congressionally-chartered High Performance Computer and Communications (HPCC) Program. In Korea, the IT networking is, in most cases, carried by a research institute funded by government, or the Electronics and Telecommunications Research Institute (ETRI) (Larson, 1995; OECD, 2000f). The typical examples ofiT networking in Korea have occurred for the development of electronic telephone switching technology (TDX) and Code Division Multiple Access (COMA) wireless technology. In the 1980s, TDX switching technology was developed by a government-supported inter-firm coordination system combining research laboratories (ETRI), manufacturers (Samsung, Hyundai, Goldstar, and Otelco), and service operator (Korea Telecom). In the early 1990s, the COMA operating technology was developed by the IT -networking system of research laboratories and manufacturers (Samsung, Hyundai, Goldstar, Maxon) after a technology transfer from a U.S. firm (Qualcomm). 131

PAGE 146

Summary and Implications To answer the question about the comparative characteristics of national policy styles for the development of the Internet across space, the three nations reflect rather different system designs and thus demonstrate the different imperatives and national policy styles for the development of the Internet. Based on the investigation of the three different regional roads to the mature use of Internet, three different policy styles are identified across space: that is, ( 1) a de regulative model of the U.S. among the "always-on" national group, (2) aredistributive model of Finland among the Scandinavian national group, and (3) a distributive model of Korea among the East Asian national group. As the non-policy determinants affecting Internet diffusion, the high intensity of information technology in the three nations is compared in view of(l) industrial distribution of IT employment, (2) research and development (R&D), (3) innovativeness of the IT sector, and (4) IT networking. The identification of the three models of policy styles for the mature use of the Internet is intended to represent the comparative differences of policy determinants of Internet diffusion, while the comparison ofthe non-policy factors intends to reveal the comparative similarities of non-policy factors in the three nations as a high IT intensity group. Thus, the main focus of this thesis is on identifying the comparative characteristic of policy determinants for the mature use of the Internet among the three 132

PAGE 147

nations, reaching toward a social optimum point by showing high Internet diffusion based on the similar and high IT intensity. In short, based on the 3-D synchronic model, the comparative characteristics (especially, the differences) of the national policy styles in the three nations can be identified in a triad of Internet policy as in Figure 4-1, which will be embedded in a diachronic analysis of policy patterns based on the 3-S model in following chapters 5, 6, and 7. Figure 4-1: Synchronic Triad oflntemet Policy in Three Nations Korea (Network) o High Stateness Ethos U.S. (Service Providers) o Free Market Ethos o Independent Regulatory Body o De-regulative Model Mature Use Of Internet 133 o IT Ministry o Distributive Model Finland (Service Users) o Welfare Society Ethos o Functional Ministry o Re-distributive Model

PAGE 148

CHAPTERS POLITICAL INITIATION AND INTERNET DIFFUSION Based on the synchronic characteristics of the three different national policy styles reviewed in chapter 4, this chapter and the following two chapters (chapter 6 and 7) investigate the diachronic characteristics of the national policy patterns for the development of the Internet in the United States, Finland, and Korea to answer the second research question. In these three chapters, three different policy issues-i.e., national information infrastructure (network variable in a triad), liberalization (service-provider variable in a triad), and universal service (service-user variable in a triad)-in the three nations are compared and contrasted across time and space to show both the similarities and differences in policy patterns for the development of the Internet. Furthermore, the diachronic relationship between the three policy issues and the three stages of Internet diffusion over time is investigated. For this purpose, in this chapter, as the first chapter of the diachronic analysis of the national policy patterns for the mature use of the Internet, the chronological relationship of the political initiation ofthe Nil to the first stage (or, initiation) of Internet diffusion in the three nations is investigated. 134

PAGE 149

Based on Easton's (1965, 1979) input-output model and Dye's (1998) policy analysis model of causes and consequences, the political initiations of the Nil in the three nations are compared and contrasted in view ofthe theoretical pattern of Internet policy-i.e., (1) the political imperatives from the network axis in a triad oflnternet policy as a policy input, (2) the higher level of the national CIO as a policy system and distributive policy as a specific policy output, and (3) the diachronic relationship between the distributive policy and the initiation of Internet diffusion as a policy impact on Internet diffusion. These comparisons and contrasts are intended to produce both comparative characteristics of policy patterns and policy implications on the governmental role in the first stage oflnternet diffusion. Political Imperative: Imperative to Distribute The policy rationale for governmental intervention or the motivation to intervene concerns whether and where lines shall be drawn between public and private responsibilities. In the first stage oflnternet diffusion, policy rationale for initiating the digitalization of the existing network can be expressed in the imperative to distribute. It is political in nature in that the initiation is started from a higher level ofthe so-called "national CIO" to enhance national competitiveness. 135

PAGE 150

Theoretical Policy Rationale: Political Distribution According to Noam's (1994) diffusion model of interactive technology, a network will not be feasible unless subsidized by external sources, either by government or by the network operator's "non-economic" willingness to accept losses in the early phases of operation because where the network is small, average cost is high and externalities small, in the first stage, or below a critical mass point. In the Internet arena, both the huge amounts of initial investment on the Internet as a social network and its characteristics of natural monopoly and network externalities caused by the unique nature of the Internet as the commons are hypothesized to bring government intervention and initiatives to digitize the existing network in the first stage of Internet diffusion. So, the prominent issues prior to marketization (i.e., natural monopoly, network externalities) and sequential government initiatives based on political goal-settings of building the information infrastructure (the "information superhighway") or "information society"61 are the comparative characteristics in the first stage of Internet diffusion. 61 The European countries prefer the use of the notion "Information Society," rather than the American notion "National Information Infrastructure" to reflect a broader concept ofthe challenge and opportunities imbedded in the Information Age (Riis, 1996). 136

PAGE 151

Individual Policy Rationale In the first stage of Internet diffusion, political imperatives from politics as an idealized will to intervene can be compared and contrasted with the empirical wills to intervene of the three nations as policy input. United States. Public Funding and Initiation. In the name of"national defense" and "scientific research," the U.S. government launched the Internet protocol suite and the Internet itself(Kahin, 1997; Drake, 1995). The initial network ofthe Internet, ARPANET ofthe U.S. Department of Defense (DOD), was formed as a computer technology for military use in the late 1960s. In those times, the so-called "Cold War" was escalating and the American decision-makers were concerned about how the country could communicate after a nuclear war (Gromov, 1996, Westphal and Towell, 1998). ARPANET was initially funded for the purpose of military usage in 1968, by substantial public investments from the DOD's Advanced Research Projects Agency (Denning and Lin, 1994). The turning point was the advent ofUSENET NEWS in the late 1970s, which resulted from the effort of a computer scientists' group, responsible for the construction of the ARPANET but excluded from connecting to the defense establishment (Giese, 1996). The advent of USENET marked the beginning of a rapid growth in the life of 137

PAGE 152

publicly accessible computer networks as well as the end of the military's control of the computer network (Giese, 1996: 129-132). The ARPANET split into military and civilian components in 1983, with the civilian component serving as the genesis of what is now known as the Internet. In 1985, the National Science Foundation (NSF) began a program to establish Internet access across the U.S. by creating a research backbone called the NSFNET. The NSFNET, upgraded after congestion ofthe ARPANET62 in the mid-eighties, was federally subsidized to provide "free access" to academic and research users before the subsequent privatization of the NSFNET on April 30, 1995 (Kubicek et al., 1997; Hallgren and McAdams, 1999). In addition, the initial development ofthe Internet network was triggered by the so-called ''National Information Infrastructure"63 (Nil) initiatives in the Clinton-Gore Administration. The Clinton-Gore Administration officially launched the U.S. Nil initiative on September 15, 1993, with the publication of The National Information Infrastructure: Agenda for Action. The U.S. Nil initiative emerged in 199211993, and was conditioned by a combination of special features of the U.S. political system with a presidential campaign starting in the context of a broad public discussion on a national 62 As we reviewed in chapter 2 (page 16), the ARPANET was almost continuously overloaded with increasing numbers of data being lost and delayed, which caused the performance of the ARPANET to befall on Hardin's (1968) tragedy ofthe commons (Hallgren and McAdams, 1999: 470). 63 Nil can be defined as a seamless web of communications networks, computers, databases, and consumer electronics that will put vast amounts of information at users' fingertips" (IITF, 1993). 138

PAGE 153

infrastructure (Gramlich, 1994; Schneider, 1997).64 The Nil quickly advanced on the policy agenda because the Clinton Administration used it as a key point in their election campaign. They were running as New Democrats who stood for strong government leadership in industrial and technological development, contrasting sharply with a laissez-faire approach ofthe Reagan and Bush Republican eras (Drake, 1995; Gramlich, 1994; Schneider, 1997). Thus, the U.S. federal government played a critical role in initially seeding the initiation of Internet diffusion by the funding and initiation. This explains why the Internet technologically originated in the U.S. can be considered a better example of "government-inspired" rather than "market-led" innovation in its initial stage (Kubicek et al., 1997). The Highway Analogy. The policy rationale ofthe active role of federal government for the initiation ofNII is analogous to the policy rationale of the active role of federal government for the construction of the Interstate Highway System (ISH) (Shew, 1993; Kubicek and Dutton, 1997; Hallgren and McAdams, 1999). Vice President Gore explained to the press what in his view the Nil had in common with a highway system:65 64 The U.S. political system can be characterized by a potentially high degree of political leadership (presidentialism), a strong but divided parliament, a pluralistic interest group system, a high degree of regulatory delegation to independent regulatory agencies, and checks and balance system (Schneider, 1997). 65 For the full text of this speech, see http://www.hpcc.gov/white-house/gore.nii.html. 139

PAGE 154

One helpful way is to think of the National Information Infrastructure as a network of highways much like the Interstates begun in the 1950s. These are highways carrying information rather than people or goods. And I'm not talking about just one eight-lane turnpike. I mean a collection of Interstates and feeder roads made up of different materials in the same ways that roads can be concrete or macadam -or gravel. Some highways will be made up of fiber optics. Others will be built of coaxial or wireless. But-a key point-they must be and will be two-way roads. Considering the general skepticism American have shown towards big federal projects based on their strong faith in the free market, the interstate highway system built in the 1950s in the U.S is one ofthe major national projects in which federal action was widely appreciated (Kubicek and Dutton, 1997). As such, the U.S. Nil was largely seen as a non-partisan issue. Finland. Over the last few decades, Finland has made a remarkable transformation from an agrarian economy to a diversified, modern digital economy. Today, Finland has become one of the world's leaders moving towards the information society: it is ranked 3rd in the world by the 2000 Information Society Index (IDC/World Times Survey, 2000), after Sweden and the U.S.; in early 2000, 65% of Finns had a mobile phone subscription and 78% of the 2.35 million households in Finland now have at least one mobile phone; and at the end of 1999, there were approximately 121 hosts per thousand inhabitants, ranking Finland second in the world after the U.S. (OECD, 2001a). The current success of Finland as a digital nation results in part from a farsighted and goal-oriented government policy. During the first half of the 1990s, 140

PAGE 155

\ continuous work was conducted on new programs aimed at promoting an information society. Extensive studies were carried out with the aim of creating a national information strategy that led in 1994-1995 to the first National Information Strategy, under the leadership of the Minister of Finance. A national information society strategy was introduced in Finland at the end of 1994 with the government policy document Finland Towards the Information Society a National Strategy. In Finland, the information society strategy concept dates back to work ofthe Information Society Advisory Board (1976 to 1991) and a review of Finland's IT and telecommunications policies performed by the OECD between 1990 and 1992 (Chatrie and Wraight, 2000). One ofthe conclusions ofthe OECD report was that Finland had reached very high levels of IT and telecommunications penetration and expertise but that the country was lacking a clear statement of strategy in these areas (OECD, 1992). At the same time, the U.S. was launching the concept ofthe information highway and the European Union was preparing a strategy for the European way into the Information Society. Consequently, the Finnish government decided to charge the Ministry of Finance with the task of preparing the first Finnish information society strategy. The purpose was to cope with primarily three challenges: integration into the global economy; overcoming the national economic depression; and the pressure for change posed by new information technologies. Especially, at that point of time, Finland was experiencing a severe national economic depression, largely as a result of 141

PAGE 156

the downfall ofthe Soviet Union to which the economy of Finland had been closely connected. Finland had to extend its relations with other nations and information communications technologies were seen as an important area with new growth potentials. Thus, the Finnish government chose to initiate Internet diffusion as the means to overcome the "economic depression." Korea. The number of Internet users in Korea has rapidly increased since the commercial Internet access started in 1994, and exceeded 3.1 million in 1998; the current number oflnternet users is 20.9 million (48.6% ofthe total population) in March 2001 (MIC, 2000, 2001). Internet diffusion in Korea was also triggered by the governmental planning of an advanced national information infrastructure with a government policy document Korea Information Infrastructure (Kll): Blueprint for Implementation in April 1994. The Korea Nil initial plan occurred at about the time the U.S. and Japan66 were announcing plans for their respective national information infrastructures (Nil). The Korean government's strong leadership in initiating the Kil can be considered to be originated from the government-led "catch-up strategy" in the broad Asian context or from the national planning strategy of economic development in the narrow Korean context in that the Nil was seen as a part of national economic policy in Korea (Jeong 66 In 1993, Japan began debating plans for a nationwide digital communications network or national information infrastructure. For further information, see West J. et al., "Back to the Future: Japan's Nil Plans" in Kahin, B. and E. J. Wilson III (eds.) (1997:61-111). 142

PAGE 157

and King, 1997; OECD, 2000d). Jeong and King (1997:112-113) articulated the catchup strategy of Korea in this way: The Kif movement arose from .... a new fear that a failure to build an information infrastructure would hurt Korea's basic industries to the point that they might not be able to compete in the global marketplace, leaving the nation farther behind the developed countries. This concern extended to production industries such as computers, telecommunications, components and semiconductors-all of which might be left behind by the global production system, with a subsequent loss of export and import substitution opportunities. In short, the Korean government's strong leadership in KII as a part of economic policy provided the tools for national competitiveness, and thus economic development, in a globalized economy. The basic assumption was that national competitiveness would arise from the ability of each entity in the economy to develop, acquire and adapt new and state-of-the-art information technologies as tools that would be available on and through information networks (Porter, 1990; Jeong and King, 1997). Thus, Korea's powerful tradition of central government coordination makes topdown planning of the KII a comparative characteristic of its initiation. Korean government's strong leadership in KII can also be found in establishing technical specifications for network standards like CDMA wireless technology and its funding by the Informatization Promotion Act. Summary The Internet, which was technologically originated in the U.S., was a federallyfunded or government-inspired innovation for the purpose of initial military usage and 143

PAGE 158

later academic and research usage. Furthermore, in the U.S., the Clinton-Gore Administration initiated the Nil project in 1993 to digitize the national network with the "highway (ISH) analogy." In Finland, the government initiated the Information Society Strategy in 1994 to digitize the national network as a means to overcome the economic depression. In Korea, the government had strong leadership in initiating the Kil in 1994 to digitize the national network with the government-led catch-up strategy or national planning strategy to enhance so-called national competitiveness. Policy Programs and Agencies In this section, the comparative characteristics of the policy agencies to implement the Nil and national development of specific plans for constructing the Nil in the three nations are briefly reviewed. Reviewing the comparative characteristics of the Nil policy programs is intended to reveal the focus of each national program from the perspective of the triad of Internet policy-i.e., whether focus is on (1) network digitalization, (2) service providers' entrepreneurial activities, or (3) service-users' equity. In addition, reviewing the comparative characteristics ofthe policy agencies to implement indicates (1) whether or not the so-called "national CIO" is institutionalized in a nation (2) funding mechanisms of digitizing the national network. 144

PAGE 159

Theoretical Policy Programs and Agencies Distributive Policy. The national goal of building the Nil can be achieved by specific policy programs of investment or subsidies on Nil, which can be classified into a distributive policy in view ofLowi's (1964, 1972) policy typology in that such policy programs show the focus of their funding mechanisms. The focus of their funding mechanisms revealed in the plan for the national information infrastructure can affect Internet diffusion. National CIO. In the initial stage oflnternet diffusion, the administrative arrangement of a national chief information officer (CIO) can likely affect Internet diffusion by envisioning the national goal of building the so-called "information society" and by initiating a national structure (Nil). Considering the multidisciplinary nature and effects of the Internet as an interactive social network, the Internet-related policy issues can be interrelated issues among the functional ministries in the era of the industrial society. In this respect, a new organizational scheme in the information society can be termed a "national CIO," and its institutionalization can affect Internet diffusion. Individual Policy Programs and Agencies In this stage, both the actual policy agencies to implement and the specific policy programs ofNII summarized in the three governmental reports (Table 4-3) can be 145

PAGE 160

compared in detail both with each other and with the theoretical policy agencies and programs. United States. Policy Program. The heart ofthe Nil initiative as set out in the Agenda for Action is a set of nine principles and goals coupled with actions items (IITF, 1993: 6-7): 1. Promote private sector investment, through tax and regulatory policies that encourage long-term investment, as well as wise procurement of services. 2. Extend the universal service concept to ensure that information resources are available to all at affordable prices. Because information means empowerment, the government has a duty to ensure that all Americans have access to the resources of the information Age. 3. Act as a catalyst to promote technological innovation and new applications. Commit important government research programs and grants to help the private sector develop and demonstrate technologies needed for the Nil. 4. Promote seamless, interactive, user-driven operation of the Nil. As the Nil evolves into a "network of networks," government will ensure that users can transfer information across networks easily'and efficiently. 5. Ensure information security and network reliability. The Nil must be trust worthy and secure, protecting the privacy of its users. Government action will 146

PAGE 161

also aim to ensure that the overall system remains reliable, quickly repairable in the event of a failure and perhaps most importantly, easy to use. 6. Improve management of the radio frequency spectrum, an increasingly critical resource. 7. Protect intellectual property rights. The Administration will investigate how to strengthen domestic copyright laws and international intellectual property treaties to prevent piracy and to protect the integrity of intellectual property. 8. Coordinate with other levels of government and with other nations. Because information crosses state, regional, and national boundaries, coordination is important to avoid unnecessary obstacles and to prevent unfair policies that handicap U.S. industry. 9. Provide access to government information and improve government procurement. William J. Drake (1997: 309-311) summarized the distinguished five aspects of the agenda from that of its predecessors: first, the agenda makes clear that the administration favors expanding competition in all segments of the digital marketplace in order to increase innovation and investment while lowering prices for consumers; second, while supporting liberalization, the agenda also proposes the establishment of public safeguards, such as promoting universal service; third, the agenda couples its call for competitive supply with attention to the demand side of the market; fourth, the administration is working to strengthen the telecommunications policy-making process. 147

PAGE 162

At the level of individual agencies, management has been improved and streamlined in accordance with Vice President Gore's "Reinventing Government" program;67 Fifth, the Nil initiative is noteworthy for the attention it gives to bringing diverse stakeholders into the policy process through the following ad-hoc task forces. Through a well-balanced agenda and the catchy metaphor ofthe "superhighway," the Nil initiatives can be considered successful in terms of focusing public attention on the massive changes in information technology and the potential for economic and social change under the circumstances of the American strong trust in market-led industrial strategies (Drake, 1995; Kubicek and Dutton, 1997). Policy Agencies. In the same year as the publication of the government Nil document, a specialized Information Infrastructure Task Force (IITF) was created, and a U.S. Advisory Council on the National Information Infrastructure (NIIAC)68 was established to bring together the major interests concerned (Drake, 1995; Kubicek and Dutton, 1997; Schneider, 1997; Kahin, 1997). U.S. government Nil initiative was managed through the IITF, an "unfunded" and ad hoc task force composed of representatives of executive branch agencies (Kahin, 67 Vice President AI Gore presented the report of his National Performance Review (NPR), titled From Red Tape to Results: Creating a Government That Works Better and Costs Less, to President Bill Clinton and the public on September 7, 1993. For the core principles of the so-called reinventing see Osborne and Gaebler (1992). 8 NIIAC was a conventional but high-profile private-sector advisory council, and was appointed by the Secretary of Commerce. Initially limited to 25 members, it was expanded to 37 members. For more information, see Kahin, B., (1997:166-167). 148

PAGE 163

1997). With the independent agencies such as the Federal Communications Commission (FCC) participating as observers, the IITF was comprised of over forty high-level representatives of all the federal agencies with stakes in the Nil. These blends of functional responsibilities and perspectives were intended to help the administration examine any given Nil issue in an integrated fashion, rather than through the prism of a single agency's mandate (Drake, 1995). The IITF was chaired by the Commerce Secretary and most staff work was done by the National Telecommunications and Information Administration (NTIA) of the Department of Commerce (Kahin, 1997). The interagency IITF wrote The Agenda for Action, with its organizational structure composed ofthree committees (i.e., Telecommunications Policy, Information Policy, and Applications) outlined at the end of the document (IITF, 1993). Given that the national political climate was skeptical of any new manifestation of government, the IITF convened on an "ad hoc" basis, and, further, the Nil initiative did not appear as a line item in the federal budget at all. As Kahin ( 1997: 151) argued, such an ad hoc and unfunded arrangement of policy agencies was conceived to avoid seeking Congressional approval. Finland. Policy Program. Considering that government in Finland initiated Internet diffusion as the means to overcome the "economic depression," the strategy document 149

PAGE 164

of Finland towards the Information Society-a National Strategy (Ministry of Finance, 1994) focused much attention on the business potentials and technology development. It highlighted three challenges facing the country: 1. External challenges linked to globalization; Finland must integrate with the open global economy. 2. Internal challenges linked to economic crisis; Finnish society needs renewal to overcome the economic depression. 3. Pressures for change arising from new information technology. Therefore, this first Information Society strategy in Finland was based on three main elements: 1. Renewal into an Information Society. 2. The development of an information industry; 3. The creation of the necessary conditions: research, know-how, development of an IT infrastructure. The strategy contained 5 lines of action: 1. IT and information networks were to serve as tools in private and public sector renewal. 2. Information industry was to become an important sector of economic activity. 3. Professional expertise in IT was to be maintained at a high overall level, with selected peaks. 150

PAGE 165

4. Everyone would have the opportunity and basic skills to use the services of the information society. 5. Finland's information infrastructure would perform competitively and be capable of providing high quality services. Furthermore, the technology and economy-oriented national strategy in 19941995 has been criticized for neglecting the "social dimension" of Internet policy i.e., the needs of different disadvantaged citizen groups as well as regional and local aspects, which caused the Finnish national program to be elaborated with a separate national document (Henten and Kristensen, 2000; Chatrie and Wraight, 2000).69 Policy Agencies. Considering that a national strategy for an information society in Finland originated from the motives for coping with the national economic depression in the early nineties, central coordination of the national strategy for an information society in Finland was located in the Ministry of Finance (Hen ten and Kristensen, 2000; Chatrie and Wraight, 2000; Sorgaard, 2000). Extensive studies were carried out with the aim of creating a national information strategy that led in 19941995 to the first National Information Strategy, under the leadership of the Minister of Finance. Following the publication of Finland towards the Information Society-a National Strategy (1994), the Minister of Finance produced a follow-up report, Finland's way to the Information Society-The National Strategy and its 69 The elaborated national policy program and agencies will be reviewed furthermore in chapter 7. 151

PAGE 166

Implementation ( 1996), which presented a brief summary of the 1994 report and an analysis of developments related to its implementation during 1995 and 1996 (Chatrie and Wraight, 2000). A relatively small unit in the Ministry of Finance exercised coordination through various inter-agency groups and networks, through preparing government decisions in the field, and through projects developing government information services (Sorgaard, 2000). The central level coordination tasks were handled by the "Public Management Department" that is organized in two units, i.e., the Information Management Unit and the Governance Policy and Public Services Unit, in the Ministry ofFinance.70 Thus, the coordination function ofthe information society in Finland was institutionalized under the umbrella of the Ministry of Finance from the perspective of public management of the IT -related budget allocation. Korea. Initial Policy Program. With the start of President Young-Sam Kim's administration in 1993, the implementation of an advanced national information infrastructure received a major policy impetus. In April 1994, the Ministry of Information and Communication (MIC) released a plan for establishing a high-speed 7 For the further information about the role and organization ofthe two units, see Sorgaard, P. "IT Coordination and Public Management Reform: A Comparison between Finland and Norway." Ministry of Finance Research Reports, Helsinki, Apr. 2000. Ppl7-29. 152

PAGE 167

information infrastructure, The Korea Information Infrastructure (Kll): Blueprint for Implementation (MIC, 1995). 71 The goal of KII was the construction of an advanced national information infrastructure consisting of communications networks, Internet services, application software, computers and operating systems, and information products and services. The KII of the 21st Century was expected to enable all Koreans to access information and communicate with anyone, anytime, and anywhere. All information and communications services in voice, data and video will be provided easily, reliably, securely, in a timely manner and cost effectively. The projected KII consisted of high-speed government and public information networks. The New Korea Net-Government (NKN-G), funded by the government, provides government agencies and public institutions, including research organizations and universities, with low-cost information and communications services. Application services and key technologies are to be developed in collaboration with industry, universities and government laboratories. When these technologies and their applications have been tested (i.e., test-bed projects such as the installation of the fiberoptic backbone network between Seoul and Daejon) and deployed over NKN-G, they will be commercialized on the New Korea Net-Public (NKN-P). 71 The initial plan for the Nil has been revised by the series of meetings of the KII Task Force since its introduction in April 1994 (Jeong and King, 1997). The initial plan for the KII is based on the English version presented in the 1995 International Conference on Computers and Communication. !53

PAGE 168

Construction and operation of the NKN-G will take place in three stages (Appendix L): groundwork (1995-1997), diffusion (1998-2002), and completion (2003-20 1 0). The NKN-G is designed to improve the efficiency of government operations and the delivery of public services. Its plan is to connect central and local government agencies and various public organizations, including schools and libraries, by the year 2015. The objective of the NKN-P is to provide interactive broadband multimedia information services to users in the private sector by wiring offices and homes with fiber-optic cables. In the early stages of the plan, the NKN-P will target urban offices and apartments that are likely to be the recipients of heavy traffic. Like the NKN-G, the NKN-P will be implemented in three stages (Appendix M). Cost estimation for IT projects is difficult because technologies themselves are advancing continuously, enabling services to be provided over networks less expensively. Nevertheless, to fulfill such an ambitious national plan, the estimation of costs is necessary because the government must determine the investment stream required as part of establishing public consensus in support of the national project, and to help the parties with a stake in the KII to prepare their financing schedules (Jeong and King, 1997). The establishment of the KII over 20 years was expected to cost 44.8 trillion won (U.S. $55.29 billion) (MIC, 1994). The NKN-G was planned to be constructed by the use of government budget and other funds from the sale of government ownership of Korea Telecom (KT), which is in the process of privatization. 154

PAGE 169

KT has projected to spend an estimated total of 8 I 1 billion won (approximately US $ 1. I billion). Elaboration. For the purpose of further promotion of IT industry under the distressed economy of the financial crisis in I 997, work on the revision of the KII plan began in March 1997, building on the five-year rolling plan for information promotion. The KII plan was further developed in June 1998 with the publication of Cyber Korea 21: An lnformatization Vision for Constructing a Creative, Knowledge-Based Nation (MIC, 1999). It was comprised ofthree parts: (I) the vision of a creative, knowledge based society, (2) key initiatives ofCyber Korea 21, and (3) promotion strategies. The four basic objectives of the national plan are: I. Early establishing an information infrastructure. 2. Increasing productivity and transparency of all economic players, including business, government and individuals, through the utilization of advanced IT. 3. Promoting new business and creating new jobs through the utilization of IT. 4. Designating competitive telecommunications products and services as key export products. For implementation ofthe Cyber Korea 21, the Korean government plans to concentrate resources between I 998-2002 on three key areas: (1) enhancing information infrastructure, (2) wiring government, business and individuals, and (3) reinforcing growth in the software and information providers industries. 155

PAGE 170

Policy Agencies. For the systematic and efficient implementation of the national plan, a high-level steering committee, the "Korea Information Infrastructure Committee," has been formed at the inter-ministerial level. The steering committee, headed by the Prime Minister, includes the ministers of related ministries and representatives from the National Assembly and the Supreme Court. The Informatization Promotion Act, enacted in August 1995, provides a strong legal basis, not only for inter-ministerial cooperation, but also for making public IT investments necessary to the success of the KII plan. The Ministry of Information and Communication (MIC) is mandated to operate the specialized IT fund i.e., the informatization promotion fund. In short, the national information infrastructure strategy (KII) in Korea was initiated and elaborated with the same policy agencies (or MIC) but separate national programs (or documents) in different time periods. Summary In the three nations, the distributive policy programs and policy agencies for constructing the Nil can be compared and contrasted in view of (I) the focus of funding (i.e., whether focus is on network digitalization, service-providers' entrepreneurship, or service-users' equity), (2) the funding mechanism for digitizing the national network, (3) whether or not the so-called "national CIO" is 156

PAGE 171

institutionalized, and (4) the practical implementing agencies to distribute the national funding. In the U.S., the comparative focus of the Nil was on the service-provider dimension (i.e., expanding competition in all segments of the digital marketplace to increase service-providers' innovation and investment) with an ad hoc and unfunded institutional arrangement of the Information Infrastructure Task Force (IITF) under the administrative umbrella ofthe Secretary of Commerce. In Finland, the comparative focus of the national strategy for achieving an information society was on the service-provider dimension (i.e., increasing serviceproviders' entrepreneurship via expanding competition) with the IT-related national budget allocation under the strong leadership ofthe Ministry ofFinance.72 In Korea, the comparative focus is on the network dimension (i.e., digitizing the national network) with the specialized IT fund (i.e., the Informatization Promotion Fund) under the leadership of the Ministry of Information and Communication. Policy Impact: Political Initiation and Internet Diffusion In this section, the diachronic relationship of the political initiation of the Nil and the first stage oflntemet diffusion is investigated. For this purpose, (1) the linkage 72 Later (1998), the focus was changed into the service-user dimension (i.e., enhancing the service-users' equity) with the IT-related R&D allocation under the strong initiatives of the Finnish National Fund for Research and Development (SITRA) with an elaborated national document. For further information, see chapter 7 (page 236). 157

PAGE 172

between the Nil and the Internet, (2) policy relevance ofthe Nil in view of the linkage between the Nil and the Internet, (3) time-matching comparison between the Nil and the first stage of Internet diffusion, and (4) government role in the first stage of Internet diffusion are briefly reviewed. Nil and Internet: Digitizing Networks The main purpose of governmental involvement in the National Information Infrastructure (Nil) is upgrading the nation's physical communication infrastructure-in other words, on integrating the traditionally divided communications conduits (voice, video, and data) into a digitized and packet-switched network. The traditional communications world relied on distinct infrastructures for each communication service. Voice travels over a national-wide, wired, circuit-switched network. Video moves over a separate system of terrestrial broadcast stations, supplemented by coaxial cable or hybrid fiber-coax networks carrying video from a cable company to all homes in a given area. Data moves on the voice network, under which the information is converted from digital to analog form and back again (Weinberg, 1999). However, from the technological perspective of network design, digitalization and packet switching have the potential to change that traditional design. One can convert the information transmitted via any communications service-whether it is voice, video, or data-into digital form. Packet switching enables the transmission of 158

PAGE 173

that digitized information across different networks without regard to the underlying network technology. This means that the digitized information corresponding to any service can be transmitted over any physical infrastructurecopper wires, fiber, hybrid fiber-coax, microwave, or direct broadcast satellite. Such logic of network design by digitizing the network was expressed in the Nil vision. With the explosion of the Internet services or the packet-switching service, the "metaphor ofthe highway" for the Nil as in the case ofthe U.S. has been overtaken by the "mythology of the Internet" as Kahin (1997: 184) argued. Policy Relevance: Nil and Telecommunications Networks The relevance of policy programs (i.e., Nil plans in the three nations) for digitizing the existing networks can be evaluated in view of the historical development of local telephone market and the ownership of local telephone service providers.73 Historically, the local telephone services in the U.S. have been provided by privately owned monopolistic companies. Before the Modification of Final Judgment (MFJ) on January 1 1984, the American Telephone and Telegraph (AT&T) company enjoyed a monopoly on long-distance service, controlling 80 percent of the local access line through the Bell Operating Companies. After the MFJ, the Bell system was broken up. The Regional Bell Operating Companies (RBOCs) provided the local telephone 73 It is not only because this dissertation focuses on the dial-up connection via a telephone line, but also because the Internet in the simplest form is just computer applications over the telephone networks. 159

PAGE 174

service, with AT&T being largely restricted to the lucrative long-distance market (Bernt and Weiss, 1993; Brock, 1994). The traditional government role in the local telecommunication market focused on preventing the provider-centered monopolistic dominance of the local facilities (i.e., the so-called essential facilities) by introducing competition in local market. In this respect, the limited government role in the Nil for facilitating the digitalization of existing networks by introducing competition among the service providers indirectly is relevant to the U.S. historical context of telecommunications development and the ownership ofthe communications network. In Finland, the local telephone services have been provided by the so-called private/public mix of telecommunications companies. While, the privately owned or customer-owned74 regional monopolistic companies (Finnet Group) provided local telephone service in urban areas, the state-owned monopolistic companies (Telecom Finland) provided local telephone service in rural areas. Thus, in Finland, the traditional government role in the local telecommunications market can be bifurcated: on the one hand, the government focused on preventing the customer-tied monopolistic dominance of the local facilities by introducing competition in local urban markets indirectly; on the other hand, government directly invested on digitizing the local rural network via a state-owned Telecom Finland. 74 In Finland, customers closely tied with the local telephone companies are called the local cooperatives. This special characteristic of local telephone companies in Finland is to be reviewed in detail in chapter 6. 160

PAGE 175

In this respect, the change of the government role in the initial national strategy for achieving an information society -i. e., the change from direct investment on digitizing local rural network to indirect facilitation of digitizing the local urban network by facilitating network competition with introducing the competition among the service providers is relevant to the Finnish historical context of telecommunications development and the ownership of the communications network. In Korea, local telephone services had long been provided by the state-owned monopoly, Korea Telecom. The traditional government role in the local telecommunications market focused on direct investment on digitizing the local network via a state-owned Korea Telecom. 75 In this respect, the active government role in the KII for the direct investment on digitizing the existing network is relevant to the Korean historical context oftelecommunications development and the ownership of the communications network. Nil and Initiation As a whole, governmental initiation can be justified by the huge amount of initial investment on the digitized network as a social network; furthermore, the empirical data on the time periods of the governmental initiation on the Nil are matched with the empirical data on the time periods of the initial stage of Internet diffusion. 75 Furthermore, the direct investment of the Korean government can be also justified in that Korea showed a comparatively low rate of digital main lines: Finland (76.8%), U.S. (71.6%), and Korea (61.8%) in 1994. For more information about the rates of digital main lines in 1991-2000, see appendix 0 (the U.S., Finland, and Korea). 161

PAGE 176

In the U.S., based on the empirical data on Internet users (Appendix N), the initial stage of Internet diffusion can be roughly estimated at between 1991 to April 1995 or before the year of the taking-off of the Internet users triggered by the privatization ofthe NSFNET. In these time periods, the U.S. government's active role can be found in (1) funding NSFNET and in (2) initiating the Nil plan. In Finland, the initial stage of Internet diffusion can be roughly estimated at between 1991 to 1994 or before the year of the taking-off of the Internet users triggered by the initial introduction of competition in the Finnish local telephone market. In these time periods, the Finnish government's active role can be found in (1) upgrading the local and rural network with the government-owned PTT and in (2) initiating the national strategy for achieving an information society. In Korea, the initial stage oflnternet diffusion can be roughly estimated at between 1991 to April 1999 or before the year of taking-off the Internet users triggered by the introduction of competition in the Korean local telephone market. In these time periods, the Korean government's active role can be found in (I) direct funding for constructing the NKN-G and the NKN-P and in (2) initiating the KII. Even though the three nations show different degrees of governmental involvement-i. e., a different focus of funding, a different funding mechanism for digitizing the national network-the role of the government in this period can be theoretically proposed as an active promoter of the digitalization of the existing national network with a distributive policy among the triad oflnternet policy. Thus, the 162

PAGE 177

political dimension ofthe 3-D synchronic model can be roughly matched with the first stage ofthe 3-S diachronic model ofthis dissertation Summary and Implications To answer the second research question and to test theoretical proposition (2a) about the diachronic characteristics of policy pattern in the first stage, policy factors affecting the initial Internet diffusion in the three nations are compared and contrasted based on the 3-S diachronic model. This comparison is intended to produce the similarities and differences of policy pattern in the initial stage oflnternet diffusion. Chronologically, time periods (1991before in the three nations can be identified to fall into the first stage oflnternet diffusion : i.e from 1991 to May 1995 in the U.S., from 1991 to January 1994 in Finland, and from 1991 to April 1999 in Korea. During these time periods in each nation, comparative focus of governmental involvement for the mature use of the Internet was on the network variable (or network digitalization) among the three variables in a triad oflnternet policy. Such governmental involvement in the initial stage oflnternet diffusion was triggered by the political initiation or goal-setting ofNII or information society vision with a policy tool of distributive policy. Thus, the comparative focus was on the network variable (or political dimension) in a 3-D model during the first stage in a 3-S model. These comparative similarities can produce a theoretical suggestion of 163

PAGE 178

governmental role as an active promoter of the digitalization of the existing national network. The comparative differences of policy pattern in the first stage are closely related with both the synchronic differences of policy styles across space reviewed in chapter 4: i.e., policy imperative, policy system, and focus of national plan. Regarding the policy imperatives, In the U.S., the Clinton-Gore Administration initiated the Nil project to digitize the national network with the highway (ISH) analogy. In Finland, government initiated the Information Society Strategy to digitize the national network as a means to overcome the economic depression. In Korea, government had strong leadership in initiating the KII to digitize the national network with the government-led catch-up strategy or national planning strategy. With relation to the focus of national plan, the comparative focus ofthe U.S. Nil was on the service-provider dimension (i.e., expanding competition in all segments of the digital marketplace in order to increase service-providers' innovation and investment). In Finland, the comparative focus of national strategy for achieving an information society was on the service-provider dimension as in the U.S. case. In Korea, The comparative focus was on the network dimension (i.e., digitizing the national network). With relation to the policy system, in the U.S., the Information Infrastructure Task Force (IITF) was an ad hoc and unfunded institutional arrangement under the umbrella ofthe Ministry of Commerce. In Finland, the Ministry of Finance had a 164

PAGE 179

leading role with the IT -related national budget allocation. In Korea, the Ministry of Infonnation and Communication had strong leadership in KII with the specialized IT fund (i.e., the lnformatization Promotion Fund). 165

PAGE 180

CHAPTER6 ECONOMIC DEREGULATION AND INTERNET DIFFUSION In this chapter, as the second chapter ofthe diachronic analysis of national policy patterns for the mature use of the Internet, the diachronic relationship between the economic deregulation (or liberalization) of the communications market and the second stage (or proliferation) oflnternet diffusion over time is investigated. For this purpose, the economic deregulation policies of the communications market in the three nations are compared and contrasted, based on Easton's inputoutput model and Dye's policy analysis model of causes and consequences: i.e., first, the economic imperatives from the service-provider axis in a triad of Internet policy as a policy input; second, the regulatory agencies as a policy system; third, the de-regulative policy as a specific policy output; and fourth, the diachronic relationship between the de-regulative policy and the proliferation oflnternet diffusion as a policy impact on Internet diffusion are compared and contrasted in view of the proposed theoretical patterns of Internet policy. This analysis is intended to highlight both the comparative characteristics of policy patterns and the policy implications on the government role during the proliferation of the Internet. 166

PAGE 181

Economic Imperative: Imperative to De-regulate The policy rationale for governmental intervention or the motivation to intervene concerns whether and where lines shall be drawn between public and private responsibilities. In the second stage of Internet diffusion, policy rationale for the proliferation of Internet diffusion can be expressed in the economic imperative to deregulate. Theoretical Policy Rationale: Economic De-regulation According to Noam's diffusion model of interactive technology, through the second stage (n1-n2)-i.e., above a critical mass point to optimum point that Noam called the "cost-sharing phase"the network users can lower their marginal cost simply by adding members. Self-sustaining growth is therefore possible by private network providers until a private optimum point where average costs increase and utility plateaus. In the Internet arena, enhancing the entrepreneur capabilities of service providers through market liberalization can be justified in the second stage of Internet diffusion in that the Internet network that was initially funded by the government to digitize the existing networks can achieve self-sustained growth. So, marketization issues, such as the network congestion caused by the undersupply of Internet service, are hypothesized to result in privatization and competition. As such, economic imperatives from the 167

PAGE 182

market, especially from the supply-side axis of Internet policy for the proliferation of Internet diffusion are the comparative characteristics of the second stage of Internet diffusion. Individual Policy Rationale In the second stage of Internet diffusion, the economic imperatives from the service-provider axis of a triad of Internet policy as a theoretical policy rationale to de regulate can be compared and contrasted with the individual policy rationales to de regulate of the three nations. United States. As we reviewed in chapter 4, the U.S. demonstrated its leading role in the liberalization of communication market even among the "always-on" national groups. The American government's role in detennining the competitive environment ofthe communications market has been limited to regulating the monopolistic dominance of the service providers because of the historical recognition of the state as a threat to personal and commercial freedom. Instead, in the U.S., the market is considered a neutral regulatory mechanism. The American policy makers strongly trust in the market function, which can be symbolized by a belief in free market ethos (Blumler, 1991). Keeping in the line with the American strong trust in free market, the emphasis ofthe national plan on the national infonnation infrastructure of the Clinton-Gore 168

PAGE 183

Administration was on the market-led industrial strategies. The Clinton Administration intended to stimulate private sector investment in information infrastructure by "telephone companies," "cable television companies," and "computer companies" rather than pouring funds directly into the telecommunications networks (McKnight and Neuman, 1995). Thus, such a market-led industrial strategy on the Nil resulted in the government-supported NSFNET being privatized on April 30, 1995. Furthermore, competition was encouraged in the local telephone market by the passage of a statutory framework of the "Telecommunication Act of 1996" (Kahin, 1997). The main events ofliberalization ofthe telecommunications market in the U.S. are summarized in Table 6-1. Table 6-1: Liberalization of U.S. Telecommunications Market 1982: Modified Final Judgment (MF J) in settlement of the antitrust case was entered. 1984: AT&T was divested and equal access (interconnection) was introduced. 1985: Customers selection of carrier was allowed and equal access was enlarged to the customers. 1989: Price cap regulation was introduced in AT&T. 1991: Price cap regulation was enlarged to the BOCs. 1992: VOD (video on demand) service was allowed to the local telephone companies. 1993: Number portability was introduced in "1-800" service. Nil action plan was released. 1996: Telecommunications Act was revised. The FCC Order on local interconnection was entered into force. 1997: The FCC Orders on universal service and access charge reform were entered into force. Source: Manishin, G. B. Telecom Industry History (1992) and Brock, ( 1998) 169

PAGE 184

Finland. By the early 1990s, the effects of the Finnish recession had been profound. Indeed, Finland is the only OECD country in which the recession of the early 1990s was as severe as the "Great Depression" of the 1930s (Manning and Shaw, 1998). Finland's GDP declined by around 13 percent per annum from 1991 to 1994. In those three years, the unemployment rate climbed from under 4 to 18 percent ofthe labor force (Manning and Shaw, 1998). Thus, it was no surprise that there were cutbacks in Finland's welfare services and the widespread introduction of user charges and the introduction of market reforms. During the initiation of national information society strategy in 1994-1995 as a means to overcome such a national economic depression, the focus of the national information society was on business potentials and technology development. Such a national strategy was paired with the two separate introductions of competition into the local telephone market. Deregulation started in 1994 by opening both the local and the domestic long distance telephone market based on the Telecommunication Act of 1992, amending the Telecommunication Act of 1987. The lack of service-based competition in the local markets-such as the incumbent local telephone providers' refusal of their network capacity to the competitors -brought in the second deregulation of the local telephone market by the Telecommunications Market Act of 1997 that replaced the Telecommunications Act of 1992. The main events of liberalization of telecommunications market in Finland are summarized in Table 6-2. 170

PAGE 185

Table 6-2: Liberalization of Finnish Telecommunications Market 1987: Telecommunications Act Administration of telecommunications transferred to the Ministry of Transport and Communications. 1988: Competition in corporate networks and data transmission partially liberalized. The new Radio Act 1990: The special rights ofthe National Board ofPost and Telecommunications abolished. Free competition in data networks and wireless GSM networks 19901991: Licenses granted to regional radio telecommunications networks. Corporate networks subject to free competition. 1992: Switched data transmission exempted from licenses. Competitive licenses to long-distance and local telecommunications. 1993: Restricted competition in long-distance and international telecommunications. 1994: Local, long-distance, and international telecommunications subject to free competition. The first licenses to service operators 1995: Competing licenses to DCS networks 1996: The amendment to the Telecommunications Act 1997: The Telecommunications Act replaced by the Telecommunications Market Act. 1998: Minor forms ofmobile telecommunications exempted from license. Transmission of international telecommunications to Finland mainly exempted from notification duty. Sources: Steinbock, (200 I :87) Korea. Considering the historical foundations ofthe high stateness ethos in Korea, it is no surprise that the more external forces from the bilateral or multilateral negotiations for opening the market or financial crisis in 1997 are discussed as the general motives for the introduction of competition in the telecommunications market 171

PAGE 186

(Cho et. al., 1996; Hong, 1998; Hyun and Lent, 1999; OECD, 2000e; Lee and Lee, 2001). As in the case of the bilateral negotiations for the opening of the telecommunication market, the U.S. efforts to open the Korean telecommunications market have been "incessant" and ''tenacious," to the extent that Korea is the only country to be twice designated (i.e., in 1989 and in 1996) by the U.S. as a priority foreign country (PFC) (USTR, 1998; Hyun and Lent, 1999).76 For instance, after its designation as a PFC in 1989, both the duopolistic system in the international and wireless service markets and the full competition in the value-added service market were introduced. As in the case of the multilateral negotiation for the opening of the Korean telecommunications market, the Uruguay Round (UR) Agreement passed in 1994 and the advent of the WTO and its Negotiating Group on Basic Telecommunications (NGBT or GBT) provided a motive for the telecommunications liberalization (Choi, 1997; Chung, 1996). For example, the UR negotiations brought the 1994 reform plan and the NGBT provided a motive for the 1995 liberalization (Table 6-3). 76 By Section 1377 of the Omnibus Trade and Competitiveness Act of 1988, the designation as a PFC by the United States Trade Representative was based on the examination of not only trade barriers, but also market potential and any measurable progress toward trade liberalization. For example, when Korea was first identified as a PFC in 1989, problems cited included discriminatory procurement practices, lack of transparency and trade secret protection (USTR, 1995). 172

PAGE 187

Table 6-3: Liberalization of Korean Telecommunications Market 1982: Korea Telecom Authority (later, Korea Telecom) created. 1990: Competition for value added network services introduced. 1990: Beginning of duopoly for international telephone service (DACOM). 1992: Competition in paging services allowed (2"d wireless pager license given). 1994: Beginning of duopoly for mobile services. 1995: Decision in introducing competition in national long distance market. 1996: 27 new service providers licensed in the following areas: PCS (3), Trunk Radio Services (6), CT-2 (11), leased line facility rental (2), international telephony (1, Onse enters international market as 3rd service provider), radio paging (1), and wireless data transmission (3). 1997: 9 new service providers licensed in following areas: local telephone service (1), TRS (4), Leased line service (2), radio paging (1), and long distance (1). 1997: Revision of classification of telecommunication services introducing new category of "special telecommunication service providers" (voice resale, Internet telephone). Source: MIC (2000) Furthermore, the financial crisis in 1997 provided yet another strong motives for the liberalization of the Korean market in general, and the liberalization ofthe telecommunications market in specific. 77 In short, with the response to the external forces for the opening of the telecommunication market, the Korean government took strong initiatives to liberalize the communications market during the 1990-1997 period of the incremental transition phase from public monopoly to competitive market structure. 77 Interestingly, Lawrence Summers, the U.S. Deputy Treasury Secretary, stated that, "In these past few months the International Monetary Fund (IMF) has done much more to liberalize Asian economies and open their markets to U.S. goods and services than many, many years of trade negotiations in the region" (Mail Business Newspaper, I 998). 173

PAGE 188

Summary Even though the introduction ofcompetition in the local telephone market was not primarily intended to enhance the Internet diffusion, the economic rationale of introducing competition in local telephone markets gave a strong impetus to the proliferation oflnternet diffusion by facilitating network competition (i.e., the network competition between dial-up connection via a telephone network and the alternative connection via a DSL or a cable network) as well as by lowering the local telephone charges. In general, the economic rationales of de-regulative policies intending to enhance the entrepreneurial activities (i.e., private investment and innovation) of service providers by introducing competition in the local telephone market can be identified in the three nations. In specific, in the U.S., the emphasis of the Nil plan was on the market-led industrial strategies based on American strong trust in the free market. The U.S. government sought to encourage private sector investment in information infrastructure by private service-providers rather than directly funding governmental (or public) resources into telecommunications networks. Stich an economic rationale was consistent with the emerging government role in the local telecommunications market focused on preventing the provider-centered monopolistic dominance oflocal facilities (TCA, 1996). 174

PAGE 189

In Finland, the emphasis of the initial national strategy for achieving an information society was on overcoming a national economic depression, which caused the focus of national strategy to be on business potentials and technology development. Such an economic rationale caused the two separate introductions of competition into the local telephone market. Furthermore, this economic rationale was intended to enhance the entrepreneurship of service-providers by introducing competition between the two separate monopolistic local telephone companies -that is, the private customer-tied monopolistic Finnet Group in urban area and the state-owned monopolistic Telecom Finland in rural area. In Korea, responding to both the external forces urging the opening of the telecommunications market as well as the economic depression, competition in local telephone market was introduced. These economic rationales in Korea was intended to enhance the entrepreneurship of service providers between private-owned new in comer (Hanaro Telecom) and the state-owned monopolistic Korea Telecom. Policy Programs and Agencies In this section, the comparative characteristics of both the policy programs for the de-regulation of the local telephone market and the specific policy agencies to implement the de-regulation policy in the three nations are briefly reviewed. Reviewing the comparative characteristics of the policy programs for the de regulation of the local telephone market intends to identify (I) the type of local 175

PAGE 190

competition (i.e., facility-based competition or service-based competition), (2) the ownership of the local facilities, and (3) pricing structure in local market in the three nations. In addition, reviewing the comparative characteristics of the policy agencies is designed to reveal (1) the forms ofthe independent regulatory bodies, (2) the degree of independence or autonomy, and (3) the division of regulatory responsibilities. Theoretical Policy Programs and Agencies De-regulative Policy. After initiating the Internet diffusion, the specific policy programs of liberalizationi.e., privatization, competition, and price de-regulation in the local telephone market-that can be classified into a de-regulative mode in Lowi's policy typology are necessary to encourage entrepreneurial activities (e.g., private investment and technological innovation) ofthe service-providers. As such, the degree of de-regulation of the service provider axis in a triad oflnternet policy in a nation can influence the comparative characteristics of the proliferation of the Internet. Independent Regulatory Agencies. By securing a fair environment for competition among the service-providers, the degree of independence of regulatory body in a nation is expected to influence Internet diffusion in the second stage of Internet proliferation. Furthermore, as Henisz (2000) argued, the number of independent branches of government with veto power is expected to foster private investment on the Internet with providing the more "policy credibility" and 176

PAGE 191

"predictablility." In this respect, the nominal "independence" of regulatory body from both the state (for example, the independence from political favoritism) and the service-providers (for example, the independence from the backdoor influence of the most powerful PTOs) was called for in the communications arena (Melody, 1997). Considering the historical development of state-owned service-providers such as public telecommunications operators (PTOs) especially in European nations (for example, France Telecom in France and Deutsche Telekom in German), a requirement for the institutional independence of regulatory function from policy function of executive branch resulted in the institutional arrangement of the so-called independent regulatory agency. The demand for independent regulatory agency emerged with the worldwide trends toward the liberalization oftelecommunications market. As a example ofthe most independent regulatory agency to date, U.S. Federal Communications Commission (FCC) has been spoken of even though an FCC-type independent regulatory agency is considered a unique product of the U.S. Constitutional system, with its elaborate division of powers among executive, legislative and judicial branches of government (Melody, 1997; Stillman, 1996: 54-59; Rosenbloom, I 998: 190). Considering that the U.S. FCC-type independent regulatory agency has relatively broad mandates and significant freedom (at least in theory) both to interpret and to enforce their mandates by being placed outside the formal executive departments with quasi-legal and executive powers to regulate prices in telecommunications markets to curb monopolistic and unfair practices by incumbent 177

PAGE 192

service-providers, few nations have permitted such a degree of independence, at least so far (Melody, 1997; Stillman, 1996). The preferred alternative could be to have the regulatory agency within a ministry, thereby putting in a sympathetic position but undermining its institutional independence. Individual Policy Programs and Agencies In the second stage of Internet diffusion, both the individual policy agencies and the specific policy programs for de-regulation of local telephone market in the three nations can be compared in detail both to each other and to the theoretical policy agencies and programs. United States. Unbundling and Resale. The Telecommunications Act passed in 1996 prohibited state and local governments to maintain any legal barriers to local entry (OECD, 1998c). The Telecommunications Act of 1996 continued three entry routes that new competitors could use-resale, unbundling (or line-sharing), and separate facilities. The aim ofthe Act was not to choose one particular mode of local competition but to establish the conditions where all three would be available and to let competitors choose the mode of entry that makes the most technical and economic sense to them (OECD, 1998c). 178

PAGE 193

The Act was implemented by three major FCC orders known as the "Competition Trilogy"the rules for (1) local interconnection or so-called local competition (FCC Order 96-325), (2) universal service (FCC Order 97-157), and (3) access charge reform (FCC Order 97 -158) designed to reform the regulatory regime (Harris and Kraft, 1997; Noll, 1998). The key feature ofthe Telecommunication Act of 1996 and its implementing orders, which was considered one ofthe main determinants of the proliferation ofthe Internet, was the openness ofthe local telecommunications network (Werbach, 1997; Oxman, 1999; Baret al., 2000). The network openness for the local exchanges is summarized below: (I) Interconnection: Incumbent Local Exchange Carriers (incumbent LECs, or local telephone companies) are required to provide interconnection to any requesting carriers at any technically feasible point. The FCC concludes that price should be based on Total Element Long-Run Incremental Cost (TELRIC) plus a reasonable share of forward-looking joint and common costs. (2) Unbundling: Incumbent LECs are required to provide non-discriminatory access to network elements on an unbundled basis at any technically feasible point. The FCC concludes that price should be based on Total Element Long Run Incremental Cost (TELRIC) plus a reasonable share of forward-looking joint and common costs. (3) Resale: Incumbent LECs are required to offer for resale any telecommunications service that the carrier provides at retail to subscribers. The 179

PAGE 194

FCC concludes that the price of resale services should be set at a discount off retail based on the costs that the incumbent LEC can avoid by selling at wholesale rather than retail. Both "resale" and "unbundling"in another words, "retail model" (Green and Teece, 1999) or "network openness" (Oxman, 1999; Baret al., 2000)represent the main characteristics ofU.S. local competition. The main pro-competitive rationales for resale and unbundling is that it may promote facilities-based competition. But it is important to understand that unbundling and resale also impede facilities competition: regulatory prescriptions for unbundling at prices that are excessively low may act against the consumer's longer run interests through the reduction of incentives for companies to install their own wired or wireless networks (OECD, 1998c).78 Those arguments against the unbundling provision, summarized in Table 6-4, can be considered one reason why local competition in the U.S. has not developed as quickly as anticipated (Harris and Kraft, 1997; OECD, 1998c):79 78 In this respect, it is appropriate that the so-called Tanzin-Dingell bill, focusing on promoting the local facilities competition especially in the digital subscribers line (DSL) was approved in House on February 27,2002. 79 One empirical example of this argument is that the total broadband penetration of the U.S. measured by DSL and cable modems as the alternative routes to the dial-up connection to the Internet was ranked as the third among 29 member nations at the end of2000, while the penetration measured in June of 2001 was ranked as the fourth (OECD, 2001 b: 14) (Appendix K). 180

PAGE 195

Table 6-4: Pros and Cons ofUnbundling Pros Cons Encourages competition by reducing the Reduces incentive for construction of economic barriers to entry by allowing competitive network facilities. new entrants to construct some components of their networks and to obtain other components from the incumbent operator. Encourages innovation, since new Undennines investment in alternative entrants can combine new technologies access networks (wired or wireless). with comQ_onents of existing networks. A voids unnecessary and inefficient Can enrich the new entrant at the duplication of components. expense of the incumbent operator. Facilitates access to rights of way, Requires detailed regulatory towers, etc., by new entrants, and avoids intervention and technical the disruption to streets and to the coordination. environment during a duplicate roll-out Provides a new revenue stream to Requires technical coordination incumbent. between operators. Source: OECD (2001 b: 18) Ownership and Pricing Structure. Local telephone service in the U.S. has been predominantly provided by "privately owned" monopolistic companies. Historically, monopolistic local exchange carriers in the U.S. have been subject to "rate of return" regulations, a policy pennitting them to recover and earn a reasonable profit on their net capital investment that is, their "rate base" and to cover their operating expenses (Harris and Kraft, 1997). In local telephone markets, rate of return regulation was usually accompanied by granting a monopoly franchise and imposing rules requiring local exchange companies to provide service to all customers in the service territory who requested it. 181

PAGE 196

In the late 1970s, state and federal regulators began to shift away from rate of return regulation toward price cap or profit-incentive regulation based on the principles of economic efficiency. Price caps typically require regulated companies to reduce prices steadily based on some measurement of average industry productivity growth. Price caps intend to improve technical efficiency by using the profit incentive to encourage regulated firms to produce a given level of output at a low cost. Furthermore, to promote universal telecommunications service by keeping prices below cost for rural subscribers, state and federal policy-makers ofthe U.S. have generally required local exchange carriers to provide service at geographically averaged rates, rather than varying rates to reflect the higher costs of serving areas with low population density. In addition, the pattern of price regulation typically included cross-subsidies where basic residential local exchange service was provided below cost, while services like business, long-distance access, local toll were priced above cost. These cross-subsidies were designed to promote subscribership by residential customers. However, they also led to a structure of prices that did not reflect different costs. The principle of geographically averaged rates was acknowledged in the Telecommunications Act of 1996 in Section 254, (a) (3): Consumers in all regions of the Nation, including low-income consumers and those in rural, and high cost areas, should have access to telecommunications and information services, including inter-exchange service and advanced telecommunication and information services; they are reasonably comparable to those 182

PAGE 197

services provided in urban areas and that are available at rates that are reasonably comparable to rates charged for similar services in urban areas. Moreover, the specific methods of reimbursements (or access charges) for crosssubsidies among the long distance companies, called interexchange carriers (IXCs), the local exchange carrier (LECs), and the customers (telephone subscribers) are mandated in the FCC's rules for access charge refonn (FCC Order 97-158). Among others, the FCC Order mandates that telephone subscribers pay a flat charge for local access. This flat Subscriber Line Charge (SLC) is paid directly to the LECs. A flat rate regime for local calls that is provided below costs basically subsidizes the Internet usage indirectly by providing an unlimited access to the dial-up connection to the Internet compared to the usage-based charging structure for local calls (Pospischil, 1998). Independent Regulatory Agencies. Among the other agencies already reviewed in chapter 4, the Federal Communications Commission (FCC) of the U.S. is a typical example ofthe independent regulatory agency. The FCC is an independent agency consisting of five Commissioners, nominated by the President and subject to confirmation by the Senate. One of the Commissioners is appointed Chair by the President. Decisions are made by simple majority rule ofthe Commissioners. Once confirmed, they cannot usually be removed from office during their five-year term. The FCC has exclusive jurisdiction over inter-state matters, as well as intra-state matters where federal legislation pre-empts state authority (Tyler and Bednarczyk, 183

PAGE 198

1993; Brock, 1998; Huber, 1997; Kerf and Geradin, 1999; OECD, 2000b; GarciaMurillo and Macinnes, 2001). The FCC is considered an independent regulator because it has important regulatory functions such as issuing licenses (fixed and mobile), oversight of license requirements, regulation on interconnection, and spectrum/number allocation, regulation on pricing (price caps) and universal service. Finland. Initiation ofLocal Competition. Full competition in the Finnish telecommunications market, including local, long-distance, and international markets, was introduced in January 1994. Until that point, the telecommunications market had been divided into two areas of operation-i.e., local services in urban southern areas and long distance as well as local service in rural eastern and northern areas (OECD, 1992; Nattermann and Murphy, 1998; Martenson, 1998). A special characteristic ofthe highly sophisticated Finnish telecommunications system is the local telephone companies, which are divided into privately owned or customer-owned companies in urban areas and state-owned companies in rural areas. The Finnish telecommunications market was characterized by statutory monopoly rights of numerous private local operators80 in urban southern areas (Nattermann and Murphy, 1998). In preparation for total liberalization (or de80 The small local operators numbered 815 in the 1930s, and by 1965 the number of local operators had fallen to only 88. By 1994, this number had again roughly halved. In 1998, the Finnet members were 48 (Nattennann and Murphy, 1998). 184

PAGE 199

regulation) of the market, the private operators joined forces to found the Finnet group of companies, or Finnet Group (by 2000, its break-up generated Elisa Communications) consisting of the remaining 46 independent private telephone companies. The local access services in the less-densely populated regions of northern Finland and the long distance market were the statutory monopoly of the governmental Administration of Posts and Telegraphs (P&T). In 1990, the P&T was transformed from a state agency offering telecommunications service into an independent business enterprise (akin to the corporate form of the postal service in the U.S.) and has remained a fully state-owned enterprise. In 1994, the two activities ofP&T-postal services and telecommunications services were separated and organized under the PT Finland holding. Its subsidiary, Telecom Finland Ltd. (its successor is Sonara Ltd.), became a stock-holding company fully owned by the government. In the local telephone market, the Finnet Group is the incumbent with its 46 former regional monopolies. These cooperatives had more than a century to build up a dense local network effectively connecting each household to the network. More interestingly, these cooperatives themselves have so far not entered the markets of each other, although the Finnet Group holds an unlimited license enabling it to operate throughout the country. This reflects the collusive conduct of the cooperatives in the local market due to cross-ownership of their subsidiaries of the Finnet Group (Nattermann and Murphy, 1998). 185

PAGE 200

The entrants into the cooperatives' local markets in the southern urban areas of Finland in 1994 were Telecom Finland (TF) and Telivo (or now Telia).81 TF's former local monopolies in the rural northern two-thirds of Finland have not attracted much attention from other operators due to low population density in these regions. Therefore, the local access markets in which competitions are expected to take place are the former monopoly regions of the cooperatives. However, compared to the 27 percent market share ofTF before 1994, the current market share of roughly 30 percent indicates that entry into the local markets has been slow at best, with Telia's entry having been even slower (Natterrnann and Murphy, 1998). On the contrary, within a week of the liberalization of the domestic long distance and local market, the Finnet Group had gained a 50 percent market share in the long-distance market. This is the basic reason why the more elaborated mechanisms for local competition were introduced in 1997. Elaboration of Local Competition. To further stimulate the competition introduced in 1994, the Ministry of Transport and Communications in 1996 started to prepare a new law for telecommunications. The Telecommunications Market Act entered into force in the beginning of June 1997, replacing the Telecommunications Act of 1987.82 The major changes related to the local competition include: 81 Telia, formerly known as Telivo, when it was a subsidiary of the electric utilities holding company, is now a part ofT eli a of Sweden (Nattermann and Murphy, 1998). 82 For an English translation of The Telecommunications Market Act, see http://www.vn.fi/vn/lm/vho/tu. 186

PAGE 201

(1) A telecommunication network operator with significant market force in local telecommunications services was required to lease subscriber connections to other telecommunication companies so that they would be priced in the same way as their own service operators. (2) A telecommunication network operator with significant market power could be imposed more stringent obligations relating to competing operations than those imposed on other telecommunications companies. For example, local incumbents (or Finnet Groups) were imposed obligations to lease their local access lines to other full license competitors (TF or Telia). (3) Obligation was demanded for telecommunications operator to separate the provision of telecommunications networks and telecommunications service from each other financially. This obligation of the so-called "internal or financial \ divestiture" between network and service provision intended to force each carrier (especially, Finnet Group) to offer its network at the same conditions to its own service affiliate as to its competitors by abolishing intra-firm cross-subsidies. The Finnish Telecommunication Act of 1987 did not require operators to lease their networks to other operators licensed to construct their own networks. Therefore, the local telephone companies, or the Finnet Groups, were not required to lease their access lines to Telecom Finland. As of June 1997, however, all operators are obliged to lease their local access lines to all full license carriers. The regulatory agency (FICORA) had determined that the ability of the cooperatives to refuse resale of their 187

PAGE 202

subscriber lines was one of the main reasons for the slow development of full-blown competition in the local telephone market. Ownership and Pricing Structure. The Finnish telecommunications market was characterized by the "cooperative ownership" of the numerous local telephone operators in urban southern areas (Nattermann and Murphy, 1998). The local operators were run as "cooperatives" owned by their respective customers in their traffic areas. Most of these telephone operators were owned by their customers while a small number were entirely municipally owned. The aim of these local operators was not to maximize profits but rather to minimize prices in accordance with the interests of the customers who were also the shareholders. However, the practice of minimizing overall prices might have resulted in cross-subsidization in small cities and rural areas, and thus have raised prices for certain customers in that the high cost of networking in small and rural areas is supposed to bring in the high local call charges. Thus, the cooperative nature of the local operators (Finnet Group) has always functioned as a price regulation mechanism for its services. Due to the limited area of coverage of any given cooperative, consumers were able to compare prices between their cooperative and others. Since the consumers were also shareholders of the company, they were able to vote for lowering prices if their own cooperatives' tariffs moved significantly above those of the surrounding operators. 188

PAGE 203

Until recently, each customer was able to purchase a share in the cooperative for each subscription line purchased from the cooperative. These shares not only gave the customer partial ownership of his or her company but also reduced the customers monthly phone bill by a flat rate of 50 FIM (approximately US $10) per month. Due to the cooperative nature of the local operators functioning as a price regulation mechanism and the similar pricing structure of Telecom Finland and the Finnet Group, the only way for a customers to lower his or her monthly telephone charges is to become a member in the cooperative. Thus, the tying of the consumers to their original local operator combined with the right to refuse leasing of lines to competitors has slowed Telecom Finland's move into the local telephone market. Such a flat rate regime for local calls caused by the cooperative nature of the Finnish local operators can be considered to subsidize the Internet usage at least indirectly by providing an unlimited access to the Internet as in the case of the U.S. Independent Regulatory Agencies. In Finland, while the Ministry of Transport and Communications is responsible for telecommunications policy function, an independent regulatory agency, the Telecommunications Administration Center (TAC, later FICORA), is responsible for regulatory function of the telecommunications sector. The head of TAC is nominated by the President, and enjoys an indefinite term of office. T AC is equivalent to an independent regulator because it has important regulatory 189

PAGE 204

functions such as issuing licenses (fixed and mobile), oversight of license requirements, regulation on interconnection, and spectrum/number allocation. Korea. Local Competition. The Korean government embarked on a structural reform after the conclusion of the WTO negotiations on basic telecommunications on February 14, 1997 with the WTO agreement entering into force in February 5, 1998 (Lee and Lee, 2001). The motivation was largely twofold: to saturate the market with as many domestic providers as possible before the market opening; and to make Korean laws and regulations conform with the scheduled commitments (Lee and Lee, 2001). In this vein, Information and Communication Minister Kang Bong-Kyun stated, "The government has selected nine future service providers in the five communications sectors as a way to introduce a competitive system among domestic companies prior to opening the market to foreigners." (The Korea Times, 1997). In June 1997, the MIC allowed new firms to enter the domestic local telephone market and long-distance telephone market. Hanaro Telecom was selected as the second local phone service-provider, and the international call service-provider, Onse Telecom, was given the third long distance service. To ensure competitiveness, the Hanaro Telecom consortium was composed of 444 companies, including the second long-distance and international telephone services 190

PAGE 205

provider (DACOM), Korea Electric Power Corp., and the big three conglomerates (the so-called chaebols), Hyundai, Samsung, and Daewoo (Hwang, 1997). In April1999, Hanaro Telecom, mainly focusing on the deployment of Asymmetric Digital Subscriber Line (ADSL) for high-bandwidth Internet services, launched its business in the local market that has been regarded as the "impregnable fortress" of incumbent Korea Telecom (Lee and Lee, 2001). By mainly focusing on the introduction of ADSL technology in the local market, Hanaro Telecom provided the incumbent PTOs (Korea Telecom) with an incentive for construction of the alternative ADSL network facilities to connect the Internet. Thus, the competition in the Korean local markets was more geared toward a facility-based compared to the service-based or resale-based competition in the U.S. local market. The regulatory regime experienced a significant change in August 1997, in accordance with the amendment of the "Telecommunications Business Act." Aside from the mitigation of foreign ownership restrictions, one ofthe most important changes was in that of the classification of service providers. Current telecommunications service providers were classified as facility-based, special, and value-added service providers. Based on both the ownership of the telecommunications facilities (owned or leased) and the types of telecommunications service (value-added or special), they were differentiated with respect to the market entry condition, which were licensing for facilities-based, registration for special, and notification for value added service providers (Table 6-5). 191

PAGE 206

Table 6-5: Korean Classification ofTelecommunications Service Providers Category F aci I ities-based Special Service Value-added Service Provider Provider Service Provider Facilities Owned Leased/Use Leased Services Fixed Telephony, Type I All value-added Telegraph, (Switched Reseller) telecommunication Leased line Voice resale, services provided serv1ces, Internet telephony, by the facilitiesMobile services, International based service And Other services Callback, etc providers specified by the Type II Minister (Switchless Reseller) Rebiller, etc Type III: In-building Communications Services Entrance Licensing Registration Notification Requirement Source: MIC (2000) Among others, the Korean government newly introduced a category of resalebased services, so-called "special services," which had never been permitted legally in Korea until then (MIC, 1999; OECD, 2000e; Lee and Lee, 2001 ). 83 In addition, one of the comparative characteristics ofthe classification of telecommunications service providers in Korea is a sub-category of leased line services among the facilities-based service-providers, mainly focusing on leasing its own facilities. This sub-category for leased line service-providers, such as Thrunet, GNG 83 As of the end of December 2000, the number of special service providers is 276 (MIC, 2001 ). 192

PAGE 207

Networks, and Powercomm, fortified the trends toward the more facility-based competition in the Korean local telephone markets. 84 A weakness in the local competition of the Korean telecommunications market is the predominance of the incumbent service provider (Korea Telecom). For instance, incumbent KT's market share of local service, measured from the number of subscribers as of March 2001, is 98.2 percent ofthe total local subscribers, while Hanaro Telecom's market share of local service is 1.8 percent ofthe total local subscribers (MIC, 2001). In this respect, the lack of provision for unbundling the local loop is frequently regarded as one of the most important causes ofthe predominance ofthe incumbent carrier in the Korean local telephone market, (Booze-Allen & Hamilton, 1998; OECD, 2000e). In particular, the OECD recommended in The OECD Review of Regulatory Reform in Korea (2000:89) that "access to raw copper (or local networks) can enable new entrants to begin rapidly competing in the local loop and provide a way to pressure the incumbent to improve efficiency and pricing structures," and that "placing time limits on access to raw copper (or local networks) provides incentive to entrants to construct their own facilities." The first recommendation on the benefits of the introduction of unbundling requirements in the local markets can be debated as reviewed in the pros and cons of unbundling considerations (table 6-4). Nevertheless, the second recommendation can 84 As of the end of December 2000, the number ofleased line service providers is 13 (MIC, 2001). 193

PAGE 208

be seen as an alternative to reduce the predominance of the incumbent service provider (KT). So, the introduction of the confined unbundling requirements with a certain "time limit" into the Korean local telephone market can be one proper alternative to finding the so-called "marketplace" in the local telephone market. Ownership and Price Structure. Consistent with the historical origins of most European local telephone companies like Deutsche Telekom and France Telecom, the incumbent local telephone company, Korea Telecom (KT) is a state-owned company, with the government owning 59% as of September 1999. The Act on Privatization of KT provides a legal basis for independent management of KT although the government retains shareholder rights. Asymmetric regulation is imposed on incumbent KT, in that its local tariffs require approval from the Ministry of Information and Communication and it is subject to interconnection requirements. KT is also at present responsible for universal service in Korea. At the end of 1995, the prior approval system for telecommunications tariffs was abolished. Under the old system, MIC had to approve all telecommunications tariffs from all operators. Now, operators are at liberty to determine their own tariffs, and any changes in tariffs have to be reported to the MIC. However, the tariffs ofKT's local telephone service require formal approval of MIC in that KT is dominant in the local telephone markets. Compared with other OECD nations, Korea's prices are well below the OECD average in both the business and residential tariff cases, including the 194

PAGE 209

fixed (i.e., monthly rental) and usage components (OECD, 2000a: 41-42). This results from government policies in the 1980s and I 990s that maintained low prices as part of universal service policy. Such an approval system ofthe Korean government was recommended to introduce a price cap system, which is consider more transparent, and not subject to political interference (Booze-allen & Hamilton, 1998; OECD, 2000e). The MIC is presently reviewing the possibility of introducing price caps (MIC, 2000). Independent Regulatory Agencies. In Korea, while the Ministry of Information and Communication is responsible for telecommunications regulation, it has a semi independent regulatory agency, the Korea Communications Commission (KCC), within the Ministry. The KCC is relatively independent from the Ministry in that commissioners in KCC are nominated by the President, and enjoy a legally guaranteed term of office. Furthermore, the KCC can make binding decisions on disputes between telecommunications operators. Nevertheless, the KCC is not equivalent to an independent regulator because the KCC is a part of the Ministry and the most important regulation such as licensing, spectrum/number allocation and price regulation are exercised by the Ministry. Summary The policy programs and agencies for the de-regulation of the local telephone market in the three nations can be examined in view of(l) the type oflocal 195

PAGE 210

competition (i.e., facility-based competition or service-based competition), (2) ownership of the local facilities, (3) pricing structure in the local markets in the three nations, and (4) independence ofregulatory functions. In the U.S., the comparative focus of the introduction of competition in the local telephone market was based on retail-based competition (i.e., resale and unbundling) with a flat rate regime for local calls provided by privately owned monopolistic companies, all monitored by a fully-independent regulatory agency (FCC). In Finland, the comparative focus of the introduction of competition in the local telephone markets was changed from the more facilities-based competition in 1994 to the more resale-based competition in 1997 with an independent regulatory agency (TAC). In addition, local telephone services have been provided by the private/public mix of telecommunications companies: The private-owned or customer-owned regional monopolistic cooperatives (Finnet Group) provided local telephone service in urban areas; the state-owned monopolistic companies (Telecom Finland) provided local telephone service in rural areas with a flat rate regime for local calls tied with the local customers. In Korea, with the semi-independent regulatory agency (KCC), the comparative focus of the introduction of competition in the local telephone market was on the more facility-based competition complemented by introducing the resale competition -i. e., the "type I" (see table 6-5, p. 192) service provider of the so-called "special service provider"in that the unbundling is not yet institutionalized. In addition, the local 196

PAGE 211

telephone services in Korea have been provided by the state-owned monopolistic Korea Telecom with the low prices approved by the Ministry of Information and Communication as part of public service. Policy Impact: Economic De-regulation and Internet Diffusion In this section, the diachronic relationship of the economic de-regulation of the local market and the second stage of Internet diffusion is investigated. For this purpose, we briefly review (I) the linkages between the de-regulation oflocal telephone and the Internet, (2) the policy impact of de-regulation in local to Internet diffusion (3) the time matching comparison between the local de-regulation and the second stage of Internet diffusion, and (4) government role in the second stage of Internet diffusion. De-regulation and Internet: Two Routes to Internet Taking off The main rationale for de-regulation or liberalization in the local telecommunications sector is that competition, privatization, and de-regulation of the price structure will enhance economic efficiency and entrepreneurial activities of service-providers. As a result, de-regulation is thought to create low local usage charge in the short run and private investment and technological innovation in the long run in the local telecommunications sector. This economic rationale is supposed to affect 197

PAGE 212

Internet take-offvia the two separate routes: i.e., (I) enhancing facilities-based competition and (2) lowering local usage charges. Competition among the service-providers is thought to promote facilities-based competition by introducing competition in local telephone market and privatization. In the long run, the access network for connecting the Internet will be upgraded. Typical examples ofthe upgraded Internet connection are the access network using the cable and the digital subscriber line (DSL). Another route is through lowering the usage charges for the telephone network. Internet access charges are composed of two elements, that is, (1) usage charges for the telephone network and (2) the Internet Service Provider's (ISP) access charge. In most nations, ISP charges typically form "less than half the total cost" of accessing the Internet (Kelly and Petrazzini, 1997) or free ISPs or flat rate (OECD, 2000a). In this respect, low or no telephone usage charges can be thought to significantly boost Internet taking-off.85 Policy Relevance The relevance of policy programs (i.e., competition in local, privatization, and local pricing structure) for enhancing the economic efficiency and entrepreneurial 85 For example, Internet diffusion has been relatively high in Gulf nations where there is no local call charge. Bahrain, Kuwait, and the United Arab Emirates all have relatively high penetration rates compared to other developing nations (Kelly and Petrazzini, 1997). 198

PAGE 213

activities of service-providers in the three nations can be evaluated in view of the mode of competition and local pricing structure. Mode of Competition. The mode of competition is thought to affect the Internet diffusion. While the U.S. and Finland focus more on service-based local competition (i.e., resale in Finland especially from 1997 or unbundling in the U.S.), Korea focuses more on facility-based competition. Especially, Hanaro Telecom, the incoming local telephone company, initiated facility-based competition with the monopolistic incumbent KT by mainly dealing with the deployment of Asymmetric Digital Subscriber Line (ADSL) for high-band-width Internet services. Furthermore, the leased line service-providers like Thrunet, GNG Networks, and Powercomm fortified the trends toward more facility-based competition in local telephone market. Thus, the more facility-based competition in local competition in Korea can be considered one of the reasons why Korea is among the first rank in the broadband access penetration to the Internet among the OECD nations in 2000 and 2001 (OECD, 2001 b) (Appendix K).s6 86 There are several alternative explanations for the success in the roll out of high speed Internet access in Korea. For example, (I) the popularity oflnternet telephony as in Dialpad service in Korea (ITU, 200lc), (2) the popularity oflnternet cafe to access the Internet, (3) the high proportion ofKoreans living in apartment buildings (OECD, 2001 b), ( 4) the technological advantages of DSL using existing copper local loops (OECD, 200lb). 199

PAGE 214

With regard to service-based local competition, there are some arguments about whether or not resale and unbundling (i.e., service-based local competition) can promote facilities-based competition as summarized in Table 6-4 (page 181 ). The arguments for the unbundling requirements focus on the economic benefits of competition such as (I) avoiding unnecessary and inefficient duplication of components and facilities, (2) reducing the economic barriers to entry by providing some components of the incumbent network, (3) encouraging innovation since new entrants can combine new technologies with components of existing networks, and ( 4) providing a new revenue stream to the incumbents by resaling the components of the incumbent network. The arguments against unbundling requirements focus on the economic costs of competition such as (I) enriching the new entrant at the expense of the incumbent operators, (2) reducing incentive for construction of competitive network facilities, (3) undermining investment in alternative access networks, and (4) requiring detailed regulatory intervention since unbundling requires technological coordination between operators. Extreme unbundling requirements can increase the economic costs of competition by ( 1) reducing incentive and investment for alternative access network and (2) increasing regulatory intervention for coordination. Therefore, the more limited unbundling requirements with specified time limits can promote greater facilities-based competition. Thus, the mode of local competition (facility-based or service-based with 200

PAGE 215

limited unbundling requirements) can affect Internet diffusion by promoting investment in alternative access networks. Local Pricing Structure. Local pricing structure is considered another economic route to affect Internet diffusion. The mono-causal and economic perspective for deregulation or liberalization argued that competition among local service-providers could promote Internet usage by lowering (or unmetered) local usage charges for the telephone network. As a whole, flat rate or low rate of local telephone charges in the three nations can be considered one of the main determinants to take off at the stage of Internet diffusion, irrespective of their origins (or causes) of flat rate or low rate of local telephone charges. Nevertheless, the origin of flat rate or low rate of local telephone charges was caused not by an economic rationale but a social rationale. The local telephone market in the U.S. and Korea can be considered a "muddle" with an inherently-distorted pricing mechanism by the non-market considerations (or sociopolitical forces) in the name of universal service as in the U.S. and public service as in Korea (Harris and Kraft, 1997).87 In the U.S., the local telephone services have been provided with a flat rate regime as a part of universal service via a rate-averaging for indirectly subsidizing local callers. In Korea, the local telephone services have been 87 Especially, rate-averaging for subsidizing the local callers can be considered one of the main reasons why the pricing mechanism in the local telephone market is distorted by the non-market considerations. For the further information, see chapter 7. 201

PAGE 216

subsidized with low prices by an approval ofthe Ministry oflnformation and Communication as a part of public service. In Finland where the local telephone pricing mechanism is less distorted in that there is no rate averaging for subsidizing local callers, local telephone services have been provided with a flat rate regime tied tightly with the local customers via a unique nature of local cooperatives. Thus, to provide a proper explanation ofthe origins of flat rate or low rate of local telephone charges, a purely economic rationale needs to be complemented with the social rationale for Internet diffusion, which will be reviewed further in the following chapter 7. De-regulation and Proliferation As a whole, de-regulation in the local telephone market can be justified by enhancing economic efficiency and entrepreneurial activities of service-providers. Furthermore, the empirical evidence on the time periods of the introduction of local competition in Finland and Korea and the privatization ofNSFNET in the U.S. is consistent with the empirical evidence of the Internet taking-off. According to the ITU chronological data (1991-2000) on Internet users per I 00 inhabitants in the U.S. (Appendix N), the second stage oflnternet diffusion can be roughly estimated after the year of the taking-off of Internet users in 1995. In these time periods, the economic de-regulation activities can be found in (1) privatization of 202

PAGE 217

NSFNET in April, 1995 and (2) introduction of local competition through the Telecommunication Act of 1996. In Finland, the ITU chronological data (1991-2000) (Appendix N) show that the second stage of Internet diffusion can be roughly estimated between 1994 to 1998. In these time periods, the economic de-regulation activities can be found in (1) the initial introduction of competition in local telephone in January, 1994 (2) elaboration of competition in local telephone with the Telecommunication Market Act in 1997. In Korea, the ITU chronological data (1991-2000) (Appendix N) show that the second stage of Internet diffusion can be roughly estimated between 1999 to 2000. In these time periods, the economic de-regulation activities can be found in competition in local telephone market (April 1994). Even though the three nations show different degrees of the de-regulation-i.e., the different mode of local competition, the different ownership and pricing structure the role of the government during this period can be theoretically identified as an efficient promoter of service-providers' economic efficiency and entrepreneurial activities among the triad oflnternet policy. Thus, the economic dimension of the 3-D synchronic model can be roughly matched with the second stage ofthe 3-S diachronic model of this dissertation. 203

PAGE 218

Summary and Implications To answer the second research question and to test theoretical proposition (2b) about the diachronic characteristics of policy pattern in the second stage, policy factors affecting the proliferation of Internet diffusion in the three nations are compared and contrasted based on the 3-S diachronic model. This comparison is intended to produce the similarities and differences of policy pattern in the second stage of Internet diffusion. Chronologically, different periods (from liberalization to social control) in the three nations can be identified to fall into the second stage of Internet diffusion: i.e., from April1995 to 1996 in the U.S., from January 1994 to March 1998 in Finland, and from April 1999 to January 2001 in Korea. During these times in each nation, the comparative focus of communications policy for the mature use of the Internet was on enhancing economic efficiency and entrepreneurial activities of service providers (or service-provider variable among the three variables in a triad of Internet policy). Based on the economic logic of liberalization, the communications market was de-regulated at this stage with the policy tools of de-regulatory pol icy, such as introduction of competition, privatization, and price de-regulation. Thus, the comparative similarities of policy pattern in the second stage can be found in its focus on the service-provider variable (or economic dimension) in the 3-D model during the second stage in the 3-S model. These comparative similarities can produce a theoretical suggestion of 204

PAGE 219

governmental role as a fine tuner for enhancing economic efficiency and entrepreneurial activities of service-providers. The comparative differences of policy patterns in the second stage are closely related to the synchronic differences of policy styles reviewed in chapter 4: i.e., policy imperative, mode of competition, and ownership and pricing structure. In the U.S., the emphasis on policy imperative was on the market-led industrial strategies to stimulate private sector investment based on American strong belief in competition. In Finland, the emphasis was on overcoming a national economic depression, which caused the focus of national strategy on the business potentials and technology development. In Korea, with the response to the external forces for the opening of the telecommunication market and the economic depression, competition in local telephone markets was introduced. With relation to the focus of competition in local telephone market, in the U.S., the comparative focus was on the retail-based competition (i.e., resale and unbundling). In Finland, the comparative focus was changed from the more facilities-based competition to the more resale-based competition. In Korea, the comparative focus was on the more facility-based competition complemented by introducing the resale competition. With relation to the ownership and pricing structure, in the U.S., local telephone service was provided with a flat rate regime for local call provided privately owned monopolistic companies. In Finland, the local telephone services have been provided 205

PAGE 220

by the so-called "private/public mix" of telecommunications companies with a flat rate regime for local calls tied with the local customers. In Korea, the local telephone services have been provided by the state-owned monopolistic Korea Telecom with low prices approved by the Ministry oflnformation and Communication as part of public service. 206

PAGE 221

CHAPTER 7 SOCIAL CONTROL AND INTERNET DIFFUSION In this chapter, as the third chapter of the diachronic analysis of the national policy patterns for the mature use of the Internet, the diachronic relationship between social control ofthe communications market and the third stage (or control of proliferation) oflnternet diffusion over time is presented. For this purpose, the universal service policy as a means for narrowing the digital divide in the three nations is compared and contrasted based on Easton's inputoutput model and Dye's policy model of causes and consequences; i.e., first, the social imperatives from the service-user axis in a triad oflnternet policy as a policy input, second, the functional ministries as a policy system, third, the re-distributive policy for the universal access to the communications networks as a specific policy output, and fourth, the diachronic relationship between the de-regulative policy and the control of Internet proliferation as a policy impact on Internet diffusion are assessed in view of the theoretical pattern of Internet policy. This analysis is intended to produce both the characteristics of policy patterns and the policy implications on the government role for the control of Internet proliferation. 207

PAGE 222

Social Imperative: Imperative to Re-distribute The policy rationale for governmental intervention or the motivation to intervene addresses whether and where lines shall be drawn between public and private responsibilities. In the third stage of Internet diffusion, the policy rationale for control of proliferation of Internet diffusion can be expressed into the social imperative to redistribute. Theoretical Policy Rationale: Social Re-distribution According to Noam's diffusion model of interactive technology, through the third stage of the interactive medium (n2-n4 ) -i.e., above an economic optimum point to an exit point-the socially optimal size (n3 ) will not be reached by itself, but by some external governmental direction through required expansion. This is because the market does not take into account the network externalities (such as enhancing social solidarity). From a social point of view, however, social welfare still increases at an economic optimum point (n2). So, a socially optimalized point is located on a certain point between an economic optimum point and an exit point. In the Internet arena, closing or narrowing the digital divide of service-users through universal service policy in the third stage of Internet diffusion can be justified in that the "social optimum" is hard to be achieved by the market itself via the invisible function of the price due to both the technological traits ofthe interactive Internet (as 208

PAGE 223

reviewed in chapter 2) and an inherently-distorted pricing mechanism in the local telecommunications market by the non-market considerations in the name of universal service (as reviewed in chapter 6). Consequently, after marketization issues, such as narrowing the digital divide, are hypothesized to result in social control of Internet diffusion. As such, the social imperatives from the community, especially from the service-user axis oflnternet policy, are the comparative characteristics in the third stage of Internet diffusion. Individual Policy Rationale In the third stage of Internet diffusion, the social imperatives from the service user axis of a triad of Internet policy as a theoretical imperative to re-distribute can be compared and contrasted with the individual imperatives to re-distribute within the three nations. Such the individual imperatives to re-distribute in communications market can be symbolized by the so-called "umbrella term" of universal service, including a wide range of policies responding to the economic, social, and political needs (Mueller, 1997). United States. In the U.S., the concept of universal service is closely related to both the origin of provider-centered system integration and the comparatively narrow economic solution ofthe natural monopoly of the telecommunications market. 209

PAGE 224

The origins of universal service in the U.S. can be tracked back to the development of the telephone network at the beginning of the 20th Century. The tenn finds its genesis in the strategy of the Bell Telephone Company (the forerunner of AT&T) to forestall competitors by pursuing a policy of "system integration" (Dordick, 1990). This strategy consisted of initially interconnecting competing local exchangers and then absorbing them into the Bell "grand system" (Verhoest, 2000). In this vein, universal service simply meant national interconnection for the service providers.88 Thus, universal implied in its origin "everywhere," rather than everyone (Dordick, 1990; Verhoest, 2000). Furthermore, as Verhoest (2000: 596) argued, the connotation of universal as "a telephone for everyone" was only added for political reasons: the AT&T credo "One System, One Policy, Universal Service"89 served to justify a national telephone monopoly to the Congress. Such a concept of universal service was loosely articulated in Section 151 of the Communications Act of 1934, "to make available ... to all the people of the United States a rapid, efficient, nation-wide, and world-wide wire and radio communication service with adequate facilities at reasonable charges." This obligation for the universality gave a statutory background for the narrow economic solution of the socalled "cross-subsidization" and "cost averaging" that contributed to universally expanding the service under the monopoly regime: in other words, the long-distance 88 In this regard, Mueller argued that AT&T did not aim to extend the service to all (Mueller, 1997). 89 This slogan was ingeniously invented by Theodore Vail, the architect of the American Telephone and Telegraph (AT&T) system or the Bell system in I 907 (Dordick, I 990). 210

PAGE 225

callers had to pay more in terms of the charges meant to subsidize the local callers by the mechanism of interconnection (i.e., reimbursements or access charges); urban local callers, who were supposed to pay less because of the low costs for installing the network in highly populated urban areas, had to pay more in terms of the charges meant to subsidize rural local callers, who were supposed to pay more due to the high costs for installing the network in scarcely populated rural areas under the homogeneous local pricing mechanism (i.e., indirect subsidization by cost averaging) called universal service. Thus, universal service in the U.S. was inherently centered on the service. provider in that the costs and revenues of service-providers are subsidized for universality. Furthermore, universal service in the U.S. was centered on the economic solution in that service users' rights to access communications resources was intended to be re-distributed by both the provider-centered pricing mechanism based on the classic economic efficiency and the notion of economic welfare (Verhoest, 2000). According to Verhoest (2000: 599), the definition of universal service is based on three pillars"affordability" (at an affordable price), "accessibility" (available to all users independent of their geographic location), and "quality of service." Furthermore, the service criteria of universal service describes an output that roughly coincides with the classical definition of economic efficiency: i.e., first, products or services would be offered at the lowest possible cost to reflect optimal production costs (affordability); second, service would be optimally distributed among the members of 211

PAGE 226

society given their disposable income (accessibility); and, third, there would be an optimal level of innovation (quality). Such basic concepts ofuniversal service were well expressed in the national infonnation infrastructure plan, the Agenda for action (IITF, 1993). Identifying nine specific principles to guide governmental action, it presented the need for promoting private sector investment to extend the universal service concept to guarantee that infonnation resources were "available to all" "at affordable prices." As evidence, the basic concept ofuniversal service was codified into Section 254 ofthe Telecommunication Act of 1996 and the FCC Order (97-157) for universal service. Finland. In Finland, the concept of universal service is closely related to the broader social solution of the natural monopolistic nature of the telecommunications market. The concept ofuniversal service in Finland can be described in tenns of social service in that service users' rights to access communications resources are re distributed by user-centered direct subsidization based on the notion of general welfare. The historical development oftelecommunications systems in Finland gives an insight toward the origin of the social service. After Alexander Graham Bell developed a practical telephone in 1876, Finland's first telephone connection was built a year later (Jussila et al., 1999; Steinbock, 2001). The telecommunications industry in Finland originated from the pre-independence era, or the Grand Duchy of Russia, when the industry was organized by Russian rather than Finnish authorities. The Finnish Senate, 212

PAGE 227

arguing that Finnish telephone activities must be left to the private sector, issued the "Gracious Declaration" (i.e., a new telephone statute) in 1886 to avoid Russian intervention (Steinbock, 1998; Turpeinen, 1997). Compared to telegraph, which was "highly centralized" and "born international" at that times, the telephone was far more decentralized and "born local." The Russian authorities could allow certain autonomy in local telephone communications, but the telegraph was a matter of national security (Verhoest, 2001; Steinbock, 2001). After Finland declared its independence in 1917, local telecommunications was left to private firms, resulting in more than 800 local telephone companies in 19341940, and causing conflicts over tariffs and interconnections among the local companies. The Finnish way to solve the natural monopolistic characteristic of the telephone market was the so-called ''third way" of a private/public mix of telecommunications (Turpeinen, 1997; Steinbock, 1998, 200 I). Between 1920 and 1949, the PTT, state-owned telephone company acquired the equipment of some 170 telephone companies, primarily in eastern and northern rural areas of Finland. Thus, the paradoxical logic of system integration caused by the natural monopolistic nature of the telecommunications market as in the U.S. case was resolved by a private/public mix of telecommunication in Finland: (1) the state-owned PIT monopoly of both long-distance telephone at the national level and local telephone in rural eastern and northern areas, and (2) the users-owned local cooperatives' (i.e., the Finnet Group) monopoly of urban southern areas. 213

PAGE 228

Given the large landmass of Finland and the low population density in the northern areas of Finland, homogeneous pricing of local access can place an unduly high burden on urban customers. In this respect, the Telecommunications Act of Finland states that operators are not required to serve all customers at the same price not related to costs.90 Thus, there are no cross-subsidies between local telephone companies and long-distance telephone companies, and no cost-averaging for indirectly subsidizing from urban callers to rural callers in Finland. Based on general welfare, the Finnish government's method to ensure universal access to telecommunications services is through direct assistance to the customers in question by a social welfare authority, or Ministry of Social Affairs and Health. This approach entails using tax revenues to directly support bill payments of citizens meeting support eligibility requirements, based on the recipients' income. The basic concepts of universal access to telecommunication services was well expressed in the revised national strategy document, Quality of Life, Knowledge and Competitiveness, and has been codified into the Act on Social Assistance and Decree in 1998. Korea. In Korea, the concept of universal service is closely related with the public or governmental solution of the natural monopolistic nature of the 90 This implies that only as long as costs for providing services to two customers are identical, prices have to be equal as well. If on the other hand fundamentally different costs are incurred (i.e., local calls in downtown Helsinki or in the northernmost region of Finland) then prices are to reflect these differences in costs (Nattermann and Murphy, 1998). 214

PAGE 229

telecommunications market. As in most European countries, the telecommunications service in Korea was provided by state-owned monopoly service providers, Korea Telecom. Furthermore, Korea's historical context reinforced the governmental provision of the telecommunications service. After telephone service started in 1886 in the Yi Dynasty (MIC, 1992), the first automatic telephone system was installed in 1935 under the Japanese colonial government (Larson, 1995). During the Korean War (1950 to 1953), 80 percent of Korea's existing telephone systems were destroyed. After the war, the first major and systematic investment in telecommunications came with the first of four consecutive five-year economic development plans, beginning in 1962. So, for most of the 1970s, telecommunications services were provided by the government. In 1982, the state owned Korea Telecommunications Authority (KTA, renamed in 1990 as Korea Telecom) was created as a corporate entity, and started to introduce competition in the telecommunications market in the 1990s. As Gamham (OECD, 1991) pointed out, there was no urge to elaborate explicit service obligations under the former public monopoly, since the state, by its nature, could be expected to act in the public interest. Hence, the concept of universal service in Korea has only recently come in to use in the 1990s, roughly coinciding with the introduction of competition in the telecommunications market. In short, the concept of universal service for public access to the telecommunications service in Korea inherently originated from the so-called public 215

PAGE 230

services provided by the government, and the tariff on local calls was regulated under the name of public services. The public telecommunications operator (PTO), or KT, had traditionally been compensated through cross-subsidization from long-distance and international services before 1995 (MIC, 1999). Summary Essentially, the evolution and diversity ofthe universal service concept originated from the updating and rationalizing of cross-subsidies in the old monopolistic era to make them compatible with competition. In the U.S., universal service originated from the needs of provider-centered system integration is inherently centered on the service provider in that the costs and revenues of service providers are subsidized for universality. Furthermore, universal service in the U.S. centered on the economic solution in that service users' rights to access communications resources were intended to be re-distributed indirectly by the provider-centered pricing mechanism based on classic economic efficiency and re distribution of economic welfare. In Finland, the concept of universal service is closely related to the broader concept of social solution to the natural monopolistic characteristic of the telecommunications market. Thus, the concept of universal service in Finland can be described as the concept of the social service because service users' rights to access 216

PAGE 231

communications resources are re-distributed by both user-centered direct subsidization and the notion of general welfare. In Korea, the concept of universal service for public access to the telecommunications service was originated from public services provided by the government. Furthermore, the tariff on local calls was regulated under the name of public services. The public telecommunications operator (KT) had traditionally been compensated through cross-subsidization from long-distance and international services. Policy Programs and Agencies In this section, the comparative characteristics ofthe specific policy instruments for universal access to the communications network (i.e., universal service program and its agency to implement) in the three nations are briefly reviewed. For this purpose, we examine the respective characteristics of ( 1) the financial incentive mechanisms (policy programs and agencies) in the U.S. to promote access to the communications network, (2) the social assistance mechanisms (policy programs and agencies) in Finland to promote access to the communications network, and (3) the public service mechanisms (policy programs and agencies) in Korea to promote access to the communications network. 217

PAGE 232

Theoretical Policy Programs and Agencies In the third stage of control of proliferation, the policy programs for the enhancing of public access to the communications network that can be classified into a re-distributive mode in view ofLowi's policy typology can affect Internet diffusion. The social goal of increasing social equity of service-users can be achieved by the policy instruments of the subsidies (or, re-distributive policy) for universal access to the communications network. As such, the modes of re-distribution in the service-users axis of a triad of Internet policy in a nation can be the comparative characteristics in the third stage oflnternet diffusion. Furthermore, a national ministry is theoretically involved in this stage in that the re-distributive policy for universal access to the communications network is considered a policy function. Individual Policy Programs and Agencies In the third stage of Internet diffusion, the three different specific policy programs and policy agencies for universal access to the communications network can be compared in detail with each other and with the theoretical policy programs and agencies. United States. Financial Incentives to Promote Access. The codification of universal service in the U.S. was indicated in Section 254 ofthe Telecommunication Act of 1996 (hereafter, 218

PAGE 233

1996 Act) wherein the FCC identified universal service as one ofthree main goals. A major component of the 1996 Act broadened the scope of existing universal service policy to include new telecommunications services and to target new groups of subscribers, such as schools, libraries, and health care providers (Prieger, 1998; OECD, 1998c). The 1996 Act articulated several principles to guide the creation ofthe new universal service policies in Section 254 (b): I. Quality and Rates: Quality services should be available at just, reasonable, and affordable rates. 2. Access to Advanced Services: Access to advanced telecommunications and information services should be provided in all regions of the Nation. 3. Access in Rural, and High Cost Areas: Consumers in all regions of the Nation, including low-income consumers and those in rural, insular, and high cost areas, should have access to telecommunications and information services, including inter-exchange services and advanced telecommunications and information services, that are reasonably comparable to those services provided in urban areas and that are available at rates that are reasonably comparable to rates charged for similar services in urban area. 4. Equitable and Non-discriminatory Contributions: All providers of telecommunications services should make an equitable and non-discriminatory contribution to the preservation and advancement of universal service. 219

PAGE 234

5. Specific and Predictable Support Mechanism: There should be specific, predictable and sufficient Federal and State mechanisms to preserve and advance universal service. 6. Access to Advanced Telecommunications Services for Schools, Health Care, and Libraries: Elementary and secondary schools and classrooms, health care providers, and libraries should have access to advanced telecommunications services .... These principles are the foundation upon which the rest of the 1996 Act provisions concerning universal service are built. The 1996 Act mandates specific support for three groups: subscribers living in rural and high-cost areas; low-income subscribers; and educational and rural medical institutions. The specific support policies for these three groups and their funding mechanisms (New Universal Service Fund) are the main areas ofthe 1996 reforms to universal service in the U.S (Prieger, 1998; OECD, 1998c; Cherry and Wildman, 1999), which are summarized below: First, the traditional programs to subsidize service to rural, insular, and high cost areas were continued with telecommunications operators providing universal service91 being able to draw compensating support from a Universal Service Fund. In this vein, restructuring of the "Subscriber Line Charge" and the "Common Carrier Line Charge" were recommended to partially transfer Universal Service Fund support costs 91 Under the new plan adopted by the FCC, the list of supported services expands to comprise ( l) single party service, (2) dual tone multifrequency signalling or digital equipment, (3) voice grade access to the PSTN, (4) access to emergency services (911), (5) directory services, (6) operator services, and (7) interexchange services (FCC Recommended Decision, 46-52, 65-67). 220

PAGE 235

to subscribers and interexchange carriers (OECD, 1998c). Also, the 1996 Act does not reform "rate averaging" practices but codifies them into the 1996 Act 254 (b) (3). Second, support systems for low-income subscribers like "Lifeline" and "LinkUp America"92designed to subsidize hook-up cost and the cost of monthly telephone bills-were revised and extended. For this purpose, it was recommended that the federal Lifeline subsidy increase from $3.50 to $5.25 per month, with additional matching of state funds to $7, and that both programs funded by the new Universal Service Fund be extended to the whole nation. Third, discounts to assist schools, libraries, and rural health care centers to connect to the so-called Information Superhighway were initiated. These discounts became colloquially known as the "E-rate" and were designed to cut between 20 and 90 percent off the monthly charges of connecting to the networks, and in some cases, some of the internal wiring costs. Fourth, the more transparent and explicit support mechanism for universal service, or new "Universal Service Fund (USF)," was introduced. Based on the principle in section 254(b) (4) of the 1996 Act, the contributors to the USF are carriers providing interstate telecommunications services (section 254(d)). However, in its Report and Order, the FCC clarified that some entities providing information service, such as the Internet and enhanced service providers, are not carriers and are not 92 Since 1984 low-income subscribers have enjoyed programs subsidizing access to the network. Lifeline Assistance reduces the monthly telephone bill of program participants by $7 in most states, and its companion program Link Up America subsidizes initial connection charges. These programs are funded by interexchange carriers (Prieger, 1998). 221

PAGE 236

required to make contributions.93 Hence, the ISPs are exempt from universal service contribution. Policy Agencies. The traditional dual jurisdictional nature of telecommunications regulation is still effective in the universal service policy area. The FCC has jurisdiction over the provision of interstate services and the states retain jurisdiction over the provision of intrastate services. For the purpose of the uniform application of universal service mechanism to both federal and state, a coordination mechanism between them is needed. For this purpose, the Federal-State Joint Board on Universal Service (Joint Board) was created under section 254 of the 1996 Act, and the FCC implements the recommendation of this Joint Board. The Joint Board issued its recommendations regarding the financing and management of a federal universal service support fund in November 1996 (Recommended Decision, 1996). The FCC established rules (Universal Service Rules, 1997) responsive to these recommendations in its Report and Order issued in May 1997. The support mechanisms necessary to achieve the universal service objective are administered by the Universal Service Administrative Company (USAC), an independent not-for-profit organization formed 93 In addition, the FCC also defined that (1) contributions be levied on end-user revenues as to the type of revenue base, (2) contributions are to be levied on both the interstate and intrastate revenues of interstate carriers to support educational institutions and health care providers, and (3) contributions are to be levied only on the interstate revenues of interstate carriers to support low-income and high-cost customers (Cherry and Wildman, 1999). 222

PAGE 237

in September 1997 under regulation of the Federal Communications Commission (FCC).94 Finland. Social Assistance to Promote Access. The method employed in Finland to ensure universal access to communications network is direct social assistance (or income support) to the customers in question from the Ministry of Social Affairs and Health (Nattermann and Murghy, 1998). The direct income support is financial assistance provided by social welfare as the last resort to secure the person's or the family's access to telephone service. The content and amount of income support, and other grounds for granting income support are prescribed in a separate "Act on Social Assistance" and Decree. In 1997, income support paid to 344,700 households having a total of 435,000 persons (11.5% of the population). In that year, households received income support for an average duration of 5.6 months. Gross expenditure for income support in 1997 totaled 555 million euro (Ministry of Social Affairs and Health, 2001 ). Income support is made up of a basic benefit and the amount of other expenses considered necessary (supplementary benefit). The basic benefit utilize tax revenues to cover the costs of food, clothing, minor health care costs, personal hygiene, the cleanliness of the home, use of local public transport, a newspaper subscription, a TV license, use of a telephone, and recreation and the pursuit of hobbies. 94 For more information on USAC, see http://www.universalservice.org. 223

PAGE 238

Policy Agencies. The Finnish method to ensure universal access to the telecommunications network is direct assistance to the customers in question from a social welfare authority, or the Ministry of Social Affairs and Health. This approach entails using tax revenues to directly support bill payments of citizens meeting support eligibility requirements, based on the recipient's income. Korea. In Korea, there are two separate policy responses to promote public access to the communications network: that is, (I) establishing more explicit cost allocation framework for implementation of universal service in a competitive environment, and (2) establishing more direct funding to promote public access to the network which will review in next section (i.e., policy trends toward closing the digital divide). Universal Service. No specific services had been clearly designated as universal services in Korea to date. But some services have been practically provided by KT for public interest and social welfare. Before 1995, since there was no funding mechanism to cover the deficit from providing public services, KT, as a public telecommunications operator (PTO), has been compensated traditionally through cross-subsidization from long-distance and international services to the deficit areas of local voice telephony service, public pay phone service, directory assistance service, and telegraph service. Between 1996 and 224

PAGE 239

1997, the NonTraffic Sensitive {NTS) access deficit contribution system was introduced to finance deficits incurred by subscriber line service, directory assistance service, administrative communications service, and maritime mobile telephone. The access deficits from these services were recouped by facilities-based service providers in proportion to the volume of their interconnected traffic. Between 1998 and 1999, the financing was differentiated by two scopes of services: i.e., (1) subscriber line service was financed by access charges on the basis of traffic volume, (2) policy-oriented services, including administrative communication service and maritime mobile, was financed by the contribution in proportion to the respective totals. At this time, a plausible definition of universal service in Korea is specified in Article 2 of the Telecommunications Business Act. This article gives a legal basis for the introduction and further development of universal service in Korea. Policy Agencies. Section 38 ofthe Korean Government Organization Act states that the Minister oflnformation and Communication (MIC) is responsible for the matters regarding telecommunications service and the initiatives on high-speed information infrastructure (KII).95 With relation to telecommunications service, the MIC has a full responsibility to (1) set up policy and framework on market entry and ownership regulation, (2) establish competition safeguards, consumer protection, and 95 MIC is also in charge of(l) R&D for information and communications technology, (2) training and education of professional workforce, (3) protection of intellectual property rights of computer programs, (4) spectrum management and broadcasting technology, and (5) postal service (MIC, 2001). 225

PAGE 240

interconnection guidelines in the marketplace, and (3) manage scarce national telecommunications resources (such as numbering and telecommunications networks). By extension, the MIC is responsible for the policy on universal service in Korea. Furthermore, MIC implements a host of policies in relation to the nation-wide infonnatization plan: i.e., (1) the Korea Information Infrastructure (KII) initiative and other supporting projects, (2) facilitation of the use of the Internet, (3) streamlining rules and regulations on electronic commerce, electronic signature and privacy protection. Summary Generally speaking, the re-distributive policy programs and policy agencies for universal access to the communications network in the three nations reflect their basic national ethos reviewed in chapter 4 and their historical development of the telecommunications market reviewed in chapters 5 and 6. More specifically, in the U.S., the concept of universal service to promote access to the communications network focused on the financial incentives with more rationalized funding mechanisms for cross-subsidization and rate-averaging under the tight supervision of the FCC and the state Public Utilities Commissions (PUCs). This system is based upon both the contextual background of American trust on the economic market (or, the economic welfare) and provider-centered historical development ofthe communications network. 226

PAGE 241

In Finland, the social service to promote access to the communication network focused on social assistance with direct income subsidization from a social welfare authority, or the Ministry of Social Affairs and Health. This system is based upon both the contextual background of Finnish welfare society ethos (or, social welfare) and user-centered historical development of the communications network. In Korea, the public service to promote access to the communication network focused on cross-subsidization and rate-averaging with tight regulation of the MIC. This system is founded on both the contextual background of Korean high stateness ethos and network-centered historical development of the communications network. Policy Impact: Social Control and Internet Diffusion In this section, we examine the diachronic relationship of universal service to promote public access to the communications network and the third stage of Internet diffusion. For this purpose, we discuss first, the linkage between universal service and the Internet, second, the relevance of policy programs to resolving the digital divide as an effect of universal service to the Internet diffusion, third, the time matching comparison between social control and the third stage of Internet diffusion, and fourth, the government role in the third stage of Internet diffusion. 227

PAGE 242

Universal Service and Internet: Solving the Digital Divide Traditionally, the concept of universal service focused mainly on universal access to the telecommunications network. This "old" concept of universal service can affect Internet diffusion indirectly in that the Internet in simplest and earliest form is just a computer application over the telecommunication network. However, as the infrastructure of telecommunications are proliferating in number96 and the alternative networks to connect the Internet via a cable modem or DSL are emerging at an unprecedented rate, there are several arguments concering the expansion of the traditional notion of universal service to newer or advanced communications services like the Internet (Mueller, 1997; Noam, 1994; Compaine and Weinraub, 1997). In this vein, after reviewing the arguments on the "new" or "expanding" universal service to the Internet, the policy trends toward social control of Internet diffusion by expanding the concept of universal service and narrowing the digital divide in the three nations are examined as they directly affect Internet diffusion. The arguments on expanding the "old concept" of universal service, which arose from the American historical context, are closely related to the U.S. Constitution. If the old universal service is expanded beyond Plain Old Telephone Service (POTS), 96 For example, ITU (200Ia) data shows that in 2000 the main telephone lines per 100 inhabitants are 69.97% (U.S.), 55.02% (Finland), 46.37% (Korea) and cellular subscribers per 100 inhabitants are 72.04% (Finland), 56.69% (Korea), 39.79% (U.S.). Thus, the development of the telecommunication network of the U.S. focuses more on the wired networks, while those of Finland and Korea focus more on the wireless networks. 228

PAGE 243

ultimately the central issue of the expanded universal service is access to what-the communications or the information content? (Compaine and Weinraub, 1997). Hence, if the new universal service means expanded communication service analogous to ''the post," Article 1, Section 8 ofthe U.S. Constitution specifically gives Congress the authority to establish and fund a postal system. On the contrary, if the new universal service means expanded information service analogous to ''the press," the First Amendment prohibits Congress from infringing in the content business. Under the situation of the technological convergence among the domains of communications and information on the World Wide Web, there are arguments both for and against expanding the traditional concept of universal service. The arguments in favor of expanding the concept of universal service include: (1) There are strong needs to narrow a gap between information haves and have-nots in an information society (Dordick, 1981; NTIA, 1999), what many have called the "digital divide," or the effect on the public welfare. (2) The access to advanced computer and information services are important in educational terms to develop the so-called "New Literacy"97 (Compaine, 1988; Papert, 1993). (3) Based on the so-called "Free Net" concept (much like the free public library system) where anyone who was interested could get controlled access to the 97 Compaine (1988: 155) posited that the New Literacy is the ability to use and work with the new computer and information systems while the Old Literacy reflected the technologies of printing and mechanization. 229

PAGE 244

Internet, a nationwide system of public access computer systems could be established (Cisler, 1984 ). (4) Among the other online services, universal access to email is needed in that email provides new opportunities for information, supports social integration of groups, provides wider resources, can facilitate citizen participation in the political process, is a catalyst to network use, and even supports global democratization (Anderson et al., 1995). On the other hand, based on the so-called "go slow" position, those opposed argue against expanding the concept of universal service: (1) With relation to the logic of "economic concerns" about the questions of who will pay and how much, if one group (presumably the economically disadvantaged) is going to get network service at or below cost, others have to pay more than their share. Such adjustments with rates is impossible without an AT&T -like monopoly (Browning, 1994; Crandall and Waverman, 2000). (2) With relation to technological "bad-timing," the technology may not yet be in the best form for universal distribution. Furthermore, there is danger that jumping in too fast can lock in a technology that soon would be superseded by a better one, e.g., France's Minitel (Compaine, 1988). (3) With relation to the logic of"harm to network," the network should accommodate every potential user, but it would require massive resources to attempt to make the 230

PAGE 245

networks accessible to all and furthermore a waste of funds on a project can never be solved (Meyers, 1994; Metcalfe, 1995). (4) With relation to the logic of"setting priorities," government can not subsidize every issue, possibly causing social equity concerns (Schrage, 1994; Browning, 1994; Crandall and Waverman, 2000).98 (5) A final argument is that "it is simply not necessary in today's society," and that "the real problem is that most people don't yet know that they want to be wired" (Sandford and Frissell, I 995).99 Among the various arguments on expanding the concept of universal service, this dissertation is based on the logic of supporting groups in that (1) the Internet is the most advanced or worldwide interactive medium causing multi-dimensional network externalities, (2) the inherent nature of Internet service is classified into the the commons, i.e., a quasi-public (CPRs) or public goods, (3) the logic of the so-called "go slow" cannot solve the distorted practices ofrate-averaging100 in local telephone market, (4) the Internet service is considered a communications service, not an information service, and (5) from the perspective of network design, the 98 Interestingly enough, Schrage (1994) argued that "the lack of subsidy for radios, televisions, and VCRs has not created a communications rich and poor, and the real problem isn't access; it's some pundits insistence that issues of social equity and economic opportunity are better shaped by investing in technology than in people; That's a bad idea and it leads to bad policy." 99 Available at http://www.wired.com/wired/archive/3.05/access.if.html. 100 The homogeneous pricing of local access caused by rate averaging would have placed an undue burden on urban customers for subsidizing the rural customers who can be a "better-off' group. 231

PAGE 246

technology for the dial-up connection to the Internet is moving toward its maturity in the technologically-advanced OECD nations. Policy Trends toward Social Control: New Universal Service and Digital Divide101 Considering the differences in the concept of the "old" universal service based on the prevailing ethos in the three nations, the policy trends toward the "new" universal service varies from one nation to another. United States. The new universal service concept applied to advanced computer networks-the concern of the digital divide that the cleavage between information haves and haves-nots could be widening-can be more important in the American context than in the other two nations' context in that the Internet was born in the U.S. and the U.S. has played a leading role in Internet diffusion (Schneider, 1997). The concept of new universal service was articulated in early 1995 at the G7 Information Summit by Vice President AI Gore. Gore said that "the Clinton Administration was committed to the goal of connecting every classroom, every library, every hospital and every clinic to the national and global information infrastructure" 101 Inagaki and Mahajan (1999: 9) posit that the idea of universal service refers to the access to physical infrastructures such as telephones and computers, while the concept of digital divide addresses underlying socio-economic inequalities that are manifested in information gaps. So, they argue that the digital divide issues (information gap) are not reducible to the lack of universal service (communication gap). Nevertheless, an access to the web (communication gap) can be a prerequisite to narrowing the digital divide (information gap). 232

PAGE 247

(Gore, 1995). This commitment was later articulated in the Telecommunication Act of 1996. In addition, the National Telecommunications and Information Administration (NTIA) conducted a series of field hearings on universal access across the U.S. in 1994. Its report called for an increase in the goal beyond basic analog voice grade service. Furthermore, from 1995, NTIA has measured the trends toward narrowing the digital divide suggesting the action agenda to implement this reduction with a series of reports, Falling through the Net. According to the NTIA report (July, I 999), the digital divide in the U.S. occurred from the criteria of race or ethnicity, income, education, and geography. While the white household, the household of the Asian/Pacific Islander, high-income Americans, and Americans with college degree are classified into the "information haves," the African-American household, Hispanic household, low income Americans, and Americans with an elementary school education are classified into the "information have-nots." Recently, the NTIA has transferred TIIAP (Telecommunication and Information Infrastructure Assistance Program) to TOP (Technology Opportunity Program) in 2000 to increase financial support allocated for the programs of bridging the digital divide, and has operated the digital divide portal site102 to provide relevant and valuable information to various sectors interested in digital divide issues. 102 Available at http://www.digitaldividenetwork.org. 233

PAGE 248

Finland. In Finland, social service to promote access to the telephone network focused on social assistance with direct income subsidization from the social welfare authorities. These subsidies are based upon the contextual background of its welfare society concepts (or social welfare) and user-centered historical development of the communications network. Thus, the American concept of digital divide that emerged from the intersection between the social system and communications technologies via a rate-averaging in the local telephone market is not as pronounced in the Finnish context. The concept of digital divide is expressed by "alienation" in the information society in the Finnish context, and the "Information Society Advisory Board" monitors the eventual alienation development based on the annual series of statistics published by Statistics Finland, On the Road to the Finnish Information Society (Statistics Finland, 1997, 1999, 2001). Furthermore, the broader concept of socio-economic inequalities in the information society was articulated in a revised national strategy document, issued in December 1998 by the Finnish National Fund for Research and Development (SITRA), Quality of Life, Knowledge and Competitiveness (SITRA, 1998). This criticism has been reflected in permitting a wider spectrum of issues related to the development of an information society in Finland. The new strategy took its point of departure that people were not only employees but also citizens and consumers (Henten and Kristensen, 2000). 234

PAGE 249

The goals of the Finnish information society initiative are (SITRA, 1998): 1. To ensure welfare, employment and income levels for citizens. 2. To ensure equal opportunities for the acquisition and management of information and for the development of knowledge. 3. To ensure conditions for entrepreneurship, quality of working life, and competitiveness. 4. To provide opportunities for human interaction and cooperation. 5. To strengthen democracy and provide opportunities for social influence. 6. To improve security, individual data protection and the citizen's status as a consumer. 7. To develop services, cultural provision and to increase international interaction. 8. To ensure Finland's attractiveness as a location for innovative enterprises. 9. To promote equality between regions. I 0. To support the objectives of sustainable development. In this period, numerous agencies were established to promote wide public discussion on the national information society strategy. In 1996, the Finnish government created a ''National Committee for Information Society Issues" (or "National Council for Information Society") to represent the private and the public sector as well as research. The chairman of the committee was the Minister for 235

PAGE 250

Administration, and the Ministry of Finance provided the secretariat.103 In the same year, the Ministry of Finance established a broad "Information Society Forum" comprising around 55 experts representing different viewpoints on the information society to assist the National Council for the Information Society. Based on wide discussion started in 1997 on the economic and technological focus of the initial national strategy, a steering group supported by the National Council for the Information Society, the Council of State, and the Finnish National Fund for Research and Development (SITRA) was formed and implemented the revision process (Henten and Kristensen, 2000). This group was chaired by the chairman of SITRA and vice-chaired by the Minister of Finance. Members of the group were representatives of both the public and private sectors. Korea. The Korean government has recently made great strides in bridging the digital divide. For example, the "Informatization Strategy Development Meeting," presided over by the President, was held in 2000 to make comprehensive measures to reduce the digital divide. In this meeting, various acting plans-such as (1) building of high-speed information networks in rural areas, (2) promotion of information havenots' ability to use IT, (3) support of the development oftechnology for accessibility of the disabled, and (4) development of valuable contents for the information poor-were announced. 103 For more information about the function of the national council for information society, see Chatrie and Wraight, (2000:83). 236

PAGE 251

The Korean Parliament also passed "Act on closing Digital Divide" (enacted in January I 6, 200 I) to support the government's efforts in solving digital divide. Under this Act, a Committee for Closing the Digital Divide and a 5-year master plan and annual implementation plans to bridge the digital divide have to made. Other major measures included in this Act are: (1) setting up telecommunications service accessibility guideline for the disabled, subsidizing the poor and the disabled for acquisition of PC or other telecommunications equipment, (2) developing access helping technologies for the disabled, (3) supporting the information service providers (ISPs) that provide valuable contents for the fishermen/farmers, disabled, elderly, and poor, (4) establishing public access centers that offers Internet access and learning opportunity to the residents in need. Thus, Korean way of controlling the proliferation oflnternet diffusion is closely related to the direct funding of public access to the Internet. Universal Service and Control of Proliferation Policy response toward the so-called "social control" of Internet diffusion for securing users' equity reflects the social recognition of underlying socio-economic inequalities in society as a whole. The three nations, based on the different driving motives or ethos, show different policy responses toward the market failure caused by the monopolistic characteristic of the telecommunications market. 237

PAGE 252

' It is premature to claim that Internet diffusion has already reached beyond the point ofthe "private optimum" in that the Internet was recently commercialized to the public in 1995, and no nation among the three nations shows over 50% of the Internet users per 100 inhabitants. Nevertheless, policy trends toward social control of Internet proliferation can be detectable in the three nations, which are suggesting that the three nations are reaching toward the third stage of Internet diffusion. In the U.S., universal service was codified into the Telecommunication Act of 1996, nominally guaranteeing the introduction of competition in the local telephone market. Given the strong support in the market function and historical origins of cross subsidization and rate-averaging for regulating the privately owned monopolistic company, policy program for liberalization (competition) needs to be tightly coupled together with policy program for social control (universal service) via the local flat-rate pricing structure. Furthermore, the recent policy trends toward narrowing the digital divide show its reaching toward the third stage of Internet diffusion. In Finland, social service was codified into the "Act on Social Assistance" and Decree in 1998. Broader concepts of socio-economic inequalities in an information society were articulated in a revised national strategy document, Quality of Life, Knowledge and Competitiveness in 1998. These two major policy trends toward social control of Internet diffusion indicate its movement toward the third stage of Internet diffusion. 238

PAGE 253

In Korea, a more refined universal service than current policy for accessing the telecommunications network is under consideration, but the policy trends toward solving the more broadly defined digital divide combining the two American concepts of communication gaps and information gaps was codified into the "Act on closing Digital Divide" in 2001. This policy trend toward social control ofinternet diffusion likewise reflects Korea's moving toward the third stage of Internet diffusion. Even though the three nations show different degrees of social control of Internet diffusion -i. e., the different recognition and re-distributive mechanisms of universal service and digital divide -the role ofthe government in this period can be theoretically proposed as a fair promoter of service users' social welfare with aredistributive policy among the triad oflnternet policy. Thus, the social dimension of the 3-D synchronic model can be roughly matched with the third stage ofthe 3-S diachronic model ofthis dissertation. Summary and Implications To address the second research question and to test theoretical proposition (P2c) regarding the diachronic characteristics of policy pattern in the third stage, policy factors affecting the control of proliferation of Internet diffusion in the three nations are compared and contrasted using the 3-S diachronic model. This comparison is intended to produce the similarities and differences of policy pattern in the third stage of Internet diffusion. 239

PAGE 254

Chronologically, different time zones (beyond social control) in the three nations can be identified to fall into the second stage oflnternet diffusion: i.e., beyond 1996 in the U.S., beyond March 1998 in Finland, and beyond January 2001 in Korea. During these periods in each nation, comparative focus of communications policy for the development and dissemination of the Internet was on enhancing equity of service users or service-user variable among the three variables in atriad of Internet policy. The social control of Internet diffusion in the third stage was characterized by the policy tools of re-distribution, such as universal service policy. Thus, the comparative similarities of policy pattern in the third stage can be found in its focus on the service-user variable (or social dimension) in the 3-D model during the third stage in the 3-S model. These comparative similarities can produce a theoretical suggestion of governmental role as a fine tuner for enhancing equity of service-providers. Such a governmental involvement for enhancing service user's equity in this stage can be justified in that the economic welfare of the second stage (i.e., economic efficiency and entrepreneurial activities of service-providers) cannot be linked with the social welfare of the third stage (i.e., general welfare or equity of service-users) by an inherently distorted local pricing mechanism. The comparative differences of policy patterns in the second stage are closely related to the synchronic differences of policy styles reviewed in chapter 4: i.e., policy imperative, mode of universal service, and policy trends toward narrowing the digital divide. 240

PAGE 255

In the U.S., universal service originated from the needs of provider-centered system integration and is inherently centered on the service provider and the economic solution. It is based on the values of classic economic efficiency and re-distribution of economic welfare. The universal service mechanism focused on the financial incentives with Universal Service Funds for cross-subsidization and rate-averaging under the tight supervision of the FCC and the PUCs. The policy programs for liberalization are tightly coupled together with the policy program for social control through the local flat-rate pricing structure. In Finland, the concept of universal service is closely related to the concept of social service in that service users' rights to access communications resources are re distributed by the user-centered direct subsidization from the social welfare authorities, the Ministry of Social Affairs and Health based on the notion of general welfare. Social service was codified into the "Act on Social Assistance" and Decree in 1998. In Korea, the concept of universal service was originated from the public services provided by the government, and the tariff of local calls was regulated under the name of public services. The public telecommunication service provider had traditionally been compensated through cross-subsidization from long-distance and international services and rate averaging with the tight regulation ofMIC. A more refined universal service is now under consideration. In the U.S., the recent policy trends toward narrowing digital divide reflect its reaching toward the third stage of Internet diffusion. In Finland, a broader concept of 241

PAGE 256

socio-economic inequalities in an information society was articulated in a revised national strategy document, Quality of Life, Knowledge and Competitiveness in 1998. These policy trends toward social control oflnternet diffusion similarly indicate the third stage of Internet diffusion. In Korea, the policy trends toward solving more broadly defined digital divide was codified into the "Act on closing Digital Divide" in 200 I. Again, this policy trend toward social control of Internet diffusion suggests Korea is moving toward the third stage of Internet diffusion. 242

PAGE 257

CHAPTERS CONCLUSION: REVIEW OF POLICY OBSERVATION This dissertation has analyzed a synchronic and diachronic relationship between communications policy and Internet diffusion to suggest the policy implications necessary for the mature use of the Internet. For this purpose, the 3-D and 3-S frameworks were fonnulated by reviewing synchronic causal studies, diachronic studies, and synthetic comparative studies. Furthennore, comparative dynamics was adapted as a research method to propose more extensive policy implications via a diachronic and synchronic comparison of policy styles and policy patterns in the three nations. In this concluding chapter, (1) the findings from a comparison of policy styles and policy patterns in the three nations are summarized, (2) the practical policy recommendations for the mature .use of the Internet are drawn from the findings from comparative analysis, (3) discussions oftheories, analytic framework (3-D and 3-S models), and validity ofthe findings, and finally (4) the future academic works are suggested. 243

PAGE 258

Summary of Findings The findings of this dissertation are summarized in accordance with both the research questions raised in chapter 1 and the research propositions proposed in chapter 3 regarding the policy styles and policy patterns in the three nations. Based on the findings about the policy styles and patterns, the policy observations are summarized in light of the 3-D and 3-S frameworks formulated in chapter 3 ofthis dissertation. Technological Factors As the technological determinants affecting Internet diffusion, the high intensity of information technology in the three nations is compared in view of (1) industrial distribution ofiT employment, (2) research and development (R&D), (3) innovativeness of the IT sector, and (4) IT networking. The deployment ofiT in the U.S. and Finland are well balanced among the three components, while that of IT in Korea is centered on the manufacturing industry. The share of IT R&D in total business R&D in Finland was just over 51%, the largest for any OECD nations, even though one might note that the total size ofthe Finnish GDP is comparatively small. The shares ofiT R&D in the U.S. and Korea are well above the OECD average (around 34%)-that is, around 40% in Korea and around 38% in the U.S. In terms ofthe industrial distribution ofiT sector R&D, the proportions ofiT R&D undertaken by IT manufacturing business in both Korea and Finland are close to 90%, while that ofiT R&D undertaken by IT manufacturing in the U.S. is around 74%. 244

PAGE 259

In terms of the ratio of R&D to value added and Gross Domestic Product (GDP), OECD (1998c) showed that Finland had the largest ratios ofR&D to value added in the IT sector. It also had high R&D to GDP ratios, while the U.S. and Korea had high ratios ofR&D to value added, both for the IT sector and the total business sector.104 -Public IT networking, combining research laboratories, academia, and industry can influence Internet diffusion by providing institutional platforms. The typical example is Finland's telecom cluster programs, where Nokia perceived itself as an infrastructure leader whose spill-over effects benefited the national economy. In the U.S., through the combined efforts of federal research sponsors (ARPA and NSF), academia, and industry, prototype networking capabilities were deployed on a national scale and popular applications were introduced especially through federally funded IT R&D programs. In Korea, a research institute funded by government, or ETRI played a leading role in the IT networking. Its typical examples occurred for the development of TDX and CDMA wireless technology. The comparison of the technological factors oflnternet diffusion intends to reveal comparative similarities of technological backgrounds for Internet diffusion in the three nations as a high IT intensity group. 104 For the purpose of comparative understanding of proportion of IT R&D in an economy, the total size of R&D and GOP of the three nations was not noted in this thesis. 245

PAGE 260

Policy Styles To answer the first research question (QI), a research proposition (PI) was fonnulated: Q I: What are the comparative characteristics of national policy styles for the mature use of the Internet (Internet policy) across space? PI: A synchronic policy style of communications policy affecting Internet diffusion in a nation can be generalized in view of its policy input, policy system, and policy output. These research question and proposition were fonnulated to show the comparative characteristics of national policy styles for the mature use of the Internet across space. The findings about the national policy styles (see chapter 4) in the three nations can be summarized as the following: (1) Among the always-on national groups, the U.S. strongly adheres to the market function (or, the economic dimension in a triad oflnternet policy) or what we have called a free market ethos, and heavily depends on the relatively and comparatively independent agency (FCC) as the basis of its regulatory system. Moreover, the U.S. has had a leading role in the de-regulation of the communications market. This synchronic and comparative characteristic can be called a de-regulative model of national policy style for the mature use of the Internet, and represents providercentered Internet diffusion. 246

PAGE 261

(2) Among the Scandinavian national groups, Finland strongly attests to the community function (or, social dimension in a triad of Internet policy), what has been referred to as a welfare society ethos. It utilizes the existing functional Ministries as an implementing policy system. Furthermore, Finland demonstrates the comparative characteristics of direct subsidization of usage. Taken in combination, the comparative characteristics represent a re-distributive model of national policy style for the mature use of the Internet, which we have termed a user-centered Internet diffusion. (3) Among the East Asian national groups, Korea strongly supports the government function (the political dimension in a triad oflnternet policy), or a high stateness ethos. Korea creates a new IT Ministry as an implementation system. Finally, Korea shows direct governmental involvement on the planning ofNII. These comparative characteristics display a distributive model of national policy style, and represent a network-centered Internet diffusion. In short, the policy styles of the three nations chosen as the most different systems represent the three different roads to the mature use of the Internet: (1) the de regulative model ofthe U.S. among the always-on national groups; (2) theredistributive model of Finland among the Scandinavian national groups, and (3) the distributive model of Korea among the East Asian national groups. Three national groups were identified in terms oflnternet performance and policy styles, but not 247

PAGE 262

geographical traits. These synchronic policy styles in the three nations were embedded in the diachronic policy patterns for the mature use of Internet. Policy Patterns To answer the second research question (Q2), the four research propositions (P2, P2a, P2b, and P2c) were formulated: Q 1: What are the comparative characteristics of national policy patterns for the mature use of the Internet across time and space? P2: A diachronic policy pattern of communications policy affecting Internet diffusion in a nation can be generalized in view of each stage of Internet diffusion. P2a: Political imperative from the state (politics) and its distributive policy by a national CIO for mature use of the Internet are the comparative characteristics in the first stage of Internet diffusion. P2b: Economic imperative from the market (service providers) and its de-regulative policy by an independent regulatory agency for the mature use of the Internet are the comparative characteristics in the second stage of Internet diffusion. P2c: Social imperative from the community (service users) and its de-regulative policy by a functional ministry for the mature use of the Internet are the comparative characteristics in the third stage of Internet diffusion. These research question and propositions were designed to show the comparative characteristics of national policy patterns for the mature use of the Internet across time and space. The diachronic and comparative characteristics of the national policy patterns (see chapter 5, 6, and 7) in the three nations can be summarized as follows using Noam's (1994) diffusion typologies: 248

PAGE 263

United States. (1) Initiation: The Internet, which was technologically originated in the U.S., was a federally-funded innovation for the purpose of initial military usage and later academic and research usage. Furthermore, the Clinton-Gore Administration initiated the Nil project in 1993 to digitize the national network with the interstate highway analogy. The comparative focus of the Nil was on the service-provider axis of a triad of Internet diffusion (i.e., expanding competition in all segments of the digital marketplace as a way to increase service-providers' innovation and investment) with an ad hoc and unfunded institutional arrangement (or inter agency) of the Information Infrastructure Task Force (IITF) under the cognizance ofthe Secretary of Commerce. (2) Proliferation: The market-led industrial strategies to stimulate private sector investment based on American strong support of the free market were codified in the Telecommunication Act of 1996. The comparative focus of the introduction of competition in the local telephone market was on retail-based competition (i.e., resale and unbundling) with a flat rate regime for local calls provided privately owned monopolistic companies. (3) Control of Proliferation: Keeping in line with the introduction of competition in the local telephone market, universal service was also codified into the Telecommunication Act of 1996. Universal service in the U.S. originated from the 249

PAGE 264

needs of provider-centered system integration. It was inherently centered on the economic solution based on classic economic efficiency and re-distribution of economic welfare. The universal service mechanism focused on the financial incentives with Universal Service Funds for cross-subsidization and rate-averaging under the tight supervision of the FCC and the state PUCs. The recent policy trends toward narrowing the digital divide indicate its movement toward the third stage of Internet diffusion. Finland. (I) Initiation: The Finnish government initiated its Information Society Strategy to digitize the national network as one avenue to overcome the effect of the economic depression. The comparative and initial focus of the national strategy to achieve an information society was on the service-provider dimension (i.e., increasing service providers' entrepreneurship via expanding competition) with the IT-related national budget allocation under the strong leadership of the Ministry of Finance in 1994. With the elaboration ofthe national strategy in 1998, the focus was changed to the service-user dimension (i.e., enhancing the service-users' equity) with the IT related R&D allocation under the strong initiatives of the Finnish National Fund for Research and Development (SITRA) in 1998. (2) Proliferation: The emphasis during this later period remained on overcoming a national economic depression, which led to a national strategy focused on the 250

PAGE 265

business potentials and technology development. The comparative focus of the introduction of competition in the local telephone market in Finland was changed from the more facilities-based competition in 1994 to the more resale-based competition in 1997. In addition, the local telephone service was supplied by the private/public mix of telecommunications companies with a flat rate regime for local calls tied with the local customers. (3) Control of Proliferation: The concept of universal service is more fairly described as one of the social services in that service users' rights to access communications resources were re-distributed by user-centered direct subsidization based on the notion of general welfare from a social welfare authority, in this case, the Ministry of Social Affairs and Health. The social service policy was codified into the Act on Social Assistance and Decree in 1998. A broader concept of socio-economic inequalities in an infonnation society was articulated in a revised national strategy document, Quality of Life, Knowledge and Competitiveness ( 1998). These two major policy trends toward social control oflnternet diffusion reflect Finland's reaching toward the third stage of Internet diffusion. Korea. (I) Initiation: The Korean government provided strong leadership in establishing the KII to digitize the national network employing the government-led national 251

PAGE 266

planning strategy. The comparative focus is on the network dimension (i.e., digitizing the national network) with the specialized IT fund (i.e., the Informatization Promotion Fund) under the leadership of the Ministry of Information and Communication. (2) Proliferation: With response to the external forces for the opening of the telecommunications market and the economic depression, competition in the local telephone market was introduced. In Korea, the comparative focus of the introduction of competition in the local telephone market was on the more facility based competition complemented by introducing resale competition. The local telephone service in Korea was provided by the state-owned monopolistic Korea Telecom with the low prices approved by the Ministry of Information and Communication as part of public service. (3) Control of Proliferation: Universal service in Korea originated from the public services provided by the government, and the tariff of local calls was regulated under the name of public services. The public telecommunications service provider had traditionally been compensated through cross-subsidization from long-distance and international services and rate averaging with the tight regulation of MIC. The more refined universal service is now under consideration, but the policy trends toward solving the more broadly defined digital divide was codified into the Act on Closing Digital Divide in 200 I. This policy trend toward social control of Internet 252

PAGE 267

diffusion suggests Korea is moving toward the third stage of Internet diffusion, even though not as rapidly as in the case ofthe U.S. Pattern Matching Comparison The chronological development of policy patterns in the respective national developments of the Internet can be compared to the chronological development of Internet users per I 00 inhabitants in each nation. In the U.S., the initial stage oflnternet diffusion can be roughly estimated between 1991 to April 1995 or before the year of the taking-off of Internet users triggered by the privatization of the NSFNET. In these time periods, the U.S. government active role can be found in (1) funding NSFNET and (2) initiating the Nil plan (September 1993). The second stage of Internet diffusion can be roughly estimated after the year of the taking-off of Internet users in 1995. In these periods, economic de-regulation activities can be found in ( 1) privatization ofNSFNET in April 1995 and (2) introduction of local competition through the Telecommunication Act of 1996. The third stage of Internet diffusion started from 1996 in that social control or universal service was codified into the Telecommunication Act of 1996, keeping in line with the introduction of competition in local telephone market. Thus, the economic de regulation and social control of Internet diffusion was simultaneously codified into the Telecommunication Act of 1996. 253

PAGE 268

As such, the case ofthe U.S. may not be matched with the theoretical policy patterns of this dissertation, that is, propositions P2b and P2c are not supported by the U.S. case. This exceptional case can be understood in that the policy programs for liberalization in the U.S. are tightly coupled together with policy program for social control through the local flat-rate pricing structure caused by cross-subsidization and rate-averaging. In Finland, we can estimate the initial stage of Internet diffusion to be approximately between 1991 and 1994 or before the year of the taking-off of Internet users triggered by the initial introduction of competition in the Finnish local telephone market. In these periods, the Finnish government assumed an active role in (1) upgrading the local and rural network with the government-owned PTT and (2) initiating the national strategy for achieving an information society. The second stage oflnternet diffusion can be roughly estimated between 1994 and 1998. In these periods, the economic de-regulation activities can be found in (1) the initial introduction of competition in local telephone in January 1994 and (2) elaboration of competition in local telephone with the Telecommunication Market Act in 1997. Social service was codified into the "Act on Social Assistance" and Decree in 1998. Broader concepts of socio-economic inequalities in an information society were articulated in a revised national strategy document, Quality of Life, Knowledge and Competitiveness in 1998. These two major policy trends toward social control of Internet diffusion demonstrate movement toward the third stage of Internet diffusion. 254

PAGE 269

In Korea, the initial stage oflnternet diffusion is estimated to lie at between 1991 and 1998 or before the year of the taking-off of Internet users triggered by the introduction of competition in the Korean local telephone market. In these time periods, the Korean government active role can be found in (1) direct funding for constructing the NKN-G and the NKN-P (2) initiating the KII ( 1994). The second stage of Internet diffusion can be roughly estimated to lie at between 1999 and 2000. In these periods, the economic de-regulation activities can be found in competition in local telephone market (April 1999). Furthermore, the comparatively late policy trend toward social control of Internet diffusion codified into the "Act on closing Digital Divide" in 200 I shows its reaching toward the third stage of Internet diffusion. Thus, the cases of Finland and Korea can be well matched with the theoretical policy patterns in the developmental stage oflnternet diffusion ofthis dissertation, that is, propositions P2a, P2b, and P2c are supported by the cases of both Finland and Korea. The diachronic relation of policy patterns to Internet users per I 00 inhabitants in the three nations can be summarized in Figure 8-1. Figure 8-1 shows several policy implications worthy ofbeing further discussed. First, all the three nations chosen as the most different systems show that critical mass points (taking-off point) oflnternet users in each nation lie at around 5 Internet users per 100 inhabitants. The theoretical meanings, generalizability, and subsequent policy implications of 5% Internet users as a taking-off point in a nation are worthy of being further investigated in future academic work. 255

PAGE 270

Second, taking-off points in the three nations are well matched with the policy trends toward liberalization. In the U.S., taking-off oflnternet users occurred twice at both April1995 and 1996, when NSFNET was privatized and local telephone competition was introduced respectively. In Finland, taking-off oflnternet users occurred twice at both 1994 and 1997, when initial local telephone competition was introduced and elaborated respectively. In Korea, taking-off point oflnternet users occurred at April 1999, when competition in local telephone market was introduced. Third, from the perspective of the time-consuming for reaching toward 40% Internet users from 5% Internet users, both the U.S. and Finland spent around 6 years as the leading nations of developing the newly-emerging technology, while Korea spent just 2 years as a catching-up nation. 256

PAGE 271

Figure 8-1: Policy Patterns and Internet Users in Three Nations u.s. Social Control 40 35 Initiation Liberalization 30 25 20 15 10 5 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 40 ... 35 30 25 20 15 10 Finland Initiation Liberalization 5 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Korea -+-Internet users per 100 inhabitants --+-Internet users per 100 inhabitants Initiation Liberalization Social Control 45 40 35 30 25 20 15 10 5 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 257 -+-Internet users per 100 inhabitants

PAGE 272

Policy Observations Based on the findings from the comparative analysis ofthe policy styles and patterns in the three nations, the policy observations about the overall comparative characteristics of the three nations can be summarized from the theoretical perspective ofthe 3-D and 3-S frameworks as the following: Policy Styles (3-D) and Internet Diffusion. The policy styles of the three nations are roughly matched with the three theoretical dimensions oflnternet policy of this dissertation. The case of the U.S. reflected the economic dimension oflntemet policy. The communications policy for Internet diffusion focused on enhancing economic efficiency and entrepreneurial activities of service providers for private investment and innovation. Private investment and innovation were achieved by an active de-regulative policy (for example, introduction oflocal competition through resale and unbundling) and privately owned monopolistic service providers with the basic recognition of the communications service as a private service. The case of Finland demonstrated the social dimension of Internet policy. Thus, the communications policy for Internet diffusion centered on enhancing users' equity. It was achieved by direct social assistance from the social welfare authorities and "private/public mix" of telecommunications companies (i.e., the private or customer owned monopolistic cooperatives in urban areas and the public monopolistic provider 258

PAGE 273

or Telecom Finland in rural areas) with the recognition of communications service as a social service. The case of Korea mirrored the political dimension oflnternet policy. The communications policy for Internet diffusion in Korea was directed on enhancing network digitalization. It was achieved by an active distributive policy (for example, the direct funding and planning ofNKN-G and NKN-P) and state-owned monopolistic service provider (KT), treating communications service as a public service. Policy Patterns {3-S) and Internet Diffusion. The diachronic relationship of communications policy to Internet diffusion can be identified in the three stages of Internet diffusion. First, the comparative characteristics of the political initiatives on national information infrastructure can influence the initial stage oflnternet diffusion by digitizing the existing communications network. Second, the comparative characteristics of the economic de-regulation by introducing competition in the local telecommunications market can affect the proliferation of Internet diffusion by stimulating facility-competition among local companies and lowing (or, flat-'rating) the charges for the dial-up connection to the Internet. 259

PAGE 274

Third, the comparative characteristics of social control in the name of universal service can affect Internet diffusion beyond the private optimum point by subsidizing the usage of the Internet. Finally, the comparative characteristics of policy styles and patterns in the three nations can be summarized in Figure 8-2. Figure 8-2: Policy Styles and Patterns in Three Nations The Mature Use of Internet Policv Pattern in Stage J Control) Public Service Universal Service Social Service (Public Provision) (Asymmetric Subsidies) (Direct Subsidies) Policv Pattern iAStage 2(Economic DereguJationt t Public PTO Private PTO Public/Private Mix (Facility competition) (Resale, Unbundling) (Cooperatives, resale) Policv Pattern it Stage 1 (Political Initiation) t t Network Digitalization Providers' Innovation Providers' Innovation -+ Users' Equity Policv Style 1 t j Government-led Market-led Community-led Distributive Model De-regulative Model Re-distributive Model I I I Korea I I The United States. I I Finland I 260

PAGE 275

Policy Recommendations and Government Role Policy recommendations can be drawn from the findings about both the policy styles in the three nations and the policy patterns in each stage oflnternet diffusion, which lead us to suggest government role in each stage of Internet diffusion. Policy Recommendations United States. The main rationale for de-regulation in the monopolistic local telecommunications sector is that introducing the competition among the service providers enhances economic efficiency and entrepreneurial activities ofthe service provider. Furthermore, as a result, the supposed competition should result in lowering or no local telephone usage charge in the short run and induce private network investment-and technological innovation in the long run. Nevertheless, the extreme unbundling requirements for securing the competition or inducing facilities-based competition brought in the comparatively and theoretically tighter regulation of the FCC and the state PUCs for both coordination and oversight over monopolistic power, which can cause lowering private motivation to invest in an alternative network. The U.S. extreme unbundling requirements can be compared to the Canadian case in that Canada shows a typical model case among the advanced always-on national groups in liberalization policy. The Canadian regulatory body decided against mandatory discounts for resellers of incumbents' services and to require only essential 261

PAGE 276

facilities limited by an antitrust-type standard to be unbundled with the time limits (CRTC, 1997).105 The extreme unbundling requirements ofthe U.S.including urban loops, tandem switching, transport, directory assistance databases and services, which are not essential facilities in Canada-can be considered to promote resale of incumbents' services and network elements at the expense of delaying or deterring facilities-based entry, especially in residential and small business markets (Harris and Kraft, 1997). Furthermore, the detailed regulatory intervention ofthe state Public Utility Commissions (PUCs) for technical coordination between operators to fulfill the extreme unbundling requirements can further impede facilities-competition. In practical terms; these differences result in more limited unbundling in Canada than the U.S. and are therefore likely to promote greater facilities-based competition in Canada. The data show that total broadband penetration of Canada (2"d) as the alternative routes to the dial-up connection to the Internet in June of2001 outranks that of the U.S. (3rd) (Appendix K). Hence, it is recommended to consider imposing more limited unbundling requirements with time limits (as in Canada) and deregulating the limits on investing in an alternative network like DSL as in the Tanzin-Dingell bill (see p.180). 105 The Canadian regulatory body required incumbent local exchange companies to unbundle access to (I) telephone numbers, (2) directory listings, (3) all local loops for five years, (4) low density rural loops permanently, and (5) transport for five years (CRTC, 1997). 262

PAGE 277

Furthermore, it is noteworthy that the flat rate local call structure in the U.S. was caused not by an economic imperative, but by policy considerations ofuniversal service (i.e., rate averaging and then for subsidizing local telephone calls). Thus, in local telecommunications market, economic welfare (or user's equity) cannot be linked theoretically with economic efficiency (or innovation caused by entrepreneurial activities) by an invisible market function of the price. In other words, competition among service-providers in the U.S. local telecommunications market cannot be linked with enhancing the serv'ice-users' welfare because of an inherently-distorted pricing mechanism by non-market considerations in the name of universal service. As a result, Internet diffusion in the U.S. is much indebted not to economic rationales but to asymmetric subsidization. Its typical examples include (1) indirect subsidization from the urban callers to the rural callers by rate-averaging, (2) indirect subsidization from the basic service providers (like LECs) to the enhanced service providers (like ISPs) by exemption ofiSPs from the access charges, and (3) indirect subsidization from LECs to ISPs by the exemption ofiSPs from universal service contribution.106 In this.respect, in the short run, the more direct subsidization like Lifeline and Link-up America needs to be enlarged. This is not only because an asymmetric or indirect subsidization can be justified only in the initial or proliferation stage for boosting a certain usage (or Internet usage), but also because an asymmetric 106 Even though it is out of the scope of this dissertation, one more crucial asymmetric subsidization is subsidizing from off-line users to on-line users by the Internet Tax Freedom Act of 1996. 263

PAGE 278

subsidization has a possibility to contribute an adverse re-distribution of the general welfare as in the case of subsidizing the rural customers who can be a "better-off." In the long run, "customer ownership of the local loop" (Schechter, 1996) as in Finland can be one of policy alternatives for the non-economic solution of inherently monopolistic nature of local loop to secure the publicity or general welfare of Internet users. Finland. The case ofFinland can be considered a less distorted solution to the inherent monopolistic nature ofthe local loop in that there is no rate-averaging in the Finnish local telephone market by public installing of telecommunications facilities in non-economic rural regions and by enhancing users' equity by direct social assistance from the social welfare authorities. Still, considering that the close tie of local companies with the customer through local cooperatives demonstrated monopolistic power in competition with the public companies, we recommend imposition of limited unbundling requirements to counter the community failure caused by the customer-tied local cooperatives in urban areas. In this thesis, the term of community failure is formulated to describe monopolistic power of local cooperatives in Finnish local competition caused by the close tie of local companies with the customer through institutional arrangement of customer as stock holder (see pages I 88-189). 264

PAGE 279

Korea. Given the direct involvement of government on both installing the KII and liberalizing the local telecommunications market by approving the local tariffs, it is recommended to treat the theoretical government-failure by rationalizing (or updating) the safe guards to secure local competition, such as imposing limited unbundling requirements with time limits and introducing a price cap system. In addition, to overcome theoretical market-failure (such as, no linkage between economic efficiency and social equity) caused by natural monopolies of the Korean local telecommunications market (as in the U.S.), it is recommended to consider policy alternatives for more directly subsidizing the service users like social assistance as in the Finnish case or Lifeline and Link-up America as in the U.S. Furthermore, considering that the privatization of the public monopolistic company (KT) is in process, "customer ownership of the local loop" (Schechter, 1996) as in the Finnish case can be one of policy alternatives for the non-economic solution of inherently monopolistic nature of local loop to secure the publicity or general welfare of Internet users. Theoretically, the main advantage of customer ownership of the local loop can be a strong linkage between economic efficiency of service providers and social equity of service-users in local telecommunications market where pricing mechanism provides weak linkage between economic efficiency and social welfare. 265

PAGE 280

Government Role in Each Stage Based on the findings and policy observations, the third research question (Q3) regarding the government role in the developmental stage of the Internet can be answered. Q3: What is the appropriate role of government in terms of the national policy styles and pattern across time and space? Setting aside an exceptional case of the U.S. where the policy programs for liberalization and social control were simultaneously introduced into the Telecommunication Act of 1996, governmental involvement for the mature use of the Internet can be theoretically justified in the developmental stage of Internet diffusion. It is because (I) Internet service can be considered the "commons" in its nature in general, (2) public access to the Internet as a social network can be a critical necessity, but not a technical luxury for the full participation of the knowledge-based information society, and (3) the three important variables (network, service-providers, and service-users) and their ultimate goals (network digitalization, economic efficiency and entrepreneurial activities of service-providers, and general welfare of service-users) cannot be linked together by an inherently-distorted pricing mechanism in natural monopolistic local telecommunications market, and (4) thus, social optimum in Internet diffusion can be achieved not by the market itself but by the policy-mix of political, economic, and social dimensions of Internet policy. 266

PAGE 281

Initiation. The three nations show different degrees of the governmental involvement in political initiation -i. e., the different focus of funding and different funding mechanism for digitizing the national network. Nevertheless, all the three nations show an active governmental involvement in national network development. Such involvement in the digitization of national network can be justified in that the digitizing of national network needs the huge amount of initial investment. Especially, installing network in high cost/low profit rural area cannot be achieved by the function of profit (price) as in the case of postal service (the post) regarded as public goods. In this regard, the role of the government in this period can be theoretically linked to a network dimension among the three axes of triad oflnternet policy, and proposed as an active promoter of the digitization of the existing national network with a distributive policy (national planning and funding mechanism). Proliferation. Within the three nations, different motives and degrees of the de regulation were seen to exist-i. e., the different mode of local competition and the different ownership and pricing structure. Nevertheless, all the three nations show active de-regulation activities in the telecommunications market. Such de-regulation activities as privatization and introduction of competition among service-providers in the second stage can be justified in that the network beyond a certain point (or critical mass point, 5% Internet users in this thesis) can achieve self-sustained growth because of the economy of scale (see Noam 's interactive diffusion model). As such, the role of 267

PAGE 282

the government during this period can be theoretically linked to the service-provider dimension among the three axes in triad of Internet policy, and identified as an efficient promoter of service-providers' economic efficiency and entrepreneurial activities with a de-regulative policy. Control of Proliferation. The three nations were shown to have different mechanism and degree ofthe social control oflnternet diffusion-i. e., the different recognition and re-distributive mechanisms ofuniversal service and digital divide. Nevertheless, all the three nations show an active governmental involvement in social control of Internet diffusion. Such an involvement in social control of Internet diffusion by providing universal service and narrowing digital divide can be justified in that distorted pricing mechanism in Internet arena cannot provide a linkage between providers' economic efficiency and users' social welfare. Furthermore, there is a strong possibility of large amount of network externalities in that the Internet in nature is interactive communications medium. In this regard, The object ofthe governmental involvement in this period can be theoretically linked to the social dimension among the three axes in triad oflnternet policy, and proposed as a fair promoter of service users' social welfare with a re-distributive policy. 268

PAGE 283

As a logical conclusion, it is recommended that the government involvement as a "fine tuner" (what Keynes indicated) of three dimensions in each developmental stage of the Internet will bring the end stage oflnternet diffusion.107 Discussion of Framework and Findings Theories and Analytic Framework (3-D and 3-S) To get a more comprehensive understanding about the government role in Internet diffusion, this thesis proposed a theoretical and descriptive framework ofthe 3-D and 3-S models with the three sets ofvariables of a triad oflnternet policy. For this purpose, the inherent nature of Internet service as the commons is reviewed based on the basic economic classification of goods and services, which brings us to the three theoretical approaches to Internet diffusion: i.e., political approach with allocation mechanism of the government, economic approach with an allocation mechanism of the market, and social approach with an allocation of the mechanism ofthe community. Based upon the theoretical perspective of the Internet service as the commons, the three sets of theories explaining the relationship of communications policy to Internet diffusion over time and space are reviewed: synchronic causal theories (across space), diachronic process theories (across time), and synthetic comparative studies (across time and space). 107 The concept of fine tuning of three dimensions in Internet economy in this thesis is analogous to Keynes' fine tuning of demand and supply in macro-economy. 269

PAGE 284

Reviewing the synthetic causal theories leads us to propose the following: ( 1) The governmental (or policy) determinants of Internet diffusion can be more important among the nations with a highly developed communications network (like OECD nations) than the non-governmental (or non-policy) determinants of Internet diffusion that can be crucial to the nations with a low developed communications network like the most non-OECD nations (for examples, around 95.6% of Internet hosts were in the OECD nations) in that communications facilities can be considered a prerequisite for the development of the Internet network. (2) As the governmental determinants (policy factors) of Internet diffusion, the policy mix among political, economic, and social policies is recommended in that the inherent nature of Internet service is not a private goods arid local communications pricing is already distorted by social (i.e., not market like universal service) considerations. Under the diachronic theories of Internet diffusion, two diffusion models are reviewed among the various theories regarding the technological innovation in that ( 1) Roger's model can give us a basic and broad overview of the chronological trends of innovative technology, and (2) Noam's model can give us an insight toward the chronological policy implications on each stage of diffusion of the interactive communications medium. Furthermore, the choice of the two diffusion models (especially, Noam's) can be justified because we are dealing the diffusion of a 270

PAGE 285

service (or Internet service) on an interactive network, but not of a consumer goods (or PC and VCR). Nevertheless, it is necessary to note that the social optimum point in Noam's model is a just relative or metaphorical point between private optimum and exit point under the limited assumption of a network with an equal-price system and negative network externalities. According to the synthetic theories of Internet diffusion, the several theories are reviewed to give an overview of research method of this dissertation. Among others, Rose's comparative dynamics-a method of comparison across time and space-is adapted as a main framework of this thesis, which leads us to formulate the synthetic 3D and 3-S frameworks with a synchronic (or across space) 3-D model and a diachronic (or across time) 3-S model. Such 3-D & 3-S synthetic frameworks can be justified in that it is intended to provide a more comprehensive explanation ofthe multidimensional policy factors of Internet diffusion by overcoming the comparative limits of both synchronic causal theories (or overlooking of the dynamic process of Internet diffusion) and diachronic process theories (or overlooking ofthe policy factors oflnternet diffusion). Thus, the literature review on synchronic causal theories, diachronic process theories, and synthetic comparative dynamics provides an insight to formulate 3-D model, 3-S model, and the combined research method ofthis thesis, respectively. Based on both the implications of the literature review and the subsequent 3-D and 3-S analytic frameworks, theoretical propositions are formulated to be tested by 271

PAGE 286

pattern matching comparison between three different communications policies (Nil, liberalization, and universal service) and three different stages oflnternet diffusion (initiation, proliferation, and control of proliferation). These propositions are largely derived from a combination ofNoam's interactive model providing policy implications in each stage of diffusion with Rogers' logistic model providing a chronological s shaped diffusion trend (Figure 3-2, page 73). Thus, considering the 3-D and 3-S analytic frameworks of this thesis, these can provide a more comprehensive overview of the relationship of communications policy and Internet diffusion across time and space. Nevertheless, the weaknesses of the 3-D and 3-S frameworks must be noted: (1) it cannot provide non-governmental factors (e.g., English proficiency) of Internet diffusion, (2) it can provide just comparative policy emphasis on each stage oflnternet diffusion, (3) even though the critical mass point can be analogous to the point of the taking off of Internet users (April 1995 in the U.S., January 1994 and 1997 in Finland, and April 1999 in Korea), the exact private optimum point and social optimum point cannot be identified in this dissertation in that the Internet users per I 00 inhabitants are reaching toward around 40-41 (34.65 in the U.S., 37.22 in Finland, 40.25 in Korea in December 2000) (Appendix N). Validity of Findings Based on Allison's (1999) Model I (the National Actor) and Model II (the Organizational Actor), the focus of comparison and contrast in this thesis was on the 272

PAGE 287

national and organizational characteristics of policy styles and patterns in the developmental stage of Internet diffusion in the three nations. So, the findings of this thesis are valid only in the organizational and national level i.e., the level of individual actor was considered as a given in relationship between communications policy and Internet diffusion. The construct validity and reliability of this thesis can be a weakness as in the most case studies method in that "a case study investigator can fail to develop a sufficient operational set of measures and that subjective judgments are used to collect the data" (Yin, 1994). To cope with the construct validity threat, this thesis calibrates the balance between specificity and generalizability by (1) employing a data triangulation of quantitative and qualitative information, (2) using the sources of evidence from documentation and archival records (especially from OEC and ITU), and (3) developing an operational set of measures regarding communications policy (e.g., Nil documents, competition, privatization, pricing structure, and universal service) and Internet diffusion (Internet host, Internet server, and Internet users). Nevertheless, the construct validity of this thesis can be regarded as a comparative weakness in that the main purpose of this dissertation is both exploratory and descriptive to build a theory to describe the contemporary Internet-related policy controversies. Internal validity requirement for this dissertation can be addressed by (1) formulating the theoretical 3-D and 3-S frameworks to test the descriptive propositions 273

PAGE 288

regarding the relationship of communications policy to Internet diffusion (2) conducting time-series or chronological analysis, and (3) using the analytic tactic of pattern-matching comparison. Finally, external validity of this thesis can be met by (l) utilizing replication logic in multiple-case studies such as a literal replication (comparative similarities) and a theoretical replication (comparative differences) and (2) utilizing the most different system design as a kind of ideal type method. Thus, this thesis can be applicable especially to the nations where the highly developed communications network is established and to the diffusion of the interactive communications technology such as broadband access to the Internet (cable and DSL) and digital TV. Future Academic Study Future academic work based on this dissertation is suggested in five arenas: (1) the theoretical policy systems in infonnation society, (2) the so-called ''third way" in communications sector, (3) local competition (unbundling) and high speed Internet access, (4) the relationship between communications service (the post) and infonnation service (the press), and (5) utilizing the 3-D and 3-S frameworks. Theoretical Policy Systems in Infonnation Society Considering the multidisciplinary nature and the effects of the Internet as an interactive social network, the Internet-related policy issues can be the inter-ministerial 274

PAGE 289

issues among the functional ministries in the industrial society. In this respect, a new organizational scheme fitted in knowledge-based information society need to be critically reviewed. The issues about a new institutional scheme in an information society raised in this dissertation 108 suggest the following: Critically reviewing the validity of the administrative criteria of the so-called "functionality" for classification of the ministerial arrangements in the era of the transformation from the industrial society to the so-called "network society." Institutionalizing an organizational scheme for coordinating intra-ministerial issues and extra-ministerial issues by expanding the concept of Chief Information Office (CIO) in Management Information System (MIS) into the concept ofNational Chief Information Office (National CIO) in Public Management Information System (PMIS). Institutionalizing a more high-level of the organizational scheme for coordinating inter-ministerial issues like "e-Minister" and "e-Envoy" as in the U.K. Critical reviewing the way of harmonizing a need for de-centralization caused by the technological characteristics of the Internet (i.e, decentralized intelligence) with a need for centralization and coordination caused by the multi-disciplinary effects of the Internet (i.e., e-Minister and e-Envoy). 108 This dissertation is intended to compare and contrast the issues about the organization scheme fitted into an information society to raise the institutional issues to be solved in the near future. 275

PAGE 290

Critically reviewing the validity of the distinction between policy function and regulatory function by establishing the independent regulatory agency in that the criteria for distinguishing the two concepts (policy function and regulatory function) is ambiguous at best, arbitrary at worst. It is arguable whether the rate averaging for subsidizing from urban callers to rural callers is an economic regulatory function or social re-distributive policy function. Third Way in Communications Sector The Finnish local urban telephone operators were run as cooperatives owned by their respective customers in their traffic areas. Most ofthese telephone operators were owned by their customers while a small number were entirely municipally owned. This ownership structure can be considered a newly emerging self-governing allocation mechanism (i.e., a small or medium range of community in CPRs situations in Ostrom's [1999] institutional rational choice situation) in the communications arena. Considering that this organizational arrangement of the local cooperatives can be an alternative to the provider-centered monopolistic organization scheme (whether it is privately owned or publicly owned) in the communications market, the issues to be reviewed critically in the near future are the following: Critical reviewing the way of harmonizing the organizational aim of maximizing profits with the shareholders' (i.e., the customers) aim of minimizing prices. 276

PAGE 291

Critical reviewing the monopolistic characteristics of local cooperatives as a price regulation mechanism for its service by tying shareholders with minimizing overall pnces. Securing safeguard measures for easing the monopolistic power of local cooperatives, or community failure. Unbundling and High Speed Internet Access The main pro-competitive rationales for resale and unbundling is that it may promote facilities-based competition, but it is important to consider that tightly defined unbundling requirements can also impede facilities competition (or, impeding high speed Internet access via DSL or cable modem). The issues raised in this dissertation are the following: Whether or not unbundling requirements encourage the incentives for construction of competitive network facilities. Whether or not unbundling requirements encourage investment in alternative access networks (wired or wireless). Conducting a cost-benefits analysis about whether or not the benefits such as avoiding unnecessary and inefficient duplication of components are greater than the costs such as detailed regulatory intervention for the technical coordination between operators. Critical reviewing the limited unbundling requirements as in the Canadian case. 277

PAGE 292

Post and Press Considering that the domains of communications service (the post) and information service (the press) are rapidly convergent together on the web, and the information service over the communications line (the contents service on web) are overwhelmingly catching up the communications service (telephone service), the issues raised in this dissertation reflect the following: How can we formulate the new universal service expanding beyond Plain Old Telephone Service (POTS): is it to communications (i.e., to the Internet) or information content (i.e., to the content on web)? If the access to the content on the web is more important in a knowledge-based information society, how can we h_armonize Article 1 in Section 8 of the U.S. Constitution (permitting governmental involvement in the post) with the First Amendment (prohibiting governmental involvement in the press)? If the new concept of universal service means access to the new communications service like the Internet, how can we coordinate the new concept of universal service, or access to the web (communications gaps) with the so-called digital divide (information gaps)? Designing the specific and generalized indicators for measuring the digital divide for the purpose of comparing national disparity among the nations. Critically reviewing the linkage between the Internet service (communications service) as a commons reviewed in chapter 2 and the content on the web 278

PAGE 293

(information service) as a commons for full participation in economic, political, and social life in a knowledge-based information society ofthe 2151 century. Utilizing 3-D and 3-S The 3-D and 3-S frameworks, developed as a tool for comparative dynamics in this thesis, can be utilized and supplemented by future academic works. Applying the 3-D and 3-S frameworks to the other developed OECD nations or developing non-OECD nations. Supplementing the 3-D and 3-S frameworks by analyzing the missing variables of this thesis like the effects of the individual actors and the cultural factors. Investigating and generalizing the 5% of Internet users in a nation as a taking-off point. Closing Remarks The main focus of this dissertation is on identifying the government policy factors of Internet diffusion time and space. In this respect, the several findings and subsequent arguments of this dissertation can be summarized in four themes: First, based on a theoretical review of its inherent nature, the Internet service can be identified as a "commons," which can be a rationale for on-going governmental involvement. 279

PAGE 294

Second, as a logical argument based on the above theoretical finding, a key success factor of Internet diffusion in these three nations as a high IT intensity group is not "non-governmental regulation (or un-regulation)" (European Commission Report, 2000), but an explicit government policy in each stage oflnternet diffusion. Third, the economic logic of liberalization i.e., the competition among service providers is supposed to facilitate to invest the alternative network and to enhance the social welfare of service-users -cannot function well in the local communications market because of the distorted local pricing mechanism by the social or political considerations like universal service. Fourth, the single and most significant factor for the development of the Internet is not ''the movement toward deregulation in the telecommunications sector" (FCC Working Papers, 1997, 1999), but the contingent policy mix of (1) political national planning for digitizing the existing network in the initial stage, (2) economic liberalization for enhancing economic efficiency and entrepreneurial activities of service-providers in the proliferation stage, and (3) social control for enhancing equity of service-users in the control of proliferation stage. Thus, from a public affairs perspective, it is clear that governmental involvement in each stage oflnternet diffusion will bring the ultimate stage of the mature use of the Internet, often overtly, other times as a fine tuner of three dimensions, but never as a non-involved bystander. 280

PAGE 295

Appendix A: Internet Hosts per 1000 Inhabitants Change Jul-95 Jul-96 Jul-97 Jul-98 Jul-99 Jan-00 (Jul 1999-Jan 2000) Finland 22.5 55.3 67.1 104.0 122.7 148.1 25.4 U.S. 14.0 26.2 37.2 78.1 1 I8.0 141.5 23.5 Iceland 25.6 40.5 52.6 77.1 I08.2 137.0 28.8 Norway 16.1 29.4 50. I 77.0 92.8 120.3 27.6 Sweden 14.6 26.3 39.7 62.1 96.0 114.8 18. 9 Canada 14.0 25.2 38.7 73.4 93.6 I 11.1 17.5 Denmark 9.0 18.8 32.0 51.1 73.4 92.7 19.3 New Zealand I2.3 21.7 42.5 49.6 59.2 88.1 28.9 Netherlands 9.9 16 3 25.5 42.0 56.9 84.8 27.8 Australia 12.2 23.3 40.6 45.5 60.2 77.3 17.1 Switzerland 10.6 17.9 25.6 41.3 59.6 76.1 16.5 U.K. 6.0 12.1 I 8.1 28.2 43.7 60.3 16.6 Austria 5.6 9.9 12.4 20.2 35.5 50.0 14.5 Belgium 2.8 5.3 10.0 19.1 36 8 49.2 12.4 Luxembourg 4.9 9.5 12.8 23.6 38.1 43.5 5.4 Ireland 3.3 7.1 I0.8 16.6 28.8 36.4 7.7 Germany 4.8 7.9 12.4 18.4 25.6 34.0 8.4 France 2.5 4.4 6.6 11.4 22.7 29.8 7.2 Japan 1.4 4.2 8.0 11.8 18.1 25.8 7.6 Spain 1.5 2.6 4.6 9.9 16.2 22.8 6.5 Italy 1.1 2.6 4.6 8.0 12.6 18.9 6.3 Korea 0.7 1.5 3.5 5.4 8.5 18.8 10.3 Hungary 1.2 2.6 3.5 7.7 10.3 14.2 3.9 Portugal 1.0 2.0 2.2 5.4 8.1 12.8 4.7 Czech Rep. 1.5 3.2 4.9 6.7 9.4 12.8 3.4 Greece 0.6 1.4 2.1 4.4 8.1 9.6 1.5 Poland 0.4 1.0 1.2 2.6 4.4 6.0 1.6 Turkey 0.1 0.3 0.6 1.0 2.6 5.4 2.8 Mexico 0.1 0.3 0.4 1.1 2.7 5.0 2.2 OECDAve. 5.9 11.2 16.8 31.4 47.1 59.3 12.2 I. These data are based on a survey undertaken by NW for the ISC. Internet hosts under gTLDs(.com,.org and net) have been reallocated according to the proportion of gTLDs registered by users in each country. The data for January 2000 was contributed by Mathew Zook. Source: ISC, OECD (2000a). 281

PAGE 296

Appendix B: Web Sites in OECD Sites by country ByccTLD Weighted by weighted by By ccTLD Domain GTLDs gTLDs Per 1000 Probable probable Inhabitants Physical economy Location or Network United States(l) 489,688 7,465,358 5,602,437 1.8 United Kingdom 832,878 895,369 1,125,788 14.2 Switzerland 87,035 96,206 132,259 11.9 Denmark 62,844 63,870 90,139 11.9 Luxembourg(2) 3,137 4,727 4,727 7.4 Iceland 2,746 2,787 4,278 9.9 Germany 745,130 793,547 940,218 9.1 Netherlands 121,249 142,442 213,018 7.7 Sweden 57,376 75,266 133,649 6.5 New Zealand 29,008 31,183 40,884 7.6 Austria 56,899 64,695 80,957 7.0 Canada 59,937 231,428 374,338 1.9 Australia 106,145 120,650 169,814 5.7 Norway 25,802 28,284 49,534 5.8 Finland 17,062 29,802 40,948 3.3 Belgium 41,200 42,000 73,092 4.1 Czech Republic 32,058 32,694 35,429 3.1 Ireland 6,334 7,073 19,800 1.7 Korea 75,949 83,273 181,490 1.6 Italy 74,219 80,148 150,607 1.3 Hungary 9,929 12,719 15,177 1.0 Poland 44,939 45,206 53,750 1.2 France 51,669 67,158 226,147 0.9 Greece 11,399 11,461 15,613 1.1 Portugal 10,239 10,342 16,541 1.0 Spain 20,125 27,121 107,316 0.5 Japan 65,829 76,436 176,925 0.5 Mexico 13,001 14,929 27,137 0.1 Turkey 11,710 14,110 58,945 0.2 .com 6,280,702 .net(3) 800,381 .org 552,241 OECD 10,798,860 10,570,283 10,160,959 9.7 I. The US includes .edu,.mil,.gov and .us but most activity occurs under gTLDs. 2. A netsizer weighting was not available for Luxembourg. 3. The data for .net is an estimate based on the December 1999 data. Source: Netcraft (www.netcraft.com), OECD (2000a). 282

PAGE 297

Appendix C: Secure Servers per 1 Million Inhabitants Sep 97 Feb98 Aug98 Feb99 Jul99 Jan 00 Iceland 36.6 36.2 47.1 100.5 104.1 161.6 United States 27.6 40.9 60.8 89.9 116.0 147.7 Australia 13.6 24.1 36.6 53.9 69.8 105.7 New Zealand 15.4 22.7 26.6 46.8 59.3 83.3 Switzerland 8.0 13.8 24.1 41.3 54.6 81.8 Canada 18.1 30.6 33.5 49.5 58.0 74.6 Luxembourg 7.2 9.5 28.4 42.2 61.0 75.1 Sweden 6.0 10.5 20.7 33.6 45.7 63.5 United Kingdom 6.0 9.2 14.0 21.3 29.5 47.4 Finland 3.9 7.4 15.7 24.8 34.8 47.8 Norway 5.2 9.5 14.5 22.7 29.3 44.6 Ireland 4.6 9.2 16.6 17.5 26.2 42.1 Austria 3.2 7.5 13.0 23.1 29.5 40.5 Denmark 2.1 5.5 10.1 13.8 21.2 33.7 Germany 1.8 4.0 6.8 13.2 19.8 29.5 Netherlands 4 8 6.8 9.4 15.5 19.4 25.8 Belgium 2.1 3.5 5.1 11.4 15.7 21.2 France 1.1 2.4 4.3 7.6 10.7 15.9 Spain 3.0 4.6 6.7 9.3 10.9 14.3 Japan 1.6 2.5 4.2 7.6 9.2 12.9 Czech Republic 0.6 1.3 2.5 4.5 8.6 10.6 Italy 1.5 2.7 3.4 5.3 7.5 9.6 Portugal 1.6 1.9 3.1 5.1 6.0 8.8 Greece 0.5 0.8 1.4 2.8 4.5 5.6 Hungary 0.7 1.1 1.9 2.4 2.6 4.3 Korea 0.4 0.7 0 9 1.9 2.3 3.4 Poland 0.2 0.2 0.7 1.2 1.6 2.5 Turkey 0.1 0.1 0.2 0.6 0.8 1.2 Mexico 0.2 0.3 0.3 0.5 0.6 1.0 OECD Average 8.9 13.6 20.1 30.5 39.6 52.2 Source : Netcraft, OECD (2000a). 283

PAGE 298

Appendix D: Internet Subscribers (January 2000) Country Largest PTO Largest Largest PTO's Total Subscribers PTO's Share% Subscribers per 100 Subscribers Australia Telstra 650,000 27 2,407,407 12.9 Austria Telekom 107,000 22 486,364 6.0 Austria Belgium Belgacom 325,000 30 1,083,333 10.6 Canada All PTOs 800,000 13 6,169,500 20.4 Czech Czech 45,000 23 199,000 1.9 Republic Telecom Denmark Tele Danmark 393,000 35 1,135,393 21.4 Finland So rena 252,500 45 564,224 10.9 (Telecom Finand) France FT 1,124,000 37 3,030,000 5.1 Germany DT 3,300,000 37 9,000,000 11.0 Greece OTE 85,983 43 199,960 1.9 Hungary Matav 51,315 45 114,033 1.1 Iceland Telecom 19,650 40 49,125 17.9 lceland(PTI) Ireland Eircom 243,000 60 405,000 10.9 Italy Telecom ltalia 1,990,000 40 4,975,000 8.7 Japan NIT 1,098,000 10 10,590,000 8.4 Korea Korea 1,970,021 18 10,860,000 23.4 Telecom Luxembourg P&T 11,411 Luxembourg Mexico Telmex 402,754 30 1,350,000 1.4 Netherlands KPNTelecom 907,000 32 2,834,375 18.1 New Zealand TelecomNZ 245,000 46 535,000 14.1 Norway Telenor 400,000 58 695,303 15.7 Poland TPSA 80 Portugal Portugal 261,000 35 745,714 7.5 Telecom Spain Telefonica 659,000 35 3,625,000 9.2 Sweden Telia 613,000 30 2,040,000 23.0 Switzerland Swisscom 322,852 33 898,000 12.6 Turkey Turk Telecom 6,638 United BT 1,300,000 18 7,400,000 12.5 Kingdom United States (I) 3,965,000 8 49,723,100 18.4 OECDtotal (2) 21,548,124 34 121,114,832 11.0 EU total 11,571,894 37,524,364 10.0 Note: (1) The telecommmunication carriers used for the US are the traditional Bell system operators. AT&T and the RBOCs, plus GTE. (2) The OECD average is a simple average. Soured: OECD (200la). 284

PAGE 299

Appendix E: Telecommunication Facilities Competition in OECD PSTN competition Local Trunk Inti. Year for Market Competition Australia C(3) C(2) C(3) Austria C(2) C(ll) C(I3) Belgium C(l1) C(l1) C(ll) Canada C(61) C(22) c Czech Republic MIC M M 2nd operators in 7 special areas. Open competition 2000. Denmark c c c Finland C(64) C(20) C(16) National long distance include 7 France C(23) C(l3) C(14) national operators and 6 multiregional. Germany c c c 2001; Consideration being given Greece M M M to 2000 for competition. 2002 Hungary M M2002 M2002 1.12.1998 Iceland M M M Ireland M M M Italy C(5) C(4) C(4) Local open in 1999 Japan C(5) C(15) C(21) Open from July 1998 Korea M C(3) C(3) Luxembourg c c c Mexico C(IO) C(l4) C(7) Netherlands C(l60) C(3) C(3) New Zealand C(3) C(7) C(l5) 2003 for long distance and Norway c c c international Poland D M M 2000 Portugal M M M Spain C(3) C(3) C(3) Sweden C(l5) C(l5) C(l5) 2006 Switzerland C(l2) C(l2) C(l2) Turkey M M M UK C(134) C(20+) C(7) us c c c Key: C:Competttion; D:Duopoly; M:Monopoly; numbers in parentheses mdtcate number of licensed operators. In a number of cases, all licensed operators are not yet active. For a number of countries licenses do not differentiate between local, national and international PSTN. Some licenses may be regional. Resellers are not included. Source: OECD, 1999a 285

PAGE 300

Appendix F: Major Telecommunications Operators in OECD Operator(s) Status Australia Telstra State owned:67% Optus Communications Privately owned Austria Post und Telekom Austria AG I 00% state owned through a holding company(25% + I share expected by end 1998) Belgium Belgacom State owned 51 %(strategic partnership with Ameritech, Singapore Telecom, Tele Da.rtinark 49%) Canada Stentor Members Privately owned Alternative long distance carriers Privately owned Czech Republic SPTTelecom State owned:51% Denmark Tele Danmark Privately owned Finland SoneraLtd. State owned Finnet Group(45 local telephone Privately owned companies) France France Telecom State owned:62% Germany Deutsche Telekom AG State owned:61% Greece OTE State owned:65% Hungary Hungarian Telecommunication State owned:6.47% Co. Iceland Telecom Iceland State owned. Corporatized as of 111197. No privatization before 2000. Ireland Telecom Eireann State owned 80% (KPN Netherlands & Tele Danmark 20% with an option for a further 15%) Italy Telecom Italia State owned:5% Japan Type I carriers(l) Minimum of33.3% state-ownership. Presently 65% state owned NIT KDD Privately owned 124 others Privately owned Type II carriers(2) Privately owned Korea Korea Telecom State owned:71.2% Luxembourg P&T Administration State owned Mexico Telefonos de Mexico Privately owned Nethelands KPN Telecom NV State owned:43.8% New Zealand Telecom New Zealand Privately owned Clear Communications Privately owned(25% owned by Television New Zealand -a State owned enterprise) Norway Telenor State owned 286

PAGE 301

Appendix F: (Cont.) Operator(s) Status Poland Telekomunikacja Polska Spolka State owned(privatization of20-25% Akcyjna(TPSA) shareS late 1998) Portugal Telecom Portugal State owned:25% Spain Telefonica Privately owned Retevison State owned:30%(fully privatized end 1998) Sweden Telia State owned Tele2(Netcom) Privately owned Switzerland Swisscom State owned(privatization to begin in 1998 but State will retain majority share) Turkey Turk Te1ekomunikasyon State owned United BT Privately owned Kingdom Cable & Wireless Privately owned Communications Kingston Telecom, Cable Privately owned Telephony and others United States Privately owned Local and InterArneritech Exchange Bell Atlantic(including Nynex) Carriers Bell South SBC(including Pacific Telesis) US West GTE and others Long Distance AT&T Privately owned and International Exchange Carriers Source: OECD, 1999a 287

PAGE 302

A lppen IX argmg s fl R .d I U tructure or es1 entia OECD sers m PSTN charging structure Local Internet Local Internet ISP fee telephony access telephony access includes unmetered local call(2) Australia Unmeterd(flat rate) Per call Per call Austria Metered Metered Seconds Seconds Belgium Metered Metered Seconds Seconds Canada Unrnetered Unrnetered None None Czech Rep. Metered Metered Units Seconds Denmark Metered Metered Seconds Seconds Finland Metered Metered Seconds Seconds France Metered Metered Seconds Seconds Yes Germany Metered Metered Units Seconds Greece Metered Metered Units Seconds Hungary Metered Unrnetered/ Seconds None/ Metered(3) Seconds Iceland Metered Metered Seconds Seconds Ireland Metered Metered Seconds Seconds Yes( off-peak) Italy Metered Metered Seconds Seconds Japan Metered Metered Units Units Korea Metered Metered(3) Units Units Luxembourg Metered Metered Units Seconds Mexico Unmetered(First I OOcalls Per call Per call free, then flat rate) Netherlands Metered Metered Seconds Seconds Yes New Zealand Unrnetered Unmetered None None Norway Metered Metered Seconds Seconds Poland Metered Metered Units Units Portugal Metered Metered Seconds Units Spain Metered Metered Seconds Seconds Yes Sweden Metered Metered Seconds Seconds Switzerland Metered Metered Units Seconds Yes Turkey Metered Metered Units Units United Metered Unmetered/ Seconds None/ Yes Kingdom Metered(3) Seconds United States Metered/Flat rate! Seconds/per calVnone Unmetered I I I. NTI offers a flat rate for late evemng for certain specified numbers for which subscribers pay a h1gher rental. The company is also expanding an unmetered option using ISDN 2.1n France for Worldonline users at off-peak times from IOL. 3. Users in Hungary and the U K have the option of metered or unmetered Internet access under the new tariff plans on offer. Korea telecom 's offer operates in major cities Source : OECD, 2000a 288

PAGE 303

Appendix H: Regulation on Universal Service (US) Existence Existence of Cost Cost Country of US funding Finding allocation Notes framework mechanism The costs of the US are Australia Yes Yes R M shared in proportion to carriers' shares of 'eligible revenue'. Austria Yes Yes Belgium Yes Yes R R Canada R R R R Czech No Republic . If it is proved that a deficit exists in the provision of US, the NT A will collect a Denmark Yes Yes R R contribution from fixed telephony company on the basis of the amount of turnover. Finland No . ART proposes the assessment of the cost of the France Yes Yes M(R) M(R) US and the level of operators' individual contributions to the Ministry. While there is a legal Germany Yes Yes R R provision for US, it has not been applied yet. Greece Yes Yes M(R) M(R) Regulator implements Ministry's decision. Hungary No . Direct subsidy from Iceland Yes No . government. Cross subsidy between services. Ireland No . Italy Yes Yes R R According to the NIT law, Japan Yes No . NIT's voice telephony service is regulated as universal service. Korea Yes Yes M M Luxembourg Yes Yes Mexico Yes No . Subsidy from access charges. 289

PAGE 304

Appendix H: (Cont.) Existence Existence of Cost Cost Country of US funding Finding allocation framework mechanism New Zealand See note ---Norway Yes No -Poland Yes No -Portugal Yes Yes R R Spain Yes Yes R R Sweden Yes No --Switzerland Yes Yes --Turkey Yes No -United Kingdom Yes No -United States Yes Yes R R .. Note: M Mm1stry, R-Independent telecommumcatiOns regulator. Source: OECD, 2000b 290 Notes The Kiwi Share Obligations are in effect a type of US requirement. Public disclosure of Kiwi Share costs are required form January 2000. Interconnection charges contribute to any such costs. Incumbent bears USO based on its license requirement. Establishment of the US is predicted in the draft of new telecommunication law. The criteria for the division ofthe net costs of universal service between operators and providers that are obliged to contribute are defined and published by ICP. Telefonica has been designated the dominant operator required to provide US until the end of2005. US being provided through a license condition on dominant carrier. US license granted on a periodic basis by tender Cross subsidy between services. US is an obligation on British Telecom and Kingston Telecom. Each telecom carrier that provides inter-state telecom services must contribute, on an equitable basis, to the provision of US

PAGE 305

Appendix I: Internet Hosts and International Digital Divide October October October October 1997 1998 1999 2000 Share of Internet Hosts(%) OECD Share 96.15 95.37 95.52 95.61 Non-OECD 3.85 4.63 4.48 4.39 Regional Distribution of Internet Hosts(%) North America 70.01 69.58 72.96 72.31 Europe 18.69 18.88 16.61 17.13 Asia 7.14 7.72 7.08 7.41 Oceania 3.01 2.58 2.05 1.89 South and Central America 0.66 0.84 0.98 1.01 Africa 0.49 0.40 0.33 0.25 Internet Hosts per 1000 inhabitants OECD 22.90 34.50 55.50 81.50 Non-OECD 0.21 0.38 0.59 0.85 Regional Penetration of Internet Hosts per 1000 Inhabitants North America 46.28 69.74 116.41 168.68 Oceania 26.81 34.76 43.84 59.16 Europe 6.13 9.45 13.41 20.22 Central and South America 0.48 0.91 1.67 2.53 Asia 0.53 0.87 1.28 1.96 Africa 0.17 0.21 0.28 0.31 (1) The geographical data include OECD countries in those regions. Source: Netsizer (www.netsizer.corn), OECD (2001a). 291

PAGE 306

Appendix J: Non-OECD Nations with Largest Internet Hosts Internet Hosts(OOO) Internet hosts per I 000 inhabitants Oct. Oct. Oct. Oct. Oct. Oct. Oct. Oct. 1997 1998 1999 2000 1997 1998 1999 2000 Taiwan 231.60 599.20 872AO 1169.20 10.65 27.32 39.42 52.83 Hong Kong 100.00 128.30 247.30 532.00 15.38 19.19 35.93 77.28 Brazil 93.10 186.40 317.10 519.40 0.58 1.12 1.89 3.09 Russian 70.60 139.80 200.90 267.20 0.48 0.95 1.36 1.82 Federation Israel 53.90 136.50 221.70 247.10 9.13 22.81 36.34 40.50 South 126.60 154.60 206.90 230.00 3.05 3.80 5.19 5.76 Africa Singapore 74.30 84.90 128.00 214.60 23.94 26.84 39.69 66.55 Argentina 17.50 48.30 126.00 200.10 0.49 1.34 3.44 5.47 China 20.10 22.30 53.60 109.20 0.02 0.02 0.04 0.09 Chile 17.10 34.10 54.10 69.90 1.17 2.30 3.60 4.65 Malaysia 30.80 40.20 58.90 66.10 1.42 1.85 2.70 3.03 Thailand 15.60 31.00 37.20 61.00 0.26 0.51 0.61 1.00 Colombia 15.40 23.70 44.70 48.40 0.41 0.60 1.08 1.16 India 6.30 14.90 32.50 47.20 0.01 0.02 0.03 0.05 Ukraine 19.10 24.30 37.40 42.80 0.37 0.48 0.74 0.84 Total of 892.00 1668.50 2638.70 3824.20 Above Total Non 1000.30 1845.10 2888.40 4150.20 OECD Top 15 as 89.2 90.4 91.4 92.1 share of NonOECD(%) Source: Netsizer (www.netSizer.com), OECD (200la). 292

PAGE 307

Appendix K: Broadband Access to Internet in OECD DSL Cable Modems Total Rank Rank User's (June (End per 100 2001) 2000) June Increase June Increase 2001 From 2001 From 2000 2000 Korea 4,205,813 53% 2,310,330 48% 13.91 I I Canada 702,267 51% 1,194,700 30% 6.22 2 2 Sweden 122,000 190% 78,400 24% 4.52 3 4 USA 3,334,491 37% 5,500,000 49% 3.24 4 3 Netherlands 97,000 547% 336,000 34% 2.74 5 6 Austria 68,800 79% 122,300 24% 2.36 6 5 Denmark 69,740 164% 54,000 32%. 2.33 7 8 Belgium 92,000 114% 140,264 37% 2.27 8 7 Iceland 4,764 143% 0 1.99 9 9 Luxembourg 6,920 0 0 1.60 10 NA Germany 780,000 290% 1.03 11 15 Japan 400,760 4,018% 784,000 25% 0.94 12 11 Switzerland 18,000 350% 46,000 70% 0.90 13 12 Finland 19,623 31% 18,000 0.73 14 10 France 177,000 177% 174,000 43% 0.59 15 16 Australia 27,000 170% 85,000 33% 0.59 16 13 Portugal 2,000 100% 55,358 120% 0.57 17 18 Norway 2,500 165% 20,500 25% 0.52 18 14 Spain 157,702 251% 25,664 91% 0.47 19 20 New Zealand 16,000 65% 1,267 93% 0.45 20 17 Italy 239,000 108% 0 0.44 21 19 UK 80,772 150% 83,750 325% 0.28 22 22 Czech Rep. 0 0 11,000 10% 0.11 23 21 Hungary 1,000 150% 7,800 160% 0.09 24 23 Poland 18,000 NA 10,000 NA 0.07 25 28 Mexico 0 0 0.02 26 24 Ireland 300 0 0 0.01 27 25 Greece 300 0 0 0 0.00 28 26 Turkey 0 0 0.00 NA NA Slovak Rep. 0 0 0 0 0.00 NA NA OECD 10,643,752 68% 11,138,333 44% 1.96 EU 1,913,157 200% 1,152,736 41% 0.82 Source : OECD (200 I b) 293

PAGE 308

Appendix L: Korean Plan for NKN-G. Stage I 1997): Groundwork Transmission Network Between 5 metropolitan areas: 622 Mbps-2 Gbps Between metropolitan areas and 7 large cities: 622 Mbps Between large cities and small-medium cities: 1550622 Mbps Switching Technology Data switching technology R&D and pilot project tests of A TM switching technology Services High-speed Local Area Networks One-stop delivery for selected government services Electronic library and museum service for selected areas Remote diagnosis service, distance learning Stage II Diffusion Transmission Network Between 5 metropolitan areas: 2.5 Gbps-tens of Gbps Between metropolitan areas and 7 large cities: 2.5 Gbps Switching Technology ATM switching technology Services Super-high-speed interconnection among LANs (above 155 Mbps) Extending one-stop delivery for government services Electronic library and museum service for extended areas Remote diagnosis service, distance learning Advanced services from geographic information system (GIS) Stage III Completion Transmission Network Networks for broadband multimedia services Automation of network operation and maintenance Switching Technology Enhanced switching enabling various multimedia services Services HDTV image information service Three-dimensional video conferencing service Super-computer application service Provision of government information by multimedia technology 294

PAGE 309

Appendix M: Korean Plan for NKN-P. Stage I (1995-1997): Groundwork Transmission Network Interconnection oflocal telephone stations with 155-622 Mbps network Switching Technology ATM multimedia services in metropolitan areas Subscriber Loop Fiber-optic cabling for large buildings with heavy traffic, Densely populated areas Services Video conferences Still-image picture phone services High-speed I resolution fax service Expansion ofiSDN services Stage II (1998-2002): Diffusion Transmission Network 2.5-10 Gbps synchronous networks Switching Technology Development of ATM test network in large cities Subscriber Loop Full-scale supply of fiber-optic cables in apartment complexes And heavily populated areas Services Commercial services using A TM decentralized switching network test services using ATM switching network Stage III (2003-2015): Completion Transmission Network Supply of optical transmission devices Switching Technology Integration of existing networks via ATM-based network Subscriber Loop Supply of fiber-optic cables to residential homes Services A TM-based multimedia services HDTV-level video exchange services 295

PAGE 310

Appendix N: Communications Service in Three Nations (1991-2000) united States (National Currency: Dollar, Area: 9,393,132 km2) Indicators Unit 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Population 10x6 253 255 258 261 263 265 268 270 273 275 Telephone lines 10x6 139 143 148 153 160 165 172 180 184 193 Telephone lines 55.18 56.13 57.39 58.88 60.73 62.18 64.37 66.54 67.30 69.97 per 100 % digital lines % 48.5 54.8 62.8 71.6 76.2 80.5 83.9 87.8 91.6 Cellular users 2.99 4.32 6.20 9.26 12.84 16.59 20.65 25.61 31.55 39.79 per 100 Annual telecom. 10x9 21 22 23 23 24 22 23 24 26 29 Investment PC 10x6 59 65 70 78 86 97 109 124 141 161 Internet hosts 10x6 0.5 0.9 I 3 6 10 21 30 53 81 Estimated 10x6 3 4 6 9 20 30 40 60 74 95 Internet users Internet users 1.18 1.76 2.13 3.26 7.60 11.30 14.93 22.20 27.17 34.65 per 100 Finland (National currency Markka Area 377 032 km2) Indicators Unit 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Population IOxJ 5029 5055 5078 5099 5177 5132 5147 5160 5165 5176 Telephone lines 10x3 2718 2742 2763 2801 2810 2842 2861 2841 2850 2848 Telephone lines 54.04 54.24 54.41 54.93 54.28 55.37 55.59 55.07 55.18 55.02 per 100 % digital lines % 42.4 50.5 62.1 76.8 89.8 100.0 100.0 100.0 100.0 100.0 Cellular users 6.35 7.64 9.63 13.25 20.07 29.27 42.02 55.15 63.38 72.04 per 100 Annual telecom. 10x6 2624 2344 2222 2898 3665 3899 4299 3934 4901 5885 Investment PC IOxJ 570 650 720 810 1200 1400 1600 1800 1860 2050 Internet hosts IOxJ II 19 33 68 216 314 487 460 462 529 Estimated IOxJ 70 95 130 250 710 860 1000 1311 1667 1927 Internet users Internet users 1.39 1.87 2.56 4.90 13.71 16.75 19.42 25.40 32.27 37.22 per 100 Korea (National currencyWon Area 98 447 km2) Indicators Unit 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Population 10x6 43.44 44.02 44.61 44.85 45.10 45.55 45.99 46.43 46.86 47.30 Telephone lines 10x6 14.57 15.59 16.68 17.64 18.60 19.60 20.42 20.08 20.51 21.93 Telephone lines 33.55 35.42 37.41 39.34 41.24 43.04 44.40 43.27 43.79 46.37 per 100 %digital lines % 46.2 53.6 58.8 61.8 63.4 65.2 66.7 68.2 73.9 80.9 Cellular users 0.38 0.62 1.06 2.14 3.64 6.98 14.96 30.19 50.03 56.69 per 100 Annual telecom. 10x9 2383 2396 2555 2920 3370 4698 7702 6277 8367 8783 Investment PC 10x6 1.96 2.50 3.05 3.87 4.85 5.99 6.93 7.83 8.51 11.25 Internet hosts IOxJ 2 4 9 18 29 66 131 203 461 398 Estimated 10x6 O.D2 0.04 0.11 0.13 0.36 0.73 1.63 3.10 10.86 19.04 Internet users Internet users 0.04 0.09 0.24 0.30 0.74 1.60 3.55 6.68 23.17 40.25 per 100 Source: ITIJ (200 I a) 296

PAGE 311

Appendix 0: Internet Diffusion in Three Nations (1991-2000) U.S. 40 35 30 25 20 -+-Internet users per 100 15 inhabitants 10 5 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Finland 40 35 30 25 20 --+-Internet users per 100 15 inhabitants 10 5 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Korea 45 40 35 30 25 --+-Internet users 20 per 100 15 inhabitants 10 5 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 297

PAGE 312

GLOSSARY AC Curve Average Cost Curve. ADSL Asymmetric Digital Subscriber Line. ARPA Advanced Research Project Agency. ARPANET Advanced Research Project Agency Network. AT&T American Telephone and Telegraph. CCR Model Community-Cultural-Ritual Model. ccTLDs Country Code Top Level Domains. CDMA Code Division Multiple Access. CIC Program Computing, Information and Communication Program. CIO Chieflnformation Officer. CPRs Common Pool Resources. CRTC Canadian Radio-Television and Telecommunications Commission. CyPRG Cyberspace Policy Research Group. DOD Department ofDefense (in the U.S.). DSL Digital Subscriber Line. DT Deutsche Telekom. EAL Economic Analysis of Law. ETRI Electronics and Telecommunications Research Institute (in Korea). 298

PAGE 313

FCC Federal Communications Commission (in the U.S.). FICORA Finnish Communications Regulatory Authority. FT France Telecom. GDP Gross Domestic Product. Gil Global Information Infrastructure. gTLDs Generic Top Level Domains. HPCC Program High Performance Computing and Communications Program. HTTP HyperText Transfer Protocol. IHS Interstate Highway System. IITF Information Infrastructure Task Force. IP Internet Protocol. ISPs Internet Service Providers IT Information Technology. ISC Internet Software. Consortium. ITU International Telecommunications Union. IXCs Interchange Carrier (after divestiture). KCC Korea Communications Commission. KII Korean Information Infrastructure. KT Korea Telecom. LECs Local Exchange Carrier (after divestiture). MFJ Modified Final Judgment (divestiture decree). 299

PAGE 314

MIS Management of Information Systems. MTC Ministry of Transport and Communications (in Finland). MIC Ministry of Information and Communication (in Korea). NGBT Negotiating Group on Basic Telecommunications. NGI Next Generation Internet. Nil National Information Infrastructure. NIIAC National Information Infrastructure Advisory Council (in the U.S.). NKN-G New Korea Net-Government. NKN-P New Korea Net-Public. NPRM Notice of Proposed Rulemaking. NPR National Performance Review. NSF National Science Foundation. NSFNET National Science Foundation Network. NTIA National Telecommunications and Information Administration (in U.S. Department of Commerce). NTS NonTraffic-Sensitive (the largest category oftelephone plant and associated expense) NW Network Wizards. OCLC Online Computer Library Corporation. OECD Organization for Economic Cooperation and Development. P&T (Finnish Governmental Administration of) Posts and Telegraphs. 300

PAGE 315

PFC Priority Foreign Country. PMIS Public Management of Information Systems. POTS Plain Old Telephone Service. PSTN Public Switched Telephone Network. PTOs Public Telecommunications Operators. PUCs Public Utilities Commissions (in the U.S.). R&D Research and Development. RBOCs Regional Bell Operating Companies. SITRA Finnish National Fund for Research and Development. SLC Subscriber Line Charge. SSL Secure Socket Layer. TAC Telecommunications Administration Center (in Finland). TCP Transmission Control Protocol. TCP/IP Transmission Control Protocol/Internet Protocol. TDX Time Division eXchange. TEKES Finnish National Technology Agency. TELRIC Total Element Long-Run Incremental Cost. TF Telecom Finland. Three-D (3-D) Three Dimension. Three-S (3-S) Three Stage. TIIAP Telecommunications and Information Infrastructure Assistance Program. 301

PAGE 316

TOP Technology Opportunity Program. TRS Service Trunked Radio System Service. UR Uruguay Round. USAC Universal Service Administration Company (in the U.S.). USF Universal Service Funding. USTR United States Trade Representative (in the U.S.). WTO World Trade Organization. WWW World Wide Web. 302

PAGE 317

REFERENCES Adams, D. and M. Hess. "Community in Public Policy: Fad or Foundation?" Australian Journal of Public Administration. 60(2): 13-23, June 2001. Adams, D and M. Hess. "National Competition Policy and (the) Public Interest." Briefing Paper No.3. National Center for Development Studies. Australian National University. 1999. Abramson, J. B. et al. The Electronic Commonwealth: The Impact of New Media Technologies on Democratic Politics. New York: Basic Books, 1988 Agranoff, R. and B. A. Radin. "The Comparative Case Study Approach in Public Administration." Research in Public Administration. No 1: 203-231, 1991. Albarran, A. B. and D. H. Goff. Understanding the Web: Social, Political, and Economic Dimensions of the Internet. Ames, Iowa: Iowa State University Press, 2000 Allen, D. ''New Telecommunications Services: Network Externalities and Critical Mass." Telecommunication Policy. 12: 257-271. 1988. Allison, Graham T. Essence of Decision: Explaining the Cuban Missile Crisis (2"d ed.), New York: Longman, 1999 Anderson, R. et al. Universal Access to E-mail-Feasibility and Societal Implication. Santa Monica, CA: Rand, 1995 Antonelli, C. "The Diffusion oflnformation Technology and The Demand for Telecommunications Services." Telecommunications Policy. 13: 255-264. 1989. Appleby, P.H. Policy and Administration. Tuscaloosa, AL: University of Alabama Press, 1949. Arthur, W. B. Increasing Returns and Path Dependency in the Economy. Ann Arbor, MI: University ofMichigan Press, 1994. Arthur, W. B. "Competing Technologies, Increasing Returns, and Lock-in by Historical Events". Economic Journal, 99(397):116-131, 1989. 303

PAGE 318

Ashford, D. E. History and Context in Comparative Public Policy. Pittsburgh, PA: University of Pittsburgh Press. 1992. Bailey, E. Economic Theory of Regulatory Constraint. Lexington: Lexington Books, 1973. Baker, R. Comparative Public Management: Putting US. Public Policy and Implementation in Context. Westport, CT: Praeger Publishers, 1994. Bar, F. et al. "Access and Innovation Policy for the Third-generation Internet." Telecommunications Policy 24: 489-518. 2000. Barlow, J.P. "A Declaration ofthe Independence ofCyberspace". 1996 (Available at: http://www.dff.org/pub/publications/John-Perry-Barlow/barlow-0296.declaration) Baron, D., and Besanko, D. "Regulation, Asymmetric Information and Auditing." Rand Journal of Economics 15: Winter, 1984. Baumol, W., et al. "Weak Invisible Hand Theorems on the Sustainability of Multi Product Natural Monopoly." American Economic Review 67, June 1977. Bell, Daniel. The Coming of Post Industrial Society. New York: Basic Books, 1973. Berg, S. and J. Tschirhart. Natural Monopoly Regulation: Principles and Practice. New York: Cambridge University Press, 1988. Bernard, H. Russell. Social Research Methods: Qualitative and Quantitative Approaches. Sage Publications, 1999. Blackman, C.R. Convergence between Telecommunications and Other Media: How should Regulation Adapt?" Telecommunications Policy Vol. 22 No. 3: 162-170, 1998. Blumler, Jay G. "The New Television Marketplace: Imperatives, Implications, Issues." In James Curran and M. Gurevitch. (Eds.) Mass Media and Society. London: Edward Arnold. 1991. Bobrow, Davis and John Dryzek. Policy Analysis By Design. Pittsburgh, PA: Univ. Of Pittsburgh Press, 1987. Bolter, Walter G. Telecommunications Policy for the 1990s and Beyond. Armonk, New York: M.E. Sharpe, Inc, 1990. 304

PAGE 319

Booze-Allen and Hamilton. Creating the Foundation for a World Class Telecommunications Service Industry in Korea. Seoul: KISDI. 1998. Brin, David. "The Internet as a Commons." Information Technology and Libraries. v14. n4: 240-3, Dec, 1995. Bressers, Hans and Mac Honigh. "A Comparative Approach to the Explanation of Policy Effects." International Social Science Journal v. 38 no. 2:267-287, 1986. Brock, Gerald W. Telecommunications Policy for the Information Age: from Monopoly to Competition. (3rd ed.) Cambridge, Massachusetts: Harvard University Press, 1998. Brock, Gerald W. Toward a Competitive Telecommunication Industry: Selected Papers from 1994. Mahwah, New Jersey: Lawrence Erlbaum Associates, Inc., 1995. Browing, J. "Universal Access (an idea whose time has passed)" Wired. 2 (09). 1994. Buchanan, J. M. and G. Tullock. The Calculus of Consent, Ann Arbor: University of Michgan Press, 1962. Bunge, F. M. South Korea: A Country Study. Washington, D.C.: Government Printing Office. 1982. Burgess, Philip M. and Florine P. Raitano. "The Other Digital Divide", Special Report ofthe Center for the New West, September 1999. Burk, Dan. "Federalism in Cyberspace", 28 Conn L Rev, 1095, 1996. Cairncross, Frances. The Death of Distance: How the Communications Revolution Will Change Our Lives. Cambridge, MA: Harvard Business School, 1997. Calabrese, A. and J. Burgelman. (eds.) Communication, Citizenship, and Social Policy: Rethinking the Limits of the Welfare State. Lanham, MD: Rowman and Littlefield Publishers, Inc., 1999. Calvert, C. "Regulation Cyberspace: Metaphor, Rhetoric, Reality, and the Framing of Legal Opinions". Comm/Ent: Hastings Communications and Entertainment Law Journal20(3): 541-566, Spring 1998. Canadian Radio-Television and Telecommunications Commission (CRTC). "Telecom Decision CRTC 97-8." Local Competition. Ottawa, Canada. 1997. 305

PAGE 320

Carey, J. W. Communication as Culture. Boston: Unwin Hyman, 1989. Castells, M. The Rise of the Network Society. Cambridge, MA: Blackwell Publishers, 1996 Castles, F. G. Comparative Public Policy: Patterns of Post-war Transformation. Massachusetts: Edward Elgar Publishing, Inc., 1998. Catinat, M. "The Reform of the Telecommunication Regulatory Frameworks: The American and European Approaches." HIIP Conference on The Impact ofthe Internet on Communications Policy. Harvard University. Dec, 3-5. 1997. Cave, Martin. "The Evolution ofTelecommunications Regulation in the UK." European Economic Review 41 : 691-699. 1997. Cawley, Richard. "Internet, Lies and Telephony," Telecommunications Policy, Vol.21, No.6, July 1997. Chatrie, Isabelle and Paul Wraight. Public Strategies for the Information Society in the Member States of the European Union. An ESIS Report for European Union. 2000. Cheng, Tun-Jen. "Political Regimes and Development Strategies: South Korea and Taiwan." In G. Gereffi and D. Wyman (eds.) Manufacturing Miracles. Princeton: Princeton University Press. 1990. Cherry, B. A. et al. Making Universal Service Policy: Enhancing the Process Through Multidisciplinary Evaluation. Mahwah, N.J.: Lawrence Erbbaum Associates, 1999. Cherry, BandS. S. Wildman. "Review of Federal Universal Service Policy in the United States." In Cherry, B. A. et al. Making Universal Service Policy: Enhancing the Process Through Multidisciplinary Evaluation. Mahwah, N.J.: Lawrence Erbbaum Associates, .1999. Cho, Shin et al. "Restructuring the Korean Telecommunications Market: Evolution and Challenges Ahead." Telecommunications Policy. Vol. 20 No.5: 357-373. 1996. Choi, B. "Political Economy ofTelecommunications Market Opening in Korea: Interplay of Domestic and Foreign Interest." Global Economic Review. 26 (2): 77. 1997. Chung, S. "Korean Response to the Delay of the WTO Basic Telecom Negotiation." Telecom. 8(9): 22-23. 1996. 306

PAGE 321

Cimatoribus, M. et al. "Impacts ofthe 1996 Telecommunications Act on the US Model ofTelecommunications Policy." Telecommunications Policy. V. 22. N.6: 493-517. 1998. Cisler, S. "Access Issues for the National Research and Education Network." In C Huff and T. Finholt (eds.) Social Issues in Computing: Putting Computing in its Place. McGraw-Hill, New York. 1984. Clark, David D. "Adding Service Discrimination to the Internet," Telecommunications Policy, Vol.20, No.3: 169-182, April 1996. Clegg, Alicia. "Telecommunications and the Internet," Telecommunications Policy, Vol.20, No.8: 545-548, October 1996. Clement, B. "The Impact of Convergence on Regulatory Policy in Europe". Telecommunications Policy Vol. 22 No.3: 197-205, 1998. Cole, D. C. and P. N. Lyman. Korean Development: The Interplay of Politics and Economics. Cambridge, Mass.: Harvard University Press. 1971. Coleman, James S. Foundations of Social Theory. Cambridge, Mass: Harvard University Press. 1990. Compaine, B. M. "Information Technology and Cultural Change: Toward a New Literacy?" In B. M. Compaine (ed.) Issues in New Information Technology. Ablex Norwood, 1988. Compaine, B. M. and M. J. Weinraub. "Universal Access to Online Services: An Examination ofthe Issue." Telecommunication Policy. Vol. 21 No.1: 15-33. 1997. Compaine, B. M. (ed.) The Digital Divide: Facing a Crisis or Creating a Myth? Cambridge, Massachusetts: The MIT Press. 2001. Cook, T. D. and D. T. Campbell. Quasi-experimentation: Design & Analysis Issues for Field Settings. Boston: Houghton Mifflin Company, 1979. Courville, L. et al. Economic Analysis ofTelecommunications. Amsterdam: North Holland, 1983. Crandall, R. W. After the Breakup: U.S. Telecommunications in a More Competitive Era. Washington, D.C.: Brookings, 1991. 307

PAGE 322

Crandall, R. W. and L. Waverman. Who Pays for Universal Service? When Telephone Subsidies Become Transparent. Washington, D.C.: Brookings Institution Press. 2000. Creswell, J. W. Research Design: Qualitative and Quantitative Approaches. Thousand Oaks: Sage Publications, Inc., 1994 Crew, M.A., and P.R. Kleindorfer. The Economics of Public Utility Regulation. Cambridge: MIT Press, 1986. Cruickshank, D. The Long Distance Competition Debate. Manotick, Ont.: Donald J. Cruickshank Associates, 1989. Cuilenburg, J. V. and Paul Slaa. "Competition and Innovation in Telecommunications: An Empirical Analysis of Innovative Telecommunications In The Public Interest" Telecommunications Policy Vol. 19 No.8: 647-663, 1995. Cuilenburg, J. V. and Paul Slaa. "Competition in the Local Loop: Towards an Anticyclical Competition Policy in Telecommunications". Telecommunications Policy Vol. 18 No.1: 51-65, 1994. Dahl, R.A., and C.E. Lindblom. Politics, Economics, and Welfare. NewYork: HarperCollins. 1953. Danielson, A. L., and D. R. Kamerschen. Current Issues in Public Utility Economics. Lexington, Mass.: Lexington Books, 1983. Davies, Stephen. The Diffusion of Process Innovation. New York: Cambridge University Press. 1979. Deaken, J. The Electronic Cottage. New York: Morrow, 1981 deLeon, Peter. Democracy and the Policy Sciences. Albany, NY: State University of New York Press, 1997. deLeon, Peter. "The Democratization of the Policy Sciences." Public Administration Review. Vol. 52, No.2.: 125-129, March/April 1992. de Leon, Peter. Development and Diffusion of the Nuclear Reactor: a Comparative Analysis. Cambridge, Mass.: Ballinger Publishing. Co., 1979. deLeon, Peter and Phyllis ResnickTerry. "Comparative Policy Analysis: Deja vu All Over Again?" Journal ofComparative Policy Analysis. No. 1:9-22. 1999. 308

PAGE 323

Denning, D. E. and H. S. Lin. Rights and Responsibilities of Participants in Networked Communications. Washington, D.C.: National Academy Press. 1994. Denzen, N. K. andY. S. Lincoln. Strategies of Qualitative Inquiry. Sage Publications, 1998. Dierkes, M. et al. Comparative Policy Research: Learningfrom Experience. Brookfield, Vermont: Gower Publishing Company, 1987. Dimock, M.E. A Philosophy of Administration. New York: Harper and Company, 1958. Dogan, Mattei and A. Kazancigil. (ed.) Comparing Nations: Concepts, Strategies, Substance. Cambridge, Massachusetts: Basil Blackwell, 1994. Dordick, H. S. The Emerging Network Marketplace. Abex, Norwood. 1981. Dordick, H. S. "The Origins of Universal Service: History as a Determinant of Telecommunication Policy." Telecommunications Policy. 14(6): 223-31. 1990. Drake, William J. The New Information Infrastructure: Strategies for U.S. Policy. New York: The Twentieth Century Fund Press, 1995. Dryzek, J. S. Deliberative Democracy and Beyond: Liberals, Critics, Contestations. New York: Oxford University Press. 2000. Dye, Thomas R. Understanding Public Policy. (9th ed.) Upper Saddle River, NJ: Prentice Hall, 1998. Easton, D. A Framework/or Political Analysis. Englewood Cliffs, NJ: Prentice-Hall, 1965. Easton, D. A Systems Analysis of Political Life. University of Chicago Press, 1979. Eisner, Marc Allen. Regulatory Politics in Transition. Baltimore, MD: John Hopkins University Press, 1993. Erikson, R. et al. The Scandinavian Model: Welfare States and Welfare Research. New York: M. E. Sharpe, Inc. 1987. Esping-Andersen, G. The Three Worlds of Welfare Capitalism. Princeton, New Jersey: Princeton University Press. 1990. 309

PAGE 324

Etzioni, Amitai. The Spirit of Community: The Reinvention of American Society. New York: Touchstone. 1994. Everard, Jerry. Virtual States: the Internet and the Boundaries of the Nation-State. New York: Routledge, 2000. Fazio, Dennis. "Hang on to Your Packets: The Information Superhighway Heads to Valley Fair". Technical Report, Minnesota Regional Network, 1995. (Available at: http://www.mr.net/announcements/valleyfair.htinl) Feldstein, M. Economic Policy During the 1980's. Chicago: University of Chicago Press, 1994 .. Feldstein, M. S. "Equity and Efficiency in Public Sector Pricing: The Optimal Two Part Tariff." Quarterly Journal of Economics 86, May 1972. Fischer & Lorenz. Internet and the Future Policy Framework for Telecommunications. A Report for the European Commission, 31 January 2000. Fletcher, A. L. "Markets and States in the Information Society: A Comparative Institutional Analysis of Telecommunications Policy in the United States and France." Ph.D. Diss. University of Georgia, 1997. Flynn, R. and P. Preston. "The Long-run Diffusion and Techno-economic Performance ofNational Telephone Networks: A Case Study oflreland, 1922-1998". Telecommunications Policy. V. 23: 437-457, 1999. Francis, J. The Politics of Regulation: A Comparative Perspective. Cambridge, Massachusetts: Blackwell Publishers, 1993 Friedland, L. A. "Electronic Democracy and the New Citizenship" Media, Culture & Society. Vol. 18: 185-212, 1996 Friedman, Milton. Essays in Positive Economics. Chicago: University of Chicago Press. 1953. Friedman, Milton. Market Mechanisms and Central Economic Planning. Washington, D.C.: American Enterprise Institute for Public Policy Research. 1981. Friedman, Milton. Capitalism and Freedom. Chicago: University of Chicago Press. 1982. 310

PAGE 325

Friedman, Milton. Why Government is the Problem. Stanford, Calif.: Hoover Institution on War, Revolution, and Peace, Stanford University. 1993. Froomkin, Michael, "The Internet as a Source ofRegulatory Arbitrage", In Kahin, B. and C. Nesson (eds.) Borders in Cyberspace. MIT Press, 1997. (Available at: http://www.law.miami.edu/-froomkin/articles/arbitr.htm) Froomkin, A. Michael and J. Bradford De Long, "The Next Economy?" 1997. (Available at: http://www.law.miami.edu/-froomkin/articles/newecon.htm) Gackenbach, Jayne. Psychology and the Internet: Intrapersonal, Interpersonal, and Transpersonal Implications. San Diego: Academic Press, 1998. Garcia-Murillo, M. A. and I. Macinnes. "FCC Organizational Structure and Regulatory Convergence." Telecommunications Policy. Vol. 25: 431-452. 2001. Garnham, N. "Information Society as Theory or Ideology: A Critical Perspective on Technology, Education and Employment in the Information Society." Information, Communication & Society. 3.2: 139-152, 2000. Ghertman, M. et al. "Regulation and Transaction Costs in Telecommunications: A Research Agenda". Telecommunications Policy Vol. 19 No.6: 487-500, 1995. Gibson, Robert. Technology & Change. 1999. (Available at: http://www .naruc.org/technolo.htm) Giese, Mark. "From ARPA to the Internet: A Cultural Clash and Its Implications in Framing the Debate on the Information Superhighway." In Lance Strate et al. (eds.) Communications and Cyberspace: Social Interaction in an Electronic Environment. Cresskill, NJ: Hampton Press. 1996. Gillett, S.E. and I. Vogelsang (eds.) Competition, Regulation, and Convergence: Current Trends in Telecommunications Policy Research. Mahwah, NJ: Lawrence Erlbaum Associates, 1999. Goodnow, F.J. Comparative Administrative Law. New York: G.P. Putman's Sons, 1893. Goodwyn. L. Populist Movement. New York: Oxford University Press, 1976. Gore, A. Earth in the Balance. Boston: Houghton Mifflin, 1992. Gore, A. "The Gil: Conditions for Success." Intermedia. 23(2): 48. 1995. 311

PAGE 326

Green, J.R. and Teece. "Four Approaches to Telecommunications Deregulation and Competition: The U.S., U.K., Australia, and New Zealand." Center for Research on Telecommunications Policy (CRTP). Working Paper Series 49. Feb, 1999. Gramlich, E. M. "Infrastructure Investment: A Review Essay." Journal of Economic Literature. 32:1176-96. 1994. Greve, Bent. (ed.) Comparative Welfare Systems: The Scandinavian Model in a Period ofChange. New York: St. Martin's Press, Inc., 1996. Gromov, G. R. "The Roads and Crossroads oflntemet History." 1996. Available at http://www.hooked.net Grossman, Wendy. Net. Wars. New York: New York University Press, 1997. Guillen, M. F. and S. L. Suarez. "Developing the Internet: Enterpreneurship and Public Policy in Ireland, Singapore, Argentina, and Spain." Telecommunications Policy. V.25: 349-371, 2001. Gulick, L.H. "Politics, Administration and the New Deal" Ann Am Acad Politic Soc Sci 169:55-66, 1933. Gurbaxani, V. "Diffusion of Computing Networks: The Case of Bitnet". Communications of the ACMV.33, N.12: 65-75, Dec. 1990. Gurr, T. R. and K. Jaggers. Polity 98 Project: Regime Characteristics. 1800-1998. 1998. Available at http://www.bsos.umd.edu/cidcm/polity. Gutierrez, L. H. and S. Berg. "Telecommunications Liberalization and Regulatory Governance: Lessons from Latin America." Telecommunications Policy. V.23: 865884,2000. Hacker, K. "Missing Links in the Evolution of Electronic Democratization." Media, Culture & Society. V. 18: 213-232, 1996. Hallgren, M. J. "Funding an Internet Public Good: Definition and Example". Computer Networks and ISDN Systems. INET '94, Prague, V.27: 403-409, 1994. Hahm, Pyong-choon. "Toward a New Theory of Korean Politics." In Korean Jurisprudence, Politics and Culture. Seoul: Yonsei University Press. 1986. 312

PAGE 327

Hahm, Sung D. and L. C. Plein. After Development: The Transformation of the Korean Presidency and Bureaucracy. Washington, D.C.: Georgetown University Press. 1997. Hallgren, Martyne M. and Alan K. McAdams. 'The Economic Efficiency oflnternet Public Goods", McKnight, L., and J, Bailey, eds., Internet Economics, Cambridge, Massachusetts; The MIT Press, 1999 Hamilton, G. G. and N. W. Biggart. "Market, Culture, and Authority: A Comparative Analysis of Management and Organization in the Far East." American Journal of Sociology. V.94 Supplement S52-S94, 1988. Hargittai, E. Holes in the Net: The Internet and International Stratification. Senior Honors Thesis. Smith College. 1996 (A vail able at Hargittai, E. Holes in the Net: The Internet and International Stratification Revisited Proceedings of the Internet Society's Internet Society Summit Meeting, 1998 (Available at http://www.isoc.org/inet98/proceedings/5d/5d l.html) Hargittai, E. "Weaving the Western Web: Explaining Differences in Internet Connectivity among OECD Countries." Telecommunications Policy. V. 23: 701-718, 1999 Haring, J. R. and D.L. Weisman." Dominance, Non-dominance, and the Public Interest in Telecommunications Regulation" Telecommunications Policy pp98-1 06, March 1993. Harris, R. G. and C. 1. Kraft. "Meddling Through: Regulating Local Telephone Competition in the United States." Journal of Economic Perspectives. V. 11 N.4: 93112, Fall, 1997. Hausman, J., et al. "The Effects ofthe Breakup of AT&T on Telephone Penetration in the United States." American Economic Review 83, May 1993. Hayek, Friedrich A. von. Economic Freedom. Cambridge, Mass.: B. Blackwell, 1991. Hayek, Friedrich A. von. The Fatal Conceit: the Errors of Socialism. Chicago: University of Chicago Press. 1989. Head, J. G. Public Goods and Public Welfare. Durham, North Carolina: Duke University Press, 1974. 313

PAGE 328

Heady, F. Public Administration: A Comparative Perspective (6th). New York: Marcel Dekker, 200 I. Heidenheimer, Arnold J. et al. Comparative Public Policy: The Politics of Social Choice in America, Europe, and Japan. (3rd ed.) New York: St. Martin's Press, 1990. Heikkila, M. and H. Uusitalo (eds.). The Cost of the Cuts: Studies on Cutbacks in Social Security and their Effects in the Finland of the 1990s. Helsinki: Stakes. 1997. Henderson, Gregory. Korea: The Politics of the Vortex. Cambridge, Mass.: Harvard University Press. 1968. Henisz, W.J. "The Institutional Environment for Economic Growth." Economics and Politics. 12,1-31, 2000. Henten, A. and T. M. Kristensen. "Information Society Visions in the Nordic Countries." Telematics and Informatics 17: 77-103. 2000. Hersen, M and D. H. Barlow. Single-case Experimental Designs: Strategies for Studying Behavior. New York: Pergamon. 1976. High, Jack (ed.) Regulation: Economic Theory and History. Ann Arbor, Michigan: The University ofMichigan Press, 1991. Hill, J. "Liberalization, Regulation, and Development: Telecommunications" Gazette 60(6): 459-476, Dec. 1998 Hindle, John, et al. (eds), The Internet as Paradigm, the Eighth Edition of the Annual Review of the Institute for Information Studies. Queenston, MD: The Aspen Institute, 1997. Hong, Sung Gut. "The Political Economy of the Korean Telecommunications Reform." Telecommunications Policy. Vol. 22 No.8: 697-711. 1998. Huber, Peter. Law and Disorder in Cyberspace :Abolish the FCC and Let Common Law Rule the Telecosm. New York: Oxford University Press, 1997. Hudson, Heather. "Universal Service in the Information Age" Telecommunications Policy 18(8):658-667, 1994. 314

PAGE 329

Hyun, D.aiwon and John A. Lent. "Korean Telecom Policy in Global Competition: Implications for Developing Countries." Telecommunications Policy. Vol. 23: 389-401. 1999. Inagaki, N. and R. Mahajan. "Emerging Issues in the International Telecommunications and Information Environment: Research Topics." TIPI International White Paper. 1999. Information Infrastructure Task Force (IITF). The National Information Infrastructure: Agenda for Action. Washington, D.C.: U.S. Department of Commerce. Sep. 15, 1993. Ingram, Helen, and Anne Schneider. "The Choice of Target Populations." Administration & Society. Vol. 23, No.3: 333-356, November 1991. ITU. Challenges to the Network: Telecoms and Internet. Genava, 1997. ITU. Challenges to the Network: Internet for Development. Geneva, 1999 ITU. Telecommunications at a Glance. Geneva, 2000. ITU. Year Book of Statistics: Telecommunications Services 1991-2000. Geneva, 2001a. ITU. World Telecommunications Indicators. Geneva, 2001b. ITU .IP Telephony and the Internet: Republic of Korea Case Study. Geneva, 2001 c. Available at http://www .itu.int/wtpf/casestudies/korea.pdf Jayakar, K. P. "Industry Structure, Regulatory Choices and The Diffusion of Telecommunications Services: What Government Can Do To Further Universal Service." Ph.D. Diss. Indiana University, 1999. Jenkins-Smith, Hank C. Democratic Politics and Policy Analysis. Pacific Grove, CA : Brooks/Cole, 1990. Jenning, E. T. Jr. "Public Choice and the Privatization of Government: Implications for Public Administration". In Ott, J. S. and A. C. Hyde and J. M. Shafritz, (eds.) Public Management: Essential Readings, Chicago: Nelson-Hall Inc., 1991. Jeong, Kuh-Hwan and J. L. King. "Korea's National Information Infrastructure: Vision and Issues." In Kahin, Brian and E. J. Wilson III. National Information Infrastructure Initiatives: Vision and Policy Design. Cambridge, Massachusetts: The MIT Press, 1997. 315

PAGE 330

Johnson, David R. and David G. Post. "Law and Borders: The Rise of Law in Cyberspace", 48 Stanford L Rev, 1367, 1996. Johnson, Dawn L. "It's 1996: Do You Know Where Your Cyberkids Are? Captive Audiences and Content Regulation on the Internet", 15 J. Marshall J. Computer & Info. L. 51, 86, 1996. Johnson, Chalmers. Miti and the Japanese Miracle. Stanford, Calif.: Stanford University Press. 1982. Joseph, Richard A. "Converging Telecommunications Technologies: Challenges facing Government and Regulators in Australia and New Zealand." Telecommunications Policy. Sep/Oct. 1993. Jussila, 0. et al. From Grand Duchy to a Modern State: A Political History of Finland since 1809. Carbondale, IL: Southern Illinois University Press. 1999. Kahin, Brian. "The Internet Business and Policy Landscape" In Hindle, John, et al. (Eds), The Internet as Paradigm, the eighth edition of the Annual Review of the Institute for Information Studies. Queenston, MD: The Aspen Institute, 1997. Kahin, Brian. "The U.S. National Information Infrastructure Initiative: The Market, the Web, and the Virtual Project." In Kahin, Brian and E. J. Wilson III. National Information Infrastructure Initiatives: Vision and Policy Design. Cambridge, Massachusetts: The MIT Press, 1997. Kahin, Brian and E. J. Wilson III. National Information Infrastructure Initiatives: Vision and Policy Design. Cambridge, Massachusetts: The MIT Press, 1997. Kahin, Brian and J. Keller (eds.). Public Access to the Internet. Cambridge, Massachusetts: The MIT Press. 1996. Kahn, A. E. The Economics of Regulation: Vol I Principles; Vol IL Institutional Issues. New York: Wiley, 1990. Kamarck, E. C. and J. S. Nye, Jr. (eds.) democracy.com: Governance in a Networked World Hollis Publishing, 1999. Kantor, A. and M. Neuharth. "Offthe Charts" Internet World. Dec. 1996. Katz, Michael L. "Ongoing Reform of U.S. Telecommunications Policy." European Economic Review. 41: 681-690. 1997. 316

PAGE 331

Katz, Raul L. The Information Society: An International Perspective. New York: Praeger, 1988. Kellerman, A. "Fusions of Information Types, Media and Operator, and Continued American Leadership in Telecommunications." Telecommunications Policy. Vol.21, No.6.: 553-564. 1997. Kelly, T. and Petrazzini, B. "What does the Internet mean for Development?" Telecom Interactive Development Symposium. Geneva, 11 September 1997. Kenda, Michael. "The Digital Handshake: Connecting Internet Backbones". Federal Communications Commission Office ofPians and Policy, OPP Working Paper series #32,Sep.2000. Kerf, M and D. Geradin. "Controlling Market Power in Telecommunications: Antitrust vs. Sector-specific Regulation-An Assessment of the United States, New Zealand, and Australian Experiences." Berkeley Technology Law Journal v.l4 i3:919 (53). Fall, 1999. Ketti, Donald F. Sharing Power: Public Governance and Private Markets. Washington D.C.: The Brookings Institution, 1993. Kiesler, Sara (ed.) Culture of the Internet. Mahwah, New Jersey: Lawrence Erlbaum Associates, 1997. Kim, JaeYoung. "Universal Service and Free Speech on the Internet after Telecommunications Deregulation: Experiences of the United States, Germany, and Japan." Ph.D. Diss. Southern lllinois University at Carbondale, 1999. Kolgerg, Jon Eivend. (ed.) The Study of Welfare State Regimes. New York: M. E. Sharpe, Inc. 1992. Kretzmann, J.P. and J. L. Mcknight. Building Communities from the Inside Out: A Path toward Finding and Mobilizing a Community's Assets. M.A.: Polity/Basil Blackwell, 1993. Kubicek, H. et al. The Social Shaping of Information Superhighways: European and American Roads to the Information Society. New York: St. Martin's Press, 1997. 317

PAGE 332

Kubicek, H. and William H. Dutton. "The Social Shaping oflnformation Superhighways: An Introduction." In Kubicek, H. et.al. The Social Shaping of Information Superhighways: European and American Roads to the Information Society. New York: St. Martin's Press, 1997. Kuhnle, Stein. "The Scandinavian Welfare State in the 1990s: Challenged but Viable." West European Politics. v23. i2: 209 (19). Apr., 2000. Kumar, Ranjit. Research Methodology: A Step-by-Step Guide for Beginners. Sage Publications, 1999. Kurxweil, Ray. The Age of Spiritual Machines. New York: Penguin Putnam Inc., 1999. Kutais, B. G. Internet Policies and Issues. New York: Nova Science Publishers, Inc., 1999. Kvist, J. "Welfare Reform in the Nordic Countries In the 1990s: Using Fuzzy-set Theory to access Conformity to Ideal Types." Journal of European Social Policy. V. 9 (3): 231-252, 1999. Laffont, J. J. and J. Tirole. "The Politics of Government Decision Making: A Theory of Regulatory Capture." Quarterly Journal of Economics 106, November 1991. Lafontaine, F. "Agency Theory and Franchising: Some Empirical Results." Rand Journal of Economics 23, Summer 1992. Laponce, J. A. Languages and their Territories. Toronto: University of Toronto Press, 1987 Larson, James F. The Telecommunications Revolution in Korea. New York: Oxford University Press. 1995. Lee, H. Y. and Lee L. C. "Korea's Telecom Services Reform Through Trade Negotiation." Telecom Policy Issue. Seoul: KISDI, 2001. Lee, Sang Hyup. "Three Essays on Telecommunications Regulatory Policies." Ph.D. Diss. University of Florida, 1999. 318

PAGE 333

Lehr, W. and Kiessling, T. "Telecommunication Regulation in the U.S. and Europe: The Case for Centralized Authority". In Gillett, S.E. and I. Vogelsang (eds.) Competition, Regulation, and Convergence: Current Trends in Telecommunications Policy Research. Mahwah, NJ: Lawrence Erlbaum Associates, 1999. Lenert, Edward M. "A Communication Theory Perspective on Telecommunications Policy". Journal of communication, Autumn 1998. Lessig, Lawrence. Code and Other Laws ofCyberspace, New York, NY; Basic Books, 1999. Levin, S.L. and J. B. Meisel. "The Internet: Regulatory Arbitrariness All Over Again". Paper presented at Harvard Conference of 3-5 Dec 1997 on The Impact of Internet on Communications Policy (available at: http://www .ksg.harvard.edu/iip/iicompol/papers/Levin.html). Levy, B. and P.T. Spiller. "The Institutional Foundations of Regulatory Commitment: A Comparative Analysis of Telecommunications Regulation". Journal of Law, Economics and Organizaation. 1 0(2), 201-246, 1994. Lijphart, A. "The Comparable-Cases Strategy in Comparative Research. Comparative political Studies, no. 8: 158-177, 1975. Lin, Ning. "Diffusion of Information Technology: A Case Study of Computer Network and the Role of Government, Industry, and Academia in Developing the Intemet/NREN." Ph.D. Diss. University of Texas at Austin, 1993. Lim, Phillip W. "The Asian Values Debate Revisited: Positive and Normative Dimensions." Korea Journal. Pp 365-384, Summer 2000. Lim, Youngil. Technology and Productivity: The Korean Way of Learning and Catching Up. Cambridge, Mass.: The MIT Press. 1999. Little, Ian M.D. "An Economic Reconanissance." In Walter Galenson. (ed.) Economic Growth and Structural Change in Taiwan. Ithaca, N.Y.: Cornell University Press. 1979. Littlechild, S.C. Elements ofTelecommunications Economics. Peter Peregrinus. Steven age. UK, 1979. Lowenstein, Ralph L. and J. C. Merill. Macromedia: Mission, Message, and Morality. New York: Longman. 1990. 319

PAGE 334

Lowi, T. J. "Four Systems of Policy, Politics and Choice." Public Administration Review. Vol. 32:298-310, 1972. Lundvall, Bengt-Ake. (ed.) National Systems oflnnovation: Toward a Theory of Innovation and Internet Learning. N.Y.: St. Martin's Press, Inc., 1992 Lyytinen, Kalle and Seymour Goodman. "Finland: The Unknown Soldier on the IT Front." Communications of the ACM. Vol. 42 No.3: 13-17. March, 1999. Mack, R. L. The Digital Divide: Standing at the Intersection of Race & Technology. Durham, NC: Carolina Academic Press. 2001. Mackie-Masen, Jeffrey K. Telecommunications: Guide to the Internet. Rockville, Maryland : Government Institutes, Inc., 1999. Mackie-Mason, J. and H. Varian. "Some Economic FAQs About the Internet," Journal of Electronic Publishing, June 1995. Mackie-Mason, J. and H. Varian. Some Economics of the Internet, University of Michigan, 1993. Mahler, A. and E. M. Rogers. "The Diffusion oflnteractive Communication Innovations and The Critical Mass: The Adoption of Telecommunications Services by German Banks". Telecommunications Policy 23: 719-740. 1999. Majone, G. (ed.) Deregulation or Reregulation?: Regulatory Reform in Europe and in the United States. New York: St. Martin's Press, 1990. Manishin, G. B. "Regulation ofthe Communications Functions ofthe Internet". Papers Presented at Fall Internet World 1996. (Available at: http://www .technologylaw .com/iworld/iworld 1.html) Mann, C., Global Electronic Commerce: A Policy Primer. Washington, DC: Institute for International Economics, 2000 Manning, N. and I. Shaw. "The Transferablility of Welfare Models: A Comparison of the Scandinavian and State Socialist Models in Relation to Finland and Estonia." Social Policy and Administration. Vol. 32, No. 5: 572-590. 1998. Maor, Moshe and Jan-Erik Lane. (ed.) Comparative Public Administration: Volume I (Analytical Frameworks and Critique) Volume II (Comparative Research). Brookfield, Vermont: Ashgate Publishing Company, 1999. 320

PAGE 335

Margolis, Michael. Politics as Usual: the Cyberspace Revolution. Thousand Oaks: Sage Publication, 2000. Martenson, G. "The impending Dismantling of European Union Telecommunications Regulation." Telecommunications Policy. Vol 22 No. 9: 729-738. 1998. Martins-Filho, C., and Mayo, J. W. "Demand and Pricing ofTelecommunications Services: Evidence and Welfare Implications." Rand Journal of Economics 24, Autumn 1993. Mason, E. S. et al., The Economic and Social Modernization of the Republic of Korea. Cambridge, Mass.: Council ofEast Asian Studies, Harvard University. 1980. Mathios, A. D., and Rogers, R. P. "The Impact of Alternative Forms of State Regulation of AT&T on Direct-Dial, Long-Distance Telephone Rates." Rand Journal of Economics 20, Autumn, 1989. Mayer, Lawrence C. Redefining Comparative Politics. Newbury Park, Cal.: Sage, 1989. Mayers, Ramon H. "The Economic Transformation of the Republic of China on Taiwan." China Quarterly. 99:500-528. 1984 McChesney. R. W. "The Internet and U.S. Communication Policy-making in Historical and Critical Perspectives". Journal of Communication. 46( 1 ): 98-125, 1996. McCloskey, Matthew J. "Bibliography of Internet SelfRegulation," Internet Law and Policy Forum, 1998. (Available at: http://www.ilpf.org/selfreg/bib4-15.htm) McKnight, J. L. The Careless Society: Community and its Counterfeits. New York: Basic Books, 1995. McKnight, Lee W. and J. Bailey. (eds.) Internet Economics, Cambridge, Massachusetts; The MIT Press, 1999. McKnight, Lee W. and J. Bailey. (eds.) "Special Issue on Internet Economics," Journal of Electronic Publishing, Fall 1995. McKnight, Lee W. and Brett Leida. "Internet Telephony and Open Communications Policy", Papers presented at Harvard Conference of 3-5 Dec 1997 on the Impact of Internet on Communications Policy, (Available at: http://www.ksg.harvard.edu/iip/iicompol/Papers/McKnight.html) 321

PAGE 336

McKnight, Lee W. and W. R. Neuman. "Technology Policy and the National Information Infrastructure." In Drake, William J. (ed.) The New Information Infrastructure: Strategies for US. Policy. New York: The Twentieth Century Fund Press, 1995. Megginson, W. L. and J. M. Netter. 'From State to Market: A Survey of Empirical Studies on Privatization." Available at http://www.faculty staff.ou.edu/M/William.L.Megginson-1/. Melody, W. H. "On the Meaning and Importance oflndependence in Telecom Reform." Telecommunications Policy. Vol. 21 No.3: 195-199. 1997. Metcalfe, B. "Beware of Mandated Access for solving the Problem of Information have-nots" Infoworld 17(20): 49, 1995. Meyers, S. J. "Universal Access? Take the High Road" Data Communications 23 (9), 1994. Milgrom, P., and Roberts, J. "Informational Asymmetries, Strategic Behavior and Industrial Organization." American Economic Review 77, May 1987. Miller, S. I. and M. Fredericks. Qualitative Research Methods: Social Epistemology and Practical Inquiry. New York, NY: Peter Lang Publishing Inc., 1994. Milward, H. B. and L. 0. Snyder. "Electronic Government: Linking Citizens to Public Organizations Through Technology." Journal of Public Administration Research and Theory. 6 (1996): 2: 261-275. Ministry oflnformation and Communication (MIC). Korea Internet White Paper. Seoul: MIC, 2001 Ministry oflnformation and Communication (MIC). The Information and Communication White Paper. Seoul: MIC, 2000 Ministry of Finance. Finland Towards the Information Society-A National Strategy. Helsinki: Ministry ofFinace. 1994. Mishra, R. The Welfare State in Crisis: Social Thought and Social Change. Wheatsheaf, Brighton. 1984. Mjoset, L. "The Nordic Model never Existed, but does It have a Future?" Scandinavian Studies. V.64 N4: 652 (20), Fall, 1992. 322

PAGE 337

MOSAIC Group. The Global Diffusion of the Internet Project: An Initial Inductive Study. Arlington, VA. Mar. 1998. Available at http://www.agsd.com/gdi97/gdi97.html Moschella, D. Waves of Power: The Dynamics of Global Leadership. New York: American Management Association, 1997. Moschovitis, Christos J.P. et al. History of the Internet: a Chronology, 1843 to the Present, Santa Barbara,CA: ABO-CLIO, Inc., 1999. Mueller, D. C. Public Choice. Cambridge: Cambridge University Press, 1979. Mueller, Milton. "Telecommunications Access in the Age of Electronic Commerce: Toward a Third-generation Universal Service Policy." Federal Communication Law Journal. V.49 N.3: 655-673. 1997. Nabseth, Land Ray, G. F. The Diffusion of New Industrial Process: An International Study. Cambridge University Press. 1974. Naisbitt, J. Megatrends. New York: Warner Publications, 1982 Napoli, P. M: "The Unique Nature of Communications Regulation: Evidence and Implications for Communications Policy Analysis" Journal of Broadcasting & Electronic Media 43(4): 565-581, Fall 1999. Nattermann, P.M. and D. Murphy. "The Finnish Telecommunications Market: Advantage of Local Access Incumbency". Telecommunications Policy. Vol. 22. No.9: 757-773, 1998. Neuman, R. W. et al. "The Politics of a Paradigm Shift: Telecommunications Regulation and the Communication Revolution." Political Communication. v10 n1: 77(18), Jan-March 1993. Niskanen, W. A. Bureaucracy and Representative Government, Chicago: Aldine Atherton, 1971. Noam, E. "Reconnecting Communication Study with Communications Policy". Journal ofCommunication. 43(3): 199-206, 1993. Noam, E. Beyond Liberalization: From the Network ofNetworks to the System of Systems." Telecommunications Policy. 18(4): 286-294, 1994 323

PAGE 338

Noam, E." Beyond Liberalization II: The Impending Doom of Common Carriage." Telecommunications Policy. 18(6): 435-452, 1994 Noam, E. Beyond Liberalization III: Reforming Universal Service." Telecommunications Policy. 18(9): 687-704, 1994 Nolan, R. and C.F. Gilson. "Managing the Four Stages ofEDP Growth," Harvard Business Review, January/February. 1974. Noll, A. M. "The Costs of Competition: FCC Telecommunication Orders of 1997 ." Telecommunications Policy. V.22. N.l: 47-56, 1998. Noll, R.G. (ed.) Regulatory Policy and the Social Sciences. Berkeley: University of California Press, 1985. Novak. T.P. and D.L. Hoffman. "Bridging the Digital Divide: The Impact of Race on Computer Access and Internet Use." Science. V. 280: 390-391, 1998 NTIA (National Telecommunications and Information Administration of U.S. Department of Commerce). "Falling through the Net: Defining the Digital Divide". July 1999. (Available at: http://www .ntia.doc.gov/ntiahome/fttn99/contents.html) Nyaw, Mee-kou, and Chan-leong Chan. "Structure and Development Strategies of the Manufacturing Industries in Singapore and Hong Kong: A Comparative Study." Asian Survey. 22:449-69. 1982. OECD. Universal Service and Rate Restructuring in Telecommunications. Paris: OECD, 1991. OECD. ICCP Reviews of Information and communications Policy: Finland. Paris: OECD, 1992. OECD. Communications Outlook. Paris: OECD, 1997a. OECD. Regulatory Reform in Telecommunications Service. Paris: OECD, 1997b. OECD. Internet Infrastructure Indicators. Paris: OECD, 1998a. OECD. Internet Traffic Exchange: Developments and Policy. Paris: OECD, 1998b. OECD. Regulatory Reform in the Telecommunications Industry: The United States. Paris: OECD, 1998c 324

PAGE 339

OECD. Measuring the ICT Sector. Paris: OECD, 1998d. OECD. Communications Outlook. Paris: OECD, 1999a. OECD. Internet Indicators: Online Component of Communications Outlook 1999. Paris: OECD, 1999b. OECD. Local Access Pricing and E-commerce. Paris: OECD, 2000a. OECD. Telecommunications Regulation: Institutional Structures and Responsibilities. Paris: OECD, 2000b OECD. Science, Technology and Industry Outlook. Paris: OECD, 2000c. OECD. Knowledge-based Industries in Asia. Paris: OECD, 2000d. OECD. The OECD Review of Regulatory Reform in Korea. Paris: OECD, 2000e. OECD. Information Technology Outlook. Paris: OECD, 2000f. OECD. Communications Outlook. Paris: OECD, 200la. OECD. The Development of Broadband Access in OECD Countries. Paris: OECD, 200lb. Ogden, M. R. "Politics in a Parallel Universe: Is There a Future for Cyberdemocracy?" Futures 26(7): 713-729, 1994 Ohno, Kenichi. "Overview: Creating the Market Economy." InK. Ohno and I. Ohno. (eds.) Japanese Views on Economic Development. Routledge, London. 1998. Okun, A.M. Equality and Efficiency: The Big Trade Off Washington, DC.: Brookings Institute, 1974. Omohundro, John T. Chinese Merchant Families in Iloilo. Athens: Ohio University Press. 1981. Osborne, David and Ted Gaebler. How the Entrepreneurial Spirit Is Transforming the Public Sector, from Schoolhouse to Statehouse, City Hall to Pentagon. Reading, MA: AddisonWesley. 1992. 325

PAGE 340

Ostas, D. T. "Postmodern Economic Analysis of Law: Extending the Pragmatic Vision of Richard A. Posner" American Business Law Journal. V .36 i I: 193, Fall 1998. Ostrom, Elinor. Governing the Commons: the Evolution of Institutions for Collective Action, New York: Cambridge University Press, 1990. Ostrom, Elinor. "Institutional Rational Choice: An Assessment of the Institutional Analysis and Development Framework." in Paul A. Sabatier (ed.), Theories of the Policy Process, Boulder, CO: Westview Press, 1999. Ostrom, V. and E. Ostrom. "Public Goods and Public Choices." In E.S. Savas, (ed.) Alternatives of Delivering Public Services: Toward Improved Performance. Boulder: Westview Press. 1977. Ouchi, William G. Theory Z: How American Business can meet the Japanese Challenge. Reading, Mass.: Addison-Wasley, 1981. Oxman, Jason. "The FCC and the Unregulation of the Internet." Federal Communications Commission OPP Working Paper series #31, July 1999. Paltridge, S. "How Competition helps the Internet." The OECD Observer. No 201: 25, August-September 1996. Papert, S. "Obsolete Skill Set: The 3 Rs." Wired 1993. Available at http://www .hotwired.com/wired/1.2/departments/idees.fortes/papert.html Parks, M.R. and K. Floyd. "Making Friends in Cyberspace" Journal of Communication, 46(1): 80-97, 1996. Parsons, T. "The Professions and Social Structure." Social Forces no. 17:457-67, 1939. Patton, M. Q. How to use Qualitative Methods in Evaluation. Newbury Park, CA: Sage, 1987. Peltzman, S. "The Economic Theory of Regulation After a Decade of Deregulation." Brookings Papers on Economic Activity-Microeconomics, 1989. Peltzman, S. "Toward a More General Theory of Regulation." Journal of Law and Economics 19, August 1976. Pestoff, V. A. Beyond the Market and The State: Social Enterprises and Civil Democracy in a Welfare Society. Ashgate. Aldershot. 1998. 326

PAGE 341

Peters, B. Guy. Comparative Politics: Theory and Methods. New York: New York University Press, 1998. Peterson, H. C. "An Empirical Test of Regulatory Effects." Bell Journal of Economics 6, Autumn 1975. Philip, G. and Hing Tsoi. "Regulation and Deregulation of Telecommunications: The Economic and Political Realities. Part I: The United States. Part II: The U.K. and other West European Countries. Part III: Japan, the Pacific Basin and Less Developed Conuntries." Journal of Information Science. V.l4 N.5:257 (8). Sep. 1988. Polinsky, A.M. An Introduction to Law and Economics. Boston: Little, Brown, 1983. Pool, Ithiel de Sola. Technologies of Freedom. Cambridge, MA: Belknap Press. 1983. Porter, Michael E. The Competitive Advantage of Nations. New York: The Free Press. 1990. Posner, R. A. "The Social Costs of Monopoly and Regulation." Journal of Political Economy 83, August 1975. Posner, R. A. "Theories of Economic Regulation." Bell Journal of Economics and Management Science 5, Autumn 1974. Posner, R. A. Economic Analysis of Law. Little, Brown and Company. 1992. Pospischil, R. "Fast Internet: An Analysis about Capacities, Price Structure, and Government Intervention" Telecommunications Policy. 22(9): 745-756, Oct. 1998. Press, L. "Notes on a Framework to Characterize the Global Diffusion of the Internet." INFO '97, Havana, Oct. 1997. Available at http://www.som.csudh.edu/fac/lpress/articles/fmwpres.htm Press, L. et al. "An Internet Diffusion Framework". Communications of the ACM 41(10): 21-26, 1998 Prieger, J. "Universal Service and the Telecommunications Act of 1996: The Fact after the Act." Telecommunications Policy. Vol. 22 No.1: 57-71. 1998. Przeworski, Adam." Methods of Cross-national Research: 1970-83: an Overview." In Dierkes, Metal. Comparative Policy Research: Learning from Experience. Brookfield, Vermont: Gower Publishing Company, 1987. 327

PAGE 342

Putnam, Robert D. Making Democracy Work: Civic Traditions in Modern Italy, Princeton, New Jersey: Princeton University Press, 1993. Putnam, Robert D. "Bowling Alone: America's Declining Social Capital", Journal of Democracy, 6(1): 65-78, 1994a. Putnam, Robert D. "Bowling Alone: Democracy in America at the End of the Twentieth Century." Nobel Symposium in Uppsala, Sweden. (August) 1994b. Putnam, Robert D. "Tuning In, Tuning Out: The Strange Disappearance of Social Capital in Ameica" PS: Political Science & Politics, Vol.28, No.4, December 1995. Ragin, C. C. The Comparative Method: Moving Beyond Qualitative and Quantitative Strategies. Berkeley, Cal.: University of California Press, 1987. Rai, A. et al. "How to Anticipate the Internet's Global Diffusion." Communications of the ACM. v.41. n.10: 97-106, Oct. 1998 Rainey, Hal G. Understanding and Managing Public Organization (2"d ed.) San Francisco, CA: Jossey-Bass Inc., Publishers, 1997 Redding, S. C. "Cognition as an Aspect of Culture and Its Relation to Management Processes: An Exploratory View of the Chinese Case." Journal of Management Studies. 17: 127-48, 1980. Rheingold, H. The Virtual Community: Homesteading on the Electronic Frontier. Reading, MA: Addison-Wesley, 1993 Richardson, J. Policy Styles in Western Europe. Winchester, Mass.: Allen & Unwin. 1982. Riggs, Fred W. Administration in Developing Countries: The Theory of Prismatic Society. Boston: Houghton Mifflin Company. 1964 Riggs, Fred W. "Agraria and Industria-Toward a Typology of Comparative Administration." In William J. Siffin, (ed.) Toward the Comparative Study of Public Administration. Bloomington: Indiana University Press. 1957 Riis, A.M. "The Information Welfare Society: An Assessment of Danish Government Initiatives Preparing for the Information Society." Paper presented in Gil Conference. Harvard University. March 1996. 328

PAGE 343

Rodger, J. J. From ci Welfare State to a Welfare Society: The Changing Context of Social Policy In a Postmodern Era. New York: St. Martin's Press, Inc., 2000. Rogers, E.M. Diffusion of Innovation. New York: The Free Press, 1983 Rose, Richard. Lesson-drawing in Public Policy: A Guide to Learning across Time and Space. Chatham, NJ: Chatham House Publishers, Inc., 1993. Rosenbloom, D. H. Public Administration-Understanding Management, Politics, and Law in the Public Sector. (4th ed.), New York: Random House, 1998 Rutkowski, T. "Internet Hosts". 1996. Available at http://www.intemetvalley.com Sabatier, Paul A. (ed.) Theories of the Policy Process, Boulder, CO: Westview Press, 1999. Sabatier, Paul A. "Towards Better Theories of the Policy Process", PS: Political Science and Politics, 24(2), (June) 147-156. Sahal, Devendra. Patterns of Technological Innovation. Massachusetts: Addison Wesley Publishing Co. 1981. SaKong, II. "Macroeconomic Aspect of the Public Enterprise Sector." In C. K. Park. (ed.) Macroeconomic and Industrial Development in Korea. Seoul: Korea Development Institute. 1980. Sandfort, Sand D. Frissell. "The Big Lie of Universal Access Subsidies." Wired 3(05) 1995. Sartori, G. "Compare Why and How: Comparing, Miscomparing and the Comparative Method." In Mattei Dogan & A. Kazancigil. (ed.) Comparing Nations: Concepts, Strategies, Substance. Cambridge, Massachusetts: Basil Blackwell, 1994. Saunders, R. J. et al. Telecommunications and Economic Development. Baltimore, Maryland: The Johns Hopkins University Press,1983. Schechter, P. B. "Customer Ownership ofthe Local Loop: A Solution to the Problem of Interconnection." Telecommunications Policy. Vol. 20 No. 8:573-584, 1996. Schmandt, Jurgen et al. (eds.) Telecommunications Policy and Economic development: the New State Role. New York, N.Y.: Praeger, 1989. 329

PAGE 344

Schneider, V. "Different Roads to the Infonnation Society? Comparing U.S. and European Approaches from a Public Policy Perspective." In H. Kubicek et al. The Social Shaping of Information Superhighways: European and American Roads to the Information Society. New York: St. Martin's Press, 1997. Schrage, M. "Let Them Read Books" Los Angeles Times Syndicate, Los Angeles 1994. Shapiro, Andrew L. The Control Revolution: How the Internet is Putting Individuals in Charge and Changing the World We Know, New York, NY; PublicAffairs, 1999. Shapiro and Varian. Information Rules. Harvard Business School Press, 1999. Sherman, R. The Regulation of Monopoly. Cambridge: Cambridge University Press, 1989. Shew, W. "Infonnation Superhighways: Is There a Role for Government?" The American Enterprise v.4: 18-22, July/Aug. 1993. Shim, J.K., et al. The International Handbook of Electronic Commerce. Chicago. CA: Glenlake Publishing Company, 2000 Shin, R. T. and J. S. Ying. "Unnatural Monopolies in Local Telephone." Rand Journal of Economics 23, Summer 1992. Sibley, D. "Asymmetric Infonnation. Incentives and Price Cap Regulation." Rand Journal of Economics 20, Autumn 1989. Sim, S. P. and S. Rudkin. "The Internet-Past, Present and Future." BT Techno/ JV. 15 N. 2: 11-23. 1997. Sorgaard, Pal. IT Co-Ordination and Public Management Reform: A Comparison between Finland and Norway. Helsinki: Ministry of Finance. 2000. Spacek, Thomas R. "Internet Evolution: some Communications Policy Implications and Guiding Principles for a new Policy Framework." 1999. (Available at: http://www.ksg.harvard.edu/iip/iicompol/Papers/spacek.htm) Spiller, P. "Politicians, Interest Groups and Regulators: A Multiple-Principals Agency Theory of Regulation or, 'Let Them Be Bribed'." Journal of Law and Economics 33, April1990. Spulber, D. Regulation and Markets. Cambridge: MIT Press, 1989. 330

PAGE 345

Statistics Finland. On the Road to the Finnish Information Society. Helsinki: University Press. 1997. Statistics Finland. On the Road to the Finnish Information Society II. Helsinki: University Press. 1999. Statistics Finland. On the Road to the Finnish Information Society Ill. Helsinki: University Press. 2001. Steinbock, Dan. The Competitive Advantage of Finland: From Cartels to Competition. Helsinki: ETLA SITRA. 1998. Steinbock, Dan. "Assessing Finland's Wireless Valley: Can the Pioneering Continue?" Telecommunications Policy. V.25: 71-100. 2001a. Steinbock, Dan. The Nokia Revolution. New York: Amacom. 200 I b. Steinmo, S. et al. Structuring Politics: Historical Institutionalism in Comparative Analysis. New York: Cambridge University Press, 1992. Stigler, G. "The Theory of Economic Regulation." Bell Journal of Economics and Management Science 2, Spring 1971. Stigler, G. and C. Friedland. "What Can Regulators Regulate?" Journal of Law and Economics 5, October 1962. Stigler, J. E. and S. Yusuf. Rethinking the East Asian Miracle. Oxford University Press. 2000. Stillman, Richard J. II. The American Bureaucracy: The Core of Modern Government. (2"d ed.) Chicago: Nelson-Hall Publishers. 1996. Stone, Deborah. Policy Paradox: The Art of Political Decision Making.(2"d ed.) New York: W.W. Norton & Co., 1997. Swan, P. "The International Diffusion of an Innovation". Journal of Industrial Economics. September 1973. Tapscott, D. The Digital Economy: Promise and Peril in the Age of Networked Intelligence. New York: McGraw-Hill, 1997. 331

PAGE 346

Tapscott, D. et al. (eds.) Blueprint to the Digital Economy: Wealth Creation in the Era of E-Business. New York: McGraw-Hill, 1998. Taylor, Lester. Telecommunications Demand: in Theory and Practice. Norwell, MA: Kluwer Academic Publishers, 1994. Taylor, W. E., and L. Taylor. "Postdivestiture Long-Distance Competition in the U.S." American Economic Review 83, May 1993. Thomas. "The Politics of Growth: The German Telephone System." In M. Renate & T.P. Hughes. The Development of Large Technical System. Boulder, CO: Westview Press, 1988 Thompson, Frederick. Regulatory Policy and Practices: Regulating Better and Regulating Less. New York, N.Y.:Praeger, 1982. Thompson, Maryann Jones. "The Demographics of Who's Online." The American Standard, March 8, 1999. Train, K. E. Optimal Regulation: The Economic Theory of Natural Monopoly. Cambridge: MIT Press, 1991. Trebilcock, M., et al. Law and Economics of Competition Policy. Vancouver: The Fraser Institute, 1990. Trochim, W. "Outcome Pattern Matching and Program Theory." Evaluation and Program Planning. No. 12: 355-366, 1989. Tufte, E. R. Data Analysis for Politics and Policy. Engiewood Cliffs, N.J.: Prentice Hall, 1975. Turpeinen, 0. Telecommunications since 1796: Telecom Finland Ltd. Helsinki: Edita. 1997. Tyler, M. and S. Bednarczyk. "Regulatory Institutions Processes in Telecommunications: An International Study of Alternatives." Telecommunications Policy. Pp650-676. Dec., 1993. UNDP. Human Development Report. New York: Oxford University Press, 1998 United States Trade Representative. 1995 Trade Policy Agenda and 1994 Annual Report of the President of the United States on the Trade Agreements Program. 1995. 332

PAGE 347

United States Trade Representative. 1997 Annual Report of the President of the United States on the Trade Agreements Program. 1998. Uusitalo, Hannu. Income Distribution in Finland: The Effects of the Welfare State and the Structural Changes in Society on Income Distribution in Finland in 1966-1985. Helsinki: Central Statistical Office of Finland, Studies No. 148. 1989. Varian, Hal R. "Economic Issues Facing The Internet" In Hindle, John, et al. (Eds), The Internet as Paradigm, the eighth edition of the Annual Review ofthe Institute for Information Studies. Queenston, MD: The Aspen Institute, 1997. Venturelli, S. S. "Political Foundations of the Information Society in the European Union: Public Space in the Information Age." Ph.D. Diss. University of Colorado, 1995. Verba, S. "Comparative Politics: Where Have We Been, Where Are We Going", in Wiarda, H. J. (ed.) New Directions in Comparative Politics. Boulder, CO: Westview, 1986. Verhoest, Pascal. "The Myth of Universal Service: Hermeneutic Considerations and Political Recommendations." Media, Culture & Society. Vol22:595-610. 2000. Vickers, J., and G. Yarrow. "Economic Perspectives on Privatization." Journal of Economic Perspectives 5, Spring 1991. Vincent Ostrom and Elinor Ostrom. "Public Goods and Public Choices". In E.S. Savas, Alternatives for Delivering Public Services: Toward Improved Performance. Boulder: Westview Press, 1977. Visco, Lgnazio. "Structural Reform in Korea after the 1997 Economic Crisis The Agenda and The Implementation." Paper presented in Conference on Economic Crisis and Restructuring in Korea. Seoul. Dec., 1999. Wade, R. Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization. Princeton: Princeton University Press. 1990. Waldo, Dwight. The Administrative State: A Study of the Political Theory of American Public Administration, New York: Ronald Press, 1948 Waldon, T. "The Industry Analysis Division's Reference Book of Rates, Price Indices, and Expenditures for Telephone Service." FCC Report. July 1998. 333

PAGE 348

Wallace, Patricia. The psychology of the Internet. Cambridge, N.Y.: Cambridge University Press, 1999. Walsh, K. Public Services and Market Mechanism: Competition, Contracting and the New Public Management, Macmillan, 1995. Wamsley, G. L. and M. N. Zald. The Political Economy of Public Organization. Lexington, Mass.: Heath, 1973. Weber, M. "Objective Possibility and Adequate Causation in Historical Explanation" in The Methodology ofSocial Sciences. New York: Free Press. 1949. Weill, P. and Marianne Broadbent. Leveraging the New Infrastructure: How Market Leaders Capitalize on Information Technology. Cambridge, MA: Harvard Business School Press, 1998. Weimer, D. and A. Vining. Policy Analysis: Concept and Practice. Upper Saddle River, NJ: Prentice-Hall, Inc., 1998. Weinberg, Jonathan. "Telecommunications Service, Access Charges, Universal Service Mechanisms and Other Flotsam of the Regulatory System." In Gillett, S. E. and I. Vogelsang (eds.) Competition, Regulation and Convergence: Current Trends in Telcommunications Policy Research. Mahwah, New Jersey: Lawrence Erlbaum. 1999. Weisbrod, B. A. The Nonprofit Economy. Cambridge Mass.: Harvard University Press, 1988. Weisman, D.L. "Asymmetrical Regulation: Principles for Emerging Competition in Local Service Market" Telecommunications Policy Vol. 18 No.7: 499-505, 1994. Wellenius, Bjorn et al. Telecommunication: World Bank Experience and Strategy. Discussion Paper 192. Washington: World Bank. 1992. Wenders, J. T. The Economics ofTelecommunications: Theory and Policy. Cambridge, Mass.: Ballinger, 1987. Werbach, Kevin. "Digital Tornado: The Internet and Telecommunications Policy", Federal Communications Commission Office of Plans and Policy, OPP Working Paper series#29, March, 1997. Westphal, H. and E. Towell. "Investigating the Future oflnternet Regulation". Internet Research. 8( 1 ): 2631, 1998. 334

PAGE 349

Whinston, A. B. et al. The Economics of Electronic Commerce. Indianapolis, IN.: Macmillan Technical Publishing, 1997. Wiarda, H. J. New Directions in Comparative Politics. Boulder, CO: Westview, 1986. Williamson, 0. Markets and Hierarchies New York: The Free Press, 1975. Willoughby, W.F. Principles of Public Administration. Baltimore, MD: Johns Hopkins University Press, 1927. Wilson, James Q. The Politics of Regulation. New York: Basic Books, 1980. Winston, C. "Economic Deregulation: Days of Reckoning for Microeconomists." Journal of Economic Literature 31, September 1993. Wolf, Charles. Markets or Government: Choosing between Imperfect Alternatives. Cambridge, Massachusetts: The MIT Press. 1993. World Bank. East Asian Miracle. New York: Oxford University Press. 1993. World Bank. East Asia: The Road to Recovery. Washington, D.C. 1999. Xavier, Patrick. "Australia's Post-July 1997 Telecommunications Regulation in International Context." Telecommunications Policy. V.21 N. 6:533-551. 1997. Yin, R. K. Case Study Research: Design and Methods. (2nd ed.) Thousand Oaks: Sage Publications, Inc., 1994. Yoon, Chang-Ho. "Liberalization Policy, Industry Structure and Productivity Changes in Korea's Telecommunications Industry." Telecommunications Policy Vol. 23:289306. 1999. Ypsilanti, D. and P. Xavier. "Towards Next Generation Regulation" Telecommunications Policy. 22(8): 643-659, Sep. 1998. Zelezny, John D. Communications Law: Liberties, Restraints, and the Modern Media. (2nd ed.) Ed. Belmont, CA: Wadsworth Publishing. 1997. Zuboff, Shoshana. In the Age of the Smart Machine: The Future of Work and Power. New York: Basic Books, Inc., 1988. 335