|
|
|
Email this citationThe Emergence of a Supranational Telecommunications Regime
Wayne Sandholtz and Alex Stone Sweet *
Political Relations and Institutions Research Group
Working Paper 2.43
November 1996
Abstract
Had we been writing this volume even ten years ago, telecommunications could have provided a case study under the heading "why EC policy-making remains in the states." The Commission's efforts to bring telecoms to Brussels had by 1986 yielded meager results. Telecommunications in Europe was a patchwork of national systems, dominated in each country by a state monopoly, the PTT. 1 Furthermore, national governments had an array of reasons for wanting to retain control of their PTTs: they were frequently an instrument of industrial and employment policies, as well as a generator of revenues that could be siphoned off for other state purposes. Nevertheless, by 1996 new EU rules were abolishing the PTT monopolies and putting in place a regime to provide for EU-wide markets for equipment, services and infrastructure. 2 The aim of this chapter is to provide a theory-driven account of these developments.
This essay speaks to one of the central theoretical controversies in the study of integration. Intergovernmentalist theory holds that supranational organizations and transnational actors exercise no independent impact on EU politics. Governments, in this view, sometimes delegate certain functions to the Commission or the European Court of Justice, which remain on a short leash. EU outcomes are the result of lowest-common denominator bargaining among the member states. Our theory of institutionalization denies these assertions. We argue that supranational organizations and transnational actors do exert an autonomous influence on EU politics. By tracing the process by which an EU telecoms regime came into being, this essay can determine the extent to which member states controlled policy outcomes, versus the degree to which the Commission, the Court, and transnational actors drove developments. The evidence is clear: the initiatives for these changes all came from the Commission, acting in alliance with various societal groups that had a stake in efficient, advanced, pan-European telecommunications. Rulings of the European Court of Justice provided legal precedents that the Commission could exploit. The Commission also attacked the PTT monopolies with Article 90 directives (that do not require Council approval), an innovation that the ECJ upheld. In other words, movement toward supranational politics occurred along all three of our dimensions (rules, supranational organizations, transnational society). Furthermore, shifts toward supranationality in one dimension tended to produce or reinforce parallel shifts in the other two.
The first section of the chapter summarizes the volume's theoretical framework as I will use it to structure the telecommunications case study. The second section looks at the "transactors" (societal groups), explaining why the PTT system harmed their interests and why European integration was their preferred solution. The third section describes the intergovenmental starting oint, in tems of the PTT system andthe nature of member state preferences. The fourth and crucial section is an account of how theinterested groups and the Commission exploited existing ECrules and institutions to integrate telecoms policy-making. It documents movement along thedimensions in two phases of telecoms policy integration. A brief conclusion closes the chapter.
The core empirical proposition of this volume is that the European Union has transformed itself from a largely intergovernmental arrangement (the founding treaties) into a supranational polity. The intergovernmental-supranational continuum is a way to make that claim more nuanced by recognizing: (1) that the distinction invoked is not a dichotomy, and (2) that specific policy domains can be located at different points along the continuum. The three dimensions (rules, supranational organizations, transnational society) suggest ways of identifying and charting movement along the continuum. We suggested, for example, that as the EU becomes more supranational, we should expect to see more informal norms as well as formal rules (legislation, jurisprudence), more domains added to the competences of EU bodies, increasing autonomy of EU organizations, and the expansion of linkages among societal actors across borders and between EU bodies and societal actors.
We also suggested that there is a logic of institutionalization: once supranational rules, organizations, and transnational society have begun to emerge, they become causal variables, pushing the EU further toward the supranational pole. Supranational rules and organizations provide the context for determining interests and behaviors thereafter. Subsequent behaviors, policy debates, and disputes take place in a context in which the range of possibilities has been shaped by past practices, legislation, and judicial precedent.
What we have, then, is a feedback loop. The question then becomes, what initiates such a loop of institutional development? We argued, taking our cue from Deutsch and Haas, that societal actors engaged in cross-border communications and exchange ("transactors") are the motor driving integration. These are the people who need European standards, rules, dispute resolution mechanisms -- who need integration. The causal mechanism is quite simple: increasing levels of cross-border transactions and communications by societal actors will increase the perceived need for supranational policy-making. In fact, the absence of European rules and coordination inhibits transactors from realizing their values. As Walter Mattli argues, state actors have their own interests; two fundamental ones are to maximize their own autonomy and to maximize the generation of wealth within their territory. In defense of their autonomy, states will frequently resist the shift toward supranational policy-making. But as they do so, they inhibit the generation of wealth within their territories by those actors that depend on European transactions. Maximizing autonomy therefore has its cost in prosperity. Since political leaders must generate prosperity to satisfy their constituencies, blocking integration for more than the short run is unsustainable. 3
Thus, though states remain powerful actors in a supranational context, commanding important resources, they do not ultimately control the integration process. Societal transactors provide the impetus for integration. And once the dynamic logic of institutionalization is underway, it shapes the context in which actors -- EU bodies, national governments, and non-state actors -- define and pursue their interests. States, in fact, are embedded in supranational politics that they can influence but neither dictate nor control.
The integration of telecommunications policy-making in the EU provides a case in which the plausibility of most of these ideas can be probed. The role of societal transactors will be played by those groups whose well-being depends on cross-border communications and exchange within Europe: telecoms users, and providers of equipment, services and infrastructures. We would expect to find that rising levels of exchange across national borders (trade, finance, production, mergers and acquisitions) increases the demand for state-of-the-art, pan-European telecommunications services, which in turn leads to demands by transactors for policy integration. We would expect transactors to exploit whatever opportunities European institutions afforded. We would further expect European organizations like the Commission and the Court to use existing rules in creative ways so as to advance integration. Member states will frequently oppose or obstruct telecoms policy integration, but they can be outflanked. The result should be movement toward supranational policy-making: new European rules, European organizations, and an expanding transnational society in telecommunications.
In fact, the telecommunications case confirms the expectations generated by our theory. By themselves, single cases do not make or break theories. They can, however, serve useful purposes in theory building and testing. Case studies are ideal for "process tracing," which allows us to verify that the apparent empirical relationships we observe are not merely correlations but do in fact embody the causal links our theory generates. We find in the telecommunications case both the outcomes and the causal relationships that our theory posits.
An explanation of the transition from state monopolies to supranational policy-making begins with the societal actors for whom the PTT monopolies were increasingly costly and problematic. Our theoretical framework sees societal actors engaged in cross-border communications and transactions as the primary engine of integration. The key to the story in telecommunications is this: as the level of cross-border telecommunications rose, Europe's patchwork of national monopolies became increasingly dysfunctional. The expansion of intra European economic activity -- exchange of goods and services, finance and investment, management of multinational firms -- entailed an increase in telecoms traffic among EU countries.
Telecommunications networks are the essential infrastructure of the modern economy. Trade depends on communication between buyers and sellers. But the importance of telecoms goes well beyond business-people using their phones and faxes. Cross-border services -- banking, investment, airline reservations systems -- require constant, real-time data communication. Multinational firms coordinating and managing their affiliates in different member states often need permanently open voice and lines. Moreover, technological change has created new and more demanding kinds of communication: high-volume data and graphics, as well as various combinations of text, data, sound, and images. In other words, as economic integration progressed, Europe saw a steady increase in the number of actors who needed to communicate across borders -- for trade, investment, finance, production coordination, and so on -- and those actors required new and more demanding kinds of intra-European telecommunications.
Figure 1 shows that cross-border calling in the EU has grown faster than both population and intra-EU trade for 1984-94 (telephone traffic data are unavailable for the period before 1983). One would expect telephone traffic to increase with the number of potential callers; in fact, cross-border calling grew much faster than population. Growth in intra-EU calling seems to track the growth of trade, though usually at a higher rate and with less dramatic swings. If we take intra-EU trade as a rough measure of economic integration, my argument would imply a strong correlation between intra-Union commerce and intra-EU cross-border telecommunications traffic. As Table I shows, the correlation between trade and telecoms for 1984-94 is very strong indeed (.95), substantially stronger than the correlation between telecoms and GDP or population (though both are also high). A pooled cross-sectional time series analysis (see Table 2) with telephone traffic as the dependent variable reinforces the close link between economic integration and cross-border telecoms. The multivariate regression produces a highly significant coefficient for intra-EU trade. The coefficient for population is also significant, but with the "wrong" sign, probably because the trade effect overwhelms population. The data analyses are consistent with my proposition that economic integration in the EU leads to increased demand for cross-border telecommunications.
In the following sub-sections I, first, briefly explain the technological revolution in telecoms to show how it opened potentially vast markets for new telecoms services and equipment; and, second, describe those actors whose emerging telecommunications needs put them at odds with the PTT system.
telephone traffic and trade, GDP, and population
| Intra-EU cross-border telephone traffic |
|
| Intra-EU trade | .95 |
| GDP | .85 |
| Population | .76 |
SOURCES:Telephone data come from the International Telecommunications Union. Trade is the annual value of both imports and exports within the European Union. Trade data are from Eurostat: External Trade 1958-1993. GDP data are from The World Tables 1970-1994, published by the World Bank. Population data are from The World Tables of Economic and Social Indicators, ICP SR #9300, The World Bank, second release. The years covered in the model are 1983-92. The countries are Belgium, Denmark, France, Germany, Greece, Ireland, Italy, Netherlands, Portugal, Spain, and the UK. N = 109.
|
Intra-EU cross-border telephone traffic |
||
| Constant | 43.32 | |
| (55.62) | ||
| Intra-EC Trade | .00449*** | |
| (.00051) | ||
| GDP | .08789* | |
| (.03549) | ||
| Population | -.0099** | |
| (.00344) | ||
| Adjusted R2 | .899 | |
| n | 109 |
Note: Entries are unstandardized regression coefficients with standard errors in parentheses. Each of the 109 observations is for a particular country in a particular year. The model is a random-effects GLS regression model and was estimated using Stata 3.01. The dependent variable is millions of minutes in phone calls to other EU nations per country per year.*p< .01
**p< .005
***p< .001
SOURCES: Telephone data come from the International Telecommunications Union. Trade is the annual value of both imports and exports within the European Union. Trade data are from Eurostat: External Trade 1958-1993. GDP data are from The World Tables 1970-1994, published by the World Bank. Population data are from The World Tables of Economic and Social Indicators, ICP SR #9300, The World Bank, second release. The years covered in the model are 1983-92. The countries are Belgium, Denmark, France, Germany, Greece, Ireland, Italy, Netherlands, Portugal, Spain, and the UK. We are missing data for Denmark in 1992 and thus the total n is 109.
2.1 Technological revolution
The telecommunications sector has been in an ongoing technological revolution for the past two decades. New transmission modes have created alternatives to the terrestrial network of copper wires. Microwave relays and satellites permit transmission without laying cable. Fiber-optic cable is now replacing coaxial cable on main lines, with enormous advantages: greater capacity (by at least ten times), greater immunity to electromagnetic interference, lower rates of signal attenuation, and declining prices (the primary material, silica, is abundant). By way of illustration: the first trans-Atlantic coaxial cable, which entered into operation in 1956, could carry 36 simultaneous conversations; the first trans-Atlantic fiber optic cable, beginning service in 1988, could carry 40,000 calls simultaneously. 4
But the decisive revolution in telecommunications technologies began with the advent of the integrated circuit in the 1970s. Advances in microelectronics (faster, more powerful chips at steadily declining prices) were at the heart of the now-famous marriage of computing and telecommunications. 5 What emerged was the possibility of digitized, electronic telecommunications switching and transmission. The upshot of the microelectronics revolution is that computers can "talk" directly to each other without having to convert their digital information into analog signals for transmission through the telephone system. In fact, any form of information -- text, images, moving pictures, music -- can be digitized and sent directly through the phone lines in "packets" of electronic "1s" and "0s". The digital revolution, still underway, has transformed huge swathes of the economy that now rely on constant, reliable, high-volume data communications: banking, insurance, and securities; airlines and travel; publishing; medicine; design, engineering, and manufacturing; distribution; retailing; and more. There has been a similar proliferation of advanced telecommunications services available to consumers: paging, messaging, cellular phones, and on-line computer services.
The technological revolution led to an explosion of new telecoms equipment and services, which meant upheaval in a vast range of markets. Producers of new devices and services wanted access to networks and customers. Existing businesses began to discover that their competitiveness depended on access to those new devices and services. For both sets of actors, the PTTs were too slow and inefficient.
2.2 The transactors
If societal transactors are ultimately driving integration, we must clearly specify who these actors were arid why they would prefer telecoms policy-making at the EC-level to the existing fragmentation. Four sets of actors supported liberalization and coordination at the European level. The key societal actors involved in constructing an EC telecoms regime were the principal equipment manufacturers, major business users, 6 new service providers, and potential operators of alternative infrastructures. This section briefly identifies their interests.
Equipment manufacturers. All of the major electronics firms in Europe produced a variety of telecommunications equipment, ranging from the large central exchanges (switches) to telephone handsets. Historically, each had cozy ties to its PTT. The PTTs purchased their network equipment from a privileged set (never more than three) of national producers. In Britain, for example, the British Post Office (which was the telecoms operator until the 1980s) bought switches only from the so-called "ring" of domestic suppliers -- GEC, Plessey, and STC -- whose shares of BPO orders were respectively fixed at 40 percent, 40 percent, and 20 percent. 7 Telecoms equipment procurement was commonly a tool of industrial policy, allowing governments to keep national champion electronics firms afloat and even to subsidize their research and development more broadly. Technical standards (again, set by the PTTs) reinforced the preference for domestic suppliers. Each national system operated on its own technical standards and it was prohibitively costly for a manufacturer to develop switches (the big ticket item) customized for each national standard. In any case, the PTTs would have ignored foreign bids.
Why did the equipment producers lose their taste for a system that had nourished them for so long? The technological revolution was again crucial. The first electronic switch entered the market in 1972. After that, the development costs of succeeding generations of faster and more capable digital switches soared. The R&D expense associated with developing a digital public exchange in the mid-1980s was as high as $1.4 billion. 8 To recoup the development costs, a maker would have to sell some $14 -$16 billion worth of switches. The problem was that no single country had a switch market of that scale. 9 Makers of terminal equipment (telephones, modems, fax machines) also faced monopsony PTT buyers, which meant they were limited to producing and selling items that the PTT was willing to offer. Furthermore, the manufacturers were largely confined to producing for their domestic market, because of the differences in standards between countries and the preferential procurement practices of the PTTs.
Business users. Large firms with extensive cross-national transactions within Europe, as well as multinationals needing to coordinate subsidiaries and facilities in different EC countries, also found the PTT monopolies increasingly unable to meet their telecommunications needs. As intra-EC commerce expanded, and as the number of transnational mergers and acquisitions involving EC firms increased, business demand for cross-border telecommunications rose. Both trade and cross-national mergers and acquisitions received a boost from the Single Market program in the mid-1980s.
Equally important, major business users were also discovering that advanced telecoms services were an element of competitive advantage. That is, low-cost, flexible, high-capacity telecoms links could give firms specific advantages over their rivals. 10 Because the United States had deregulated telecommunications earlier and more completely than any other country, American companies had access to a greater variety of the latest equipment and services, all at lower prices. European firms feared that the telecoms advantage gave U.S. competitors an edge. In addition to needing a greater volume of cross-border communications, European companies also needed better access to the most advanced services.
Modernization of the networks was part of the problem -- since the PTTs operated in a monopoly setting, the innovations came more slowly and at higher prices than in the United States. Liberalization (opening the market to competition) was also a piece of the problem. In fact, all of the PTTs were modernizing, and in the United Kingdom the public monopoly had been abolished and competition introduced in 1981. But the overarching difficulty, especially given the rising demand for state-of-the-art, cross-border communications, was the fragmented condition of Europe. Even when PTTs introduced new services, they did so without coordination. The new services on offer were incompatible from country to country, which meant that they didn't really exist for a user needing telecoms links between, say, Copenhagen and Barcelona. 11 To establish a dedicated line linking those two cities (between for example a factory in one and a design center in the other) required separate arrangements with each PTT in between. Correcting malfunctions could become a nightmare. A policy paper published by the Roundtable of European Industrialists, whose membership includes many of Europe's largest multinationals, highlighted the problems caused for business by diverse national equipment approval rules, the incompatibility of services across borders, and the lack of coordinated planning for new services and networks. 12
Thus modernization and liberalization could, in theory, happen within each national telecoms market and still not address the need to facilitate cross-border communications. As it turned out, major telecoms users went to the Commission to push for both liberalization and EC wide coordination.
New service providers. Telecommunications has seen a proliferation in the number and variety of services that can be offered over the network. For decades, the telecoms system was simply a medium for carrying telephone conversations. It was the microelectronics revolution, and the marriage of computing and communications, that triggered an explosion in the number and variety of services that could be offered. The possibility of data communications -- the transmission of information from one computer to another via the telecoms network -- opened an immense new market for services geared toward businesses. It became possible to transmit voice, text, data, and images within and between firms. "Teleconferencing" (with video) became possible, as did computation services, the operation of "private" circuits (linking two or more sites within a firm), real-time inventory and distribution tracking, and so on. An array of services for consumers also opened up, including message storage and forwarding; on-line information; "home" banking and travel reservations; paging and mobile telephony. The next stage will integrate voice, text, data, images, and moving pictures in a "multimedia" digital network serving households; it will be able to deliver ordinary telephony, on-line services, music and movies on demand, and who knows what else.
The proliferation of potential services has attracted a variety of enterprises, large and small, hoping to offer new services via the telecoms system. It is difficult to classify these newcomers to telecoms. Some are established computer companies planning to offer various data services; others are media companies (publishing and entertainment, for example) preparing various on-line services. Many are "startups" hoping to exploit a particular niche. Service providers do not need to own or operate the network itself; they only need to be able to connect to it. The problem in Europe was that the PTTs controlled the approval process for connecting to the public network. In essence, companies hoping to offer services in competition with the PTT had to apply to the PTT for permission to do so. The PTTs began, of course, in a position of monopoly regarding the provision of telecoms services, and only gradually began to license alternative services in specific niches. New entrants also wanted to meet the growing demand for advanced trans-border services within Europe. Naturally, they saw the incompatibilities among national systems as a major obstacle to offering pan-European services. Firms hoping to provide new services, or to compete with the PTTs in existing services, thus joined the band supporting liberalization and coordination at the EC level.
Alternative infrastructures. The ultimate potential competitors for the PTTs were actors who could build and operate telecoms networks. The source of PTT power was their control of the infrastructure. The availability of alternative infrastructures would open competition in equipment and services. In recent years, a variety of industries have stepped forward with plans to build and operate telecoms networks parallel to the existing monopolies. Owners of various kinds of networks have teamed with engineering and technology companies to plan alternative telecoms infrastructures. Railway operators were among the first to seek to build telecoms networks. In each country, the railways own right-of-way linking the principal towns and cities -- precisely the nodes for the "trunk lines" of a telecoms network, as well as the most profitable telecoms markets. For instance, Mercury, the first company licensed to build infrastructure to compete with British Telecom, laid its fiber-optic cable along the tracks of British Railways.
Mobile (cellular) telephone operators (where these are not the PTTs) have generally been obliged to connect to the PTT's terrestrial network; they have pressed to be allowed to establish other terrestrial connections. The cable television (CATV) operators are natural competitors for the telecoms monopolies. Cable TV networks own both distribution networks and links to individual households. Furthermore, television cables have greater capacity (bandwidth) than existing telephone lines and are therefore more immediately useable for the integrated multimedia telecommunications of the near future. Finally, water and electric utilities have also entered the fray. They too hope to piggyback on the widespread distribution networks that they already own.
For each of these groups -- equipment manufacturers, major users, providers of new services, and owners of alternative infrastructures -- domestic reforms were unacceptably slow and piecemeal. In any case, even domestic liberalization was not enough for transactors who needed markets of European scale (suppliers of equipment and services) or for those whose businesses required efficient communications across Europe (telecoms users). These transactors consequently turned toward Brussels, for what they really needed was EC-level liberalization and harmonization. The Commission struck alliances with these various groups to push a liberalizing agenda under an emerging EC regulatory regime.
In this section I lay out the initial conditions in European telecommunications policy-making. The first part reviews the chronology of telecoms reform in the member states. The second part briefly examines the complete absence of telecoms policy-making at the European level during the first few decades of the EC's existence.
3.1 The PTTs and the member states
In European countries, the seemingly untouchable status of the PTTs had its roots in the long-standing conception of the telephone system as a natural monopoly and branch of the state. Since building the network (trunk lines, switching systems, and "local loops") entailed massive capital outlays, only a single provider could achieve the necessary economies of scale. Each additional user connected to the network reduced the average cost per line. Furthermore, the greater the number of households and businesses connected, the greater became the value of the system to each of them (because there was a greater number of telephone communications possible). Given the technological and economic realities, the case for natural monopolies was a logical one (indeed, competing networks became feasible only with the advent of alternative technologies).
In Europe, the telecommunications monopolies became a branch of the state. Governments wanted to ensure complete national coverage, with the obvious political pressure to have uniform pricing even for remote and rural areas (which, if charged on a cost basis, would pay vastly more for telephone service than the urban centers). Governments also took an interest in the national security implications of the networks. Finally, there was a deeper historical and institutional logic to the public telecoms monopolies. When countries undertook the building of telegraph systems, the postal service was the obvious agency to own and operate the lines. After all, the postal service had branch offices all over the country, and telegraph messages were seen as an alternative to letters. The military, of course, also perceived the importance of an efficient telegraph system under state control. When the telephone made its appearance, it was seen as an alternative to the telegraph. The postal service already had a network of wires, so the state added telephones to the post and telegraph service, creating the classic European PTT (post, telegraph and telephone agency).
The PTTs, generally housed within a ministry of posts and telecommunications (MPT), maintained complete control over the national telecommunications systems. The employees of the PTT thus had civil servant status and could be expected to resist any changes that might affect their security of employment or government pensions. The PTTs owned the network infrastructure, provided all of the services, and supplied the terminal equipment (phones, modems, and so on). The ministries responsible for owning and operating the telecoms network were also charged with regulating the sector (establishing standards and equipment approval procedures) and setting rates. Furthermore, governments siphoned the profits from the telecoms monopoly, frequently to offset losses in the unprofitable postal service, or to subsidize the high-technology electronics industries (as in France), or even to augment state revenues in the general budget. Furthermore, because the PTTs were major employers (frequently the largest single employer in a country, as in Germany), governments could use the posts and telecoms administrations as a tool for employment policy.
Given the sacrosanct status of the PTTs, not to mention the concrete financial and policy benefits they provided to governments, it is not surprising that telecoms liberalization has proceeded slowly in almost all EU member states. In fact, the timing of telecoms reforms has analytical significance. For the Commission to have acted merely as an agent of the member governments, it would have to be the case that the states had clear preferences that shaped and delimited the Commission's mandate. If the intergovernmentalist account is to be plausible, member-state preferences for the policies ultimately adopted had to exist prior to the emergence of Commission initiatives. In fact, however, European governments in the early 1980s were just beginning to grope for new models; there were no national telecoms reform strategies in place (excepting Britain) until the latter half of the 1980s. The Commission had defined the key elements of its plan for EU-level liberalization and coordination as early as 1979, before states really knew what they wanted. The following paragraphs briefly review the timing of events in the member states.
The United Kingdom is the only EU state where telecoms reform occurred early, and thus the only state whose preferences for liberalization (if not an EU-level regime) were chronologically in step with the Commission's thinking. The Thatcher government separated British Telecom (BT) from the British Post Office in 1981 and privatized BT in 1984. The government authorized the first rival network operator, Mercury, in 1981 and created an independent regulator (Oftel) that same year. The terminal equipment market was opened to competition in 1984. The UK has been the lone state supporting the majority of the Commission's proposals to liberalize telecoms markets. But given the deep ambivalence toward European integration in both the Thatcher and Major governments, the UK has not been a source of leadership and initiative for EU telecoms reform.
Liberalization has been later and slower in France. The terminals market was partially open by the early 1980s; the Chirac government opened the market for value-added services (VANS) and licensed a second cellular provider in 1987. The Direction Generale des Telecommunications (DGT), an agency of the French government, was still the overwhelmingly dominant telecoms supplier (including cellular). Its operational arm was reorganized and renamed France Telecom in 1988. France Telecom remained under the ministry of posts and telecommunications, and was only separated from La Poste in 1990. It retained its monopoly in the basic domains (network provision, telephone and telex). The French government in 1996 has been promising to establish a fully independent regulator and to privatize France Telecom; it has also signaled that it will introduce competition in the provision of networks and basic telephony as of January 1998, as required by EU rules. 13
In Germany, the Bundespost was the entrenched monopoly. The European Commission compelled the Bundespost to end its monopoly of cordless telephones in 1985. Until 1986, the Bundespost was the sole supplier of modems, and even then it retained its monopoly on the provision of the first handset. The study of reform options did not begin until 1985 when the Witte Commission began its work. The Witte report in 1987 recommended reforms that fell short of full deregulation or privatization. Even those proposals were watered down in the government bill that passed in July 1989, separating a (renamed) Deutsche Telekom from the ministry of posts and telecommunications. The 1989 bill liberalized the terminals market (including first handsets). A second cellular operator did not begin service until 1990. The German government has promised to privatize Deutsche Telekom in advance of the full liberalization of networks and services that will take place, as required by the EU, at the beginning of 1998. 14
The Netherlands began to consider telecoms reform in 1981, but nothing happened until a telecoms reform bill passed in 1986, converting the PTT into a state-owned holding company separate from the postal service. The reform liberalized the markets for terminals and "enhanced" services (VANS) starting in 1989. Belgium was more reluctant. The Belgian reform bill passed only in 1991, creating Belgacom as a state-owned enterprise with a monopoly on basic telephone service. Even then, the government imposed restrictive conditions on the licensing of new VANS and on the use of leased lines. In Denmark, the liberalization plan prepared in 1986 followed the guidelines emerging at the EC. Opening of the terminals market began in 1986 and was complete by 1990. A second cellular operator was authorized in 1990, but Denmark actually reduced internal competition in other telecoms market segments to prepare for the increased European competition being mandated by the EU. Various restructuring plans emerged in Italy in the late 1980s, but liberalization was not on the agenda. A second cellular operator has been licensed but has made little headway against an obstructionist PTT. 15 Telecoms liberalization was not on the agenda in Greece, Ireland, Luxembourg, and Portugal during the period in which the Commission was issuing its proposals.
For purposes of immediate comparison, and previewing the accounts to follow, we can briefly review what the Commission proposed and when. The Commission issued its first major policy statement on telecoms in 1979. It declared that the overarching objective should be the creation of a European market for telecommunications equipment and services. Commission proposals in 1980, 1983, and 1984 fleshed out the goal of a Community-wide telecoms market with specific measures. This brief chronological sketch of Commission and national telecoms reform initiatives makes clear that the Commission was generally ahead of the member states. Even the UK's liberalization of telecoms followed the Commission's 1979 statement on pan-European telecoms markets by some two years. Most of the member states did not begin contemplating domestic liberalization until the mid- and late-i 980s or later.
3.2 The intergovernmental origins, 1957-77
The Treaty of the EEC makes no mention of telecommunications at all, not even as one of the domains for potential policy coordination in the indefinite future. It simply was not seen as a common market issue. Subsequent Council directives opening public procurement to competition in the EC excluded the public utilities, namely telecoms, along with water, transport, and energy. 16 The telecoms sector was also considered to be beyond the reach of EC competition rules. The PTTs were thought to be protected by Art. 90(2), justifying exemptions for the provision of public services, and Art. 222, regarding the special status of national laws on ownership (which indirectly sheltered state-owned telecoms monopolies). As we will see, it took some key ECJ decisions and considerable creativity in interpreting the rules on the part of the Commission in order to establish an EC competence in telecommunications.
In 1957 the six member states established a secretariat outside of the EEC to coordinate postal and telegraph policies. The Six subsequently considered creating a European postal and telegraph union, either within the EEC (under Art. 235) or as an independent body. They chose neither option, deciding instead to form an organization outside of the EEC with broader European membership. The decision was due in part to French reluctance to increase supranational authorities in the EC, but also to a desire to include the UK, which was an important international telecommunications player. The result was the CEPT 17 (founded in 1959), a forum in which post and telecommunications administrations (not their governments) could set non-binding interconnection standards and tariff rules for cross-border traffic, leaving each PTT fully autonomous within its own national territory. The members of the CEPT were (and still are) the national PTTs. The PTT ministers of the EC met for the first time in 1964 (primarily to harmonize postal rates). They did not meet again until 1977. Meanwhile, in 1968, the Commission floated a proposal to form a postal and telecommunications committee to handle harmonization of technical standards (under Art. 100). It withdrew the proposal in 1973 because of a lack of member-state interest. 18
The purpose of this section is to show how telecoms moved from the intergovernmental pole toward the supranational one. By tracing the process by which that movement occurred, I show that the causal effects posited by our theory were in fact at work: transnational actors who needed European-level telecoms reform exploited the political channels that EC institutions provided. Community organizations themselves became causal factors by establishing alliances with societal transactors (the Commission) and by interpreting EC rules in ways that allowed the Community to expand its domain into telecommunications (the Commission and the ECJ). In particular, the Commission and the ECJ exercised rule-making powers in ways that the member-states did not mandate, control, nor approve. I divide the process into two phases.
4.1 Commission initiatives: the first phase
Telecommunications joined the EC agenda via industrial policy, the Commission's RACE program creating the breakthrough. Starting in the late 1970s, the Commission produced a stream of proposals for the telecoms sector, culminating in its comprehensive Green Paper of 1987. Though the legislation proposed by the Commission had to pass through the Council of Ministers, I will argue that the Commission had a substantial impact on the emergence of EU telecoms policy-making during this period. Of course the need for Council approval of legislative proposals imposed constraints on the Commission; no one argues that the Commission is unconstrained. But for the member states, the early 1980s were a period of substantial uncertainty regarding telecoms reform. Though the United Kingdom had liberalized rapidly, none of the other member states had settled on a new model to replace the PTT system. Indeed, many states were unconvinced that they had to give up the PTT model, considering instead that the PTTs could be revamped to provide the new services that users demanded at reasonable prices. Because the national governments were in a mode of policy adaptation, their preferences were neither clear nor fixed. In this setting, the Commission could influence the formation of preferences and policies. 19 The Commission: (1) had an impact on the ideas and models that shaped policy-making at both the national and the EU levels; (2) supplied "focal points" for multi-lateral policy-making; and (3) established alliances with important corporate players that in turn could lobby national governments. In short, though the legislative output for this period was intergovernmental in form, it was nevertheless shaped in significant ways by EU organizations and transnational society.
During the 1960s and 1970s, as the EC member states were actively concocting technology policies, the Commission floated a series of proposals for EC industrial policies in the high-technology sectors. These plans produced meager results. 20 But by the late 1970s it was clear that the national champion strategies to promote high-tech industries had failed. The Commission began to find a more receptive audience for its industrial policy proposals, especially in electronics and telecommunications. The Commission also had an entrepreneurial figure in Etienne Davignon, who took a special interest in the information technologies. Davignon created the Information Technologies Task Force (ITTF) within the Commission, a small group that, aided by technical reports commissioned from outside consultancies, developed expertise in the sometimes arcane details of telecommunications technologies and regulation. The ITTF drafted the document that Davignon presented to the Council in November 1979: European Society faced with the Challenge of new Information Technologies: A Community Response (COM (79) 650). Four of the paper's six points related to telecoms. The overarching goal was to be the creation of a European market for telecommunications equipment and services. The Council asked for specific proposals, and the Commission complied in 1980 with its first draft legislation in telecoms. It would require that the PTTs open bidding to all EC manufacturers for "new" terminal equipment and for 10 percent of their annual purchases of network equipment. Even this modest proposal languished until November 1984, when the Council passed it as a non-enforceable recommendation.
But the Commission's objective -- a liberalized EU market for telecoms -- had crystallized, and the next few years would see a series of Commission efforts to find politically feasible means of achieving that goal. Its case for an expanded Commission role in telecoms had at its core the threat facing European manufacturers from the U.S. and Japan. For European firms to compete, they needed R&D support plus a home market of European proportions. The latter could only be achieved on the basis of common technical standards for future networks. These ideas came together in a set of "action lines" submitted to the Council in September 1983. 21 In addition, Davignon and the ITTF created a Senior Officials Group for Telecommunications (SOGT) in November 1979, with a mix or representatives from ministries of industry and economics and from the PTTs. The SOGT approved a program for the EC based on the Commission's action lines. It focused on creating a Community telecoms market via standards, type approvals, and public procurement; coordinated planning of future networks, especially digital networks, mobile telephony, and broadband; and technology development through R&D cooperatlon. 22
At this point, the Commission made its key move. Following a pattern it had successfully pioneered in the ESPRIT program, the Commission called on the 12 largest EC electronics firms to help it draft an R&D program. 23 Working through the summer of 1984, the companies produced a massive, detailed plan for collaborative R&D leading to the next generation of telecommunications technologies. This served as the basis for the Commission's proposal of a multi-year RACE program ("R&D in Advanced Communications technologies in Europe"). The program would fund specific projects involving companies, universities or laboratories from at least two EC countries. The EC would pay half the costs; the participants would contribute the other half With all of the national champions in the telecoms equipment industry urging their governments to support the program, the Council approved the RACE Main Phase in 1987, with EC funding of ECU 550 million for five years. It similarly approved the Commission's proposed RACE Phase II in 1991 as well as a subsequent extension. RACE projects have involved some 350 organizations in 225 projects, with a total investment of about ECU 2.5 billion. 24 RACE was an important breakthrough for several reasons. First, it established a significant Community role in telecommunications. Second, it marked the Commission's emergence as an actor in telecoms, the Commission having designed and administered a complex Community telecoms R&D program with a substantial budget. Third, it cemented the Commission's ties with the most important firms in the equipment sector, thus constructing the core of a transnational coalition of actors supporting the Commission's objectives. Indeed, the Commission stood out as the political body with the most progressive agenda for telecoms liberalization and EC-wide coordination. It therefore became the leader to which diverse liberalizing interests would rally. From an analytical point of view, RACE is important because it was not the product of member-state initiative or direction. The Commission and its industry partners designed RACE and sold it to the states.
Collaborative R&D, then, was the Commission's entry into telecoms policy-making. Ironically, in telecommunications positive integration (common industrial policy) preceded negative integration (removing barriers to the common market); for the EC overall, the order has generally been exactly the reverse. In any case, the Commission's grand objective had always been a liberalized and harmonized EU-wide market in telecoms. The effort to open the EC's telecoms market started in earnest just as RACE commenced. It clearly drew energy from the broader "1992" program to complete the internal market. But one should bear in mind that creating a single market in telecommunications was not and is not at all the same sort of endeavor as opening a single market for beer or toys. In telecoms, with the exception of the UK, it meant inventing competition, not just increasing it marginally. Opening an EC telecoms market required terminating a deeply rooted system of state monopolies based on a public utility philosophy and a civil service ethic.
The effort to create a Community market in telecoms included two main prongs. The first was to abolish monopolies by permitting new entrants to compete. The second was to establish EC standards; otherwise the result might be competition in national markets that remained unintegrated. The Commission's famous Green Paper on telecommunications mapped out an agenda for both dimensions. 25 The principal policy recommendations were:
- Open competition in a Community market for terminaal equipment, with the possible exception of the first handset.
- Open competition in the provision of services, with the exception of voice telephony.
- Rules requiring the PTTs to grant fair network access to competing providers of services.
- European standards to ensure the "interoperability" of equipment and services anywhere in the Community.
- Splitting network operation and regulatory activities into separate administrations.
- Application of EC competition rules to the telecoms sector, to prevent abuses of dominant position and other anti-competitive practices.
Private sector actors responded enthusiastically to the Commission's proposals. Major users weighed in with support for the Commission's liberalizing agenda. The International Telecommunications Users Group (representing European national users' groups, in which large firms have a strong presence) endorsed all of the Green Paper recommendations. Its only reservations were that the Commission had sometimes not gone far enough, as in reserving the sale of the first handset for the PTTs. 26 The Roundtable of European Industrialists, whose members included many of Europe's largest multinationals, strongly supported the Commission's plans. "Among users," their report declared, "the will exists to bring the EEC's objectives to fruition." 27 The policy proposals of the industrial association, UNICE, whose membership included both equipment manufacturers and large users, were also exactly parallel to the Commission's in the Green Paper. 28 The liberalizing coalition forming around the Commission now included equipment makers and major telecoms users. The Council passed a (non-binding) Resolution in June 1988 endorsing the Green Paper as a blueprint for EU telecoms liberalization.
In addition to market opening, the Commission had early recognized EU standards as crucial to opening a European market in telecoms. The Commission took advantage of the emerging new approach to harmonization, based on mutual recognition. The ECJ's Cassis de Dijon was a crucial underpinning for mutual recognition, which also became the heart of the 1992 single market program. In telecoms, terminal equipment has always required certification that it conformed to the relevant technical standards before it could be sold. The PTTs traditionally controlled this function. The Commission proposed applying the mutual recognition approach to "type approval" of terminal equipment. The Council agreed on a limited basis in 1986 and then approved a comprehensive directive on mutual recognition of type approval for terminal equipment in 1991. 29 The result is that a network of laboratories throughout the EC is now authorized to test equipment for conformity to EC standards. Once a device is approved by any one of the labs, the certification is valid throughout the EC.
The Commission also succeeded in altering the institutional arrangements for establishing European telecoms standards. In the 1987 Green Paper the Commission had proposed establishing an independent European Telecommunications Standards Institute (ETSI). The move was designed to sidestep the CEPT, which the Commission regarded as too closed and dominated by the PTTs. After approval by the telecoms ministers in 1988, ETSI came into being. Representatives of the network operators (still primarily the PTTs), equipment manufacturers, and telecoms users participate in working groups to establish technical standards. Many of the standards so created are voluntary. But the Commission can also assign ETSI to prepare European standards. Once ETSI standards have been approved by the Commission and published in the Official Journal, they are binding in the EU. One of the most successful European standards has been the GSM standard for digital mobile telephony, which has become the dominant world standard.
To summarize this section, beginning in 1978 (before even the British liberalization of telecoms) the Commission articulated the goal of a Community-wide market for telecommunications equipment and services. By the late 1 980s, a number of measures proposed by the Commission -- in R&D, procurement, type approval, and standards -- had taken shape in the form of legislation approved by the Council. The fact that the legislation took the form of intergovernmental agreements in the Council does not contradict a substantial, independent Commission role during this period. The Commission provided "focal points" for agreements among the states. It mobilized major equipment producers (RACE) and telecoms users (the Green Paper) that supported its plans for Community-wide telecoms markets. Perhaps most important, the Commission helped shape the ideas and models that influenced telecoms policy-making during a period when member-state preferences were unclear and in flux. Indeed, it seems plausible to argue that during this period, rather than taking directions and mandates from the member states, Commission proposals influenced thinking and policy-making in the member-states. Though empirically validating that hypothesis would require detailed field research in the member countries, there is interesting confirming evidence from Germany.
The Witte Commission convened in 1985 to study the questions of telecoms reform. Its deliberations were parallel to the Commission's work in preparing and drafting the 1987 Green Paper. The reforms proposed in the 1987 Witte report substantially mirrored the substance of the Green Paper. One analysis of the German process argues that important influences ran from the Commission to Germany. To begin, Herbert Ungerer, an official in DG XIII of the Commission and one of the principal architects of the Green Paper, was one of the experts reporting to the Witte Commission. In their analysis of German telecoms policy networks, Schneider, Dang-Nguyen and Werle identified communication links between the Commission and most of the major German actors, including ministries, political parties, interest groups, and telecoms firms. Furthermore, 61 percent of policy actors in Germany rated Commission influence on German telecoms reforms as strong or very strong, 17 percent considered it weak, and only 6 percent said the Commission had no influence. 30 I suspect that during a period of great uncertainty in the member states, there were similar effects in other national policy processes.
4.2 The Court and Commission as rule-makers: the second phase
In the latest phase, beginning in 1988, the Commission clearly acted autonomously to push telecoms liberalization faster than the member states were prepared to go. Through a series of "consultations," the Commission has mobilized support for its program from a widening circle of influential societal actors. Relying on ECJ precedents, and supported after the fact by new ECJ rulings, the Commission has unilaterally legislated the breakup of PTT monopolies. The legal instrument for this effort has been Article 90 of the Treaty. The salient features of Article 90 directives are that they do not require Council assent and are enforceable in national and EU courts. If one could explain Commission behavior in terms of tracking the lowest common denominator position of the member states, there would be no need for procedures that bypassed the Council. The Commission turned to Article 90 procedures precisely because the member states were dragging their feet on implementing even the liberalization measures to which they agreed. But the extent of Commission autonomy is even greater than it immediately appears. After the ECJ affirmed the first pair of Article 90 directives, the Commission leveraged the Council on later proposals by threatening further use of Article 90. In other words, Council votes thereafter were in a purely formal sense intergovernmental bargains. But they were agreed in the shadow of the Commission's legal powers.
Enforcement of the Community's competition rules is one of the domains in which the Commission possesses substantial autonomy. The Treaty authorizes the Commission to police the internal market in order to prevent company practices that inhibit competition. What the Commission and the ECJ together have done is to extend the application of competition rules to state telecommunications monopolies. The key tool has been Article 90 of the Treaty of Rome, which covers the application of the treaty rules to the public sector. Article 90(1) declares that member states cannot allow state enterprises or undertakings to which the state has granted "special or exclusive rights" to engage in practices that violate the Treaty's competition rules (Articles 6 and 85 - 94). But the treaty also contained clauses that had traditionally been interpreted as exempting the telecoms monopolies: Article 222, which states that the Treaty does not prejudice member state laws concerning the ownership of property (indirectly shielding public monopolies); and Article 90(2), which provides limited derogations from the competition rules for "services of general economic interest" and revenue-producing monopolies, as long as they were not harmful to the Community interest.
The European Court of Justice gradually expanded the application of competition rules to the public sector. A key decision for the telecoms sector came in Italy v. Commission (Case 41183, commonly referred to as the "British Telecom" case). Italy challenged a Commission decision directed against British Telecom, for anti-competitive pricing of telex services. The British government intervened on the Commission's side. The ECJ ruling supported the Commission's application of the competition rules (specifically, Art. 86 regarding abuse of dominant position), and denied a defense based on Articles 222 and 90(2). The decision established that public undertakings must comply with competition rules and further noted that member states did not possess discretion in the application of Article 90(2). 31 The Court later confirmed the application of Article 86 to state monopolies in its Tele-Marketing decision (Case 311/84), and extended this principle with its decision in Macrotron (Case C-4 1/90, regarding a state employment agency). In this case, the ECJ held that Articles 86 and 90(1) applied "where undertaking in question, merely by exercising the exclusive rights granted to it, cannot avoid abusing its dominant position," and in addition ruled that a legal monopoly had a dominant position by definition. Finally, the Court established the direct effect of Article 90 (which initially was held to apply only to states) in ERT (Case C-260/89). 32 The Commission exploited this emerging jurisprudence by issuing two far-reaching directives under Article 90 (which meant that Council agreement was not required). The Commission did so knowing that powerful private actors -- manufacturers, service providers and major users -- frilly supported their liberalizing agenda. These groups had been mobilized during the extensive consultations that the Commission initiated for the issuance of its 1987 Green Paper. The May 1988 Terminals Directive declared an end to monopolies in the provision of terminal equipment, opening the market to competition. It also required the member states to establish bodies for type approval (certification) independent of the network operators. France (supported by Belgium, Germany, Greece, and Italy) challenged the directive in the ECJ. The Court affirmed the Commission's authority to issue Article 90 directives, affirmed its power to apply Article 90(1) in a general way vis-a-vis the states (rather than merely in regard to specific abuses), and affirmed the suppression of exclusive rights for state monopolies (following Dassonville). 33 The Commission has subsequently enforced the terminals directive against member states reluctant to end monopolies, inducing Ireland to abolish the monopoly on the first handset and Denmark to open the market for private branch exchanges. 34
The Commission followed the terminals legislation in June 1989 with an extremely controversial directive opening the services market. The member states were clearly not prepared to go as far as the Commission wanted, namely, to liberalize all services except basic voice and telex. The Commission had been preparing an Article 90 directive since 1988, but it nevertheless sought to minimize friction with the member states. The Commission and the Council struck a compromise in December 1989. The Commission agreed to withhold formal issuance of the services directive until the Council had passed into law the Commission's proposed directive on Open Network Provision (ONP). The intent of ONP was to create common rules of access and technical interfaces, thus guaranteeing that equipment can connect and that services can operate anywhere in the EU. For its part, the Council reached political agreement on the ONP directive, with some details to be hammered out in the following months. Both directives took effect in June 1990 after some additional compromises between Commission and Council. The relationship between the two EU bodies in this episode was not simply one of member states in control of a tightly-reined Commission. Rather, Council and Commission each held sources of leverage, and policies were the outcome of bargaining between the two -- not just among the member states. For instance, the Commission brandished its authority to issue (or delay) the Article 90 services directive (to which several member states objected) to push member states to approve the ONP directive (under Article 1OOa). In order to expedite passage of the ONP directive, the Commission could then agree to minor amendments (regarding the length of the transition period and conditions that states could attach to some service providers, subject to later Commission review). 35 The resulting ONP directive was therefore the product of Council voting, but it was a Council voting under a Commission threat.
The Services Directive of June 1990 (90/388/EEC) opened the market for enhanced or "value-added" services starting immediately and the market for basic data communications from January 1993. Basic voice telephony was reserved for the PTTs. Equally important, the Services Directive required the member states to separate regulatory functions from the PTTs. Finally, the directive called for an assessment of the services markets with respect to the goals of the directive by 1992; this review would be the Commission's occasion to bring pressure to bear on the voice telephony monopoly reserved for the PTTs. Belgium, France, Italy and Spain challenged the Services Directive in the ECJ. In its decision, the Court again affirmed the authority of the Commission to issue general regulations pursuant to Article 90(3). 36 Finally, in a handful of preliminary rulings on Article 177 appeals, the ECJ has reinforced the implementation of the Article 90 directives. The Belgian government had accused private parties of violating national rules by using or marketing devices that had not been approved by the national certification body. The accused parties sought enforcement of the Terminals Directive. The Court ruled against the Belgian government for failing to establish an independent agency for type approvals. In two cases from France, the Court ruled that the regulatory office established within the ministry of posts and telecommunications was not sufficiently independent. Both sets of cases confirmed the direct effect of the Commission directives. 37
The movement toward supranational politics here can hardly be exaggerated. The Commission and the Court essentially created a Commission power to abolish state monopolies. These normative and institutional changes shaped subsequent politics. For instance, the review of the services markets called for in the Services Directive took place as scheduled in 1992. Again, the Commission invited responses to its October 1992 report on the telecoms services market. In response, over 80 organizations submitted written comments and more than 110 offered oral responses at a series of hearings organized by the Commission. Groups providing input in the consultations included most of the PTTs, consumer and business user associations, telecoms equipment manufacturers, service providers, potential suppliers of alternative infrastructures, consultancies, and individual companies not directly involved in the telecoms sector (like Volkswagen). The Commission objective was clearly the complete liberalization of voice telephony in the EU (it had already drafted a directive on the application of ONP to voice telephony). Its report stressed the fragmentation of national markets as a competitive disadvantage for EC users and the price distortions resulting from the continuing monopolies in voice telephony. 38 However, the Commission recommended as politically feasible in the near term the opening of competition in international voice calls within the EU, with national calling to be liberalized later.
Naturally, the PTTs weighed in against competition in voice telephony. But the clear message from other groups was that liberalization of voice telephony should proceed as quickly as practicable. Furthermore, the ECJ in November 1992 (as the services consultation was underway) issued its ruling upholding the Commission's Article 90 Services Directive. Emboldened by the court decision and by the support of the private groups, the Commission in the spring of 1993 dropped its cautious approach (to liberalize voice service only between, not within, member states) and proposed full liberalization of voice telephony by January 1998. 39 Given the array of forces in favor of liberalized voice telephony, the member states could do little other than assent. Accordingly, they passed a July 1993 Council Resolution on full liberalization of all services (including voice telephony) by January l998. 40 The Commission converted the Council's non-binding statement of principles into law with a new Article 90 directive in December 1995. 41
Despite the legal breakthroughs, progress on liberalizing the service markets was uneven and unsatisfactory to the Commission and to many private actors. The PTTs (except in Britain) retained their monopolies of the telecommunications infrastructure -- the network -- and could therefore control access to the network by potential services competitors. Eliminating the network monopoly would liberalize services markets, as network operators would have to compete to attract service providers and customers. Liberalization of infrastructure became the next element in the Commission campaign. Article 90 remained the instrument of choice because the member states remained opposed to liberalization as the Commission envisioned it. For instance, as late as March 1993, the French were willing to consider the liberalization of telephone services -- over a ten to fifteen year period. The French also believed that only some services should be deregulated and that, in any case, infrastructure should remain a monopoly. 42
The Commission set out its agenda for the liberalization of infrastructure in a series of Green Papers. Each of the Green Papers involved consultations like the earlier ones. The consultations thus served to inform and mobilize groups that supported liberalization. The first piece in this set was the 1990 Satellite Green Paper, which pointed out the link between infrastructure competition and services competition. 43 There existed considerable scope for liberalization in satellite communications (telephone, telex, data and other services relayed by satellite). Only the United Kingdom had a completely liberal market, whereas in France and Italy satellite services were a PTT monopoly. Other member states had restrictive licensing systems. Ml of the member states maintained monopolies or quasi-monopolies with respect to terminals (dishes) capable of receiving and sending signals. A number of member states, led by France and Belgium, strongly opposed the Commission's initial draft of the green paper on satellites, which proposed complete liberalization of satellite terminals and guarantees of fair access to satellites for service providers. Indeed, the Commission was forced to reconsider its initial proposals. 44
Nevertheless, the Commission eventually obtained its preferred policy, via an October 1994 Article 90 directive on satellite terminals and services. Subsequent to the Satellite Green Paper and its poor reception, the ECJ affirmed the Commission's first two Article 90 directives on terminals and services. Technically, the satellites directive amended the earlier two directives, which had already been upheld by the ECJ -- another way of solidifying the new law. 45 For the satellites directive, the Commission consulted with the member states on its content and even obtained Council approval of the directive prior to issuing it. 46 But again, the Council considered the satellite directive in the knowledge that the Commission had the power to issue it unilaterally and that the ECJ had already supported the Commission twice.
The second green paper was the April 1994 Green Paper on Mobile and Personal Communications, which emphasized the theme that infrastructure competition would promote the liberalization of services by neutralizing the PTT network monopoly. Mobile (cellular) networks would serve as an alternative to the monopoly ground-based networks. 47 The European standard for digital mobile telephony (GSM) was a world leader. The technology had the potential for users to operate their cellular phone anywhere in the EU. That potential had not been exploited because of national fragmentation. The Commission position was that member states should license competing mobile networks in an open and non-discriminatory mariner, and that mobile operators should be allowed to build their own terrestrial infrastructure, thus enabling them to connect to mobile or ground networks in other countries without having to go through their home PTT. The consultations following the release of the green paper drew written responses from over 70 associations and organizations; more than 100 organizations expressed their views orally at Commission hearings. The groups submitting comments covered the full range of telecoms players: users, manufacturers, network operators, service providers, trade unions. Once again, though the PTTs opposed liberalization of mobile communications, the vast majority of other respondents fully supported the Commission's proposals.
By June 1995, the Commission had in hand a draft Article 90 directive for liberalizing mobile communications. There was considerable scope for introducing competition: only the UK provided for open competition among mobile providers. In the rest of the EU, licensing of mobile networks was severely restricted, and mobile operators were frequently prohibited from utilizing any terrestrial infrastructure other than that supplied by the national telecoms monopolies. 48 Furthermore, most member states -- Italy, Belgium, Ireland, Spain and Austria in particular -- charged the second mobile operator an "entrance fee" that the state monopoly is not required to pay. 49 As with satellites, the new law would amend the original services directive. It provided for liberal licensing of competing mobile operators and ensured that mobile networks could connect to each other and to the public network. The Commission adopted the Article 90 directive liberalizing mobile communications in early 1996.
Completing the set of green papers was the 1994-95 Green Paper on the Liberalisation of Telecommunications Infrastructure and Cable Television Networks, issued in two parts. Even before the Green Paper was issued the member states were voicing strong objections to the deregulation of alternative infrastructures. At the conclusion of a September 1994 Council meeting, a statement on alternative infrastructures was deleted from the final Conclusions because the member states could not agree. Only Germany and the UK favored deregulation of infrastructures in 1998; Belgium, Denmark, Greece and Portugal opposed the whole idea. 50
But the Green Paper was extremely ambitious. It proposed full liberalization of infrastructures in 1998, which would mean permitting new entrants -- cable TV operators, mobile operators, railroads, water and electric utilities -- to build and operate telecoms networks in direct competition with the PTTs. Services already liberalized (everything but voice telephony and telex) would be deregulated on alternative infrastructures in 1996. 51 Consultations opened by the Commission in October 1994, and lasting through March 1995, drew inputs from an immense array of actors. Over 100 submitted written contributions and the hearings involved more than 125 groups. In addition to the usual players, the Infrastructures consultations attracted cable TV industry associations, broadcasters, film makers, and advertising industries. The interest of the latter groups stemmed from the Green Paper's proposal to permit the provision of telecoms services via cable television networks. The attraction of cable TV networks for liberalizing telecommunications is that they already possess the crucial "local loop," that is, the connection to households. Furthermore, the greater bandwidth of TV cables (as compared to traditional telephone wires) would in principle be well-suited for advanced telecoms services (high quality sound, images, moving pictures). As with the other consultations, aside from the PTTs, the various interested parties supported the Commission 5 objective of full infrastructure liberalization. In fact, the association of employers' federations, UNICE, released an additional endorsement of the Commission's agenda on the eve of the November 1994 Council meeting that was to discuss the first part of the Green Paper. 52
At that November meeting of the Council, the battle lines were clearly drawn. Four states supported the Commission's two-stage plan (France, Germany, Netherlands and the UK, soon to be bolstered by Sweden and Finland). But the remaining eight states were absolutely opposed to the first stage (virtually immediate deregulation of infrastructures for services already liberalized). As a result, the Resolution issued at the conclusion of the Council mentioned only the 1998 deregulation of all infrastructures. In response, competition commissioner Karel van Miert, declared that the Council's reluctance would not stop him from considering an Article 90 directive on the first stage. 53
In fact, even before it issued its Green Paper on Infrastructures, the Commission was holding out the possibility of leapfrogging a foot-dragging Council with an Article 90 directive. 54 Immediately after the November 1994 Council meeting that debated the Green Paper the Commission prepared a draft Article 90 directive that would open cable television networks for the provision of liberalized telecoms services. 55 At the time, only the UK had liberalized cable networks for the carrying of telecoms services. But reaction from the Council in early 1995 was strongly negative: most of the delegations to the relevant working group opposed the early liberalization of cable TV for telecoms. 56 The Commission nevertheless pressed ahead, issuing the Article 90 cable TV law (amending the 1990 services directive) in October 1995, to take effect as of January 1996. 57 The cable TV directive was only part of the infrastructure liberalization project. The Commission's ultimate objective was complete infrastructure liberalization, which was the subject of a February 1996 Article 90 directive. The latest directive mandated that services liberalized already (in essence, all services except voice telephony and telex) could be offered on any infrastructure from July 1996, and that complete liberalization of all services and infrastructures would begin January 1998 (with limited transition periods for member states with less-developed networks). In addition, the directive set out basic principles for the regulatory framework that would govern the new regime of open competition, particularly in the sensitive areas of interconnection (conditions and fees for service providers to connect to existing networks), licensing (conditions for authorizing new market entrants) and provision of universal service (means of financing and guaranteeing access to some minimum level of service to all residents). 58 The PTT monopoly on networks was coming to an end, as cable televison operators, railroads, utility companies, mobile networks and satellite systems were allowed to compete with the taditional telecoms operators. In principle, service providers would not have to rely on the PTTs for access to customers (though in practice, there will be a lag before alternative networks are built and functioning).
A decade ago, telecommunications in the EU was a patchwork of state monopolies. Governments clearly preferred to retain control of the PTTs, for a number of reasons: the historic public utility, natural monopoly ideology surrounding telecoms in Europe; the capacity to use the PTTs as instruments of industrial and employment policies; and the revenues generated by the telecoms monopolies. Furthermore, the Treaty made no mention of telecommunications and there was a standing presumption that certain articles protected the PTTs from EU competition rules. Yet today the liberalization of telecoms is being driven primarily at the EU level. In the language of our continuum, telecoms policy-making has moved substantially from the intergovernmental pole toward the supranational.
Intergovernmentalist theory cannot explain this movement. The intergovernmentalist account sees no independent role for the Commission, the ECJ and transnational groups. The EU telecoms regime would therefore be the result of a string of compromises each of which embodied the preferences of the most reluctant state. But as the empirical record shows, the key steps in liberalizing telecoms at the EU le continuum, telecoms policy-making has moved substantially from the intergovernmental pole toward the supranational.
Intergovernmentalist theory cannot explain this movement. The intergovernmentalist account sees no independent role for the Commission, guaranteeing access to some minimum level of service to all rethe ECJ and transnational groups. The EU telecoms regime would therefore be the result of a string of compromises each of which embodied the preferences of the most reluctant state. But as the empirical record shows, the key steps in liberalizing telecoms at the EU level consisted of unilateral Article 90 directives from the Commission. The Commission resorted to Article 90 precisely because the member states were, with the UK being the only regular exception, resisting liberalization. If liberalization were proceeding at the pace of the slowest member states, there is no question that telecoms reform would be far less advanced than it is today.
In contrast, our theory of institutionalization explains the empirical record; things happened as we would expect. The fragmented PTT system was increasingly costly both for the increasing number of transactors (users) who relied on cross-border communications and for the new telecoms players who hoped to meet their telecoms needs. These groups -- equipment manufacturers, service providers, business users, consumer groups, and providers of alternative infrastructures -- rallied to the Commission's cause in the series of consultations. Those groups pressing for liberalization were the transnational society of telecommunications.
Our theory also reserves a place for supranational organizations that can have an independent effect on EU politics and policy-making. For a variety of reasons -- having to do with historical path dependence, the costs of supervision, and formal EU law 59 -- the member states cannot always control EU organizations like the Commission and the ECJ. In telecoms, the Court and the Commissions were clearly not simply doing the bidding of either the most powerful of the most reluctant member states. As the empirical record shows, Commission initiatives were almost invariable more ambitious (in terms of liberalizing) than most member states were willing to contemplate. That is, the member states could not have been dictating their preferences to the Commission because the Commission was always ahead of them. Even the British telecoms reforms came two years after the Commission had formulated the objective of a pan-European telecoms market. The case for the Commission acting independently of member-state preferences is absolutely clear when the Commission utilized its rule-making powers under Article 90 (liberalization of terminals, services and infrastructures). But even in the first phase (from RACE to the fin because the Commission was always ahead of them. Even the British telecoms reforms came two years after the Commission had formulated the objective of a pan-European telecoms market. The case for the Commission acting independently of member-state preferences is absolutely clear when the Commission utilized its rule-making powers under Article 90 (liberalization of terminals, services and infrastructures). But even in the first phase (from RACE to the first Green Paper) the Commission had an autonomous impact.
Of course it is true that the key legislative pieces in the first phase were voted by the Council of Ministers. But intergovernmental agreements can also be shaped substantially by supranational processes. As the empirical account showed, the Commission's initiatives in the first phase came before member-state preferences took shape. In fact, most of the member states did not define their preferences regarding telecoms reform until the latter half of the 198Os. The Commission's initiatives had an impact on member-state thinking and preference formation, by shaping ideas and models, providing focal points for EU decision-making, and mobilizing influential societal groups. The Commission's impact on policy-making was akin to education, or persuasion. More generally, if the interactions and discourses of the EU significantly shape state preferences, then interstate bargains are not what intergovernmentalists assume them to be.
The telecoms case shows the interactions among the three dimensions of our continuum: supranational rules, supranational organizations, and transnational society. EU law, driven in crucial instances by the autonomous rule-making powers of the Commission and the ECJ, has brought an end to the PTT era. The liberalization of telecoms markets is underway. But a frilly competitive EC telecoms market poses broader challenges in a context of decentralized regulatory authority. Three main issues stand out:
-
Interconnection,
to ensure that the multiplying networks and services will be technically compatible everywhere, and that connection fees will not be a barrier to new entrants;
-
Licensing,
to guarantee that whatever procedures are used to license new providers are not discriminatory or anti-competitive; and
- Universal service, to provide access to a minimum set of services to all users at affordable prices on an equal basis, to make sure that less profitable customer groups (like those in remote or rural areas) would not be abandoned or forced to pay radically higher tariffs.
Note *: Wayne Sandholtz, Dept. of Politics and Society, University of California, Irvine. Back.
Note 1: The only exception being the United Kingdom, which liberalized telecoms in 1981. Back.
Note 2: "Equipment" refers both to network equipment (the lines and switches) and terminal equipment (devices connected to the network by users, including telephones, modems, faxes, local branch exchanges, and so on). "Services" refers to the various types of communications offered on a network (voice telephony, data communications, mobile communications, forwarding, paging, voice mail, databases, and so on). "Infrastructure" refers to the network itself, including the tasks of maintenance and management. The basic infrastructure historically consisted of the network of lines with copper wires running to each house or business. Today there are parallel and linked infrastructures: satellites, microwave transmission, cellular networks, and cable television networks (which can be used for telecommunications). Back.
Note 3: Walter Mattli, "Explaining Regional Integration," Working Papers on International Society and Institutions, Global Peace and Conflict Studies Program, University of California, Irvine, forthcoming. Back.
Note 4: Herbert Ungerer, with Nicholas P. Costello, Telecommunications in Europe (Brussels:Commission of the European Communities, 1988), p.64. Back.
Note 5: A variety of neologisms capture the resulting hybrid: information technologies, telematics, informatique. Back.
Note 6: "Users" refers to consumers of telecoms equipment and services; many large enterprises are important consumers of advanced services and equipment. Back.
Note 7: Rob van Tulder and Gerd Junne, European Multinationals in the Telecommunications Industry (Amsterdam: Universiteit van Amsterdam, 1984), p. 46. Back.
Note 8: Godefroy Dang Nguyen, "Telecommunications: A Challenge to the Old Order," p.108. Back.
Note 9: van Tulder and Junne, European Multinationals, p.70. Back.
Note 10: See Francois Bar and Michael Borms, From Public Access, p.4. Back.
Note 11: For instance, packet-switched data networks nominally using the X.25 standard could function at only 4.8 kilobits per second for transborder traffic, compared to 48 kilobits per second domestically in the mid-1980s. See Roundtable of European Industrialists, Clearing the Lines: A User's View on Business Communications in Europe (Paris: European Roundtable, 1986), pp.-12. Back.
Note 12: Roundtable of European Industrialists, Clearing the Lines: A Users' View on Business Communications in Europe (Brussels, 1986), pp.16-19. Back.
Note 13: Eli Noam, Telecommunications in Europe (New York: Oxford University Press, 1992), pp.163-66; Financial Times, 10 January 1996, p.3. Back.
Note 14: See Noam, Telecommunications in Europe, pp.90-98; Financial Times, 10 November 1995, p.14. Back.
Note 15: Noam, Telecommunications in Europe. Back.
Note 16: The directives in question were: Council Directive 7 1/3O5IEEC, Concerning the Coordination of procedures for the Award of Public Works Contracts; and Council Directive 77/62/EEC, Coordinating Procedures for the Award of Public Supply Contracts. See Wolf Sauter, The Relationship Between Industrial and Competition Policy under the Economic Constitution of the European Union, with a Case Study of Telecommunications, Ph.D. thesis, Law Department, European University Institute (Florence, 1995), p.188. Back.
Note 17: Conference europeenne des administrations des postes et des telecommunications. Back.
Note 18: This paragraph relies primarily on Volker Schneider and Raymund Werle, "International regime or corporate actor? The European Community in telecommunications policy," in Kenneth Dyson and Peter Humphreys, eds., The Political Economy of Communications: International and European Dimensions (London: Routledge, 1990), pp.86-88. Back.
Note 19: See Sandholtz, High-Tech Europe, chaps. 2, 8. Back.
Note 20: Sandholtz, High-Tech Europe, chap. 4; Sauter, The Relationship Between Industrial and Competition Policy, pp. 78-83. Back.
Note 21: Communication from the Commission to the Council on Telecommunications: Lines of Action, COM(83) 573. Back.
Note 22: Communication from the Commission to the Council on Telecommunications COM(84) 277. Back.
Note 23: For an analysis of the politics behind ESPRIT see Wayne Sandholtz, "ESPRIT and the Politics of International Collective Action," Journal of Common Market Studies, Vol. 30 (1992). The so-called "Roundtable" of telematics firms were Bull, Thomson, and CGE from France; Siemens, Nixdorf, and AEG from Germany; GEC, ICL, and Plessey from the U.K; Olivetti and STET from Italy; and PHilips form the Netherlands. Back.
Note 24: Commission of the European Communities, RACE 1993 (Brussels, 1993); and RACE 1995 (Brussels, 1995). Back.
Note 25: Towards a Dynamic European Economy: Green Paper on the Development of the Common Market for Telecommunications Services and Equipment, COM (87) 290. Back.
Note 26: George G. McKendrick, "The INTUG View on the EEC Green Paper," Telecommunications Policy, vol.11 (1987), pp. 325-29. Back.
Note 27: Roundtable of European Industrialists, Clearing the Lines: A Users' View on Business Communications in Europe (Brussels, 1986), pp.16-19. Back.
Note 28: La Tribune, 24 January 1987; Financial Times, 23 January 1987. Back.
Note 29: Council Directive on the initial stage of the mutual recognition of type approval for telecommunications terminal equipment, 86/361/EEC; Council Directive on the approximation of laws of the Member States concerning telecommunications terminal equipment, including the mutual recognition of their conformity, 91/263/EEC. Back.
Note 30: Volker Schneider, Godefroy Dang-Nguyen, and Raymund Werle, "Corporate Actor Networks in European Policy-Making: Harmonizing Telecommunications Policy," Journal of Common Market Studies 32;4 (December 1994), 473-98. Back.
Note 31: See Wolf Sauter, "The Telecommunications Law of the European Union," European Law Journal, vol.1(1995), p.101. Back.
Note 32: Sauter, The Relationship Between Industrial and Competition Policy, p.169. Back.
Note 33: Case C-202/88, France v. Commission, March 19, 1991. Back.
Note 34: European Report, No.1628, 10 November 1990, p.3. Back.
Note 35: European Report, No.1560, 3 February 1990, p.11; European Report, No.1593, 9 June 1990, p.3. Back.
Note 36: Joined Cases C-271-90, C-28 1/90, and C-289/90, Spain, Belgium and Italy v. Commission. Back.
Note 37: Joined Cases C-46/91 and C-93/91, Procureur du Roi v. Lagauche and others, Evrard; Case C 69/91, Ministere Public v. Decoster and Case C-92/91, Ministere Public v. Taillandier. See Sauter, "The Telecommunications Law of the European Union," p.105. Back.
Note 38: COM (93)159, Communication on the Consultation on the Review of the situation in the telecommunications services sector. Back.
Note 39: Eurecom, vol.5, no.5 (May 1993), p.1. Back.
Note 40: 93/C2 13/EEC, Council Resolution of 22 July 1993 on the review of the situation in the telecommunications sector. Back.
Note 41: European Report, No.2112, 1 March 1996. Back.
Note 42: European Report, No. 1843, 13 March 1993. Back.
Note 43: COM (90) 490, Towards Europe-wide systems and services: Green Paper on a common approach in the field of satellite communications in the European Community; Council Resolution, December 19, 1991. Back.
Note 44: European Report, No.1624, 24 October 1990, p.1. Back.
Note 45: Commission Directive 94/46/EC of 13 October 1994 amending Directive 88/301/EEC and Directive 90/388/EEC in particular with regard to satellite communications. Back.
Note 46: European Report, No.1984, 15 October 1994. Back.
February 1990, p.11; European Report, No.1593, 9 June 19 Note 47: COM (94)145, Towards the Personal Communications Environment: Green Paper on a Common Approach in the Field of Mobile and Personal Communications in the European Union, April 27, 1994. Back.
Note 48: European Report, No.2052, 24 June 1995. Back.
Note 49: European Report, No.2067, 16 September 1995. Back.
Note 50: European Report, No.1980, 1 October 1984; No.1988, 29 October 1994. Back.
Note 51: COM (94) 440, Green Paper on the liberalization of telecommunications infrastructure and cable television networks. Part One: Principles and Timetable; and COM (94) 682, Green Paper on the liberalization of telecommunications infrastructure and cable television networks. Part Two: A common approach to the provision of infrastructure for telecommunications in the European Union. Back.
Note 52: European Report, No.1994, 19 November 1994. Back.
Note 53: European Report, No.1994, 19 November 1994. Back.
Note 54: European Report, No.1980, 1 October 1994; No.1988, 29 October 1994. Back.
Note 55: European Report, No.2004, 24 December 1994. Back.
Note 56: European Report, No.2025,15 March 1995. Back.
Note 57: Eurecom, Vol.7, No.10 (November 1995), p.1. Back.
Note 58: European Report, No.2112, 1 March 1996. Back.
Note 59: See the chapters by Pierson, Pollack February 1990, p.11; European Report, No.1593, 9 June 19 and Sandholtz/Stone on these points. Back.
