Universitat Pompeu Fabra

Department of Political and Social Sciences

Ramon Trias Fargas, 25-27

08005 Barcelona (Spain)

During the nineties, public telecommunications policies in Europe became increasingly more complex with the progressive dismantling of state monopolies and the establishment of a competitive market, giving shape to a new multilevel government system. This was a common occurrence in the field of European politics (Scharpf, 2001). At the end of the nineties, the emergence of this type of multilevel government in the sector was characterised by the presence of a combination of joint decision-making factors among member states (the creation of a market) and elements of mutual adjustment (the privatisation of operators), which led to a sophisticated area of decision-making on multiple levels. In this sense, and despite the fact that during the entire decade the liberalisation of telecommunications markets in the member countries was a process which was strongly promoted (and in some aspects even brought about hierarchically) by the European Commission, in each country there were priorities and specific intervention logics regarding the numerous aspects of government in the sector. In some countries, for example, during the period of market transition an intense political struggle appeared in order to pre-define the profile of the principal operators. In almost every country, the mechanisms to introduce competition (interconnection fees, granting of licences, regulation of prices, etc.) were adjusted in order to introduce national policy objectives. Even the creation of new public institutions to regulate telecommunications services demonstrated that there was a great degree of variation as far as the countries’ degree of autonomy and capacity for action was concerned.

Since the eighties, there have been two important tendencies which have influenced the remarkable transformation of the telecommunications sector throughout the world. Firstly, technological changes (digitalisation has led to convergence in the areas of mass media, computers and telecommunications) and secondly, the strong tendency towards globalisation, with the internationalisation of numerous markets and the predominance of large transnational companies. We now form part of an environment in which technology is becoming increasingly globalised. Products and services are becoming internationalised and the intervening role of public powers is changing, with new means of intervention being developed and objectives being redefined. As one can imagine, this is neither a completely linear nor a unidirectional process.

The very multilevel structure of the form of government in the sector which emerged in Europe during the nineties could be interpreted, in functionalist terms, as an adaptation to the complexities and demands of the changing environment (Schneider, 2002). A more sophisticated means of government undoubtedly offers European governments greater flexibility and at the same time greater security with which to face quickly changing technologies and unexpected business decisions, two phenomena which are becoming increasingly common within the global context. Hence, national governments can endorse the maintenance of some elements of their style of regulation and state intervention in order to tackle new challenges brought about by internationalisation and new technologies. In this, the national governments share a set of characteristics with the other members of the European Union as regards the functioning of the market (Levi-Faur, 1999).

Undoubtedly, this situation adapts well to the characteristics of a policy sector where the role of institutional factors with a temporary dependency (path-dependency) was very important. At the same time, the demands for change were also very strong within the environment. The long tradition of public administrations specialising in telecommunications management was a largely influential factor, which seriously conditioned the process of institutional change within the sector in the majority of European countries. In this way, the European Commission was able to take the initiative to bring about change in the sector during the nineties, largely because the member states were quite limited in independently enforcing change in their policy- making (Thatcher, 1999). However, this does not mean that sectorial 'path dependency' will not continue to have a relatively strong effect in each country, and a clear demonstration of this is precisely how each country individually adjusted to the formation of the market during the definition of the new economic and institutional environment.

Included in the principal issues which national governments tackled individually during this period of market transition were: a) preferences on ownership (public or private) of the dominant operator, b) how to deal with their situation of predominance in the market, c) how to design a new framework of public intervention in the sector, and d) what type of institutional reforms to encourage in the public sector itself (to be able to bring in new policies). Another issue was how to tackle the situation of strong competition amongst corporate and financial groups, which were, at times, relatively opportunistic in their search for quick earnings in a fast-growing sector (Thatcher, 2000, Hulsink, 1999). As we can see, we are dealing with a wide range of issues, which in many aspects were limited (but not completely defined) by the criteria and regulations put forward by the European Commission (and adopted by member states in their national legislation).

The context of change in policy-making and institutional innovation on multiple levels which came about in the European telecommunications sector during the nineties constitutes the referential framework of the case study at hand. We are going to concentrate on analysing quite a limited aspect of this set of transformations which we have outlined. Basically, our objective is to study the formation and subsequent transformation of the network of actors present in this policy sector throughout the market transition process which took place in the European Union in the latter half of the nineties. More specifically, we are going to be studying the case of Spain. Our aim is to investigate its specific characteristics as a means of providing evidence for the permanence of national decision spaces in this sector with a particular agenda (made up of interests and power relationships existing within the network of actors). The sector selected themes and intervention priorities at the same time as it took European directives on board, given that it formed part of the multilevel telecommunications government structure which was being formed in the European arena.

Despite its limitations, the concept of public policy networks is very useful in differentiating and describing the group of public and private organisations which intervene in the public policies sector and the relationships between them (Marin/Mayntz, 1991, Dowding, 1995). Within this perspective, we can observe how actors have an interdependent relationship, how they interact and which institutional frameworks come about as a result of the stability of specific relationships existing in the network (Laumann/Knoke, 1987). This does not mean that all actors have the same power within the process. Each one's power depends mainly on the resources they possess and on their importance for the other actors (Scharpf, 1978). This interdependence is not static, but should be defined rather as a continually developing and changing process, which actually generates the actors involved in the interaction. For this reason, the concept of interdependence is a key factor in the constitution and maintenance of policy networks (Aldrich, 1979 Hanf/Scharpf, 1978, Klijn/Teisman, 1991). As a result of these interdependent relationships, actors interact with each other in mutual games and strategies, whereby they try to obtain certain favourable results, which is why they must take into account the action strategies of the other actors. Generally, we can find three distinct aspects which are tackled through this type of analysis: the formation (and transformation) of the policy networks, the delimitation of the subsystems of interaction among actors, and the contribution to explaining decision- taking on public policies (König, 1998).

In our case, knowing the basic characteristics of the network of actors and observing how this transformed during the market transition process gives us a better understanding of the pattern of development and change in public telecommunications policies in Spain during the nineties. It is our belief, however, that our analysis does not aim to provide an explanation for the political decisions taken during this time, but rather to place how the decision process came about in context, by setting out why certain policy options were finally adopted, while other options were not even contemplated. We have based our study on the field of work developed by authors on the development of telecommunications policies in Spain (Jordana et al, 1997, 2000). With an interval of almost four years between them (winter 1996 and autumn 1999), two separate surveys on telecommunications policies were carried out on representatives from public and private organisations involved in the telecommunications sector in Spain. The surveys covered the attitudes, preferences and opinions of the organisations' representatives concerning the regulation and development of telecommunications policies in Spain. They also asked for information on their relationships with the other organisations of the sector. We are therefore going to identify the stages both before and after the liberalisation of the Spanish telecommunications sector, taking two distinct moments in the development of the Spanish sectorial telecommunications network.

A specific characteristic of the Spanish telecommunications sector within the European context has traditionally been the absence of a public administration to take charge of its management. The operator, Telefónica, had a monopoly on the sector but it acted as a private company (which had a public concession). During the dictatorship, when the government acquired the controlling shares of Telefónica from ITT, the organisation and management system of the company did not change in any way whatsoever and the system remained unaltered until the 1990s (Jordana, 1998). Therefore, as a rule, the government fully delegated the defining of sector policies to the operator, who in turn set its own objectives, with some considerations of a public nature. However, its action was basically focused on criteria of a corporate nature (for example, there were no universal service objectives). It was only in the mid-eighties that the government started to establish its own organisation to define its telecommunications policy, which was separate from Telefónica's. Some years were to pass, however, before the new public bodies were consolidated and formulated some of their own criteria, which were distinct from those of the monopoly Telefónica.

This account of Telefónica and the Spanish government is significant in that it indicates the separate origins of public organisations responsible for policy-making in the area of telecommunications regarding the controlling operator, Telefónica. It was not, however, a complete separation, as there was a strong relationship between them and the professionals in the sector who almost always came from prestigious telecommunications engineering schools with a strong corporate identity. However, we consider that this configuration, to a large extent, facilitated the creation of new institutions in the nineties, which were more autonomous, and also led to the state assuming certain control of the construction of the market. In other terms, it can be pointed out that the absence of a strong 'path-dependency' in the institutions within the sector contributed to the reinforcement of public action (by increasing the capacity for action by the state) during the period of policy change.

Our hypothesis regarding the structuring of the sector's policy network in Spain during market transition is that the institutions’ aforementioned configuration permitted the construction of a new system for the mediation of interests which was somewhat more pluralistic and far removed from the customary closed-up and patrimonial traditions of southern Europe. At the same time, it allowed the government to prevent a capture situation regarding the incumbent operator Telefónica which, despite this, did not stop trying to achieve this end during the entire period. The entry of new business and social interests to the policy network during the transition process was limited but sufficiently significant for new policy objectives to appear. These new objectives were associated with how the country adapted to the new situation of the sector. This also gave rise to the appearance of a more complex institutional environment and a wide utilisation of the room for manoeuvre, which the common European regulations concerning the sector permitted.

Our intention is to use the analysis of the sector's policy networks during this period to provide different factors to allow us to carry out an empirical test of the hypotheses used in the precedent interpretation. More specifically, we have concentrated on analysing three basic aspects of the nature of the policy network: a) firstly, we wish to analyse how the centrality of the state institutions within the network had developed in order to examine if the state’s capacity for action had been reinforced throughout the process; b) secondly, we wish to observe how power and influence was distributed within the realm of the policy network in order to characterise the new system of mediation of interests emerging in the sector once political change had come about; and c) finally, we wish to identify how the central ideas of collaboration and conflict were evolving in the emerging policy network, observing the distribution of resources, information and values among actors.

We are not endeavouring, therefore, to use the analysis of policy networks to explain how their make-up affects the process of public decision-making, beyond establishing a generic relationship between the initial weakness of this and the capacity of the government to push policy change forward in the sector. In this sense, we consider that the structure of the initial network facilitated a specific range of policy options, in that a coalition of actors with the capacity to veto government initiatives did not exist. This does not, however, explain the government's decisions. For this reason, we are interested in closely identifying how the structure of the network was transformed during this period of intense policy changes. Our aim is, thus, to observe the basic beliefs which existed regarding the sector and how the different actors' perceptions of the concept of influence changed.

In more general terms, we can state that our examination concentrates on analysing how a policy network changes as a result of a profound change in both the policy orientation and the institutions of a policy sector, brought about fundamentally by reasons alien to the very dynamics of the network. It is, therefore, a matter of evaluating how these dynamics of change are transformed into variations within the existing relationships of interdependence, leading to new games and relationships, and also redistributing the relationships of power and influence existing in the heart of the network. The appearance of new actors in the 'policy network' is a key element in understanding the transformation which took place during the period of policy change. Our aim is not so much to investigate the reasons for the appearance of new actors on the scene, but to see how the network of actors was reshaped, which once formed, was able to reach a situation of long-lasting stability. One may think that the appearance of new actors on the scene was not an intentional result of government action given that its decisions had another type of reasoning, closer to that of acting as a short-term go-between for different types of interests wishing to participate in the sector (Pierson, 2000). In this sense, the creation of institutions and the steps taken during the period of policy change had an unexpected effect - the formation of a new policy network, which was to lead to the subsequent stabilising of the type of decisions and sector policies in the country, formulating thus a peculiar national adaptation to the multilevel telecommunications government in Spain.

At first glance, the Spanish public telecommunications policies network could be characterised by the types of organisations forming part of the network and by placing the active participants in distinct categories. As active participants we consider all the organisations which are responsible for managing, co-ordinating or controlling, in some form or other, the creation or distribution of symbolic or material values of the sector (Laumann/Knoke, 1987). The classification we carried out established eight distinct types of organisations: political parties, public administrations, telecommunications operators, business and user associations, trade unions, large companies from other sectors, the mass media and finally, consultants and investigation centres.

The first group of actors consists of the different political parties with a capacity to intervene in the legislative process. The positions of the different political groups on the telecommunications sector varied considerably throughout the decade. Up to the mid-eighties, the area of telecommunications policies was an eminently technical subject for political parties, and was present neither in political speeches nor in their regulation priorities. However, from 1995-1996 onwards, the degree of involvement of political parties in sector-related issues increased considerably, although they did not actually formulate well-defined positions. On the one hand, they increasingly realised the importance the sector had for economic development and the intense integration of the audio-visual and telecommunications sectors. However, their position was a key factor in regulating the sector (by adapting community directives) given that the government did not have an absolute majority in Parliament during those years. These facts led to their acquiring a certain protagonism during the market transition period, which started to wane once the sector policies left the parliamentary arena.

Another group of actors forming part of the telecommunications policy network is that made up of public bodies which had direct or indirect power in the sector. When the first questionnaire was sent out, at a time coinciding with the last months of the socialist government (whose term of office ended in March 1996), the public actors of note were the Presidency of the Government and the Ministry for Public Works, Transport and the Environment (MOPTMA). Within the MOPTMA, there were specialised units such as the General Secretary for Communications and the General Management of Telecommunications. Together with these, the Ministries for the Economy and Industry also played an important role as actors who influenced the definition of policy-making in the area of telecommunications. For quite a time, there was an internal struggle within the socialist government regarding the rhythm of introduction of the market in the sector. While the Ministry for the Economy’s aim was to rapidly advance the opening up of the market to promote competition in the economy, the MOPTMA defended the strategy of slow liberalisation, allowing time for Telefónica to modernise itself and to readjust its rates.

However, by the end of 1994, the positions of the public actors regarding the decision to adjust the process of liberalisation to the predominant rhythm in the European Union had converged. Telefónica was initially opposed to this decision, but it quickly realised its advantages, especially within the context of the growing internationalisation of the company.

The second questionnaire coincided with the end of the first term of the government of the 'Partido Popular' and therefore reflected a group of relatively different public actors. Since mid-1996, there had already been a reorganisation on an administrative level, with the transfer of powers related to telecommunications policies to the Ministry for Public Works, and in the same year the CMT (Commission for the Telecommunications Market) was created, which was a semi-independent regulating body. Apart from this, the economic Ministries maintained their interest in this policy sector. Regional administrations also assumed some new responsibilities regarding the extension of new telecommunications networks throughout the country. In this context, the orientation of the Ministry for Public Works had a certain continuity with regards to the previous positions of the socialist government. It aimed to develop a controlled process to liberalise the sector, which was adapted to European rhythms, and to try not to adversely affect Telefónica's business interests in its international expansion process. However, it actively introduced a new policy element in its attempt to structure the Spanish market by encouraging new national operators.

The third group of actors consists of telecommunications operators. Among these, two large groups with opposing interests must be distinguished. The first was the monopoly Telefónica, which was already firmly established in the Spanish market and which had a strategy aimed at maintaining its position in the sector. The second group was all the new operators attempting to enter the market. Telefónica's influence in the public area was enormous, and, in some aspects, as the differentiation between public and company interests became more pronounced, this influence became more and more evident at a public level. There was, therefore, some confrontation between the company and public bodies during the market transition period from 1996 to 1998, which demonstrated the existence of tension regarding the design of the sector. In contrast, the new operators demonstrated a wide diversity of strategies. While some began to operate immediately in the sector during the transition years, taking advantage of the liberalised spaces (and even exploiting some poorly regulated areas), other operators concentrated on developing strategies of a more political nature, looking for politico-corporate go-betweens. They expected that these strategies would allow them to benefit from certain favouritism from the public powers.

Both strategies converged following formal liberalisation at the end of 1998 (having produced winners and losers among the followers of both strategies). In general, their positions coincided in the defence of asymmetric regulation as a way of competing against Telefónica.

Another group of actors is made up of business associations. In general, the associative structure of the sector was very fragmented, with small or medium-sized associations generally connected to very specific subsectors. Some associations were even in confrontation because of the type of activity they carried out (e.g. electronic goods importers vs. exporters). The period studied was highly volatile as far as organisations were concerned, with some small organisations disappearing and other new ones coming into being. The rapid technological development of the sector is one of the more evident reasons for this volatility. For example, the associations connected to the subsector of Videotext, which was quite active in 1996, had all disappeared by 1999, while by this very year, new Internet-related associations had already come into existence. Apart from this, the large corporate intersectorial associations did not have a very belligerent position regarding sector policies due to, for one, the different opinions the companies themselves had on this subject. In contrast, during the transition period, there was a very active group of business users, who, as important telecommunications clients, shared a common interest in the introduction of competition as a formula towards achieving large savings.

The trade unions had a strong power of representation within Telefónica, which they by and large maintained during this entire period. Within the rest of the sector, and especially in the newly emerging operating companies, the presence of trade unions was virtually non-existent. Their positions tended to be relatively conservative as far as the transformation of the sector was concerned and they considered that the process of sectorial change was too fast and radical and that it could endanger existing jobs in the sector. Their opposition to the privatisation of Telefónica was limited - as it was already a private company, both legally and historically, its complete privatisation did not imply a large conceptual transformation. In contrast, the majority of workers in the company had held company shares for some time. As far as job losses were concerned, the expansion of the sector and the growth of the company meant that there were no traumatic job losses during the entire transition process.

Another significant group of actors in the Spanish sectorial network within the corporate sector and connected to the area of telecommunications was the group of equipment manufacturers, services companies and large financial groups. These companies, each with its own business strategy, showed a strong interest in participating in the deliberation processes on sector policies during the entire period studied. In some cases, it was a question of identifying investment opportunities in a fast-growing sector and in other cases, it was a question of looking for privileged contacts to take advantage of new regulations which were being introduced. At times, these companies entered the policy networks for a short time in order to obtain more information to be able to better form their strategies.

Another of the identified groups is the mass media. Their interest in participating in the 'policy network' was based on their perception of the growing process of convergence between telecommunications and the media. In particular, the large media groups of the country joined debates and formed business and political alliances with other actors in the policy network, including the public actors themselves. This gave rise to a large politico-mass media confrontation concerning digital TV via satellite during 1997 and 1998

Finally, as regards the group of consultants and investigation centres, it must be recognised that their presence as actors with a marked identity was very limited. This was especially so during the early period, when numerous actors were trying to enter the network, or to just gather information on this. Consultant companies exercised the task of spreading and intermediating basic information on the sector and they also introduced comparative data from other countries. The role of juridical consultant companies should be highlighted, as it was significantly important as a guide to new companies wishing to enter the market during the transition period towards competition. However, as the group of actors matured, the role of consultant companies became increasingly limited, and their purely technical function took over. It should be mentioned that the presence of think tanks, or specialised investigation centres was entirely residual as the only centre specialised in such issues, and financed by Telefónica, was dismantled as soon as the privatisation of the company was finalised in the year 1997.

In this section, we are going to concentrate on the degree of centrality of state institutions within the policy network. Our aim is to revise the hypothesis concerning the strengthening of state institutions as a whole during the liberalisation process of telecommunications in Spain during the advancing formation of the network of actors.

The first factor we used to evaluate the strength of state institutions was the opinion of the actors who made up the policy network about the degree to which the regulating institutions depended on the dominant operator, Telefónica. We were interested in observing how the actors' perceptions had evolved. If, in both surveys, the majority considered that the level of dependency was very high, it could be thought that during the process of change there was an important degree of a capture on the part of public institutions. This would probably reduce its centrality and its capacity to manage the public policies of the sector.

In the first survey in 1996, there was a widespread perception that there was a relationship of dependency between the 'Secretaría General de Comunicaciones' (MOPTMA) and the Telefónica monopoly. 88% of sectorial agents believed that there was a total (37%) or partial (57%) dependency between the regulator and Telefónica. This concept was even widely recognised by representatives of the public organisations themselves. Added to this opinion, was the widespread perception that the defence of Telefónica's interests had a braking effect on the development of telecommunications in Spain. This idea was to be found in all types of organisations, with percentages exceeding 50% in the majority of cases, although it was trade unions and political parties that least criticised Telefónica's role. In the second survey at the end of 1999, this evaluation of dependency between the regulating bodies and Telefónica had changed substantially. It was still high (71% of those questioned thought there to be a total or partial dependency) but it had fallen in relation to 1996 and moreover, the reduction was higher among actors with a more direct involvement in sector policies, i.e. the operating companies, the business sector and the public administration itself.

This perception of the lack of independence of the agents responsible for the regulation of telecommunications was linked to actors' evaluations of a further two factors: the resources available to the public administration to carry out its tasks and the degree of definition of its policy objectives. The perception of the resources available to public organisations, in economic terms and in their capacity to generate information, is an important factor for generating confidence in the power of public intervention as well as in how to calibrate their autonomy.

In the answers to the 1996 survey, there was a marked difference of opinion between those actors who considered that the Spanish regulating bodies had enough resources and those who thought that the resources were insufficient. Oddly enough, among those who were of a negative opinion were the operators and the consultant companies, precisely those organisations which had more information on the sector. Lack of resources was a factor that obstructed independent action by the regulator as it made it dependent on the information provided by the actual companies in the sector, in particular Telefónica. In 1999, an important change was detected in the evaluation made by the actors of public administration resources. The majority of actors in the sector (68%) considered that the public institutions had sufficient resources to fulfil their function. Here, the change of opinion was clear even in the case of the telecommunications operators and sector analysts. In this sense a positive evaluation of action by the new semi-independent regulator and its capacity to control the sector was reflected.

As far as the evaluation of the degree of definition of the state institutions' objectives was concerned, it can be understood that this issue also reflected a strengthening of these. As the objectives were becoming clearer, actors were able to implicitly recognise the state's capacity to impose them. In this sense, the changes in the actors' answers between 1996 and 1999 are highly relevant. In 1996 we can observe a clear two-way split in the answers to this question: the percentage of people questioned who held opposing opinions was very similar - 41% classified the objectives as clear and 59% as confusing. Political parties, operators and business sectors were more critical. In contrast, at the end of the nineties, the evaluation of the definition of the regulator's objectives was still negative (about 40% of those questioned still held that the objectives were clear) but the response was more subtle with a predominance of the answer 'not very clear' as opposed to the answer 'confusing'. Apart from this, we can see that the groups that had criticised the definition of objectives of the public institutions had changed. Now all the operators had critical opinions, while the business sectors and the actual administration evaluated the definition of public objectives more positively.

There is a final factor regarding the centrality of public institutions. Although it was distinct from the state institutions, it should be pointed out that the actors' perception of the role played by European institutions, especially the Commission, was a conditioning factor in the liberalisation of telecommunications in Spain. In the 1996 survey, there was almost absolute unanimity: 92% of those questioned considered the European Commission's action to promote the opening up of the telecommunications market in Spain to be an important or determining factor. This opinion was extendable to the different groups of actors, and was very strongly perceived by the operators, the administration itself and the unions.

At the end of the nineties, this perception of the influence of European institutions on Spanish telecommunications policies still existed but it was much subtler. The proportion of those questioned who valued the role of the European institutions as determining had fallen substantially, especially among the group of operators and public administration. Aside from this, this change in the actors' perceptions still coincides with the previously expressed thesis on the redefinition of the policymaking spaces within the sector regarding the structure of the multilevel government that was being formed at the end of the nineties.

The second hypothesis which we wish to test is related to the transformation of the type of 'policy network' present in the telecommunications sector in Spain during the nineties. Therefore, if we were to classify the situation existing at the beginning of the nineties we could talk about the existence of a very closed 'community policy', made up exclusively of state agencies (Ministries and Telefónica), where one of the agencies (Telefónica) had a certain capture capacity over the other public agencies. Aside from this, Telefónica held a special status as it was formally a private company and acted as such in many aspects. On the other hand though, it was controlled by the government, which held 33% of the shares (the rest were widely scattered, circulating mainly on the stock exchange and in pension funds) and nominated its president. In line with Van Vaarden's classification (1992) we could talk about an exclusive-type Statism, although including the peculiarities already mentioned.

Our argument is that at the end of the nineties the policy network in the telecommunications sector had changed completely. We were before a model of sector government of quite a pluralist nature, where private companies and special interest groups had a strong participation, although there was no delegation of public authority, basically due to the fact that they had to dramatically reduce the definition of public matters in the sector as a whole. There was intense conflict within the sector among different private actors, fundamentally as far as the regulation policy issues were concerned, and there were also distinct visions within the public realm. All in all, it was a much more complex panorama, although it was not a completely open model of pluralism. We found many limitations in the sector's public debate issues in that many interests were filtered and excluded before they were considered by the decision centres of the policy network. In this sense, we can state out that only a reduced group of actors had a significant and stable degree of influence regarding decisions concerning policy making in the sector.

The network of actors examined in the 1996 survey was heavily divided. On the one hand, there was the 'policy community' made up of Telefónica and the other state agencies which formed the heart of the network controlling almost exclusively the pace of market liberalisation. On the other hand, there was a wide group of very different organisations that tried to introduce various initiatives to influence the characteristics of the market transition of the sector. However, the influence of these organisations was very limited, with the exception of some business and finance groups, which traditionally already had access to the core of political power. The policy community was not as strongly united as it had been at the beginning of the nineties, despite the close collaboration which had been established internally between Telefónica and the Spanish public administration on many issues since the early nineties (the covering of a universal service, the management of European programmes, the development of technical regulations, etc..). Some government sectors were somewhat reticent to accept the strategy defended by Telefónica, and were relatively open to new arguments put forward by some actors from the business scene. Basically, the essential point was the question whether the government should continue to fully protect and help Telefónica as it had been doing up to then in order to facilitate its conversion to a large multinational, or whether it should participate actively in the creation of the new market, considering that Telefónica was already sufficiently capable of resisting the introduction of competition.

The organisations wishing to enter the policy network were very diverse. The majority, however, had one characteristic in common: their new interest in the sector. With the exception of the operators (basically foreign, apart from Telefónica) almost all of these actors were going through a process of intense learning of all the relevant technical, political and economic aspects, as their starting point was far removed from the traditional structure of the telecommunications network. Some actors tried to approach the policy community with more intensity with a view to increasing the fracture between political figures and Telefónica (for example, by demonstrating that Telefónica was not giving all the information it had to the government). These actors, who were few and far between, were often (business) users associations or some form of media. The 1996 survey was carried out just after the parliamentary debates on the cable telecommunications law which took telecommunications liberalisation issues into a much wider arena, filling even the front pages of the newspapers. This 'big bang' in telecommunications policies evidently attracted the attention of numerous politically sensitive actors. Some were looking to find a strategic position during the time of formation of new power spaces (or at least this is how it was perceived, given that there were some certain exaggerated expectations in some cases) with the hope of obtaining future yields. For many actors in general, it was a matter of learning the dynamics and identifying where the other members of the network's positions lay.

The policy network examined in 1999 stood out because of the rupture of the traditional relationship between the government and Telefónica. Although both were the most powerful actors, in contrast to the situation existent in 1996, they were now in a situation of non-cooperation, with a lasting conflict regarding the asymmetric regulation policy applied by the government and in particular the CMT. Later, Telefónica formed a strategy to put pressure on and threaten the government while the government avoided direct confrontation with the operator through the application of contention strategies. The complete privatisation of Telefónica is a key factor to explain this change in behaviour, although it must be recognised that it was the government which had changed its strategic line with regard to the market transition process.

Thus, the group of actors with the greatest influence in policy making in 1999 presented a more complex structure. Firstly, due to the presence of the CMT, which had certain room for autonomous manoeuvre and secondly due to the presence of other well established operators such as Retevisión and Airtel or the business groups behind these new operators (and also behind Telefónica) like some banks, foreign operators, and also large electric companies, which gave them large influential power. This group was much less united than the previous group of 1996, and there was continuous tension about how to continue modulating the market transition process. In general terms, there was consensus on the establishment of a model of limited competition, so as not to overly facilitate the entry of new companies in search of new market opportunities. However, there were diverging opinions: for example, the position of the CMT was more in favour of opening up the market than the Ministry. Apart from this, the new liberalisation framework promoted by the Commission made it impossible to completely 'close up' the market to the operators already established there and new business initiatives were continually appearing in search of a niche in the market.

It should be highlighted that the new policy network, in spite of its increased degree of pluralism, was quite limited to business interests of different types and to public institutions, but this did not result in an overly pluralistic collective action, providing greater access to other types of actors such as trade unions, users or even political parties. The evaluation of the 'Consejo Asesor de las Telecomunicaciones', a government advisory organ made up of a wide group of representatives from telecommunications economic and social interest groups was an example of the difficulties many actors had to influence policy decision-making processes.

In 1996, 40% of those interviewed considered that it was a useless institution, with no interest, in that the policy decisions went no further than there, while in 1999 it can be observed that its interest had been re-evaluated, in particular by a number of professional and social sectors which stressed its usefulness in providing them with access to contacts and information relevant to sector policies.

More specifically, it is interesting to observe how other actors evaluated the positions of the political parties. In 1996, when Parliament began to debate the different legislative initiatives for liberalisation, only 45% of actors considered that political parties had elaborate opinions on sector policies and there was a predominance of negative opinions about their capacity, especially among operators and business associations. Later, in the 1999 survey, when the liberalisation laws had been passed in Parliament, opinions of the parties were even more negative. Another example of the policy network beginning to close up with regards certain groups of actors is the case of the 'Comunidades Autónomas' (CCAA), as the regional governments are known as in Spain. In this case, 35% of the actors questioned in 1996 were in favour of increasing the CCAA's participation in telecommunications regulative issues, while in the 1999 survey this wave of opinion had almost disappeared. For this reason, it can be considered that following liberalisation, the CCAA would have had difficulty in finding allies to increase their presence in the core of policy decision making.

Network measures of actor’s reputation for power and influence can help us to understand better the structure outlined in the previous paragraphs. If we observe figures in annex 1, we will find out that the number of actors having a power reputation bigger than 2 (maximum value was 4) is much lower in the 1996 than the 1999 policy network. In fact, in 1996 three actors reached a value higher than 2: Telefonica, The Ministry, and PSOE, the party in charge, the rest of the actors’ values were quite lower. However, it is possible to observe that the 1999 network has ten actors with a value for power reputation bigger than 2. These configurations are coherent with the hypothesis we set forth previously, claiming that the network structure changed in a way that allowed more actors to be relevant in influencing telecommunications policy. It is very remarkable the existence of a internal diversity of quite powerful actors in the public sector, a new situation that challenges the previous hierarchical role of the government. We can also illustrate these two different policy network configurations using a visual representation of the network structure, in which we interpreted a link when a actor evaluated another actor as having the maximum power (4). This allows us to construct a matrix about subjective power perceptions within the network.

Throughout the decade, as is very evident from the two periods when the survey was carried out on members of the policy network, the issues forming the debate became progressively more diverse and complex. In the first half of the decade, there was just one horizontal split through the entire network of actors, forming two large coalitions - the issue was liberalisation and its rhythm and intensity. The actors of 1996 were, in their majority, critical of the rhythm and content of the opening up and liberalisation of the sector. 66% of the organisations interviewed considered that the reforms which had been brought in were too cautious. The group made up of users, business associations and groups of businesses hoping to enter the Spanish market mainly launched criticism. The trade union associations maintained that the opening up was too radical, in agreement with Telefónica and the government, whose criticism of the liberalisation process was very moderate. There were, therefore two clear groups with issue-related preferences: government, dominant operator and workers with preference for maintaining the status quo, and in contrast, equipment manufacturers, business associations, users, and new potential operators with preference for policies related to increased competition and an opening up of the market.

By the end of the nineties, the opinion of the actors present in the sectorial network about the liberalisation process had changed considerably. Now, 75% of the sectorial agents considered that the change towards liberalisation in Spain was adequate, and this predominant opinion was repeated throughout the different groups of actors. Only the trade unions continued to be critical of the process. The central conflict which the positions of the actors in the policy network had constituted had now disappeared - the different agents had accepted the idea of a fully liberalised market as inevitable, and there were only some divergences with a sector of operators and business groups, who were in favour of an even greater liberalisation of the market. Meanwhile, in the years of market transition, new issues had begun to appear, such as the discussion as to whether the government should give prior shape to the business structure of the market or not. As the market was being liberalised, the issues of regulation were becoming increasingly sophisticated and many more specific issues appeared, each bringing about different stances and generating a certain diversity in the central points of conflict and co-operation. Each of these issues brought together different coalitions of actors, leading to certain variability in the relationship patterns among actors belonging to the policy network.

Among the different issues which constituted central points of conflict within the sector, we can highlight the issues of horizontal regulation in the audio-visual and telecommunications sector, where opinions were clearly divided both in favour of and against the introduction of some regulation on this issue. We can also highlight price regulation issues, where there was intense confrontation between different operators, each of whom looked for support among the other actors in the policy network. Thus, for example, in 1996 there was a strong majority in favour of using the regulation of interconnection fees to prevent new monopolies while in 1999 there were many more opinions that were only partially in favour of this objective. Another area of conflict was the definition of a universal service, especially as far as the provision of Internet services was concerned. Here, the sector was very divided, with almost fifty-fifty in favour of and against the need to offer universal access to the Internet. In general, the operators were very opposed to these issues, whereas the groups of users and service companies, for distinct reasons, saw many opportunities in these policy options.

Dissatisfaction with the level of public debate on the development of telecommunications policies in Spain was an aspect which was clearly stated both in 1996 and 1999. The majority of the groups shared this opinion, with the exception of political parties and public administrations. Significantly, actor dissatisfaction increased as the policy network was consolidated. In 1996, 30% considered that the level of discussion had been 'insufficient' and in 1999 this had risen to 57%. One factor that can help to understand this trend is that in 1999 the debates were more technical and linked to regulating measures while in 1996 policy options were discussed at length. However, this dissatisfaction also confirms the non-existence of forums specialised in public debate or indeed think tanks with capacity to make discussion on sector policies more dynamic.

A key subject of discussion on sector policies during the second half of the nineties was the question of state intervention to facilitate the formation of a market with a predominant presence of national operators. In the 1996 survey, 71% of actors in the sector considered it to be very important that, in the face of growing globalisation within the telecommunications market, the government should procure that the ownership of telecommunications companies was linked to national interests. All types of organisations defended this general opinion, although the operators and business sector were less enthusiastic. Regarding the question as to whether the state should actively intervene to prevent private companies with a predominantly foreign capital from controlling the principal telecommunications operators in Spain, the answer was yes. 66% of those interviewed approved of state intervention in this area and just over a third (35%) rejected any type of intervention in the capital make-up of telecommunications operators. As was to be expected, the private sector was slower to accept the legitimisation of state intervention (only 16% accepted), while public organisations were clearly those most in favour of these measures (79% approved). Despite some differences, it is evident that there was basic consensus on how to direct policies regarding this issue. In this sense, the operation promoted by the government to form a new public operator, Retevisión, to compete with Telefónica, in the form of a duopoly during the two years prior to complete liberalisation (December 1998) must be understood within the context of generalised apprehension about the possible loss of corporate control of the sector by Spanish business groups. In the second survey at the end of the nineties, the sector's evaluation of this issue had changed considerably, splitting the basic consensus that had existed. Many actors ceased to support this type of policy, and only 41% of the sector though it necessary to continue controlling the link of telecommunications companies to national interests. The issue of the introduction of some protectionist policies in specific subsectors also divided the sector in the same way, and division on this issue remained constant throughout the period from 1996 to 1999.

To conclude our analysis of cooperation and conflict patterns within the sectorial policy network, we have summarised in two tables the stances of the actors present in the Spanish telecommunications sectorial area in 1996 and 1999, years in which, we recall, basically coincide with the beginning and the end of the most intense period of market liberalisation. These tables show the degree of interest (high, medium or low) of the different groups of actors in a selection of particularly conflictive or controversial issues regarding telecommunications in Spain, regardless of the specific orientation of each group, ie. whether they were in favour or opposed to the issue in question. In line with this examination, a more detailed analysis could lead us to analyse the emergence of certain issue networks formed as the area of telecommunications policies became more complex and diverse. However, for the moment, we aim only to identify the core issues or highly intense interest areas for some groups of actors between 1996 and 1999, with the aim of discussing the development of issues of debate which formed and united the policy network between 1996 and 1999. As we have already said, our hypothesis is that in line with liberalisation, the subjects of debate in the policy network have become more sophisticated and characterised by greater diversification and specialisation in the areas of interest.

Apart from the grouping of interests shown in table 1, we can also detect two groups of actors - according to their areas of interest - who were present during the first stage of the liberalisation process in the sector. On the one hand, the regulating institutions, the dominant operator and the trade unions showed a high interest in issues related to the protection of employment, and promoting initiatives associated with protectionist ideas or the defence of national interests. On the other hand, users and business groups showed a high interest in initiatives associated with increased competition and the regulation of rates. To a certain degree, therefore, we could talk about the beginning of a split in the policy network, with one side oriented more towards issues of a business nature and the other more oriented towards purely regulation themes.

If we observe the grouping of interests in the survey of 1999, we are before quite a different situation. By the end of the nineties, the interests of certain actors had changed significantly. Instead of finding a split into two groups in which the actors and interests coincided, we now find a regrouping of the majority of actors on regulation and competition issues, while interest in issues related to market structure and protectionist policies has reduced. In other words, it can be considered that following the liberalisation of the market, the policy network had reduced the range of policies and means of intervention forming the main interests of the actors present in the policy network. However, as we have seen, there were many internal differences among the actors and we can find more points of conflict and co-operation in comparison with 1996. This is still, however, a sign that the policy network's make-up became increasingly stable once the market had been liberalised. The range of policies which hold the interest of the policy network is still focused, however, on those policies which the national states can continue to bring in within the multilevel government of the sector, now transformed to a European scale.

In this sense, from 1999 onwards, we suppose that the transformation of the network has already become more difficult, following the important changes which appeared during the regulation transition period. In any case, the large size and the importance of the sector may lead to the formation of some issue networks, breaking off somewhat from the central telecommunications policy network, including new more specific actors with a certain capacity to promote decisions on specific problems and issues. Internet-related issues are clearly one of the areas which fulfil these type of conditions.

As we have seen, at the beginning of the nineties, the structure of the sectorial policy network in Spain did not offer incentives to modify the barriers to entering the Spanish telecommunications market, a factor which favoured Telefónica alone. Why did the policy network change between 1996 and 1999, during the years of market transition? Why did it change, as we have shown, towards a limited pluralism? Evidently, an initial and very generalised explanation is the importance of external influence, which we could identify with the entire new regulation framework promoted by the EC. From this point of view, therefore, it would be necessary to explain that the demands to eliminate the monopoly and introduce competition provoked the rupture of the initial core group of actors, in the exclusive-type Statist model and facilitated the entry of other actors to the policy network.

Even so, this explanation does not give us a sufficiently convincing reason to indicate why some actors and not others entered the policy network with more force, or why this new policy network acquired an internal dynamism which was less cohesive, with the appearance of persistent tension and conflict. It is necessary to find other more explanatory arguments. An argument complementary to that of external influence would be the existence of changes in the nodes of the network, brought about or by intentional or non-intentional strategies by the actors (some successful and others not so) - to move from one position to another and achieve greater benefits.

This line of explanation gives food for thought but it comes up against the problem that it presupposes that the rules of the game were clearly laid down and that everybody had a certain idea of the results of their strategies (always with the doubt caused by the ignorance of the actions of others). However, this was not the case during the years of changes in sector regulation. In the market transition process, the new rules of the game were still being formed, and it was not certain that the old ones were still in play. It could even be remarked upon that it was a typical case of doubts that are common in the processes of political transition.

It is necessary then to test out another explanation - in this case, that of the introduction of institutional variables. From this angle, we could consider that the introduction of new institutional rules during the period 1996-1999 was the critical variable to explain how the policy network took shape during the second stage. If the new institutional rules had been distinct (we are not referring to the rules of the functioning of the market, but rather to the rules which constituted it) the make-up of actors with a greater influence in the policy network would also have been different. From this perspective, we can identify the appearance, at a key moment, of institutional innovation in Spain as far as the telecommunications sector was concerned, with the coming to power of the 'Partido Popular' in mid-1996. This could be classed as a turning point (rapid change in the key rules of the game). This moment, because of the combined impact of the different decisions contemplated, will have important effects on the structuring of the network of actors and for years to come.

More specifically, these above-mentioned decisions consisted of the set of measures which the 'Partido Popular' took immediately following its coming to power. This was a time when Telefónica was still without a new leading power, and had therefore, temporarily lost its capacity to foster the privileged relationship with the government (which had also evidently changed). At this precise moment of change of people and teams, what brought about the change was, more than a clearly determined political option or a different government ideology, the fact that the central actors of the network (including Telefónica) had been inactivated during the first few months of change of government. This permitted other actors, who had up to then had substantially less influence, to introduce their ideas in the key decision areas, achieving thus, that the new government accepted to quickly announce a set of measures (which had immediate effect) for the institutional formation of the government's telecommunications regime. These measures were basically the creation of the CMT as a regulating agency, the granting of a telephone licence to Retevisión (a public company which distributed television signals and had a very reduced value then) to establish a duopoly during the transition towards liberalisation, the introduction of the beginning of asymmetric regulation against Telefónica, as well as their own complete privatisation.

The conversion from a government executive law to a formal Parliament law took over a year. As Telefónica had returned to its traditional position, re-establishing its link with the government, it pressured against these new measures. However, the decisions had already taken and implemented, and this brought about a set of new events, which in the years to come, contributed to forming the structure of the policy network, as can be seen in the survey carried out at the end of 1999.

In the first place, with the direct support of the government, the creation of Retevisión as a new player of importance in the telecommunications scene introduced an enormous change with respect to the previous stage. A company now existed which had to be protected and given great attention and support given the fact that it was just beginning and it had to find business partners who could provide long-term stability for the development of the company. For obvious reasons, these partners had to be different from those who had formed around Telefónica, and who were going to take charge of its strategic management during its privatisation (two banking entities - BBV and La Caixa). In order to find partners, the government had to concede practically everything it had already permitted to Telefónica's partners (through Telefónica), ie. more direct access to decision-taking in telecommunications policies. Apart from this, the creation of new public regulating bodies complicated the process of governmental decisions. Basically, we are referring to the creation of the CMT, the Spanish semi-independent regulating agency. The positions of this agency did not always coincide with those of the Ministry.

Therefore, it was the creation of Retevisión and the emergence of new partners, together with the creation of the CMT that brought about an increase in the number of actors with real influence within the policy network and the fact that its internal decision- making was subject to more tension. With the advance of the liberalisation process, the other actors in the policy network found that they had to face different decision-making cores and develop more complex strategies to be able to have influence, or at least to be visible in the public scene. However, they were not faced with a very open network: the dominant actors had an extremely predominant business orientation, and their values were of a technological and economic nature.

Aldrich, H. A. (1979), Organizations and environments. Englewood Cliffs, New Jersey. Prentice: Hall.

Dowding, K. (1995), "Model or Metaphor? A Critical Review of the Policy Network Approach", Political Studies, 43, 136-158.

Hanf, K., / Scharpf, F. W (1978), Interorganizationalpolicy making: Limits to coordination and central control. London: Sage.

Hulsink, W. (1999), Privatisation and liberalisation in European Telecommunications, London: Routledge.

Jordana, J. (1998), "La política de telecomunicaciones en España : del monopolio a la transición al mercado" in Gomà, R. / Subirats, J. (Eds) Políticas Públicas en España. Barcelona: Ariel.

Jordana, J. / Domingo, T. / Macià, S. / Salvador, E. / Sancho, D. (1997), Opiniones y actitudes sobre las políticas de telecomunicaciones en España 1996. Barcelona: Departament de Ciencies Polítiques i Socials, Universitat Pompeu Fabra.

Jordana, J. / Ruiz, C. / Sancho, D. / Sancho, D. / Sanz, A. (2000), Opiniones y actitudes sobre las políticas de telecomunicaciones en España 2000. Barcelona: Departament de Ciencies Polítiques i Socials, Universitat Pompeu Fabra.

Klijn, E. / Teisman, G. (1991), "Effective policy making in a multi actor setting:networks and steering". In Veld / Schaap / Termeer / Twist (Eds .), Autopoiesis and configuration theory: New approaches to societal steering (pp. 99-112). Dordrecht: Klawer.

König, T. (1998), "Modeling Policy Networks", Journal of Theoretical Politics, 10 (4), 387-88.

Laumann, E. / Knoke, D. (1987), The organizalional state: Social choice in national policy domains. Madison: University of Wisoonsin Press.

Levi-Faur, D. (1999), "Governing Dutch telecommunications reform: state-business interactions in the transformation of national policy regimes to (European) embedded policy regimes", Journal of European Public Policy, 6 (1), 102-122.

Marin, B. / Mayntz, R. (1991), " Introduction : Studying Pollicy networks" In Marin, B. / Mayntz, R. Eds. Policy networks : Empirical Evidence an Theoretical Cosiderations. Frankfurt: Verlag / Westview Press.

Noam, E. (1992), Telecommunications in Europe, Oxford: Oxford University Press.

Pfeiffer, G. (1990), Telecommunications in Germany: an Economic Perspective. Francfurt: Springer.

Pierson, P. (2000), "The Limits of Desing: Explaining Institutional Origins and Change", Governance, 13 (4), 475-499.

Scharpf, F. W. (1978), "Interaganizational policy studies. Issues, concepts and perspectives". In K. Hanf and F. W. Scharpf, Interorganizational policy-making: Limits to coordination and central control,. London: Sage, pp. 345-370.

Scharpf, F. W. (2001), "Notes Toward a Theory of Multilevel Governing in Europe", Scandinavian Political Studies, 24 (1) 1-26.

Schneider, V. (1993), "Games and Networks in Large Technical Systems: The Case of Videotex", in Fritz Scharpf (Ed.), Games in Hierarchies and Networks. Frankfurt: Campus.

Schneider, V. (2002), "The Institutional Transformation of Telecommunications Between Europeanization and Globalization", in J. Jordana (Ed.), Governing Telecommunications and the New Information Society in Europe, Glasgow: Edgar Edward (in print).

Thatcher, M. (1999), The Politics of Telecommunications, Oxford: Oxford University Press.

Van Waarden, Franz (1992), "Dimensions and Types of Policy Networks", European Journal of Political Research, 21, 29-52.

Thatcher, M. (2000), "Cross-national Linkages and Regulatory Reform in Europe: The Case of Telecommunications", Annual Meeting of the American Political Science Association, Marriot Wardman Park, August-September, 2000.