Subnational Sweden, the national state and the EU

ABSTRACT

The article examines to what extent the position of the Swedish state vis-à-vis the European Union (EU) has been mediated by domestic factors, and in particular how this has affected its subnational authorities, since Sweden joined the EU in 1995. It examines whether path-dependency prevented Sweden from converging with other member states. The article then examines the relevance of subnational mobilization for the relationship of the Swedish state to the EU. It provides an overview of bottom-up activism, subnational capacity building and top-down influence of the EU over conditions at the local and regional levels in Sweden. It is argued that the traditionally strong position of subnational authorities in Sweden, its institutional and administrative culture and a favourable economic situation have mediated EU influence over the Swedish state. This has also made it possible for local and regional government to bypass the state in its activities vis-à-vis the EU.

KEYWORDS:

• Sweden
• decentralization
• regionalization
• subnational authorities
• regional government

Introduction

Sweden joined the European Union (EU) in 1995, during a window of opportunity created by the breakdown of East European communism and a financial crisis. Until then, membership had been ruled out as incompatible with the Swedish policy of neutrality as the EU was seen as a part of the Western camp. However, the extensive financial crisis that hit Sweden in the early 1990s clearly illustrated how difficult it was for a small country to act alone in an increasingly globalized world.

The first time an official intention to join the EU was mentioned was in October 1990, when the Social Democratic government announced its aim to apply for membership, as one of several measures to counteract the crisis. It was thought that Sweden could no longer be an independent player and it was therefore necessary to join forces with the rest of Europe. This position was overwhelmingly supported by most of the political parties and confirmed by a majority of the citizens in a referendum in 1994. However, the next step in integration, to replace the Krona with the Euro, was rejected in a referendum in 2003 (Hilson 2008; Östberg and Andersson 2013).

More than twenty years after joining, Sweden has adjusted to the role of an established, reliable and pragmatic member of the EU (Miles 2005; Tallberg 2010; Michalski 2013; Jacobsson and Sundström 2016; Statskontoret 2016a). Staying outside the Eurozone does not seem to have limited the country’s influence in Europe. A recent study of networking in the EU suggests that Sweden continues to have access to decision-making, bargaining power and network capital (Naurin and Lindahl 2010). It is a net financial contributor giving EUR 2.5 billion in 2015, but receives support for agriculture and for its sparsely populated terriories. The Swedish population’s knowledge about the EU has increased (Utredningen om delaktighet i EU 2016). They have also become more EU-friendly. Immediately after the referendum to join the EU, Swedish citizens were among the most reluctant among the member states, but support has gradually increased, and from 2001 there has been a steadily growing majority in favour of Swedish membership (Oscarsson and Bergström 2016). Indeed in May 2017, despite the recent European crises, 55% were for and 19% against membership, which is among the highest levels recorded. However, the level of support varies in different parts of the country. Originally, most people living in the north wanted to leave, but there is now a clear majority in all parts of the country in support of Swedish membership. Among the political parties, most were already in favour at the time of the referendum to join, and the Environmental Party has recently changed its position and is now pro-EU. Only two parties want Sweden to leave – the Left Party and the populist Sweden Democrats (in Parliament from 2010). Despite the growth of the latter party, among citizens as well as policy-makers, Sweden has become well integrated into the EU.

In a well-used textbook on Sweden and the EU (Tallberg et al., 2010), the role of local and regional government is hardly mentioned. However, since both local and regional government have a strong position in the Swedish polity, we would expect them to be part of this integration with the EU (Sellers and Lidström 2007). This article aims at investigating subnational mobilization in Sweden vis-à-vis the EU. It will focus on two issues: First, to what extent and in what way is the position of Swedish subnational authorities affected by Swedish membership in the EU? Second, to what extent have subnational authorities mediated the position of the Swedish state in a way that restrains it from converging with other member states?¹ Although Sweden has been a centralized unitary state from the sixteenth century, central control has always been balanced by strong local communities and included collaboration between local and national levels. In the seventeenth century, the peasantry gained representation in the national assembly and at times supported the King against the nobility. With the establishment of local self-government in the nineteenth century, and in particular with the growth of the welfare state, local government became increasingly tied to the state as implementers of national welfare policies. Hence, characterizing Sweden as a ‘decentralized unitary state’ is highly appropriate (Loughlin and Peters 1997). Although local government has significant powers of its own, in financial terms its main role is to carry out public social, health and education welfare policies. Central government has been able to rely on local politicians to implement these policies in an effective way (Sellers, Lidström, and Bae 2018).

Swedish membership of the EU may, potentially, disturb this relationship. Rather than just a two-level game, local government is now much more part of a system of multi-level governance, i.e. ‘a system of continuous negotiation among nested governments at several territorial tiers’ (Marks 1993, 392), in which ‘supranational, national, regional, and local governments are enmeshed in territorially overarching policy networks’ (Marks 1993, 402–403). Although the Swedish welfare system remains within the competences of national policy-making, the EU may offer alternative channels for local government, which can undermine not only the links between local and national levels of government, but may also have consequences for the position of the state.

The theoretical basis for this study is a historical-institutional perspective that emphasizes path-dependencies and continuity. When faced with new situations, actors are expected to look at existing practices and solutions rather than to try something completely new. However, under certain circumstances, such as external pressure or internal collapse, a window of opportunity opens for more substantive change (Peters 2012). The key question in this article is whether Sweden joining the EU has been such a formative moment, for subnational authorities as well as for their relationship with the state. Complying with the theoretical framework of this Special Issue, Europeanization is seen as a process of EU influence over national matters, although it is recognized that there are other ways of using this concept represented in the Europeanization literature (Börzel 2002).

The remainder of the article is organized as follows. Firstly, the system of subnational governance in Sweden is briefly described. This is followed by two sections that analyze subnational mobilization vis-à-vis the EU, i.e. how local and regional government influence and are influenced by EU decision-making, and the extent to which these activities have mediated the impact of the EU on Sweden. A final section summarizes the general picture and addresses the question of whether the position of the state has been undermined or supported by the mobilization of subnational authorities. The basis for the analysis is previous empirical research into these matters combined with studies of government documents. It is concluded that both top-down and bottom-up mediation has occurred in Sweden. There has been a European influence with consequences for decentralized institution-building but the EU has also triggered significant bottom-up activism. This may be seen as a path dependency exerted by units of subnational government with strong legitimacy, but is further emphasized by their capacity building and by institutional and administrative culture.

Subnational authorities in Sweden

Local government in Sweden consists of two tiers – 290 municipalities and 21 county councils/regions).² Using a system of proportional representation, the councils are elected for four-year terms. It is compulsory for councillors to be members of political parties, and more than 90% of them represent the same parties as those in Parliament. Political decision-making is collective so there are no mayors or any other politicians with independent decision-making powers. Municipalities and county councils/regions are financed by a proportional income tax and by grants from central government. A tax equalization system evens out differences between affluent areas and those where incomes are lower and the needs for local public services are greater (Lidström 2016).

The municipal level has more resources and a wider range of functions than county councils/regions. They are also closest to the citizens and their decision-makers are better known by the general public. They vary in size from Bjurholm with 2460 inhabitants to Stockholm with 939,000 inhabitants. In order to be able to carry out more responsibilities in, primarily, the welfare sector, municipalities have been twice amalgamated into larger units, the first time in 1952 and the second during a period in the early 1970s. Major municipal functions include primary and secondary education, social services, day nurseries and care of the elderly. They are responsible for local infrastructure, including streets, parks, recreation and culture. They are also in charge of land-use planning and building permits.

The intermediate tier of elected government is currently undergoing a transformation. Until the late 1990s, this consisted of county councils with responsibility for, principally, healthcare. Indeed, at this time a uniform Scandinavian type of welfare county existed, that could be found also in Denmark and Norway (Torfing, Lidström, and Røiseland 2015). Partly inspired by Swedish EU membership, the two regions of Västra Götaland and Skåne were established at the end of the 1990s after amalgamating county councils. These also received additional responsibilities for regional development functions from the county administrative boards, which are central government administrations. There have been several attempts to extend this initiative to a comprehensive regional amalgamation reform, but these have all failed (Lidström 2010a). However, what has succeeded has been the transfer of regional development responsibility to the county councils but without any border changes between them. The transfers are followed by a change of name to regions. The smallest unit is Jämtland with 127,000 inhabitants and the largest is Västra Götaland with 1.6 million inhabitants. As the strengthening of the intermediate level of government has clearly been facilitated by Europeanization, we will return to this matter later.

In addition to directly-elected municipalities and county councils/regions, there are also subnational state authorities with an interest in the development of the region. As mentioned above, each county area has a county administrative board, which is the central government unit for general administration and supervision at regional level. For example, they perform control and support functions as regards social services and environmental protection. Contributing to regional development has previously been one of their main tasks but this is gradually being transferred to the county councils/regions. Other central government actors at regional level include universities and colleges of higher education and regional branches of national administrative agencies, for tasks such as transportation, labour market policy, forestry and agriculture.

Subnational attempts to influence EU institutions

The Local Government Act,³ which set the legal framework for both municipalities and county councils/regions stipulates that local government activities have to concern the local community or territory. Despite this rule, attempts to influence EU decision-making have generally been seen as legal and within the remit of local authorities, even if they include activities carried out in a different country. Even when the legality of an activity has been questioned, it has often been possible to find ways around the objection. Indeed, international activities of local government have even resulted in an amendment of the Act to make it possible for local government to send surplus material abroad as international aid (Lindquist, Lundin, and Madell 2016).

National policy-makers have not explicitly restricted subnational governments from being active in relation to the EU. There could, potentially, be a conflict of interest if local government expressed views or carried out activities that differed from official Swedish policy. However, as yet no such case has occurred. Neither has any government issued any policy on how local government should behave as regards such relations.⁴ On the contrary, governments have mainly been facilitators. For example, the county administrative boards, i.e. the extended arm of the state in the regions, are often involved as partners in EU related activities. Members of the government have also taken their own initiatives. For example, a Social Democratic minister of infrastructure and regional affairs set up a reference group of regional politicians who met to discuss regional and national development policies vis-à-vis the EU. The Minister in charge of European Affairs in a non-socialist government expressed support for regions that are active vis-à-vis the EU (Lindh 2006).

However, opposing voices have also been heard, even from very important policy-makers. Göran Persson, who was the Social Democratic Prime Minister 1996–2006, has been critical of the international activities of county councils and municipalities. He has suggested that they should focus on their main welfare responsibilities, rather than on extravagant lobbying in Brussels. Also, he has been afraid too-strong regions would undermine the unitary character of the Swedish state, which in the long run could be a threat to the universal welfare system (Svenning 2005). However, he did not initiate any restrictions along these lines during his time in power. Neither has any other government.

This would suggest that the established position and democratic legitimacy of subnational government has functioned as a factor that has mediated the influence of the EU on the Swedish state. Due to its responsibility for providing public welfare, Swedish subnational government is among the strongest in Europe. Total local government spending (both municipalities and county councils/regions) represents 25% of GDP and 44% of total public expenditure (Dexia 2008). However, it also enjoys comparatively high levels of democratic legitimacy. As subnational government is positioned between the citizens and a superordinate level of government it derives its legitimacy from both these sources (Lidström 1998). A comparative analysis that takes into account both these components of local government legitimacy, suggests that Swedish local government enjoys a high degree of legitimacy in relation to both its citizens and the state. This is a condition, which it shares with local government in the other Nordic countries, Estonia and the Netherlands (Lidström and Baldersheim 2016).

There are no signs that there has been a change in the democratic legitimacy of local government in Sweden following membership of the European Union. Indeed, citizens’ support for local government has increased, as previously mentioned this has also been a period during which Swedish citizens have become increasingly EU-friendly. Neither does a high degree of democratically legitimate local government seem to undermine the position of the state vis-à-vis the EU. In fact, the general pattern in the previously mentioned comparative study is that countries with above average legitimate local government systems are also among the strongest members of the EU, for example by being net providers of resources to the EU. These countries include the Nordic member states, France, Germany, Austria and the Benelux countries. Notable exceptions are Italy and the UK, who are net providers of resources but have local government systems with weak legitimacy.

Even before Sweden became a member of the European Union, subnational authorities were active on the international scene. For several decades, twinning with local authorities in other countries has been a common way of establishing links with local government elsewhere in order to promote cultural exchange or express friendship. For many years, subnational authorities have been members of international organizations. However, once the government expressed the intention that Sweden should join the EU, such activities increased significantly. Also, they became more focused, aiming at learning how the EU system functions and exploring ways of influencing decision-making. Hence, by the time Sweden became a member, subnational authorities were already on the scene and well prepared (Jerneck and Gidlund 2001). One of the lessons from the initial years was that in order for bottom-up activities to effectively influence EU policies, several different strategies had to be combined.

These activities were the first steps of a capacity building strategy of subnational authorities. One part of this was to improve the skills and knowledge of EU matters among staff and politicians. Specialized staff were hired and specific politicians were given roles as international representatives of the subnational authority. Municipalities and county councils/regions that wanted to take on such a role needed to be large, i.e. it is not usually something that a small municipality will have the resources to do on its own. EU specialists in different subnational authorities established networks and met with each other, constituting a kind of policy community. It has also meant that EU-oriented activities are primarily elite-driven. This is supported by a recent study, indicating limited knowledge about the EU among local officials and politicians. Only 36% of the local councillors consider themselves to know enough about how EU affects their municipality to be able to discuss it with their electorate. (Utredningen om delaktighet i EU 2016).

Subnational capacity building has made it possible for local and regional governments to be effective when pursuing their interests vis-à-vis EU decision-makers. Several parallel strategies have been used. The most costly for a municipality or a county council is to attempt to influence EU decision-making entirely on their own. For example, a county council may lobby an individual decision-maker in the EU bureaucracy to ensure that an application for financial support is in line with the requirements of the EU and to remind the official of its existence. Capacity building may also include collecting information and developing networks. However, due to the costs, such individual activities are reserved for very specific circumstances when actions are required that cannot be coordinated with others.

Therefore, it is much more common for subnational authorities to work together using a variety of different channels. One such channel would be the information centres that represent Swedish local and regional interests in Brussels. These started to emerge very soon after Sweden became a member of the EU. For example, the two northernmost counties, Västerbotten and Norrbotten, set up the North Sweden European Office that represents the county administrative boards, the county councils, the municipalities, business interests and the universities in the north. It was established in 1997 and monitors northern Sweden’s interests in EU decision making and pursues the regions’ issues in areas such as regional development, transportation and research. There are currently twelve offices in Brussels that represent Swedish local and regional interests.

The information centres work both up-stream and down-stream. Up-stream strategies aim at exerting influence at an early stage of the decision-making process, for example on a new EU directive or a revised cohesion policy. Such strategies may also include arranging seminars that offer information to EU decision-makers and increase knowledge about EU matters in the home regions. Down-stream lobbying, on the other hand, concerns decisions that have been taken by the EU institutions and focuses on influencing their implementation. This may include activities to secure resources (Jerneck and Gidlund 2001). A comparative study of regional lobbying in Brussels has observed that the Nordic and UK offices pursue different strategies than their counterparts from federal and regionalized states. While the Nordic and UK offices are looking for funding opportunities and to collect information, those from federal and regionalized states are more into traditional lobbying in order to influence decisions (Lein-Mathiesen 2004).

Another example of joint activities between different subnational authorities is the holding of conferences on EU-related matters at home. Local and regional representatives meet for one or two days to discuss common challenges in relation to the EU and to discuss strategies on how these can be met. It is also an opportunity for learning about current EU matters. One example is Europaforum Norra Sverige which is arranged jointly between the four northernmost counties in Sweden, and which brings together local, regional, national and European politicians and representatives from business and the county administrative boards. Subnational authorities are also members of international organizations that aim at influencing EU decisions. One such is the Association of European Regions (AER) in which twelve Swedish regions are full members, and which pursues a general regional interest. Another is the Conference of Peripheral and Maritime Regions (CPMR) which has nine Swedish regions as members and which highlights the specific conditions of the peripheries of Europe.

In addition to their own activities, either on their own or as members of international organizations, subnational authorities may also act by delegating powers to national organizations. Perhaps the most important organization in this respect is the Swedish Association of Local Authorities and Regions (SALAR), which is regarded to be one of the strongest local government associations in the world (Feltenius 2016). All Swedish municipalities, county councils and regions are members of this organization which has 440 employees and considerable resources. It works both nationally and internationally to enhance the interests of local government. This includes monitoring EU-related activities, capacity building and training of its own members, and attempts to influence EU decision-making. The Association has its own information office in Brussels and is also represented in European organizations that have a role in relation to the EU, such as the Council of European Municipalities and Regions (CEMR). The Swedish delegation to the EU Committee of Regions is selected by the association, although formally appointed by the government. SALAR also tries to influence the Swedish government’s stance on EU matters (Svenska Kommunförbundet/Landstingsförbundet 2001; Sveriges kommuner och landsting 2007). Hence, Swedish subnational authorities have developed their capacity to influence decisions in the European Union. This capacity is a mediating factor that underlines the position of subnational authorities in the Swedish relationship with the EU.

Swedish subnational politicians and officials were initially hesitant in the meeting with what was seen as an administrative culture that was different from Sweden’s, for example by being more hierarchical and secretive in contrast to the tradition of transparency and openness that is a feature of the Swedish political-administrative model. Transparency and open government has a long historical legacy in Sweden. The first Freedom of Information Act was passed in 1766 and despite occasional temporary setbacks public access to most official documents became a principle in Swedish public administration from the early 1800s. Currently, official documents also include email correspondence between agencies. A further expression of openness is that public sector managers are prohibited from trying to discover the identity of public employees who leak information to the press (Hall 2016).

Although there have been instances when access to information has been restricted, it turned out that the fear had been exaggerated and that instead, Sweden contributed to spreading the practice of open government within the EU (Hillebrandt, Curtin, and Meijer 2014; Hall 2016). The first ten years was a learning period but eventually the Swedish local and regional representatives adapted to the Brussels game. Over the years, the input by Swedish subnational authorities into the EU decision-making process has become more professional and sophisticated. It has also moved from being primarily concerned with extracting resources to a more general interest in EU matters (Lindahl and Berg 2007).

EU impact on subnational authorities in Sweden

A good reason for subnational authorities to be active in relation to the EU is that they are affected by decisions taken by its institutions. In this section we will discuss four such expressions of Europeanization – on the everyday activities of subnational governments, on regional development policies, on the style of policy-making and on regional institutionalization.

EU laws and regulations have a profound impact on the everyday activities of subnational authorities in Sweden. As an early study from the Swedish Agency for Public Management noted, this was initially a surprise to many Swedish politicians, who thought that municipalities and county councils would largely be exempt. It was expected that the EU was about market-related activities, and would not affect the provision of social, health and educational welfare, which are the main tasks of Swedish local government. However, this has turned out to be wrong (Statskontoret 2005).

SALAR has tried to measure the role of the EU vis-à-vis local government in quantitative terms, by assessing how usual it is that matters that relate to the EU affect issues that are on council agendas. The conclusion is that approximately 60% of the issues on the municipal council agendas, and 50% on the county council agendas concern matters which are influenced by the EU. The difference between municipalities and county councils is mainly a result of differences in functions. Healthcare issues dominate the county council agendas, which is an area where the EU has only a limited role (Sveriges kommuner och landsting 2010).

In the analysis, SALAR makes a distinction between two types of influence from the EU – legal influence and political/cultural influence. Legal influence concerns matters that are subject to EU regulation and directives, such as rules for public procurement, regulation of public support to private firms and environmental regulation. Political/cultural influence, on the other hand, concerns areas of responsibility where the EU has no regulatory powers, but aims at providing opportunities for learning through exchange of ideas and best practice. The best-known example of this role is the Open Method of Coordination, applied within the area of social inclusion and social protection. The member states are asked to set up a three-year strategy within these areas and as many of the welfare functions are carried out by local government these are inevitably involved in this process.

A second impact of the EU concerns regional development policies. For many Swedish regions, especially in the rural and peripheral areas, the financial support from Structural Funds and in particular the European Regional Development Fund is the most significant impact of the EU. As required, these resources are complemented by money from subnational authorities and private businesses. During the current programme period (2014–2020), slightly over a third of this support goes to the four northernmost counties. It is noteworthy that support for the union among the general public in these areas has increased in parallel with continuing EU funding.⁵ However, whether regional development funding from the EU actually contributed to this change in opinion has not been investigated empirically.

The structural funds have contributed to changing the focus of Swedish regional policies. Sweden’s membership in the European Union coincided with a major shift in its regional policies. Until the middle of the 1990s these were mainly redistributive, aiming at compensating the most disadvantaged areas of the country. Policy-making was highly centralized with a decisive role for central government. The local and regional tiers were mainly receivers rather than active agents (Hörnström 2010, 2013). At regional level, it was the central government agency, i.e. the county administrative board, that was in charge of regional development policies. Elected units of government at local and regional level were not expected to interfere in this policy area, which was seen as being reserved for the state. Indeed this may be seen as an extension of a Social Democratic Welfare model, aiming at evening out differences not only between individuals but also between different parts of the country (Loughlin and Peters 1997).

However, these policies were transformed around the middle of the 1990s. In line with the ideas of New Regionalism (Loughlin 2007), responsibility for regional development was decentralized to the regional level. Instead of being compensatory, the new policy emphasized that all regions had a responsibility to develop their own assets, in cooperation with other public and with private sector interests in the region. The region should be competitive on its own terms, rather than dependent on state support. Clearly linked to an increasingly globalized economy, the end of the era of the post-war expansion of the public sector and the arrival of a neo-liberal agenda, this shift in policies was also encouraged by the EU (Tanzi and Schuknecht 2000; Lidström 2007). The structural funds explicitly targeted the regional level and emphasized horizontal cooperation and co-financing. During the first two decades, committees headed by the county administrative board representatives and involving local and regional politicians took decisions on allocation of resources from the structural funds at regional level. However, from 2014, these decisions have been transferred to a central government agency – The Swedish Agency for Economic and Regional Growth (Tillväxtverket), which suggests that there have also been instances of recentralization in the Swedish relationship with the EU.

A third expression of EU impact is the adoption of new styles of decision-making among subnational actors. An important part of the institutional and administrative culture in Sweden, in use at least from the 1930s, has been to involve different, affected interests, in particular trade unions and employers’ associations, in public decision-making (Gustafsson 1995). This has often been referred to as neo-corporatism and has been used at local level too (Lewin 1994). With the EU membership, horizontal coordination has been even more pronounced. The structural funds system of regional development has relied on the partnership principle, with joint decision-making between various interested parties at local and regional levels and also joint mobilization of resources for shared projects (Warleigh-Lack and Mccallion 2012). Along similar lines, regional growth agreements were introduced to Sweden in the late 1990s as a new policy tool, inspired by similar agreements in other EU countries. In order to receive central government co-funding, consensus on development goals had to be reached among different regional actors.

Compared to traditional neo-corporatism, the partnership principle is more concerned with joint mobilization of resources and is mainly confined to the field of regional development. Another difference is that neo-corporatism was more strongly directed by the state whereas partnerships are more loosely connected networks of independent actors. However, both the old and new styles have obtaining consensus as a main goal. They were also both criticized for encouraging agreements behind closed doors and were problematic in terms of democratic accountability (Hudson 2005).

A final example of European influence over subnational governance is the strengthening of the directly-elected regional level of government in Sweden. Before Sweden became a member of the EU, the very existence of the intermediate level of government, i.e. the county councils, was questioned. Voices were heard from both left and right claiming that Sweden was too small to have two levels of subnational government, and that the municipalities in cooperation could just as easily handle self-governing tasks that required larger territories, such as responsibility for health centres and hospitals. This was, by the way, already the model in use in Finland. Development policies were a central government responsibility, partly carried out through its long arm in the counties, i.e. the county administrative boards.

With membership of the European Union, the focus of the debate shifted completely and very quickly. The voices suggesting an abolition of the intermediate level quietened. As member of an EU that had adopted the Maastricht Treaty three years earlier and emphasized the notion of a Europe of the Regions, it was now claimed that Sweden needed a stronger level of regional government in order to fully be able to exploit the potential of membership. Although the Swedish division into counties had been debated since the 1960s and was seen as outdated (Krantz 2002), it was not until after Sweden had become a member of the EU that it was possible to mobilize sufficient support for at least a partial regional reform. However, although the membership of the EU had triggered a reform process, the strengthening the regional level of government was also supported by other societal changes such as a stronger specialization of health care and an emphasis of the role of the regions in economic development

The division into counties, originating from the seventeenth century, was particularly badly adapted to modern patterns of commuting and mobility in two areas – West Sweden and Skåne. The first initiatives came from affected county councils and municipalities, in a process that eventually led to amalgamations of county councils in western Sweden into the Region of Västra Götaland in 1998 and in southern Sweden into the Skåne region in 1999. Apart from the traditional county council functions of healthcare, these were also given the responsibility for regional development from the county administrative boards (Lidström 2010b).

The establishment of the two regions encouraged further change. On two occasions, large-scale reform for the whole country, along the same lines as that implemented in West Sweden and Skåne was initiated. In 2007, the Parliamentary Committee of Responsibility (Ansvarskommittén 2007) and in 2016 two politically appointed analysts (Indelningskommittén 2006) proposed similar comprehensive regional reforms. The remaining county councils would be replaced by fewer but larger regions and regional development functions would be transferred to these from the county administrative boards. However, none of the proposals has been realized as they did not receive sufficient parliamentary support. The first was effectively blocked by the Moderate (Conservative) party, the leading party in the non-socialist coalition 2008–2014 which was strongly against comprehensive reform, and the second was abandoned when it was obvious that a parliamentary majority was lacking. One reason for this hesitant attitude from several parties was the public’s reluctance for regional reform (Nilsson 2016). Nevertheless, the final outcome of this gradual, piecemeal and fragmented process of regional institutionalization in Sweden, that lasted over twenty years (Lidström 2010b), seems to be a comprehensive but limited reform. By 2019, the responsibility for regional development will have been transferred from county administrative boards to all county councils, and these will be called regions instead of county councils, but without any mergers.

Through the emphasis on decentralization and self-government, the regionalization reforms are well in line with a key feature of the Swedish political-administrative model. However, the particular emphasis on the regional level, rather than the municipal, is a novelty in Swedish decentralized governance. In this respect, the EU contributed to changing the reform agenda, and paved the way for what is finally ending up as a comprehensive move of responsibility for regional development to elected regional governments.

Conclusions: Subnational authorities, mediating domestic factors and the state

Over the last twenty years, Sweden has become a well-integrated member of the European Union, on all levels. Popular support for membership is stronger than ever and all but two major political parties want the country to remain in the union. Swedish membership of the European Union in 1995 could be regarded as a potential critical juncture, opening up for changing the Swedish state in a way that would make it more similar to other member states. The analysis in this paper has focused on the role of subnational authorities vis-à-vis the EU, and it suggests that these have been able to mediate the European impact on the Swedish state, in several respects.

First and foremost, due to the traditionally strong position of subnational authorities in the Swedish polity, the impact of the EU on the Swedish state has been mediated in significant ways, underlining the role of path dependency as a mediating factor. Linked to its central role in the Swedish welfare state, local and regional governments are among the most powerful in the world, at least in financial terms. They have also considerable political and fiscal autonomy, such as an unlimited right to set local taxes without needing to consult with central government or other local authorities.

Local and regional governments have been able, independently and without restrictions from central government, to pursue their interests vis-à-vis the EU. As they are directly affected by EU regulation and are expected to implement EU decisions, they have a clear interest in influencing how these are shaped. Subnational authorities carry out their own activities and are represented by various institutional interests with their own channels to the EU institutions. This has largely taken place with the tacit consent of both Social Democratic and non-socialist governments. There have been no cases where central government has criticized or interfered in the local and regional activities that have been carried out in relation to the EU. In addition, extensive decentralization of functions and resources for regional development to subnational authorities has taken place. The regional level of government has been partly reorganized and empowered so that it can be responsible for these functions and be a stronger actor in an EU setting. The strong position of subnational authorities is also supported by the norms and values of the citizens and is reflected in the way intergovernmental relationships are shaped, not least through the strong position of the Swedish Association of Local Authorities and Regions. This suggests that path-dependency, rather than change, has been at work. The strong legitimacy of subnational authorities has been supported by the EU policies of regionalization but has also been in the interests of subnational governments themselves. Hence, it represents a combination of top-down and bottom-up mediation.

Subnational authorities have also been strengthened by the systematic capacity building that local and regional actors have been involved in, even from the years before Sweden joined the EU. Local and regional governments have been keen to learn how the EU system functions and how to gain access to EU decision-making. This bottom-up activity has further improved the position of subnational governments vis-à-vis the EU and contributed to mediate the impact of the EU on the Swedish state.

Apart from the mediating role of local and regional government, there are also features of the Swedish institutional and administrative culture that have turned out to be significant. For example, the emphasis on consensus, expressed in a tradition of neo-corporatism, made it easy to adopt the EU way of decision-making through partnerships, where various local and regional interests are involved in joint decision-making. The several hundred year history of transparency and open government, which some feared would be under threat with EU membership, has instead contributed to introducing such policies in other EU member countries. Again, these reflect a long tradition, which has remained largely unaltered in the meeting with the EU administrative culture and is in this sense a mediating factor from the bottom and up.

A final mediating factor of relevance in the Swedish case is the economic situation in Sweden after joining the EU. The decision to apply for membership in the European Union was triggered by a severe banking and financial crisis in the early 1990s. Since then, GDP has grown steadily, with the exception of a drop in 2009. Nevertheless, compared to most other EU countries, Sweden went through the 2008–09 financial crisis very smoothly, and definitely without any drastic reductions in subnational spending. Although the public sector’s share of GDP has decreased from 62% in 1995 to around 50% today, the local and regional share of the public sector has increased (Statskontoret 2016b). Therefore, the economic situation in the country has not been a reason for the government to reduce funding to local and regional authorities or to interfere in its self-government, which has been the case in many other EU member states, as shown by a recent comparative research project on the impact of the crisis on local affairs (Kuhlmann and Bouckaert 2016). Such pressure on the government has not occurred in Sweden The final question to be addressed is to summarize whether the mediating position of subnational authorities has weakened or strengthened the state in relation to the EU. We have identified several cases where the EU has contributed to decentralized decision-making in Sweden. Main examples are regional institution building, more decentralized regional development policies and the possibilities for subnational authorities to bypass the state when dealing with EU matters. However, EU influence has been significant, but has been complemented by other external and internal drivers. For example, regionalization corresponded with dominating ideas of neo-liberalism and new regionalism that also developed over the same period. There have also been domestic reasons for establishing stronger regional institutions, such as a need to restructure the healthcare system. However, the pressure from the EU has not undermined the position of the state. On the contrary, local and regional self-government seems to have been an asset, and not a problem, for the state in relation to the EU.

Although decentralization is the main tendency, there are also some noteworthy instances of recentralization. During the first decade of Swedish membership of the EU, the county council/regional level of government was involved in making decisions on the allocation of resources from the EU structural funds. These functions were later transferred to a central government agency at national level. The main reason was to obtain better policy coordination. But on the whole, path-dependency has been strong, and Sweden has remained a decentralized unitary state.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1 The article builds on and further develops the analysis carried out in Lidström (2010a) which summarized EU impact on Swedish regions after the first ten years of membership. Notably, research on these matters was more extensive during the first ten-year period than in more recent years.

2 Regions are county councils that have been given additional responsibility for regional development. There has been a gradual transfer of such functions from central government to county councils during the last 20 years. By 2019, all county councils will have been transformed into regions.

3 The Local Government Act 2017:725.

4 During the period 1995–2017, Social Democratic governments ruled Sweden for 14 years (three of which in coalition with the Environmental Party), and non-Socialist governments ruled for 8 years.

5 According to Statistics Sweden that carries out regular surveys of the Swedes’ views on EU membership, 55% were for and 19% were against in May 2017. In the four northernmost countries, the corresponding figures were 48% for and 26% against.