Democratic Institutions Without Democratic Content?-New Regionalism and Democratic Backsliding in Regional Reforms in Sweden

The aim of this article is to examine trends of democratic backsliding associated with the long standing reform work on regional institutions and policies in Sweden. To this end, democratic backsliding is conceptualized in a different manner compared to conventional understandings. By doing so, the article highlights a missing aspect in the research on democratic backsliding that concerns how well-intended reforms designed to strengthen democratic institutions can also harbor non-democratic consequences. In Sweden, a new political arena was created when the former county councils were transformed into so-called called regions in 2019. As part of this, the regions have been assigned responsibility for both health care and regional development planning. The overall research problem to be analyzed in this article focuses on the relations between the policy objectives for democracy and regionalist ideas of economic growth that both were central concerns in the reform processes. The results highlight how the governing rationalities in the regional reform processes have changed during the period between 1990 and 2020. The original conception of creating a mini-version of a liberal and representative democracy have turned into a form of democratic backsliding privileging economic goals. The economic rationalities that permeate the political sphere today close the space for articulated different interests and opinions-a dimension that we argue is crucial for any democratic society. We draw two main conclusions: First that the neoliberal aspect of governing is missing in the analysis of democracy at the regional level, resulting in a descriptive discussion of democracy that tend to ignore the effects of the particularly strong emphasis on economic growth. Secondly, that there is a lack of a discussion on democracy that takes the regional level into account, i.e., that the sub-national level should be regarded and thus discussed as a distinctive level of democracy.


INTRODUCTION
The aim of this article is to examine and discuss the puzzling trends of democratic backsliding that is associated with the reform work of the directly elected Regional Councils in Sweden. Despite the articulated ambition of well-thought-out reform policies to strengthen democratic values at the regional level, the trend seems to have gone in the opposite direction. Thus, we are interested in finding out how firm democratic ambitions could be replaced in the Swedish regional reform process more or less unnoticed by both political stakeholders and researchers.
The reform work was completed in 2019 with the creation of a new political level of decision-making, where the former County Councils are now called Regions with a widened responsibility that include both health care and regional development. The process started already in the 1990s and was sparked by the preparation of the Swedish membership of the European Union. At that time, in the wake of the Maastricht Treaty, the reform agenda was partly focused on a renewal of the political institutions at the regional level in the member states. The reforms were launched under the visionary slogan of a new Europe as the Europe of the Regions. The most significant ideas in this vision were to stimulate economic growth and contribute to address what has long been debated as a democratic deficit in the institutional structure of the European Union (see Sharpe, 1993;Bullmann and Vergleichende, 1994;Keating, 1998;Larsson et al., 1999;Keating, 2008).
This visionary orientation was ultimately based on theories of economic development associated with regionalism. A "new" regionalism was proclaimed and contrasted with an "old" (and obsolete) version. While the old regionalism denoted regional development as activities in a fixed territory and within a subnational administration, the new regionalism emphasized vertical networks and open territorial borders within a globalized economy. The metaphor of an "old container" was replaced by a dynamic and boundless innovation system. New concepts were added to the regional development vocabulary such as regional entrepreneurship, trans-regional investments, cross-border networks, subnational lobbying, multi-level governance (Gren, 1999;Hettne and Söderbaum, 2000;Amin, 2004).
The reform process in Sweden restructuring the regional system came to be lengthy and included a large number of investigations and political controversies. In parallel to the reform agenda in the European Union, the Swedish reform processes have had two key policy goals (Johansson, 2000): First, to contribute to a need for increasing economic growth in the Swedish economy. Secondly, to strengthen democratic values by transferring political responsibility for regional development from state authorities to directly elected regions. The main idea has been that development issues, such as labor market, business development, infrastructure, public transports, cultural policies, should be anchored in a regional context with decisive influence for citizens and political parties in a representative or liberal model of democracy. It is worth noting that even the Swedish reform agenda were inspired by the new regionalist paradigm (Gren, 2002).
The regional issue in Sweden should also be seen as part of development processes that spread globally from the 1970s onwards through what came to be called "devolution" and which in research was analyzed as and "fiscal decentralization" (Bahl, 1999;Oates, 1999;Kee, 2003). The main idea is that the responsibility for policies in the public sector should be organized to optimize various functions in the political system. The purpose is generally stated to be to make the public sector more efficient. For example, the Treaty on European Union, Maastricht Treaty of 1992, proclaimed the principle of subsidiarity, i.e., public policies and their implementation should be assigned to the lowest, most effective level of government.
We believe that arguments about economic efficiency have dominated research on devolution and fiscal decentralization. However, we claim that the field of research has never problematized how policy tasks should be organized within the framework of democratic institutions. It has often been concluded that decentralized tasks, for example at the regional level, should be handled within the framework of second-order versions of representative democracy at the national level. In the Swedish regional reforms, the ambition has been to strengthen both efficiency and democratic values simultaneously. A critical analytical task in this article is to examine the realism of that assumption.
The regional reforms have been implemented incrementally since the end of the 1990s and the overall conclusion in regional research in Sweden is that the policy goal concerning democracy has not been accomplished (Johansson, 2005;Hudson and Rönnblom, 2007;Johansson and Rydstedt, 2010;Johansson et al., 2015). On the contrary, the well-founded intentions to strengthened liberal democratic values have so far resulted in tendencies that could be described as a version of democratic backsliding.
Although being a broad concept, democratic backsliding is mainly used when analyzing threats to the institutions of liberal democracy from right-wing populism or other autocratic movements, or with Nancy Bermeo: "At its most basic, it denotes the state-led debilitation or elimination of any of the political institutions that sustain an existing democracy" (Bermeo 2016:5).
But although democracy still fails in more "classical manners" though open coup d'états or straight forward dismantling of democratic institutions, contemporary political developments include more subtle versions of democratic threats, or with David Runciman, "Democracy could fail while remaining intact" (Runiciman 2018:4). In his book How Democracy Ends (2018) Runciman points to more silent challenges to democracy through undermining democracy from within. Here, conspiracy theories and filter bubbles accompanied by increasing digitalization orchestrated by global tech companies play central roles in dismantling democratic institutions. Following Runciman, we argue that there are more silent versions of severe challenges to democracy, where democracy is undermined invisible from within, and that the concept of democratic backsliding also could encompass these trends.
While we do not see democracy in Sweden being straight forward challenged by a coup d'état, voting fraud or harassment of the opposition, we thus believe that there are more silent aspects in need of recognizing when discussing democratic backsliding. Here, we are especially concerned with the contemporary rationalities of governing, where the rationalities of the market have replaced the rationalities of the political sphere, a development that could be conceptualized in terms of a neoliberal hegemony. Through conceptualizing neoliberalism as a form of governing, not as ideology or policy (Larner, 2000), we argue that the processes of de-politization that this shift enhance risk transforming democratic ambitions into administrative practices (Mouffe, 2013).
Hence, in this paper, we use the concept of democratic backsliding in a slightly different way, focusing on how ambitions of strengthening regional democratic institutions have failed and relating this failure to the implicit but permeating rationality of economism in the governing of politics (Brown, 2015). The reforms have not led to any renewal of the role of political parties or of increased civic interest or participation. Instead, the policy is characterized by being consensual, by decision-making processes that have moved into non-transparent or elitist negotiation networks, by stronger influence for civil servants, by governing as projectification, and an increasing use of procured consulting services (Öjehag-Pettersson, 2015;Olivius and Rönnblom, 2019;Scott, 2021).
We argue that the regional reforms in Sweden contains a contradiction between the ideals of democracy and the ideals of economic growth, a contradiction that not have been acknowledged in earlier research on regionalism, and a contradiction that could be conceptualized through discussing democratic backsliding in relation to the regional reforms with a focus on the more silent changes of governing rationalities.
The overall research problem to be analyzed in this article focuses on the relationships between the policy objectives for democracy and regionalist ideas for economic growth. We will problematize the liberal democratic values in relation to the ideas of the new regionalism, asking if the development of democratic values associated with a liberal, representative democracy are inherently incompatible with the basic ideas of the new regionalist paradigm.
I the next section, Conceptualizations, we will give a deepen description of the conceptual framework of the Swedish regional reforms. We will then give a brief overview of the historical development of regionalism and with an in-depth orientation of the development and content of the "new regionalism." We assert that the Swedish regional reforms after 1990 are inspired by policy ideas of the new regionalist paradigm. In Regional democracy and methodological nationalism section we problematize what is meant by democracy at the regional level and we highlight some possible explanations behind that democracy at the regional level in Sweden indicate democratic backsliding. In The reform process in Sweden 1990-2020 section, we give a description of the reform work on the regional issue in Sweden during the period 1990-2020. In this section we will bring together problems associated with the new regionalism in relation to a liberal and representative model of democracy. In the following New regionalism and the rationalities of rule section, we will delve deeper in the problem on the tendencies of democratic backsliding through the introduction of a, in this context, new analytical framework, governmentality. In the concluding Discussion: New regionalism and democratic backsliding section, we discuss how a governmentality framework could enhance the discussion of democratic backsliding in the analysis and point out some significant issues that need more attention in future research on politics at the sub-national level.

CONCEPTUALIZATIONS-REGIONALISM AND REGIONAL DEMOCRACY
In this section, the theoretical perspectives of regionalism and the conditions for democracy at the regional or subnational level are presented and discussed. The idea is to examine the relationships between regionalism as planning theory and the prerequisites for building democratic institutions at the regional level as it has been discussed in earlier research.

Regionalism
Regionalism has a multifaceted history with roots in American urban planning with visionary planners as Clarence Stein, Patrick Geddes, and Lewis Mumford. The early regionalists emphasized the city's role as a driving force for economic development. Still, it was made clear that it depended on incorporating the resources and natural values of the surrounding ruralities into the planning processes. Central values were social equality and the region as an identity-based community (Wheeler, 2002;Talen, 2006).
As early as the 1920s, Lewis Mumford developed ideas about the content and significance of regional development policies. Mumford criticized how the central government managed and controlled the development of society at the regional and local levels. He argued that regional planning controlled from above could not take sufficient account of how people live their daily lives. The type of regionalism that Mumford developed focused on a bottom-up perspective with human needs, environmental factors, and aesthetic values in focus. The mainstay of the new regionalism at this early age was to balance the relations between urban environments and the surrounding ruralities. (Mumford, 1938, ch VI). Ideas on regional development planning gradually diffused and were institutionalized in the political systems for community planning worldwide. After the Second World War, the regions became part of policy development at the national level. What in many countries, not least in Scandinavia, came to be named regional or localization policies. In social science research, a new discipline -regional sciencewas created, which analyzed economic aspects of regional development with mainly quantitative methods. Regional policy planning was driven from above based on a growing need to plan the distribution of economic values from a geographical and national perspective (Wheeler, 2002). In the Scandinavian countries, societal development was characterized by urbanization and correspondingly rapid depopulation of rural areas. Regional policy in Sweden was introduced during the decades after 1945 in a conflictual relation between stimulating business development in urban areas versus the desire to slow down the ongoing depopulation of rural regions (Elander, 1978;Johansson, 1991;Svensson, 2000).
With the changes in regional policy initiated by the EU with the Maastricht Treaty in 1992, the basic policy ideas for regional development also change. With its central governing and territorially designed policy, the former regional policy now came to change towards ideas on network relations, decentralized governing, or devolution, and a growing interest in regional identities. In research on fiscal decentralization, these development features are analyzed to express both functional and political needs for renewing and streamlining public policy from a socioeconomic perspective (Goodwin et al., 2005). Economic growth and political decentralization are emphasized more strongly in policy development than before (Sharpe, 1993;Keating, 1998;Wheeler, 2002). In social science research, this development is seen as the new regionalism compared to the old regionalism. The old regionalism was based on a centralized and territorially fixed policy. In addition, the democratic legitimacy was based on political decisions at the central governmental level, albeit with a limited influence for political parties and social movements in each region (Gren, 2002;Hidle and Leknes, 2014).
The ideas of the new regionalism paradigm do have some similarities with early regionalism. Mumford, for example, was critical of the controlling power of nation-states concerning regional development planning. There are also some other common denominators. The new regionalism contains a more normative approach compared to the ideals that characterized the regional science approach shortly after 1945. New regionalism mainly pleaded for a market-fixed economic policy and to some extent, for the importance of cultural identities and aspects of sustainability in urban areas. However, the differences in developmental thinking between Mumford's regionalism and the new regionalism in the 1980s are to be described as fundamental. The new regionalism focuses on a changed economic policy and a need for increased European integration (Keating, 1998;Wheeler 2002). The dominant policy orientation was to stimulate economic growth in strengthening the competitiveness, fiscal efficiency, and entrepreneurial thinking in Europe. The sick men of Europe (mostly referred as a metaphor for the economic situation in the United Kingdom, Germany, and Italy in the 1980s) needs to be cured by an entirely new economic agenda promoting the free movement of goods, capital, services, and people. In addition, rural areas are essentially missing as a part of regional development efforts in the new regionalist thinking (Ortiz-Guerrero, 2013).
The basic idea of the new regionalism is a notion of transforming the economy from an industrial society to an information and service society. The societal analysis is based on the assumption that the world has undergone an accentuated economic, social, and political globalization. The dominant conclusion is that the transformation of society has moved in the direction of increased importance for both global and regional processes, while the nation-states are continually losing ground. The consequences are: • that industrial activities take place in regionally based but territorially unbounded network relations • that the room for information and service-based production is increasing • that there are increasingly rapid business transfers both within and between regions. • policy ambitions to highlight cultural identities in the region In sum, the planning ambitions mainly concern five policy areas: • To invest in infrastructure systems to stimulate competitiveness in business operations • To highlight the importance of larger cities as a driving force for economic growth • To invest in policies to improve the "innovation climate" in each region • To promote cultural and creative planning in efforts to make regions economically flexible and creatively vibrant • To let development processes be controlled from below by local and regional actors However, the ideas of the new regionalism never came to formulate any distinct and innovative ways of approaching issues on democracy. The ability to create new forms of political representation, legitimacy, and accountability can be described as weakly developed. The liberal, representative model of democracy and a nationally defined citizenship came to live on as a framework for development processes in regional networks. According to several researchers, this marks a tension between the territorially determined political order and the various forms of organizational entities in the new regionalism paradigm (see, e.g., Amin, 2004;Painter, 2008;Syssner, 2011), or as one of the researchers states: many of the proposals for reform-possibly in the hope of being taken seriously-appear as mini versions of representative democracy, in the form of proposals for elected regional assemblies, accountable regional elites, and incorporation of the interests of different groups within the assemblies. Again, there is nothing wrong with this, but what is on offer is an imitative model of democracy, rather than an opportunity for a different and more expanded politics of place (Amin, 2004, p. 37).
We argue that this is a fundamental weakness in the ideas that gained ground in the EU system from the 1990s on developing a Europe of the Regions. The new regionalism has never presented a coherent idea for pursuing reform policies to ensure fundamental democratic values in regional development planning (cf. Painter, 2008 values. The new regionalist paradigm is mainly silent on issues how citizenship is to be defined, how values such as participation, transparency and, accountability can be secured. In the next section, we will delve deeper into ideas associated with democracy in regions.

Regional Democracy and Methodological Nationalism
Problems associated with democratic issues at the regional level in unitary state systems have had a minor place in political science research (Hendriks et al., 2012). However, it should be noted that research on regional democracy in federal states can be described as extensive . As Amin states in the quote above, the basic observation is that democracy at the regional level is considered a mini-version of traditional representative democracy. The challenges are about how democratic values can be linked, not only to the formal institutions at the regional level but also to institutions in cross-border cooperation and, for example, in informal arrangements as network and partnerships in regional policies (cf. Loughlin, 2001;Syssner, 2011). Increasing interaction and intertwined interdependencies between levels in the political systems received increasing attention in social research during the 1990s. Developments within the EU, both its enlargement and changes in the political institutions, were linked to the concept of multi-level governance (Bache and Flinders, 2004). Part of this research has touched on issues of democratic values in multi-level systems. Some of these studies have, in turn, been devoted to the role of the regions as a democratic intermediate level in the political system (see a compilation of articles in Däubler et al., 2018). The overall conclusion of this research is that issues of democracy at the regional level have only received scant attention in both public policy development as well as in social science research: Yet while we have gained better knowledge of the causes behind regionalisation, our systematic understanding of how this development affects the quality of democratic representation lags behind (Däubler et al., 2018, p. 542).
The hitherto weakly developed research on democracy at the regional level has been regarded as methodological nationalism in a dominant part of political science research: the tendency within political science to focus on the nation-state as the main unit of analysis in studying social and political life, and, in consequence, to neglect the region as a unit for political analysis (Jeffery and Schakel, 2013, p. 299, se även;Tatham and Mbaye, 2018).
This kind of critical remarks means that traditional theories and methods for studies of the institutionalization of democracy in a nation-state context have also become a guiding model for studies of democracy at the regional level. The democratic values of the regions have been seen as a mini-version of the national and representative model of democracy. The typical example is that general elections to regional assemblies have been reduced to what has been called "second order" in relation to the national parliaments as "first order elections" . The methodological perspectives have to a large extent, been locked into territorially defined regional boundaries. Political Science studies of democracy at the regional level have to a large extent, regarded the regions as down-scaled nation-states and have so far not regarded the region as a distinctive institution in society. Against this background, some political scientists have pleaded to create a "regional political science" .
We have to look to other disciplines, mainly Sociology and Human Geography, to find alternative approaches. This kind of approach is often based on the increasing degree of globalization and that the political analysis should be extended beyond a nation-state context. The analyzes of the development of regions are seen as an expression of a societal development characterized by complex structures which, among others, Jessop (2002) have described as multi-centric (several centers of power), multi-scalar (takes place in interaction networks at different levels) and multi-temporal (different time horizons). In this type of complex social structures, the regions do not constitute a mini-version of the national political systems but a distinctive political institution.
Not only this. The development of the regions and the regional reforms implemented in many countries have been anchored among regionally-based actors. In many cases, regionalization has been negotiated as a conflictual issue versus the actors representing the central government (Johansson et al., 2015;Niklasson, 2016). Another issue has to do with what should be included in a regional citizenship (Syssner, 2011;Mitander, 2015;Piccoli, 2018). How should the demos of regional democracy be delimited? Who is affected by policies carried out in cross-border regions or other kinds of regionally organized networks? And who is left out?
In sum, we argue that the ambitions to expand democracy to the regional level should be discussed beyond being something more than a mini-version of democracy at the national level. We agree with other regional researchers (e.g., Jeffery and Schakel, 2013) that a developed regional Political Science is needed to analyze the conditions for democratic legitimacy and accountability at the regional level, and that this is relevant for the analysis of the Swedish regional reforms from 1990 onwards, but we also argue for a more elaborated analysis of the prerequisite for democracy bringing the economization of governing into the analysis.
The reforms in Sweden received their most essential impulses from the discussion within the EU about a Europe of the Regions (Johansson, 2000). Even in Sweden, the regions were seen as a multi-level system and a part of the new regionalist movement (Gren, 2002). The ambitions of the reform work to strengthen democracy in the new regions in Sweden were formulated within a representative model of democracy, at the same time as a marketized rationality of governing politics also was introduced. Although there were some alternative models of democracy, mainly ideas about municipal federal systems, the main track has been to start from the already existing institutions at the regional level. Thus, the central concept of democracy in the reform efforts has been to let existing county councils be the core of democracy in the new regions, without discussing how the on-going transformations of how these institutions are governed are addressed, leaving the democratic ambitions in the reform fairly under-analyzed.
In the next section, we will present an overview of' the Swedish regional reforms and how they are related to the articulations of regional democracy. The analytical interest is directed towards deepening the understanding of the problems associated with the ideas of the new regionalism in relation to a liberal and representative model of democracy.

THE REFORM PROCESS IN SWEDEN 1990-2020
The Swedish system of local government comprises traditionally of two tiers. The first consists of 290 municipalities (kommuner) and the second of 21 regions (previously named county councils-landsting). The county councils were established in 1862 to provide various services, and later on, the most important task became to take responsibility for the health care system in Sweden. There is no hierarchical relationship between the regions and the municipalities. There is also a county administrative board (länsstyrelsen) led by the county governor (landshövdingen). The county administrative board is the central government's unit in each region with tasks to control and support the implementation of national legislation. From 2019, when the county councils came to be called regions, they were given responsibility for both healthcare and regional development (Regeringen, 2021).
The European debate on regions was introduced in Sweden at the beginning of the 1990s, and the first Government Commission was set up in 1991. The Commission conducted a forward-looking study, and in retrospect, we can conclude that the proposals made by the Commission became formative for the continued reform process. To a large extent, the Commission had been initiated as part of the preparations for Swedish membership of the European Union. The inquiry is based on the debate on a Europe of the Regions, and several development features are associated with the ideas of the new regionalism. The formative significance of the Commission consisted of considerations in two parts (Johansson, 2000): • the future division of county borders • the responsibility for the policies on regional development The Commission proposed reducing the number of counties in Sweden. The then 24 counties should be reduced to at least half as many. The Commission focused on development trends associated with growing labor market regions and the capacity to compete with larger regions in the rest of Europe. The issue of a new county division came to be intensely debated for many years but has only resulted in two new regions being formed through county mergers; Skåne and Västra Götaland. There has been a general lack of political agreement between the political parties in Sweden about the county division (Lidström, 2010). Many regional representatives of parties and interest groups have objected to having to merge with other counties. The Swedish regional map has thus, with two exceptions, been unchanged since 1810 (Lidström, 2011). The Commission's analysis of the responsibility for the regional development work also became controversial. Three different types of models were presented for further discussion (Johansson, 2000): I. Maintenance of the already existing system of a regional-state responsibility for regional development policies II. A changed model based on ideas of regional self-government and letting the directly elected regional councils be responsible III. An innovative model called inter-municipal co-operation with a responsibility of the regional development placed on federations consisting of municipalities and county councils in each region The issue of the responsibility came to be a recurrent and conflictual part of the reform work throughout the period up to 2019. Doubtless, the Commission's three models resemble the contradiction analyzed in regional research and discussed among practitioners/politicians between the old and new regionalism. The whole reform agenda came to deal with the institutional alternatives to the old regionalism represented by Model I. Model II, and Model III represents a new kind of regional development policy based on a bottom-up strategy within horizontally organized networks. Either the reform intentions were directed towards a directly elected assembly or inter-municipal cooperation, the proponents of the reform work are used visionary rhetoric concerning a new "political concert in Europe" versus the old nation-state container of territorially fixed regions (Gidlund, 1993;Jönsson et al., 2000;Gidlund and Jerneck, 2000).
The reform development was based on a reform strategy that permitted try-outs of different models in different counties. In some counties, the state was still responsible for regional development policy. In other counties, the responsibility was placed on the county councils, and still, in other counties on inter-municipal cooperations were responsible. Thus, the regional situation was asymmetrical and "messy" (Johansson, 2000;Mccallion, 2008;Niklasson, 2016). There were mainly two issues that were debated.
First, in a lengthy debate on one of the most central ideas of the new regionalism, namely to shift the responsibility for regional development policy from the central state to a bottom-up anchorage in regionally-based networks. Representatives of the state, mainly among the governors (landshövdingarna), opposed these ideas. In Sweden, the governors are the central government's extended arm in the counties, were critical of transferring the regional development work to bodies anchored in local and regional contexts. The governors believed that this would lead to growing inequalities between the regions and pleaded for a continued presence of the central government in the regions (Niklasson, 2016). Briefly put, the governors represented a traditional perspective that we early have described as the old regionalism. Secondly, the Swedish regional debate initially (a few years towards the end of the 1990s), revolved around a critique articulated by representatives of the municipalities. Municipal politicians around Sweden wanted to avoid being subordinated to a strengthened role for the county councils. The constitutional construction in Sweden statutes that the county councils, have not an overriding position in relation to the municipalities (Lidström, 2011, p. 5). This construction means that county councils and municipalities have different tasks and are not in a relationship of superiority and subordination. The municipal politicians saw a risk that the regional reforms would lead to a shift in power in favor of the county councils and with a weakened role for the municipalities in the regional development work (Johansson, 2004;Johansson et al., 2015).
In addition to the above-mentioned aspects of power relations, the political arguments regarding this issue also address another significant aspect. In line with ideas on the new regionalism, representatives of the municipalities argued that regional development work should not be organized in a traditional and territorially based model. Alternatively, the regional development work should be coordinated in interorganizational and cross-border negotiation networks where municipalities and the region could interact with other actors within the framework of what came to be called an intermunicipal co-operation model (see above). Inspiration was taken from a try-out project conducted in one so-called pilot county-the Regional Federation in Kalmar County. During a period between 2003 and 2014, the Kalmar model also spread to other counties and was during a short while applied in more than half of the regions in Sweden. The inter-municipal co-operation model was organized as a kind of a local federation where the municipalities in a region, in association with the county council, formulated the regional development policy based on network arrangements, partnerships, and negotiated agreements between actors in the private and non-profit sector (Mörck, 2008;Niklasson, 2016).
The debate on the inter-municipal cooperation model was intense, and representatives of most political parties saw benefits for the efficiency but at the same time experienced worries from a democratic perspective. Due to the institutional structure of the inter-municipal model as an indirectly elected assembly, there was a critique on shortcomings concerning transparency, participation, and accountability. This development also illustrates how ideas in the new regionalist paradigm of flexible territorial organizations and networks are difficult to reconcile with fundamental democratic values (Johansson et al., 2015;Torfing et al., 2015). As a result, the reform work came to change direction in that the regions should be organized as directly elected regional councils. Thus, after 20 years of debate, the regions in Sweden came to be established as miniversions of a representative democracy (Larsson, 2016).
Reforms in Swedish politics on the regions came from the early 1990s onwards to be inspired by the ideas of the new regionalism. It is a strengthened responsibility for the regions that will increase the competitiveness of the Swedish business and industry and contribute to economic growth. Moreover, developments in Sweden show a shift in responsibility for regional development policy from the central-state to the regions. In sum, the reform debate was, from the beginning, directed towards economic growth and that this, in turn, should be founded on two types of political values; 1) that regional policy should have a democratic character and to be based on the specific societal context that exists in each region and 2) that policies should be designed with a high degree of regional autonomy. The objectives of regional development policy were thus given-what was debated were the procedures; what would the county borders look like and who would be responsible for the policy-making and so on. There was almost no debate about the content of creating a new innovation policy or an entrepreneurial spirit in the new regions.
Hence, the political content was taken for granted and not included in the discussion on democracy. In other to perform a more in-depth analysis of what we regard as tendencies of democratic backsliding in the development of Swedish regionalization reforms, focusing on the more silent forms of governing that permeate these processes, we have chosen to turn to the Foucauldian framework on governmentality.

NEW REGIONALISM AND THE RATIONALITIES OF RULE
So far, we have discussed the restructuring of regions by focusing on specific reforms and the historical development. To make sense of this analytically, we will introduce the governmentality framework, originated in the work of Michel Foucault and his lectures at College de France during 1978 and 1979 (Foucault 2007;Foucault 2008), and show how this form of analysis can help us interpret tendencies of democratic backsliding in the politics of regional development in Sweden.
Central for the governmentality framework is the conception of neoliberalism as a way of governing. First, Foucault (2007; argues that "to govern" is an activity that encompasses a wide range of circumstances well beyond those traditionally associated with the state, or other forms of formal politics. Indeed, in its broadest sense, governing can be identified as "the conduct of conduct." In other words, it captures the multifaceted array of practices where the goal is to rule, control, monitor, guide, steer, and direct or otherwise decide the fate of both objects and subjects in the world. In fact, Foucault (2007; argues, governing does not have to be associated with conscious decision making. Rather, quite often, subjects tend to govern themselves which means that rule can operate through other means than sovereignty and punishment. In addition to being an activity that can be found across societies, from the smallest of details to the grandest of plans, governing is also nested in mental and cognitive registers. Thus, this governmentality rests upon what Foucault labels political rationalities. Such rationalities can be understood as the underlaying, internal logic that must be in place in order to rule a given domain in a particular way and not another. Another way of expressing the function of political rationalities can be to think of them as the underlaying logic for a particular way of governing. In other words, to make them visible scholars may ask "what forms of thought, knowledge, expertise, strategies (and) means of calculation" (Dean, 2010:42) are nested in the ways that rule is articulated? Thus, according to Bacchi and Goodwin (2016: 42) political rationalities are "the rationales produced to justify particular modes of rule" and they function as ways that make any form of activity thinkable to both rulers and the ruled. These diagrams of power "draw upon the theories, ideas, philosophies, and forms of knowledge that characterize our intellectual heritage" (Bacchi and Goodwin, 2016:43) which means that most contemporary policy making in liberal states involve the social and human sciences quite extensively. It is against this notion of political rationalities that neoliberalism should be understood according to Foucauldian scholars (Larner, 2000). This stands in some contrast to more common interpretations of neoliberalism as an ideology or as an economic doctrine. While not disregarding the importance of such conceptualizations, to identify neoliberalism as a political rationality highlights other aspects. Thus, it can be understood "as a normative order of reason" (Brown, 2015:10) that gradually started to materialize three or 4 decades ago, then continuing to expand before now dominating the foundation for how we think about governing, thereby dictating its (im) possibilities. As such, its expressions are varied across the world, and Foucauldian scholars are often wary of making the mistake to assume that neoliberalism is a matter of uniformly rolling out or rolling back state policy in similar ways in all countries or governing sectors (Peck, 2013). Rather, it's concrete and tangible manifestations are likely to be different in South Korea compared to Sweden, Argentina, or Ghana (Ong, 2007;Brown, 2015). That being said, understood as a rationality, neoliberalism shares a number of traits that are remarkably consistent for most contexts, including the governing of regions in Sweden.
Here, we will not discuss the actual processes of how the rationalities of governing have changed nor will we detail the complex interplay between them and prominent actors and events. Instead, we sketch the contours of three interrelated themes in the neoliberal rationality that now dominate liberal democratic governance (Rönnblom et al., 2021). As such, they illustrate a comprehensive reprogramming of how to govern and why. Thus, these themes also highlight deviations away from the liberal rationality that traditionally underpin democratic states.
Importantly, the following presentation is not to be understood as empirical generalizations of how neoliberalism manifests everywhere and every time. Rather, we present the three themes and their relations as an exemplar, a figuration that is neither general nor singular. Following (Massumi, 2002:17) and others we argue that writing through examples or illustrations provides us with a way discussing abstract developments and relate them to more tangible events. Therefore, for each of the themes we exemplify a number of changes in the governing of Swedish regions that we argue can be understood in a new light by positioning them against this backdrop of neoliberal rationality.

The State and the Economy
We begin with the first theme that we have called the state and the economy, a theme that also functions as a starting premise for the other themes as it has to do with the general role and function of economy in relation to governing. With the rise of neoliberalism during the mid-20th century, the distinction between a political or a governing sphere of society and an economic one starts to break down. In terms of political rationality this is very important since governing now, in essence, becomes modeled on the same principles as those that have been in place for a long time with respect to private business. Along with (Foucault et al., 2007), we argue that this is not best understood as a simple extension of liberal principles, but rather as a remodeling and indeed an important break away from them. New Regionalism then, is an expression of this neoliberal political rationality that thoroughly economizes the state. Indeed, as Brown (2015:62) claims, "the political rationality of the state becomes economic in a triple sense: the economy is at once model, object and project." This means that economic principles first and foremost become the ideal model for how the state operates, internally as well as externally. At the same time, the economy, and particularly growth, is turned into the primary object of the state's concern. Therefore, the state begins to facilitate a project of unlimited marketization of virtually all human domains and conduct (Brown, 2015). Thus, the neoliberal state is an active one. However, unlike the social democratic or classic liberal states, where the state could activate certain functions of the economy to realize particular goals, the neoliberal state is activated on behalf of the economy.
In particular, neoliberalism as an order of reason, puts the state hard at work to install competition and competitiveness as the primary internal logic for the economy. Importantly, this is necessary precisely because unlike in the context of private business, competition is not easily applied as a governing principle in all contexts. Thus, while neoliberalism strives to naturalize competition and competitiveness, it does so based on the realization that competition is not a naturally occurring feature of all human activity.
Competition's number one virtue in neoliberal rationality is its promise of growth. This expands the primary object of governing, the economy, and then enables new ways for governing both subjects and objects. Growth is in this way turned into the premise of any form of social policy. Indeed, to focus on growth and the economy, can be said to be the one primary social policy of the neoliberal order of reason. If managed properly the quest for growth will set free the entrepreneurial spirits of all humans and bestow them with a means to prosper by their own hand. Conversely, neoliberal subjects of governing need to be activated precisely by competition. States, regions, cities and humans alike need to compete for their place in this order. In other words, this focus on growth through competition is very important. As it is installed as a governing principle throughout society it replaces other economic logics such as for instance exchange and even consumption. Moreover, to wholeheartedly govern through competition also means an acceptance of inequality and ruin, not as an unfortunate side effect, but rather as a matter of principle. This goes for regions as well as humans.
In terms of New Regionalism and the reform process pertaining to regions and regional development in Sweden we find these aspects of growth, competition and competitiveness to be a salient feature. Indeed, it is striking how growth through competition is articulated as the by far most important value, even as the focus has been on democratizing the regional level of government (see Rönnblom, 2008;Johansson and Rydstedt, 2010;Säll, 2014). In fact, competition and competitiveness are often articulated in such ways that the present situation resembles war times. Therefore, those who read through Regional Development Documents issued by the various regions over the past decade will find a lot about "developing spearhead technology," "to gather the ranks," "mobilize forces," "to act strategically" and "to organize a united frontier." Similar formulations are also common in EUpolicy documents as well as national policy documents designed to govern regions in Sweden and other countries. Together they signal a militaristic urgency that in some ways only can be interpreted as "compete or die" (Öjehag-Pettersson and Mitander, 2020).
At the same time, other democratic values in regional development tend to be premised upon this one and only social policy to produce growth through competition. Gender equality, sustainable development as well as issues of urbanization and rural stagnation are often referred to, yet often in such ways that they need to be addressed though growth. Indeed, they are often viewed as potential means to become competitive if they can be handled correctly (Öjehag-Pettersson, 2015).
This also illustrates the general ambition to activate the regions on behalf of the economy. Regions should compete, survive and prosper for the main purpose of driving growth. Importantly, this is not a hands-off approach from the state in a simple sense. Rather, while it hands over certain freedoms and responsibilities to regions it still governs them and requires from them a focus to produce competitiveness.

The Truth of the Market
The second theme of neoliberal rationality that we find important may be labeled the truth of the market. While the market and the economy were already hailed among classic liberalism and, indeed also embraced by social democrats, its "truth" and legitimacy were founded in legal arrangements and restricted to certain domains of society. Neoliberal rationality transposes the foundation for the market as a site of truth with respect to human behavior, desire and capacity. Rather than being based on juridical claims and legal argumentation the market gains the status of natural law. As this happens, it becomes the model for good governance, for behaving correctly and indeed for "reality" as such. To deny markets and market functions is under neoliberal reason to position oneself beyond logic in the sense of not accepting reality. This is important because it is not just a matter of behaving irrational, it is to not accept the truth of the world. Therefore, as neoliberal political rationality prevails, "market principles frame every sphere and activity, from mothering to mating, from learning to criminality, from planning one's family to planning one's death (Brown, 2015:67)." Closely linked to the triumph of the marketization project is the notion of consensus. Given that the economy is the main focus, that growth is best achieved through competition and that competition is naturally expressed through markets there is very little room for political difference. Again, with the market as a form of reality principle in place, those who do not accept it disqualify themselves as non-realists. Thus, the prevailing market produces consensus in a double way. First in the sense just described, that we must collectively believe in the market, and secondly, also in the sense that political squabble and difference is best eradicated in the name of competitiveness. A divided region is a region that does not realize its potential. All interests should unite to prosper in the harsh global competition.
This topic of consensus is also very prominent in the ongoing reformations of the Swedish regional structure. While, as discussed above, there certainly have been controversies and political debates connected to issues of how to construct new regions in terms of geographical borders as well as in terms of "who does what" with respect to different governing bodies, the actual content of regional politics has been more or less without discussion. Regional politics was always about producing growth, and particularly during the past two decades about fostering competitiveness (Säll, 2014).
In fact, one of the most prominent aspects of how to govern regions and how they could best be guided through the fierce global competition that they inevitably are understood to be part of has been to form a regional leadership. This has often been identified as comprised by a range of stakeholders rather than politicians. Businesses of the regions, institutions of higher education and representatives for different special interest have continuously been understood as the most competent actors and they should therefore have a significant say in regional issues. Often driven and backed-up by different forms of knowledge and research conducted at university research centers and think tanks, such leadership have formed in networks, clusters, innovation systems or quadruple-helix formations. While leading politicians are often involved in such networks, they are certainly not a leadership informed by traditional democratic values such as accountability and legitimacy. Rather, it is a leadership of experts from and of the market that are understood to be well equipped to interpret and act upon the very same market in order to produce a competitive region (cf Rönnblom, 2008).

Economic Man and the Enterprise Society
The third and final theme that we wish to draw attention to may be called economic man and the enterprise society. While the notion of economic man has been around for a long time, we are again witnessing a replacement in terms of features and principles. Most prominently, and perhaps most strikingly, is that this figure who acts on self-interest, rational choice and economic incentive, in tandem with the market, becomes the model for all human activity. It effectively vanquishes the Aristotelean notion of humans as a fundamentally political animal. Thus, the room for any homo politicus, equipped with ends, means and virtues that are so important for liberal democracy becomes severely limited. Likewise, neoliberal rationality also reduce what (Foucault et al., 2007) calls the subject of rights, or homo juridicus, to a subordinated figure that is constantly overflowed by the traits of the relentless expansion of markets and homo oeconomicus.
Additionally, as homo economicus starts to exhaust the human experience, the governing subjects under neoliberalism become a form of capital, labeled simply human capital. As such homo oeconomicus primarily seeks profit and interest through his decisions and actively tries to calculate the life trajectories where capital appreciation is best realized. In this sense, human capital replaces labor as the image of how, and through what means, actors can realize themselves and be governed as a resource. At the end of this process emerges a figuration of human beings that sits uneasy in collective circumstances such as even among the most basic political demos that seek to assert their sovereignty (Brown, 2015).
Finally, the democratic subject as human capital is also associated with entrepreneurship. This endeavor is the most reasonable activity for economic man, and in the same way that human capital in a sense replaces labor, Brown (2015) argues that entrepreneurship replaces production. Indeed, Foucault (2007:147) calls this an "enterprise society" and points out how it represents a situation where even the consumption focus, so vital to classical versions of liberal rationalities, is rendered less important. Rather, under neoliberal rationality, workers are turned into entrepreneurs as the form of enterprise, along with the market, is multiplied throughout the social body. The guiding image here then, is not the classical market-place where actors barter and exchange commodities. Nor is it the super-market where desire is satisfied through consumption (Foucault et al., 2007). Rather, the enterprise society is one where humans everywhere conduct themselves in the form of a small business, always looking for ways to enhance their productivity and increase their (human) capital value.
In terms of regional development, we argue that over the past decades it has been filled with various initiatives, policies and rhetoric concerning the importance of the entrepreneur. These contain, not only a manifest celebration of entrepreneurship, but also a salient conviction that subjects both should and do already act as entrepreneurs that live to enhance their human capital.
We argue that this neoliberal theme is most prominently visible in the many variations of the so-called creative economy that has been an important policy narrative in urban and regional development for some time now. Inspired by the theories of Richard Florida (2002) regions have tried to foster environment and a "climate" that caters to the most creative, innovative and successful segments of the populations. This creative class is understood to be very important for fostering competitiveness and it has functioned as an ideal image for who to attract as well as how to form the present population through education and incentives. Moreover, the creative professionals are thought to be highly mobile. As a class of people, they set an example for the modern age as they constantly seek to relocate to regions or cities where their human capital can best be developed. In this process, they are not only looking for high salaries or wealth, but rather opportunities to realize themselves as innovators and entrepreneurs. In other words, they resemble the ideal citizen of Foucault's enterprise society as they look for ways to develop their own and their family in terms of human capital.
However, the creative class is just one iteration of this theme where the entrepreneurial subject stands at the center of regional development. Indeed, over the past decade or two, regions have followed various policy initiatives that seem somewhat interchangeable. Innovation systems, triple (or quadruple) helix, clusters, creative economy or smart regions are different and suggest various pathways towards competitiveness, yet they all revolve around innovations and entrepreneurs.
To sum up, we believe that discussing the Swedish regional reforms through the lens of political rationalities, have made an explanation of why the democratic ambitions articulated in these reforms have been left out of both political debate as well as in new regionalism research visible. The economic rationalities that permeate the political sphere today close the space for articulated different interests and opinions-a dimension that we argue is crucial for a democratic society. When the democratic institutions have been degraded into assisting the market in order to reach economic growth, and when democratic ideals like justice and solidarity are turned into effectivity and competition, democracy risks being watered down into pure administration of the goal of economic growth. We risk ending up with democratic institutions without a democratic content.

DISCUSSION: NEW REGIONALISM AND DEMOCRATIC BACKSLIDING
This article analyzes the relationships between policy objectives for democracy and regionalist ideas for economic growth. The case of the reform work launched at the regional level in Sweden between 1990 and 2019 illustrates that policies of neo-regionalism seem to be adversarial in relation to objectives to constitute a classic liberal and representative democracy. We have above discussed three interrelated themes in neoliberal rationality that have significantly affected how to reform and enhance democratic values at the regional level: 1. In the themethe state and the economypolitical rationality has become economic, with Wendy Brown, in a triple sense; as a model, object, and project. In terms of regional reforms, we find it obvious how growth through competition is essential in regional development policies. Not only this; competition and competitiveness are often articulated in such ways that the present situation resembles war times. 2. In the themethe truth of the marketthe rationality of governing illustrates the foundations for the market as a site of truth with respect to human behavior. The market has gained the status of natural laws. Concerning the development of regional leadership, it has become "self-evident" to involve business stakeholders rather than democratically elected politicians. The content of regional leadership is a onesided promotion of neo-regional business models such as innovation systems, an entrepreneurial spirit, quadruplehelix formations, etc. 3. In the themethe Economic man and the enterprise societythe main character acting in the regional development arena is a figure who acts on self -interest and as a rational actor and has replaced other figures such as homo politicus or homo juridicus. In regional politics, the importance of the entrepreneur has been at the center of the debate, not the least in policies on creative industries and the fostering of a climate that gathers the most creative and innovative actors in the region.
We draw two main conclusions from these rationalities of rule: 1) That the neoliberal aspect of governing is missing in the analysis of democracy at the regional level, resulting in a fairly descriptive discussion of democracy that ignores the democratic effects of the solid emphasis on economic growth. 2) That there is a lack of a discussion on democracy that takes the regional level into account, i.e., that the sub-national level should be regarded and thus discussed as a distinctive level of democracy.
An overall reflection is thus that an analysis of democracy at the regional or sub-national level needs to recognize rationalities of governing as well as be regarded as a distinctive level of democracy. Here, we believe that the more silent conceptualization of democratic backsliding is useful in making subtle challenges to democracy visible in the analysis, especially concerning how economic rationalities risk, with Mouffe (2013), taking the political out of politics, but also as an analytical strategy to move beyond treating regional democracy as a mini-version of nationally designed democracy.
Regional democracy, or democracy at the meso-level, is located at the intersection between local politics and politics at the national and international levels. Hence, politics at the regional level is in several respects characterized by dealing with multi-level political problems. Based on ideas in traditional regionalism, the institutionalization of democracy at the regional level needs to take into account at least two kinds of political issues; 1) the relations to societal processes at the local level and 2) the relations between urban and rural development.
First, the reform-work in Sweden illustrates the significance of considering issues anchored in local communities. For this reason, the reform work was carried out for 10-15 years by setting up inter-municipal federations in more than half of the regions in Sweden. Due to the conventional thinking of democracy as based on a representative model, the reform work came to move away from the inter-municipal model. The democratic objections that were directed at ideas on inter-municipal arrangements were in several respects relevant and worth considering. However, this change meant that the new regions partly lost their support and legitimacy with the municipalities. We argue that the distinctiveness of the region as a democratic institution rests in its anchoring in the local community in order to contribute to a policy that extends beyond economic growth and entrepreneurship. Moreover, cooperation within the inter-municipal federations shows the need for regional coordination of local policy issues such as climate change, public transport, social equality, and health issues.
Secondly, which was a central starting point in the regionalism of Lewis Mumford, the relationship between urban centers and surrounding ruralities is in many ways a crucial issue. In neoregionalism, the big cities were seen as the driving force for economic development in the entire region. A positive development in the urban centers creates "trickle-down effects" even to the benefit of rural areas. Although Mumford also regarded the urban environments as essential nodes for societal development, he emphasized balancing the allocation of resources between urban centers and the surrounding rurality. The fixation on economic growth in neo-regionalism has been an obstacle to developing perspectives and strategies on the relations between urban and rural areas.
Finally, we want to give a methodological reflection for regional Political Science. The development of the regions has been given much space in Political Science research in recent decades, not least regarding European politics. We agree with the criticism that other Political Scientists have raised; that research on the distinctiveness of regional democracy has received little attention, one example being that neo-regionalism never has developed a coherent theory on the relations between local communities and regional development. The methodological toolbox has all too much been based on what some researchers call methodological nationalism . In the Political Science analysis, the regions have been regarded as mini-versions of nation-states, and analyzes of democracy have often been overshadowed by a focus on the nation-state level as the norm and framework of understanding democracy. In many ways, the political analysis has thus lost several dynamic aspects of regional development policy when more subtle dimensions of power related to aspects like economization but also time and space tend to disappear in the analysis.

DATA AVAILABILITY STATEMENT
The datasets presented in this article are not readily available because The article uses and have not generated a new dataset. Requests to access the datasets should be directed to jorgen.johansson@spa.gu.se.