Is the propensity of a local government to reform dependent on its local autonomy? Evidence from Switzerland

Transcription

1 Is the propensity of a local government to reform dependent on its local autonomy? Evidence from Switzerland Paper for the IRSPM Conference 2015, Panel C101 Special Interest Group in Local Governance, Birmingham, March 30 th -April 1 st, 2015 Nicolas Keuffer, PhD candidate IDHEAP Swiss Graduate School of Public Administration University of Lausanne Unit of Swiss Administration and Institutional Policies IDHEAP Building CH-1015 Lausanne +41(0) nicolas.keuffer@unil.ch This work is in progress, please do not cite without contacting the author. Abstract This contribution is focused on local autonomy, the propensity of a local government to reform and the relations between these two concepts, which have rarely been examined. We consider local autonomy as a necessary institutional resource for local governments to have, so that they are able to take the initiative to reform. Drawing on the literature, we suggest a multidimensional concept of local autonomy making the distinction between seven dimensions. Regarding the reforms we distinguish six different types depending on their respective legitimacy orientation: changes towards more direct participation and representative democracy increasing input legitimacy, the supply of services by local governments based on the guidelines of the New Public Management or of Governance and functional and territorial reforms such as inter-local governments collaborations and mergers, whose aim is to improve both input and output legitimacy. The analysis are conducted in Switzerland, aggregating on the cantonal level the results of a huge set of data stemming from the National Survey of Secretaries conducted in 2009 in every Swiss local government. Swiss municipalities offer an excellent context to test the relations between local autonomy and local government reforms since this federalist country has considerable differences as far as the size and importance of municipalities in the 26 constituent units, as well as where its perception and the cultural and politico-administrative setting, are concerned. Additionally, the Swiss case is interesting when making an international comparison because of its municipalities small size and far-reaching autonomy and the importance of direct democracy. In first place we find that the differences in local autonomy degree among cantons can be explained on one hand by the language spoken area where the canton is located and on the other hand by the size and hence we identified three clusters. Secondly, differences regarding Swiss municipalities propensity to reform can be explained by the different factors separately. For instance, the intensification of IMC and of considering NPM reforms is higher in German-speaking and larger municipalities although concrete plans for a merger are more present in Italian and French-speaking and smaller municipalities. Finally, it is impossible to confirm that local autonomy as a whole concept is positively correlated with the propensity of local governments to reform. This speaks for an even further diversification of local autonomy and a clarification of the specific operationalization of local government reforms conducted in Switzerland. 1

2 1. Introduction This contribution is focused on the relations between local autonomy and the propensity of a local government to reform, which have hardly been examined (Lidström 1998, Wolman 2008, Kersting and Vetter 2003, Vakkala and Leinonen 2014). Facing a growing complexity of tasks and a decrease in the resources at their disposal to provide them, as well as the trend of globalization and the increasing level of citizens demands, many European local governments have had since the 1990s the propensity to launch reforms (Kersting and Vetter 2003). We consider a reform as a change process which has different phases such as the discussion, the decision (when the authorities decide to adopt a particular reform), the practice and the results (Pollitt 2002). The focus is here on the propensity of local government to reform, i.e. the intensions to launch a change project without taking into consideration the implementation processes, the practice or its results. We also make the assumption that leading managerial and political actors of local governments must possess a certain degree of legal discretion, decision-making power, scope of tasks to perform and personal and financial resources in order to launch modernization reforms. In other words, local governments must possess a necessary degree of autonomy to more or less lead to the introduction of an institutional change process. This idea of local autonomy as a precious asset for local governments is the subject of the European Charter of Local Self-Government, whose preamble expresses in this respect that local authorities [are] endowed with democratically constituted decision-making bodies and possessing a wide degree of autonomy with regard to their responsibilities, the ways and means by which those responsibilities are exercised and the resources required for their fulfilment (Council of Europe 1985). Drawing upon its tangible criteria and the literature, we suggest seven dimensions in order to measure objectively local autonomy, as an institutional setting. Being closest to the citizens, local governments are considered to be in an extraordinary position to answer to the citizens demands and problems. Indeed, on one hand, local level is often used as a playing ground and laboratory of reform (Vetter 2007). On the other hand, local governments proximity give them the opportunity first to guarantee an efficient and effective delivery of public services and second to integrate citizens into the democratic decision-making process (Goldsmith 1990). But the emphasis on either the local government s first function of responding satisfactorily to the collective preferences of its citizens (vehicle of local democracy) or on the second function of giving the possibility to citizens to exercise democratic control over the decisions of the polity (bureaucratic efficiency) consists in an efficiency-democracy dilemma (Pierre 1990: 38, Sharpe 1970). This can also be seen as the two essential parts of the local policy making process. Indeed, Easton (1953) first made the distinction between inputs into the political system, such as votes, demands, interests or identities of the citizens, and outputs which referred to the actual decisions and actions of governments. Drawing on Easton s model, Scharpf conceptualized government legitimacy as rooted in either input-oriented or output-oriented democratic thought (1970, 1999). As a result, local government reforms, if they had the ambition to increase the two forms of legitimacy at the same time, would come into contradiction (Dahl 1994, Mouritzen 1989, Dahl and Tufte 1974). In practice, it should therefore be better for the local government to not neglect either of them in order to promote the stability of the democratic system but rather to make a choice. Theoretically, distinctions should be made 2

3 among changing activities according to the objectives they pursue: internal reorganisations increasing local democracy (input), the supply of services by local governments (output) based on the guidelines of the New Public Management (NPM) or of Governance (both vertical and horizontal networks), and territorial reorganisations such as inter-local governments collaborations and mergers, whose aim is to improve throughput. Having output and input-oriented local government reforms as dependent variables and the multidimensional concept of local autonomy as an independent variable, the first question that this contribution aims to answer is to what extent is the propensity of a local government to reform dependent on its autonomy? The perspective adopted thus consists in considering the potential consequences of local autonomy. For Wolman and Goldsmith, this leads to ask more fundamental and wide questions about local autonomy: Do local governments in urban areas have autonomy in the sense that their presence and activities have independent impacts on anything important? Do local governments politics affect the well-being of their citizens? (1990: 3). Indeed, by improving either the government by the people or the government for the people (e.g. Kübler and Ladner 2003: 137) local government reforms may have a positive impact on the citizens day-to-day life. However, according to Wolman (2008) it is necessary to go further by looking more precisely at the causes of the different local government reform activities (see also Kuhlmann and Wollmann 2014). This investigation falls therefore within a comparative local government framework in such way as to identify the key factors leading to a higher propensity for local government to reform. The second purpose of this study is then to explain the differences between respectively varying degrees of local autonomy and the more or less high propensities scope of local governments to reform. In order to answer these two research questions, the analysis are conducted in Switzerland both rely on data of municipal level stemming from the conducted 5th National Survey of Secretaries of every Swiss local government and aggregate them on the cantonal (regional) level. In general Switzerland is more known for its stability. This is most certainly caused by a kind of cultural scepticism towards general change as well as system-inherent institutional impeding factors (Kriesi 1999). Nevertheless, since the second half of the 1990s cantonal and local governments have been overwhelmed by the execution and the financing of public tasks and have sluggishly lost their autonomy. At the lowest level especially, Swiss municipalities faced a number of problems and issues. These are for instance an increasing complexity and variety of tasks, although their capacity for action and independence were falling (Ladner et al. 2000, Ladner 2001), as well as the increasingly strong citizens demands for the efficiency and effectiveness of public action (Geser 1997). In response to these challenges, Swiss municipalities have amplified considerably their reform activities. Fitting into a large European local governments change trend (e.g. Kuhlmann 2010), Swiss municipalities also launched reforms in order to improve the efficiency of their administrations and to strengthen democratic participation (Kersting and Vetter 2003). Moreover, the reform of allocation of tasks applied in 2008 created a new form of multi-level governance by involving distinct levels of government and stressing the importance of cooperation instead of coercion and control (Van Kersbergen and Van Waarden 2004). In parallel the role of municipal level in the Swiss intergovernmental cooperation (Scharpf 1977) has been legally guaranteed through both the introduction of art. 50 in the federal Constitution and the ratification of the European 3

4 Charter of Local Self-Government (Council of Europe 1985). For all these reasons we make the initial postulate that in the 2000s Swiss municipalities launched rather proactive problemsaving reforms. Being a federalist country with considerable differences as far as the size and importance of municipalities in the 26 constituent units, as well as where its perception and the cultural and politico-administrative setting are concerned (Horber-Papazian 2006), the Swiss local level offers in addition an excellent context to explain the different relations between local autonomy and local government reforms. This contribution will therefore be structured as follows. First, we shall draw upon the theories to formulate hypothesis explaining the relations between local autonomy and the propensity of a local government to reform as well as their differences. We will then present the dimensions retained to measure local autonomy and their application on the Swiss case. In the following section we will show the different types of local government reforms and see to what extent these have been launched in Swiss municipalities. Finally we shall connect local autonomy with the propensity of Swiss municipalities to launch reforms. This contribution ends with a discussion of the results and some concluding remarks. 2. Theories, hypothesis and methodology Local government reforms are intentional, conscious and planned particular processes whose goal is to change the institutional system and that the local government has undergone. They have to do both with a government at the local level, which can be defined as the activity of process of governing or governance, a condition of ordered rule, those people charged with the duty of governing or governors, and the manner, method or system by which a particular society is governed (Finer 1970: 3-4) and with a local government as the political institution which usually takes place at the lowest territorial level of a country. In this respect, European local government institutional systems have four criteria in common (Lidström 1998): a local government has a clearly defined territory, executes a certain amount of autonomy, has authoritative power over its citizen and has directly elected decision-makers and/or municipal assemblies. The European Charter of Local Self-Government stabilizes the fact that it is a precious asset if a local government benefits from local autonomy, which denotes the right and the ability of local authorities, within the limits of the law, to regulate and manage a substantial share of public affairs under their own responsibility and in the interests of the local population (Council of Europe 1985: art. 3, 1). European and Swiss local governments, facing a growing complexity of tasks and a decrease in the resources at their disposal to provide them, as well as the trend of globalization, the reorganization of the political systems in Central and Eastern Europe and the increasing level of citizens demands, have had in the 1990s the propensity to launch reforms, most of which were New Public Managementinspired reforms (e.g. Wollmann 2004, Ladner et al. 2000, Hendricks and Tops 1999). In this perspective, this accumulation of local government reform activities was primarily a product of pressures coming from the deep mutations of the political, economic and social context. In Switzerland, for instance, local reforms were launched to improve the effectiveness of local government administrations, and the vertical allocation of tasks between the different federalist levels has been revised to take more into account the role of local governments 1. 1 Amongst others local governments must be consulted in order to provide grants through the contracts between Confederation and the cantons (Federal Assembly, 1990: Subventionsgesetz, art. 19.2). 4

5 Both the introduction of art. 50 in the Federal Constitution and the ratification of the European Charter of Local Self-Government have reinforced the recognition of the local tier in the Swiss state structure. Furthermore the speeding up of globalization, with increasing mobility and new communication technologies, has led to a greater integration of local governments (Klöti 2000). Finally, the pressure on the Swiss local governments tasks to perform and financial situation has decreased during the twenty-first century s first decade (Ladner et al. 2013). For all these reasons, we make the assumption that Swiss local governments have certain abilities in the 2000s to launch proactive problem-solving reforms. In other words, from the development theory perspective, the strategy of local governments has been to use their different resources, social and political capital and autonomy to incrementally mutate their institutions so that these could be in line with the aspirations of their citizens and not only be a response to the economic situation (Ladner 2011). In this respect, a study conducted in Switzerland shows that for about half of the nearly 180 investigated local government reforms of the project managers a direct link between the financial situation and the reform project was denied (Ladner 2009). In the literature, a variety of approaches exist which try to explain the factors that lead to reforms and give more or less capacity to the political institutions to change (for a more complete list, see Wagschal 1999): First we have the system theories, which give the political systems the capacity to adapt according to the environment s demands and the respective evolutions of other political systems (e.g. Easton 1953); Second, the actors theories identify the motivation of several actors as the necessary factor to reform: party theories stress the influence of the party composition of the government on the policy choices (e.g. Schmidt 1996), or public choice theories (e.g. Boyne 1998) consider that the motivation of a politician is to maximise their own interests which can be the same as the collective interest or only the will of being re-elected; Finally, we find rather anti-reform actor theories, which suppose that actors are so involved in politically connected and integrated institutions and that mutability is almost impossible (e.g. Scharpf 1978). Following Ostrom (1991), we consider here that the rational choice explanations of the actors motivations based on calculated selfinterest are complementary with the institutional analysis. Hence the analytic frame is drawn on the new institutionalism because, on one hand, of its interpretation of the different constituents and underlying logics of development of the institutions. Indeed it emphasizes the importance of the relative autonomy of political institutions because these are more than simple mirrors of social forces (March and Olsen 1984: 739). On the other hand, new institutionalism relies on different aspects for the explanations of the reasons for the reform s introduction and degrees of implementation (e.g. Kuhlmann and Wollmann 2014). In essence, three main variants of new institutionalism have been put forward and have also included an actor-centred approach (e.g. Hall and Taylor 1996, Peters 1995): - Historical institutionalism is based on the assumption that strategies and choices of the actors are predefined by institutional arrangements established for the long term (Peters 1999). Institutions are considered to be fairly persistent and the options for reforms are limited by the historical institutional path-dependence of the politicaladministrative system (e.g. Peters et al. 2005). As a result, the degree of propensity to reform as well as their practical effects depends on the pre-existing institutional setting and patterns of actions. Institutional incremental developments can nevertheless occur, 5

6 mainly in a critical juncture period. However, as Hall and Taylor (1996) note, historical institutionalism is often unable to have well-developed responses to what precipitated such critical junctures. - Rational choice institutionalism has more points of tangency with historical institutionalism than typically assumed in that institutional transformation resulting from deliberate actor strategies can also lead to long term structural change (Thelen 1999). Accordingly, reform processes should lead to very different results in different contexts, depending on preferences, strategic calculations and freedom of choices of the constellations of actors, which should be then considered as an independent variable (Kuhlmann and Wollmann 2014). In turn, institutions are seen as limitations of bounded-rational choices. Actor constellations, their resources for action, orientations and strategies are structured by institutional rules and procedures (Mayntz and Scharpf 1995). - Sociological institutionalism sees institutions as cultural phenomena that provide cognitive scripts and normative frames that could enhance the legitimacy of the organisation and its participants (Gorges 2001). Regarding reform processes and transformations, sociological institutionalists consider that formal organizational changes are accompanied by cognitive and socio-cultural adaptation processes and by an alignment of the qualification skills of the actors. It allows for an examination of the question whether formal rules are actually incorporated into operative routines, or if they exist only on paper. Additionally, since new formal institutions can face cultural risks of rejection, institutionalization is often the product of a simple adjustment process leading to institutional isomorphism (DiMaggio and Powell 1991). This study then draws on the reforms explanations of the three different varieties of institutionalism. With the conceptual triad of polity, politics and policy (e.g. Palonen 2003), it considers local autonomy as an institutional setting of local government (polity) for which framework conflicts between actors (politics) may lead to the formulation and implementation of local government reforms (institutional policy). As mentioned supra we assume that Swiss local government reforms in the 2000s were launched in a pro-active perspective and in a good economic situation. As a result we do not on one hand take into account the pressures related to the scope of tasks to perform and to the financial situation and on the other hand we consider local autonomy as a necessary resource to organize and/or to implement local government reforms. Indeed, several authors showed the positive relationship between financial, social, economic and political resources and the strength of reforms, especially because failures can be better managed (Ladner et al. 2000, Brunsson and Olsen 1993, Walker 1969). Like Kersting and Vetter, we make the assumption that local autonomy as an institutional setting may affect the reform processes at the local level and thus formulate the first hypothesis that local government reforms depend on the degree of local autonomy with local autonomy being related to stronger reforms with regard to efficient local service delivery as well as higher citizens integration in local policy-making (2003: 22). Hypothesis 1: The higher the degree of local autonomy, the higher the propensity of a local government to reform. 6

7 When talking about own resources at the disposal of a local government, the size plays a particularly important role. Indeed, it is generally admitted in the literature despite certain practical biases that should be taken into account (e.g. Boyne 1995, King and Ma 2000) that bigger local governments have more resources than smaller ones since they can benefit from economies of scale: a situation where the political boundaries are tighter than the scale of production which minimizes unit costs, calls for a larger size of the group up to the point where economies of scale are exhausted (Dafflon 2006: 284). Furthermore, basic administrative costs do not weigh as heavily in bigger than in smaller local governments (e.g. Council of Europe 2001). We thus expect that the size of the population has a positive impact on the propensity of a local government to reform. Hypothesis 2: The bigger the local government, the higher its propensity to reform. As the sociological and historical institutionalism approaches suggest, the environment of a local government may also have an influence on the propensity to reform, and this for several reasons. First, financial or economic crises may to a greater or lesser extent impact on the local governments depending on their path dependent institutional settings which have been constituted through a cultural long term period (Ladner 2011). Second, since the propensity to reform may be influenced by cantonal incentives it is necessary to take into account bigger areas than only local governments (e.g. inter-cantonal conferences). Indeed, the ideas of reforming and underlying potential additional financing expectations are evoked in the framework of institutions for collective actions (e.g. Ostrom 1990). Finally, a reform s diffusion depends on context factors like class, geography, climate, ethnicity, language, culture, economic conditions, demography, technology, ideology and religion (March and Olsen 1984: 735). Consequently local governments which have common features may be more receptive through the diffusion of new values and norms and will develop quiet similar institutional settings. We then make the hypothesis that the propensity to reform is higher in municipalities located in same language areas because reform discourse diffusion may go over the cantonal boundaries and have the same orientations in a similar cultural context. Hypothesis 3: The more a local government comes from the same cultural part as others, the more their propensity to reform may be the same. The last hypothesis is related with actor-centred institutionalism. The stronger the interests, political preferences, wills and skills of local actors the weaker the path-dependent institutional setting (i.e. the local autonomy) may be. Leadership may also have the capacity to gather or to free up necessary resources by creating for instance coalitions (Sabatier 1988). Local actors may either be leaders in the administration, newly elected executive members, who are the major players in local politics (Geser et al. 2012), or citizens through direct participation, which is very developed in Switzerland (Ladner 2009). We will focus more precisely on the two latter types of actor, by examining also whether the party affiliation may influence the propensity of a local government to reform. Hypothesis 4: The more active the local government s actors, the higher the propensity of a local government to reform. Since mobilized factors are very diverse in nature, these four hypotheses will be tested in the framework of a comparative research design. Using an exploratory strategy of analysis i.e. 7

8 certain univariate analysis will be made between the different dimensions, for which we cannot expect the exact respective relation, the comparative approach of this investigation should give the opportunity to look more precisely at the differences in local government reform activities, which until now were rarely discussed (e.g. Kersting and Vetter 2003). In this respect, Switzerland represents a quasi-experimental context to analyse questions relative to local autonomy, local government reforms and the relations between the two because of the differences in terms of institutional settings at the local level. This contribution will use a most different systems design (Przeworski and Teune 1970, Meckstroth 1975) since there is no other country where the differences between municipalities are so important regarding certain variables that we need to isolate. In return, the advantage of such a research design compared to cross-national studies lays in the fact that Swiss local governments are more homogenous with regard to many variables concerning federal laws or cooperative federalist territorial organisation. The analysis will be conducted both at the local and on an aggregate level using the 26 cantons as units. For some variables, we will have to rely on data at a municipal level using the 2350 or so Swiss municipalities as units of observation. Looking at each municipality is particularly important for aspects of local autonomy and local government reforms since we do not only expect differences between municipalities from different cantons (for example in the case of legal and political aspects) but also between municipalities from the same canton (like in the case of their size and economic and financial situation). Since there are no official statistics available on Swiss government reforms at the local level, we have to rely largely on survey data. The data presented here stem from the fifth nationwide survey of local secretaries, as they are the ones with the best insight into local politics (see methodological appendix). This survey also provides us with the control variables we mentioned in the hypothesis as well as the elements necessary to measure both objectively and subjectively local autonomy. But that is the subject of the next chapter. 3. Local autonomy, empirical theories to measure local autonomy and Swiss municipalities autonomy Local autonomy and empirical theories to measure local autonomy Local autonomy is important in the way a state organizes its territory and provides services to its population. This concept is therefore often mentioned in many reports, both on a global 2, European 3 and Swiss level 4. Three major issues related to local autonomy can be highlighted at this stage. First, as suggested by the mentioned references, the local government finds itself in a strong normative perspective, according to the general idea that the local autonomy can be viewed as the stumbling block of a real democracy where decisions taken meet citizens direct demands (e.g. Tocqueville 1968, Mill 1991). In addition, it may increase policy diversity and lead to policy innovation and through its possible impact of political interest and citizen education at the local level, it also acts as a democratic countervailing power regarding 2 See for instance the Gold Report by United Cities and Local Governments (UCLG 2008). 3 See for instance Local authority competences in Europe by The European Committee on Local and Regional Democracy (CDLR 2007). 4 See for instance the 4 th cantons monitoring Autonomie communale entre illusion et réalité by Avenir suisse (Rühli 2012). 8

9 the higher levels of government (Goldsmith 1995). It is thus the subject of a Charter, ratified by 47 member states of the Council of Europe (statement on ), whose various tangible criteria are regularly evaluated and subjected to recommendations 5. In this regard, we get to our second point. Local autonomy in the academic field knows problems of definition: "Local autonomy means many different things to different people" (Clark 1984: 205), which also implies terminological ambiguities. In this respect, Wolman et al. define local government autonomy as a system of local government in which local government units have an important role to play in the economy and the intergovernmental system, have discretion in determining what they will do without undue constraint from higher levels of government, and have the [ ] capacity to do so (2008: 4). Third, the literature is not very specific when it comes to defining tangible dimensions to measure empirically the different degrees to which local government may be characterized as autonomous (Hansen and Klausen 2002). Writing in the early 1980s, Clark (1984) suggested a theoretical framework to clarify the meaning of local autonomy. In reference to the two principles of power derived from Bentham he defines local autonomy with two specific powers: initiation and immunity. Initiation is the competence of local government to carry out tasks in the local government s own interests. By contrast, the power of immunity means the possibility for the local government to act without being under the control of higher levels of government. Combining the two principles of local power Clark identified four ideal types of autonomy. Under Type 1 autonomy local governments have both the powers of initiative and immunity from higher levels of government. Type 4 autonomy, on the contrary, characterizes local governments which are administrative arms of higher tiers of the state in the sense that they hold no power of initiative and are subject to a strong control. Type 2 autonomy can be described as decentralized liberalism. It allows the local government to act in its own interests, but makes their decision subject to a control by higher levels of government. Finally, Type 3 autonomy also has limited autonomy in a sense that the local government enjoys no powers of local initiation but has no fear of higher tiers of the state because of their immunity. Since the power of initiative is crucial according to Clark, this latter type holds less autonomy than Type 2 (Clark 1984). This theory is the most developed one from a classic political science perspective on local autonomy. The approach is based on a constitutional and legal understanding of central-local relations ( freedom from ); consequently, the focus on local autonomy is mainly top-down, analysing to what extent higher levels of government delegate tasks and concede competences, without paying attention to the real capacities of the local government to act and thus express its local identity (Pratchett 2004: 363). Attempting to deal with the neo-marxist arguments of relative autonomy, Gurr and King concentrated themselves not only on the limits imposed by the higher levels of governments upon a local government but also on a multitude of local factors: the autonomy of the local state in advanced capitalist societies at any given historical juncture is a function first of its 5 It can be mentioned the extent to which the principle of local autonomy is enshrined in written constitution and law (art. 2), the extent of powers and responsibilities which authorities the closest to citizens appertain (art. 4), the value of local governments consultation in the decision making process (art. 4), as well as the degree of their influence capacity and judicial remedy (art. 5, 8 and 11), and finally the importance of financial, administrative, fiscal and human resources they actually have at their disposal (art. 6 and 9). 9

10 relationship with local economic and social groups, and second of its relationship with the national or central state (1987: 56). These two sets of relationship are summarized into two dimensions and together they determine the degree of local government s autonomy. Type 1 autonomy thus depends on local economic and social factors. It is more concretely constrained by the extent of the effective revenues which can be extracted from the local economy, the capacity of economic actors to control the local political agenda and the presence of local political organizations and social movements able to resist or reshape the local policies implemented (Gurr and King 1987). To ensure its perpetuation, a local government should be able to count on the local economy as well as on local taxes. With the slow decline of the local economy, local governments became more and more financially dependent on higher levels of government. The financial constraints of the Type 1 autonomy can be overcome, but in return higher levels of government increase their control through the financial resources granted to the municipalities. As a consequence Type 2 autonomy decreases (Gurr and King 1987). Indeed, Type 2 autonomy concerns the extent to which a local government can pursue its interests without being limited by constitutionally-specified constraints, strict objectives accompanying subventions and national political pressures on policies (Gurr and King 1987). To identify how Type 2 autonomy can be used analytically, Goldsmith (1995) summarizes the limits imposed on local governments by higher levels of government in five headings. First, a local government s autonomy depends on its legal situation: the constitution and laws determine its competencies as well as the control range of the higher levels of government. Second, the range of functions delegated by higher levels of government has to be taken into account. The third factor stresses the fact that the more tasks a local government is responsible for, the higher its autonomy and this, of course, under the condition that it holds discretion 6 to perform these functions. The forth heading expresses the idea that functions cannot be performed without financial resources. Here, it is the financial competences that are relevant (e.g. the ability of the local government to set its own tax rates). Finally, the degree of influence which a local government is able to exert over higher levels of government is also an important factor. This political influence expresses itself through both an indirect and a direct access to national decision-making (Page 1991). Indirect influence should be observed for instance through local government interest groups or associations. On the other hand, direct forms reflect formal individual relationships between local and higher levels of government s representatives. These two ways of vertical influence on central governments can have real consequences on policies implemented on a local level. But again, it seems to some authors that local governments should have resources mainly financial to be truly autonomous (Pratchett 2004, King and Pierre 1990). More recent researches try in a comparative perspective to measure more systematically the degree of local autonomy of a large number of countries and subnational tiers (Hooghe et al. 2010, Wolman et al. 2008, Sellers and Lidström 2007). It is interesting to note that the dimensions used to measure the degree of local autonomy are different. Comparing local government autonomy across the U.S. states Wolman et al. (2008) define for instance local 6 Discretion refers to the ability of actors within local government to make decisions about the type and level of services it delivers with the formal statutory and administrative framework for local service delivery, and about how that service is provided and financed (Page and Goldsmith 1987: 5). 10

11 autonomy in terms of three dimensions: local government importance, local government discretion and local government capacity. Examining variations among regional authorities across states, Hooghe et al. (2010) distinguish between elements concerning the extent to which a regional government has authority over those who live in this territory (self-rule) and the influence of regional governments to shape national decision making (shared-rule). This distinction supports the local governments typologies of Page and Goldsmith (1987) and of Goldsmith and Page (2010) in terms of the range of functions delegated to them, the discretion they have to fulfil them and the access they have to national decision-making. Many other authors have taken into account local autonomy in their classification of the local government systems from a comparative perspective by in particular distinguishing functional, territorial and political profiles (Kuhlmann and Wollmann 2014). Among many others 7, the well-known typology of Hesse and Sharpe (1991) made the distinction between three different groups of Western industrialized democracies by using political, structural and functional criteria. This review of literature among the theories of local autonomy, the theories to measure local autonomy as well as the comparative researches on local governments systems and the typologies which emanate from them give us the opportunity to develop a comprehensive map (see Figure 1). By trying to differentiate other researches on local autonomy amongst various local autonomy s dimensions and approaches, it is interesting to see that all the combinations exist (see the approaches configurations, in appendix). We can also note that traditionally researches on local autonomy are mostly focused on formal rules, i.e. the legal or constitutional autonomy granted to local governments as well as fiscal decentralisation rules and the existence of legal remedy (Legal framework: B). Legal regulations may also attribute local governments competencies. But some researches concentrate on the way local governments use practically their financial own-revenue to deal with the scope of tasks they are responsible for, their expenditures as well as their outputs (Importance: A). Another domain of investigation consists in focussing on to what extent local governments, in order to bring up these services autonomously, have their own administrative means (local leadership, management competence, administrative capacity) and have sufficient room to decide what they want to do and how (discretion, local referenda, social and economic horizontal independence) (Capacity: C). Finally, some authors focus on two political aspects. On one hand, it is interesting to see the way local governments are organising democratically their decision-making bodies and their legitimacy (local democracy, local identity). Since it may have to do with a higher local government (e.g. direct democracy), we place this aspect on the bottom-right side of the map (shared-rule). On the other hand researches are focussed on the relation with the higher tier (vertical independence), seeing for instance to what extent local governments have the ability to influence the higher political levels through formal or informal upward access (D: Access). 7 See e.g. Dreier (1994), Lane and Ersson (1987), Swianiewicz (2014), Humes and Martin (1969), Batley and Stoker (1991), Huber and Stephens (2001), Esping-Andersen (1990), Lidström (2003), Bennett (1993), Bours (1993), Goldsmith (1992), John (2001). 11

12 Self-rule Figure 1: Comprehensive map of local autonomy s dimensions and approaches Financial and legal criteria Importance Legal framework Shared-rule Capacity Access Economic, social and political criteria Source: inspired mainly by Clark (1984), Gurr and King (1987), Page (1991), Goldsmith (1995), Wolman et al. (2008) and Hooghe et al. (2010). Having the different elements of local autonomy depicted by theory as well as the possible data to measure these elements in mind, we suggest to make the distinction between legal, financial, administrative, economic, political and social aspects of autonomy giving local governments the possibility to govern themselves independently and to look at the relationship between the municipalities and the higher tier (Ladner and Keuffer 2014): 1. Legal autonomy (B): The criteria of local governments legal position are the specification in the legislation of the higher tier(s). We are particularly interested in to what extent they guarantee the existence of the local governments and grant them freedom when it comes to the organization of their political system. 2. Financial autonomy (A): In general, local governments obtain the financial resources to fulfil their task from transfers from higher levels and trough taxes and fees. As for the transfers, it is not only important what percentage of the total income stems from transfers but also to what extent these transfers are conditional and linked to specific tasks and activities. As for the tax raising power, the questions are what kind of tax municipalities are able to levy, to what extent they are able to decide on the tax rate independently and again the part of their expenses they cover themselves. An additional element which could be taken into account here is the local governments possibilities to take up loans and to get into debt. 12

13 3. Policy autonomy (A): Most importantly is, of course, the scope of tasks local governments are concerned with. To what extent local governments deal with, for their inhabitants, crucial tasks such as welfare, health, schooling, land use planning, etc.? To some extent this can be measured by the percentage of public expenditures which are spent by local governments. 4. Administrative autonomy (C): To what extent do local governments have their own personnel administer and fulfil their tasks? It makes a difference whether civil servants in charge of local activities are paid by higher tiers or remunerated by the local government: a powerful administration can be an important element of a local government s system capacity. 5. Decisional autonomy (C): The range of tasks local governments are concerned with is one thing; another thing is whether the local governments dispose of any competences when it comes to deciding on the fulfilment of these tasks. Can they take decisions themselves or are they simply executing decisions taken on higher level? 6. Social autonomy (C): Restrictions on the local authorities to take decisions can also come from within the local government with regard to interested groups, important tax payers or entrepreneurs preventing the authorities to decide independently. 7. Vertical influence (D): The question here is, to what extent are local governments able to influence decisions on a higher political level? Another possibility is through representation in decisional bodies on a higher level or through organizations representing the interests of the municipalities. This aspect also contains the range of control the higher level disposes of. Is this control concentrating on lawfulness or does it also contain political aspects? We will test this conception empirically using Swiss municipalities as an example. In Switzerland, it is more common to use the term of municipal autonomy than local autonomy (Federal Council 2003). Considering the relatively strong position of Swiss municipalities, their autonomy has also been a recurrent matter of interest. In the Swiss context, municipalities are subject to a very strong sentiment of belonging to the higher tier, which is the regional one (the 26 cantons). Consequently the 26 different cantonal legislations have to be taken into consideration. According to Meylan et al. (1972) and other studies on the different dimensions of local autonomy (e.g. Fiechter 2010) however, local factors must be also taken into consideration when measuring the effective local government s leeway 8. Various studies in the Swiss context highlight the multidimensional nature of local autonomy and the difficulty when defining practically the actual degree of local autonomy or differentiating regional governments according to it (Geser 1997, Ladner et al. 2000, Fiechter 2010). To Horber-Papazian and Jacot-Descombes (2014), local autonomy is divided between the competences decisional and implementation and the capacities administrative, financial and operational. An attempt to capture the simultaneous variation of the cantonal and intra-cantonal local autonomies can be seen in the concept of perceived autonomy. In Swiss surveys the municipal secretaries are asked to indicate the degree of autonomy their municipality disposes of on a scale from 1 to 10 (Ladner 1994). 8 Pierre has also defined local autonomy in this sense: local autonomy is defined as consisting of the maximization of local government policy-making powers, revenue-raising powers and implementation capacities. is constrained by central state policy and actions, local and national economic trends and local political and social interests (1990: 38). 13

14 Studies trying to classify Swiss cantons according to local autonomy are most often of a legal nature. Giacometti (1952) divides the cantons into three different groups, depending on whether the municipalities have a large, medium or small autonomy. Other comparisons were conducted by Meylan et al. (1972). For them there only is real autonomy under two conditions: having sufficient resources to perform the tasks and having enough decisionmaking power. Consequently the authors base their comparisons on self-established socioeconomic and financial typologies of Swiss municipalities. In the next subsection, after giving some descriptive information on the Swiss municipalities, we will apply the objective concept empirically on Switzerland. We will then give the results of perceived autonomy mostly thanks to the data stemming from the 2009 monitoring of Swiss municipalities (see the Appendix on methodology). Additionally, we will develop a typology explaining the differences through cluster analyses. Swiss municipalities and Swiss municipalities autonomy Switzerland is a highly decentralized federal state which comprises three territorial levels: the Confederation (national), 26 cantons or half-cantons (regional) and the municipalities (local). In direct contact with citizens, the municipalities (in the political sense 9 ) are an important pillar of the political system and political culture in Switzerland and their autonomy has a high political value (Linder 1999). As we will see later on in this subsection, apart from their far reaching political autonomy most outstanding for the Swiss municipalities is their small size and the fact that until recently there have been no territorial reforms. As by January 1 st, 2014 Switzerland counts municipalities. Between 1850 (the Swiss nation state was founded 1848) and the 1990s, the number of municipalities remained relatively stable. The small decrease from 3205 to 3021 (FSO 2002) was mainly due to the incorporation of neighbouring municipalities into cities. Since the 1990s amalgamations have become rather popular with some cantons actively promoting and financially supporting them (Steiner 2003). Between 2003 and 2013 the total number of municipalities decreased from 2842 to 2396 (see Table 1). Some cantons have a few hundreds municipalities, others have less than ten. Given the small size of the country and the larger number of municipalities, their comparatively small size is another significant feature (Ladner 2002). The average number of inhabitants per municipality was in 2012 (see Table 2). The median was meaning that half of all Swiss municipalities have less than inhabitants. There are again remarkable differences between the cantons. In general, the French-speaking cantons 10 except Geneva (GE) have smaller municipalities than the German-speaking cantons. 9 It is important to note that all cantons have general purpose municipalities on their territory (Einwohnergemeinde or politische Gemeinde, commune, comune), but some cantons also have special purpose municipalities: the ecclesiastical, bourgeois, educational or other special types of municipalities. 10 French-speaking area is constituted with following cantons: GE, NE, VS, VD, FR and JU; TI is the only canton of the Italian-speaking area; the other cantons are situated in the German-speaking area. 14

15 Table 1: Evolution of the number of municipalities by canton between 2003 and 2013 (ranked by the number of municipalities located in the canton) 3-19 municipalities and more BS 3 UR 20 NE LU GR AI 6 AR 20 TG 80 SO AG OW 7 GL 29-3 JU VS FR NW 11 SZ 30 BL 86 ZH 171 TI ZG 11 SH SG VD GE 45 BE Source: (own calculation). Table 2: Municipalities size per cantons in 2012 (ranked by the average) Canton Size of the municipalities Canton Size of the municipalities Average Median Average Median BS SH GL AG ZG AR GE AI ZH BE SG TI OW VS SZ VD LU SO NW UR NE FR BL JU TG GR CH Source: (own calculation). The legal autonomy Swiss municipalities enjoy depends to a very large extent on the cantons. As a result of this, their autonomy varies considerably from one canton to another. Some of the cantons mention their municipalities by name in their constitution which means, for example, that an amalgamation needs an amendment of the constitution. This automatically leads to a popular vote and needs the agreement of the majority of the citizens in this canton. In other cantons, at least the right of existence is granted which also prevents the cantonal authorities from amalgamating municipalities against the will of their inhabitants. Cantonal constitutions and cantonal laws also regulate the municipalities competences when it comes to their tasks and their political systems. In some cantons there are very few prescriptions whereas in others, the form of the legislative body, the size of the authorities and the electoral system, for example, are defined by the cantonal level. According to Fiechter s analysis of the legal documents of all 26 cantons, the legal autonomy is highest in BS and BL and lowest in VS (Fiechter 2010: 61f, Table 3 in appendix). Regarding financial autonomy, we concentrate on the financial equalization system since all Swiss municipalities have rather similar competences to create their own income through income tax where they can fix the tax rate themselves. The data stem from an expert survey conducted in all Swiss cantons (Rühli 2012: 154). If the financial resources of a larger number of municipalities depend on a considerable amount of transfers from higher level or from other municipalities their financial autonomy is low. High scores on this dimension are found for the cantons of GR and BL, low scores for TG and OW. 15

16 The municipalities policy autonomy considers the scope of their activities. This can be measured through the municipalities expenditures compared to the expenditures of the municipalities and cantons altogether. Additionally, we look at the different tasks the municipalities are involved in. The more tasks and the higher the expenditures, the more autonomous the municipalities are supposed to be. ZH and GR, for example score high on this dimension whereas BS where almost all tasks are in the hand of the canton scores very low. The administrative autonomy of the municipalities within a canton can be measured through the percentage of civil servants being employed by the municipalities. For a measurement on municipal level we have to look for instance at the number of local civil servants for one thousand inhabitants. A high degree of administrative autonomy is found in the canton of AG; a low degree in the cantons GE and BS where local tasks are mainly administered through the cantonal authorities. The decisional autonomy measures the difference between the decisional power and the financial responsibilities. These data stem from an expert survey conducted in all Swiss cantons (Rühli 2013: 164f.). In the canton of GL municipalities seem to have a very high political autonomy; in the canton of ZH where the scope of the municipalities is large, the decisional autonomy is low. Another important element is to what extent local authorities are exposed to pressure from other actors in the municipalities such as interest groups, entrepreneurs or parties (social autonomy). We find considerable differences between municipalities. Indeed, a high degree of social autonomy is found in AI; a low degree in FR. Not all cantons are heard on higher political level with the same degree (vertical influence). In some cantons, municipalities have the possibility to do this through the means of direct democracy (i.e. the local authorities have the opportunity to launch initiatives and referendums). Another mean to measure this is the perceived influence on various bodies and institutions on cantonal level, such as the cantonal government, the cantonal parliament, cantonal committees and the cantonal administration. The municipalities seem to be very influential in BS and AI and considerably less influential in TI. A simple aggregation (means) of the different dimensions allows for a first overall ranking of the cantons when it comes to the empirical autonomy of their municipalities (see Table 3, in appendix). High scores are found for AR and LU and low scores for FR and GE (see Figure 1, in appendix). Regarding now the self-perceived local autonomy i.e. local autonomy subjectively perceived by all the Swiss municipalities chief administrative officers in 2009 it is very interesting to observe that in general the results are very in line with the empirical autonomy the Swiss municipalities dispose of. Indeed, the cantons of GE and FR are still in the lowest although the latter has this time a better rank and the best cantons ranked are also more or less the same (see Figure 2, in appendix). These similarities, in addition to the existing correlations between the different variables (see Table 4, in appendix), confirm that the seven dimensions used capture well the complexity of the local institutional reality. Furthermore, these analysis, which are conducted at the municipal level (i.e. without aggregation) show that the factors of size (population) and culture (language) seem to matter in order to explain the 16

17 differences. We can respectively see in Figures 3 and 4 (in appendix) that French and Italianspeaking local secretaries perceived their own autonomy in a much more unfavourable way than the German, and that local autonomy s perception is gradually higher when the municipalities are bigger, although it slightly decreases in cities with inhabitants. With the aim of testing these explanation tracks as well as bringing up to date the existing typologies for Swiss local autonomy, we made two cluster analysis with comparable results: the first one for cantonal aggregated local autonomy (see Figure 5, in appendix) and the second one for local secretaries self-perception of local autonomy (see Figure 6, in appendix). In this latter dendrogram, clusters are clearer. Indeed, it is easy to identify 3 groups: - The French and Italian-speaking cantons: GE, JU, NE, TI, VD, FR plus SO and BL (8 cantons whose mean is under 4.12/10); - The North-Central cantons (which have a relatively high average of population): GL, NW, LU, SG, BE, ZH, AG, SH and SZ (9 cantons whose mean is 5/10); - The Eastern cantons (which have a relatively high average of population): AI, AR, UR, GR, OW, TG, ZG, plus BS and VS (9 cantons whose mean is between 5.56 and 7/10). If we set BS aside in the analysis of objective local autonomy, since it has only three municipalities (see Table 1), we can also observe three distinct groups at the number 10 of rescaled distance which widely coincide with the presented typology (see Figure 5, in appendix). As far as the consequences of local autonomy are concerned, we focus on the results local governments can achieve in regard to different constitutional, economic or political resources ( freedom to approach, according to Pratchett 2004: 365). The next section deals therefore with the propensity of local governments Swiss municipalities in particular to reform and the explanation of its differences. 4. Local government reforms and the propensity of Swiss municipalities to reform Local government reforms are currently on the agenda throughout the world (Caulfield and Larsen 2002). As we said above, European local governments have increased their reform activities since the 1990s in different aspects that could be conceptualized. Considering administrative reforms, as institutional policy, Kuhlmann and Wollmann make for instance the distinction between external (or outward-looking) and internal (or inward-looking) administrative reforms. ( ) Administrative reforms focusing on external relations (external institutional policy) are aimed at changing functional and/or territorial jurisdictions, membership rules and relations between organizations at different levels or sectors. ( ) Internal administrative reforms, in contrast, are concerned with changes in the distribution of responsibilities and resources within administrative organizations and between internal administrative units as well as the reorganization of decision-making and cooperation rules (2014: 37). Swiss municipalities have been left untouched by reforms until the 1990s (Ladner 2005). Several proposals form Swiss institutional reform typologies. A possible classification 17

18 of reforms is for instance provided by Germann (1996), who made the distinction between five categories: territorial, functional, administrative, financial and management reforms. Since we make the distinction between the input and output-oriented legitimacy affectation of the reforms (Scharpf 1999) and because we focus on the propensity of Swiss municipalities to reform in the 2000 s their reform activities mirrored almost perfectly the European pattern (Ladner and Steiner 2005), we will rather refer to the categories mentioned by Kersting and Vetter (2003): Table 5: Classification and legitimacy orientation of Swiss municipalities reforms Reforms of the Local Administration Legitimacy Local Political Reforms Legitimacy Functional and Territorial Reforms Changes towards more Direct Participation (5) - Reform of allocation of tasks - Abolition of the municipal parliament - Intensification of Intermunicipal cooperation (IMC) (1) - Concrete plans for a merger of municipalities (2) Management Reforms - Considering New Public Management (NPM) reforms (3) - Considering (horizontal) governance reforms (4) Inputoriented Outputoriented Changes in Local Representative Democracy (6) - Increasing of the municipal s competences - Change in the electoral system of the executives from majoritarian to proportional Throughputoriented The widest reform launched in Switzerland has been the reform of the reallocation of tasks, applied in It caused the transition from an executive to a cooperative federalism, taking into account the three government levels. This functional reform of the vertical governance (Hooghe and Marks 2003) had therefore some formal impacts on Swiss municipalities, such as the introduction of art. 50 in the federal Constitution 11. However, it is difficult to show the effective consequences because data are missing and its introduction at the federal level has only indirectly led cantons to reform their allocation of tasks with the municipalities 12. We will thus not take it into account here. Few territorial reforms have occurred in Switzerland in comparison to other European countries. In contrast to other European countries, Swiss municipalities have never implemented comprehensive regional reforms. Until 2003, their number has declined very little (see Table 1) and the intensification of IMC still represents the most frequent change even if IMC has always been very developed in Switzerland (Steiner 2003). Regarding the responses relative to the propensity of Swiss municipalities to spread more IMC, we notice that they have strengthened in three quarter of the municipalities between 2005 and 2010, especially in big municipalities (see Table 6, in appendix). With reference to the merger The autonomy of the communes is guaranteed in accordance with cantonal law. 2. The Confederation shall take account in its activities of the possible consequences for the communes. 3. In doing so, it shall take account of the special position of the cities and urban areas as well as the mountain regions. (art. 50) 12 Very often, the respective cantons have had the initiative to reform their allocation of tasks with the municipalities at the same time when they were reviewing the cantonal constitution (for a list, see Ladner 2011). 18

19 activity, whose reasons in favour are often related to more efficiency and effectiveness in service provision and to the strengthening of municipal autonomy (Steiner 2002), the results show that merger reforms are nowadays less discussed or launched in big municipalities and in German-speaking cantons (Table 7, in appendix). Most of them have rather taken place in rural cantons with a large number of small municipalities (TI and FR, for example). The guiding principles of NPM consist in a package of reforms affecting the entire political administrative system. This makes a categorization more complex. Even though it could limit the reality we will focus on the structural, procedural and leadership changes in Swiss municipal administrations (Steiner and Kaiser 2013). According to Kersting and Vetter (2003), most of the local government reforms in Switzerland occurred precisely in the field of management reform strategies. Previous researches showed that some municipal administrations have introduced management instruments recommended by the NPM discourse but that did not have the NPM appellation (Steiner 2000, Ladner et al. 2000, Ladner 2005a). To avoid this phenomenon, we focus on the introduction of both individual NPM measures and NPM-labelled project. We observe that the propensity of Swiss municipalities to have the initiative of NPM reforms is more developed in the Germanspeaking part of Switzerland and in big municipalities (see Table 8, in appendix). We call governance reforms changes which relate to the relationship between sectors, in particular the public and private ones, and to the idea of governing without government (Rhodes 1996: 652). In our survey it includes the outsourcing, the public-private partnerships (PPP) and the introduction of a competition between the external service providers. It is then related to a kind of horizontal decentralization (Bennett 1989). In Switzerland this has been welldeveloped for a long time. Regarding the ten last years, the propensity of municipalities to launch governance reforms is primarily correlated with the size (see Table 9, in appendix). Switzerland is without a doubt the model country in terms of direct democracy. At the local level citizens have two main ways to bind political decisions whose existence and conditions vary from a canton to another (Ladner and Fiechter 2012): the procedures of direct democratic participation (referendums and initiatives) and the direct democratic assemblies (instead of a parliament). As a result of this far development of direct local democracy, changes are not frequent in Switzerland in European comparison (Denters et al. 2014, Kersting and Vetter 2003). For instance, mayors are in every municipality directly elected or local referenda have difficulty to replace traditional initiatives. Therefore, we concentrate on the abolition of the parliament (for a direct assembly) as an aspect of change towards more direct participation and this on both the increase of local parliament s powers and the modification of the executive election process from the majority to the proportional as changes in local representative democracy. Regarding the latter aspect, the results of analyses show that in small municipalities in particular, a proportional system may lead to higher local democracy (Ladner 1996). Our results show that only seven Swiss municipalities had the propensity to abolish their parliament in the last ten years. As could be expected, more than half of these reforms took place in municipalities that have between 1000 and 5000 inhabitants (see Table 10, in appendix). Regarding the propensity of Swiss municipalities to change for more representative democracy in the last ten years it is difficult to notice a clear tendency. It seems that the initiatives of reform have mostly occurred in big municipalities and in municipalities that are located in the Italian-speaking area (see Table 11, in appendix). 19

20 Finally, the correlation matrix (see Table 12, in appendix) between the context variables and the six different legitimacy-oriented reforms gives us two types of information. On one hand we observe that NPM reforms are most strongly correlated with Governance reforms and participation reforms with participation reforms. The legitimacy orientation seems therefore to matter. A negative (not significant) correlation seems to exist between IMC and merger reforms. This could refer to the consideration of IMC as an alternative to mergers. On the other hand, we can note that correlations are not very strong between the explanatory variables and the various types of reform. Nevertheless, we can say without a doubt that the bigger the municipality, the higher the initiative to launch managerial reforms. 5. Local autonomy and the propensity to launch local government reforms in Switzerland This section aims to test the relations between the propensity of Swiss municipalities to have the initiative of the six different types of reforms, as dependent variables, the seven dimensions of local autonomy as well as its self-perception, as independent variables, and the control variables. In order to match with the hypothesis, the latter are threefold: context variables (population size through the mean and the median), cultural variables (language and confession) and actors-centred variables (percentage of members of the Social Democrats within the local executive and citizen s participation at executive elections) because they may have an important role to have the initiative of reforms. In the univariate analysis made (see Table 13, in appendix) all the variables are aggregated on cantonal level. The correlations coefficients based on the data we used so far, are not overwhelming. There are for instance no significant results for total local autonomy. Considering one by one the reforms, there are at least some correlations that we can explain and connect to our assumptions. First, the IMC is viewed as a good alternative to mergers by the members of the Social Democrats but demand means in the long run. It is therefore also positively correlated with the freedom of organization and existence (legal autonomy) and financial autonomy. Second, it is then quite logically that the propensity to have concrete plans for a merger is on the contrary negatively correlated with legal and perceived autonomy. Initiatives of mergers have been higher these last ten years in Italian and French-speaking cantons, so much so that a link with the cultural variable can be made. Third, results show that local autonomy can be generally viewed as a necessary resource to consider NPM reforms, thus the higher the municipalities, the more they may consider NPM-oriented changes. It could be a surprise that language area is not significant in the sense that previous researches on Swiss municipalities have shown that the NPM discourse was more widespread in the German-speaking cantons (e.g. Ladner and Steiner 2003). Since the NPM discourse stresses on the private-sector of management practice (e.g. Hood 1991), it is also quite astonishing that it is positively correlated with the percentage of members of Social Democrats within the local executive. Fourth, small Swiss municipalities which have many tasks to achieve with respect to their discretion and influence may have the propensity to consider governance reforms, such as outsourcing and developing PPP. This kind of horizontal decentralization may cost money to the municipality and therefore demands financial autonomy. Finally, changes in local democracy are positively correlated with policy autonomy and with the electoral participation 20

21 in the executive elections. Indeed, citizens have to be strongly involved in the electoral system from majoritarian to proportional. Consequently, the first hypothesis cannot be confirmed. Indeed, by declining local autonomy in seven dimensions and reforms that the municipality may have the propensity to get into six different types, we observe that results are diverse, be it regarding the independent or the dependent variables. The results related to the second hypothesis are also contrasting. Since the bigger the Swiss municipality, the higher its consideration of NPM reforms, it seems that it is rather small municipalities that will consider the governance reforms. Our results do not allow us either to confirm the relationship between cultural and dependent variables (hypothesis 3). Finally, it appears that the activity of local actors, especially Social Democrats executives, may lead to the initiative to reform regardless of the institutional setting. The fourth hypothesis is therefore partially confirmed. 6. Discussion and conclusion The aim of this contribution is to focus on local autonomy, the propensity of a local government to reform and the relations between these two concepts. We have considered local autonomy in a positive perspective, as a necessary institutional resource for local governments to be able to have the initiative to reform. Drawing on the literature, classified in a comprehensive map of local autonomy s dimensions and approaches (see Figure 1), we identified seven different dimensions covering aspects such as their legal position, their financial self-reliance, their scope in terms of tasks they are responsible for, the size of their administration, their influence when it comes to taking decisions compared to their financial involvement, their independence compared to other actors in the municipality and their influence on decisions of higher political level. Using data about the Swiss municipalities aggregated on cantonal level we are able to show that these dimensions are interrelated with each another. The correlations, however, are rather weak indicating that they cover a wider spectrum of local autonomy than a single variable could do. Nevertheless we believe that the seven dimensions provide a better picture of the real state of autonomy municipalities dispose of. It could be also interesting to make some combinations with the dimensions according to the four different identified approaches. By trying to explain the differences in local autonomy between the Swiss cantons, the cluster analysis made both at the municipal and at the aggregated cantonal level gave good results in order to get the existing Swiss local autonomy typologies up to date. We found that the most prominent explaining factors are firstly the language spoken area where the canton is located and secondly the size, thus enabling us to identify three clusters: the French and Italianspeaking cantons, the North-Central cantons and the Eastern-cantons. Regarding the reforms, we were inspired by the classification of Kersting and Vetter (2003) and made the distinction between six different types depending on of their respective legitimacy orientation. The results of statistical tests performed at the local level show that the differences between municipalities can be explained by different factors respectively. For instance, the intensification of IMC and of considering NPM reforms is higher in Germanspeaking and big municipalities although concrete plans for a merger are more present in Italian and French-speaking and small municipalities. We can thus make at this stage four 21

22 remarks. First, this legitimates the choice of drawing the different explanations on the three variants of institutionalism. Second, reforms are very different from one type to another. Third, there are no existing standards to launch reforms: on the contrary, we can assume that the reforms are initiated when several factors interact with each other. Finally, it is important not to underestimate the role of the cantons, which contribute significantly to the objectives and incentives to reform municipal operations. Using data stemming from the Local Secretaries Survey 2009 has also finally allowed us to test the correlations between local autonomy and the propensity of a local government to reform. Results are not overwhelming and allow only to partially confirming the hypothesis made. This can be a consequence of the quality of the data used, especially regarding the reforms of local democracy. Studying the changes in citizens participation is difficult since Switzerland has highly developed terms and conditions of direct democracy highly developed. Differences among cantons exist (e.g. only the cantons of GE and NE both have a mandatory local parliament) but they seem to be immutable. Furthermore, it is difficult to find a real direction: is it the local autonomy that could have a positive impact on the propensity of local governments to reform or are the reforms leading to more local autonomy? It would be also interesting to conduct further statistical tests, such as linear multivariate regressions. Furthermore, more reflection is needed when it comes to the aggregation of the different dimensions into an overall measurement of local autonomy and the local government reforms. Can the values for the different dimensions be added to a single index; do the dimensions have to be weighted? Finally, there is an additional challenge while looking at the Swiss case. Indeed, for some dimensions discussed in this contribution, for example the legal framework, there is little or no variance among the municipalities within a canton. Nevertheless, we believe that in the perspective of the study by Kersting and Vetter (2003) which compared the perception of the influence of objective input and output pressure on local government reforms across Europe, this contribution brings additional elements. First, we have not relied on qualitative descriptions of reform activities from different experts but tried to rely on data. Second, we have used this data to measure some operationalized concepts. The results show that the dimensions of autonomy do not give the same results. This speaks for an even further diversification of local autonomy. We think that Kersting and Vetter, taking into account only the two dimensions of functional responsibility and local discretion, are not sufficiently in phase with the complexity of local autonomy reality. Finally, trying to test the relations between local autonomy and local government reforms within a federalist country like Switzerland meets the same challenges as a similar endeavour for a comparison between countries. One of the advantages is the availability and the comparability of the data. For an international comparison it is doubtful whether similar data is to be found for a larger number of countries. This calls for the development of internationally applicable concepts of local autonomy and local government reforms and the ways such data can be presented and used. Regarding local autonomy we are in the process of developing a standardized codebook of eleven dimensions that will be applied to 39 European countries from 1990 to 2014 including changes 13. Documenting reforms which account for more or less autonomy will give us some pieces of information about the triggers and institutional 13 In the framework of the European Commission study on Self-rule Index for local authorities in the EU submitted by Prof. Dr. Andreas Ladner from the University of Lausanne (Switzerland). 22

23 consequences. However, we are of the opinion that a more qualitative approach is needed to understand the differences between formal and effective aspects and the fine motivations of the local actors and therefore to understand in more further detail the recent developments of the local government s level, which has to be more examined. Keywords: local governance, local government reforms, local autonomy, comparative analysis, Switzerland 23

24 References BATLEY, Richard and STOKER, Gerry (1991): Local Government in Europe. Trends and Developments. Houndmills: Macmillan. BENNETT, Robert J. (1993): Local Government in the New Europe. London and New York: Bellhaven Press. BENNETT, Robert J. (1989): Territory and Administration in Europe. London and New-York: Pinter. BOURS, Adriaan (1993): Management, tiers, size and amalgamations of local government. In: BENNETT, Robert J. (eds.). Local Government in the New Europe. London and New York: Belhaven Press, BOYNE, George A. (1998): Public Choice Theory and Local Government. A Comparative Analysis of the UK and the USA. Palgrave Macmillan. BOYNE, George A. (1995): Population Size and Economies of Scale in Local government. In: Policy and Politics, 23 (3), BOYNE, George (A.) (1993): Central Policies and Local : the Case of Wales. In: Urban Studies, 30 (1), BRAUN, Dietmar (2000): Public Policy and Federalism. Aldershot. BRUNSSON, Nils and OLSEN, Johan P. (1993): the Reforming Organization. London: Routledge. CAUFIELD, Janie and LARSEN, Helge O. (eds.) (2002): Local Government at the Millenium. Wiebaden: Springer. CDLR European Committee on Local and Regional Democracy (2007): Local competences in Europe. Prepared with the collaboration of Gérard Marcou Professor at Paris I Panthéon-Sorbonne University, Head of the Research Network of Local Government of Europe (GRALE). COUNCIL OF EUROPE (2001): Relationship Between the Size of Local and Regional Authorities and their Effectiveness and Economy of their Action. Report by the European Committee on Local and Regional Democracy (CDLR), December COUNCIL OF EUROPE (1985): European Charter of Local Self-Government. (consulted in January 2015). CLARK, Gordon L. (1984): A Theory of Local. In: Annals of the Association of American Geographers, 74 (2), DAFFLON, Bernard (2006): The Assignment of Functions to Decentralized Government: From Theory to Practice. In: AHMAD, Ehtisham. and BROSIO, Giorgio (eds). Handbook of Fiscal Federalism. Cheltenham: Edward Elgar Publishing Limited, DAHL, Robert A. and TUFTE, Edward R. (1974): Size and Democracy. Stanford California: Stanford University Press. DAHL, Robert A. (1994): A Democracy Dilemma: System Effectiveness versus Citizen Participation. In: Political Science Quarterly, 109 (1), DENTERS, Bas, GOLDSMITH, Michael, LADNER, Andreas, MOURITZEN, Poul E. and ROSE, Lawrence E. (2014): Size and Local Democracy. Edward Elgar. DENTERS, Bas and ROSE, Lawrence E. (2005): Comparing Local Governance: Trends and Developments. Basingstoke/New-York: Palgrave Macmillan. DIMAGGIO, Paul J. and POWELL, Walter W. (1991): The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields. In: American Sociological Review, 48 (2), DREIER, Volker (1994): Kommunalpolitik in Europa Einheit durch Vielheit? In: WEHLING, Hans-Georg (eds.). Kommunalpolitik in Europa. Stuttgart: Kohlhammer,

25 EASTON, David (1953): The political system: An Inquiry into the State of Political Science. New York: Alfred A. Knopf. ESPING-ANDERSEN, Gosta (1990): The three worlds of welfare capitalism. Princeton University Press. FEDERAL ASSEMBLY (1990): Bundesgesetz über Finanzhilfen und Abgeltung (Subventionsgesetz, SuG) vom 5. Oktober 1990 (Stand am 1. Januar SR FEDERAL COUNCIL (2003): Message relatif à la Charte européenne de l autonomie locale du 19 décembre 2003 (03.079). Bern. FSO Federal Statistical Office (2002): Eidgenössische Volkszählung 2000 Bevölkerungsentwicklung der Gemeinden Neuchâtel: FSO. FIECHTER, Julien (2010): Politische Gemeinden und lokale Autonomie in der Schweiz. Eine Studie zur normativen Bedeutung und empirischen Erfassung der Gemeindeautonomie und ihrer Ausprägung im kantonalen und lokalen Vergleich. In: Cahier de l IDHEAP, 251. FINER, Sammy E. (1970): Comparative Government. London: Allen Lane, Penguin. FLEURKE, Frederik and WILLEMSE, Rolf (2006): Measuring local autonomy: A decision-making approach. In: Local Government Studies, 32 (1), FLEURKE, Frederik and WILLEMSE, Rolf (2004): Approaches to Decentralization and Local : A Critical Appraisal. In: Administrative Theory & Praxis, 26 (4), GERMANN, Raimund E. (1996): Administration publique en Suisse: L appareil étatique et le gouvernement. Volume 1. Stuttgart/Vienne: Paul Haupt Berne. GESER, Hans, MEULI, Urs, HORBER-PAPAZIAN, Katia, LADNER, Andreas and STEINER, Reto (2012): Les membres des exécutifs dans les communes suisses. Glarus: Rüegger. GESER, Hans (1997): Zwischen Aufgabenzuwachs und Autonomieverlust: Neue selbstbehauptungsstrategien der Gemeinden im Zeitalter vertikaler Politikverflechtung. In: GESER, Hans: Online Publikationen: Gemeinde und Verwaltung. Zürich, GIACOMETTI, Zaccaria (1952): Die rechtliche Stellung der Gemeinden in der Schweiz. In. BRIEDEL, Marcel (eds.). Die direkte Gemeindedemokratie in der Schweiz. Zürich: Polygraphischer Verlag, GOLDSMITH, Michael and PAGE, Edward C. (2010): Changing government relations in Europe: from localism to intergovernmentalism. London: Routledge. GOLDSMITH, Michael (1995): and City Limits. In: JUDGE, David, STOCKER, Gerry and WOLMAN, Harold (eds.). Theories of Urban Politics. London: Sage, GOLDSMITH, Michael (1992). Local Government. In: Urban Studies, 29 (4), GOLDSMITH, Michael (1990): Local : Theory and Practice. In: KING, Desmond S. and PIERRE, Jon (eds). Challenges to Local Government. London: Sage, GORGES, Michael J. (2001): Blind Alley. New Institutionalist Explanations for Institutional Change: A Note of Caution. In: Politics, 21 (2), GRODECKI, Stéphane (2007): Les compétences communales comparaison intercantonale. In: TANQUEREL, Thierry and BELLANGER, François. L avenir juridique des communes. Pratique de droit administratif. Genève, Zurich, Bâle: Schulthess Médias Juridiques SA, GURR, Ted Robert and KING, Desmond S. (1987): The State and the City. London: Macmillan. HALL, Peter A. and TAYLOR, Rosemary C. R. (1996): Political Science and the Three New Institutionalism. In: Political Studies, 44 (5), HANSEN, Tore and KLAUSEN, Jan E. (2002): Between the Welfare State and Local government. In: Local government Studies, 28 (4),

26 HARISALO, Risto (2011): Kunnallinen itsehallinto: epätodellinen haave vai todellinen mahdollisuus? In: HAVERI, Arto, STENVALL, Jari and MAJOINEN, K.aija (eds). Kunnallisen itsehallinnon peruskivet (Cornerstones of Municipal Self- Government). Acta 224. Helsinki: The Association of Finnish Local and Regional Authorities, HEINDRIKS, Frank and PIETER, Tops (1999): Between democracy and efficiency: Trends in local government reform in the Netherlands and Germany. In: Public Administration, 77 (1), HESSE, Joachim J. and SHARPE, Laurence J. (1991): Local Government in International Perspective: Some Comparative Observations. In: HESSE, Joachim J. (eds.): Local government and Urban Affairs in International Perspective. Baden-Baden: Nomos, HOOD, Christopher (1991): A public management for all seasons? In: Public Administration, 19, HOOGHE, Liesbet, MARKS, Gary and SCHAKEL, Arjan H. (2010): The Rise of Regional Authority: A Comparative Study of 42 Democracies ( ). London: Routledge. HOOGHE, Liesbet and MARKS, Gary (2003): Unraveling the Central State, but How? Types of Multi-Level Governance. In: American Political Science Review, 97 (2), HORBER-PAPAZIAN, Katia and JACOT-DESCOMBES, Caroline (2014): Communes. In: KNOEPFEL, Peter, PAPADOPOULOS, Yannis, SCIARINI, Pascal, VATTER, Adrian, HAUSERMANN Silja (eds). Handbuch der Schweizer Politik, HORBER-PAPAZIAN, Katia (2006): Les communes. In: KLÖTI et al. (eds.). Manuel de la politique Suisse. Zürich: NZZ Verlag, HUBER, Evelyn and STEPHENS, John (2001): Development and Crisis of the Welfare State. Parties and Policies in Global Markets. University of Chicago. HUMES, Samuel and MARTIN, Eileen H. (1969): The Structure of Local Government: a comparative survey of 81 countries. The Hague. JOHN, Peter (2001): Local Governance in Western Europe. London: Sage. KERSTING, Norbert and VETTER, Angelika (eds.) (2003): Reforming Local Government in Europe: Closing the Gap between Democracy and Efficiency. Opladen: Leske + Budrich. KING, Desmond S. and PIERRE, Jon (1990): Challenges to Local Government. London: Sage. KING, David N. and MA, Yue (2000): Local authority size in theory and practice. In: Environment and Planning C: Government and Policy 18, KLÖTI, Ulrich (2000): Regieren im verflochtenen dreistufigen Föderalismus. In: KNOEPFEL Peter and LINDER, Wolf (eds.). Verwaltung, Regierung und Verfassung im Wandel. Gedächtnisschrift für Raimund Germann. Basel: Helbling & Lichtenhahn, KRIESI, Hans-Peter (1999): State formation and nation building in the Swiss case. In: KRIESI, Hans-Peter (eds.). Nation and national identity: the European experience in perspective. Chur: Verlag Rüegger, KÜBLER, Daniel and LADNER, Andreas (2003): Local government reform in Switzerland. More for than by but what about of? In: KERSTING, Norbert and VETTER, Angelika (eds.). Reforming Local Government in Europe: Closing the Gap between Democracy and Efficiency. Opladen: Leske + Budrich, KUHLMANN, Sabine and WOLLMANN, Helmut (2014): Introduction to comparative public administration. Administrative Systems and Reforms in Europe. Edward Elgar. KUHLMANN, Sabine (2010): New Public management for the classical continental European administration: modernization at the local level in Germany, France and Italy. In: Public Administration, 88 (4), LADNER, Andreas and KEUFFER, Nicolas (2014): Local autonomy, a multidimensional concept: How to measure local autonomy, how to explain different degrees of local autonomy and what about its effects? Paper presented at the 2014 EGPA Annual Conference in Speyer (Germany), September

27 LADNER, Andreas, STEINER, Reto, HORBER-PAPAZIAN, Katia, FIECHTER, Julien, JACOT- DESCOMBES, Caroline and KAISER, Claire (2013): Gemeindemonitoring 2009/2010. Bericht zur fünften gesamtschweizerischen Gemeindeschreiberbefragung. KPM-Schriftenreihe Nr. 48. LADNER, Andreas and FIECHTER, Julien (2012): The Influence of Direct Democracy on Political Interest, Electoral Turnout and Other Forms of Citizens Participation in Swiss Municipalities. In: Local Government Studies, 38 (4), LADNER, Andreas and STEINER, Reto (2005): Reforming the Swiss Municipalities: Efficiency or Democracy? In: KRIESI, Hanspeter, FAREGO, Peter, KOHLI, Martin, ZARIN-NEJADAN, Milad (eds.). Contemporary Switzerland: Revisiting the Special Case. Houndmills: Palgrave Macmillan, LADNER, Andreas and STEINER, Reto (2003): Die Schweizer Gemeinden im Wandel: Konvergenz oder Divergenz? In: Swiss Political Science Review, 9 (1), LADNER, Andreas (2011): Die wichtigsten institutionellen Reformen zwischen 1970 und 2010 in Bund, Kantonen und Gemeinden: Welche Rolle spielt die konjunkturelle Lage? In: SOGEL, Nils C. (eds.). Des politiques au chevet de la conjoncture. Lausanne: Presses polytechniques et universitaires romandes. LADNER, Andreas (2009): Auslöser, Ziele und Auswirkungen von kommunalen Reformprojekten. In: KNOEPFEL, Peter (eds.). Réformes de politiques institutionnelles et action publique. Bern: Haupt, LADNER, Andreas (2005): Switzerland: Reforming small autonomous municipalities. In: DENTERS, Bas and ROSE, Lawrence E. (eds.). Comparing Local Governance: Trends and Developments. Basingstoke/New-York: Palgrave Macmillan, LADNER, Andreas (2005a): NPM und die Gemeinden. In. LIENHARD, Andreas, RITZ, Adrian, STEINER, Reto and LADNER, Andreas (eds.). 10 Jahre New Public Management in der Schweiz. Bern: Haupt. LADNER, Andreas (2002): Size and direct democracy at the local level: the case of Switzerland. In: Environment and Planning C: Government and Policy, 20, LADNER, Andreas (2001): Collaborations et fusions dans les communes suisses. In: HORBER-PAPAZIAN, Katia (eds.). L espace local en mutation. Lausanne: presses polytechniques universitaires romandes, LADNER, Andreas, ARN, Daniel, FRIEDRICH, Ueli, STEINER, Reto and WICHTERMANN, Jürg (2000): Gemeindereformen zwischen Handlungsfähigkeit und Legitimation. Bern: Institut für Politikwissenschaft und Institut für Organisation und Personal. LADNER, Andreas (1996): Majorz oder Proporz: Die Auswirkungen des Wahlverfahrend auf die politische Partizipation in den Berner Gemeinden. In: Swiss Political Science Review, 2 (4), LADNER, Andreas (1994): Finanzkompetenz in der Gemeinde - Ein Überblick über die Praxis. In: Finanzen der öffentlichen Hand Nr. 5 "Finanzföderalismus". Bern: Emissionszentrale der Schweizer Gemeinden, LANE, Jan-Erik and ERSSON, Svante O. (1987): Politics and Society in Western Europe. London: Sage. LIDSTRÖM, Anders (2003): Kommunsystem i Europa [Local government systems in Europe]. Malmö: Liber. LIDSTRÖM, Anders (1998): The comparative study of local government systems a research agenda: In: Journal of Comparative Policy Analysis: Research and Practice, 1 (1), LINDER, Wolf (1999): Schweizerische Demokratie Institutionen, Prozesse, Perspektiven. Bern, Stuttgart, Vienna: Hautp. LIPHART, Arend (1999): Patterns of Democracy. Government Forms and Performance in Thirty-Six Countries, New Haven: Yale University Press. LIPHART, Arend (1984): Democracies. Patterns of Majoritarian and Consensus Government in Twenty- One Countries, New Haven: Yale University Press. MARCH, James and OLSEN, Johan (1984): The New Institutionalism: Organizational Factors in the Political Life. In: Amercian Political Science Review, 78 (3),

28 MAYNTZ, Renate and SCHARPF, Fritz W. (1995): Der Ansatz des akteurszentrierten Institutionalismus. In: MAYNTZ, Renate and SCHARPF, Fritz W. (eds.): Gesellschaftliche Selbstregelung und politische Steuerung. Frankfurt am Main and New-York: Campus, MECKSTROTH, Theodore W. (1975): I. Most Different Systems and Most Similar Systems : A Study in the Logic of Comparative Inquiry. In: Comparative Political Studies, 8, MEYLAN, Jean, GOTTRAUX, Martial and DAHINDEN, Philippe (1972): Communes suisses et autonomie communale. Lausanne. MILL, John S. (1991): Considerations on Representative Government [1861]. Oxford: Oxford University Press. MOURITZEN, Poul E. (1989): City size and citizens satisfaction: two competing theories revisited. In: European Journal of Political Research, 17 (6), MUELLER, Sean (2011): The Politics of Local : Measuring Cantonal (De)centralisation in Switzerland. In: Space and Polity, 15 (3), NORTON, Alan (1994): International Handbook of Local and Regional Government: A Comparative Analysis of Advanced Democracies. Aldershot: Edward Elgar. NORTON, Alan (1991): West European Government in Comparative Perspective. In: BATLEY, Richard and STOCKER, Gerry (eds.). Local Government in Europe. Trends and Developments. Houndmills: Macmillan. OECD Organisation for Economic Co-operation and Development (2012): Fiscal Federalism Studies Measuring Fiscal Decentralisation Concepts and Policies. Korea Institute of Public Finance. OSTROM, Elinor (1991): Rational Choice Theory and Institutional Analysis: Toward Complementarity, American Political Science Review, 85, OSTROM, Elinor (1990): Governing the commons: The evolution of institutions for collective action. Cambridge: Cambridge University Press. PAGE, Edward (1991): Localism and Centralism in Europe. Oxford: Oxford University Press. PAGE, Edward and GOLDSMITH, Michael (1987): Central and Local government Relations. London: Sage. PALONEN, Kari (2003): Four Times of Politics: Policy, Polity, Politicking, and Politicization. In: Alternatives: Global, Local, Political, 28 (2), PETERS, Guy B., PIERRE, Jon and KING, Desmond S. (2005): the Politics of Path Dependency: Political Conflict in Historical Institutionalism. In: the Journal of Politics, 67 (4), PETERS, Guy B. (1999): Institutional Theory in Political Science. The New Institutionalism. London and New-York: Continuum. PETERS, Guy B. (1995): Political institutions: old and new. In: GOODIN, Robert and KLINGEMANN, Hans- Dieter (eds.). A New Handbook of Political Science. Oxford: Oxford University Press. PIERRE, Jon (1990): Assessing Local. In: KING, Desmond S. King and PIERRE, Jon (eds.). Challenges to Local Government. London: Sage, POLLITT, Christopher (2002): Clarifying Convergence: Striking Similarities and Durable Differences in Public Management Reform. In: Public Management Review, 4 (1), PRATCHETT, Lawrence (2004): Local, Local Democracy and the New Localism. In: Political Studies, 52, PRZEWORSKI, Adam and TEUNE, Henry (1970): the Logic of Comparative Social Inquiry. New York. RHODES, Rod A.W. (1996): The new Governance: Governing without Government. In: Political Studies, XLIV,

29 ROLLA, Giancarlo (1988): : a guiding criterion for decentralizing public administration. In: International Review of Administrative Sciences, 64, RÜHLI, Lukas (2013): Irrgarten Finanzausgleich. Kantonsmonitoring 5: Wege zu mehr Effizienz bei der interkommunalen Solidarität. Zürich: Avenir Suisse. RÜHLI, Lukas (2012): Gemeindeautonomie zwischen Illusion und Realität. Kantonsmonitoring 4: Gemeindestrukturen und Gemeindestrukturpolitik der Kantone. Zürich: Avenir Suisse. SABATIER, Paul A. (1988): An Advocacy Coalition Framework of Policy Change and the Role of Policy- Oriented Learning Therein. In: Policy Sciences, 21 (3), SCHARPF, Fritz W. (1999): Governing in Europe: effective and democratic? Oxford University Press, Oxford. SCHARPF, Fritz W. (1978): Die Theorie der Politikverflechtung: ein kurzgefasster Leitfaden, In: HESSE, Joachim J. (eds.). Politikverflechtung im föderativen Staat. Studien zum Planungs- und Finanzierungsverbund zwischen Bund, Ländern und Gemeinden, Baden-Baden, Germany: Nomos. SCHARPF, Fritz W. (1977): Public organization and the waning of the welfare state: a research perspective. In: European Journal of Political Research, 5 (4), SCHARPF, Fritz W. (1970): Demokratietheorie zwischen Utopie und Anpassung. Konstanz: Universitätsverlag. SCHMIDT, Manfred G. (1996): When parties matter: A review of the possibilities and limits of partisan influence on public policy, European Journal of Political Research, 30 (2), SELLERS, Jeffrey and LIDSTRÖM, Anders (2007): Decentralization, Local Government, and the Welfare State. In: Governance: An International Journal of Policy, Administration and Institutions, 20 (4), SCHARPE, Lawrence J. (1970): Theories and Values of Local Government. In: Political Studies, 18, SELLERS, Jeffrey and LIDSTRÖM, Anders (2007). Decentralization, Local Government, and the Welfare State. In: Governance: An International Journal of Policy, Administration and Institutions, 20 (4), SELLERS, Jeffrey (2006): Comparing Local Governance in Developed Democracies. Selected Indicators. Online Working Paper. SHARPE, Lawrence J. (1988): The growth and decentralisation of the modern democratic state. In: European Journal of Political Research, 16 (3), SHARPE, Lawrence J. (1970): Theories and Values of Local Government. In: Political Studies, 18, STEINER, Reto and KAISER, Claire (2013). Administration communale. In LADNER et al. (eds). Manuel d administration publique Suisse. Lausanne : Presses polytechniques et universitaires romandes, STEINER, Reto (2003): The Causes, Spread and Effects of Intermunicipal Cooperation and Municipal Mergers in Switzerland. In: Public management Review, 5 (4), STEINER, Reto (2002): Interkommunale Zusammenarbeit und Gemeindezusammenschlüsse in der Schweiz. Bern/Stuttgart/Wien: Haupt. STEINER, Reto (2000): New Public Management in Swiss Municipalities. In: International Public Management Journal, 3, STEPHENS, Ross (1974): State centralization and the Erosion of Local. In: the Journal of Politics, 36 (1), STOCKER, Gerry (2010): The comparative study of local governance: towards a global approach. Symposium on Chinese Social Capacity Building in International and Comparative Perspective, Zhejiang University, Hangzhou, China. STOCKER, Gerry (2006): Comparative local governance. In, RHODES, Rod A.W., BINDER, Sarah A. and BROCKMAN, Bert A. (eds.). The Oxford Handbook of Political Institutions. Oxford, UK, Oxford University Press,

30 SWIANIEWICZ, Pawel (2014): An Empirical Typology of Local government Systems in Eastern Europe. In: Local Government Studies, 40 (2), THELEN, Kathleen (1999): Historical Institutionalism in Comparative Politics. In: Annual Review of Political Science, 2, TOCQUEVILLE, Alexis de (1968): Democracy in America [1835]. London: Fontana. UCLG - United Cities and Local governments (2008): Decentralization and local democracy in the world. First global Report by United Cities and Local Governments. U.S.A. VAKKALA, Hanna and LEINONEN, Jaana (2014): Local autonomy and municipal management: institutional stability or independent local governance? Paper presented at the 2014 EGPA Annual Conference in Speyer (Germany), September VAN KERSBERGEN, Kees and VAN WAARDEN, Frans (2004): Governance as a bridge between disciplines. EJPR, 43 (2), VERHOEST, Koen, PETERS, Guy B., BOUCKAERT, Geert and VERSCHUERE, Bram (2004): The study of organisational autonomy: a conceptual review. In: Public Administration and Development, 24 (2), VETTER, Angelika (2007): Local Politics: a resource for democracy in Western Europe? Local autonomy, local integrative capacity, and citizens attitudes towards politics. Lexington Books. WAGSCHAL, Uwe (1999): Schranken staatlicher Steuerungspolitik: Warum Steuerreformen scheitern können. In: BUSCH, Andreas and PLÜMPER, Thomas (eds.). Nationaler Staat und international Wirtschaft. Baden- Baden: Nomos, WALKER, Jack L. (1969): The Diffusion of Innovations among the American States. In: American Political Science Review, 63, WOLLMANN, Helmut (2004): Local Government Reforms in Great Britain, Sweden, Germany and France: Between Multi-Function and Single Purpose Organisations. In: Local government studies, 20 (4), WOLMAN, Harold (2008): Comparing local government systems across countries: conceptual and methodological challenges to building a field of comparative local government studies. In: Environment and Planning C: Government and Policy 26, WOLMAN, Hal, MC MANMON, Robert, BELL, Michael and BRUNORI, David (2008): Comparing Local government Across States. GWIPP Working paper. WOLMAN, Harold and GOLDSMITH, Michael (1990): Local autonomy as a Meaningful Analytic Concept: comparing Local Government in the United States and the United Kingdom. In: Urban Affairs Review, 26 (1),

31 Appendix Methodology Most of the data we use (for measuring social autonomy, vertical influence, perceived autonomy, interest in local politics, turnout at local elections and debt increase) stem from the 5th National Survey of Local Secretaries conducted between October 2009 and February There are several good reasons for our choice of the local secretaries as the target group for the questionnaire: First, because of their position at the intersection between the political and the administrative sphere. The local secretaries lead the administration and at the same time take part in the meetings of the local executives. Therefore they possess both administrative and political knowledge. Second, the function of the local secretary can be found in all municipalities. Third, unlike the members of the local executive, the secretaries are usually not affiliated to a political party. Hence, we can expect them to have a more objective (administrative) view on the functioning of their municipality than the members of the executive would have. Fourth, the large amount of small municipalities makes it a reasonable choice to target the local secretaries since even in the smallest municipalities the position of the secretary exists. Moreover, in the political and administrative institutions of small municipalities, the local secretary often is the only person fulfilling his functions on a part- or fulltime employment basis. Fifth, the local secretaries had proven in the past to be a reliable target group for local surveys. The high response rates and the quality and reliability of the collected data in the previous surveys of 1988, 1994, 1998 and 2005 have shown that the choice of the local secretaries as target group is as good as it gets. The questionnaire was sent to the local secretaries of all 2,596 Swiss municipalities and 1,497 have responded to all (or most of) the questions (57.7 per cent). There is some variance in the response rate when looking at the linguistic regions: German speaking municipalities participated more often in the study (60.4%) than the French speaking (55.3%) and especially the Italian speaking (44.5%) municipalities. Further data analysis shows that the response rates are also on comparable levels between the municipalities of different cantons and of different socioeconomic categories. The high number and the equal distribution of respondents lead to the assumption that our results are not only valid for the participating municipalities but also to a high degree for all Swiss municipalities. 31

32 Approaches configurations (own classification), based on the comprehensive map s essay of local autonomy s dimensions and approaches (Figure 1) A-B-C-D: Hooghe et al. (2010), Fleurke and Willemse (2004); A-B-C: Stephens (1974), Verhoest et al. (2004); A-B-D: Goldsmith (1995), Page and Goldsmith (1987), Goldsmith and Page (2010), Hesse and Sharpe (1991); A-C-D: Wolman et al. (2008); A-B: Norton (1991, 1994), Clark (1984); A-C: Pierre (1990), Vetter (2007), Harisalo (2011); A-D: Braun (2000); A: OECD (2012), Grodecki (2007); B-C: Rolla (1998), Sellers (2006), Sellers and Lidström (2007); B-D: Page (1991); B: Liphart (1999, 1984), Sharpe (1988); C-D: Gurr and King (1990), King and Pierre (1990), Stoker (2010, 2006); C: Fleurke and Willemse (2006), Denters and Rose (2005); D: Pratchett (2004), Boyne (1993), Mueller (2011), John (2001). 32

33 Statistical results Table 3: Different Dimensions of Empirical results (CH cantonal means) Legal Financial Policy Administr ative Decisional Social Vertical Influence ZH BE LU UR SZ W NW GL ZG FR SO BS BL SH AR AI SG GR AG TG TI VD VS NE GE JU CH (mean) All variables are transformed to a scale reaching from 0 (lowest autonomy) to 100 (highest autonomy). Total (mean) Sources: Legal autonomy: Fiechter (2010: 90); Financial : Rühli (2012: 154); Policy : Fiechter (2010: 90); Administrative : Fiechter (2010: 90); Decisional : Rühli (2013: 164f.); Social : Local Secretary Survey 2009; Vertical Influence: Local Secretary Survey

34 Figure 1: Cantonal local autonomy profiles (high: AR, LU and low: GE, FR) Legal Fiscal/Financial Appenzell A. Rh. Vertical Influence Policy (Scope) Political/Decisi onal CH (mean) AR Legal Fiscal/Financial Vertical Influence Genève Policy (Scope) Political/Decisi onal CH (mean) GE Administrative Social Administrative Social Legal Fiscal/Financial Vertical Influence Luzern Policy (Scope) Political/Decisi onal CH (mean) LU Legal Fiscal/Financial Vertical Influence Fribourg Policy (Scope) Political/Decisi onal CH (mean) FR Administrative Social Administrative Social 34

35 GE JU VD NE TI FR SO BL BE ZH Total SH AG SZ SG GL NW LU UR VS GR AI AR TG OW BS ZG Figure 2: Self-perception of the local autonomy according to the canton Source: question 18a; 1 = no autonomy; 10 = high autonomy. N = Table 4: Different Dimensions of Local : Correlation Matrix Legal autono my Financi al autono my Policy autono my Admini strative autono my Decisio nal autono my Social autono my Vertical influenc e Total autono my Perceiv ed autono my Legal autonomy 1.403* ** * ** Financial.403* * autonomy Policy autonomy * * Administrative.487** * ** **.621** autonomy Decisional * autonomy Social autonomy.449* ** **.639**.618** Vertical influence *.766** 1.419*.561** Total autonomy.784**.367*.378*.797** **.419* 1.509** Perceived ** **.561**.509** 1 autonomy N= * significant at 0,05 (one-tailed); ** significant at 0,01 (one-tailed); without BS. Source: cf. Table 3, in appendix. 35

36 Figure 3: Self-perception of the local autonomy according to the language area French-speaking part Italian-speaking part German-speaking part Total Source: question 18a; 1 = no autonomy; 10 = high autonomy. N = =1401. Figure 4: Self-perception of the local autonomy according to the population size (8 categories) Source: question 18a; 1 = no autonomy; 10 = high autonomy. N = =

37 Figure 5: Dendrogram using Ward Linkage (Rescaled Distance Cluster Combine) for cantonal aggregated local autonomy Figure 6: Dendrogram using Ward Linkage (Rescaled Distance Cluster Combine) for secretaries self-perception of local autonomy 37