Eran Razin a & Anna Hazan b a Department of Geography, The Hebrew University, Jerusalem, Israel b Department of Public Policy and Administration,

Transcription

1 This article was downloaded by: [Hebrew University] On: 27 March 2014, At: 13:55 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: Registered office: Mortimer House, Mortimer Street, London W1T 3JH, UK Local Government Studies Publication details, including instructions for authors and subscription information: Attitudes of European Local Councillors towards Local Governance Reforms: A North South Divide? Eran Razin a & Anna Hazan b a Department of Geography, The Hebrew University, Jerusalem, Israel b Department of Public Policy and Administration, Sapir College, Israel Published online: 06 Feb To cite this article: Eran Razin & Anna Hazan (2014) Attitudes of European Local Councillors towards Local Governance Reforms: A North South Divide?, Local Government Studies, 40:2, , DOI: / To link to this article: PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the Content ) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.

2 This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sublicensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at

3 Local Government Studies, 2014 Vol.40,No.2, , Attitudes of European Local Councillors towards Local Governance Reforms: A North South Divide? ERAN RAZIN * & ANNA HAZAN ** *Department of Geography, The Hebrew University, Jerusalem, Israel, ** Department of Public Policy and Administration, Sapir College, Israel ABSTRACT The impact of governance attributes on frequently overlooked attitudes of local councillors towards local governance reform agendas is assessed, based on the MAELG survey of approximately 11,000 councillors in 15 European countries and Israel. Eighteen policy statements are grouped into six reform agendas, followed by analyses of their explanatory factors. Europe s north south divide is found meaningful for the understanding of variations in attitudes towards reform, being most profound in notions of democracy, participation and devolution. Southern Europe, including postcommunist Europe, shows clear support for new democratic mechanisms, whereas northern European councillors reveal more reserved, perhaps weary, attitudes towards reforms. Councillors may provide the correct statements in the south, because less has been reformed there so far, knowing that change would anyway be difficult. Substantial variations between different northern sub-groups could be linked to the memory of nondemocratic regimes in Germany and Austria and to the influence of existing systems, deeply embedded in local political cultures. Nevertheless, the impact of present institutional practices is far from universal. Councillors in liberal regimes are not proprivatisation, and cautious attitudes towards reform are shared by northern decentralised and centralised countries. KEY WORDS: Councillors, north south divide, local government reform, local government typologies, Europe Introduction Debates over local government reforms are almost perpetual (AAAPS 1895, Jones 1942, Razin and Smith 2006, Kersting and Vetter 2003), involving mayors, upper level/central state politicians and bureaucrats, as well as the Correspondence Address: Professor Eran Razin, Department of Geography, The Hebrew University, Jerusalem 91905, Israel. msrazin@mscc.huji.ac.il 2013 Taylor & Francis

4 European Local Councillors and Local Governance Reforms 265 media and diverse interest groups. The voice of local councillors, however, seems to be rarely heard in these debates. Major reform discourses concern intergovernmental relations (division of tasks, political autonomy and local finance) and territorial-structural reconfigurations (amalgamations, cooperation mechanisms, etc.). However, reform agendas that more directly concern local councillors are changes in democratic mechanisms and in the local government power triangle: mayor, elected council and senior bureaucracy (Marcou and Wollman 2009). The prime objective of our article is to provide comparative insights on councillors views on reform agendas, based on the MAELG (Municipal Assemblies in European Local Governance) survey conducted in 15 European countries and in Israel. Attitudes towards reform agendas are identified through levels of agreement with various statements, mainly concerning new public management (NPM)/privatisation issues, local democracy, formal/informal democratic practices and explicit attitudes towards reform, focusing mainly on democratic traits of local governance. Our main argument is that the national context has a prime influence on attitudes of councillors towards local government reform. Cross-national differences could be attributed to path dependent political traditions and local government systems, such as the major role of referenda in Swiss politics, and centralism, reflected by the appointment of mayors (Burgomaster) and the high dependency of local government on central grants in the Netherlands. A fundamental issue in predicting attitudes of councillors is whether these attitudes reflect a tilt towards existing practices in their countries, or to the contrary, antagonism towards present realities and a desire for change. Such a desire is assumed to be more common among councillors than among other actors in the local government power triangle mayors and bureaucrats because councillors have lower stakes in the existing system and particularly those in opposition could be expected to more forcefully argue for change. An assessment of attitudes of European mayors towards reform (Egner and Heinelt 2006) revealed that institutions do matter: attitudes of mayors were influenced by the institutional arrangements in which they operate and by their particular perceptions towards their task, both related to a classification of European countries into local political leadership types (Heinelt and Hlepas 2006). Our aim is to identify the impact of broad contextual attributes on cross-national variations in attitudes towards local governance reform. We argue that types of local governance systems, defined by central local government relations and welfare state regimes, as well as levels of economic development and even the simplistic north south distinction, influence these attitudes of councillors. Councillors in centralised systems could either endorse centralised practices or favour decentralisation and new democratic practices. Councillors in more liberal welfare states could either be supportive or antagonistic towards privatisation and NPM agendas. Councillors in

5 266 E. Razin and A. Hazan countries that passed radical political and economic transitions in recent decades (post-communist Europe) could also be distinct in their support or rejection of further reforms, democratisation and implementation of NPM practices. Level of economic development and experiences with nondemocratic regimes in recent history could influence attitudes, the former particularly towards NPM reforms and the latter towards the introduction of various democratic practices. Thus, the national governance context is expected to markedly influence attitudes of councillors, although it is difficult to predict the direction of these influences a priori. The effect of local governance types could be associated with a broader European north south divide. A division between north-western Europe and Southern Europe has rather predominated the narrative on the construction of boundaries in Europe, a further distinction being between well-established north-western Europe and the transitional east (Eder 2006). Using the north south divide as an explanatory factor could be simplistic, perhaps anachronistic. However, it may still be relevant, particularly in attitude surveys, in which expressed perceptions blur detailed specificities of each country. The north south divide is expected to reflect gaps in levels of economic development, but also in political values and even in behavioural aspects associated with practices of responding to questionnaires. The north is represented in our case by the well-established north-western European countries and the south by central-eastern post-communist Europe and the Mediterranean region. The article begins with a discussion of reform agendas and the role of local councillors in these agendas, followed by classifications of governance systems that could be related to councillors attitudes towards reform. We classify the 16 countries into eight sub-types a detailed classification that may be further reduced into fewer groupings based on the empirical results. The empirical analysis begins with the examination of responses of councillors to 18 statements that imply attitudes towards reform agendas and to directly proposed reforms. The statements are then grouped into three reform agendas, the third one further divided into three sub-agendas. These form six dependent variables of multivariate regression analyses that examine whether the country typology remains a significant explanatory variable when personal attributes and intra-national contextual attributes are taken into account. The MAELG survey was carried out in 16 countries between 2007 and It included approximately 11,000 councillors from about 800 municipalities with more than 10,000 inhabitants a large sample that enabled all forms of multivariate analysis. Results proved very stable, unaffected by a substantial number of cases that were either added or deleted in various versions of the dataset. 1 It should be acknowledged that identifying the impact of contextual attributes could prove difficult, because lengthy opinion closed-format mail questionnaires may produce similar correct responses by councillors who practically have rather diverse

6 European Local Councillors and Local Governance Reforms 267 attitudes. Nevertheless, a careful evaluation of the large set of questions that compose the dependent variables is expected to reveal meaningful results for understanding the role of major, but understudied, actors in local government reform discourses. Reform agendas and councillors Local government reforms refer to several interrelated agendas: 1. Performance: improved efficiency, effectiveness of service provision (including planning and economic development) and good government. Such have been, for example, the motives for the 1998 mega-city reform in Toronto (Sancton 2010) and the 1998 amalgamation reform in Greece (Hlepas 2003); the former aimed at saving money and the latter at effectiveness and good government. 2. Social justice: reduced intra-local and inter-local disparities. Such has been the 2000 reform in South Africa that followed the work of the municipal demarcation board (Cameron 2004). 3. Democratic qualities/enhanced citizen participation, substantially influencing, for example, reform discourses in post-communist Europe in the 1990s (Baldersheim et al. 2002). Whereas much of the reform discourse in the 1990s and 2000s concerned competitiveness in the global economy and explicit or implicit neoliberal visions, maintenance of democratic legitimacy has also been regarded as a prime objective for reform (Kersting and Vetter 2003). The above mentioned reform agendas can be achieved through action at several spheres (Razin 2004): 1. Territorial/structural: consolidations, establishment and abolition of upper-tier governments, special-purpose governments, joint municipal organisations and alike. 2. Decentralisation/centralisation: modifying central local government relations in the functional (division of tasks), political (local autonomy) and fiscal (local government finance) spheres. 3. Democratic: electoral methods, modes of participation, and the balance of power in the triangle: mayor councillors senior bureaucratic gatekeepers (Zinger and Dana 2009). 4. Administrative: internal organisational changes within the local authority, including implementation of NPM practices (Wollmann 2003), outsourcing, public private partnerships, privatisation and alike. Reforms have an effect on the status of councillors particularly when opting for more personalised electoral systems, offering councillors new management tools, or conversely introducing more direct modes of citizen

7 268 E. Razin and A. Hazan participation that lessen the significance of representative democracy (Kersting and Vetter 2003). The emphasis given to new modes of active democracy versus the minimal discussion of the status of councillors (for example: Gabriel 2000, Burns 2000) demonstrates the marginality of councillors in reform discourses. For example, the suggested response to the failure of traditional representative institutions has been an emphasis on urban leadership and community involvement (Haus and Heinelt 2005). Councillors are frequently mentioned as posing problems, because of excessive fractionalisation of local party systems, rather than being a fundamental pillar of local decision-making. A few studies did assess the impact of reforms on local councillors and the attitudes of councillors towards reform. A Danish reform opting a new leadership role for elected councillors, as an alternative to market-oriented NPM, did not lead to unequivocal outcomes (Hansen 2001). Councillors also assessed that the 2007 amalgamation reform in Denmark led to a decrease in their influence vis-a` -vis top bureaucrats, except for those regarded as leading councillors (Kjaer et al. 2010). A rare older English survey of councillors focused on the tension between their private lives and municipal function (Barron et al. 1991). A study based on the MAELG data, on the effect of a Dutch reform targeted at empowering the council, did identify an impact of enhancing accountability activities of councillors (de Groot 2009). Issues included in the MAELG questionnaire refer to the democratic and administrative spheres of reform: separation between politicians and bureaucrats; decentralisation within local government, such as to neighbourhood organisations; local referenda; opinion surveys; consultation forums; user boards; attitudes towards participatory democracy; direct election of mayor, reducing the number of councillors, and NPM/liberal agendas of outsourcing, privatisation and competition. The lack of questions on the territorial/structural and decentralisation spheres can be attributed to the less direct implications of these on the status and action of councillors. Typologies of local government systems Although the conceptual and methodological challenge of comparing local government systems (Wolman 2008) is beyond the scope of our study, we do not simply employ one of the existing classifications in the European literature (Heinelt and Hlepas 2006). Rather, we opt for a detailed classification that focuses on attributes of the local government system and the welfare state regime, without defining a few types that are preassumed to influence councillors attitudes. Local government systems can be defined according to several centralisation/decentralisation scales (Razin 2000, 2004): territorial (consolidation versus fragmentation), functional (who does what?), political (local autonomy) and fiscal (local government

8 European Local Councillors and Local Governance Reforms 269 finance). Four extreme models, based on integrated centralisation-decentralisation and modernity-traditionalism scales, have been proposed: 1. A functionally decentralised comprehensive Western welfare state, either fiscally and politically centralised (the Netherlands), or decentralised (Scandinavian countries). 2. A politically and fiscally decentralised self-government model (the USA). 3. Traditional centralism (Arab countries). 4. A traditional decentralised system (Ghana). A later classification (Razin 2007) retains the centralisation-decentralisation scale, but instead of the simplistic modernity-traditionalism scale, it incorporates the broader scale of welfare state regimes. The USA serves as a prototype of a decentralised liberal regime, whereas the Netherlands is at the other end, representing a centralised comprehensive welfare state. Scandinavian countries, and to a lesser extent Germany, represent decentralised comprehensive welfare states whereas the UK is perhaps a close example of a centralised liberal regime (Razin 2007). Sellers and Lidstrom (2007) further emphasise the welfare state regime context, in their typology of local government systems. Their typology is based on two dimensions local capacities (functional decentralisation) and top-down supervision (political autonomy), defining four extreme types, from autonomous local government, characterised by high capacity and low supervision, to state monopoly, characterised by low capacity and strict topdown supervision. This typology is linked to the welfare state regime context: six types (Huber and Stephens 2001) that can be roughly reduced for most purposes into three: 1. A liberal welfare state regime, characterised by limited public provisions, mainly targeted to provide a minimal safety net for the poor (the USA and to an extent also Canada, the UK and Ireland). 2. A Christian-Democratic/conservative welfare state regime, characterised by extensive welfare mechanisms, but with no imperative for egalitarian provision and outcomes. Such regimes emerge in middle European and Napoleonic contexts, the latter being more administratively centralised and politically decentralised, with a flavour of client patron local relations. Southern Europe is a variant of the conservative welfare state regime, characterised by Napoleonic, client patron orientations in a context of somewhat lower levels of well-being. 3. A social-democratic/egalitarian comprehensive welfare state regime (Scandinavia). An additional developmental/informal regime, defined by Sellers and Lidstrom (2007), largely reflects lower levels of economic development,

9 270 E. Razin and A. Hazan rather than a unique political-ideological regime. Linking these regimes to the local government typology, Sellers and Lidstrom emphasise that decentralisation/empowerment of local government, associated with territorial centralisation, is a precondition for the egalitarian welfare state. The egalitarian/social democratic regime is characterised by high local capacity and moderate supervision, whereas the liberal regime is characterised by low supervision (with the exception of the UK and Ireland) and medium to low local capacity. Such combinations of various dimensions of centralisation-decentralisation, welfare state regimes and levels of economic development influence reform agendas and attitudes of councillors towards these agendas, together with more specific path dependent attributes of the system; that is, historically specific explanations for the course of institution-building and the evolution of civic culture (Wollmann 2000, Almond and Verba 1963). Some typologies of local government systems indeed emphasise a more path dependent approach, such as the Hesse and Sharpe (1991) typology, distinguishing Franco, Anglo and North-Middle European groups. Another typology that is perhaps more directly concerned with the status of councillors, relates path dependent attributes of political leadership, horizontal power relations (within the local authority) and vertical local central relations (Heinelt and Hlepas 2006). In strong mayor forms, mayors act as executive mayors, that is, they serve as chief executive officers (CEOs), except for in Franco type systems, in which the mayors are political mayors that primarily represent the communities. In weak mayor forms (committee leader form and collective form), mayors are termed collegial leaders, whereas in council manager forms, mayors are termed as ceremonial. Table 1 indicates relatively homogenous levels of economic development among the MAELG survey countries. All except the three post-communist countries and Israel are highly developed, and even among the four transitional economies two Israel and the Czech Republic are approaching the developed economies in terms of gross domestic product (GDP) per capita. Norway strikes out with a particularly high GDP per capita, but it is unclear whether this affluence (and lack of EU membership) differentiates attitudes of councillors in Norway from those of neighbouring countries. Western and Southern Europe include a variety of unitary, federal and intermediate government structures, expected to mainly influence intra-country variations, whereas most relevant for inter-country variations could be the welfare state regime. Cross-national variations in the level of centralisation, referring to territorial, functional, political and fiscal dimensions, cannot easily be collapsed into a single centralisation-decentralisation scale. The UK and the Netherlands are the most centralised among the MAELG survey countries, except for the functional dimension. Notable centralism characterises also the Czech Republic (except for the territorial dimension) and to some extent also Israel. Decentralisation characterises Sweden (except for the territorial

10 European Local Councillors and Local Governance Reforms 271 Table 1. Attributes of MAELG survey countries Country GDP per capita (CIA, The World Factbook) US$ 2009 estimates Government structure Welfare state regime Political leadership types (Heinelt and Hlepas 2006) United Kingdom Highly devel. 35,400 Unitary (within England and the devolved national administrations) Intermediate-liberal Diverse Sweden Highly devel. 36,800 Unitary Egalitarian Committee/collegial leader Norway Highly devel. 59,300 Unitary Egalitarian Collective/collegial leader Germany Highly devel. 34,200 Federal Conservative/egalitarian Strong executive mayor Austria Highly devel. 39,400 Federal Conservative/egalitarian Strong executive mayor Switzerland Highly devel. 41,600 Federal Conservative-intermediate Collective/collegial leader The Netherlands Highly devel. 39,000 Unitary Conservative Collective/collegial leader Belgium Highly devel. 36,600 Federal Conservative Collective/collegial leader France Highly devel. 32,800 Unitary Conservative Strong political mayor Italy Highly devel. 30,200 Unitary (several special autonomous stratus) Spain Highly devel. 33,700 Unitary, leaning towards federal Conservative southern Euro. Strong political mayor Conservative southern Euro. Strong political mayor Greece Highly devel. 32,100 Unitary Conservative southern Euro. Strong political mayor Czech Republic Transitional, 25,100 Unitary Intermediate Collective/collegial leader Poland Transitional, 17,800 Unitary Intermediate Strong executive mayor Croatia Transitional, 17,600 Unitary Intermediate Collective/collegial leader Israel Devel.-transitional, 28,400 Unitary Intermediate Strong executive mayor

11 272 E. Razin and A. Hazan dimension), and to an extent also Germany, Austria and Switzerland. The other countries display an intermediate mix in the centralisation-decentralisation scale. Cross-national comparative research tends to emphasise a convergence of welfare state regimes and governance practices, associated with trends such as economic globalisation and European integration (Wollmann 2000, Schubert et al. 2009). This convergence is argued to diminish the significance of the above mentioned typologies of welfare state regimes and local government systems, but does not eliminate path dependent specificities. In light of these trends we refrain from choosing one of the previously defined typologies as our explanatory variable and suggest an initial division of the 16 MAELG countries into eight detailed sub-types that could be grouped into fewer types based on our empirical findings. Three sub-types represent extreme models: 1. Centralised liberal: Great Britain a liberal welfare state characterised by substantial centralisation (territorial and fiscal centralisation, coupled with moderate levels of functional decentralisation), unlike the decentralised liberal US model. 2. Centralised comprehensive (conservative) welfare state: the Netherlands, characterised by territorial, political and fiscal centralisation. 3. Decentralised egalitarian welfare state: the Scandinavian prototype (Sweden and Norway in the MAELG survey), characterised by decentralisation, except in the territorial dimension. In between these three extremes are several sub-types, all being more decentralised variants of the comprehensive welfare state; the latter two possessing some ingredients of either decentralised liberal or Southern European systems: 1. A balanced Central European comprehensive (conservative) welfare state: Germany and to an extent Austria in both federalism could also have a balancing effect. 2. A more decentralised and liberal variation of the balanced comprehensive (conservative) welfare state: Switzerland. 3. A functionally centralised politically decentralised Napoleonic conservative welfare state: France and Belgium. a. Two additional sub-types represent transitional contexts: 4. The Southern European conservative welfare state: Spain, Italy, Greece more clientelistic, with somewhat lower levels of well-being than the six Western European models specified above. 5. The transitional sub-type: post-communist countries and Israel. These are economies in an intermediate level of development. Among them,

12 European Local Councillors and Local Governance Reforms 273 the Czech Republic is most centralised functionally, politically and fiscally, enabling extreme territorial decentralisation, whereas Poland is more balanced. Israel is rather centralised functionally and politically, less so in the territorial and fiscal dimensions. Their welfare state regime is a product of a mixture of influences of socialist roots, liberal sentiments, and EU comprehensive welfare state values. Countries of this sub-type share substantial similarities in their local political cultures. It is unclear whether such a detailed typology is meaningful in explaining attitudes of councillors towards reform, but starting with this typology enables the relevant distinctions to be accessed. At the other end of the spectrum of typologies stands the simplified north south divide. This divide has been attributed to sociopolitical culture, economic development, public service provision, and law and policy enforcement, being suggested in diverse fields such as citizen s views on health care services (Mossialos 1997). Assumed to distinguish between more functionally and politically decentralised Northern Europe and functionally weak and parochial local governments in Southern Europe (Page and Goldsmith 1987), it has been later argued that the significance of the north south distinction has diminished (John 2001, Goldsmith and Page 2010). Proposed, for example, as an explanatory factor for variations in urban leadership in Europe, actual patterns were found to be more complex than this distinction suggests (Getimis and Grigoriadou 2005). The proliferation of NPM practices, the growing emphasis on inter-local and inter-regional competitiveness, the rise of an intermediate tier of local government, the incorporation of post-communist countries into the new Europe, and other trends of north south convergence, accompanied by greater diversity within north and south, have been argued to blur the role of the north south distinction in explaining intergovernmental relations. The relevance of this distinction has been criticised, for example, in an examination of the application of European environmental policies (Borzel 2000). Nevertheless, the north south divide is used as a discoursive device by different actors at the intra-national (Gonza lez 2011) and European scales. It is perhaps simplistic and outdated, but it could be premature to assume that it is no longer relevant in influencing governance practices or attitudes. Associating cross-national variations in attitudes towards reform to local government sub-types Responses to 18 questions on attitudes towards reform are summarised in Table 2. The index presented in the table was calculated as follows: first the proportion of those who support the particular principle (agree and strongly agree, or desirable and highly desirable) in each country was divided by the

13 274 E. Razin and A. Hazan proportion of respondents who support the same principle in all 16 countries. Then, the proportion of those who reject the particular principle (disagree and strongly disagree, or undesirable and highly undesirable) in each country was divided by the proportion of respondents who reject the same principle in all 16 countries. This second measure is not identical to the first because some respondents chose a neutral reply. The two measures were combined by subtracting the second from the first. General attitudes towards the need to substantially reform local government are rather mixed, but more respondents do not see a need for major reform than those who see such a need. Among all respondents, over one-third is unsupportive of the need for major change, whereas slightly less than one-third is supportive (Table 2). The need for reform is acknowledged in particular in Norway and Israel, whereas the Dutch and Polish lead the opposition for major change. Table 2. Attitudes of councillors towards reform in general and selected democratic and administrative reforms by MAELG countries Separation Adherence of Decentralisation between bureaucracy to of local govt. politics and politically to promote administration defined goals participation Support for reform Devolution to neighbourhood organisations Centralised liberal UK Centralised comprehensive NL Decentralised egalitarian Norway Sweden Balanced comprehensive Germany Austria Decentralised-liberal balanced comprehensive Switzerland Napoleonic conservative France Belgium Southern European conservative Spain Italy Greece Eastern European/transitional Czech R Poland Croatia Israel Total (per cent) Agree 31.3% 51.8% 76.8% 57.6% 39.7% Disagree 38.6% 30.0% 8.4% 14.9% 30.3% (continued)

14 European Local Councillors and Local Governance Reforms 275 Most respondents slightly over 50 per cent support the separation between political decision-making and day-to-day administration and task fulfilment. Leading in their support for such separation are Greek and Polish councillors, whereas France (with its Napoleonic strong political mayor system) leads the opposition to separation between politics and administration, followed by the UK and Norway. A vast majority agree that local bureaucrats should as far as possible stick to politically predefined goals, although in Poland and the Czech Republic this principle is opposed by one-third of responding councillors. Indeed, in all four transitional countries the support for such strict subordination of bureaucrats to politically defined goals is weaker than average. A majority of councillors regard decentralisation of local government as necessary to involve citizens in public affairs. Substantial opposition to this statement is evident only in Switzerland and the Netherlands (about one- Table 2. (Continued) Support Support for an Support Support for surveys to monitor Support for co-decisions Support for citizen Support for local advisory for binding public procedures consultation for user referenda referendum referendum opinion with citizens procedure boards Centralised liberal UK Centralised comprehensive NL Decentralised egalitarian Norway Sweden Balanced comprehensive Germany Austria Decentralised7liberal balanced comprehensive Switzerland Napoleonic conservative France Belgium Southern European conservative Spain Italy Greece Eastern European/transitional Czech R Poland Croatia Israel Total (per cent) Agree 45.8% 51.5% 48.4% 57.8% 49.0% 77.4% 31.7% Disagree 21.7% 18.9% 26.5% 13.9% 26.0% 5.9% 35.1% (continued)

15 276 E. Razin and A. Hazan quarter of respondents). These countries represent two extremes: in Switzerland the lower support in this statement perhaps reflects the fact that such decentralisation is already in place, whereas in the Netherlands it reflects the adherence to centralised practices. Views over the devolution of responsibilities to neighbourhood organisations are, however, diverse. Support is broadest in the transitional economies Croatia, Israel, Poland, Czech Republic and in Greece, where it does not imply that such devolution actually occurs. Councillors Table 2. (Continued) Support for participatory democracy, including concerned local actors Direct elections of mayors Reducing the number of councillors Support for contracting out and privatising Support for competition between service providers Support for public private partnerships Centralised liberal UK Centralised comprehensive NL Decentralised egalitarian Norway Sweden Balanced comprehensive Germany Austria Decentralised-liberal balanced comprehensive Switzerland Napoleonic conservative France Belgium Southern European conservative Spain Italy Greece Eastern European/transitional Czech R Poland Croatia Israel Total (per cent) Agree 59.7% 57.3% 21.7% 35.0% 56.0% 41.1% Disagree 20.0% 27.1% 51.1% 41.7% 20.0% 26.3% Notes: The table presents the following index: A/B minus C/D. A high positive index indicates relatively strong agreement with the principle, a high negative index indicates relatively strong disagreement, zero indicates attitudes that are similar to the average among all respondents in the 16 countries. A Proportion of those strongly agreeing or agreeing with the statement in the particular country. B Proportion of those strongly agreeing or agreeing with the statement among all respondents in the 16 countries. C Proportion of those strongly disagreeing or disagreeing with the statement in the particular country. D Proportion of those strongly disagreeing or disagreeing with the statement among all respondents in the 16 countries.

16 European Local Councillors and Local Governance Reforms 277 from Switzerland, Norway, the Netherlands and Belgium lead the opposition to devolution of powers to the sub-municipal level. Councillors express positive attitudes towards local referenda about 50 per cent support referenda of various types. Support for referenda is exceptionally strong in Greece and rather strong in post-communist Europe. Expectedly, in Switzerland, where referenda are a major component of democratic process, nearly all support binding referenda. French councillors express strong support for non-binding referenda. Opposition to referenda is most evident in the Netherlands, reflecting its centralised traditions, followed by the UK and Norway. It should be noted that support for referenda and direct participation in different spheres of policy-making tends to be higher in countries that experienced non-democratic regimes in the twentieth century (except for the unique Swiss case), whereas in countries where democratic values have been stable referenda has not been regarded to be important as a means to consolidate democratic values. Four statements refer to additional modes of participation. A majority of respondents support frequent surveys to monitor local public opinion and co-decision procedures, where citizens can make binding decisions on certain local issues. A vast majority of over 75 per cent support citizen consultation procedures. However, when it comes to granting citizens more concrete powers the councillors are less enthusiastic: transferring the powers of scrutiny over municipal services to user boards is opposed by a majority of respondents. Expressed attitudes that favour the different modes of citizen participation are particularly strong in Southern Europe and in transitional countries. Greece, Spain and Croatia stand out in strong support in all modes of participation, perhaps reflecting their experiences with non-democratic regimes. Perhaps for similar reasons, councillors in the transitional/eastern European countries show marked support, relatively to other countries, to user boards. Councillors in Europe north and west of the Alps are more reserved in their support of these modes. Similar to their attitudes towards referenda, the British, Dutch and Scandinavians, as well as French and Belgians, also tend to be less supportive than others of these modes of participation. Dutch councillors show strong support only to citizen consultation procedures, whereas Germans and Austrians strongly support co-decision procedures. A positive sentiment towards participatory democracy political decisions should not only be taken by representative bodies but be negotiated together with the concerned local actors is expressed in various Southern European and transitional countries: Czech Republic, Croatia, Greece, Poland, Israel and Spain. The two Scandinavian countries Norway and Sweden stand out with a relatively high proportion of councillors (about one-third) that do not favour participatory action beyond representative bodies, and the Netherlands also has an above average proportion of councillors that do not seem to emphasise direct democracy.

17 278 E. Razin and A. Hazan Most respondents support direct election of mayors, but wide crossnational variations are evident in attitudes towards this issue, apparently also influenced by the practice in each country. Croatia, Israel, Greece, Poland, Italy Germany and Switzerland strongly support the direct election of mayors. Among these only the Swiss do not have strong executive mayors. A clear majority against the direct election of the mayor is evident primarily in the UK, the Netherlands, Norway and Sweden systems characterised by a relatively weak collegial mayor. A large majority of councillors, led by the French (with its over 36,000 communes), oppose reducing the number of members in the council. Poland and Israel are the prime exceptions in both a clear majority supports reducing the number of councillors. In Israel this attitude apparently reflects an image that a large number of councillors makes it more difficult to form a coalition, and a desire that a smaller number of councillors would enable giving them an allowance for their work. NPM reforms are controversial but two of three associated statements do enjoy a majority of supporters. There is a small majority for the notion that there are only few benefits from outsourcing and privatisation, but a clear majority supports public private partnerships and an even wider majority supports competition between service providers. Expressed support for competition-privatisation is marked primarily in transitional countries: Poland, Israel, and to an extent also Croatia and the Czech Republic. A negative sentiment towards forms of privatisation is particularly evident among councillors from Germany, Austria and Switzerland. Notably, although Britain is the origin of the modern application of public private partnerships, these mechanisms receive relatively little support among British councillors. The substantial cross-national variations in attitudes of councillors towards reform agendas could have complex explanations, relating also to the status of councillors in each country, and to specificities of local government systems, but Table 2 does reveal some interesting patterns. The UK s centralised liberal sub-type is characterised by a lack of enthusiasm for reform. The centralised nature is accepted by councillors who do not show particular support for major change, either in the relations between the political and the bureaucratic rank, in introducing local referenda and other modes of citizen participation, or in the direct election of mayors. These views are in line with the stable democratic experience and the lack of participatory reforms in Britain (Kersting et al. 2009). Interestingly, the liberal element of this sub-type is not received with enthusiasm by councillors, who do not express clear support for public private partnerships. Attributes of the Dutch centralised comprehensive welfare state are reflected in the responses of councillors, and show a strong sentiment towards the existing centralised system. Councillors support separation between politics and administration, in line with a recent Dutch reform, and

18 European Local Councillors and Local Governance Reforms 279 the subordination of bureaucrats to politically defined goals. Responses also show lack of interest in decentralisation within the municipal level, opposition to local referenda and to the transfer of powers to joint frameworks with citizens beyond consultation procedures. Councillors also oppose the direct election of mayors and mildly support NPM, competition and privatisation principles. Views of councillors in the decentralised egalitarian Scandinavian subtype are surprisingly similar to those of councillors in the centralised Dutch model. These show a sentiment for clear separation of politics and administration, opposition or lack of interest in local referenda and in other modes of direct participation, opposition to the direct elections of mayors, and support of privatisation (in Sweden). These views are seemingly not in line with overriding values of decentralisation and egalitarianism. They reflect perhaps confidence in the stability and consensus over the present system that does not necessitate major reforms, indicating that political attitudes in centralised Netherlands and decentralised Scandinavia are not that different. The balanced comprehensive German/Austrian sub-type differs from the above three types. Attitudes of councillors are more positive towards local referenda, other modes of participation, and the direct election of mayors, but somewhat negative towards liberal NPM steps of privatisation. These views indicate an emphasis on principles of local democracy and the welfare state. As the label balanced indicates, attitudes in this sub-type lean towards the average and are rarely extreme in supporting or opposing particular notions. The more decentralised and liberal variant of the Swiss balanced sub-type is indeed rather similar to the balanced comprehensive sub-type in attitudes of councillors, although these are influenced by the specificities of Swiss decentralised federalism and direct democracy. Attitudes of councillors in the Napoleonic conservative French/Belgian sub-type are diverse, rarely extreme in supporting or opposing particular notions, except for the marked opposition (particularly in France) to the separation between politicians and the fulfilment of administrative tasks. Although regarded in the past as part of Europe s south, in the local government literature (Page and Goldsmith 1987), responses to the MAELG questionnaire (Table 2), as well as broader patterns of development, reveal greater resemblance to Europe s north. Southern European (Spain, Italy and Greece) and East European/ transitional (Czech Republic, Poland, Croatia and Israel) sub-types are clearly distinct in attitudes expressed by councillors, compared to those of the six sub-types of north-western Europe. Councillors in countries of both sub-types express clear support in separating politics and administration, in decentralisation at the local level, in local referenda (particularly in Southern Europe), in other modes of participation, and in direct election of mayors. Support in all modes of decentralisation and democratisation is

19 280 E. Razin and A. Hazan exceptional in Greece and Croatia, whereas Italy and the Czech Republic are moderate versions of these sub-types. Positive sentiments towards NPM/efficiency are also marked particularly in transitional countries. These show strong support for privatisation and competition. The only three countries where councillors expressed support for reducing the number of councillors (a step associated with efficiency) are of these subtypes: Poland, Israel and Italy. Thus, councillors in transitional countries and in Southern Europe express clear support in most proposed modes of decentralisation, democratisation and privatisation. It can be questioned whether these views reflect practices and real political change or a desire for change that is not necessarily perceived as feasible, for example in Greece (Hlepas 2003), or perhaps even an acknowledgement of attitudes that seem politically correct and not necessarily a real urge for their implementation. A multivariate perspective In order to conduct a multivariate analysis of factors explaining councillors attitudes, 16 of the 18 statements were grouped into three dependent variables, each representing a reform agenda. The first two separation of politics and administration, and privatisation and competition are associated with notions of NPM, and the third includes the broad spectrum of statements that concern democracy, participation and devolution. Each of the three dependent variables is the mean score given by each respondent to the statements grouped in this variable. The third dependent variable democracy, participation, devolution was further broken into three separate dependent variables, one indicating support for decentralisation and devolution, the second, support for referenda, and the third, support for new forms of participation (Table 3). Two of the 18 statement were not included in any of the dependent variables: a general statement of support for reform, and reducing the number of councillors. These statements, not clearly related to any of the major reform agendas, are excluded from the multivariate analysis. Means of the dependent variables for each of the country groupings clearly reveal the particularly marked support for all sorts of democratic and participatory mechanisms in transitional and Southern European countries (Table 3). A particularly reserved attitude towards such democratic and participatory mechanisms is evident in the Netherlands and in Scandinavian countries, as well as in the UK. Swiss councillors have a particularly positive attitude towards referenda. German and Austrian councillors are far more positive towards new forms of participation and referenda than the rest of north-western Europe, their responses reflecting perhaps their experiences with a non-democratic regime, making it easier to introduce direct democracy (Kersting et al. 2009), and deterring councillors from giving responses that could be interpreted as indicating weak democratic values.

20 European Local Councillors and Local Governance Reforms 281 Table 3. Dependent variables by country groupings (means). Dependent variables support for: Separation of politics and administration Privatisation, competition Notions of democracy, participation, devolution Decentralisation, devolution Referenda New forms of participation United Kingdom The Netherlands Norway, Sweden Germany, Austria Switzerland France, Belgium Transitional a South b Notes: a. Czech Republic, Poland, Croatia, Israel; b. Spain, Italy, Greece. The country typology is significant for all six dependent variables at the level. Support for separation of politics and administration 2 3 in Table 2. Support for privatisation, competition in Table 2. Support for broader notions of democracy, participation, devolution 4 14 in Table 2. Support for decentralisation, devolution 4 5. Support for referenda 6 8. Support for new forms of participation 9 13.