Introduction

Since the 1950s, local governments in developed countries have played a prominent part in the provision of public welfare services. Over the past decades, local governments have faced pressures to produce higher value public services at lower costs. Especially the smaller communities have experienced their territorial scale as no longer compatible with the increasing scale of production required to efficiently provide for high standard public services. Overall, three strategies are in use to cope with the problem of scale. The first is territorial reorganization that results in large-scale local government, a strategy that was followed in some countries (i.e. Sweden, United Kingdom, Australia), but that has met with steep resistance in many others because of the value attached to citizen participation and the self government of small communities. The second strategy is the outsourcing of service delivery to large-scale private (or public) companies; this strategy was traditionally in use for the provision of capital intensive services (electricity, waste management), but has spread to a large variety of public welfare services, due to the rise of neo-liberal ideologies and the New Public Management Movement (OECD 2002). This article focuses on the third strategy: inter-governmental or, more specifically, inter-municipal cooperation.

The joint provision of public services by local governments creates economies of scale and scope and thus offers possibilities to overcome scale-related production obstacles, to meet the rising expectations of citizens and to achieve cost-efficiencies (Bish and Ostrom 1973; Gyapong and Gyimah-Brempong 1988; Morgan and Hirlinger 1991; Herweijer 1998: 150; Hulst 2000: 2–4; Airaksinen and Haveri 2003: 9; Hepburn et al. 2004: 14–16; Shresta 2008: 1–9). Nowadays, inter-municipal cooperation with respect to the provision of public services is a widespread phenomenon throughout the Western world (Stoner 1964; ACIR 1985; Dafflon and Perritaz 2000; CDLR 2001; Agranoff and McGuire 2003; Warner 2006).Footnote 1 Waste disposal, water provision, social (security) services, fire brigades and emergency services, public transport, environmental protection, public health and education constitute the policy sectors where local governments frequently cooperate. In spite of being relatively widespread, inter-municipal cooperation has hardly been subject to international comparative research. This article reports on research that took to analyse the background, characteristics and performance of institutional arrangements in a number of countries, viz. Belgium (Flanders), Germany, Italy, Finland, France, the Netherlands, Spain and the United Kingdom (Hulst and Van Montfort 2007).

With respect to inter-municipal cooperation on service delivery three findings are notable. First, there is considerable variety in institutional arrangements. Arrangements differ with respect to their composition (only local governments or a mix of municipalities and other public and/or private partners), their scope (the cooperation includes one service or a range of services) and the degree of organizational integration (the service is carried out by a separate joint organization or it is organized through agreements). Second, there are considerable differences between the countries included in the research with respect to the dominant forms of cooperation. Third, the picture is not stable. Over the past decades shifts have occurred on the scope, the composition and the degree of organizational integration of arrangements for cooperation. This article sets out to identify the factors that determine the institutional design of cooperation and thus understand how these shifts can be explained. We start with a brief outline of the concept of inter-municipal cooperation and the theoretical framework used in the research. For a more extensive discussion we refer to Hulst and Van Montfort (2007a: 12–16).

Theoretical framework

As a preliminary note, the concept of inter-municipal cooperation is used in a broad sense and includes all arrangements where local governments cooperate with each other, with other public authorities or with private institutions. For the sake of simplicity we will frequently just refer to cooperation between municipalities. With respect to the concept of cooperation, two features are basic. The first is that the interactions between the cooperating partners concern a common task or goal and enjoy some degree of institutionalisation. Although in many situations occasional cooperation between public entities occurs, our analysis is limited to cooperation that shows more or less stable patterns over time and concerns certain standing issues, tasks, services or policies.

The second feature concerns the relation between the institutions of cooperation and the constituting municipalities and other participants. We reserve the concept of inter-municipal cooperation for institutional arrangements that formally depend on local government (and other participants) for their establishment, decision-making and continued existence. Therefore, we exclude authorities, foundations or companies that have a statutory independence and mandate, even if local government is somehow involved as a cofounder, shareholder or member of a supervising board or council.

Our theoretical framework is built on basic concepts of sociological and historical institutionalism. Central to our approach is the notion that behaviour and decisions result from the interaction between institutions and actor rationality. In the context of inter-municipal cooperation we assume public and private actors to be goal-oriented and rational, albeit in a bounded way, pursuing their interests and making use of their powers to achieve their goals. But at the same time, we assume that institutions provide meaning and influence the way actors define their interests and preferences (cf. Berger and Luckmann 1966; March and Olsen 1987; DiMaggio and Powell 1991; Hall and Taylor 1996). Moreover, we assume that the institutional context determines the opportunities and constraints actors face and sets the rules of the game for actors pursuing their goals. While existing institutions thus shape the strategies and conduct of actors, these in return can crystallize out in new institutions (Thelen and Steinmo 1992, 10; Scharpf 1997: 38; Scharpf 2007). We thereby assume that path dependency plays a role: strategic choices made at one point in history constrain the range of possibilities for later strategic choices as much as existing institutions leave their tracks in new institutions (North 1991; Tilly 1994; Hay and Wincott 1998).Footnote 2

This theoretical approach implies that the national institutional context is an important factor in the explanation of the preference structures of the actors involved and of the presence and characteristics of cooperative arrangements. We distinguish three different elements of the institutional context: the formal structure of the state, the administrative culture, and legislation and incentive structures that specifically relate to inter-municipal cooperation. The first element, the formal structure of the state, generally laid down in a country’s constitution and in organic legislation, includes the number of administrative tiers, the distribution of responsibilities between the different tiers, the scope and autonomy of local government and the number and size of the municipalities (Agranoff and McGuire 2004: 496–501; UNDP 2006: 17–21). These features to a great extent determine the need for and possible advantages of cooperation.

The second element, the administrative culture comprises sets of values, norms, informal rules and traditions relating to the state, its political organization and its public administration. It is generally assumed that these normative elements are not only reflected in the formal state structure, but also affect the relations and interactions between the public and private sector and between public authorities themselves (cf. Loughlin and Peters 1997; Hulst 2000: 34–35; Loughlin 2001; Visser 2002). While there is ample evidence for the presence of different, long-lasting state traditions, we should also bear in mind that states have not been immune to global trends that emerged over the past decades. Western and Northern European countries have, albeit in different degrees, embraced the ideas of New Public Management (Hood 1995; Pollitt and Bouckaert 2000) and of governance through public-private networks (Rhodes 1996; Stoker 1998; Agranoff 1998), representing values and policy styles that are at odds with at least some of the existing state traditions. We assume that the dominant administrative culture will bear upon the preference for strategies to cope with the need for large-scale service delivery, e.g. outsourcing to the private sector or inter-municipal cooperation.

The third and last element of the national institutional context we assume relevant is the way upper level government — central or regional government — approaches the issue of cooperation. In most countries included in our research legal frameworks to regulate cooperation are present and in a number of countries central governments use financial incentives to promote (certain forms of) inter-municipal cooperation. Legislation and incentives define opportunities and constraints for cooperation and make some institutional arrangements more attractive than others, sometimes banning others (Parrado Díez 2006b: 8–13; UNDP 2006: 26–27, 29–32; Osterrieder et al. 2006: 31–32).

We further assume that the preferences of local government and the presence and characteristics of cooperative arrangements are not only influenced by the national institutional context but also by ‘environmental factors’ (Morgan and Hirlinger 1991; Carr and LeRoux 2005: 18; UNDP 2006: 8). Social-economic, technological and demographic circumstances can bear on the need for cooperation: rising scales of production, urbanization and increasing market pressures may force local governments to extend the domain of cooperation to new services. Environmental factors may also influence the preference for certain cooperative arrangements. When budget cuts are imminent local governments’ priority may be to seek forms of cooperation that ensure a low cost provision of services. Highly dynamic environments can induce municipalities to opt for the more flexible forms of cooperation.

Our central assumption is that the combination of the goals and preferences of local government, the national institutional context and external factors determines whether local governments use inter-municipal cooperation as a strategy to cope with the issues of scale, and if so, what types of institutional arrangements emerge. In the same way, we assume that shifts in the institutional arrangements for cooperation are somehow related to changes in the institutional context, external factors and preference structures of local government.

In the next sections we will successively discuss the three dimensions of institutional design of inter-municipal service delivery mentioned above and try to explain the shifts we have observed along these dimensions: the scope of cooperation, the composition of cooperative arrangements and the degree of integration.

Scope of cooperation: from single-purpose to multi-purpose organizations and vice versa

The scope concerns the range of tasks and activities included in the cooperative arrangement. We distinguish between single-purpose and multi-purpose arrangements (Hulst and Van Montfort 2007c: 216). A clear example of single-purpose arrangements is formed by joint authorities in their own right that engage in social services, refuse collection or music education. Multi-purpose authorities or arrangements, on the other hand, integrate a number of tasks or activities, up to the point where joint authorities assume the form of a quasi-regional government responsible for a wide range of public services and planning activities.

Shifts go both ways

Taking stock of the scope of the inter-municipal service organizations in the countries included in our research, we can make two observations. The first is that in different countries different patterns have developed: in Flanders, Finland and Germany single-purpose organizations dominate the picture, in Italy organizations are for the major part multi-purpose, and in France, Spain and the Netherlands substantial numbers of both types are present.Footnote 3 The second observation is that in a number of countries there are considerable shifts in preference for the two types of organizations. Moreover, these shifts are not unidirectional.

Our research shows that in some of the countries where single purpose organizations dominated the picture over a length of time, after some point multi-purpose organizations became more and more popular. In France, single-purpose entities traditionally dominated the picture, but over the last decade, there is a substantial growth of multi-purpose organizations. Between 1995 and 2006 the single-purpose SIVU saw their number decrease from 14,490 to 11,739; in the same period more than 1500 new multi-purpose organizations—Communautés de communes, Communautés urbaines and Communautés d’agglomeration—were set up.Footnote 4 Also in Finland the traditional way to organize inter-municipal cooperation was to establish single purpose joint authorities (Kuntayhtymä) that took care of specific services on behalf of the municipalities, especially in the fields of health care, care of the disabled and education. Over the past decade however, new joint authorities have been set up with a multi-purpose character. Between 1997 and 2004, 62 multi-purpose inter-municipal arrangements for sub-regional development emerged; they nowadays cover the bigger part of the country. More than 25% have a formal status and actually provide services themselves. A radical new multi-purpose (pilot) model is found in the Kainuu-region, where nine municipalities formed a joint authority responsible for a series of services (education, health care services, welfare for drug abusers, environmental health care, child welfare) and for economic planning and development (Haveri and Airaksinen 2007: 51 et passim). In the Netherlands, between 1985 and 1990, the number of single-purpose arrangements dropped from 1,500 to 659; the bulk of the other arrangements were integrated in 95 multi-purpose joint authorities (Pröpper et al. 2005: 23–24).

The opposite shift, from multi-purpose to single purpose organizations, can be observed in Spain. After the introduction of the 1978 Constitution there was a sharp rise in the establishment of cooperative arrangements and multi-purpose organizations dominated the picture. In 1998 there were 882 mancomunidades, most of them multi-purpose in character and 428 consorcios, for the larger part single-purpose. Around 1990 the tables turned and more and more single purpose organizations were established. In 2007 the number of consorcios had spectacularly risen to 1014, while the number of mancomunidades had grown at a much slower rate and did not reach a 1000.Footnote 5 The case of Netherlands presents a second example. While 95 multi-purpose organizations were created between 1985 and 1990, in 2005 more than half of them had been dissolved; the number of single-purpose organizations kept stable (Pröpper et al. 2005: 23–24).

Institutional context: legal stipulations and incentives

If the formal structure of the state and other elements of the institutional context induce local governments to cooperate for the delivery of public services, what determines whether they set up single-purpose or multi-purpose organizations or arrangements? For an explanation we turn to the opportunities and constraints created by the legal frameworks and incentive structures directly related to inter-municipal cooperation. It is possible to distinguish between three different situations: one in which legislation and policies of upper level government aim to prevent the establishment of multi-purpose organizations, explicitly forbidding them, prescribing single-purpose arrangements or somehow creating incentives for the latter; one in which upper-level government does exactly the opposite, imposing constraints on single-purpose service organizations; and one in which upper-level government leaves it to the municipalities themselves to choose the form they prefer.

Belgium (Flanders) and the Netherlands present examples of the first situation. Both countries have at one time in history issued legislation containing restrictions on multi-purpose service organizations. The Dutch 1950 Provisions Act stipulated that every service, task or joint interest municipalities wanted to cooperate on, required a separate arrangement or organization. The main reason for this absolute ban on multi-purpose arrangements was to prevent the proliferation of autonomous, multi-purpose regional authorities. There was a general fear that such authorities would develop into a new administrative level and hollow out local autonomy (Hulst 2000: 7). In Flanders, the 2001 Framework Decree on Inter-Municipal Cooperation introduced four new administrative forms to replace the existing intercommunales. While the intercommunales could have multiple objectives, the service association must limit its activities to one well-determined service, the mandated association to one or more clearly stated tasks in one or in functionally related policy areas and the inter-local and project associations to a project of limited scope. The restrictions on the scope of inter-municipal cooperation are meant to restore the democratic supervision by the local councils (De Peuter and Wayenberg 2007: 27–28). The dominant position of single-purpose organizations in the Netherlands between 1950 and 1985 and the shift from multi-purpose to single purpose organizations in Flanders over the past years are clearly related to the legal constraints concerning inter-municipal cooperation.

The second situation was found in France (after 1992) and in the Netherlands (after 1985). Both countries introduced legislation to promote the establishment of multi-purpose joint authorities, preferably integrating all services and tasks in a rural region or urban agglomeration. Prior to these interventions, in both countries a large number of single-purpose organizations had developed over time, which organized different groups of municipalities that varied from service to service. It had resulted in fragmented and obscure administrative systems that lacked built in mechanisms for integrated policy-making and coordinated service delivery (Hulst 2000: 10–11; Hulst 2005: 107–108; West 2007: 72–74). The Dutch strategy was primarily one of coercion. The 1985 Joint Provisions Act simply ordered provinces to divide their territory into functionally coherent regions and obliged municipalities to integrate their cooperative arrangements into one multi-purpose joint authority (Hulst 2000: 12–14). The French followed a piece-meal approach using a mix of instruments. From 1992 on, through subsequent legislation, a series of new administrative forms were introduced for the urban areas, with different sets of compulsory functions and their own financial resources (West 2007: 78–81). Moreover, central government introduced the concepts of aire urbaine and agglomeration (for the urban areas) and pays (for the rural areas), referring to functionally coherent territories, and has been pushing local government to organize their cooperation on that basis (West 2007: 88). Amongst other things, it introduced criteria and procedures for the identification of functionally coherent territories and set out to close contracts with the administrative boards of the pays, entailing financial resources for social and economic development (Hulst 2005: 108).

While the main purpose of the Dutch project was to rationalize inter-municipal cooperation and improve transparency for the sake of democratic accountability, in France the promotion of multi-purpose authorities for service delivery (and planning) seems to serve as a means to create a fourth national territorial institution alongside the region, the departement and the municipality (West 2007: 89). Whatever the central government motives in play, the legal frameworks and national policies that were introduced in the Netherlands and France in 1985 and 1992 respectively, can explain the shift from single purpose to multi-purpose cooperative arrangements.

If the third situation occurs, one in which upper-level government leaves it to the municipalities themselves to choose the institutional design for cooperation, the preference structure of local government is likely to be the determinant factor, a factor that we will address now.

Local preferences

To gain insight in the role of local preference structures for the institutional design of cooperation, we must dwell upon some of the features of the different forms of cooperation.

Cooperation opens the way to higher quality services at lower costs, but cooperation itself also brings along extra costs compared to service delivery by individual local governments. In most of the countries included in our sample the establishment of a joint organization requires drawing up statutes that contain provisions concerning its aim, the decision-making authority attributed by the municipalities, the governing bodies, the funding of the organization, the modification of the statutes and the dissolution of the organization, amongst others. In some countries the establishment of the organizations and their statutes are subject to approval of upper level government. In Flanders, for example, the establishment of a joint service organization (dienstverlenende vereniging), one of the heavier forms of cooperation, is surrounded with extensive safeguards and requires the drawing up and approval of a series of documents (De Peuter and Wayenberg 2007: 25–26). Apart from the start up costs, there are also costs related to the government and management of the organization. The joint character of the organization generally implies the presence of general boards composed of delegates from the local councils and executive boards with three or more members. Moreover, decision-making at the supra-municipal level involves periodic consultation of and reporting to the local councils, activities that require administrative support at the inter-municipal level.

Assuming that local governments seek cost efficient service delivery—this will be the case especially when local resources are scarce—start up and management costs constitute drivers for the establishment of multi-purpose service organizations, because it enables them to share management costs for different services. Moreover, once a joint authority is established, it is easier to transfer new services to an existing organization than to set up a new single-purpose organization. The Spanish case provides a good example. After the transition to a democratic regime, land marked by the 1978 Constitution, for the major part microscopic local government with few technical and economic resources was not only granted formal autonomy for the first time in history, legislation also stipulated a series of basic public services to be rendered. The vast majority of the 8,100 municipios—7,000 of which have a population fewer than 5,000—was not in a position to render these services on a stand-alone basis. Consequently, the number of mancomunidades, public bodies set up and governed by groups of neighbouring municipalities, rose sharply. The major part of the now 998 mancomunidades Footnote 6 is multi-purpose and there are many that cover all the services and tasks that local governments in a certain area cooperate on, thus profiting from a common administrative infrastructure (Nieto Garrido 2007: 178).

While considerations of cost-efficiency are a driver to set up multi-purpose organizations or to integrate new services into existing organizations, our research shows that once these organizations are in full operation, issues of manageability and control arise. In the first place, the more services are integrated into one organization, the more difficult it is to manage the organization as a whole, to coordinate and create synergy between the different services and to manage the separate services adroitly. In the second place, big multi-purpose organizations sometimes prove to be a threat to the very municipalities that established them. Spain and the Netherlands present examples of countries where considerations of manageability and control have led to a shift from multi-purpose to single purpose organizations.

The Spanish consorcios constitute an administrative form especially designed to institutionalize vertical intergovernmental relations (Nieto Garrido 2007: 182). However, there are an increasing number of single-purpose consorcios that are purely inter-municipal; in 1998 they made up about 8% of a total of 428, in 2006 purely inter-municipal consorcios amounted to more than 10% of a total of 1014. There are strong indications that the complicated governance structures of the multi-purpose mancomunidades hinder the effective management of the organizations and that local governments seek the simpler, professional management structures of the smaller and more transparent single-purpose consorcios (Font and Parrado Díez 2001). In the Netherlands there are examples of municipalities dismantling multi-purpose organizations in order to improve the management of services and to regain local control over inter-municipal authorities that in their view had gotten out of hand.Footnote 7 In the province of Noord-Brabant, the participating municipalities abolished four out of six multi-purpose joint authorities that organized cooperation on a number of public services (health care, emergency services) and planning tasks (public housing, spatial planning). Subsequently, services were organized in single-purpose organizations or in multi-purpose entities integrating only closely inter-related services.

When choosing between single and multi-purpose organizations, apart from considerations of start up costs, management costs and control, there is also the issue of scale. Every service has an optimum scale of operation, i.e. a scale on which the service can be rendered in the required variety and quality at the lowest production costs (cf. Ostrom 1976). If a service requires a variety of specialized personnel and capital goods, the minimum scale of operation will be relatively big. On the other hand, some services also have a maximum scale of operation given the characteristics and availability of the production goods involved. Fire stations must be able to reach seats of fire within a certain time limit and can thus only service a limited territory. The integration of different services into one multi-purpose organization generally implies that all services are delivered to the same group of municipalities, and therefore on the same territorial scale. For some of the services the multipurpose organization will be under or over-bounded (cf. Bennett 1993: 7–8). Put in another way, if an optimal scale of public service production is a dominant consideration for the partners involved, they will tend to set up different single-purpose organizations, so that each service can be delivered at its own optimal scale.

France and the Netherlands present good examples of the relevance of scale considerations on the institutional design of cooperation. In both countries cooperation between municipalities has a long history and has gradually intensified over time. In France local government was free to opt for single-purpose (SIVU) or multi-purpose organizations (SIVOM); single-purpose organizations for a long time dominated the picture. The same communes generally participate in a number of single-purpose organizations at the time, which frequently organize different groups of communes and thus operate on different territorial scales (Hulst 2005). In the Netherlands the 1500 single-purpose arrangements that were created between 1950 and 1985 covered more than 30 different services and showed substantial variation in composition and scale (Hulst 2000: 10). The logic of the appropriate scale came fully to the fore when central government tried to rationalize inter-municipal cooperation and issued the 1985 Joint Provisions Act prescribing that all cooperation should take place in one multi-purpose public body organizing the same group of municipalities and operate on the same scale. Local government vehemently protested and successfully lobbied for escape clauses allowing them to engage in cooperative arrangements on a smaller or bigger scale, if the service in question so required. In 1990, despite the legal stipulations aimed at the establishment of multi-purpose organizations, 754 single-purpose cooperative arrangements still remained (SGBO 1991). Even so, over the past decades Dutch local government has not stopped to lobby for the elimination of all restrictions concerning single-purpose organizations. In the end it was successful: in a 2006 revision of the Joint Provisions Act the restrictions were lifted.

Composition of arrangements: from horizontal to vertical forms of cooperation

The composition of cooperative arrangement refers to the participating actors. Cooperation can be restricted to municipalities, but local governments can also extend their cooperative arrangements to include other public authorities (upper level government, specialised government agencies) or private sector organizations (foundations and companies) (Hulst and Van Montfort 2007c: 212). We limit our discussion to the shift between horizontal arrangements that organize cooperation between municipalities and vertical arrangements that integrate municipalities and upper level government.

Two types of vertical arrangements

In almost all countries included in the research there is a strong presence of cooperative arrangements that involve only municipalities (Hulst and Van Montfort 2007c: 213). However, in a number of countries institutionalised cooperation between local government and upper level government institutions is gaining popularity. This is the case in Spain, France and the United Kingdom. We can distinguish between two types of vertical arrangements. The first concerns standing organizations set up to provide for the actual delivery of public services, where local governments cooperate with each other and with upper level local or regional government. The second concerns contract like arrangements, which primarily aim to improve service delivery at the local level through coordination and performance management. Vertical contracts are generally closed between local government(s) on the one hand and central and/or regional government on the other.

The Spanish consorcios and the French syndicats mixtes are examples of the first type. Syndicats mixtes are standing organizations with a board and staff of their own. Their number has risen substantially over the last decade, from 1107 in 1995 to 2749 at the end of 2006.Footnote 8 There is a wide variety of syndicats mixtes, some including the departement and a number of multi-purpose and single-purpose inter-municipal organizations, some also including the regional government in place. They cover a wide range of services, from public utilities such as electricity, water provision and public transport to education and tourism. The number of Spanish consorcios rose from 428 in 1998 to a total of 1014 in 2006.Footnote 9 As mentioned in the previous section, the share of purely inter-municipal consorcios has risen, but the vast majority of conosorcios still organizes cooperation between different levels of government, and is thus vertical in character. In Spain, the cooperation between local governments and the provinces is the most popular of all possible vertical combinations. They cover more than 30 different services (Font et al. 1999:12; Parrado Díez 2006a: 7).

The second type of vertical arrangements concerns contractual agreements. They include the Contrats de ville, Contrats d’agglomeration and Contrats de pays in France and the Local Public Service Agreements in the United Kingdom. Contrats de ville offer additional financing or influence over central policy-making in return for greater coherence in local actions (Hall and Mawson 1997; Sullivan 2004: 10; Parrado Díaz 2006b: 9). The first generation Contracts de ville (1993–1997) resulted in a total 214 contracts covering 1.300 neighbourhoods, which dealt with themes such as housing, transport and employment (DIV 2000: 3–4). The Contrats de ville have been substituted by a new type of contracts for the period 2007–2009, the Contracts urbains de cohesion sociale. In June 2007 nearly 490 contracts were signed covering 2200 neighbourhoods.Footnote 10 Although legislation allows for Contrats de ville between the central state and individual cities, more than 70% organize cooperation with a number of municipalities.Footnote 11

In the United Kingdom, Public Service Agreements are used across central and local government as a mechanism to drive up performance and quality of services. Local Public Service Agreements (LPSA) were first set up in 2000 and were extended to all upper tier authorities in 2001. The purpose of the agreements is to support local authorities in tackling underperformance and to help achieve specific strategic aims over the life of the agreement. Financial rewards, paid directly by the government, are associated with the achievement of the targets agreed upon (Parrado Díez 2006b: 9; Kelly 2007: 202–203). Since 2005 the system of LPSAs is complemented with Local Area Agreements (LAAs). The primary purpose of an LAA is to simplify funding arrangements from central government to councils. Furthermore, the program is designed to promote closer cooperation between providers of public services, with the aim that these arrangements will allow increased flexibility for local innovations in service delivery to fit local circumstances (Kelly 2007: 203–205). From 2001 the Government Offices signed 144 LPSAs with upper tier local authorities. Two rounds of LAAs followed, both rounds including more than 60 agreements.Footnote 12 In areas with a two-tiered local government, all local entities involved are co-signers of the agreement and coordination between them is thus one of the elements. But more so, the agreements aim at the coordination of service delivery between individual local governments and other providers, public and private.

Institutional context: legal stipulations and incentives

The increasing popularity of the first type of vertical arrangements, the service organizations like the Spanish consorcios and the French syndicats mixtes, cannot be ascribed to obligations following from new legislation. In both countries legislation has since long made it possible to set up cooperative arrangements between local government and upper level government entities. In fact, the syndicats mixtes were introduced in 1935, and a decree from 1955Footnote 13 explicitly stated that public bodies of different administrative levels could take part.Footnote 14 In Spain, the 1955 Reglamento de Servicios de las Corporaciones Locales laid down rules for the consorcios. Public bodies such as mancomunidades, provincial authorities and central government bodies were allowed to participate. Since the 1978 Constitution, the regional autonomous authorities, the Comunidades Autónomas, can also be partners in a consorcio (Nieto Garrido 2007: 172). While the legislation in both countries contains criteria and procedures for the establishment and management of the joint organizations, it is up to the partners to decide if they want to engage in this type of cooperation. To some extent, the growth of vertical service organizations has to do with incentive structures. In Spain there are substantial differences between the regions with respect to the popularity of consorcios, and the preferences of the Comunidades Autónomas seem responsible for this, as in some regions funds are linked to the preferred form of cooperation (Font and Parrado Díez 2001; Nieto Garrido 2007: 175).

With respect to the rising popularity of the second type of vertical arrangements, the contracts in France and the United Kingdom, the financial incentives created by the respective central governments seem to be the main explanatory factor. In both countries, the vertical arrangements came into existence after central government had introduced a financial support system. In the United Kingdom the LPSAs were backed up by pump priming grants and performance reward grants (Parrado Díez 2006b: 9). In the period of the first generation LPSAs (2001-early 2004), central government offered a maximum performance reward grant of 2½% of a year’s net budget for achieving the negotiated targets in full. A scaled-down grant was given for achieving a large portion of the expected improvement (Kelly 2007: 203). The French programs for the different types of contracts imply financial transfers of many billions of euros from central and regional governments to local government in the form of credits and tax exemptions.

In both countries the horizontal coordination of service delivery and regeneration policies is one of the explicit goals of the vertical arrangements. Cooperation between municipalities and public and semi-public agencies to overcome the territorial (France) and functional fragmentation (UK) in practice serves as one of the conditions for closing the contracts (Wassenberg et al. 2006: 19, 39). Local government and other partners at the local level do not have much choice. They either commit themselves to cooperate with each other or stand to loose substantial financial resources.

Local preferences

While the increase in vertical contractual arrangements implying cooperation between local institutions and upper level governments can be explained by the financial incentives created by central governments, the growth of the number of vertical service organizations in Spain and France, the consorcios and the syndicats mixtes, cannot. There seem to be three reasons for local governments to engage in vertical service organizations.

A first reason to create joint authorities that integrate public entities on different administrative levels is the need for vertical coordination between interdependent or sometimes overlapping jurisdictions of the actors involved. In France, syndicats mixtes organizing cooperation between inter-municipal authorities, departments and regions are found in sectors like water management and public transport, policy domains on which all entities have some competency. In Spain, examples are consorcios that provide public transport and consorcios that act as chambers of commerce. In these cases, the consorcios provide the organizational setting for the execution of joint responsibilities of local governments and provinces.

A second reason relates to the scale of cooperation. This seems to be especially relevant to explain the growth of certain types of single-purpose syndicats mixtes in France. In 1999, the Chevènement law introduced a series of new administrative forms for inter-municipal cooperation, but both the old (SIVU and SIVOM) and new joint authorities (Communauté urbaine, Communauté d’agglomeration, Communauté de commune) were often not large enough for the effective delivery of certain public services. This was particularly the case for waste management (waste-to-energy plants or recycling centres) and the provision of electricity, which must operate on a scale much larger than an agglomeration or a rural inter-municipal authority (West 2007: 87).

A third reason to create vertical service organizations is to gain access to resources from upper level government, either financial resources or technical know how. This factor seems especially relevant for Spain. Spanish legislation stipulates that the provinces support and cooperate with local government on legal, economic and technical issues (Nieto Garrido 2007: 170). Therefore it stands to reason that Spanish local governments seek cooperation with the provinces (frequently) and the regional government (occasionally) to provide services like specialised fire brigades (forest areas), water provision, waste management, and cultural and educational facilities. For small communities scant of resources, upper level government has more technical know how and financial resources to offer than equally poorly equipped neighbouring local government.

Organizational integration: from standing organizations to contractual agreements

The degree of organizational integration refers to the distinction between standing organizations and lighter forms of cooperation such as agreements. A standing organization represents the most organizationally integrated form of cooperation. It implies the integration of activities formerly carried out individually by municipalities into a new organization jointly run by the municipalities and any admissible participants. Agreements constitute a less integrated form of cooperation. They can organize service delivery between the partners, for example, by outsourcing the activities to a participating municipality or private company (Hulst and Van Montfort 2007c: 219).

Towards lighter forms of cooperation

With the exception of the UK, in all countries included in the research, cooperating municipalities make use of standing organizations for service delivery. It is often the dominant way to provide basic services like water distribution, waste disposal, fire brigades, public health or environmental protection. In a number of countries however, standing organizations are losing their popularity compared to cooperative arrangements that are organizationally less integrated. Less integrated forms include light associations without a staff of its own and agreements that either regulate service delivery by one public or private partner to the other partners or that coordinate the service delivery of different partners. Typical examples of light-touch associations are the inter-local and project associations in Flanders. After the introduction of these new administrative forms, within a few years a total of 69 light associations was established, partly substituting the former heavier intercommunales. They can be found in a variety of policy sectors (De Peuter and Wayenberg 2007: 26–27).

With respect to service delivery agreements we can distinguish between public-public and public-private agreements. The first category includes service agreements, where one municipality purchases the service from a neighbour or agrees to some form of exchange of services between the partners involved. In Finland this type of arrangements is becoming more popular and is replacing joint service delivery through standing organizations (Haveri and Airaksinen 2007: 49–50). In the Netherlands the so-called ‘centre municipality construction’ is quite frequently used in the delivery of social security services, the processing of grievances by a local ombudsman and the granting of environmental licenses (Pröpper et al. 2005: 19; Hulst and Van Montfort 2007b: 147–150). These arrangements are a specific manifestation of the present-day concept of ‘inter-governmental shared services’ (Korsten et al. 2004: 57–59).

The second category of agreements includes contracts that regulate the outsourcing of public services to private companies. Typical examples of inter-municipal cooperation that in the end leads to outsourcing to private companies we can find in France. There, new inter-municipal communities set up for public service delivery, transfer this responsibility to a larger syndicat mixte, which in turn contracts out the service to a private utility. In this way, it might be said that services are delivered through a complex nexus of contracts (West 2007: 87).

Institutional context: legislation and administrative culture

In some of the countries included in our sample legislation is or at some time was restrictive with respect to the institutional form of cooperation, banning certain types or imposing heavy juridical forms. Restrictions applied in the Netherlands, where the 1950 legislation with respect to cooperation stipulated that cooperative arrangements under private law could only be used in special cases. Belgium is an example of a country that imposed a heavy juridical form for inter-municipal cooperation, until the promulgation of the 2002 decree, which introduced several lighter forms. In these situations, a shift from heavy to light forms of cooperation can be directly attributed to the abolishment of restrictions or impositions that local governments find dysfunctional and to the preference of local government for simple and flexible institutions for cooperation, which we will discuss in the next section.

In most cases however, national legislation leaves local governments substantial leeway for the organization of their cooperation. Nowadays, Finnish, French, Dutch, Belgian and Italian legislation, for example, offers local government a range of different administrative forms to choose from, and there are no absolute bans on outsourcing the delivery of public services to the private sector. In the absence of legal restrictions, the shift to lighter, less bureaucratic forms of cooperation seems to be related to changes in the administrative culture. Due to increasing pressures on the performance of the state, because of globalization and European integration, the major part of Western European countries have to some extent adopted the paradigm of New Public Management. They introduced elements of marketization and have moved away from bureaucracy as the prime organizing principle for the public sector (Hughes 2003: 15). Contracting, both within the public sector and between the public and the private sector, is pervading modern Western European administrative organization (Lane 2001).

France, the UK and Finland provide good examples. Although France has a long history of providing public services via contract, nowadays long-term relationships between government and private providers are increasingly being substituted by short term competitive contracting (West 2005). Moreover, inter-governmental contracting, between the state, the regions and local government, has gradually substituted the centralistic planning procedures that were en vogue (and not very successful) before 1980. In the United Kingdom central government has combined performance budgeting with contract agreements with local governments (see section 4.1). Governance through networks and contractual relations has become a dominant characteristic of British local government (Rhodes 1996; Stoker 1998). Also Finland is one of the countries where NPM ideas are put into practice quite drastically. NPM themes as decentralisation, lightening of bureaucracy, empowerment of market mechanisms, decreased regulation and a shift from input budgeting to a stronger focus on results, have become part and parcel of the administrative culture (Haveri and Airaksinen 2007: 47–48). They have not only stimulated inter-municipal cooperation in general, but also seem reflected in the growing preference for contract like agreements.

Although it would require more detailed research to show that the fashion of New Public Management actually influences the preferences of local government for contract like arrangements and joint outsourcing, there seems to be an undeniable correlation. The countries that are considered forerunners in the modernization of public administration—the UK, the Scandinavian countries, France, the Netherlands, among others (Hood 1995)—are also moving to lighter and less integrated forms of cooperation. In Spain considered to be operating in the rearguard with respect to the adoption of New Public Management practices (Hood 1995) there is no significant shift towards the contractualization of inter-municipal (or inter-governmental) service delivery.

Local preferences

Although the New Public Management reform undeniably has elements of a fashion, and contractualization is used because it is the ‘modern way’ to govern, it also addresses some genuine concerns, motives and priorities of local governments. Even if light forms of cooperation or contracts would not be fashionable, there would be reasons to prefer it to more bureaucratic forms of cooperation.

In the first place there is the consideration of cost efficiency. Using contractual agreements, municipalities can avoid the start-up costs and costs related to the governance and management of a joint organization and still create the same economies of scale. As mentioned earlier, formal procedures for the establishment of joint organizations, joint management, and provisions for control and accountability of local councils involve additional costs. Moreover the establishment of a new organization for service delivery requires building up knowledge, skills and organizational routines. If small communities contract out service delivery to one of the municipalities in their region, they can profit from the experience and specialist knowledge built up by service delivery agencies of their big neighbour. Cost-efficiency also plays a role in inter-municipal cooperation that results in the outsourcing of public services to private firms. Local governments can evidently hope to get a better deal from private companies if they join forces and strengthen their bargaining power vis-à-vis the private sector and if they purchase large quantities of the services in question. Typically this kind of cooperation for outsourcing takes place in sectors where large scale-production yields efficiencies, such as electricity provision; drink water provision and public transport and road maintenance and repair (Wilson and Game 2006).

A second motive for using contractual arrangements instead of standing organizations for service delivery is the need for flexibility. Once standing organizations have been established, it often requires formal decision-making procedures involving all the partners in cooperation to adapt to new circumstances. Moreover, diversification, expansion or reduction of the services involved cannot be accomplished without substantial interventions in the standing organization. Contractual agreements are, on the other hand, relatively flexible, although they will generally have a fixed term and cover a number years, especially when the service rendered requires investments from the part of the producer. But adjustment is generally a simple matter of agreement between the parties involved. This flexibility makes them appropriate instruments in the case of new services, tasks or projects with a temporary character or services that have to be rendered in a dynamic environment. Typically in Finland the contractual arrangements are in use for projects and services with a temporary character (Haveri and Airaksinen 2007: 50).

The case of Flanders illustrates the importance local government attributes to flexible cooperative arrangements. As mentioned earlier, recent legislation introduced administrative forms of cooperation that were lighter than the existing intercommunales. Local government has eagerly taken up the opportunity to organize their cooperation in less integrated arrangements. But an evaluation of the implementation of the new legislation shows that municipalities are not at all satisfied with the new administrative forms, because they still have to follow all kinds of rules and regulations for their establishment and modification. They feel that they can’t tailor cooperation to their specific needs and that they face considerable costs if they want to expand or intensify cooperation or adapt it to new circumstances (De Peuter and Wayenberg 2007: 31–32).

Conclusions

Inter-municipal cooperation is a widespread phenomenon throughout the Western world. Our inventory research of the institutional arrangements for cooperation in eight European countries not only shows that these arrangements vary with respect to the scope of cooperation, their composition and organizational integration, but it also brings to the fore that shifts occur with respect to the dominant type of institutional arrangements. In this article we have tried to unravel the factors that can explain the existence of different forms of inter-municipal cooperation and of shifts between them; shifts between single-purpose and multi-purpose arrangements, between horizontal and vertical forms of cooperation and between standing bureaucratic organizations and contractual agreements. But explanations cannot be reduced to simple factors. It is the interaction between external factors, the institutional context and the preference structures of local government that in the end determines the pattern of cooperation and the shifts therein. In the previous sections we have focussed on the individual factors with occasional attention for their interaction. In this concluding section we will return to the complex interrelations between the different factors.

With respect to the scope of cooperation it seems clear that the national legislation and incentive structures play a prominent role. If these prescribe or favour one administrative form—a single-purpose or a multi-purpose arrangement—over the other, it will have an immediate impact on the patterns of cooperation. On the other hand, our research shows that if external, e.g. political factors induce local governments to put high value on cost efficiency or on arrangements that create the optimal scale of service delivery, they tend to mould legislation to their own interests or circumvent it. Thus we can explain why in the Netherlands, long after the Joint Provisions Act of 1985 imposed the use of multi-purpose arrangements, a large number of single-purpose arrangements survived. Local government used all available legal loopholes to keep single-purpose arrangements alive because these enabled them to cooperate with varying groups of municipalities and thus provide single services on different scales. It also explains why in Flanders we sometimes find a series of formally single-purpose arrangements, as required by law, which de facto are one and the same multi-purpose institution with a common staff and board. This way the local governments reduce management costs and create better conditions for synergy (De Peuter and Wayenberg 2007: 32.

If national legislation leaves it to the municipalities themselves to organize their cooperation in single or multi-purpose arrangements, local government will generally seek arrangements that best meet the actual local preferences. But elements of the institutional context or environmental factors can alter the preference structure. Cost-efficiency may become very important for microscopic local government that is confronted with a sudden expansion of the local service domain and lead to the predominance of multi-purpose organizations (the Spanish case after the 1978 Constitution). On the other hand, a change in the political culture emphasizing the importance of local control over cooperative arrangements can eventually lead to the dismantling of multi-purpose organizations, of which some examples are present in the recent Dutch practice.

With respect to the composition of cooperative arrangements elements of the formal state structure set the stage. The large scale and the strong financial dependency of local government in the United Kingdom can explain the near absence of purely-inter municipal cooperation until now, as well as the fact that cooperation is predominantly vertical in character. But other elements of the institutional context, i.e. pressures from central government to improve efficiency, and external factors such as the growing economies of scale of IT and communication services, have actually induced local governments to set up inter-municipal call centres and to collaborate in revenue collection and benefits pay (Kelly 2007: 208).

In the same way the organizational integration of cooperative arrangements and the shifts that can be observed therein, are the outcome of a combination of elements of the institutional context, local preferences and external factors. In most European countries included in the research, inter-municipal cooperation with respect to service delivery took the form of standing public organizations governed by the participating municipalities. A more dynamic environment, a revaluation of the private sector as a possible producer of public services and more emphasis on value for money have resulted in a growing popularity of contract like arrangements between municipalities and in the (joint) outsourcing to the private sector. Then again, we may not preclude that in some European countries the tables may turn again, if public accountability and control move up in the preference structure of local government.

We have summarized the different factors and their interrelationship in Fig. 1. We think that the case for the power of our model to explain the presence and shifts in the patterns of inter-municipal cooperation has been sufficiently made. At the same time we do not pretend to have identified all relevant external factors and elements of the institutional context or to have unravelled all interrelationships between the different factors. This will evidently require (a lot) more research.

Fig. 1
figure 1

Explaining the institutional design of cooperative arrangements

Full size image