Why such a different choice of tools? Analysing recent local government reforms in Denmark and Norway

ABSTRACT

Local government amalgamation reforms are politically demanding ventures because potential benefits are often diffuse and long term, while costs are concentrated and immediate. We investigate the role of national political actors in forming alliances and choosing policy tools in such demanding reform contexts. Empirically, we compare the Danish amalgamation reform, characterised by the use of authoritative government tools and a nationally directed amalgamation process, and the Norwegian reform, which primarily used softer tools that involved substantial autonomy at the local level. Our analysis is built on a rich set of qualitative data. We show that differences in the pro-reform alliances established by the two national governments help explain the different choices of government tools for carrying out the local government reforms. A strong pro-reform alliance, as was the case in Denmark, lent legitimacy to the use of authoritative tools.

KEYWORDS:

• Amalgamation reform
• government tools
• policy choice
• coalition building
• territorial reform
• tools mix

Introduction

Administrative reforms are usually unpopular among voters and taxpayers because they offer no immediate gains and they are difficult to implement because they affect the distribution of political and economic benefits (Christiansen and Klitgaard 2010; Aberback and Christensen 2014). Reforming local government structures is the ultimate example of changes that entail immediate costs for numerous stakeholders in return for long-term, diffuse and uncertain benefits. Undeniably, merging municipalities is a highly politicised and controversial answer to the widely recognised need to modernise administrative arrangements in today’s welfare states (Baldersheim and Rose 2010).

Government tools are the foundation of public institutions and refer to governments’ capacity to ‘get things done’ (Hood 1983; Salamon 2002; Acciai and Capano 2020). Focusing on government or policy tools helps us understand how problems and solutions become connected in the actual policymaking processes (Béland and Howlett 2016). Acknowledging the significance of tools, we ask why two national governments with similar reform ambitions and political-institutional settings ended up choosing very different tools to carry out their amalgamation reforms. More specifically, we studied the Danish and Norwegian amalgamation reforms in 2002–2007 and 2014–2017, respectively. In both countries, the reforms were initiated due to a perceived need to form larger, more robust municipalities via large-scale amalgamations. In Denmark, the aim of the reform was to modernise the public sector and increase its effectiveness by establishing larger, stronger and more sustainable municipalities (Regeringen 2004). Similarly, the Norwegian government’s initial motivation was to form ‘sustainable and economically robust municipalities’ (Prop. 95 S 2013,2014, 29).

Despite similar reform ambitions, the actual policy output reduced the number of municipalities by 64% in Denmark and by 17% in Norway. The differences in policy output can be partially explained by the fact that amalgamation processes were authoritative in Denmark and more voluntary involving softer government tools in Norway (e.g., Baldersheim and Rose 2010; Askim et al. 2017; Kjærgaard et al. 2020). However, we pose the following questions: Why did the two national governments choose such different government tools to begin with? Seeing that contemporary approaches to intergovernmental relations have increasingly emphasised the importance of bargaining, compromising and networking (Salamon 2002; Radin 2012), why did the Danes choose the traditional command-and-control approach that accentuates the authority of the national government? Correspondingly, why did the Norwegian government not do the same?

While deducing the relationship between implemented tools and reform outputs in amalgamation reforms is relatively straightforward, it is more difficult to understand why specific tools were chosen in the first place. Such choice partly depends on environmental factors and the institutional context and is partly related to individual preferences and goal achievements (Capano and Lippi 2017). When framing our analysis, we will – in accordance with Baldersheim and Rose’s (2010) modelling of the politics of territorial choice – combine insights from different literatures in political science to capture the complexity of the relevant variables. The dependent variable, government tools, will be described according to the basic classification of government or policy tools (Hood and Margetts 2007; Howlett 2011). As different tools prioritise effectiveness, efficiency, equity, manageability and legitimacy differently, the choice of tools is inevitably political and requires state-craftsmanship (Anderson 1971; Salamon 2002). To analyse the mastery involved in choosing suitable tools for highly politicised undertakings, such as amalgamation reforms, we will concentrate on the pro-reform alliances formed by central government actors during the preparatory process. In addition to focal insights from the tools literature, we will draw on elements of the Advocacy Coalition Framework (ACF), which emphasises coalition building as a distinct aspect of contentious policy processes (Sabatier 1998; Pierce et al. 2017). Our argument is that the tools devised by the Danish and Norwegian national governments depended not only on the available resources but also on how the two governments used these resources to form pro-reform alliances. More specifically, we believe that the comparative strength of the pro-reform alliances accounts for the tools that the two national governments had at their disposal for implementing the amalgamation reforms.

The vast distances and large, sparsely populated areas in Norway make the idea of larger and similarly sized municipal entities across the country less viable than in the smaller, more densely populated Denmark. We also know that centre-periphery cleavages are more predominant in Norway than Denmark (Hansen and Kosiara-Pedersen 2018). However, these clear differences do not explain why the two countries’ national governments decided to use very different tools in their amalgamation reforms. Theoretically, the Norwegian government could have accounted for geographical variation and distinguished between the tools chosen for central and peripheral municipalities while still authoritatively steering the reform process using compulsory means. Thus, it remains a puzzle why two relatively similar countries with relatively similar political reform intentions ended up choosing such different government tools. We investigated this issue empirically by using a rich set of qualitative data that included government documents and an extensive number of elite interviews with national-level political and administrative actors from both countries.

The article is organised as follows: First, we consider our theoretical approach and supplement the literature on government or policy tools with analytical insights from the literature on the ACF. Then, we present our comparative method and data and analyse how the two governments selected their respective tools during the initial phases of the reform processes. Finally, we conclude by summarising and discussing our findings.

Theoretical approaches

Amalgamation reforms, like most public sector reforms, are complex endeavours (Pollitt and Bouckaert 2017). Multi-level reforms in which national and local units act interdependently are particularly complex (Howlett and Del Río 2015; Howlett, Vince, and Del Río 2017; Weible and Carter 2017). To carry out such reforms, a mix of government or policy tools is needed. Among the many categorisations of tools, some approaches deal with intergovernmental relations specifically (Radin 2012; Acciai and Capano 2021). However, according to Hood (1983; Hood and Margetts 2007), the tools concept is generic, and his so-called NATO scheme, which identifies the four basic resources that a government may have, is a good fit for our analyses of local government reforms. In the context of amalgamation reforms, nodality denotes the informational tools based on the resource governments have due to their central position in initiating and implementing reforms. Authority refers to governments’ legal basis for regulatory and coercive actions, which involves the use of legal regulations that force certain behaviours in the reform process. Treasure enables governing via financial means – that is, using available financial instruments and relying on incentives to exhort municipalities to amalgamate. Finally, organisation represents governments’ ability to use force to act directly. Taking these four types of resource as our point of departure, we will show that in Denmark, the reform tools involved the use of force by the national government, while in Norway, the tools were considerably softer, which made the reform voluntary.

Explaining why specific kinds of tools are chosen is the central theoretical issue in contemporary literature on government or policy tools (Capano and Lippi 2017). Traditionally, the literature has been dominated by institutional explanations, such as available resources, organisational capacity and the characteristics of the political system, such as the decision-maker’s national policy style and organisational setting (e.g., Linder and Peters 1989; Schneider and Ingram 1990; Hood and Margetts 2007; Howlett 2011). However, in the two studied cases, the issue is that governmental actors operating in relatively similar institutional settings, with comparable political preferences, and pursuing the same kind of outcome made different decisions regarding the tools for implementing the local government reform. Knowing that the selection of government tools depends on the reciprocal influence between decision makers’ preferences and contextual constraints (Capano and Lippi 2017), actor-based explanations are needed to understand differences in tools choice.

Policy is formulated by means of the designed instruments, and the choice of tools is inherently political (Salamon 2002; Hood and Margetts 2007; Howlett 2011). In the classic texts on policy measures by Anderson (1971), the practice of government is described as ‘the art of the possible’ and governments’ choice of tools is a matter of ‘statecraft’. Without specifying the elements of this state-craftsmanship, tool choice is defined by Anderson as the art of making choices based on the possibilities available at a specific time and in a specific context (see also Capano and Lippi 2017). Our aim is to show that the differences in the tools used in the two amalgamation reforms can be explained by the differences in the strength of the pro-reform alliances formed by the two national governments. Due to the need to disentangle policy subsystems in this early phase of the policy process, in which tools are developed and advocated (Béland and Howlett 2016), we will draw on some analytical elements offered by the Advocacy Coalition Framework (ACF) to structure our analysis. According to the ACF, advocacy coalitions denote groups of policy actors that, over time, share core policy beliefs, coordinate their behaviour to influence the policy process, and strategically exploit their resources to change existing policies (Sabatier 1998; Nohrstedt 2011; Pierce et al. 2017; Jenkins-Smith et al. 2018). The actors we investigated did not share a common belief system consisting of basic values, causal assumptions and problem perceptions as assumed in the ACF. Rather, the alliances we investigated were ad-hoc. However, seeing that coordinated behaviour is the defining feature of coalitions (Sabatier 1998; Pierce et al. 2017), we investigate how governmental reform owners and opposing actors identified allies and formed influential alliances.

According to the ACF, besides beliefs, essential elements in coalition formation include perceived influence and others’ resources, interests, and trust (Jenkins-Smith et al. 2018). Based on this insight, we investigate the capacity that alliances have to engage in various activities and achieve influence over the choice of tools depending on available resources. Such resources may consist of political resources, such as public opinion, mobilising capacity and political leadership, and/or formal-legal resources, such as decision-making authority, informational advantages and available financial means (Weible 2006; Weible et al. 2011).

In addition, we know from the literature on administrative reform that institutional leadership plays a facilitative role during the process by intervening at specific moments to make critical decisions to implement reforms (Boin and Christensen 2008). This kind of leadership, known as policy-brokerage in the ACF literature and as policy entrepreneurship in the agenda-setting literature, is important for steering alliances towards policy victories as well as attracting additional resources and garnering support for a policy. Such leadership may be performed by elected politicians and civil servants and involves channelling information between stakeholders and coalitions supporting a policy (Kingdon [1984] 2014; Sabatier 1998; Jenkins-Smith et al. 2018).

To summarise our theoretical starting points, we expect the interplay between actors and available political and formal-legal resources to condition the strength of the pro-reform alliances in Denmark and Norway – that is, the strength of the pro-reform alliances is partly decided by the resources available and partly by how these resources are used and political leadership is performed by the actors involved. Depending on the strength of the pro-reform alliances, they have different bases for choosing government tools. It is likely that only a strong pro-reform alliance can legitimise the use of authoritative tools. Therefore, the assumption is that the pro-reform alliance in Denmark was stronger than in Norway, which is why the national government in Denmark chose more authoritative tools for steering the merger process compared to the Norwegian government.

Methods and data

Our empirical analysis was designed as a cross-case comparison based on the logic of the ‘most similar system’ design (Gerring 2007). Denmark and Norway are two Nordic countries with relatively similar political-institutional systems. Both countries have implemented ambitious amalgamation reforms, but via two very different reform policies. Although the institutional contexts in Denmark and Norway are not completely identical, the two countries are ideal for a controlled comparison because the differences in the two governments’ choices of tools for implementing the amalgamation reforms cannot be caused by basic constitutional/institutional variables (such as differences in welfare systems, national governments, ministerial rules or local government systems) but must be the result of differences in pro-reform alliances and their use of available resources.

Indeed, Denmark and Norway are unitary, parliamentary and multi-party states with relatively similar welfare systems (Esping-Andersen 1990) that belong to the Western Nordic administrative tradition characterised by ministerial rule (monism) (Lægreid 2018). Both countries have a three-tier government system with elected bodies at all levels: municipality, county/region and nation. The two countries represent the Nordic type of local government system, which involves a high level of decentralisation via multi-purpose entities that enjoy significant levels of local autonomy (Ladner, Keuffer, and Baldersheim 2016). Both countries represent consensus systems, which makes gradualist reforms more likely (Pollitt and Bouckaert 2017). One institutional difference is noteworthy, however, as associations of local and regional governments are separated in Denmark but form one single organisation in Norway.

Our empirical basis consisted of a rich set of qualitative data based on government documents and elite interviews with national-level political and administrative actors from both countries. For Denmark, secondary data based on documents and elite interviews were available from former studies (including Christiansen and Klitgaard 2008, 2010; Blom-Hansen et al. 2012). For Norway, our analysis was based on primary data from available documents and 13 elite interviews carried out in 2017–2018. The authors manually coded the data according to the distinction made in the theoretical part, between political and formal-legal resources emphasising the ways in which these resources were used by involved actors to build pro-reform alliances.

The tools used in the two reforms

The Danish reform was based on two initiatives launched simultaneously by the national government: amalgamations and a reshuffling of tasks between the three government tiers. The latter included substantial decentralisation of tasks from the regional to the local level of government (Christiansen and Klitgaard 2010). Regarding governments’ use of tools, the Danish amalgamation reform was directed via strict political guidelines proposed by the national government. The goal was clear: new municipalities with preferably 30,000 inhabitants and at least 20,000 inhabitants. The national government relied on its authority and left very little autonomy to local governments in deciding whether to amalgamate or not. Rather, an ultimatum was given to find partners to merge with to fulfil the inhabitant requirement. While local governments could choose whom to merge with, the national government made it clear that should local negotiations fail, it would impose the necessary amalgamations (Christiansen and Klitgaard 2008, 152–154). The national government engaged in organisation by appointing a mediator, but it did so only in the few cases when the agreements suggested by the local governments did not comply with national guidelines (ibid.). Informational tools were used in the form of a major media-covered presentation of an extensive Commission Report a number of months prior to announcing the national government’s specific reform ambitions. The national government did not use any direct economic incentives. The merging municipalities were responsible for the amalgamation costs; however, according to the government, these expenses were fully compensated for by the economies of scale enabled by the amalgamations.

In the Norwegian reform, the parliament decided in 2014 to start the process of changing the territorial structure, with the transfer of new responsibilities to larger municipalities being decided a year later. The national government used its authority to impose the process upon local governments, making it obligatory to investigate the potential for merging with neighbouring municipalities and, on this basis, to make an explicit decision in the local council on whether to merge. The central government reserved the right to force locally disputed amalgamations if such amalgamations were desirable according to national or regional needs. The threat was politically indecisive and perceived by local governments to be rather vague. However, an important part of the government’s reform toolkit was information about the advantages of larger municipalities. The main arguments referred to improved professional capacities and competences, higher quality of services and stronger local democracy due to the lesser need for strong supervision from the central government (Prop. 95 S 2013,–2014). Contrary to the Danish case, the Norwegian government actively used economic incentives to facilitate voluntary amalgamations. All direct costs connected to the amalgamation process at the local level were covered by central-government grants. For a substantial period of time, the municipalities that wanted to amalgamate also received compensations for the central government grants they would lose due to increased size (covering certain expensive costs in small municipalities). Most importantly, the municipal income system was changed, with reduction in basic grants for municipalities that could have reduced their administrative costs by merging but decided not to (the so-called voluntary small municipalities) (Klausen, Askim, and Vabo 2016). An important feature of the Norwegian amalgamation reform involved the use of county governors as the government’s own organisational resources at the local level. The county governors’ main tasks were to control and supervise sub-national governments and to serve as a sort of liaison between the two government levels. Although they had a formal role on behalf of and reported to the national government, the county governors were not directly instructed to push for amalgamations by the Ministry of Local Government and Modernisation. The role that county governors played, therefore, did not represent the use of force by the national government (see also Klausen, Askim, and Christensen 2021).

Norway and Denmark used a different repertoire of government tools. In Norway, local government capacity was enhanced by information, economical incentives and country governors’ supervision of the amalgamation process. While the Danish reformers relied on authoritative tools in the reform process, the Norwegian government invoked the use of force as a threat to prevent the stalling of local processes. In Denmark, the authoritative tool, supported by the use of information to underline the need for reform, worked as an imposition, leaving the local governments with no choice. By contrast, in Norway, authoritative tools were used only for setting the general agenda of the reform process – that is, to amalgamate municipalities – while, in practice, leaving it up to the municipalities whether to amalgamate or not.

Explaining the different uses of tools

In Denmark the government tools revolved around a nationally steered merger process, while the Norwegian reform was implemented as a significantly less governmentally steered endeavour. Our argument is that this difference in the tools used to implement the reform may be explained by differences in the governments’ pro-reform alliances. In the following section, we describe the actual alliances established in each country. Then, to explain the different tools used by the two governments, we examine how central actors used political and formal-legal resources to form alliances.

The different pro-reform alliances

In both countries, the reforms were initiated by minority coalition governments. In Denmark, the amalgamation reform was not part of the programmeproposed by the coalition government of 2001 formed by the Liberals and Conservatives. The reform proposal was prepared by a small group that consisted of top ministers, top civil servants, a top representative from the Danish People’s Party (the minority government’s parliamentary partner), and the CEO of the local government association (Local Government Denmark, KL). Other interest groups, such as the regional government association (Association of County Councils, DR), were not included and received no information regarding the government’s intentions (Christiansen and Klitgaard 2008, 107–117). With Denmark being a political system based on consensus, it is generally expected that the two large parties in Danish politics, the Social Democrats and the Liberals, should agree on issues involving substantial policy changes. However, this was not the case with the decision to reform the country’s territorial structure (Christiansen and Klitgaard 2008, 116–133). The formation of the diverse pro-reform alliance was surprising, as the agrarian Liberals and the Social Democrats were expected to cooperate against the amalgamation reform due to their belief in the democratic virtues of small municipalities and the local strongholds for the two parties. The pro-reform alliance was headed by the Liberal Party (the prime minister’s own party), with the Social Democrats and the DR opposing the reform (ibid.).

In Norway, the two parties that formed the minority coalition government in the fall of 2013, the Conservative Party and the Progress Party, both argued in favour of larger municipalities in their party programmes and election manifestoes. The need for a reform was also included in the government agreement with the Christian Democratic Party and the Liberals. Immediately after the government took office, they organised a meeting with all party leaders in the parliament for consultation on the reform plans. However, none of them were formally invited to participate in further discussions and binding agreements before the reform proposal reached the parliament more than half a year later. The same was true for the Norwegian Association of Local and Regional Authorities (KS). In contrast to the KL in Denmark – and in spite of the fact that KS saw the amalgamation reform as a viable solution to the scale-related governance problems (see decision at the Convention in 2012) – KS was not included in the pro-reform alliance.

As illustrated in Figure 1, the actors backing the two governments in the process of introducing the reform tools were quite different. While the Danish minority government formed an alliance with pivotal parties in parliament as well as with the local government association, the pro-reform alliance in Norway was limited and fragile. Moreover, in Norway, an active alliance against the reform’s content was formed. Although the Labour Party had acknowledged the need for a reform in its own manifesto before the 2013 election and had initially supported the government’s ideas and plans, it ended up opposing the reform initiative. By taking a clear position against the reform initiative, the Labour Party formed an alliance with the agrarian Centre Party, which was, and always has been, against amalgamations. The two parties joined forces in opposing the reform, together with one of the trade unions, Fagforbundet, which represented the majority of the local government employees.

Figure 1. The alliances formed (alliances are illustrated by circles; significant engagements are emphasised in bold).

The use of political and formal-legal resources

As discussed in the theoretical section, political and formal-legal resources are essential in forming alliances to support a reform. Below, we examine how the two national governments used their potential resources to establish pro-reform alliances and show why the differences in these pro-reform alliances can explain the differences in the government tools employed in the two reform processes.

Integration of political parties as a political resource

The multi-level character of the political parties is a characteristic feature of the relationship between central and local governments in the Nordic countries (Sellers, Lidström, and Bae 2020). For governments, having allies in the local branches of their party or parties constitutes a potential resource, and we found that political parties’ integration at and ties to the local level were essential for determining the strength of the pro-reform alliances in the initial reform process, when the government tools were chosen.

In Denmark, the minority government actively integrated the Danish People’s Party in the reform preparation process and thus worked strategically to ensure a pro-reform majority in parliament. In addition, the largest government party initiating the reform, the Liberals, generally had strong support in the municipalities, and at the time of the reform, the party had the largest number of mayors in office. To ‘soften’ the party colleagues at the local level, the Minister of the Interior held several meetings with and visited affected municipalities to reduce potential conflict around and resistance to the reform (Christiansen and Klitgaard 2008, 103–105).

In Norway, the minority government launched the reform based on the government agreement with the Christian Democratic Party and the Liberals. None of these parties were particularly keen on the reform; consequently, they were not particularly strong allies for the government. Furthermore, the parties that held most mayor posts – the Centre Party and the Labour Party (Hansen 2014) – were not included in the pro-reform alliance. Neither the Conservative Party nor the Progress Party was particularly strong at the local level, and, therefore, none of the political parties responsible for the reform were capable of bridging the local and central interests regarding the amalgamation reform. The lack of mediating power meant that local governments could be easily provoked by the use of authoritative government tools. This was likely one of the reasons behind the Norwegian government’s reluctance to use force. In addition, although the Labour Party had pursued an amalgamation reform (in 1995–1996) when in power, it turned against the 2014 reform as the process unfolded. The Labour Party has historically been a pro-amalgamation party in the national parliament. However, there was internal opposition within the Labour Party from the mayors and local party members who did not want amalgamations in their geographical areas.

Although local versions of parties can be very different from the national ones in both countries, the institutional tradition of local self-government within the parties seems to have been stronger in Norway than Denmark (Ringkjøb 2004) and created a gap between the pro-reform alliance at the national and the local level. At the same time, in Denmark, a mutual understanding of the need for reform across government levels could have made the use of force unnecessary. In any case, the solid bridge between levels probably placed the national government in a position to use authority when choosing reform tools.

Local government associations as a political resource

In both Denmark and Norway, local governments form interest associations, which aim to influence relevant national policies (KL and KS, respectively). In contrast to the Norwegian reform, the Danish government acted strategically to turn the KL into an ally. In the Danish case, there was a strong, integrated and enduring relationship between the national authorities and local (KL) and regional (DR) government associations. The two aforementioned associations were involved in most decisions that impacted the sub-national levels (Blom-Hansen et al. 2012; Christiansen and Klitgaard 2010). The government had initially left its preferred conclusion unclear and sought to avoid provoking local governments by refraining from suggesting specific criteria for amalgamations or changes in task responsibilities. Behind the ‘veil of vagueness’, however, central government politicians continually discussed the reform with single local governments and the KL (Christiansen and Klitgaard 2010). The decision of the Minister of the Interior, Lars Løkke Rasmussen, to draft the reform in a narrow group, which included the KL and excluded the DR, was essential for gaining support for the reform. As discussions on the further decentralisation of tasks formed an integral part of the reform, the minister secured broad support from the KL and from many local politicians and mayors by offering the municipalities significant new responsibilities (Blom-Hansen et al. 2012). The weak solidarity between municipalities and counties is explained by the fact that the two kinds of sub-national authorities are organised into separate associations in Denmark. Both politically and publicly, the counties had few defenders, and the KL, as part of the pro-reform alliance, contributed to the diminishing of county responsibilities (Christiansen and Klitgaard 2008, 112–113).

This differences in institutional structure thus meant that strategic alliance building by playing the two sub-national government associations against each other was a viable option only in Denmark. In contrast to the position held by the KL in Denmark, the KS’s influence on central government decisions was more restricted. Formal contact between the central authorities and the KS ensured dialogue and information exchange, but KS were not invited to further participate in the reform process. Soon after taking office, the minister of Local Government and Modernisation presented the reform plans to the director of the KS. The KS accepted the plans, expressed the wish to be involved in the process and encouraged the minister to work for a broad reform compromise in parliament. KS was an active partner in an unsuccessful attempt to introduce voluntary municipal amalgamations in 2003–2005. The experiences of this unsuccessful attempt may explain the caution towards a renewed partnership both on the part of the national government and the KS.

Again, the local government association as a political resource was available to the Norwegian government in quite a different way than was the case in Denmark. While in Denmark the separation between KL and DR made an alliance with the KL possible for strengthening the pro-reform alliance, there were no signs of active alliance-building efforts in the initial phases of the Norwegian reform process. Due to its former experiences of national reform efforts, the KS was not eager to engage in close cooperation with the government. Even if the Norwegian government could have successfully built such an alliance, the relatively weak position held by the KS would have lent only limited legitimacy to the use of authority and the choice of coercive government tools.

Former amalgamations as a formal-legal resource

Regarding the formal-legal resources possessed by the two governments, Denmark and Norway are both unitary states in which the parliament decides the structure of the local government. Historically, amalgamations were forced through by the Norwegian parliament in an extensive reform in the 1960s and again in the 1980s, when the municipal borders of several cities were expanded (Klausen, Askim, and Vabo 2016). However, a parliament decision in 1995 established the principle of voluntariness. Since then, only locally initiated amalgamations had been approved, and the parliament had been reluctant to intervene in structural local government issues. In Denmark, on the contrary, some voluntary amalgamations had taken place since the last national reform in the 1970s, but the reluctance to intervene in local government structures had not been present to the same extent in the Danish parliament as in Norway. Therefore, due to different historical trajectories, there was a significant difference in the two governments’ explicit will to use authoritative tools to force amalgamations.

Decision timing as a formal-legal resource

Similar to the local government structure, the national government in both Denmark and Norway possesses the formal authority to decide on the level of decentralisation between government tiers. In Denmark, a reshuffling of tasks between the three government tiers formed an integral part of the policy decisions regarding the reform and was used strategically by the pro-reform alliance to compensate local governments for merging, which was important for local politicians and the KL (Blom-Hansen et al. 2012). In Norway, the government was reluctant to define new tasks for the larger municipalities. The discussion took place at a rather abstract level, and the decision was to be made after the territorial structure had been decided on. Although timing was a formal-legal resource that could have been exploited by the Norwegian government, no incentives in the form of significant tasks were offered to make the amalgamations more desirable for local governments.

The reason why new tasks were not assigned to the local level during the initial phase of the reform may be explained by a lack of reform enthusiasm among the sector-specific ministers in the Norwegian government. The fear of losing sector-specific responsibilities may explain the ministers’ reluctance to embrace the reform. A further reason why the Minister of Local Government and Modernisation struggled to garner support for the amalgamation reform among the ministers may be that several of them were promoting their own ministries’ administrative reforms at the time and arguing for larger entities in higher education, health care and the police force. Logically, there were obvious links between the amalgamation reform and these other reforms, which were also concerned with administrative scaling and territories. The reason to avoid uniting the ministers by making such links explicit and in the reform process may be attributable to a strategic intent to downplay the common centralising tendency that characterised the reforms. Moreover, the Norwegian reform was characterised not only by a lack of support from the ministers but also from the prime minister. While the reform was actively backed by Anders Fogh Rasmussen, the Danish prime minister, his Norwegian counterpart, Erna Solberg, promoted the reform’s potential gains without expressing strong commitment. This may also have been due to the need to not draw attention to the administrative centralisation tendency inherent in the other major reforms put forward by the government of 2013.

Thus, while in Denmark the Minister of the Interior played the ‘new tasks’ card to strengthen the pro-reform alliance and gather support for the use of authoritative government tools, the reluctance among the other ministers to participate in the pro-reform alliance in Norway meant that the minister of Local Government and Modernisation could not use this resource. The government tools chosen in the Danish case to force amalgamations did not provoke most local actors, while the basis for forcing amalgamations was missing in the Norwegian case. Even within the Norwegian government, the fact that the amalgamation reform had to share political attention with other major administrative reforms intervening in local governments likely made it difficult to secure support for authoritative and coercive tools in the local government reform.

Discussion and conclusion

To better understand the Danish (2002–2007) and Norwegian (2014–2017) amalgamation reforms, we set out to explain why national governments in relatively similar political-institutional contexts and with relatively similar perceptions of the need for a territorial reform chose very different tools for implementing these local government reforms. In the Danish case, the reform policy consisted of two elements: amalgamations and new tasks for the new local governments. After the decision was made in the parliament, the process was steered at the national level, a minimum threshold for municipality size was introduced and local negotiations about borders were organised using this authoritatively decided standard. In the Norwegian case, the issue regarding the decentralisation of tasks was resolved a year after the local processes had started. The tools used emphasised voluntariness. Minimum size requirements were not implemented by the national government to direct local considerations, and the reform process became highly decentralised, with local governments deciding whether to amalgamate.

Geographically, Norway and Denmark differ in ways that are likely to have had a significant impact on the solution to the scale-related problem presented by the national governments. However, this do not explain why the Danish governments decided to force structural changes upon local governments, while the Norwegian government did not express a clear goal and did not authoritatively steer the reform process. Our analysis confirmed that the use of authoritative tools was significantly influenced by the comparative strength of the pro-reform alliances that the Danish national government was able to establish. Data showed that differences in the possession and use of political and formal-legal resources by the two national governments resulted in different opportunities for building pro-reform alliances.

Sufficient support for the reform likely made it possible for the Danish government to adopt authoritative government tools and forcefully implement the amalgamations. Defining the integration of political parties as one of the political resources available for the two national governments, we found it plausible that the solid bridge between the reform-initiating Liberal party (together with the Conservative Party) and the local party branches made it possible for the Danish government to use authoritative tools to forcefully implement the territorial restructuring of local governments. Furthermore, it was especially fortunate that the prime minister and the Minister of the Interior represented the Liberal party, which was traditionally opposed to amalgamations. By visiting municipalities and increasing the support for change before the reform was launched, the Minister diminished local resistance and reduced opposition to the use of national regulations. Thus, as a political leader, the Minister of the Interior acted as a policy broker, or entrepreneur, in building the reform. Although in Norway the national parties also have local branches, the two reform-initiating parties, the Conservative Party and the Progress Party, did not have a particularly strong position at the local level. Consequently, they had little chance of winning over opposing local governments, which often had mayors from the Centre Party or the Labour Party. For the Norwegian government, it was impossible to use the integration of political parties between levels as a political resource, and the highly contested understanding of the need for a reform likely restricted the possibility of using authoritative tools.

Likewise, investigating local government associations as a potential political resource, we found that the Danish government played their cards very well in making the KL their ally. It is likely that the shared understanding between the national government and the KL, fostered via close contact before the reform was launched, contributed to legitimising the use of authoritative government tools. In the Norwegian case, however, the KS was not directly involved and expressed no support whatsoever for forcefully implementing amalgamations. Due to earlier experiences, the KS was not eager to contribute, and the national government did not put much effort into making the KS an ally. Even if the government had done so, as the local government association in Norway is considerably weaker than in Denmark, an alliance with the KS would probably not have helped much in legitimising the use of authoritative tools.

Turning to the formal-legal resources possessed by the national governments, Denmark had an advantage regarding the possibility of using authoritative tools due to the parliament’s history of forcefully implementing amalgamations. While the Norwegian parliament agreed on the principle of voluntariness in 1995, there was no such principal against the use of authoritative government tools in Denmark. While history is hard to change, the timing of decisions regarding new responsibilities was another formal-legal resource that could have been used by both governments to increase potential support for the reforms. However, while the minister responsible for the amalgamation reform in Denmark used the opportunity to strengthen local governments by offering them new tasks as the territorial changes were being introduced, this was not the case in Norway. Here, the possibility of transferring new tasks to local governments was likely hindered by the fact that the amalgamation reform had to share political attention with other reforms affecting local governments. Without new tasks to offer, the arguments for why municipalities should merge were weakened. Once again, the result was that the authoritative government tools chosen in the Danish case were unlikely to be seen as controversial by local actors, while the same basis for forcefully implementing amalgamations was lacking in the Norwegian case.

To summarise, it is likely that the use of authority to gain strict control over the amalgamation process in Denmark was possible because the national government used available political and formal-legal resources in a strategic way to form a strong pro-reform alliance. This alliance legitimised the use of authoritative tools to implement the reform. In Norway, on the contrary, the same political and formal-legal resources were not available, and, potentially because of that, we did not find signs of strategic leadership behaviour aimed at establishing a broad pro-reform alliance to strengthen the reform idea among local governments. Without such support, it was likely impossible for the Norwegian government to even think of using authoritative tools to forcefully implement amalgamations in local governments.

In this article, we have investigated highly politicised local government reforms by focusing on how actors formed pro-reform alliances because such alliances represent a basis for the kinds of tools that governments can use. Using our theoretical and comparative approach, we have contributed to the literature on local government amalgamation reforms by offering new empirical insights. Though we study the choice of policy tools, the findings indicatively touch upon the adjacent question of the effects of policy tools and the wider question of viability of public sector reforms. We know from the reform literature that major reforms often fail or disappoint because problems are complex, which makes many solutions possible and thereby open up for conflicts both over how to define problems and how to solve them (Aberback and Christensen 2014). However, our study suggests that with a strategic use of available policy instruments also complex reforms like local government reforms might be implemented (see also Blom-Hansen et al. 2012; Christiansen and Klitgaard 2010). To better understand the success and failure of reforms, the kind of actor-oriented framework discussed in this article should be further developed to explain what reform-initiating actors can achieve in given contexts – including studies focusing more explicitly on the effects of various reform strategies.

In terms of theory, the article contributes to the literature on the choice of policy instruments. While many recent studies have attempted to conceptually grasp how policy instruments are chosen or formulated by policymakers, there has been very little empirical research on the subject so far (Capano and Howlett 2020). Combining different theoretical frameworks from the literature on public policy, our article contributes to the tools literature by offering an actor-based framework for explaining why reform-initiating national governments are restricted in the kinds of tools that they may use when introducing politically demanding reforms, such as local government amalgamation reforms. Among the few empirical studies of policy choice, there have been recent examples of scholars explaining the selection of certain government tools in fighting the COVID-19 pandemic based on theoretical arguments from various literatures – for example, on national governments’ state capacity, the logic of appropriateness, policy styles and political games (Capano et al. 2020). However, though the choice of tools is a vital part of the policy-making process, further empirical studies are needed to investigate different theoretical approaches for explaining the use of certain (mixes of) government or policy tools.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The work reported here was funded by The Research Council of Norway under the Programme on Democratic and Effective Governance, Planning and Public Administration (DEMOS), grant [255111].

Notes on contributors

Signy Irene Vabo

Signy Irene Vabo is Professor of Political Science at the University of Oslo Department of Political Science, Norway, where she has special responsibility for local government studies. Her research interests include local government and governance, local democracy, and comparative studies.

Anne Lise Fimreite

Anne Lise Fimreite is Professor of Political Science at the University of Bergen Department of Politics, Norway. Her research interests include local government studies, multilevel reforms and political behaviour.

Kurt Houlberg

Kurt Houlberg is Professor of Political Science at Vive, the Danish Centre for Social Science Research. His research interests include local government policy, local government reform, public policy, public administration and contracting out.