Governance challenges of different institutional logics and modes of organising: a Norwegian case study of municipal water supply

ABSTRACT

This article explores how agencification, in terms of increasing arm's length governance, may influence governance control with municipal service-provision if the principal and the agent develop different institutional logics. The basic expectation is that structural separation between principal and agent will challenge the principal’s control of the agent, and that this challenge will be accentuated if the structural change also leads to cultural separation, conceptualised as different institutional logics. Our findings lead to the conclusion that the relationship between formal structure and institutional logics is reciprocal; structural separation creates a separation in culture and identity, thereby accentuating the perceived distance between principal and agent, which stimulates their sense of belonging to different types of organisations. Still, development of appropriate control mechanisms is in its infancy. Formally, output control exists, but receives scarce political attention. Informal control may also work depending on adherence to a common community logic.

KEYWORDS:

• Local government
• governance control
• institutional logics
• water supply
• corporatisation
• agencification

Introduction

During the last couple of decades, corporatisation has emerged as a central trend in reorganisations of municipal service-provision (Ferry et al. 2018; Voorn, Van Genugten, and Van Thiel 2017; Aars and Ringkjøb 2011; Grossi and Reichard 2008). Despite that, empirical and theoretical research is still in its infancy and our knowledge is therefore quite limited, especially concerning issues on governance and control of all types of arm’s length bodies (ALBs), including municipally owned companies (MOCs) (e.g., Krause and Van Thiel 2019; Van Genugten, Van Thiel, and Voorn 2020; Overman 2016; Overman, Van Genugten, and Van Thiel 2015).

The ideal-type of a public-democratic system may be likened to a continuous chain of governance where citizens (the people) hold and delegate political authority to elected representatives, who delegate this further to bureaucrats and professionals responsible for implementing political decisions, which then are being controlled and evaluated by the elected politicians, who finally stand accountable for their actions and achievements to the citizens exercising their power through periodic elections. In practice, this chain has several weak links, impairing the actual influence of citizens and/or their representatives (Pierre 2009). Notably, public administration has grown in size and professionality, shifting power from elected politicians to bureaucrats and professionals. This is especially true in local government where most politicians are part-time laymen (Mouritzen and Svara 2002). Corporatisation impairs this chain further by formally cutting the direct, hierarchical link between political (and administrative) authority and service-provision. The question is, which mechanisms (if any) can compensate for this loss of direct governance and secure public-democratic accountability and control?

Studies indicate that local government has failed to keep up with the growth in MOCs in terms of developing and implementing adequate steering mechanisms, giving rise to new governance challenges and dilemmas, not least related to upholding democratic principles (Van Genugten, Van Thiel, and Voorn 2020). Although arm’s length governance may show beneficial economic and organisational effects, the political effects seem to be more uncertain (Voorn, Van Genugten, and Van Thiel 2017; Overman 2016). In this article, we seek to reduce this knowledge deficit by addressing implications of different organisational forms and institutional logics for municipal governance control with service-providing entities.

We do this by exploring and comparing the governance challenges of three modes of organising water supply in Norwegian local government. The three modes of delivery are the integrated in-house model, the externalised inter-municipal company and finally a borderline form, the in-house semi-autonomous municipal firm. These forms align with Type 0, Type 2 and Type 1 respectively in Van Genugten, Van Thiel, and Voorn (2020) typology of local arm’s length bodies.

Our basic idea is that increasing structural-formal distance between principal (the political-administrative power centre of the municipality) and agent (the service provider), will weaken the principal’s control over the agent (Andersen and Torsteinsen 2017). However, distance may also be conceptualised as a cultural phenomenon in terms of different institutional logics (Thornton, Ocasio, and Lounsbury 2012). We expect that local politicians and municipal administrative leaders are carriers of political and bureaucratic logics (Argento, Culasso, and Truant 2016; Olsen, Solstad, and Torsteinsen 2017), while engineers of the municipal water service are embedded in a professional logic associated with the task of providing high quality water to the local community. The cultural distance exits therefore irrespective of the organisational mode, but through increasing structural-formal distance, we expect that the cultural distance between the political-bureaucratic and the professional institutional logics will be accentuated. Agencification (Van Thiel 2012; Verhoest et al. 2012), agentification (Torsteinsen and Van Genugten 2016; Verhoest et al. 2004) and delegation (Overman 2016) conceptualise this process in general terms while externalisation (Citroni, Lippi, and Profeti 2013; Argento et al. 2010; Grossi and Reichard 2008), corporatisation (Ferry et al. 2018) and outsourcing/contracting out (Bel and Warner 2008; Ohemeng and Grant 2008; Bel and Costas 2006) denote more specific variants of it. In this paper we use the term agencification.

Our research question is:

To what extent and how does agencification of a municipal service influence the principal’s governance control with the service-providing agent in a situation where the principal and the agent are dominated by different institutional logics?

This article contributes to the research on implications of corporatisation in local government service-provision for public governance (Krause and Van Thiel 2019; Van Genugten, Van Thiel, and Voorn 2020; Argento, Culasso, and Truant 2016). We use data from three case studies, Case North with an integrated in-house provision (Type 0), Case West with a borderline organisational form (Type 1) and Case South, which is an inter-municipal company (Type 2).

First, we describe the municipal water service before explaining our conceptual framing. After describing our methodological approach, we present our findings and end with a discussion.

Background

In Norway, municipalities are the prime providers of water. According to Statistics Norway (2018), there are approximately 1 100 municipal waterworks in the country, providing 4.4 million Norwegians (85%) with drinking water. The majority of municipalities own and operate their water works directly while some of them own and run them indirectly through 24 municipal companies, of which 14 are organised as inter-municipal companies, seven as in-house firms and three as limited companies.¹ In addition to water supply, most municipal water providers also take care of sewage. Finally, municipal water supply in Norway does not have a commercial orientation like for instance in Italy (Argento, Culasso, and Truant 2016) or UK (Bakker 2005).

Conceptual framing

An important general premise for this paper is that organisational design matters. It matters in terms of behaviour and performance, but also in terms of less visible features like cognition, identity, and culture (Scott 2014). In what way it matters is not always obvious and needs to be explored empirically (Krause and Van Thiel 2019; Van Genugten et al. 2020; Voorn, Van Genugten, and Van Thiel 2017). During the last 30–40 years, New Public Management (NPM) has set its mark on public management reform in many countries around the globe (e.g., Andrews and Van De Walle 2013). One of the most basic characteristics of NPM is arm’s length governance (Van Genugten, Van Thiel, and Voorn 2020) or agencification (Van Thiel 2012; Verhoest et al. 2012). Agencification may be defined as a process whereby local government (or any other authority) disaggregates its service-provision into more or less autonomous operative, often single-purpose entities and regulates the relationship between itself and these bodies by various control mechanisms, e.g., contracts or quasi-contracts related to output (Torsteinsen and Van Genugten 2016; Pollitt et al. 2004). The motivation for agencifying service-provision is to encourage and improve performance economically and organisationally through increased autonomy (Overman 2016; Citroni, Lippi, and Profeti 2013; Argento et al. 2010; Grossi and Reichard 2008). Autonomy may be defined as the delegated capacity of an organisational entity to make and implement decisions without (too many) constraints from the authority centre on the exercise of those capacities. The more the authority restricts the entity’s exercise by structural, political, legal, financial or interventionist measures, the less the actual autonomy becomes (Verhoest et al. 2004). The flipside of this definition is control. Control may be conceptualised as ‘a process by which an authority and/or other stakeholders influence the input, process, and output of another entity through formal and/or informal mechanisms’ (rephrased from Krause and Van Thiel 2019, 188). Input (e.g., goals, mandates, resources) and process (e.g., legal, professional, financial, behavioural) controls are usually associated with traditional public administration, while process and especially output control, the latter an offspring of NPM, target ALBs (Krause and Van Thiel 2019). Recent research indicates that in a commercial environment output control of MOCs supplemented by trust work best (ibid.)

Agencification implies building borders between the authority and the operator, intended at reducing authority intervention in daily operations and facilitating operator specialisation, effectiveness, and efficiency (Voorn, Van Genugten, and Van Thiel 2017). If agencification remains in-house, the distance is short and border crossing may be relatively easy, both formally and informally. This implies that entities are granted managerial autonomy but not separate legal personality. However, if entities gain legal autonomy, the direct connection to the democratic governance chain is replaced by indirect governance through owner assemblies and corporate boards, thus increasing the authority-operator distance and making border crossing more formal. This is especially true if they have two or more owners, causing inter-owner coordination formalities and problems and impairing individual owners’ governance control (Voorn, Van Genugten, and Van Thiel 2019; Bel and Sebö 2018; Blåka 2017; Sørensen 2007). Finally, outsourcing service-provision to entities owned by private or other public authorities (ownership autonomy) will increase the distance further, often implying control through formal contracts (Bel and Warner 2008; Ohemeng and Grant 2008; Bel and Costas 2006). However, we do not include this type of entities in our study. In sum, agencification may be conceived as a variable defined by discrete values linked to the three forms of operator organisation. In this study, Case West holds managerial autonomy (Type 1), while Case South holds both managerial and legal autonomy (Type 2). Case North have limited managerial autonomy (Type 0).

While agencification concerns the structural-formal features of organising municipal service-provision, the institutional logics perspective focuses on the more intangible aspects. Thornton, Ocasio, and Lounsbury (2012) define institutional logics as:

The socially constructed historical patterns of cultural symbols and material practices, assumptions, values, and beliefs by which individuals produce and reproduce their material subsistence, organise time and space, and provide meaning to their daily activity. (Thornton, Ocasio, and Lounsbury 2012, 51)

Institutional logics are relatively stable meta-systems that make their influences felt at macro, meso and micro levels, i.e., the societal, the field, the organisational and the individual levels. They present different scripts and prescriptions for appropriate behaviour and performance, sometimes overlapping, other times compartmentalised or conflictual. Thornton, Ocasio, and Lounsbury (2012), building on ideas originally developed by Friedland and Alford (1991), have identified seven ideal types of institutional logics: family, community, religion, state, market, profession, and corporation. Although it is no easy task to associate these meta-categories with observable empirical types, they may serve as inspiration in trying to identify logics of public service-provision organising (Olsen, Solstad, and Torsteinsen 2017). We conceptualise the state logic as two separate but closely connected logics, politics (democracy) and bureaucracy (hierarchy), and the profession logic as engineering expertise.

Agencification reforms may accentuate the differences between two institutional logics if they are separated in-house (segmentation, e.g., Type 1) or if one of them is moved to another, external organisational entity (segregation, e.g., Type 2) (Skelcher and Smith 2015). Thereby, the structural differentiation may contribute to and legitimise cultural differentiation, thus enlarging the normative and cultural-cognitive distance between the authority and the operator, identifying themselves with separate institutional logics (Argento, Culasso, and Truant 2016; Scott 2014; Thornton, Ocasio, and Lounsbury 2012).

Organisational forms are embedded in and thus carriers of various institutional logics; they are not neutral, value-free constructions Thus, when NPM-inspired corporation and market logics are introduced into public service-provision through corporatisation, they add a cultural-informal division to the structural-formal, thereby creating an extra barrier for municipalities’ corporate governance of MOCs. Further, while municipalities are multi-purpose organisations, MOCs are often single-purpose entities, which delimits their attention to their task and their organisation. However, to the extent that both authority and operators adhere to a community logic, evaluating water as a public common good, governance and control of MOCs may become easier due to common values and mutual trust (Argento, Culasso, and Truant 2016) Table 1.

Table 1. Summarise our conceptual framing and illustrate the connections between the core concepts.

Typology →	Type 0	Type 1	Type 2
Definition	Traditional in-house delivery	In-house, semi-autonomous delivery	External, autonomous delivery, but under local government ownership
Example	Vertically integrated bodies/units	Bodies integrated politically, administratively disintegrated, e.g., in-house firm	Disintegrated bodies, often organised as public companies, e.g., inter-municipal company
Autonomy: -managerial -professional -legal	Professional	Managerial Professional	Managerial Professional Legal
Logics: -politics (democracy) -bureaucracy (hierarchy) -profession -community* -corporation -market**	Governance overall dominated by political and bureaucratic logics. Profession logics dominate operative service delivery.	Political logic dominates governance of service, while profession and corporation logics dominate within the service-providing body.	Profession and corporation (to some extent market) logics dominate the service-providing body. Political and bureaucratic logics important mostly as institutional context
Control: -input (I) -process (P) -output (O) -trust (T)	I: Investments through budgetary grants, operations through fees, decided annually by municipal council. National regulations. P: Frequent testing of water quality. Financial reporting. Intervention from municipal CEO possible. O: Performance reporting growing, primarily financial indicators. User feedback important. T: High, related to professional expertise.	I: The same + appointment of board. P: Same testing regime. Intervention from municipal council possible. O: Performance reporting through annual reports to owner. User feedback important. T: High, related to professional expertise.	I: Operations through fees, decided annually by municipal councils. Company agreement and appointment of owner assembly. Strategic decisions and large investments may require sanctions from owners. P: Same testing regime. O: The same as Type 1. T: High, related to professional expertise.

*Community logic important for all organisational modes because water is seen as a common good.

**Market logic not very relevant because water is a non-commercial commodity in Norway.

In line with this, we expect that:

(1) Increasing agencification will lead to increasing normative and cultural-cognitive distance between the authority and the operator, thus accentuating their differences.

(2) This development will decrease the principal’s control with the agent, unless compensated for by adequate governance mechanisms.

Method

We collected data for this article mainly through case studies of the three organisational forms relevant in this sector: the integrated in-house service, the semi-autonomous in-house municipal company, and the inter-municipal company. Additional criteria used for selecting these cases were regional distribution and medium-sized² municipal owners. We acknowledge that a selection of three cases cannot claim to be representative. Instead, our intention is to explore perceptions, ideas, and identities. The data consist of 1–1.5-hour interviews, documents, and the organisations’ self-presentation on their home pages. In addition, we conducted an interview with the CEO of Norsk Vann, an interest and competence organisation for the water industry in Norway, owned by Norwegian municipalities, MOCs, and some cooperative private waterworks.

All interviews took place in the interviewees’ offices and were recorded. Most of the recordings were transcribed, and the researchers actively used the recordings to control their written notes. In total, we interviewed 21 persons in the period May 2017-February 2018, using a semi-structured interview guide. To paint a broader picture, we selected interviewees at the strategic, administrative, and operative levels, including shop stewards.

Case North – the integrated in-house service

In Case North, comprising 25,000 inhabitants, the municipal water service is firmly integrated into the local government organisation. We interviewed six persons, including the Labour Party mayor and the municipal CEO. The operational interviewees were the head of the largest technical affairs department, The Operations and Development Service (O&D), the heads of the two sub-entities for VA³ projects and VA operations, and finally, the shop steward for the engineers’ union. The home page of the municipality is not very easy to use and contains scarce information about the service. The information on the water service in the municipality’s Annual Report (2016) covers less than one and a half of 92 pages in total.

Case West – the semi-autonomous in-house company (KF)⁴

The municipality with approximately 25,000 inhabitants, established the KF in 2007. We interviewed the Conservative Party mayor and the municipal CEO, the chair of the KF board, and finally, the CEO of the KF and the shop steward of the engineers’ union. The home page of the municipality is easy to use and contains basic information about the KF and the VA service. The company does not have a separate email address or homepage. The Annual Report (2016) of the municipality is 100% digital and contains little information about the KF. The KF produces its own approximately 50-page annual report, which it presents to the board and the owner, but it is not available on the municipality’s home page.

Case South – the inter-municipal company (IKS)⁵

This case includes an inter-municipal company owned by four municipalities. We have limited ourselves to studying the company and its largest owner (51%), an urban municipality with approximately 31,000 inhabitants. In total, the municipalities have almost 125,000 inhabitants, to whom the IKS provides VA services. We interviewed four persons in the IKS, including the CEO and the shop steward of the engineers’ union, and four persons in the municipality, including the mayor. The IKS is responsible for main pipelines and facilities, while the municipality takes care of its own facilities and network of pipelines distributing water to and collecting sewage/wastewater from consumers within its territory. This means that there is an interface between the two systems that may give rise to coordination problems and even controversies. Both the municipality and the company have their own separate homepages, both easy to use, informative and with links to each other. However, the Annual Report (2016) of the municipality does not contain much information about VA services, reflecting the almost total transfer of responsibility to the inter-municipal company. We were not able to find the annual reports of the company on its home page.

Findings

In this section, we present our empirical findings from the three cases.

Case North – the integrated in-house service

In 2002 the municipality in this case introduced the agency model, a business inspired organisational model, disaggregating four formerly general-purpose departments (education/kindergartens; health/social affairs; technical/infrastructure affairs; culture) into 56 single-purpose service-providing entities and granting them far more managerial autonomy than they had had before. The head of the O&D remembers this as ‘a drastic and positive change’. Today, the number of agencies has been reduced to 37, mainly through merging some of them into larger organisational entities, though without diminishing their autonomy. Performance measurement and quasi-contracts have been introduced to supplement traditional control mechanisms. The VA service consists of three sub-units. In addition to the two already mentioned, the VA Projects and the VA Operations functioning as ‘purchaser’ or planning units, there is a ‘provider’ unit placed in a separate location. However, larger assignments are outsourced to private firms. The VA service employs approximately 15 persons, of which 12 are engineers, and user fees, based on the full cost recovery (FCR) principle, finance its activities. This gives the VA service a financially secure and privileged position, not having to ‘fight’ with other municipal entities in the annual competition for budgetary grants. On the other hand, the municipal council exercises full input control by annually fixing user fees and making investments decisions.

The interviews convey an impression of high satisfaction with the VA service, also corroborated through the regular user surveys presented in the annual reports of the municipality. According to the mayor, ‘this is an invisible service; nobody thinks about it until it stops working’. Except for one example, none of the interviewees can remember that local politicians have ever raised questions or commented on the service in the municipal council or the executive committee. Further, we asked if there were any tensions between the engineer-dominated VA service and other services inside the O&D agency. Neither the head of the agency nor the managers of the two planning/purchaser VA units could verify that, except for misunderstandings and negative comments that sometimes arise because of structural borders and physical distance. To the extent that there were internal tensions, they emanated from the privileged position that the FCR- principle gives the VA service, not because of engineers vs. others. This indicates that the service enjoys high trust and great respect for its professionalism and consequently, neither politicians nor top administrative leaders try to delimit its managerial or professional autonomy.

The municipal CEO, on the other hand, said that a few years ago, ‘the VA service wanted to take care of everything themselves, including procurement’. Due to some ‘unfortunate episodes’ and some ‘tendencies to Rolls Royce’, he terminated this practice by centralising all procurement in the municipal organisation to his office. This event illustrates how the municipal CEO exercised control by intervening directly into a process to avoid possible drift away from hierarchical governance. He further expresses a general scepticism towards ALBs:

KFs and IKSs evolve their own interests beyond their core mission. They want space, want to develop their own profile, create their own agendas. This could cause friction. Even traditional in-house entities may go astray, but then I have the power to intervene.

The mayor, however, is not principally anti-corporatisation, but when it comes to infrastructure and especially the VA service, she thinks ‘that this is so important that the municipality needs to have full control’. As to the actual extent of the CEO’s power, the Norwegian municipal CEO has a comparatively strong and law-based position (LGA) to reorganise and move people and money, provided he/she respects legal regulations, agreements, and norms of decent behaviour.

Further, although the engineers hold a clear professional identity and are aware of the important role they play within the VA service, they do not express anything that indicates a strong desire to separate themselves from the municipal organisation. Quite the opposite. When asked about possible corporatisation of the VA service, practically all interviewees reject that idea. Only the shop steward mentions that a KF may be a viable alternative. Otherwise, the opinion among the interviewees is clearly and explicitly anti-corporatisation. Also, the O&D head argues that efficiency and effectiveness are important, and ‘this is more easily obtained through internal coordination’.

The findings in this case indicate minor if any tensions between an engineering logic and political and administrative logics. The engineers command a type of specialised competence that is not always accessible to ‘outsiders’, thereby granting them some ‘protection’ from interference from top political and administrative levels. As long as the services are provided and the quality and costs are reasonable, politicians are satisfied, and the engineers retain their autonomy. However, the engineers of the two co-located VA units suggested that they would like to break loose from the O&D agency and establish their own agency, moving to a new location and integrating the provider VA entity. The concentration of most municipal engineers in a new separate VA agency would mean introducing a structural division that could create a more marked cultural division between the engineering (the operator) and the political-administrative (the authority) logics within the municipal organisation, hence challenging established governance routines.

Case West – the semi-autonomous in-house company (KF)

In 2003–2006, the municipality in this case replaced the traditional department model with an agency model, like the one adopted by the Case North municipality in 2002. In this process, the municipal council decided to corporatise the VA service, organising it according to the KF model regulated by the LGA. The KF type of municipal company is not legally autonomous and consequently, it is an in-house form. Nonetheless, it has its own CEO, its own board appointed by the municipal council and it operates beyond the authority of the municipal CEO. According to the mayor and the chair of the KF board, important motives for choosing the KF model for the VA service were ‘inaccurate practising of the FCR-principle’ (i.e., suspected illegal cross-subsidisation) and the need to create an attractive professional environment able to recruit engineers.

Among those we interviewed, the attitude towards organising the VA service as a KF is mainly positive, though with one exception, the municipal CEO. Although he describes the company as well run, he is principally sceptical about corporatising municipal services, claiming that ‘coordination and governance is more optimal with a unitary line of authority’. He also expresses a fear that municipal corporations will become self-sufficient, neglecting that they are parts of something bigger.

Although positive towards the KF-model, also the mayor has some concerns:

There is something mental going on when organising people at arm’s length. The KF may become so occupied by its own tasks that it loses sight of the whole. A subculture emerges very rapidly. Even politicians appointed to the board, soon become more board members than members of the municipal council and the executive committee. The KF is an instrument for the owner, who has the responsibility for the whole organisation and the whole community.

The municipal CEO explains that he tries to mitigate possible divisive cultural drift in the KFs by including their CEOs in his regular although informal meetings with the internal agency managers. Somewhat surprisingly, the mayor claims that the municipality has better political control with the KFs than with internal service providing agencies, this in contrast to the CEO’s perception. The explanation given by him and by other interviewees is that the KF of VA services excels in corporate governance, i.e., project planning, economic management control, performance measurement and reporting. The CEO of the KF explains that while the KF produces an annual report of 50 pages, the comparable internal urban operations agency (located in the same building) presents a two-page brief. Despite reservations, the municipal CEO considers the way this KF is run could serve as a benchmark for the governance of internal agencies.

The KF enjoys a high level of managerial autonomy and a medium level of legal autonomy. Since citizen satisfaction with the VA service is high, local politicians do not give the KF much attention. According to the chair of the board (an active local politician from the same party as the mayor), he and the CEO of the KF dominate the board meetings: ‘The other members are mostly silent, and I understand that they find it difficult to vote against a proposal supported by both the CEO and the chair’. To the extent that this is true, it aligns with findings from the UK (Ferlie et al. 1996) and statements from the mayor and the CEO. When asked about his role in discussing matters concerning ‘his’ KF in the municipal council and the executive committee, he admits that he participates fully ‘without hesitation’. This implies that he openly plays the role as advocate of the KF vis-à-vis the owner, a practice criticised by the mayor but applauded by the CEO of the KF: ‘I will miss it if politicians disappear from our board. These persons are extremely competent when it comes to VA matters; they are our spokespersons in the municipal council and the executive committee’. The CEO is therefore highly negative of a proposal from the national government to ban local politicians from serving on municipal corporate boards.⁶ ‘If this proposal is passed, we should revise our organisational form’, he adds. These statements indicate that even in a situation with moderate agencification the company may operate almost as if it had full legal autonomy, an experience verified by a frustrated municipal CEO. As of today, the KF CEO expresses a ‘stronger identity with the company than with the municipality, I feel affiliation and pride’, although he has an explicit awareness of the KF’s loyalty, responsibility, and public purpose. He describes how he sometimes must ‘jack down’ his engineers, keen on ‘gold-plating’ the technological solutions at the risk of too high costs. Also, the shop steward, representing the engineers’ union, underlines his strong association with the KF: ‘I am not a municipal employee, I am employed by the company’. At the same time, he thinks that ‘the KF should keep a low profile, avoid standing too much out from the owner organisation, pay attention to the others.

Summing up, the VA services seem to enjoy a stronger and more independent position vis-à-vis its owner after it became a KF. Its managerial autonomy is more clearly defined, and although its legal autonomy is medium, in practice it resembles the autonomy of municipal companies with a separate legal personality. The engineering culture seems more dominant, and employees, including managers, have developed an exclusive organisational identity, associated with the KF, but at the same time combined with a clear awareness of the KF’s public purpose and public ownership.

Case South – the inter-municipal Company (IKS)

The company was established in 1974 to handle and clean sewage and wastewater as part of a national government-initiated project (1973–1982) to save a large lake from damaging pollution, threatening to make it unfit as the main drinking water source for a populous region. This historical background is important, as the project became very successful and gave the company a high status and legitimacy regionally and nationally, a standing that seems to characterise the company even today. In 1986, the company took over water supply and waste management, and in 2002, it became an inter-municipal company. Finally, in 2016, the municipalities decided to move waste management out and reorganise it into a separate inter-municipal company.

The fact that an IKS, formally holding both managerial and legal autonomy, is a collaboration between several municipalities may increase the distance between each municipality and the company, making ownership control more fragmented and thus more challenging (Voorn, Van Genugten, and Van Thiel 2019; Sørensen 2007). Another fact, that each municipality is a twin-principal authority comprising a municipal council and a CEO (Mouritzen and Svara 2002), may aggravate this fragmentation. The question of ownership governance and control has been a subject of continuous debate among the owner municipalities since the IKS form was introduced in 2002. From 2002 to 2009, the board largely consisted of political representatives. Both the company CEO and the mayor claim that these politicians often gave more attention to the interests of their respective municipalities than to the interests of the company, interestingly the opposite concern of the mayor in Case West. Thus, in 2009, the board became ‘professionalised’, i.e., local politicians were replaced by ‘non-political experts’. Currently, the owners formally govern the IKS through the assembly of representatives (AOR).⁷ The AOR has eight members altogether, i.e., two politicians from each municipality, who also serve on their municipalities’ councils and/or executive committees.

The mayor and the company CEO’s explicit concern for ‘company interests’ indicates support for the IKS’s autonomy and the emergence of a separate corporate identity, both associated with a corporation logic (Thornton, Ocasio, and Lounsbury 2012).

There is a strong engineering logic in the company, reflecting the legacy of the environmental protection project and encouraged by supporting institutional surroundings. The head of the Water Investments Programme for the IKS (and the four owner municipalities) emphasises the well-qualified and supportive engineering milieu, focused on regional innovation and development: ‘It’s an important reason I’ve been here for so long […]. I cannot imagine working in a municipality’. This and other similar statements indicate a blending of the profession and corporation logics, mutually galvanising each other. This mindset also underlines the cultural distance between the company and the municipality’s own administration. However, when the company started to commercialise some of its inventions through separate limited companies, the owners ordered a halt (Eierskapsmelding 2019, 39), an example of process control.

Both the municipality’s technical administration and CEO seem to be characterised by a culture and a logic that deviate from those of the company.

The technical head is critical of both the company’s community role, its innovation focus, and its strong and independent engineering environment:

The company believes it should be a social actor and participate in many areas of society. This is problematic. The company has far more employees than us, with higher education, they are much more theoretical, the company has higher costs, and its solutions are gold-plated.

He believes that neither the municipal CEO nor he have the power and authority they should have, thus indicating his adherence to the bureaucratic (state) logic. The municipality is not able professionally to go against plans from the company, and as a technical head, he feels manipulated and powerless: ‘[…] you are a purchaser without power over the quality and price of what you purchase […]’. Politicians formally representing the municipalities on the AoR, often become the company’s spokespersons in the municipal executive committees. It may be worth noting that the technical head not only speaks from a purely bureaucratic logic but also from a market logic. He explicitly uses the principal-agent model as a reference frame and mentions his role as a purchaser, a reasoning close to a market logic.

The mayor expresses a much more positive view of the company than the technical head and the municipal CEO do. He emphasises the positive aspects of having an IKS with a strong technical engineering milieu: ‘The biggest benefit from the company is that you are able to have a professional environment and a professional environment that has a brand nationally’. He accepts that the VA-company should be technologically at the forefront as well as having a community-building function: ‘Technological innovations […] are not only important but necessary to develop areas and industries’. These arguments seem to align with a community logic.

The mayor does not share the municipal CEO’s and the technical head’s criticism of the company, thus demonstrating the municipality’s twin-principal authority: ‘The company has been more successful in thinking long-term than the municipal CEOs have been. For the municipal CEO, it is simply a question about power and control’, he claims. Rather, he strongly justifies the professional autonomy of the company and he supports the existing organisational form.

The mayor does not think that the form of the company alone results in a democratic deficit, lack of political governance and transparency. According to the mayor, there is relatively large political consensus in this municipality about the organisational form. The problem is rather that several municipalities are involved. Thus, he addresses the inter-owner coordination problems we mentioned in our conceptual framing (Voorn, Van Genugten, and Van Thiel 2019; Sørensen 2007). He supports the current form of ownership governance through the AoR and a ‘professional’ board that works for ‘the best of the company’. Accordingly, he seems to support important elements in the corporation logic.

To sum up, the IKS enjoys a high degree of both managerial, professional, and legal autonomy, strongly supported by the mayor of the majority owner. This support in combination with the IKS’s high standing seems to protect the company and its engineering profession from the rather harsh criticism of the municipal CEO and his technical head, who would prefer to subordinate the company to stronger process control by its owners.

Discussion

In this article, we have presented three cases of municipal water provision representing different modes of organising and different types of institutional logics, labelled Case North, Case West, and Case South. The modes of organising are described as degrees of structural agencification, illustrated empirically by the integrated in-house service (as a baseline), the semi-autonomous in-house company and finally the autonomous inter-municipal company. The ‘strength’ of agencification thus runs from weak (Type 0), through medium (Type 1) to strong (Type 2). Simultaneously, the presence of multiple institutional logics within the institutional field may lead to increasing cultural differentiation (Thornton, Ocasio, and Lounsbury 2012), thus accentuating the structural differentiation (Skelcher and Smith 2015). The stronger the agencification, the more carriers of different institutional logics seem to gain room for developing and cultivating their special identities and practices, thus challenging the governance capacity and control of local authorities. However, the top administrative (the municipal CEO) and political leaders (the mayor) seem to disagree whether, to what extent and how this is a problem. While the latter seem to rely on trust and maybe control by output, the former advocate more input and process control through re-establishing direct bureaucratic/hierarchical governance.

A basic motive for writing this article was our assumption that there is a tension between the state institutional logic of political and administrative municipal leaders and the professional logic of the municipal engineers. Our findings do not corroborate this expectation.

In Case North, the integrated model, we cannot document any convincing signs of tension between the two groups. Instead, these logics seem to complement each other under the hegemony of the public logics, thus exemplifying the assimilated hybrid type of Skelcher and Smith (2015). The relationship is, as our study shows, characterised by trust and mutual respect for each other’s competence, tasks, and roles. However, the CEO may and has intervened when he finds it necessary, using formal authority for process control.

As we move to Case West, the situation changes slightly. To the extent that we can talk about tensions at all, it does not follow the expected state-profession fault line, but instead emerges along the organisational-institutional cleavage between the municipality and the KF. The municipal CEO, although principally opposed to organising municipal service provision outside the hierarchy of the municipal authority, acknowledges that the firm is well run, thus signalling process- and output-based trust. Case West seems to align with the segmented hybrid type (Skelcher and Smith 2015) where the profession logic gets its own compound inside the state logic ‘property’.

In Case South, the institutional cleavage is even more visible. Here the harshest critics of the IKS are the engineers inside the owner municipality. Consequently, engineers do not constitute one professional group unified by the same institutional logic across structural differentiation. Instead, as our study indicates, institutional logics are influenced and partly modified by formal structures. We observe this, first, inside the municipality where the professional logic seems to be dominated by the political and bureaucratic logics, thus aligned with the assimilated hybrid type of Skelcher and Smith (2015). Second, with the single-purpose IKS, a segregated hybrid type emerges (Skelcher and Smith 2015), indicating increased professional autonomy.

According to the institutional logics perspective (Thornton, Ocasio, and Lounsbury 2012), organisational forms are materialisations of institutional logics, and thereby, logics are conceptualised as independent variables shaping organisations, work operations and behaviour. Consequently, agencification, including corporatisation seem to reflect dissemination of neo-liberalism and the emergence of market and corporation logics in public administration.

Our findings indicate, however, that the relationship between structure and culture seems to be reciprocal rather than unidirectional. Although institutional logics may influence organisational design, Thornton, Ocasio, and Lounsbury (2012) acknowledge that agents have partial autonomy, implying some manoeuvring space when they design organisational structures and processes.

Consequently, formal organisations are more than mere logics; they are power structures designed to promote or inhibit certain strategies and practices. Although influenced by existing institutional logics, organisational design may facilitate innovations in terms of compartmentalisation, assimilation, blending or blocking of logics adapted to the purpose and tasks of municipalities, including their needs for governance control (Skelcher and Smith 2015).

How then do our findings align with our two expectations?

Increasing agencification will lead to increasing normative and cultural-cognitive distance between the authority and the operator, thus accentuating their differences.

Most of our interviewees describe a development in line with our expectation. In contrast to municipal engineers in Case North and Case South, engineers in the agentified entities KF and IKS do not perceive themselves as municipally employed. Instead, they conceive themselves as employed by an independent business-like expert corporation, with which they identify explicitly. Municipal engineers, on the other hand, define their professional logic as assimilated into the dominating political and administrative logics of their municipalities (Skelcher and Smith 2015). Thus, although all engineers are carriers of a common professional logic, organisational affiliation seems to be stronger, provided their professional competence and autonomy are respected. Further, municipal leaders, especially the CEOs describe how corporatisation reduces communication and distances MOCs from their owners, formally, normatively, and culturally-cognitively. Finally, while the KF and IKS CEOs identify themselves with their companies, they still express an allegiance to their public purpose, indicating a community logic.

Agencification will decrease the principal’s control with the agent, unless adequately compensated for by adequate governance mechanisms.

Traditionally, Norwegian local government has governed service provision through input and process controls, but also through informal communication and trust. CEOs seldom intervene in providers’ exercise of professional autonomy. Nonetheless, general output control has been exercised, primarily through elections, mass media, and political activism. However, with agencification and corporatisation, output control has become more emphasised, formalised, and detailed. MOCs’ annual reports and municipalities’ ownership reports exemplify this. Despite increasing access to performance information, MOCs’ activities and output attract limited attention from municipal councils and CEOs. While local politicians, as part-time laymen, usually do not have sufficient capacity and competence, maybe neither the interest to exercise appropriate control, the CEOs’ governance mandate normally does not include MOCs. In addition, whereas CEOs express scepticism towards agencification, mayors are more positive. Thus, the municipal twin-principal authority is divided, which barely adds strength to owner control. Also, when politicians appointed to MOC boards (and also to the AoR) act as their spokespersons vis-à-vis the owner municipality(-ies), control may be further impaired. Finally, as a tangible, uncontroversial utility service water supply easily becomes de-politicised, often evaluated by purely technical indicators.

In sum, the development of control mechanisms, appropriate for handling the democratic challenges of agencification still seems to be in its infancy in Norwegian local government. The reliance on formalised output control of distanced agents may be fragile, especially if politicians’ trust is approximate or blind.

Conclusion

Our study, with only three cases in one country, obviously has its clear limitations, especially concerning possibilities for empirical generalisation. On the other hand, the article may have some theoretical value for the study of implications of corporatisation in local government for governance control, trying to raise some critical questions regarding the application of the highly popular institutional logics perspective. Although we find it useful and theoretically stimulating, our main objection to some of the literature is its seemingly unidirectional linking of the independent variable ‘institutional logic’ (idea) and the dependent variable ‘organisational forms, work practices and behaviour’ (practical world of power, material resources etc.). In our view, this is a dynamic-interactive process in which the influence may go in both directions. Finally, the use of the institutional logics perspective for exploring corporatisation in local government is slowly emerging, and we would like to see more research done along these lines.

We recommend that future research, in addition to including more cases should include other municipal services and various institutional contexts. Especially, we still lack studies concerning corporatisation of ‘soft’ municipal services, a paradox since these are the most prevalent in local government, especially in the Nordic region. Further, to what extent does the FCR-principle ‘protect’ services from the usual budgetary competition, thus reducing political attention and control and stimulating segregated hybridity, i.e., different cultures. Finally, to the extent that agencification leads to de-politicisation, what are the implications of corporatisation for democratic and public governance control with local service-provision? How and to what extent can trust be a reliable supplement to output control?

Disclosure of potential conflicts of interest

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

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Additional information

Notes on contributors

Harald Torsteinsen

Dag Magne Berge is an associate professor of political science and public administration in the Faculty of Business Administration and Social Sciences at Molde University College, Norway. In his research he has investigated the relations between politics, state, and industrial development in addition to innovations and organisational change in the public sector. He has recently published in Public Administration (2021) and Business History (2020).

Harald Torsteinsen is a professor of political science and public management in the Department of Social Sciences at UiT, The Arctic University of Norway. His research interests include organisational and political innovation and reform in local government, especially corporatisation of service provision. He has recently published in Public Administration (2021), Public Money & Management (2019) and Local Government Studies (2017).

Notes

1. Information from Norsk Vann, 21 January 2021.

2. Statistics Norway (2018), category 5, i.e., 20,000–50,000 inhabitants.

3. VA, abbreviation for ‘vann og avløp’, meaning water and sewage/wastewater.

4. KF, abbreviation for ‘kommunalt foretak’, meaning municipal firm or company. See the Local Government Act (LGA), chapter 9.

5. IKS, abbreviation for ‘interkommunalt selskap’, meaning inter-municipal company. See the Inter-Municipal Companies Act.

6. Proposed amendments in the LGA.

7. The assembly of representatives (in Norw. ‘representantskapet’) resembles the annual general meeting of a private law limited company.