Government steering and government disruption: co-operation between government and municipal actors in a state-initiated school improvement programme from the municipal actors’ perspective

ABSTRACT

This article focuses on the co-operation between the Swedish National Agency for Education and school actors at the municipal level, examined from the latter’s perspective, within the context of a state-initiated school improvement programme, namely Co-operation for the Best School Possible (CBS). Co-operation between different levels of the school system is a neglected but essential aspect to analyse in a decentralised system such as Sweden’s, which is showing signs of re-centralisation. Empirically, the article is based on interviews with local actors in a small municipality participating in CBS. The interviews were part of a case study, and the analysis was guided by the theory of soft governance, Vedung’s concepts of sticks, carrots, and sermons as policy instruments, and Weick’s concept of sensemaking. Sensemaking, evident in the case study as a retrospective communicative notion, was employed to capture the local actors’ stories of how they perceived CBS. The connection with past experiences also played a part in their sensemaking since a clear history exists and was noted between the state and municipal levels. In conclusion, the analysis shows that CBS used sticks, carrots and sermons to steer municipal school actors towards the right path as regards school improvement.

KEYWORDS:

• Co-operation
• school improvement programme
• soft governance
• policy instruments
• sensemaking

Introduction

A certain discourse has been underway for some time now on schools facing a crisis in many countries (Nordin, 2014b), prompting some governments to act. Some examples in this regard include the United States (see, e.g. Coburn et al., 2016 discussing the No Child Left Behind Act and Common Core State Standards) and the United Kingdom (see, e.g. Cameron, 2010 discussing the Secondary National Strategy). Sweden’s situation is similar. According to some international studies, such as the Programme for International Student Assessment (PISA), the academic performance of students in Sweden’s schools has declined over the last decade (Andersson et al., 2018). In response to criticism appearing in domestic and international reports, the Swedish government launched, in 2015, a school improvement programme called Co-operation for the Best School Possible (CBS), which aims to turn around low-performing schools. The government commissioned the Swedish National Agency for Education (SNAE) to implement CBS in order to raise knowledge outcomes and increase equivalence within and between schools (Utbildningsdepartementet [Swedish Ministry of Education], 2015, p. 1). The programme has been costly, and it has been quite demanding for the schools involved, particularly in terms of the time and effort they need to invest in it (Ärlestig & Johansson, 2020). Yet, little is known about how CBS and similar programmes affect schools. Ascertaining the perspectives of local municipal actors concerning large-scale school improvement programmes is important for several reasons and on several fronts. Local actors (a) ‘do’ school, (b) are responsible for implementing national reforms and (c) work within the school organization; and they are thus a significant voice in one of the largest institutions that affect people’s lives. School improvement is a major societal and political issue that influences people in real ways and has an impact on the national economy in the race to become a strong knowledge society (Earl et al., 2005; Paulsen et al., 2014; Säfström, 2005).

This article focuses on one important component of the large-scale CBS programme initiated by the Swedish government, namely co-operation. This concept can be seen as part of the new governance regime wherein networks, partnerships and co-operation are considered necessary tools for the act of governing. Co-operation has been called the new work form in the welfare state (Danermark & Kullberg, 1999) and a solution to the wicked problems faced while delivering welfare services (Haveri et al., 2009). The word co-operation is often seen in a positive light, and it has a positive connotation, even in research (Lindberg, 2009). Normative advice on how to benefit from co-operation is common. That said, it has also been referred to more negatively as an organizational trend, ‘a rationalized myth’ (Johansson, 2011 see Meyer & Rowan, 1977), a political fashion word and an educational political ideal (Dahlstedt & Hertzberg, 2011, pp. 149–150). Co-operation can be considered as both an organizational matter and a political concern that relates to a certain way of governing (Chaib, 2018). This article departs from the latter and views co-operation as part of the shift in focus from government to governance (Cardini, 2006; Watson & Drew, 2017).

According to Rönnberg (2011) the state’s role as an actor has been largely under-researched concerning changes in the education system in the last 30 years. Decentralization in Sweden has led to a hollowing-out of the state’s authority and control. The increase in global pressure resulting from testing and league tables also exemplifies how the state has been hollowed-out in its capacity to steer education centrally. However, this hollowing-out has led to a need for the state to fill in by, for example, introducing an inspectorate (in 2008), expanding national testing, and not least increasing targeted state subsidies (Lundström, 2018; Nordin, 2014a; Rönnberg et al., 2019; Rönnberg, 2011). The SNAE has ‘become one of the government’s most important means to exercise governance over education by such use of target financial funding’ (Rönnberg, 2011, p. 695).

The power relations in state-initiated co-operation between state and municipal actors in a school improvement programme are central to this article. Power is an important aspect of investigation when it comes to co-operative programmes. However, research on direct co-operation through physical meetings between government and municipal actors in education appears scarce, and little is known about how state agencies and local educational authorities ‘interact in decentralized and deregulated education systems’ (Nordholm, 2016 see also Glaés-Coutts & Nilsson, 2021). Thus, this article seeks to address such knowledge gaps by exploring how municipal actors perceive the intended co-operation between themselves and the SNAE in a state-initiated school improvement programme.

The study presented in this article draws on the concept of sensemaking proposed by Weick et al. (2016) to examine how co-operation can be perceived from the perspective of local actors. Further, it analyses the complex power relations and interactions between a state educational agency and local school actors with reference to soft governance and relevant policy instruments to understand the interactions between state and local levels in school improvement programmes such as CBS.

This article aims to illuminate how co-operation between state actors and municipal actors (headteachers, local education authorities [LEAs], and local politicians) in a government-issued school improvement initiative is perceived from the municipal actors’ perspectives. The questions that guided the study are as follows:

• How do the municipal actors (headteachers, LEAs and local politicians) in the given case make sense of co-operation with the SNAE in CBS?

• How can the municipal actors’ stories be understood and put in a wider perspective by using, as analytical lenses, concepts of soft governance and relevant policy instruments?

Hence, the article contributes to extant research on actors in subordinate and superordinate positions, as it draws on the theories of soft governance and sensemaking, as well as relevant policy instruments as central concepts. The local actors’ perceptions of co-operation significantly contribute to the knowledge about large-scale school improvement programmes.

Theoretical framework

The framework adopted for the study presented in this article combined theories on government steering and sensemaking. More precisely, the concept of soft governance and different policy instruments were referenced in combination with the concept of sensemaking proposed by Weick et al. (2016). These central theories are presented in more detail below.

Governance – a complex relationship between the state and municipality

Much has been written about the apparent shift from government to governance. This shift centres on the change from hierarchy to networks (Rhodes, 1997; Salamon, 2002; Zehavi, 2012). There is no consensus on the definition of governance (Hupe & Pollitt, 2010; Pollitt & Hupe, 2011); however, it is understood to cover the whole range of institutions and relationships involved in the process of governing (Pierre & Peters, 2000). Fredrickson (2005, in Hupe & Pollitt, 2010) used governance to describe how the state steers through grants and contracts, which is illustrative for this article. Inherent to the notion of governance is a paradox central to education: on the one hand, there is decentralization, while on the other, there is increased control from the governing power via, for example, audits (Cardini, 2006; Keddie, 2015; Lingard & Sellar, 2012; Paulsen & Høyer, 2016). Although the state’s role in steering school may have changed, it has not necessarily lessened (Bell et al., 2010; Rönnberg et al., 2019). Hierarchy still plays a vital role in the act of governing (Pierre & Peters, 2000 cf. Shadow of Hierarchy, Héritier & Lehmkuhl, 2008).

Much of governance theory thus suggests a movement away from ‘hard’ to ‘soft’ policy instruments. ‘Policy instruments are hard or soft with respect to the degree of government intrusiveness and coercion involved in the use of a specific instrument’ (Zehavi, 2012, p. 244). Hard governance stands for laws and regulations, and soft governance is aimed more towards influencing how people think. Nevertheless, as Moos (2009) suggested, soft governance ‘influences agents in much deeper ways. While these methods of influence might seem softer, or more educational, the effects of soft influence are harder and more profound’ (p. 399). Soft governance has become increasingly important in the era of decentralization where the state steers at a distance using, e.g. social technologies that build on the premise of choice. Examples of social technologies are routines, methods and work forms that can be used to influence both the behaviour and thoughts of school actors (Moos, 2009, p. 402). These types of social technologies, which can be seen as best practice advice, have hidden built-in decisions and influences, and they are often displayed as neutral tools to use.

Policy instruments

The dichotomous division into hard and soft governance has also received criticism (Zehavi, 2012). To nuance this dichotomy, policy instruments can be used when analysing governance styles. Numerous definitions of policy instruments exist. In this article, a policy instrument is regarded as ‘a deliberate structured effort by governors to solve a policy problem by modifying actions of the governed’ (Brukas & Sallnäs, 2012, p. 3). It is important to bear in mind that policy instruments are not neutral devices; rather ‘they are value-bearing policy outcomes manifesting the underlying aims and power structures’ (Brukas & Sallnäs, 2012 see Lascoumes & Le Galès, 2007). Therefore, analyses of policy instruments could reveal more about the relationship between the governing and the governed ‘than accounts of motives, or later discursive rationalizations’ (Lascoumes & Le Galès, 2007, p. 9). The power dimension is, therefore, an important factor in Lascoumes and Le Galès’s interpretation of policy instruments.

Zehavi (2012) pointed out that the hard and soft dyad does not easily fit the important instrument of finance. In this article, the trichotomy of carrots, sticks and sermons developed by Vedung (2016) has been used to describe how the government can influence municipal actors, in this case, with economic instruments (carrots), different types of regulation and subsequent sanctions if not followed (sticks) and lastly information (sermons). Carrots are usually considered softer instruments than the more coercive sticks. However, economic incentives have their disadvantages: they are costly compared with the other two instruments and can neither forbid nor command as regulations can (Vedung, 2016). Sermons are similar to carrots in that they are generally not coercive. Instead, the government tries to influence municipal actors through education, persuasion and knowledge transfer. All modes of governance can however include both soft and hard instruments (Wilkoszewski & Sundby, 2016).

Although the dichotomy of soft and hard governance can be perceived as too crude when analysing the steering arrangements utilized by the government, the distinction of soft techniques proposed by Moos (2009) that builds on the premise of choice adds a layer that highlights the possibility of influencing municipal actors more deeply. Being forced to do something for money or for regulatory purposes is different from doing something out of one’s own perceived conviction. This leads us to the last theoretical piece of the framework: sensemaking. How do local actors make sense of the co-operation and steering arrangements within CBS?

How to make sense of the acts of governance

Sensemaking is a retrospective process wherein action and interpretation are at the core (Weick et al., 2016). ‘To focus on sensemaking is to portray organizing as the experience of being thrown into an ongoing, unknowable, unpredictable streaming of experience in search of answers to the question, “what’s the story?”’ (Weick et al., 2016, p. 410). In the results section of this article, the focus will be on capturing local actors’ stories of how they make sense of a new state-initiated programme of turning around schools. Sensemaking is also often connected to events that disrupt the every-day lives of people in an organization, and the CBS initiative seems to fit that description.

Sensemaking is many things, but in this article, it is discussed as retrospective, social and systemic and organizing through communication (Weick et al., 2016). It is in retrospect that people make sense of the experiences they have had. Notably, sensemaking is influenced by several social and systemic factors, including the social relationships within and between municipal actors, as well as previous encounters with, for instance, government agencies and their representatives. Lastly, communication is a central element in sensemaking. It is through communication that members of an organization form a view of situations and the people encountered. The answer to the question ‘What’s the story?’ thus emerges ‘from retrospect, connections with past experience, and dialogue among people who act on behalf of larger social units’ (Weick et al., 2016, p. 413). In this article, sensemaking is thus used to explore municipal actors’ experiences of CBS and has bearing on the methodology of the analysis.

Weick et al. (2016) cautioned against the tendency to exaggerate agency when discussing sensemaking, which ignores the fact that ‘people internalize and adopt whatever is handed to them’ (Weick et al., 2016, p. 417). Sensemaking can also be seen as naïve, compared with uses of critical theory (Weick et al., 2016). To summarize, three different tools were used in the analysis. Sensemaking was mainly used to put the informants’ stories in perspective, and the theories of soft governance and Vedung’s policy instruments were used to interpret the results and place the case at hand into a wider perspective.

The CBS programme – background

As previously mentioned, there are claims that there is a discourse on school systems facing a crisis in many countries. The Organisation for Economic Co-operation and Development (OECD) reports and tests such as PISA have prompted governments to act. An investigation of the co-operation and power relations between different levels of the school system in Sweden is particularly interesting since (a) ‘Sweden went from having one of the most centralized to one of the most decentralized education systems in the Western world’ in the early 1990s (Lundahl, 2007, p. 121); (b) Sweden’s education system has been very reform-dense since the 1990s (Lindgren et al., 2016; Nordin, 2014a); and (c) the government invests much money into in-school development projects. There are, for example, over 70 different grants that schools can apply for (Riksrevisionsverket [Swedish National Audit Office], 2017). In connection to CBS, the SNAE alone has grown substantially in terms of both its activities and number of employees since 2017 due to an increase in budget, which in many parts is connected to CBS (Ärlestig & Johansson, 2020). In Sweden the responsibility for schools is divided between the state and local responsible organizers and this is an issue that has been debated extensively ever since the decentralization reforms. The reforms have been followed by more indirect governance via performance steering, control, inspection and economic instruments such as targeted state subsidies (Hudson, 2007; Lundström, 2018; Rönnberg et al., 2019; Rönnberg, 2011). In both domestic SOU [Swedish Government Official Reports] (2014), p. 5; 2016:38) and international reports operation and Development (2015), there are now formulations stating that the state should take more responsibility for Swedish schools, e.g. by providing and financing professional development for teachers and school management, and that it should strengthen the responsible organizers via government support and co-operation. One weakness in the system, pointed out in a commission report, is the ‘failing capacity and responsibility of many responsible organizers’ (SOU [Swedish Government Official Reports], 2017:35, p. 14). The decentralization reforms have created tension between the state and municipalities as well as between different local actors (Jarl & Pierre, 2018; Nihlfors & Johansson, 2013).

The CBS programme involves co-operation between three different parties, the SNAE, the participating responsible organizers (and local schools) and the higher education institutions (HEIs) that offers the national programme for headteachers. The Swedish Schools Inspectorate can also be included, since a selection of the responsible organizers that can be offered to participate in CBS is provided to the SNAE by the Schools Inspectorate. The selection is made based on regular supervision conducted during the previous supervisory period (SNAE, 2016).

The trends of decentralization, co-operation and more control discussed previously are identifiable in CBS, for which schools are selected based on audits by the Schools Inspectorate. Co-operation between the different levels in the school system is emphasized in policy documents concerning the CBS programme (Skolverket [Swedish National Agency for Education], 2016 see also Kronqvist Håård, 2021). The school improvement initiatives within the programme should be based on the needs of individual schools and be context-specific (SOU [Swedish Government Official Reports], 2016). The programme lasts three years for each school. The CBS process often starts with a mapping phase spanning approximately six-months. The participating schools and LEA, together with the SNAE representatives, conduct a situational analysis to pinpoint the improvement initiatives the school needs, which subsequently results in an action plan. Examples of initiatives stated in the action plans include developing teaching through the SNAE’s pre-produced web education material in assessment and special education and training in systematic quality assurance work as well as anchoring and implementing routines and plans. The SNAE is actively involved in the mapping phase, whereas in the ‘action’ phase, HEIs and consultants take a more central role. In-service professional development initiatives provided by the SNAE are also part of the CBS programme, as can be seen in the examples of initiatives listed above. The participating HEIs’ primary role is to implement certain initiatives, often focusing on systematic quality assurance, in collaboration with the schools and headteachers concerned. This article, however, concentrates on the relationship between representatives of the SNAE and local actors.

There is also a closure phase wherein a final report on the school improvement initiatives performed within the programme and plans on how to move forward is written by the local actors. The report is then submitted to the SNAE. The last two phases overlap to some extent. The research project described in this article was conducted in the final stage of the action phase, which overlapped with the closure phase (Table 1). The study took seven months to complete.

Table 1. Approximate phase structure in CBS.

Method

This article is based on data from a qualitative case study in a small, rural municipality with fewer than 15,000 inhabitants (Sveriges Kommuner och Landsting [Swedish Association of Local Authorities and County Councils], 2016). A case study provides the opportunity to greatly illuminate an individual case. It is not possible to draw general conclusions from an individual case, but the case description should rather be seen as an exemplary case (Bryman, 2016), and the primary purpose of the case study is ‘to understand the meaning of a certain phenomenon or experience’ (Merriam, 1994, p. 30). It is the strategies behind the actions taken and words spoken that are the focus, not the intentions of the actor or the true meaning of the actions taken (Alvesson & Sköldberg, 2008).

The participating schools were compulsory schools, and the pupils were thus aged between 6 and 16 years. The LEA was also part of the initiatives. The study began in the autumn of 2018 and concluded in the spring of 2019. The municipality that took part in the study was in the final stages of the three-year CBS process (Table 1). Thus, the municipality had the experience of the whole process, which was the pre-determined criterion for choosing a municipality. Since the establishment of CBS is relatively recent (2016), only a certain number of municipalities that met the criterion for the study were available. The final selection was based on convenience (Bryman, 2016; Trost, 2012). I had good access to the municipality and could thus return to it over time. The data collected consisted of meeting observations, interviews and local and national documents concerning CBS. Since the aim of this article was to depict how local actors made sense of their relationship with representatives of the SNAE in the CBS process, only the interviews were used. Five headteachers, three LEA employees (superintendent, quality manager and administrative support for the CBS process) and two local politicians were interviewed, meaning that local actors with different functions and roles were included in the study. These 10 actors had been the most involved leading municipal actors in the CBS process. Of the interviewees, six had been involved in the CBS process from start to finish. One of the headteachers interviewed was more peripheral to the process, as their school had not received remarks from the Schools Inspectorate. The schools in the municipality that did not obtain remarks from the Inspectorate’s supervisory reports could to some extent cherry-pick the initiatives they wanted to take part in the CBS process, meaning that they were not obliged to participate in all of the various school improvement initiatives.

Of the 10 interviews, seven were conducted face to face and three were done by telephone. The interviews lasted between 30 and 90 minutes. The interviews with the local politicians were generally shorter than the other interviews as those participants were not involved in the work with CBS on a daily basis. The interviews were semi-structured, and the questions concerned the interviewees’ perceptions of the process of the CBS programme from start to finish, with a focus on co-operation in the programme and events that stood out in the process. Among the questions asked were the following: ‘How was your experience of the mapping phase?’ and ‘How was your experience of the co-operation in the CBS programme?’. For the most part, the interviewees told their stories about CBS chronologically, from the mapping phase to the closure phase. The interviews were transcribed in the form of readable text and not verbatim (Bryman, 2016). All quotes were translated from Swedish to English. The interviews were analysed thematically inspired by the steps in ‘concentration of meaning’ described by Kvale and Brinkman (2014, pp. 246–249). The transcripts were read several times to gain a sense of the whole interview. The next step was to establish the natural unities of meaning in the interviewees’ stories, bearing in mind the focus on perceived co-operation with the government actors. Longer utterances by the interviewees were reduced to shorter expressions. These units were then condensed into three central themes namely identity, communication, and government disruption (Table 2).

Table 2. Concentration of meaning.

Examples of natural unity of meaning	Central theme
It was great that we got this offer … even if it was for being the worst in class.	Identity
Once we had carved out the areas we needed to improve, we could have tailor-made initiatives at the school unit level. However, over time I came to understand that there were ready-made concepts.	Communication
We’ve talked about government disruption instead of government steering. Is it what we need right now? I think we would have come just as far by working on our own.	Government disruption

The interviewees received both oral and written information about the study, including information on how the data would be stored and processed. They were also informed that the data collected would only be used for research purposes and that they could withdraw at any time (Swedish Research Council, 2017). The interviewees gave their written consent to participate in the study. Interviews contribute knowledge that is constructed in the meeting between the interviewer and the interviewee, where trust is an important issue (Resnik, 2018). This means that respondents can express their views differently under other circumstances. Furthermore, researchers conducting qualitative studies might have difficulty guaranteeing complete anonymity (Van den Hoonaard, 2003). These issues are discussed below, both from an ethical and a methodological perspective, since my interpretation of the participants’ statements might affect the results. Before beginning the interviewing process, I observed several meetings in the municipality to get familiar with the case and the local actors. This enabled me to be more attuned in the interviews and create a more open atmosphere. There was always the risk that the local actors would feel inhibited by the fact that others in the municipality might recognize their statements and would therefore give more positive statements in the interviews. To counter this risk of bias, I started with the observations and read up on the case before the interviews. Alvesson (2003) warned about several romanticized and neopositivistic views in interviews. There are many pitfalls in both conducting and analysing interviews. For example, an interview can be seen as identity work or political action by the interviewee. By relying on sensemaking and a critical perspective in analysing, I tried to avoid some of these pitfalls at the analysis stage. Moreover, the word headteacher and the acronyms LP (for local politician) and LEA (for local educational authority) were used with a number to maintain the participants’ integrity.

Results and discussion

The results are presented per the identified central themes of identity, communication and government disruption. Identity is evident in the material regarding low self-esteem among the local actors due to poor results and school inspections. Communication can be seen as inflexible, and there is a strong rhetoric of choice for the local actors from the SNAE in the material. Finally, government disruption is evident in the interview data, since locally identified improvement work was put on hold. In each section that follows, the results are discussed using sensemaking to put the local actors’ stories in perspective; this is followed by an analysis using the theoretical lens of soft governance and policy instruments to better understand how power is exercised among the different levels in the CBS programme. The results are presented descriptively to capture the stories of the municipal actors.

Identity

The municipality, or more precisely the politician responsible for the school, was quick to accept the offer to participate in CBS. ‘My spontaneous response was, of course, we should be part of this’ (LP1). According to LEA1, the SNAE gave the municipal actors a period of one work week to consider the offer to participate in CBS, which the headteachers and LEA felt was inadequate. Not everyone in the management team was behind the decision take part of CBS:

In fact, it felt like there were no alternatives. A bit like that. It is linked, as it so often is these days, to state subsidies and the municipality receiving money. The results are not brilliant, so somewhere you want to both get the results up and get some of this money cake, so there was not much reasoning behind it.

(Headteacher 1)

This suggests that saying no is not an option when money is involved and when the school’s identity is associated with poor academic results. Other headteachers pointed out the lack of any real discussion within the management team about whether they should participate. CBS was presented more or less as a done deal by LP1. There were also other concerns about being able to decline the offer: ‘What happens if we don’t get on board? What will the media say? Here, we have received an offer due to our results’ (LEA1). Both these quotes show that the actors felt that there was no real discussion about whether or not the municipality would take part in CBS. In addition, the quotes provide two reasons, external and internal, for being unable to say no; inability to turn down money and fear of bad press in case of a refusal. The question is, can you decline an offer from the SNAE when your school results are poor?

The responsible local politician also mentioned ‘being worst in class’ (LP1), which added to the feeling of not being able to say no when the offer of participation came from the SNAE.

A municipality that has such bad – far from good – results, that sees a lot of problems with the school we have – then, you cannot say no when the National Agency for Education comes and says that we want to help you get better. It’s not possible. (LP1)

This further illustrates a self-esteem decline that the municipality suffered after the school inspections. Moreover, as one headteacher said (Headteacher 5), the offer to participate in CBS was not very inspiring, as it clearly stated that the programme was aimed at schools with poor results and schools that were considered to have difficulties in making improvements on their own (Utbildningsdepartementet, 2015; Skolverket, 2016). This could lead up to the expressed identity of being ‘worst in class’.

Social and systemic factors can both play a part in the municipal actors’ sensemaking of the initiative (Weick et al., 2016). There was a clear history between the government agencies and the municipal actors, as one of the politicians stated, when the Schools Inspectorate or the SNAE calls ‘it’s usually not any good news’ (LP1). The municipal actors’ attitudes and sensemaking of the CBS initiative could have thus been affected by their previous experiences with the government agencies. Their sensemaking could also have been said to be influenced by their feeling of being ‘worst in class’. Sensemaking is social and systemic, influenced by both external and internal forces such as the media, government agencies and public opinion. The talk about bad school results may also imply low self-esteem among the municipal actors; therefore, they could have been more susceptible to external influences from the wider system.

Concerning the steering arrangements, the statements by the municipal actors showed signs of being in a subordinate position, at first in relation to the Schools Inspectorate. The relationship between the Schools Inspectorate and the school was one of hierarchical supervision with the threat of sanctions or sticks according to Vedung (2016). The SNAE promises help to self-help; however, the hierarchical relationship with the SNAE was visible in the statements of the municipal actors. The short time the municipal actors were given to consider CBS also points to a hierarchical order, which can add to feelings of inferiority. Seeing oneself as worst in class can be grounds for receptiveness to external steering and self-regulation to avoid direct punishment (e.g. penalty payments imposed by the Schools Inspectorate). The way the local actors referred to themselves is part of a discourse of comparison and standards (Perryman et al., 2018).

Communication

Many interviewees discussed perceived problems in communication with the SNAE. The municipal actors did not feel that their voices were heard, and there were many comments on the agency being ‘square’ and inflexible in discussions about different improvement initiatives. The municipal actors felt that the information provided by the SNAE in the initial phase of the initiative was ambiguous. The municipal actors wanted an overview of the entire CBS process, but the SNAE could not give any clear answers. According to LEA3, representatives of the SNAE heading the process said that they had ‘laid the rails while the train was running’, and the LEA personnel found the process to be very unclear since the instructions could change from one day to the next. The SNAE lacked a sound project model for CBS, which caused great irritation at the start.

Contact with the SNAE decreased as the CBS process moved forward. Reports were to be compiled and submitted by the LEA over time. However, the feeling of laying the rails as the train was running did not go away. LEA2 reported that the municipality had received documents and templates late from the SNAE; indeed, this continually last-minute approach had caused frustration on the part of LEA. Even in the final stages of the programme, the process model had yet to be completed. Instructions on how to write the final report on the different improvement initiatives undertaken in the municipality had still not been finalized when the SNAE representatives visited the municipality to instruct them on how to write the report. A downside of being one of the first to take part in CBS was that instructions about the process could change at any point.

The headteachers and some members of the LEA spoke a great deal about the promise of tailor-made initiatives from the SNAE in the initial mapping phase; however, the end result was ready-made packages of school improvement initiatives.

I said on a few occasions that this feels like we should sit and guess things in a package that you have already packed with some ready-made initiatives. Just say what it is you have, and we’ll do it. (Headteacher 1)

Another headteacher stated that the SNAE had a clear agenda and that the schools were being nudged in a particular direction. However, the opinion of another headteacher on the mapping phase was largely positive. Nonetheless, the packages of ready-made school improvement initiatives were evident: ‘It was very exciting with the mapping process. Yet, I came to understand over time that there were ready-made packages’ (Headteacher 5). A headteacher criticized the CBS-related work, saying it had led to ‘a wedge being struck between the lower grades and the higher grades’ (Headteacher 3). The school wanted a whole-school perspective, from pre-school to Grade 9, when working with the different initiatives, but this was denied by the SNAE. This situation meant that the school had, for example, different in-service training modules provided by the SNAE for the teachers regarding assessment. One of the headteachers was slightly more positive when referring to communication with the SNAE during the initial mapping phase. Headteacher 5 described the representative of the SNAE who had helped them in the mapping phase as ‘very guiding, which was good … [they] kept us on the right track at all times’.

Generally, both the LEA and headteachers expressed satisfaction with the different school improvement initiatives. They said that they had received solid training, although it was not always what they would have chosen. Nevertheless, there was a desire for a greater openness on the part of the SNAE – what it offered was ‘what Sweden needs’ (Headteacher 4), not necessarily what the mapping phase revealed at the different schools and at LEA. One of the headteachers concurred:

They come here and basically have a ready-made package, and [they] could say that Sweden needs this. Do this. It could almost have been an order because I think we will end up there anyway regardless of whether we have the analysis [in the mapping phase] or not.

(Headteacher 3)

These statements suggest that the headteachers were left disappointed because they had been led to believe that the school improvement initiative would be tailored to meet the needs of their school, as had been outlined in the initial mapping phase. Instead, they felt that the professional development had been predetermined and had more of a general character of what is currently seen as ‘good’ school improvement. They did not oppose the professional development provided to them per se, but rather the process leading up to the decision of which professional development to choose. The headteachers and LEA disputed the talk of the initiatives being tailor-made; ultimately, they were steered towards the ‘right’ choices for school improvement. The choices were then accepted without much protest, but the representatives at the LEA and the headteachers both reported feeling cheated.

As a retrospective communicative notion, sensemaking can be seen in some of the examples discussed here. Promises of tailor-made initiatives could be heard from several of the municipal actors, and stories of government steering carried much resemblance between the different interviews. This could point to the ‘story’ being produced through communication between the municipal actors, especially the headteachers, and being built in retrospect (Weick et al., 2016).

CBS is promoted as a co-operative programme. There is a rhetoric of choice from the SNAE; however, the SNAE demonstrates its power by making the local actors choose the ‘right’ school improvement or by simply offering current in-house training from the SNAE; thus, a form of sermon (Vedung, 2016). Technologies, such as in-house training, are soft as they aim to inform and educate (Kirsten & Wieland, 2017; Moos, 2009). They aim to deliver the best way to ‘do school’ and subject the local actors to the norms of the SNAE. Although CBS promises to offer tailor-made solutions, those offered are what is now seen as best practices. Delivering information late and steering how time should be spent and on what in the municipal schools can also be seen as demonstrations of power. The one who has the information is the one who is in charge. When information is provided late, there is little room for those at the local level to manoeuvre.

Government disruption

We have talked about government disruption instead of government steering [laughter]. … And here we thought that we had a plan which we told the National Agency for Education, but they were not interested in it. Now we are going to do this instead, they said, because you know nothing. No, but we have done this, and the Schools Inspectorate thought it was good, but now we’re not even allowed to do it like that. (LEA1)

The Schools Inspectorate placed demands on the municipality as a consequence of the inspection, which formed the basis for the subsequent selection of municipalities and schools to be offered CBS. The demands of the Schools Inspectorate did not go hand in hand with the SNAE’s school improvement plans. Ultimately, the LEA was left frustrated by the lack of communication between the Schools Inspectorate and the SNAE and told the government agencies that they needed to resolve the issue. The municipality could not work with conflicting requirements from the agencies. This communication failure resulted in the Inspectorate ignoring its demands. The interviewees talked about the fact that the work done by the municipality in response to the criticism made in the inspection was not considered. This is an example of a lack of communication, which also demonstrates that government steering can be disruptive when communication between the state agencies and the municipal level is unclear.

The overall criticism concerning the interaction with the SNAE was that it was time-consuming and cancelled out all other improvement work.

I think it’s good that there is a combination – an external influence is needed, but at the same time, you have to have time for the internal work as well. … When the National Agency for Education comes, there is no time for anything else; it’s just external influence all the time. (LEA1)

Two headteachers clearly stated that their own development ideas had been set aside to make room for CBS. They also looked forward to being able to steer development work in their schools on their own: ‘Now I’m going to dig into the things that have been put on hold for three years’ (Headteacher 3). Other headteachers were more positive about taking part in the CBS programme, despite feeling that the initiatives had perhaps not been ‘spot on’ in terms of what areas required immediate improvement.

At a more general level, there was much talk in the interviews about the lure of government subsidies, which also has bearing on CBS, since the SNAE finances the CBS programme. ‘It’s some sort of Soviet steering, I usually say – by that, I mean government subsidies’ (Headteacher 3). The question raised in connection to the subsidies was whether they were what was needed at that time.

Now we come to a core issue here: many people think that government-financed steering is a problem.In what way?Because the municipalities are forced to do things that may not be right in the process where you are. Instead, it becomes a blunt tool for the state to say in some way that this is what you need. Yes, but not right now. But if you don’t take this, you won’t get any money. Ok, we’ll do it then. We should have done that, but you do it and maybe what you need will come in the form of some initiative later. (LEA1)

LEA1’s statement demonstrates that government steering via economic incentives is conspicuous in the CBS programme and the everyday life of the LEA and of headteachers while also becoming a normative instrument for what is considered ‘good’ school improvement.

The question that seems impossible to answer – Where would we be without CBS? – also featured in what both the LEA and the headteachers said. Some interviewees attributed yielded results to the internal work, not CBS per se. There was an element of stress related to CBS: there was no time for CBS, and it was simply crammed in with the rest of the work, which resulted in the feeling that other important work was being put on hold, as demonstrated in some of the interviewees’ statements. As one headteacher claimed, the school did not have time to work on the criticism by the Schools Inspectorate, which provided the grounds for their school’s selection for CBS. Instead, other important factors, such as improved staff conditions and filling vacancies helped the school improve its ranking during the years of CBS.

However, government disruption was not only viewed as disruptive: the interviewees also spoke about how schools needed the government. In response to the issue of tailor-made initiatives, one headteacher stated, ‘It was not what we asked for, but the initiative we got was good’ (Headteacher 4). In general, there was a high level of trust in the SNAE to provide ‘good’ school improvement initiatives. ‘I have learned that there is basically nothing that the National Agency for Education comes up with that is bad. Everything is good. Then we just have to decide whether or not to be involved, should we be in this or not and do we have time for this so that we can do it really well’ (Headteacher 3). The critical issue, which was mentioned repeatedly, was timing and being able to choose what and when.

The SNAE’s presence and CBS-related work can be seen as a disruptive event (Weick et al., 2016). It undoubtedly breaks with the ongoing work of the municipality, which can of course be said to be the intention. These disruptive events spark communication between the municipal actors, which can also be observed from the similarities in their stories, as previously discussed.

Turning to the steering arrangements, the carrots, namely monetary incentives for participating in certain improvement initiatives, are visible in CBS. The hidden influence, however, lies in guiding the local actors to what is considered ‘good’ school improvement, in setting the agenda and in real-time management (Moos, 2009). There were several indications in the study that the SNAE influenced the municipal actors by setting the agenda, giving best practice advice and controlling time. CBS was time-consuming. It dictated how time – and how much time – should be spent in the schools. The municipal actors showed great trust in the quality of the school improvement initiatives given by the SNAE. However, what the municipal actors asked for in the study was more openness and freedom when determining what and when regarding school improvement initiatives.

The belief that nothing bad comes from the SNAE points to the issue of real power, and this can also be seen as an indication that influence from the national agency affects the local actors much more deeply, or as Moos (2009) puts it, more than regulatory instruments might do. The SNAE controls the agenda and how time is spent. The impression of choice is necessary for a co-operative programme. With choice comes responsibility for the decisions that are subsequently made.

Concluding remarks

The concluding section departs from the questions posed in the introduction with special reference to the second question and the issue of policy instruments, the first being discussed in the results section. The final section also includes some of the study’s limitations and suggestions for future research.

There are signs that the CBS example provided here can be seen as an attempt by the state to regain control and steer the municipality towards the ‘correct’ way of doing school improvement. The perception, primarily among the headteachers and the LEA, is of being pushed in the ‘right’ direction regarding the choice of professional development within the CBS programme, preferably towards the SNAE’s already existing national in-service web education material. This nudging can be seen as steering through soft governance (Moos, 2009). Instead of imposing hard laws and regulations, the national agency has opted to steer the municipal actors towards the ‘right’ path for school development via sanctioned routines and methods (Moos, 2009; Perryman et al., 2018), provided in in-service web education or via approved consultants or higher education institutions. This approach is a way for the SNAE to ensure that the municipal actors receive the education they require and ultimately to try to change both the behaviour and thinking of the municipal actors (Perryman et al., 2018). Using the concepts of sticks, carrots and sermons (Vedung, 2016) in a slightly ironic tone, the schools have already experienced the sticks through the inspections by the Schools Inspectorate. The SNAE enters the pen with carrots – money for school improvement and tailor-made initiatives – so the benevolent state comes along with a helping hand for the troubled municipality. The nudging in the ‘right’ direction regarding the different improvement initiatives that were chosen can be seen as a disguised form of sermon. The schools and LEA do not really know what is best for them, so the SNAE needs to give them a push in the right direction so that they can get the sermons they need.

This article indicates in many respects that CBS promotes ‘what Sweden needs’ through professional education programmes provided by the SNAE, which leads to the question of whether school improvement in the case of CBS can be seen as a means for the state to flaunt strength and re-establish control. Is school improvement about what the school needs to do or what Sweden needs to do to appear resolute in its efforts to compete to be a leading knowledge society (Säfström, 2005)? This situation points to an intricate relationship between the level of the state and that of the municipality; however, it does not point to a reduced role of the state in school issues, which previous literature on the decentralization process has highlighted (Rönnberg et al., 2019). Adolfsson and Håkansson (2021) also pointed out that CBS as a policy phenomenon is in line with the recentralization movement. They raised questions regarding the relationship between the state and the local school actors. However, as Pollitt (2005) said, both decentralization and centralization have merits, and although this article takes the point of view of municipal actors, in the study, this group also recognizes the need of the state to steer them. The question at hand rather concerns the sincerity in that relationship – if trust is to flourish in co-operative initiatives between the state level and the municipal level, perhaps the cards need to be played a little more openly.

This article sheds light on illuminate how the co-operation between the state and municipal actors (headteachers, LEAs and local politicians) in a government-driven school improvement programme is perceived from the municipal actors’ perspectives and analyses these perceptions by drawing on the concepts, among others, of soft governance and certain relevant policy instruments. CBS gives the pretence of being a co-operative programme that builds on mutual trust and respect; however, the actions of the SNAE, as this article has highlighted, can be interpreted as being those of steering and power-wielding. This article does not delve into intended or unintended objectives why the SNAE does what it does; it only discusses how SNAE’s intentions can be interpreted, which is not the same as discussing its real intentions. CBS can be said to concern itself only to educating the municipal actors through monetary incentives and expert knowledge, not co-operation in a real sense (cf. Adolfsson & Håkansson, 2021). The point made in the article is not that the exercise of steering and wielding of power is always repressive; it can be productive and beneficial too. However, a key point of the article certainly is that the exercise of power and use of knowledge are not neutral actions. The extent to which local actors can act is limited, and the steering is not a neutral action involving reciprocal co-operation and decision-making. The co-operation in the given case appears to have been on the terms of the national agency. What this example of CBS indicates is that the steering within CBS resembles much the traditional governing by the state via carrots, sticks and sermons. It is, however, cloaked in co-operation as a political ideal in the educational system, which aligns with the current streams in public governance (Cardini, 2006; Dahlstedt & Hertzberg, 2011).

Power analyses are complex, and one article alone cannot grasp all of its relevant aspects. The focus of this article is on power within a hierarchical institutional context, and it has exclusively highlighted the perspectives of the municipal actors. In so doing, the voices of the state-level actors have gotten excluded. Further empirical and policy research is necessary to critically examine how large-scale school improvement programmes affect power relations between state-level actors and local actors. Moreover, more qualitative studies (e.g. those based on interviews and observations) are needed to deepen our understanding of the complex relationships in programmes such as CBS. Finally, as briefly mentioned above, it would be of great value to conduct a study on co-operation that considers the perspectives of all parties involved.

Acknowledgments

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Disclosure statement

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