The behaviour and interactions of government officers play a pivotal role in shaping public services, making it essential to understand the actors involved and their respective roles within governance processes (Adolfsson and Alvunger 2020; Frederickson et al. 2015). The number of officers employed in the Swedish public sector has increased in recent years (Hall 2012; Laxgård 2025; Statistics Sweden 2025). More specifically, the number of school leaders within local education administrations and schools has grown, driven by the expansion of school units, the broadening of leadership responsibilities, and enhanced administrative oversight (Ärlestig and Leo 2023). A decade ago, Johansson and Nihlfors (2014) conducted a survey of chief education officers in Swedish municipalities, asking them to rank the influence of various actors. That study did not account for a group of school leaders positioned between chief education officers and principals – middle managers such as school form managers and school area managers.¹ This omission was likely due to their limited presence in Swedish municipal administrations at the time.

Since then, this category of school leader has contributed to the aforementioned increase in public sector officers. These individuals are granted authority by the municipality, which acts as the local organiser and provider of education. While both chief education officers and principals are also mandated by the municipality, they additionally hold authority conferred by the state through legislation. Principals received a more explicit and strengthened mandate in the 2010 revision of the Education Act. Chief education officers’ roles were reaffirmed and clarified in an amendment to the same act in 2018 (The Swedish Code of Statutes 2010:800). According to Roos, Johansson and Svedberg (2022), the legislators’ ambition was to clarify mandates and responsibilities and to better manage the intricacies of Sweden’s diverse and decentralised educational system. From a traditional, rules-based perspective on governance, administrative power is inherently hierarchical. Such power (Hysing and Olsson 2012; Weber, Gerth and Mills 1958; Wilson 1887) is frequently attributed to chief officers. To borrow a metaphor from Pink Floyd, officers become “just another brick in the wall”, each brick occupying a fixed hierarchical position. The benefit of clearly defined hierarchical structures lies in their capacity to facilitate accountability and promote transparent, predictable decisionmaking processes. Within this framework, middle managers appear to possess limited power, operating in the narrow spaces between public officers who hold stronger mandates conferred by the state.

However, power is not solely derived from one’s hierarchical position (Bengtsson 2012; Dahl 1991). Indeed, expertise and knowledge constitute significant sources of power for junior officers (Peters, Erkkila and von Maravić 2016). In the absence of a legally regulated mandate, middle managers may seek and utilise alternative sources of power. Their largely unregulated and embedded role within municipal structures could, paradoxically, confer considerable power. Positioned between elected councillors and teachers in schools, they operate at a significant distance from both groups. The behaviour of public officers may be shaped more by their roles than by their formal positions (Magee and Frasier 2014). Roles refer to “conceptions of appropriate goals and activities for particular individuals or specified social positions” (Scott 2014, p. 64), highlighting the importance of normative expectations. This, however, raises concerns about accountability, transparency, and predictability. The absence of formal regulation may render middle managers more susceptible to institutional values and relational dynamics, such as the institution’s logic of appropriateness, which influences perceptions of suitable behaviour (March and Olsen 1984). Different sources of power may yield varying degrees of success depending on the context, as shaped by prevailing norms.

In this article, I seek to describe and analyse the power exercised by middle managers within four local education administrations across three Swedish municipalities, drawing on qualitative interviews with chief education officers, middle managers, and principals. The aim is to deepen understanding of their role in local education governance, given that they are neither chief officers (Huang and Villadsen 2023; Högberg 2007) nor street-level bureaucrats (Lipsky 2010). Neo-institutional theory provides valuable analytical tools for exploring the interplay between institutional and organisational principles in public administration. I examine the forms of power exercised by middle managers in Swedish local education administrations, the institutional perspectives that underpin these practices, and the sources that legitimise their power. First, the empirical foundation rests on three forms of power – power over, power to, and power with – which illuminate the diverse ways in which education officers exert power. Second, the analysis is guided by three institutional perspectives – regulative, normative, and cultural-cognitive – based on Scott’s (2014) institutional pillars, which help explain why middle managers may or may not exercise certain forms of power. While the regulative perspective situates power within formal structures, the integration of normative and cultural-cognitive dimensions offers a more nuanced understanding. Third, the sources of power link these perspectives to the empirical data, illustrating, for example, how personal experience can serve to legitimise power in specific contexts.

Research on Swedish public administrative leadership is surprisingly limited in scope (Blom 1994; Madestam, Sundström and Bergström 2018; Vogel and Masal 2015). Primarily focusing on Sweden at the local level, previous research on education officers has focused on Swedish local education administrations (Håkansson and Adolfsson 2021; Liljenberg and Andersson 2023; Nordholm 2016), the dynamics of power in educational settings (Adolfsson and Alvunger 2020; Hudson 2007; Johansson and Ärlestig 2022) and the actors who exercise such power (Grimm, Norqvist and Roos 2021; Johansson and Nihlfors 2014; Ståhlkrantz and Rapp 2020; Xia, Shen and Sun 2020). Studies have demonstrated how local education authorities collaborate with schools and how national governance influences local governance (Håkansson and Adolfsson 2021; Jarl, Nordholm and Wermke 2024; Liljenberg and Andersson 2023). Research has also examined the relationship between officers and principals (Andersson and Liljenberg 2020; Liljenberg and Andersson 2020) and the distribution of power within educational organisations (Grimm, Norqvist and Roos 2021). Furthermore, we know from previous research that there are no formal regulations or directives governing the competencies of chief education officers; administrations appoint chief education officers and middle managers, and it is thus they who determine their competencies (Liljenberg, Ärlestig and Nordholm 2023). Additionally, there is research on the role and expectations of chief education officers (Bredeson, Klar and Johansson 2011; Jarl, Nordholm and Wermke 2024; Johansson and Nihlfors 2014; Rapp, Aktas and Ståhlkrantz 2022; Roos, Johansson and Svedberg 2022; Ståhlkrantz and Rapp 2020). A few studies have taken an interest in middle managers themselves (Antonsson 2024a; Liljenberg, Nordholm and Ärlestig 2022; Liljenberg, Ärlestig and Nordholm 2023), yet the bulk of existing research either does not take them into account or merely considers them as part of the local education administration. Research on the role of middle managers in education is limited, although there are a few recent exceptions (Adolfsson and Alvunger 2020; Antonsson 2024a; Svedberg and Ärlestig 2024). However, these studies do not explore the power of these managers.

Notably, previous research indicates that when principals in Swedish schools seek support from the municipality, they are more inclined to approach middle managers rather than the chief education officer (Adolfsson and Alvunger, 2020), which may inadvertently grant greater influence to the former.

In the Swedish education system, municipalities hold significant power for adaptation and implementation (The Swedish Code of Statutes 2017:725). Local self-governance is pronounced (Ehn 1998; Lumby 2013; The Swedish Code of Statutes 2017:725; Woods 2016). Power is shared between councillors and municipal officers, as well as among the officers themselves (The Swedish Code of Statutes 2017:725). Decisions are made by politicians and local councillors, while officers and professionals prepare, concretise, implement and follow up on these decisions (Jarl, Nordholm and Wermke 2024; Johansson, Lindgren and Montin 2018). Municipalities vary in demography and size; this, coupled with the freedom to decide how to organise public services, results in different ways of organising local administrations (Antonsson 2024b). Not all municipalities have administrations dedicated to education, but those that do sometimes have middle managers between the chief education officer and principals (Blom 2019; Johansson et al. 2016).

Figures 1 to 4 illustrate the organisational structures of the four local education administrations across the three municipalities included in this article.² Three administrations have one layer of middle managers between the chief education officer and principals; one has two layers. LEA A1 and LEA A2 are part of the same municipality. Each municipality has one or more education committees composed of indirectly elected councillors with delegated decision-making authority. LEA A1 reports to two separate committees: one responsible for preschools and another responsible for compulsory schools. LEA A2 is accountable to a joint committee overseeing both upper secondary education and labour market policy. Both LEA B and LEA C have a single committee responsible for all forms of schooling.

Over the years, a substantial and expanding body of literature has examined different perspectives on power (see, for instance, Bachrach and Baratz 1962; Dahl 1957; Lukes 1974; Weber, Gerth and Mills 1958). Much of this literature focuses on the power dynamics between politicians/local councillors and government officials (Svara 2006; Weber, Gerth and Mills 1958; Wilson 1887). From a new institutionalist regulative perspective, power resides within formal structures, influencing the nature of power. Note that Figures 1–4 illustrate formal structures within each administration but do not necessarily reflect the actual roles undertaken by the officers. Adding normative and cultural-cognitive perspectives to the regulative perspective provides a more comprehensive understanding of why middle managers may or may not exercise various forms of power. These three perspectives constitute or support institutions (Scott 2008; Scott 2014), and they legitimise the exercise of power. National education governance relies on regulative arrangements, while the local level depends more on normative and cultural-cognitive elements (Adolfsson 2024). The three perspectives have also previously been used to study educational leaders (Adolfsson 2024; Adolfsson and Alvunger 2020). They can be described as follows:

Public officers can have power over others and the power to produce outcomes or resist change (Galiè and Farnworth 2019; Pansardi 2012; Pansardi and Bindi 2021). Hierarchical governance depends on top-down control, similar to power over, where power is exerted through laws, regulations, authority, and command systems (Lu, Qiu and Wu 2024; Meuleman 2021). However, hierarchies may falter due to rigidity, goal drift, or poor decisions (Lu, Qiu and Wu 2024). In contrast, power with involves people working together to reach shared goals (Galiè and Farnworth 2019; Pansardi 2012; Pansardi and Bindi 2021). Bureaucratic decision-making is characterised by “dialogue, inclusion, negotiation, and shared power” (Berger 2005, 6). In welfare bureaucracies, teams can function as performative tools, cushioning structures, and so-called “moral corsets”, referring to co-learning and moral pressure to comply with team norms (Jacobsson and Hollertz 2021, 244). Teamwork prioritises unity in thinking and acting over a breadth of competencies, expanding normative control and establishing acceptance of restricted discretion (Jacobsson and Hollertz 2021). Collaboration can be a key strategy for middle managers, allowing them to work within or across teams to solve challenges and improve common understanding (van Niekerk and Jansen van Rensburg 2022). Yet, when norms are too constraining, resignation may be the only option (Jacobsson and Hollertz 2021).

Power can be exercised through various resources (Giddens 1979), including the acceptance of others (Grimm, Norqvist and Roos 2021; Haugaard 2018). Those exercising power and leadership require such acceptance, making the creation of authority a shared effort (Woods 2016). Holding a leadership position means being in authority, but not necessarily being recognised as an authority (Grimm, Norqvist and Roos 2021; Haugaard 2018) – a distinction that requires sources of power beyond formal position. For public officers, professional expertise can serve as one such source (Svara 2006), along with the ability to alter others’ incentives and reputations (Dowding 2019).

In sum, the middle managers in this study may possess various forms of power. Power over refers to formal authority exercised over others. Power to involves being in a position of authority and being recognised as such. Power with denotes shared authority. The findings also shed light on the absence of these forms of power – highlighting instances in which middle managers lack power over, power to, and power with.

This study is situated within a single-country context. It is based exclusively on interviews conducted over 13 weeks in 2023 with education officers from four local education administrations across three Swedish municipalities. Given the aim of exploring perceptions of power, qualitative interviews were deemed an appropriate methodological approach. The interviewees are comprised of chief education officers, middle managers, and principals in order to explore perceptions of the middle manager role from the perspective of middle managers themselves, as well as those of their superordinates and subordinates. I carried out a total of 18 interviews, each lasting between 35 and 97 minutes (Table 1).

Informed consent was considered in the research process. The participants received written information along with a brief oral explanation and were asked to complete a consent form for participation as well as a separate form relating to GDPR compliance. I conducted individual semi-structured interviews, primarily at the respondents’ workplaces, with a few held via Zoom. Regardless of their position, the role of the middle manager remained central throughout the discussions.

Participants were asked about power both directly and indirectly (see Appendix I).⁶ From the outset, I adopted the new institutionalist theory as the overarching framework for the study. The categories – power over, power to, and power with – are empirically and theoretically derived. As I encountered these concepts, I recognised their relevance to the material and found them useful for presenting the findings. Additionally, institutional perspectives emerged organically throughout the course of the analysis.

I recorded the interviews and kept notes during the process. The interviews were transcribed shortly afterwards using electronic transcription, followed by manual verification. Triangulation was not employed. I have utilised qualitative content analysis, and the data were coded using NVivo, where the material was organised into the categories power over, power to, and power with. This proved to be a delicate task, as certain excerpts could reasonably fit into more than one category and therefore required careful consideration. In some cases, the same piece of information was coded under two categories. For example, the following quotation (not included in the article) by chief education officer Carl can be coded as both power to and power with: “It applies to appointments too. They are not allowed to appoint a principal by themselves without my involvement.” This statement suggests that middle managers possess the power to appoint principals, but only insofar as they exercise this authority power with their superordinate.

The interviews were conducted in Swedish, and quotations have been translated into English. When translating, I aimed to remain as faithful as possible to the original quotation, while ensuring the result adhered to the conventions of the English language.

Both a theoretical and an empirical paradox emerge from this article. In brief, the theoretical paradox lies in the fact that none of the three pillars alone can adequately explain the power of middle managers. Rather, they must be considered collectively, not least since the pillars encompass contrasting elements, such as regulations and norms. The empirical paradox concerns state regulations. In its efforts to assert control over education governance, the central government has introduced regulations targeting both principals and skolchefer. The local level has consequently responded by appointing other officers – who are not subject to state regulation – to oversee governance processes and carry out the necessary work.

Theoretically, the paradox lies in the fact that none of the three institutional pillars alone is sufficient to explain the power exercised by middle managers, nor to reliably categorise groups of middle managers in distinct categories of administrative power. Take the regulatory pillar, for instance; although it remains highly relevant, its explanatory capacity has somewhat diminished in the 2020s. The traditional legal-bureaucratic model is based on uniform and enduring rules that govern the relationship between public administration and citizens. Officers are expected to act with neutrality, integrity, and impartiality. However, the rigidity of this model represents a significant weakness (Rothstein, 2023). While it is well-suited to routine tasks in stable environments, it proves inadequate when faced with challenges that demand innovation and adaptability (Petersson, 2021), rendering it increasingly ill-suited to developments within the Swedish educational governance system in the 2020s.

In theory, the regulative, normative, and cultural-cognitive perspectives legitimise the exercise of different forms of power. The findings in this article provide examples of how middle managers exercise power over, power to, and power with, though the statement carries nuances. The forms of power can shift depending on the situation and are sometimes constrained. There is a need to look beyond the regulative lens. In this article, I show that normative and cultural-cognitive perspectives are vital for understanding how power is exercised by middle managers in Swedish local education administrations. Notably, when multiple dimensions govern institutions, complexity inevitably arises. I demonstrate that when middle managers act beyond their formal hierarchical roles, their exercise of power can be legitimised with recourse to any of the three institutional perspectives, which may serve as interpretive frameworks at best. These perspectives help explain the structures intended to constrain actors; however, individuals may still make decisions that challenge or circumvent these constraints.

Figures 1 to 4 illustrate that the number of middle management layers between chief education officers and principals ranges from none to two. Fewer layers may facilitate clearer lines of accountability. In the findings section, I highlight the opaque distribution of power in LEA B. While accountability is strengthened when “authority, mandates, responsibility, rules, standards, goals, and expectations” are clearly defined and consistently applied, ambiguity undermines its effectiveness (Olsen, 2014: 107). Operating amid role ambiguity and multiple stakeholders, middle managers may make poor decisions without consequences – or be held accountable for decisions they did not make. This risk can be mitigated if institutions succeed in aligning middle managers’ norms with those of the municipality (DiMaggio and Powell 1991; Hall and Taylor 1996).

Empirically, while the regulation of principals and skolchefer has, to some extent, reinforced state control, my findings indicate that it has also had significant implications at the local level. In effect, the state’s governance dilemma has merely been relocated to the municipal level, at a time when the traditional legal-bureaucratic model no longer functions as the sole governing mechanism. The emergence of often-powerful middle managers has introduced additional complexity into the governance structure – not primarily at the central level, but locally. In other words, the state’s attempt to reduce complexity has, paradoxically, increased it at the local level. The growing number of middle managers is, in itself, a paradox. Furthermore, Table 1 illustrates that, in certain cases, the formal mandate of the skolchef does not reside with the chief education officer, but has instead been delegated to middle managers within the local education administration. This introduces an additional layer of complexity, as it cannot be assumed with certainty that the state mandate is held by the most senior officer. Additionally, the expanding presence of middle managers has given rise to uncertainties regarding the scope of their power, as well as pressing concerns about conflicting expectations and accountability.

First, let us examine the opportunities and challenges middle managers face in attempting to exercise power over, power to, and power with, and how regulative, normative, and cultural-cognitive perspectives shape their ability to enact these different forms of power.

Echoing previous research, I demonstrate in this article that while responsibilities are often delegated downwards, power tends to remain concentrated at the top. Middle managers are assigned tasks by chief education officers, but do not always hold the mandate to exercise power over subordinates. Nonetheless, they possess a degree of budgetary power, formally held by the municipal executive committee but sometimes delegated. Despite lacking a clearly defined legal mandate, their hierarchical position and financial power render them powerful actors. In this context, the principal’s formal mandate becomes comparatively less significant. For some middle managers, budgetary control is a key source of power; for others, collaborative forms such as power with are more central. In some cases, the absence of budgetary responsibility can even be experienced as a relief. Overall, the type of power exercised by middle managers varies and is shaped by local norms and cultural-cognitive factors.

Furthermore, in this study, I highlight variations in perceptions and interpretations of power, particularly concerning who has the power to act. One contested area concerns schools’ internal organisation. As recounted, chief education officers believe middle managers can do more, while middle managers argue that principals control internal organisation and that principals’ tangible power circumscribes their own. Although middle managers are hierarchically above principals, their interpretation of their role can constrain their actions. Thus, multiple institutional pillars are at play. While certain regulations position middle managers above principals, others – such as those concerning defined mandates or the absence thereof, as well as normative and cultural-cognitive perspectives – may serve to invert the hierarchical structure entirely.

In line with previous research, middle managers also exhibit varying degrees of power with, shaped by a context informed by regulative, normative, and cultural-cognitive perspectives. The findings present contrasting examples: one chief education officer enforces a strict hierarchical structure, while another promotes a flatter, consensus-driven approach. The data suggest that close collaboration fosters broad co-determination, whereas hierarchical models provide clearly defined mandates within more limited domains. Furthermore, shared power can either facilitate equitable distribution or obscure underlying imbalances. Given the varied contexts in which middle managers operate, their performance may differ across environments. A manager undervalued in a hierarchical setting might thrive in a collaborative one, while someone suited to hierarchy may struggle elsewhere.

Second, let us consider how expectations sometimes come into conflict. The types of power middle managers exercise are linked to their roles. They face conflicting expectations from both chief education officers and principals, and risk subordination to both, as each holds a stronger state mandate. While chief education officers believe middle managers could assume greater authority, middle managers often feel constrained by principals’ mandates. For instance, consider the example of middle manager Alex, who is willing to relinquish his power in favour of the principals, even when this contradicts the opinion of chief education officer Alice. These perceptions shape the kinds of power middle managers can exercise, echoing previous research stating that actors’ perceptions influence governance outcomes. Middle managers must also navigate which institutional logic – regulative, normative, or cultural-cognitive – takes precedence. Middle managers often experience role conflict, such as uncertainty over whom to follow and whether to prioritise broader local school governance or their specific school form or area.

Third, let us consider the issue of accountability within this context. As previously noted, in response to emerging needs, many municipalities have appointed non-regulated officers to take on responsibilities that fall outside the scope of formally regulated roles. Building on this article’s findings – that power may not reside with the one initially assumed – the issues of accountability and predictability become pressing. Although accountability and predictability are common features of the regulative pillar, the influence of normative and cultural-cognitive perspectives raises concerns about whether these aspects are adequately safeguarded. Middle managers’ actions are partly shaped by personal interests, whether focused on children’s welfare, career progression, or external agendas. When they deviate from established practice, perform poorly, or act with questionable intent, their relatively unregulated power can lead to unpredictable outcomes for professionals and citizens, potentially causing greater disruption than more tightly governed actors. However, suppose power rests with competent and well-intentioned officers. In that case, their involvement can be beneficial – provided they exercise their power wisely and in ways that serve the best interests of the municipality. Nevertheless, the overall difficulty in determining who, in reality, holds power within the system remains a significant challenge, as it complicates the vital issue of accountability – a fundamental principle of any democratic framework.

In this article, I have adopted an exploratory rather than a comparative or generalising approach. In doing so, I have laid the groundwork for a more extensive comparative study to be undertaken in future research. To conclude, I demonstrate that middle managers possess various forms of power, and that both the scope and the manner in which this power is exercised vary considerably. In the discussion, I also highlight two paradoxes – one theoretical and one empirical – which underscore the complexity inherent in the governance of education and its scholarly examination. What remains clear is that middle managers are powerful officers, far from being “just another brick in the wall”.

No specific funding was received for this article.

The author declares that there is no conflict of interest.