Breaking the Vicious Circle of Escalating Control: Connecting Politicians and Public Employees through Stewardship

Abstract

Politicians applying general rules as a reaction to local failures has contributed to mushrooming control in the public sector, which has in turn spurred higher transactional costs and motivation crowding among public employees. Drawing on a qualitative case study in a Danish municipality, this article explores the prospects and challenges for politicians of breaking the vicious circle of escalating control by adopting stewardship ideals into their leadership of the public employees. The results show that stewardship offers new opportunities for politicians, enabling better diagnosis of control problems, more robust control solutions, as well as a pronounced mobilization of employee support for those solutions. However, political competition, political discontinuity after elections, scandals in the press, resistance in the administration, and more diffuse decision-making processes pose potential challenges for politicians striving to tackle the problem of escalating control through stewardship.

1. Introduction

Bureaucratic control constitutes the very foundation of democracy and of reliable systems of governance (du Gay 2000; Weber 1952). However, in the wake of New Public Management Reforms, the tendency for control to mushroom, create problematic side effects, or produce higher transactional costs when implemented in practice is increasingly discussed as a growing challenge in the public sector (Bozeman and Feeney 2015; De Jong 2016; DeHart-Davis 2009). Critics claim that not only has NPM failed to effectively de-bureaucratize classical bureaucracy, but has added to the problem by introducing overly simplistic, competitive performance measurements. An extensive focus on outputs intended to produce more local autonomy has, instead, spawned new layers of control, as performance goals and procedures for evaluating them mushroom (de Bruijn 2002; Budd 2007).

The bureaucratic ideal of a clear demarcation between the administration and politics, coupled with neoliberal ideals of monitoring at “an arm’s length” have contributed to high power distance between politicians and public employees (Hood 1991; Pollitt and Bouckaert 2011). According to agency theory, principals should aim to use hands-off institutional power, and refrain from using personal forms of power, since the self-interested agent is not expected to respond to prosocial or intrinsically motivated arguments (Davis et al. 1997; Hernandez 2012; Schillemans 2013). Hence, direct interaction between political leaders and local employees is not encouraged or even considered relevant. However, critics have problematized the inherently conflictual basis of principal–agent theory, which is blamed for eroding trust between the political and administrative arenas, spurring excessive control, which in turn crowds out motivation among public employees (Bernstein et al. 2016; Frey 1994; Le Grand 2003; Schillemans 2013). The classical practice of politicians handling local mistakes with the introduction of more general rules, has led to a vicious circle of escalating control: Time spend on citizens is increasingly converted into time spend on excessive and demotivating control, which in turns increases the likelihood of local mistakes which politicians again have solved by increasing control.

Against this backdrop, stewardship theory has drawn growing scholarly attention as an alternative approach to the study of governance. Whereas agency theory rests on the idea that all actors pursue rational self-interest and extrinsic goals, stewardship theory suggests that public employees are fundamentally intrinsically driven and are likely to pursue collective goals, acting as stewards of their principal’s interests (Davis et al. 1997; Schillemans 2013). Unlike in agency theory, high power distance and detailed performance control are seen as smothering for the self-regulation of prosocial stewards. Instead, stewardship ideals recommend reducing institutional control, enhancing employee discretion, and encouraging principals to develop hands-on strategies for connecting and building trust with them (Schillemans and Bjurstrøm 2019).

Although empirical studies on stewardship and the way it transforms leadership are growing (Torfing and Bentzen 2020; van Dierendonck 2010; Hernandez 2007; Schillemans 2013), research so far has almost exclusively focused on administrative leadership (Caldwell et al. 2008; van Dierendonck 2010; Hernandez 2007; Schillemans and Bjurstrøm 2019). However, what about politicians, who, despite their primary role as democratic leaders of citizens, certainly play a role in decisions which may spur mushrooming control of public employees? According to Schillemans, politicians should provide the direction and conditions for organizational development towards stewardship, but beyond such broad recommendations, it is not clear how stewardship may be relevant to political leaders trying to solve problems of mushrooming control. The ambition of this article is to contribute to filling this research gap by addressing the following research question:

The study sets out to explore how ideals of stewardship may assist politicians in breaking vicious circles of escalating control, as well as the prospects and challenges that this implies for their leadership.

First, drawing on agency theory (Aucoin 1990; Verhoest et al. 2004) and stewardship theory (Davis et al. 1997; Schillemans 2013; Schillemans and Bjurstrøm 2019), a theoretical framework for studying political approaches to public employees is developed. Second, the case selection and methods are accounted for, and the framework is then applied to a case study of a Danish municipality where politicians are actively trying to solve problems of escalating control by adopting stewardship ideals in their relationship with public employees. After that, the analysis is presented, showing how politicians integrate stewardship ideals into existing agency practices, to solve problems of escalating control. The analysis points to several opportunities and challenges for politicians engaging in stewardship practices, which involve a more direct and interactive relationship with public employees. The study’s limitations as well as implications for practice and future research agendas are discussed, before the conclusion is presented.

2. Theoretical Framework

First, general characteristics of political leadership and its distinctiveness from other forms of leadership are outlined. The somewhat neglected relationship between political leaders and public employees is also briefly discussed, followed by an outline of central features of agency theory and the contrasting principles of stewardship theory, to lay the foundation for the theoretical framework that will be applied later in the analysis.

2.1. Political Leadership

Political leadership shares several characteristics with other forms of leadership in that it is a reciprocal, contextual process that involves not only the leader herself, but also her potential followers (Bolden et al. 2011; Parry and Bryman 2006). However, political leadership is also distinct in several ways: political leaders are democratically elected representatives who are given a mandate for a limited period. Hence, they are, by definition, vulnerable to deselection and must maintain their voters’ trust to retain political power. Politicians operate within a party government structure based on negotiation and the formation of position and opposition (Haus and Sweeting 2006).

Democratic ideals and the way political leadership is carried out are not constant but evolve in an interplay between context and societal challenges (Bentzen et al. 2020; Torfing et al. 2020). Still, some characteristics recur in several definitions of political leadership. One is agenda setting, which involves analyzing the situation at hand and choosing the problems most in need of political attention (Greasley and Stoker 2008; Kellerman 2015; Leach and Wilson 2002). A second characteristic is the task of developing robust solutions to diagnosed problems (Gissendanner 2004; Kotter and Lawrence 1974; Leach and Wilson 2002; Tucker 1995). Finally, political leaders must mobilize support among affected and involved actors, who rely on that support to implement solutions or responses to perceived problems (Kotter and Lawrence 1974; Svara 1990; Tucker 1995).

A main difference between political leaders and administrative leaders is their domain of leadership. Traditionally, political leaders are primarily seen as leaders of communities and citizens, whereas administrative leaders lead public organizations. Although the relationship between politicians and administrators at the central level has received extensive attention in several important studies (Hansen and Ejersbo 2002; Lee and Raadschelders 2008; Lee 2006; Mouritzen and Svara 2002; Svara 2006), the relationship between politicians and public employees in local-level public organizations remains radically understudied. One compelling reason for this is the inherent democratic ideal of the separation between politics and the administration, which emphasizes accountable decision-making processes (Lee and Raadschelders 2008; May and Winter 2009; Weber 1947). New Public Management Reforms anchored in ideals of agency have further increased high power distance and lack of political contact with local actors in public organizations (Greve et al. 2016; Pollitt and Bouckaert 2011). Hence, politicians are generally seen as inhabitants of a separate arena of leadership that may be dependent on, but is not tailored towards, public employees. Hence, there has been no real reason for politicians to build a relationship with public employees or to seek leadership of them.

However, problems caused by escalating control require a reconsideration of existing agency practices such as high power distance between politicians and local-level employees. The classical dynamic of politicians seeking to solve local failures by introducing general rules has exacerbated the vicious circle of escalating control, and existing agency practices have only served to reinforce this. Hence, the many grave consequences of mushrooming control as well as its side effects such as creaming, parking or other fending mechanisms (Lipsky 2010), call for new approaches by political leaders, who must reconsider their existing agency practices.

In the following, we turn first to agency theory, and then stewardship theory, to construct a theoretical framework for analyzing how politicians try to prevent escalating control through their approach to local-level employees.

2.2. Agency Theory

Principal-agency theory is the most influential perspective in explaining how to secure accountability in the delegation of human services in public organizations (Brehm and Gates 1999; Kearns 1996). Rooted in the field of economics, agency is essentially a conflictual theory, which presupposes that actors are self-serving and will pursue their own interests when possible (Eisenhardt 1989; Maggetti and Papadopoulos 2018). Hence, the “model of man” underlying agency theory is that of the self-interested actor who maximizes personal gain (Donaldson and Davis 1991; Schillemans 2013).

A principal–agent relationship can be defined as a contract in which a principal engages another person (the agent) to perform a task on their behalf, which involves delegation of some autonomy (Jensen and Meckling 2012). Agency theory has traditionally assumed that there is an inherent conflict of interest between the principal and the agent because the two parties’ goals are incongruent. Hence, without the agent’s control, the principal’s goal is unlikely to be attained (Caers et al. 2006; Eisenhardt 1989). Adding to the principal’s problem is informational asymmetry, which occurs when the principal lacks knowledge about the agent’s characteristics and efforts. Hence, the agent is expected to use this advantage in the pursuit of egocentric goals, making it prudent for the principal to limit potential losses by developing control tools to keep the agent in check (Jensen and Meckling 2012; Schillemans and Bjurstrøm 2019).

Although principals are forced to delegate some degree of autonomy to agents, numerous accountability systems are installed to delimit this autonomy actively through controls. Hence, a central challenge for principals is to gain information about possible straying by the agent, and to enforce correction when needed (Dunn and Legge 2001; Jensen and Meckling 1976; Waterman and Meier 1998). In other words, principals aim to reduce risk through control.

This requires systems of external monitoring and correction, which are designed to ensure that agents keep their focus on organizational goals and avoid agency drift (Caers et al. 2006; Fox and Hamilton 1994; Schillemans and Busuioc 2015). External monitoring also enables the use of rewards and punishments (carrots and sticks) designed to support extrinsic forms of motivation, to which agents are expected to respond positively.

Agency theory recommends high power distance in formal, hierarchical relationships, since agents are not expected to respond responsibly to offers of shared power. Therefore, principals primarily draw on formal, institutional power to keep agents in check, since the latter are not seen as responsive to softer, personal forms of power (Donaldson and Davis 1991). Institutional power is characterized by being anchored in formal roles in the hierarchy, enabling principals to legitimately reward or sanction agents through incentives or coercion (Davis et al. 1997; Frost and Moussavi 2011). Hence, agency involves the use of formal power bases, which are granted and withdrawn from the principal, according to roles and organizational membership (Davis et al. 1997). In the case of politicians as principals, formal power is anchored in the sovereignty of them having been democratically elected.

2.3. Stewardship Theory

Stewardship theory, which has roots in psychology and sociology, was first introduced by Davis, Schoorman and Donaldson as an alternative to agency theory in the private sector (Davis et al. 1997). More than a decade later, the notion of stewardship was taken up in public sector debates about leadership (Dicke and Ott 2002; Schillemans 2013; Van Slyke 2007).

Although stewardship theory is equally preoccupied with the question of how to secure accountability when a task is delegated from a principal to a subordinate, it entails fundamentally different assumptions about employees’ motivation, the extent to which principals’ and subordinates’ interests collide, and the type of leadership that is called for. Hence, a steward of an organization is someone who demonstrates a commitment to the best interests of the organization, and who willingly subjugates his/her own personal interests to the protection of others’ long-term welfare (Hernandez 2012; Schillemans 2013). Stewards place a higher utility on serving the ongoing needs of others and preserving collective resources rather than on ensuring personal gain (Hernandez 2012).

Hence, stewardship theory proposes that there are conditions under which subordinates are likely to willingly serve collective, prosocial goals and act as stewards of their principals’ interests (Contrafatto 2014; Donaldson and Davis 1991), so that principals’ and stewards’ goals become congruent or overlapping. Consequently, the foundational conflict between principals and subordinates upon which agency theory relies, fundamentally fades or even disappears, which also relieves the need for external monitoring (Block 2013; Caers et al. 2006). Rather, principals aim to support bounded self-regulation through involvement and the development of shared goals and norms (Davis et al. 1997; Lawler 1992; Schillemans 2013). However, principals aims to reduce risk by controlling agents, they aim to absorb risk by building trust with stewards (Luhmann 2017; Mayer et al. 1995). When stewards make mistakes, this is likely to be due either to lack of competences or insufficient alignment with the principal, rather than a question of self-serving behavior. Hence, correction as a management tool is not very useful and may even undermine stewards’ motivation. Instead, principals must rely on empowerment and acknowledgement of stewards through training, developmental challenges and collective learning (Davis et al. 1997; Hernandez 2012).

Because stewards are intrinsically motivated to comply with organizational goals, institutional power need not be exercised in the same way. Using the coercive power of extrinsic rewards may even undermine stewards’ prosocial motivation (Frey and Jegen 2001; Frey and Oberholzer-Gee 1997; Tosi et al. 2003). Instead, principals rely on more informal, personal power forms, such as referent or expert power (Davis et al. 1997; Hernandez 2007). Personal power cannot be granted, but is always developed over time and requires investment in terms of building and maintaining relationships with stewards. Although slower to develop, personal power can be sustained over longer periods, and is less dependent on hierarchical legitimacy: while institutional power can be given and taken away swiftly, personal power takes time to grow but is also more robust in the face of changes in formal roles (Davis et al. 1997).

Since stewardship fundamentally relies on relationships and shared goals between principals and stewards, low power distance is vital. When principals reduce power distance, this prevents stewards from distancing themselves from their principals, and helps to foster bonds of loyalty and trust. When power differences are toned down, this also allows stewards to internalize organizational values and goals through dialogue with principals, decreasing the need for external monitoring and control (Schillemans 2013). Finally, reducing power distance enables a transition in which formal power holders distribute or share power among stewards.

2.4. Agency or Stewardship: Subsidiary or Complementary?

There have been many attempts to champion either agency theory or stewardship theory as the “best way” to conduct governance. Although recent results of empirical studies show partial support for stewardship, empirical findings do not unequivocally confirm the superiority either of agency or stewardship theory (Daily and Dalton 1994; Davis et al. 1997; Donaldson and Davis 1991; Schillemans 2013). Rather, research stresses the need to approach agency and stewardship as parallel theories which, depending on cultural, contextual and situational factors, interplay in various combinations. Hence, stewardship should not necessarily be seen as a replacement for agency theory, but as a competing “model of man” which may modify, balance or transform existing practices of agency (Caers et al. 2006).

Below, in

Table 1. Contrasting ideals and practices of agency versus stewardship in the political leadership of public employees.

	Agency Approach	Stewardship Approach
Political principals’ perception of employee motivation and goals	Conflict of interests between principal and agent.	Congruent or overlapping interests between principal and steward.
	Extrinsically motivated agents/ Egocentric goals/Lower order needs (Economy, security)	Intrinsically motivated stewards/collective goals/Higher order needs (Self-actualization, acknowledgement)
Management style of political principal	External monitoring and control delimiting delegated autonomy (Reducing risk by enhancing control)	Involvement and bounded self-regulation enabling delegation of extensive autonomy (Absorbing risk by building trust)
Leadership style of political principal	Correction and rewards	Empowerment and acknowledgement
	(Incentives and coercion)	(Training, personal development and collective learning)
Power base of political principal	Institutional power	Personal power
	(Democratic sovereignty)	(Referent and expert)
Power distance	High power distance (Power is kept from agent)	Low power distance (Power is shared with steward)

, the contrasting ideals and practices of political leadership based on agency versus stewardship are outlined. This conceptualization will also guide the analysis which, inspired by previous findings, will focus on the parallel use of agency and stewardship practices among political leaders, and the opportunities as well as challenges involved in a combined approach.

3. Methods

The empirical data were collected in Copenhagen municipality which, since 2012, has actively pursued a “Trust Reform” aiming to actively reduce excessive control of public employees to avoid unnecessary transactional costs and the crowding out of employee motivation. The reform involved numerous initiatives and changes at multiple levels of the organization and ushered in changes for various actors, including politicians. This study, however, delimits itself to exploring changing ideals and practices among politicians regarding their leadership approach towards the public employees. Hence, the wider scope and implications of the trust reform lies beyond the scope of this study. In 2013, a politically decided “Code of Trust” was introduced, outlining new contours for political leadership that emphasized the need to build trust with public employees, expand their autonomy, and avoid over-regulation and detailed control. A budget agreement from 2015 reconfirmed the high priority of the reform, and underlined that trust between political leaders and public organizations was vital to prevent escalating control. The pronounced ideals behind the reform, which resonate with the notion of stewardship, make it an ideal case for studying politicians who are attempting to solve problems of excessive control by adopting stewardship ideas in their approach to public employees (Flyvbjerg 2006).

Copenhagen municipality, with 45,000 employees, is the largest public organization in Denmark, which makes it far from a representative case. However, the size of the municipality, the distance between politicians and employees, and intense focus by the press, enable a study in an environment in which patterns, challenges and opportunities regarding escalating control are expected to be pronounced.

Denmark is known as high-trust country, and Danish municipalities are highly decentralized and granted extensive autonomy (Ladner et al. 2016; Houlberg and Ejersbo 2020). This provides an interesting case in which politicians have relatively extensive room for maneuver for pursuing stewardship ideals in their endeavors to solve problems of escalating control. However, the extreme nature of this context must, of course, be taken into consideration when discussing the limitations of the results.

Politicians with a clear formal leadership role, such as being a mayor or committee leader, receive special attention in this study. However, the question of political leadership of public employees is also discussed by political members of committees, and by all elected politicians at the city council. As such, political leadership is seen as a process relevant to all politicians, although formal political leaders undertake a special responsibility.

Public employees are broadly defined as both frontline workers and leaders at the local level of public organizations. Hence, administrators and leaders at the central level in the administration, who traditionally collaborate directly with politicians, are excluded from the definition of public employees.

The study triangulates several methods, combining documents, semi-structured interviews, focus groups and observations. Interviews with politicians, as well as local-level actors, have been given special emphasis, in light of the research question guiding the study. However, administrative personnel also provided important information about changes in political approaches to solving escalating control, as well as challenges associated with this development. Observations provided tacit knowledge as well as first-hand insights into political behavior. Documents constituted important sources of knowledge about formal political agreements and explicit policies regarding reforms to de-escalate spiraling control. Types of documents selected, actors interviewed, and events observed are outlined in

Table 2. Overview of data.

Empirical Data	Type of Data	Total Number
Documents	Official policies on the trust reform (3) Folders/written communication to local level (3) Manuscript/slides for political speeches aimed at the local level (4) Internal agendas/recommendations/policy drafts (4)	14
Interviews	Political leaders (3) Top administration (3) Employees in administration (6) Local leaders (6) Union representatives (3)	21
Focus group interviews	Employees (2 focus groups with 4 employees in each group)	2
Observations	Political speech at internal conference for leaders (2) Meetings at city council (2)	4

, below. The empirical data were coded according to the central signifiers of agency and stewardship outlined in Table 1. In addition, a more open coding framework was used to identify opportunities and challenges for political leaders engaging in new initiatives inspired by the idea of stewardship. All empirical data were coded and analyzed in Nvivo.

4. Analysis

4.1. General Trust in Stewards—While Spotting and Handling Agents

The trust reform triggers new political debates about public employees and their fundamental motivation. Previously, politicians have been quick to sanction local mistakes by introducing general sanctions in the whole sector. A central shop steward explains: “Every time one person made a mistake 900 people had to face the consequences”. This general suspicion of employee tendency to slack, if given the chance, develop towards a perception of public employees towards being more pro-socially oriented, with the trust reform.

“You need to trust that all employees are doing the best they can and vouch for that”

(Politician)

This change in perception is most vivid when local mistakes or problems are brought to the attention of the political level, who must decide how to interpret such difficulties. The Code of Trust stipulates that: The political answer to critical cases in the press must be focused initiatives towards the concrete, local problem. This requires exploring the source and the extent of the problem, and securing help for leaders in the areas involved—while maintaining trust in the rest of the organization (Code of Trust). Hence, politicians agree that employees are generally motivated to comply with political decisions, and mistakes are therefore attributed to lack of competence rather than lack of prosocial motivation.

However, politicians still underline that although most public employees are trustworthy stewards, there will always be a few “bad eggs” who are driven by egocentric goals. In such cases, political leaders draw on practices anchored in agency, including sanctioning and increased control. However, the new ideals of stewardship modify previous practices by isolating “bad cases” from the “good crowd” and treating them according to their motivational basis. Identifying agents from stewards requires politicians, who invest more time in contacting their organization as well as close collaboration with the administration, who can provide in depth knowledge and data about local actors.

“We had a case at a Home Care Center where some of the employees cheated with registering work hours. That case fundamentally raised questions about the whole profession. But I kept saying ‘These employees are isolated cases’. I simply refused to scold all the stalwart ladies working in this sector because of a few bad eggs. Yes, there have been some misdemeanors, but all in all our employees are doing well”.

(Mayor)

This pattern of generally expecting employees to be working loyally towards organizational goals, while still spotting, isolating and sanctioning ego-centrically driven agents, shows how stewardship is used to modify, rather than replace, existing practices of agency.

4.2. Reducing, Differentiating and Co-Creating Control

A central aspect of the reform emphasizes the need to remove, reduce or simplify existing politics, strategies, measures and control. Hence, a clear empirical pattern is a development towards less external monitoring and control with the ambition of delegating more autonomy to the local level. This ambition is outlined in the Code of Trust:

“Politicians must have the courage to let go of detailed control. Instead, as much local autonomy as possible should be given to leaders and employees in order to enable them to execute their professionalism and focus on their core tasks”

(Code of Trust)

The new political ambition to avoid undermining local autonomy by introducing more intensive control systems is also pronounced by politicians themselves: “If we mean this seriously, we must try to remove all excessive documentation and regulation” (Speech at internal seminar by mayor).

Previously, the political focus has been primarily on securing control and documentation as a foundation for service delivery and political decision-making. Now, politicians have become more conscious of how their use of regulatory strategies affects the trust, autonomy and resources of frontline workers in the organization. Concretely, several projects have been initiated with the aim of identifying and reducing redundant rules and documentation systems. Rather than reducing risk by enhancing control, politicians now aim to build trust with public employees as a new means of absorbing risk.

Although the introduction of new forms of control, regulation or documentation requirements is certainly associated with more careful political reflection, external monitoring is still seen as a vital and legitimate strategy in political leadership.

“The trust reform does not mean that all control is bad. Most existing documentation requirements cater to a legitimate need to reduce risk for citizens. It is important that we maintain that protection—for everybody’s sake”

(internal speech—Mayor SFO)

Hence, elements of monitoring and control in a classical agency approach still play a central role in politicians’ management style. Although external control is still maintained to some degree, a central impact of stewardship has been the growing political awareness that the need for regulation may vary significantly in different areas and contexts of an organization:

“We (politicians) have had a tendency to standardize regulation too much. Now there is a development in the direction of differentiating control according to areas and also according to local needs”.

(Committee leader in the area of Social Care)

In other words, political leaders regulate in a more differentiated manner, and scale their control in tune with local contexts. For example, budget control of experienced leaders, who have a long history of managing their economy responsibly, may only be carried out once a year, while new leaders who are still learning may be subject to control four times a year. Hence, control resources are allocated according to the concrete context, risk, task and organizational need, rather than being implemented evenly across the organization.

An interesting development is the increasing tendency for politicians to involve the local level in the design of control systems. This is, for example, the case when regulation or control systems are co-created in collaborative processes with local-level leaders and employees, who are invited to express their needs, concerns and ideas, which are then included in the design of control systems. This extensive involvement in designs of control, not only provide politicians insight into local-level aspects of control, but also allow local actors to develop ownership of solutions and to internalize control.

4.3. Empowering Stewards While Sanctioning Agents

Although empowerment of local stewards is still a relatively new area of political interest, several respondents observe a growing political focus on the need to supporting the public organization better in its self-regulation.

“Usually, political leaders have always paid attention to citizens and things that might catch the eye of the press. The ‘machinery’ aspects like the development of leaders or frontline workers—that was left to the administration. That has changed”.

(Union Representative)

Although local empowerment is still primarily considered a task for administrative leaders, political leaders have refocused their attention. Discussions about how to create the right framing of the administrative organization are now accompanied by reflections on the need to build and support employees’ competences so that they can navigate more autonomously. For example, a political leader in the elderly care area emphasizes the need to accompany the removal of detailed control with extensive writing courses for personnel to help them develop a more nuanced professional language.

In addition to a growing political focus on building the needed competences and capacity among public employees, another growing political concern is how to support and help leaders and employees through the transformation connected with self-regulation: “It is crucial that we take care of employees in the leap from rules to autonomy” (Political leader at meeting). This and other examples show how politicians include strategies of training, professional development and organizational learning as central elements in the empowerment of public employees. Several politicians explained that building competences and empowering the local level are prerequisites for reducing control. “It is no use setting them free if they don’t know how to use their freedom” (political leader). Hence, empowerment is used actively to build the competences needed to navigate with extensive autonomy.

Another development is the growing acknowledgement of local leaders who succeed in transforming their leadership towards ideals of trust. Rather than rewarding them financially, politicians publicly acknowledge such leaders as good examples, co-perform with them at conferences, and give them credit for results achieved. “That really made me proud. Standing there with the mayor who told everybody about our results” (Local leader).

Although the political focus on empowerment and acknowledgement certainly reflects the growing employment of a leadership style inspired by stewardship, previous practices (especially correctional ones) dominated by agency, have not been abandoned. In fact, several actors stress that trusting local actors with more autonomy requires harsher sanctions when trust is repeatedly broken.

”Some think that the trust reform means that there are no sanctions. But you lose trust if you don’t’ sanction leaders who do not respect collective goals”.

(CEO)

Hence, maintaining general trust in public employees requires constant attention, in particular to leaders who are not willing or able to handle more extensive autonomy. Politicians, CEOs and leaders at the local level all stress that trust that is repeatedly broken will result in sanctions and ultimately firing.

“People have often told me that the word ‘sanctions’ does note rhyme with ‘trust’. However, I actually think it does. Because if you do not sanction employees who cannot honor the fundamental values of the common good, you undermine trust”.

(Politician)

Although all actors stress the importance of a fair and transparent process when sanctions are applied to identified agents, a common opinion is that greater autonomy also comes with harsher sanctions. Hence, the new political leadership style of empowerment and acknowledgement inspired by stewardship is seen as fundamentally dependent on the parallel enforcement of agency practices of sanctioning.

4.4. Supplementing Institutional Power with Personal Power

Although political leaders have historically predominantly relied on coercive and formal power in their relationship with public employees, the trust reforms, in line with stewardship ideals, has produced a growing emphasis on more personal and rhetorical forms of power.

“The signals you send, the rhetoric you use, the discourse you use are incredibly important to how they act at all levels. Is it going to be about trust or is it going to be about control and rules to protect yourself?”.

(Committee Leader in Social Care). Especially local leaders notice the new style of attention from political leaders, which is perceived as acknowledgement

Political leaders now acknowledge that political signals, if interpreted as lack of trust, may contribute to an escalation of local rules and control as a means of protection.

“If employees are to trust you, you also need to show that you trust them. In order to do that you need to consider the personal signals you send in order to define a new direction”

(Mayor of Copenhagen)

Respondents at all levels note a new development in which political leaders become more visible for the local employees, by figuring in internal communications to leaders or employees, and by giving speeches at internal conferences about their goals and ambitions in relation to trust. “It is very unusual to see the mayor appear at such arrangements—talking passionately about trust to us” (leader in health care). There are also examples of political leaders who succeed in introducing new metaphors as a way to boost their personal power. For example, one political leader introduces the metaphor of “turning the pyramid up-side down”, as a symbol of the desired change in focus. Thus, rather than making public employees focus on serving the top of the hierarchy with detailed performance information, attention should be paid to supporting frontline workers in their meetings with citizens and users. This metaphor has spread, and is being used extensively by local employees and leaders as a political beacon of development.

Although the use of informal, personal and rhetorical forms of power is certainly intensifying, this does not mean that power forms associated with agency have been abandoned. Several political leaders stressed that paradoxically, they have had to draw on formal power bases to force the administration to align with the political ambition to reduce control:

“If I didn’t have the formal power as politically elected leader, I could not have succeeded in creating this movement towards trust. Especially at the beginning. I have had to control things in detail in order to ensure that the administration was onboard”.

(Mayor)

Hence, political leaders increasingly engage in communication with public employees, staging themselves as ambassadors of trust and drawing on more personal forms of power. However, the use of personal power cannot replace existing forms of institutional power among political leaders. Rather, new, stewardship-inspired power forms are used in combination with classical agency power bases.

4.5. Creating Arenas with Lower Power Distance

The lowering of power distance marks perhaps the most radical change in the political leadership of public employees. From using almost exclusively external, hands-off monitoring in their management approach, political leaders now prioritize more hands-on dialogue with the local level of the organization.

“If you want the employees to trust you, you need to get out and talk with them”

(Committee leader in Social Care)

Increased interaction is pursued in various ways, but all involve more direct, face-to-face dialogue and a tendency to prioritize informal meetings. This includes coffee meetings, visits and dialogue meetings with leaders, union members and frontline workers in their local working contexts. One political leader spends a week’s “internship” shadowing a health care assistant, to gain insight into how control systems affect her daily work. This leader explained that the hands-on experience had inspired a dramatic change to what was previously a very detail-oriented control system, illustrating how personal experiences at low power distance can inform and inspire political principals in new ways. Local actors also noticed the lower power distance:

“It is tremendously important! I was here when politicians didn’t want anything to do with employees. Didn’t even want to talk to them… I’m surprised how much they now engage in dialogue with us about the way we experience control”

(local leader)

Political leaders also participate directly in collaborative processes in which they work closely together with the administration, local leaders, union members and frontline workers to develop new solutions to existing control or documentation problems. Such long-term collaborative processes with local actors allow politicians to experience first-hand interaction with employees and to gain knowledge about their concerns, hopes and suggestions. The increased use of lower power distance among political leaders is also spurred by an ambition to create ownership over decisions.

“You have to deepen collaboration with professionals and bring them in earlier in the process, so they are not stuck in the role of critics. They need to be engaged in defining the problems and developing the solutions”.

(Politician)

Although the political leaders are still the formal decision makers, they choose to involve local actors extensively in decisions about the design of control systems, in order to build commitment and avoid decoupling (Lipsky 2010).

Although the active reduction of power distance by political leaders is certainly experienced as a radical difference, it is, as with the previously mentioned aspects of stewardship, used in combination with high power distance in classical arenas such as city council meetings. Hence, political leaders select or create certain arenas in which they reduce power distance and aim to share their power with local actors, while still maintaining high power distance in other arenas.

Table 3. Empirical patterns showing how politicians balance and adapt stewardship and agency practices to solve problems of spiraling control.

	Empirical Patterns of Stewardship and Agency
Political principals’ perception of employee motivation and goals	Employees are generally seen as trustworthy, pro-socially motivated stewards, working loyally to promote the politicians’ goals. Problems and mistakes are handled locally with support and help, while the rest of the organization is not affected by the introduction of general rules or control. At the same time, politicians in collaboration with the administration, single out and handle “bad eggs” (self-serving agents), who act counter to organizational goals.
Management style of political principal	Politicians actively strive to reduce exessive control and delegate as much autonomy as possible to local stewards. Politicians share an official ambition to reduce excessive regulation that erodes local autonomy. Control is differentiated according to local needs and risks, and is increasingly designed in collaboration with local actors. At the same time, external monitoring and control are viewed as legitimate and necessary to secure accountability and to keep self-serving agents in check. i.e., leaders must still document, benchmark and live up to several professional, legal and economical standards.
Leadership style of political principal	Politicians engage more actively in empowering local stewards through training, facilitated learning and professional reflection, enabling them to navigate with extensive autonomy. i.e., politicians promote courses and platforms for organizational learning. Good local leaders are praised and acknowledged by political leaders, rather than economically rewarded.At the same time, correction and sanctioning of local agents who do not respect common goals or values is seen as a vital precondition for preserving general trust.
Power base of political principal	Politicians increasingly use personal forms of power in their communication and interaction with local stewards. Politicians act as ambassadors of the trust reform and aim their personal communication directly at local actors. Politicians increasingly use metaphors and discursive means to inspire local actors to pursue political goals. At the same time, politicians draw on their institutional power bases to defend desired directions or values against disloyal agents. i.e., institutional power is used to force the reluctant administration to comply with the political ambition to de-crease control.
Power distance of political principal	Politicians increasingly seek face-to-face dialogue with stewards and invite them to share knowledge and ideas. i.e., politicians attend local meetings, invite local actors to coffee meetings or engage in “internships” at the frontline of the organization. In some arenas, political principals actively share their formal power with local actors e.g., by inviting them to co-create control solutions in direct face-to-face collaboration. At the same time, fundamental decisions about reforms designed to stop escalating control are taken top down by political principals, with high power distance from agents.

, below, summarizes the empirical patterns according to which politicians balance agency and stewardship practices in their approach towards employees.

4.6. Stewardship: Opportunities and Challenges for Political Leaders?

The growing influence of stewardship approaches in politicians’ approach towards public employees fosters both new opportunities and challenges when it comes to solving problems of escalating control.

The use of lower power distance and more direct collaboration with public employees offers several benefits for political leaders. In several cases, political leaders explain that they have gained new knowledge and insights through hands-on interaction that has profoundly changed their perception of control problems. “I never would have believed that they spent that much time on useless documentation if I have had not been out there and talked with them” (politician). Numerous accounts from both politicians, administrators and leaders in institutions, consistently connect new stewardship practices with better problem diagnosis as a foundation for political agenda setting.

The trend of engaging local leaders and employees in collaborative processes also provides new opportunities in terms of finding robust solutions. For example, a group of employees, shop stewards and local leaders was invited to be part of a working group tasked with developing a new and more meaningful system of control in homecare over a period of two years. Although this required a massive investment of time in developing the system, politicians, administrators and local actors all experienced that it led to a greatly improved solution: the new control system is described as much better in terms of supporting professional task quality. Such and other empirical examples indicate that lowering power distance and sharing power with local actors assists politicians in developing more robust control solutions in their political leadership.

The most remarkable benefit, however, is that stewardship appears to be an important driver in mobilizing support and engagement among public employees. A political leader explains: “Involving them in questions of control is the right thing to do. But it has also created much greater commitment to decisions”. This experience is backed up by several stories from local leaders, who emphasize that implementing decisions about control based on employee involvement is much smoother. A leader in a social institution says: “I really use this to inspire my employees: the politicians actually back us up! Let’s show them what we can do in return!” Numerous empirical examples support the finding that politicians benefit from cultivating followership and support among public employees, who might otherwise decouple, derail or modify political decisions about control which they do not find meaningful (Britt 1991; Lipsky 2010).

Drawing on Tucker’s understanding of political leadership, introducing aspects of stewardship into the relationship between politicians and local employees leads to more accurate diagnosis of problems of control, the development of more robust solutions and, not least, remarkable mobilization of support among public employees (Tucker 1995).

One concern voiced both among politicians and administrative leaders is that the political climate can create situations in which politicians feel pressured to return completely to modes of agency and abandon stewardship approaches. This can occur, for instance, when the opposition uses critical cases to increase pressure on political leaders, who must balance their loyalty towards voters with their loyalty towards public employees.

“Although the politicians in my committee are on board with this trust reform, there are limits to their loyalty. There will always be somebody who can see the personal benefit of a quick win if a scandal occurs”.

(Politician)

Several actors also voice the concern that competition among politicians is bound to increase when elections are imminent. This inherent dynamic, as well as changes in the group of incumbent politicians in the wake of elections, will potentially affect continuity in approaches designed to de-escalate control of public employees. Hence, discontinuity in the political arena will always constitute a challenge, which can lead to a political re-prioritizing of ambitions to downscale control. Although replacement is a natural part of the democratic arena, the “Code of Trust” appears to function as an important anchor of stewardship in the turbulent competition surrounding selections.

Although administrative leaders have traditionally acted as gatekeepers between politicians and public employees, arenas of lower power distance create a much more blurry zone of interaction between politicians, administrative leaders and public employees (Alford et al. 2016). Although this enables new opportunities in terms of strengthening political leadership, direct interaction also opens “loop-holes” for circumventing the hierarchical line of command in the administration. A leader in Rehabilitation explained:

“One of my physiotherapists was in a dialogue group with the mayor. And she tells the mayor that quality is worsening because the trust reform has led to downsizing in our institution. Then I am contacted by my boss, who starts questioning me. But the reality is that our budgets have been cut, and although this is not directly related to the trust reform, it is my job to make these tough decisions”.

This and other examples illustrate that decision processes become less linear when politicians interact directly with public employees. Although some employees embrace these possibilities actively and seek dialogue with politicians, others find it difficult to navigate in these new, complex processes and fear that they are undermining the classical virtue of transparency and accountability in political decision-making. Although this challenge cannot completely be eliminated when politicians engage more directly in dialogue with public employees, awareness and critical reflection about such dynamics among politicians is considered vital to avoid unruly decision processes.

A commonly mentioned challenge is resistance or veto point in the administration, which is challenged in terms of power and competence under the new reform. “I have met massive resistance in the Administration”, says a political leader who, in the early days of the reform, ended up replacing a CEO due to his/her unwillingness to collaborate as recommended by the trust reform. When politicians scale back on political goals and control systems, this also reflects on the administration, whose staff must redesign and sometimes reduce regulation at the local level: “An administrator like me prefers to have control. We live off numbers and paragraphs. Letting that go is not easy” (Administrator). In addition, the new ambition to ensure involvement requires that administrators can facilitate dialogue between a multitude of actors, rather than just preparing political cases to present to committees. “They need to unlearn the old ways and acquire tools to manage the new trust-based approach”. (Committee leader). Hence, a shift towards stewardship in the political approach towards public employees requires the simultaneous development of administrative competences and willingness to back up political ambitions to reduce and co-create control. This development is spurred with educational activities among administrators, a Code of Trust aimed specifically at the administration and in some cases recruitment of new profiles more in line with stewardship ideals.

Opportunities and challenges for political leader are summarized in

Table 4. Opportunities and challenges for political leaders attempting to de-escalate control by balancing agency practices with new stewardship approaches.

Opportunies for Political Leaders

Challenges for Political Leaders

Better understanding and diagnosis of control problems through involvement of employees’ knowledge and experiences. Development of more robust control solutions that enjoy greater ownership among employees. Better implementation of political decisions due to enhanced mobilization of resources among public employees.

Scandals and elections enhance political competition, potentially challenging loyalty to decisions to de-escalate control. Lower power distance creates diffused processes, which in some cases challenge transparency and accountability in decisions about control. Administrative resistance, as well as lack of competence to facilitate co-creation of control, block political endeavors to stop spiraling control.

below.

5. Discussion

It is of course not possible to generalize the empirical results from a study of a purposefully selected singular case that ‘grew’ in highly fertile soil. However, statistical generalization of results was never the purpose of this article. Instead, it sought to explore and document the feasibility of applying principles from stewardship theory to political ambitions to break vicious circles of escalating control.

So, what can we learn from a unique case of a Danish municipality? Of course, the institutional and cultural context must be taken into consideration when reflecting on the broader relevance of the theorizing: Denmark is a high-trust country with a long cooperative tradition which has expectedly paved the way for a stewardship approach towards the public employees. Danish municipalities are also characterized by an extensive task portfolio, comprehensive autonomy as well as high levels of political decentralization and formal political power regarding the administration, which, even compared to other Nordic countries, provide Danish politicians ample room for pursuing ideals of stewardship, (Ladner et al. 2016). However, most dimensions of stewardship such as dialogue, face-to-face interaction and involvement entail soft forms of power which politicians across municipal contexts may draw on in some form. Although practices and conditions will unquestionably vary considerably across countries and sectors, the prospects of stewardship appear especially relevant in public sectors, anchored in pronounced agency practices (i.e., the Anglo-Saxon countries) after decades of NPM (Pedersen and Löfgren 2012).

The prospect of fertilizing existing practices of agency with new ideas based on stewardship shows promising potential in the case presented here. However, such active intervention by politicians in the administrative arena is certainly not without dilemmas. An important question is whether direct interaction between politicians and employees could pose a threat to democratic legitimacy (Koppenjan et al. 2011; Mayntz 2003; Sørensen and Torfing 2016). Could strengthening politicians’ relationship with public employee risk weakening their primary relationship with citizens? A similar critical question is whether the less linear and more diffuse processes entailed in a stewardship approach could blur transparency in decision-making processes and ultimately erode accountability (Contrafatto 2014; Papadopoulos 2007)? Although the results of this study point to many possible benefits for political leaders, such as empowering, collaborating and building trust with public employees, issues of accountability require careful consideration about the necessary conditions for sharing formal power. Although bringing public employees into the scope of political leadership may not simplify politicians’ tasks, this enhanced complexity is arguably unavoidable if political solutions are to be robust. One of the obvious reasons for this is the tremendous resource that public employees constitute in the public sector. Not only is their knowledge and knowhow extremely valuable in understanding the challenges and possibilities for de-escalating excessive control but, more importantly, their followership is decisive for how politically decided control systems are implemented.

The empirical insights generated by this study support existing research which suggests that stewardship and agency must be approached as parallel and interplaying strategies. The results of this study certainly endorse the idea that stewardship can modify, balance and supplement practices of agency in a relatively constructive interplay. However, the empirical material also offers examples of employees perceiving political signals and practices as mixed, fostering confusion about politicians’ motives. For example, some employees interpret all control as detrimental to trust, which in a few cases has left employees suspecting that they are involved in a political pseudo-process. Hence, offers of involvement may counter ambitions to build trust if employees perceive influence as superficial or even staged (Bentzen 2020). In such cases, offers of trust may be regarded as a poisoned chalice, and are unlikely to strengthen political leadership in the sense of engagement, ownership and resource mobilization among public employees (Bentzen 2019; Skinner et al. 2014; Steen et al. 2018). Hence, many relevant questions about the compatibility of agency and stewardship require further research, as do the conditions for politicians creating a constructive interplay between the two.

The case of Copenhagen municipality provides a source of inspiration for other public service organizations who may want to adopt some of the key principles and practices associated with stewardship theory. For practitioners, the results of this study can be used to reflect upon existing relations between political leaders and public employees, as well as on the prospects of de-escalating burdensome and demotivating control, by adopting stewardship ideals. Although local translation to specific contexts will always be required, the study may inspire politicians to explore new pathways for avoiding dynamics of escalating control in the public sector.

6. Conclusions

Mushrooming control, spurred by the combination of classical rule and neoliberal performance monitoring, constitutes a growing problem in public sector organizations, struggling with excessive transactional costs and the erosion of motivation among public employees. Although political leadership is generally considered separate from such administrative problems, the classical dynamic of politicians responding to local failures by tightening general rules exacerbates the problem of spiraling control. Drawing on a qualitative case study of a Danish municipality, this study shows how politicians can use stewardship ideas in various ways to balance existing practices of agency in their approach to public employees: Personal forms of power are used to supplement institutional power; high power distance is reduced in selected arenas; external monitoring and control is reduced, reshaped and increasingly co-created; and politicians engage, support and encourage local employees rather than sanction them. The study shows several prospects for politicians who are adapting stewardship ideals into existing practices of agency: closer and more interactive relationships with local employees enable them to better diagnose problems regarding control, develop more robust control solutions, and secure support for those solutions among public employees. However, engaging in stewardship relations with public employees also raises challenges for politicians, who face the dilemma of balancing their loyalty between citizens and public employees when elections or critical cases spur competition among politicians. The lower power distance associated with stewardship also creates challenges in terms of diffuse decision-making processes, which potentially threaten accountability. Finally, the study points to administrative resistance and insufficient competence as barriers to political ambitions to break vicious circles of spiraling control.

This study contributes to existing research by showing how stewardship theory can provide new, complementary avenues for politicians trying to break vicious circles of escalating control in the public sector. Research addressing the somewhat neglected relationship between politicians and local-level actors in public organizations is extremely scarce, adding to the value of these findings. The results also offer empirical insights into the parallel applications of agency and stewardship practices which, despite apparently competing ideals, appear to interplay constructively in this case. However, further research is needed to assess the potential, as well as the dark sides, of employing stewardship in political leadership practices across more diverse national and sectorial contexts.