Bureaucrats Under Illiberal Rule. How Governments Subdue Bureaucrats and How They Fight Back
By Dániel G. Szabó
Universal health care, the moon landing, restoring the ozone layer ‒ civil servants are capable of incredible things: universal health care in Europe is organised mainly by state-run hospitals; the NASA putting a man on the moon was a government agency; the frontline of mitigating the effects of climate change is very much within governments where the civil service drafts, implements, and supervises laws to protect people from floods and draughts. But total oppression and mass killings in the twentieth century would have also been impossible without bureaucracies. The civil service is capable of creating hell on Earth: dictators are empowered by bureaucrats, and mass deportations and killings were organised by government officials in the twentieth century. With his concept of the “iron cage”, Max Weber refers not only to capitalism but also to bureaucracies: highly efficient forms of organisation that shape and restrict many aspects of our lives, leaving us with little room to break free.
Modern government bureaucracies are influential, powerful, and enormous institutions. Hundreds of millions work for governments everywhere to regulate, control, monitor, and provide. No wonder that many TV shows portray the civil service in the broad sense: crime series follow government-employed detectives, medical dramas take place in state-run hospitals, and the award-winning series Chernobyl immortalised an administrative effort so big that it might have contributed to bringing down the Soviet Union. Whatever one calls it, the civil service, the public administration, or the bureaucracy, is the executive branch of the executive branch. All policies, the most humane as well as the ugliest, are implemented by it. It affects every day of our lives, including what we eat, how we commute, what our working conditions are, how we are taken care of when sick, and how we die. Who runs the World? Mid-level bureaucrats.
1. The Underreported Bureaucracy
The civil service has enormous power, yet surprisingly little is being said about its responsibility: its ability to empower or to tame the good as well as the evil in us. Public administrations’ roles in a democratic system are relatively understudied. Democracy and the rule of law are on the back foot in several aspects and in many parts of the World. Much has been said and written about the role of populist politicians, parliaments, courts, and civil society – but much less about the role of bureaucrats. The civil service is consequential because government bureaucracies are not mere executors of government policy. The relationship is complex, with bureaucrats pulling or pushing back on government policies. Sometimes bureaucracies tame government policy – such as many policies of the first Trump presidency that were held back or even sabotaged by “adults in the room”. At other times, the civil service empowers governments for better or worse. The civil service can help protect democratic values and the rule of law, or accelerate democratic decline.

Governments are elected, civil servants not. In a democratic form of government, elected politicians, not bureaucrats, should have the final say. In this blog post, I do not wish to exalt all bureaucrats who oppose or sabotage government policy: that can very well be extremely dangerous and undemocratic. But the relationship between the government and the bureaucracy is worth a closer look. This blog post analyses the literature on the relationship between bureaucracies and governments from two perspectives: what governments aim to achieve through bureaucracies and how they do so (Taxonomy of Government Approaches and Strategies), and how bureaucracies respond (Taxonomy of Bureaucratic Reactions). This blog post investigates how the works of different leading authors give different names to similar phenomena, and how they complement each other. It concludes with final thoughts, including ideas for further research.
Where We Are
In the multi-layered power relation of governments and bureaucracies, I understand the government in the narrow sense: elected politicians and their closest colleagues, the political branch. By bureaucrats, I mean the non-political, permanent employees of the executive branch who have wide-ranging expertise from law to economics, from engineering to medicine, and from nuclear physics to sewage management.

First, I focus on governments: What do they want to achieve (approaches), and how do they want to achieve it (strategies)? Second, I examine the reactions of bureaucracies. My analysis below is an extended version of the threefold analytical framework suggested by Michael W. Bauer, B. Guy Peters, Jon Pierre, Kutsal Yesilkagit, and Stefan Becker (shown on the right).
It is important to note that most of the papers analysed in this blog post examine the relationship between governments and bureaucracies in the context of democratic decline and analyse what populist and autocratic governments aim to achieve through bureaucracies and how they do so. Regardless of this focus of the academic literature, much of the following might be true in any organisational power relationship and thus has broader implications. For example, whatever change a government wants to achieve in a bureaucracy – dismantle it, or make it more efficient – it will do well to target its resources, personnel, or norms.
2.a Taxonomy of Government Approaches (Table 1) and Strategies (Table 2)
Table 1 below is a summary of terminology for what a government might want to achieve regarding the bureaucracy brought forward by some leading authors (see below). One way is sidelining, referring to the option in which a government distrusts the bureaucracy and, through patronage, replaces incumbent officials or degrades them into business-like technocrats. Another avenue is to substitute the career public service with a parallel structure. I consider sabotage a similar option insofar that it also reflects a strong distrust in the bureaucracy. Sabotage happens when a populist government aims to limit the established bureaucracy’s capacity to counteract this new government. Dismantling is very similar to the above strategies; in Bauer-Becker’s view, it reflects a situation in which a populist government sets aside bureaucratic barriers to its own power.
Ignoring the public service might stem from several reasons: either populist governments might not be interested in the technical aspects of governing at all, or they might believe they can govern on their own, without the “ineffective bureaucrats”. Paradoxically, this approach is likely to empower the public service, because the government has to go on somehow, and without effective political leadership a bureaucratic government might emerge. B. Guy Peters and Jon Pierre refers to this latter scenario under empowering the bureaucracy, explaining that bureaucracy-bashing populists, once in power, realise that if they want to do something, they need the public service. Hence, populists conclude that effective and empowered bureaucracies contribute to achieving the populist policy goals. Seemingly different, but in effect similar, is capture: populists capture the bureaucracy, shape it to their image and make it efficient in carrying out their policies.
An interesting scenario is what is called using the bureaucracy or strange bedfellows, when a populist government’s desire to govern, but lacking the necessary expertise, may overcome its distaste for the career public service. In this scenario neither the government, nor the bureaucracy really trusts the other, but they cooperate. Bureaucrats face a dilemma of maintaining control, while executing policies against their beliefs. A softer form of a similar approach is reform, when a government faces an established and strong bureaucracy and the most it can achieve is reforming the bureaucracy to tilt its operations toward the desired outcome.
| Bauer-Peters-Pierre-Yesilkagit-Becker | Peters-Pierre | Bauer-Becker |
| Sidelining | Sidelining | Sabotage |
| Dismantling | ||
| Ignoring | ||
| Empowering | Capture | |
| Using | Strange Bedfellows | Reform |
Table 2 below summarises government strategies. These are the terms used in various scholarly articles to describe the methods by which governments might achieve the goals stated in Table 1. Changing the structure of bureaucracy is often in the form of centralisation to reduce autonomy. Authoritarian leaders cannot change the entire public service; therefore, as a second-best option, they try to control the existing system as much as possible, or to tame a hostile “system” of the establishment they are fighting against as populists.
Another option is to target resources: changing the budget or personnel numbers obviously changes the power dynamics and might tame opposition. Changing the personnel is another way to exert control over the bureaucracy, which might come in the form of purges of staff, and patronage, or hiring loyal supporters. Both tactics often result in the loss of expertise within the public service and the politicisation of governing, where political loyalty becomes more important than expertise.
Norms of the bureaucracy might often be changed to suppress dissent and create an enforced loyalty. Specific sub-categories can be identified by meddling with the norms of public service. Changing the laws is an easy choice in the hands of incoming governments, both regarding the decision-making processes or the day-to-day business of civil servants. Even subtle changes make a significant difference: suppressing formalities and guaranteeing power and participation for bureaucrats might seem minor, but can have substantial consequences. Breaking the laws means the violation of existing norms, for example, in the form of guidelines or internal norms, which run contrary to laws or constitutional guarantees. As the running joke among Hungarian civil servants tells: “What is the highest source of law? The department head’s decision.” Bending the laws is an alternative that provides a less costly but more unstable solution than the previous two options. When laws are ambiguous, a different interpretation can play in the hands of a new government. An eminent example of threatening to use the law is in the domain of people management, for example, to threaten bureaucrats with punitive sanctions. An example is using unwritten, unofficial threats to dismiss or launch disciplinary action.
Lastly, reducing accountability of government action means cutting back societal participation, parliamentary and other external controls, reducing transparency and contacts with partners and the media.
| Bauer-Peters-Pierre-Yesilkagit-Becker | Peters-Pierre | Lotta et al. |
| Structure | Centralisation | |
| Resources | ||
| Personnel | Patronage, Politicisation of governing | |
| Norms | Changing the laws, Breaking the laws, Bending the laws, Threatening to use the law | |
| Accountability |
2.b Taxonomy of Bureaucratic Reactions (Table 3)
Bureaucrats, on the other hand, react to the governments’ actions. Table 3 summarises the classification of their strategies in the literature.
Working or loyalty means carrying out one’s tasks either because of a sincere belief, neutrality, or out of fear. In the realm of norms, this might take the form of changing rules: bureaucrats, if they see an opportunity, might ally with parliamentarians to change laws, or with superiors to change internal norms. This is sincere work in pursuit of what they understand as the public good.
Shirking might mean finding the easiest way to survive in the public service, or it might be politically motivated, changing policy goals to better fit certain values. Bending the rules is a way of exercising this approach. For example, alternate ways to disseminate information might help to involve non-governmental organisations even if a new government does not want it to happen.
Sabotage, also known as guerilla government, refers to the deliberate frustration of the implementation of certain government policies. In relation to norms, this may occur when bureaucrats break the law to participate in public events aimed at reversing adverse changes, despite a prohibition on such participation. Bureaucrats might justify breaking one norm to protect another. Another activity resulting in sabotage is refusal to do illegal activities requested by superiors: a refusal to sign documents, authorise procedures, or participate in meetings or activities.
Voice is an option where bureaucrats try to convince superiors or raise concerns. This can come in the form of threatening to use the law, for example, by explaining the possible personal legal consequences for the leadership. Activating the justice system is another, formalised form of voice and means turning to courts to decide conflicts between the civil service and the political leadership.
Finally, Exit is an obvious choice: leaving the civil service or requesting transfer to another unit if changes are unbearable.
| Bauer-Peters-Pierre- Yesilkagit-Becker | Hirschman | Brehm-Gates | Schuster- Mikkelsen- Correa- Meyer- Sahling | Lotta et al. |
| Working | Loyalty | Working | Changing | |
| Shirking | Shirking | Bending | ||
| Sabotage | Sabotage | Sabotage and Guerilla government | Breaking the laws, Refusal to do illegal activities | |
| Voice | Voice | Threatening to use the law, Activating the justice system | ||
| Exit | Exit |
Richness Through Interviews and Case-Studies
Theoretical research is fundamental for providing a vocabulary and a framework for discussing the issues at hand. The book Democratic Backsliding and Public Administration is particularly enlightening in summarising the key aspects and examples of the public service’s role during democratic decline. Having gone through some of the leading literature on bureaucrats under illiberal rule, I found two aspects worth more reflection.
First, any control of an organisation – both for malevolent and benevolent aims – is done by similar means. For example, changes in structure, resources, personnel, norms, and accountability can not only have devastating effects, but also be the areas to target to achieve a more democratic and rule-of-law-friendly public service. Therefore, what these scholarly works describe are not specific tools of autocratic governments, but general control and management mechanisms for organised labour.
Second, another avenue for research might be based on the strong interrelation among the three aspects above: government approaches, government strategies, and bureaucratic reactions. Therefore, they might be analysed within a more synthesised framework. Exit, sabotage, voice, and loyalty are currently used to understand bureaucratic reactions, but the same could be applied to analyse government strategies as well by looking at what the government wishes to achieve (to encourage bureaucrats to exit the civil service; the government might want to sabotage the bureaucracy; might voice concerns publicly over the bureaucracy; or might try to induce loyalty in bureaucrats). Similarly, structure, resources, personnel, norms, and accountability are used to analyse government strategies. Still, the same frameworks might well be used to analyse bureaucratic reactions as well (how the civil service reinforces its structure, resources and personnel to oppose government pressure; how it sticks to norms to stand its ground; how bureaucrats reach out to media and civil society to entrench government accountability).
As a methodological suggestion, surveys and interview-based research are indispensable and often fun to read. These very work-intensive studies were enlightening because of the breadth of real-life experience they contained. Often, they included the experience of dozens or hundreds of respondents working in the civil service under a populist government. No theory can catch up with quotes like this:
Deputy state secretaries are really like idiots. There is no work plan, and everybody deals with whatever happens to be the most critical issue to solve at that given second. They are all scared to death, that’s why. They organize an interministerial coordination meeting. But there is no official invitation, no agenda, so there are twelve people talking all kinds of bullshit. After 45 minutes, two of them stand up and say, “Sorry, we think we came for another meeting.” and they leave. This is an average working day at the ministry.
Case studies offer another way to learn about governmental strategies and bureaucratic reactions. A particularly enlightening book for me was Phantoms of a Beleaguered Republic: The Deep State and the Unitary Executive, on how President Trump tried to influence and subdue the US administration under the theory of “the unitary executive” which claims that Article II of the US Constitution vests the entirety of executive powers with the President. The book is extremely rich in examples, both on what has worked and what has not, to protect the rule of law and depth (that is, deep thinking, foresight, and sincere attempts by career bureaucrats to find real solutions) in civil service work.

Under autocratic governments, it might be hard to get to talk to bureaucrats: access to them might be limited by law, contact information scrubbed from websites, and their willingness to speak to researchers or answer questionnaires might be hampered by repressive measures or suspicion. An interesting avenue for research might be to use court files. A high-level political corruption case – to the surprise of many – was prosecuted recently in Hungary, and interrogation reports leaked to the press. The detailed questioning of public servants offers a glimpse into the attitudes towards power and norms of some civil servants after a decade of illiberal rule, such as the following excerpts:
I would like to add that I completed a training on professional integrity around 2018, and a conclusion emerged in me that this deeply problematic and unlawful practice [which I was talking about earlier] should be reported, but, on the one hand, when I realised that I could not report anonymously because my IP address could be traced back, and, on the other hand, the responsible person was [implied in the case], no one did anything.
It was very strange to me, because I did not receive an official document, instead [a colleague] showed me her phone with a picture of a sheet of paper with the list of cities and persons in rows. She said to me that I have to pass on these names [with positive recommendations]…[The colleague] said she wants no one to know that the list came from her, so she asked me to take a photo of her phone[’s screen] with my phone (on hand with the author).
Analysis of this and other court documents might be a new avenue to get access to civil servants when other, more common avenues, such as interviews, are limited or impossible to pursue.
In this blog post, I summarised and analysed some of the leading scholarship on bureaucrats under illiberal rule and compared leading authors’ taxonomies on government approaches and strategies towards the bureaucracy, and bureaucratic reactions to these actions. It is discussed how governments might try to influence and subdue the public service and how it can fight back or resist. Despite the immense importance of the topic – governments in our societies cannot do anything without the intermediation of an army of bureaucrats – it is relatively understudied. Among the closing thoughts, I suggested two ways to synthesise this taxonomy, argued for the value of surveys and interview-based research, and suggested using court documents if other methods of accessing civil servants are not feasible.
About the Author
Dániel G. Szabó is Head of Department at the 8th District of Budapest, responsible for urban development. Previously, he worked at the Municipality of Budapest, coordinating the City Council’s decision-making, IT, and participatory governance with a staff of 140. Before that, he worked on the rule of law and the independence of the judiciary at the Hungarian Helsinki Committee and at Democracy Reporting International. He has a background in election law, having served on the Hungarian National Election Commission. He was a re:constitution fellow in 2022/23 and holds an LL.M. in Comparative Constitutional Law from the Central European University.
Citation: Dániel G. Szabó, Bureaucrats Under Illiberal Rule. How Governments Subdue Bureaucrats and How They Fight Back, in: TRAFO – Blog for Transregional Research, 08.01.2026, https://trafo.hypotheses.org/63684
OpenEdition suggests that you cite this post as follows:
Forum Transregionale Studien (January 8, 2026). Bureaucrats Under Illiberal Rule. How Governments Subdue Bureaucrats and How They Fight Back. TRAFO – Blog for Transregional Research. Retrieved January 22, 2026 from https://doi.org/10.58079/15gvg


