Why bureaucracy matters, part 5: The indispensable knowledge machine
In a noisy information world, bureaucracy ensures decisions are evidence-based, transparent, and grounded in reality.
First published in the Mandarin.
What if we stopped treating bureaucracies as dysfunctional engines of red tape and started seeing them as institutions that make knowledge? What if they provide us with the sort of situational knowledge that helps us tell truth from noise?
When exaggerated claims about “mass migration” circulated, Australia’s chief statistician, David Gruen, stepped forward to clarify what the ABS figures do (and do not) show. Gruen’s message was clear: for informed debate and sensible policy, someone must uphold the conditions where the truth can be recognised at all. In a time when we are all saturated with misinformation, providing a clear statement of the facts from a trusted source is a public service.
What follows is an argument for seeing bureaucracy as a social technology for sensemaking. When it works, bureaucratic craft transforms data and information into a shared understanding. They handle complex details reliably and serve as a mediator between general rules and individual experiences.
When bureaucracy becomes clumsy or brittle, we see public service craft for what it is: a shock absorber of the tension between rules and reality. It is the ongoing negotiation between formal procedures and an unpredictable world.
This tension — between rule enforcement and our experience of frustration — is not a flaw in the system; it is the system.
Three ways of knowing, one craft
Max Weber’s portrait of ideal bureaucracy helps identify three types of knowledge that collectively make up the craft of public service.
Expertise. Public organisations are home to a remarkable concentration of specialists — engineers and epidemiologists, veterinarians and meteorologists, lawyers and linguists, economists and ecologists. Expertise is the ability to explain and predict: to know why.
Experience. Equally important is the practical know-how gained through practice. Finding workable solutions to problems cannot be solved by theory alone. Problems require judgment that practitioners develop through repeated exposure to uncertainty, failure, and correction.
Rules. Finally, there is knowledge of law, procedure, and protocol: how to reason from administrative principles, document and justify actions, and ensure consistent treatment of similar cases. How to know what is relevant and how to apply the rules fairly and in context.
A mature bureaucracy combines these three approaches. It values specialists, cultivates practical wisdom, and remembers that rules are there to support fair action. The skill lies in blending these three types of knowledge.
The engine of bureaucracy is tension, not harmony
Bureaucracy is often judged by a standard it cannot meet: seamless harmony between abstract rules and real-life situations. Novelist Terry Pratchett offers an insight into this in practice.
Authority rarely descended to street level. Vimes found it better to look at Authority for orders and then filter those orders through a fine mesh of common sense, adding a generous scoop of creative misunderstanding and maybe even incipient deafness if circumstances demanded.
Pratchett’s ‘fine mesh of common sense’ shows that sometimes frontline discretion is a principled, explainable adaptation. Look for authority, care for people, and filter the situation through a mesh fine enough to support a good decision.
Too little discretion makes the system rigid; too much causes it to lose coherence and trust. The challenge is to find the right balance while upholding the principles and applying the rules.
Seen from this angle, familiar bureaucratic dysfunctions appear differently. A system that tries to know everything about everything drifts into an information trap where monitoring becomes obsessive and reporting replaces action.
It suggests that everyone could benefit from a bureaucracy with lighter oversight and more nuanced judgment. It offers an alternative perspective on why investing in the development of APS capabilities is important—specifically, improving public servants’ ability to balance rules with real-world complexities.
We built this bureaucracy
If bureaucracy sometimes feels frustrating, it is because it reflects our own conflicting demands. As Herbert Kaufman noted in his book Red Tape: Its Origins, Uses and Abuses (1977), the frustrations and dysfunctions of bureaucracy are not the accidental leftovers of poor design; they are the residue of values we refuse to let go.
We want compassion: public services that recognise vulnerability and respond with care. We want fairness: rational, consistent, and reviewable decisions. And we want protection against abuse of power: multiple layers of audit and scrutiny that reduce the possibility of misconduct. Each value adds red tape.
The uncomfortable implication is that cutting red tape isn’t just an abstract debate about which rules we prefer or dislike. It requires a trade-off based on our core values. The real question isn’t ‘How do we have fewer rules?’ but rather ‘Which values matter most and in what situations?’.
Bureaucracy as sensemaking
Public organisations also have a capacity for sensemaking. They develop institutional habits that turn uncertainty into clear patterns and actions. They absorb messiness and transform uncertainty into action.
Sensemaking is a vital part of public service craft. It involves clearly explaining why a policy or decision is justified. It also means recognising when rules are inadequate or incomplete, and explaining why, along with available options. Moreover, it includes coordinating, cooperating, and collaborating not just to solve the problem but also to promote collective learning.
If bureaucracy is about the types of knowledge it generates, maintains and navigates, then sensemaking is the outcome.
A different conversation about productivity
Discussions about productivity often reduce public institutions to factories that should process messy human issues more quickly. In non-market sectors such as health, education, police, and public services, the main goal is to make sound decisions amid uncertainty. The central question of productivity moves from ‘How do we cut costs?’ to ‘How effectively do these organisations transform information into legitimate and timely actions?’.
Different types of productivity questions might be asked:
How effectively is data collected, stored, analysed, and disseminated? Are the implications of the analysis clearly explained?
Is decision-making open and traceable? Do public documents show their process? Can an informed outsider follow the path from facts to a decision?
When frontline staff face situations where the rules don’t quite fit, do they tend to make similar judgments for similar reasons? Are they aware of the boundaries of their discretion when applying the rules?
When things go wrong, do they surface quickly? Do small failures turn into major issues? How well does the institution learn?
Does the community trust the system to be predictable and explainable, even when outcomes are disappointing?
None of these can be reduced to a single dashboard. As consulting firm McKinsey, in their Five Big Tests for Australia’s Productivity Agenda:
Part of this is due to measurement. A longer, healthier life, or a child who can read fluently, is plainly valuable, yet these gains do not show up on profit-and-loss statements, making productivity appear flat even when outcomes improve.
Maybe we’re asking the wrong questions, making the wrong comparisons, and using the wrong measures.
Why this matters now
The information environment is becoming more difficult. Generative tools blur the line between what is created and what is real. Our attention is overwhelmed, and we are bombarded by manufactured outrage.
In such circumstances, our collective ability to tell fact from ideology and truth from conspiracy is quickly becoming a crucial societal skill. We cannot rely on the leading technology platforms to do this for us, nor can we leave it to politicians.
The 2025 Trust in the Public Service Annual Report concludes that ‘Overall trust in Australian public services has increased significantly in the past year…’. Maintaining this trend may require the public service to stand for the truth, as David Gruen observed.
You don’t want to be seen to be taking sides; I mean, you are taking sides in the sense that you’re trying to take the side of truth, and you’re trying to take the side of explaining what the statistics say and what they don’t say.
Bureaucracy, at its best, is where the discipline of fairness runs alongside the discipline of truth. It is a place where we are impartial towards the evidence and pragmatic when considering competing interests. This process can be slow because it aims to be accountable not just for outcomes but also for reasons—what was known, how it was considered, and why it justified action. The slowness is not always, as seen in productivity terms, a waste; it often reflects the cost of making good decisions on issues with far-reaching effects.
Sometimes, delays are pointless, reporting serves no real purpose, rules are confusing, and opacity conceals poor behaviour. The solution isn’t to cut costs or bypass red tape based on the loudest voices, but to follow G.K. Chesterton’s advice: ‘Don’t ever take a fence down until you know the reason why it was put up’.
Before attempting to improve public sector productivity or dismissing it as a ‘drag’ on the economy, first understand its purpose. And, continue investing in human skills that consistently bridge the gap between rules and reality.
