Administrative Burden: Is ‘Cutting Compliance Costs’ Enough?

23/01/2026

The Hidden Architecture of Bureaucratic Burden

This series examines a deceptively simple question: what does it actually cost to navigate the state? Not the cost to the government; specifically, the forms processed, the staff employed, and the systems maintained, but the cost to the person standing in the queue, filling out the application, trying to understand what has changed and their new obligations and entitlements might be. Drawing on a decade of international scholarship whilst remaining grounded in Aotearoa’s particular institutional arrangements, each post explores a different dimension of administrative burden: the learning costs of discovering what exists, the compliance costs of following the rules once you know them, and the psychological cost of the stress, the stigma, and the erosion of dignity, that accumulate in the space between a person and the service they need or the rules they need to follow.

Today, we open the series with the Ministry for Regulation’s flagship agenda and ask whether “cutting compliance costs” is the right frame for the problem. By the end, you will understand why collapsing these three categories into a single measure produces policies that appear efficient in Cabinet papers and Regulatory Impact Statements, whilst potentially failing the people they were designed to serve, and why that failure falls hardest on those with the fewest resources to bear it.

The Ministry for Regulation has made “cutting compliance costs” the centrepiece of its reform agenda, and its guidance on cost recovery draws on the Treasury’s 2017 framework. Both, understandably, focus on compliance costs and conventional cost-benefit analysis. These are measurable, defensible, and politically legible categories. I use them myself in my practice. And, I teach it at the post-graduate level.

But I find myself increasingly wondering whether this frame captures the full weight of what citizens and businesses actually experience when they encounter the state.

The compliance cost of filling out a form is real. So too is the time spent discovering the form exists in the first place, or the quiet dread that accompanies the fear of getting it wrong. These are different kinds of costs, and combining them into a single measure may obscure more than it reveals.

Which raises a question: Is there a better way to think about this?

It turns out there is. Over the last decade, scholars, including Pamela Herd and Donald Moynihan, have developed a more sophisticated architecture for understanding bureaucratic friction.

Their foundational 2015 article proposed that the “costs” of dealing with government are not monolithic. They come in at least three distinct forms, and recognising them as separate is essential if we are to grasp what actually happens when a person tries to navigate the state.

Learning costs

Learning costs are the effort required simply to discover that a programme or the rule exists, to determine whether you might be eligible, to understand how the application process works, or to understand the new behaviour the state now requires of you.

Before you can comply with anything, you must first find out what to comply with. A benefit that is technically available but practically invisible imposes such high learning costs that many eligible citizens never apply. For a regulated business, this is the time spent keeping pace with constant changes to policies, regulations, and interpretive guidance, or figuring out which of sixty-six different Building Consent Authorities’ readings of the Building Code applies to your project.

Compliance costs

Compliance costs are what the current frameworks already capture: the time, money, and effort spent following the rules once you know what they are. The form-filling, the document-gathering, the hours on hold, the fees for certified copies, the cost of professional advisors, and the productive hours lost to doing what the state requires.

But I would add that most cost-benefit analyses that I am asked to peer-review in Aotearoa and Australia suffer from weak counterfactual scenarios that overstate benefits, inadequate treatment of uncertainty, including limited sensitivity analysis, optimism bias that ignores implementation, mechanical discount rate selection, and a lack of analysis that seriously considers downside scenarios.

Psychological costs

Psychological costs are where the framework becomes genuinely illuminating. These are the stress, frustration, stigma, and erosion of dignity that accompany these processes. The anxiety that a single mistake on a form could have devastating consequences. The shame of being made to prove, repeatedly, that you are deserving of help or assistance. The frustration of dealing with a system that treats you as a number rather than a person, or that seems designed to make you give up.

I think where I am landing is that these three categories reveal the fundamental flaw in our current cost-benefit models. A clunky online portal might be celebrated as an “efficiency gain” because it reduces an agency’s administrative costs. But that analysis ignores the learning costs imposed on someone who is not tech-savvy, as well as the psychological burden carried by someone terrified of making an error. It measures the cost to the state whilst ignoring the burden shifted onto citizens or businesses.

This is not merely an academic distinction. It has consequences for real people. When we design systems that ignore learning and psychological costs, we produce policies that look elegant in Cabinet papers but fail in practice. Benefits go unclaimed by eligible people. Regulations are inconsistently followed because no one truly understands them. And a quiet alienation takes root among citizens who come to feel that the state is not, after all, on their side.

The administrative burden research documents this failure across multiple countries and policy domains. It shows that the problem is not simply inefficiency, but rather that the burdens fall hardest on those with the fewest resources. The very people these services were designed to help find themselves least able to reach them.

Over this series, we will examine how they manifest in healthcare, social welfare, and business regulation. We will consider evidence on how burdens fall unequally, how they can be gendered, racialised, and regionalised, and how they fall disproportionately on smaller businesses. We may even confront the uncomfortable possibility that, sometimes, the burden is not an accident but a tool used to limit access.

This series is not about finding easy answers. It is about asking better questions about access and who pays, and whether the current cost-benefit analysis framework truly reflects the full cost of the regulatory state.

Next month, we return to the foundational 2015 article that laid out this framework, and ask what it means to see administrative burden not as an accident of poor design, but as a political choice.

Author’s Note

A quick author’s note on what to expect from this series. Like Te Rā Whakamana, this series will bite off one idea at a time. The posts will be plainer and shorter than some of my other work. The fifty peer-reviewed articles spanning a decade of scholarship are rich and layered, and the core insights are not complicated. They have simply not been applied to Aotearoa. My aim is to make the ideas accessible, one at a time, so that practitioners and policymakers in Aotearoa can decide whether to use them.

References

Moynihan, D., Herd, P., & Harvey, H. (2015). Administrative burden: Learning, psychological, and compliance costs in citizen-state interactions. Journal of Public Administration Research and Theory, 25(1), 43–69. https://doi.org/10.1093/jopart/muu009