The machinery of change
How political directions filter through a system built for routine
“The machinery of government” is a phrase you come across a lot in political- and public management papers. As a metaphor, it takes ministers and their ministries, department, agencies and crown agencies as cogs in a machine that either work well together to produce policy, or don’t.
I believe it’s a favoured phrase because it implies intent, an engineering mind behind how everything clicks into each other. And if it doesn’t work like greased lightning, you can get your toolbox out and fix it. The usefulness of the metaphor comes to an end for me when I encounter people who display a tendency to think of people in the same way: when they are part of a system like a workplace structure, they should understand their role as a cog in the engine, do their job as it says on paper, and shut up.
Still, whether we want to find a “squeaky wheel” in a machine, or we want to look for patterns of behaviours that play out against our interest and intent, it’s a compelling picture. So today I want to write about change and reform in the public service from a perspective that probably peeked through on previous posts, but that deserves a more focused discussion.
For that, I need to wade into an area that I’m not entirely confident in: politics. While political leaders are a part of public mangement, it still feels like a strange position because as an academic – like public servants – I don’t see my work as political. I’m interested in a public service that can do the best for the people it serves, and the methods we have to do that. But every now and then I get asked for my opinion on how different well (or badly) political coalitions in government have handled the public service. And it would be ignorant to claim that politics don’t colour my opinion and my individual bias.
I do think my doctoral work has something to add to the mix, as the basis of it are two big chunks of data - each covering a different administration in NZ: I have a dataset of the 486 (reported) restructures from 62 wider public service institutions that happened between 2018 and 2021 (Labour-led), and a survey on the restructure experiences of over 1,400 public servants and union members from June 2024 (National-led).
Each dataset is one window into public management at the time, and while there’s always a lot more to each story, I’ll break down some basic observations I can make of each. For the sake of the bigger argument, this will mean some solid simplification - grain of salt please.
Growth and cuts
Do y’all remember the “550 million contractor blowout”? That was 2018 when the sector’s contractor spent was interpreted as droves of public servants cruising on waves of successive contracts, where they should have been regularly employed – and much cheaper – workers. Don’t ask me if that was a fair interpretation in hindsight, honestly, but I do recall thinking it made sense at the time when the Labour-led administration removed the hiring cap for public servants in June/July 2018. In the three subsequent years the total workforce in the public sector increased by 19%. unclear is – to what extent was this “dark” public servants moving onto the official payroll as intended, or did the combined force of employees and contractors balloon as a whole?
I won’t answer that question here (and I suspect some people may have strong feelings about that), but this development was (one of the) reasoning for the National-led government to require between 6.5% or 7.5% blanket budget cuts of public service institutions in 2024, based on the proportion of their previous growth.
You could simplify down the developments of the last two governments as: one grew the public service, and the other one cut it down. The language being used here tends to be around “right-sizing” – no matter what side you stand on. Whatever “right” means…
But the thing is:
The machinery that processed either budget or increases or cuts is the same. So if we don’t think it can do one of them efficiently – why do we think it can deliver the reverse?
Perhaps there are more material differences at play? When I look at some tendencies that I can see in my evidence base, it comes down to this:
Under Labour-led government, there is quite a lot of “strategic” restructuring with lines between directorates and teams re-drawn, quite a bit more specialisation of roles especially in digital- and IT teams, but also more investment in frontline and operational teams and the back-office workers that inform them (e.g. data specialists). You also find trends where role titles across larger teams are being removed and replace with more generic titles - a typically swinging pendulum like centralisation and de-centralisation of functions that characterises all restructuring.
Despite the fact that I coded every SINGLE job that was moved (or not moved) in these restructures (yes, I know…), I can’t report massive drifts like loads more middle-managers being hired or expansive Comms teams taking reign. There are certainly examples of that, but there are also examples of the opposite. If I was to make a judgement, it’s the vapidity, abstractness and ambiguity of the restructure motivations, and the shocking lack of explicit connection between operating model and structure changes that I find concerning. But overall, the growth just seems to be that - growth of people working in more or less the same trends as before.
Judging from what I can compare in the National-led years, the principles remain the same under a different name: restructures refer to the need for budget reduction and “right-sizing” explicitly - but very soon after framing that, they return to the same efficiency-silo-streamline language where the changes are supposed to “enable, empower, deliver” the new strategies. Lines between directorates are re-drawn, functions are combined, but in some examples the focus seems to be on trimming FTE’s wherever possible and leaving broader structures in place.
It really seems to be that plain: whether a government provides more generous budgets or cuts them - the logistics and logic by which that is being enacted are the same. And that’s the issue in its link to political direction:
You can neither chuck more money at a public service issue and expect it to be fixed, nor can you impose cuts and expect “waste” to be eliminated until you reach some sort of “efficiency” equilibrium.
I’m not saying that either ministers actually believed it was that simple. It’s entirely right that political leaders work with the public institutions in their portfolio to make policy in their estimation. And importantly, it’s right that the same politicians leave it up to the discretion and expertise of organisational leaders to translate their directives into practice. That’s the fundamental separation of powers on which our democracy lives.
But while we want to think of the machinery of government as something rational, evidence- and expertise-driven, stacks and stacks of academic studies show us that it isn’t.
There are the cycles of routines, expectations, incentives, echo chambers and habitualised behaviours that bring their own dynamics, each knocking over dominoes that fall through each blessed tier, no matter how well-intentioned leaders are.
And while I say this is not necessarily “rational” – it is nevertheless predicable. In fact, the dynamics of how change impulses of any nature are processed through the machinery of change play out in pretty clear patterns.
The machinery of change
I believe it deserves a phrase on its own. It’s related to the machinery of government, but it’s like one of those parts in an engine that every impulse gets filtered through in flash, and that ultimately has flow-on effects to its every action.
We pay oddly little attention to how much resource this machinery of change “consumes” on its own as it works. We treat it as some kind of yaddayadda blackbox, like some kind of pokey where we put in a coin in the top and expect more coins to drop out the buttom – but the one-armed bandit is a hungry sod.
When it comes to change – particularly the kind that is initiated by some sort of political directive - the main mechanism that inevitably comes to play is the restructure (in all its euphemistic names) – my dear old hobby horse.
I’ve written before about how the dynamics between leaders and affected staff play out and stabilise – rather than alter – the institutional settings. For today’s argument, let me summarise that our machinery of change in Aotearoa has three main characteristics:
It’s top-down managerialist: While I’m certain that managers care about the impacts that change has on their staff, and they try to limit any “damage” – the design of the change is an exclusive managerial prerogative. In about a third of instances, staff are asked for their pain points and solutions at an early point, but the processing, prioritising and solution-process happens behind closed boardroom doors. I’ve written more on why that is limiting change efficiency here.
Supporting activities to connect structural- and practical change are not included: As I’ve detailed here, any strategic management handbook, and every successful transformation we’ve case-studied carefully couches structures changes in thorough process & policy reviews, launches of new tools, training and governance reviews. By and large, we do not.
Whether the intended goals were met, and what other (unintended) effects a restructure had, remains largely unmonitored: Implementation plans end after the structure is “in place”. As I reported before, only 7 out of 486 restructures performed any kind of structured follow-up. The others referred to “informal” assessments and broad metrics like staff satisfaction surveys.
Last week I wrote about a notable exception in IR’s near-decade transformation – but conversations I’ve had since only highlighted the special conditions surrounding this example, which has also not been without criticism.
The biggest cog is setting the pace
I maintain that New Zealand has an unusually high pace of institutional change due to our single-chamber, 3-year cycle, direct ministerial-relations setting, to name a few. That circumstance, perhaps more than anything, shapes how change happens. Because:
In a world where change is constant, the machinery is refined for fast intake from the top, for quick processing and making-the-best-of, given the circumstances.
Seriously, the way we manage change could not be set up better to constantly respond to political demands and process new requests. Especially our public leaders are self-selected to be able to handle and convey it. And they spend a lot of time and effort to fulfill that role.
So while the hopper of the engine is fine-tuned like a foruma-one pitstop, what about the lower stages of the machine, and the output? With few checks and balances on the intersection to practice, long-term effects, and a pesky lack of metrics, the action-reaction chain of this machine becomes pretty darn murky very fast. The personal strive for personnel, at the same time, is all too real.
To circle back to our examples, and to put a fine point on it:
I’m really not certain if a bigger investment in the public service, filtered through the machinery of change, leads to the improvements and outcomes that the leading party wants to see.
And I’m even less certain that broad budget cuts that simply come with a “spare the frontline” disclaimer lead to the elimination of “waste” and leave a lean-mean-public-service-machine. With the complexities that come with this beast of a system, it’s more than likely that left-pocket-right-pocket trade-offs happen at best, and at worst, that such directions lead the public service to spring a whole bunch of new holes through which tax-payer money drains away - like in this claim.
Not broken
Make no mistake, the machinery is set up from where the power hails: it’s a reflection of the power structures and how direction flows from the top. We have no news on what the new Public Service Act will bring yet, but considering that the statements to far speak of CE responsibilities, role clarifications and appointments (a pretty high score in restructure-language-bingo), we can expect that a potential new act will operate on that top end of the chain of command, yet again, rather than paying more attention to the engine humming underneath and its systemic outcomes.
I realise that I over-simplified large parts of the relationship between ministers and the public service – there are plenty more fishhooks to untangle for another day. But for today, the sentence that keeps coming to my mind is by Lynne Cazaly:
“We need to change the way we lead change.”
And I’ll add:
If we want systemic change – we should recognise the machinery not as something that is broken, but as something that performs exactly as intended.
And if we don’t like what it does, then it’s up to us to find the cogs that need to be replaced by different methods. And more than anything, we need to find ways for conveyer belts to transfer continuously between those who hold decisions, and those who get churned through this washing machine only to find themselves in the next cycle.
So what do I think about how different administration handle the public service? I begins to feel to me like political leaders truly are the only ones in our system that could turn their attention to how the cogs are put together, before throwing a decrease or increase of funds into the same hopper and expecting a different outcome than the previous administration had. And that’s where I’ll leave that.