If we know better, why don't we do better?

The Common Good Project - Public Policy Roundtable

7 September 2026

---

Womenjeka

I would like to acknowledge the Traditional Custodians of the land on which we are meeting today, the Wurundjeri Woi-Worung   people of the Kulin Nation – and pay my respects to their elders past, present and emerging. I pay my respect and express gratitude to all Aboriginal and Torres Strait Islander people in the room today.

I have never ever started a speech with a quote - but today I will - and maybe not in the way you might expect.

Who has heard this quote before? "Do the best you can until you know better. Then when you know better, do better".

I confess I am a massive Maya Angelou fangirl. I think everything that comes out of her mouth is GOLD -

But when I first came across that quote, I remember feeling underwhelmed.

I didn't have that moment where the earth moved beneath me like I did with all her other quotes.

In fact, my reaction was something closer to:

"Well ... isn't that obvious?" Of course if you know better, you do better.

But coming here today and having the chance to reflect on the topic of our conversation I suddenly reflected that in public life – we don’t! We know better and we keep paying to know even more and confirm what we know - but for some reason we don’t DO better.

In fact the wonderful Ms Angelou’s quote has been gnawing at my heels my whole life and I didn’t even know it: as a bureaucrat, as an oversight body, whether in child protection, youth justice, disability, mental health, Indigenous Affairs, in aged care – AND it even irked me as the bright eyed bushy tailed lawyer who came out of university wanting to save the world; thinking anyone who held power in those roles did too.

Two decades later, I came to feel like nobody wanted to do better even if they knew better.

But now, as the old cynic inside me has thankfully curled up and died, I have asked myself: was it because the powers that be didn't want to do better? Because it consumed more energy to change than fight the flames of staying the same?

Or was it something else?

THAT - is what this speech is about

This speech is about the persistence of failure despite knowledge

So thank you for indulging me this conversation, and if you will: this therapy session.

“We must do better.”

It’s a phrase I have heard uttered by every shade of government in every jurisdiction after Royal Commissions, inquiries and independent reviews. I have heard it after scandals, tragedies and public failures; I heard it straight after we delivered our Review of the National Agreement on Closing the Gap when I was a Productivity Commissioner.

It sounds entirely reasonable, right? In fact, it sounds so reasonable that we rarely stop to examine it.

But over the years I have become increasingly uneasy whenever I hear it.

Not because I disagree with the sentiment, but because in many areas of public policy, we already know what better looks like.

We have shelves full of reports. We have inquiries that tell us what went wrong, evaluations that tell us what works, and international comparisons demonstrating that different outcomes are possible. In aged care, we even have a parallel model in our own country that exceeds the outcomes and cost-effectiveness of our mainstream model.

Importantly, we have communities and people with lived experience explaining – repeating over and over - what would make a difference.

Yet, after all that work, we still find ourselves persisting with systems that keep delivering the same problems.

So if we know better, why don’t we do better?

That question has followed me through child protection, Closing the Gap, youth justice and aged care.

The longer I have worked in public policy, I am convinced that our central challenge is NOT a shortage of evidence.

I think we have become exceptionally good at diagnosing problems, but considerably less comfortable redesigning the systems that produce them.

I saw this most clearly in aged care. The evidence showed that helping people remain independent at home was not only better for older people, it was often significantly cheaper than waiting until decline had already occurred and responding through crisis care.

In other words, the human answer and the economic answer pointed in exactly the same direction. Yet we continued making early support harder to access and crisis support easier to fund.

And what really began to puzzle me was that even sustained public scrutiny didn't seem to change the trajectory. Governments were being publicly criticised for aged care waitlists, questioned in Parliament, subjected to Senate Inquiries and even another round of Four Corners. They were paying a political price for the status quo, while the status quo was also costing them more money.

And here is what completely flummoxed me as Inspector-General of Aged Care: If changing course creates better human lives, costs less for taxpayers, and if sustaining the status quo leaves governments being set on fire in the media, then what exactly is stopping change?

More than that, I began to wonder whether something bigger was at stake. Because if governments repeatedly acknowledge the evidence, repeatedly promise change, and repeatedly struggle to deliver it, sooner or later people stop asking questions about the policy... They start asking questions about whether government itself is capable of learning.

But before we get to exploring that, I think we need to understand why change is so difficult in the first place.

One of the patterns I have observed throughout my career is that our appetite for diagnosing problems often exceeds our appetite for changing the systems that produce them.

A scandal emerges - an inquiry is established - a Royal Commission exposes failures that leave us all shaking our heads and asking how this was allowed to happen. Governments promise to do better.

But when the report finally arrives, often containing a hundred or more recommendations, the discussion shifts remarkably quickly. Instead of asking

"What would it take to redesign this system?" we ask "What's the lowest-hanging fruit?", " and "Which recommendations can we say we are already adopting?" 

Reform gradually becomes an exercise in adapting the existing system rather than questioning whether the system itself needs to change.

I do not say that cynically. These discussions often involve dedicated public servants trying to navigate fiscal constraints, legislative processes, electoral expectations and the daily pressure of keeping a system operating.

But the effect is profound.

What happens is that the ambition narrows – and the focus shifts from redesign to adaptation.

Recommendations are translated into project plans. Project plans become work streams. Progress becomes something reported against milestones. A new oversight committee may be established, training introduced, reporting strengthened and a strategy published.

Some of it may be worthwhile - but the existing system remains fundamentally recognisable.

There is an important difference between administering recommendations and delivering reform. When we lose sight of that distinction, a government can complete the process while the underlying problem survives - and the outcomes do not substantially change.

What I find useful about Mazzucato is that she forces us to stop looking only at what governments say they want. She asks us to look at how they've designed the system to achieve it: what I refer to as the theory of change.

Outcomes don't emerge by magic. They emerge from the incentives, relationships, rules and institutions we've built around them. If we want different outcomes, inevitably we have to be brave enough to dispense with the one we’ve got

Procurement, loans, grants, subsidies, tax incentives, bailout terms, IP arrangements, water and property rights - these aren't administrative plumbing, they're where direction either gets embedded or doesn't.

And if we aren’t prepared to re-engineer these things, then most often we end up with a system designed to make the outcome we're seeking harder to achieve.

Aged care is the perfect example of this: The public purpose of aged care post-

Royal Commission is now legislated: Dignity, connection and independence.

But if funding rewards transactions, performance frameworks measure outputs, and regulation directs enormous energy towards demonstrating compliance, those arrangements will shape what the system delivers. It can inadvertently make a mockery of the policy and legal intent.

Seen through this lens, aged care is not simply an argument about whether government or markets have failed. It is an example of governing arrangements that do not consistently support the outcome they are meant to deliver.

Let me return to where I most often hear the annual refrain that "we need to do better": Closing the Gap. The National Agreement began with an unusually honest recognition: that previous efforts were not changing outcomes and that governments THEMSELVES would need to work differently if the gap was ever to close.

But when Commissioner Mokak and I travelled the country conducting the Productivity Commission review, what we heard was not a shortage of ideas.

Communities understood the barriers, the trade-offs and the solutions.

Governments had actually recognised that something needed to change and had committed to shared decision-making. But what many communities described was a gap between the aspiration and the practice.

Often, the way decisions were made remained largely unchanged. Business as usual acquired new language, but remained business as usual.

And that, I think, is where Closing the Gap intersects so powerfully with Mazzucato's work. The issue is not simply what outcome government says it wants. It is whether the way power is exercised, decisions are made and relationships are structured, actually makes that outcome possible.

If the way power has been exercised helped create the problem, we should not be surprised that the problem persists when the distribution of power remains largely unchanged.

In many ways, Closing the Gap is the perfect example of the broader conundrum I've been describing.

None of what I have just described is hidden knowledge; the problem itself has been described in the text of a National Agreement. We can't say governments haven't acknowledged it.

Yet progress remains frustratingly slow. Which forces us to confront a much harder question: if knowledge isn't the problem, what is?

This is where the literature on policy inertia becomes useful.

Once systems become established, entire ecosystems form around them. Infrastructure, legislation, budgets, contracts, reporting frameworks and professional identities become connected to existing arrangements.

Over time, the system accumulates weight.

Each new decision is shaped by the decisions already made. Reform is not designed on a blank page. It must contend with investments, expectations and interests that have grown around the status quo.

Evidence never arrives on an empty field.

I do not think most ministers set out to ignore evidence and I do not think most public servants set out to preserve inefficient systems. Many of them agree that change is needed.

The problem is that evidence enters systems with incentives already attached.

Successful prevention is often invisible.

A crisis that never occurs rarely attracts attention.

A child who does not enter youth detention because we supported them early with the right intervention, is not a headline. An older person who remains independent at home for another year does not generate a press conference.

Perhaps the surprising question is not why evidence struggles to prevail. Perhaps it is why we imagine evidence could automatically overcome accumulated infrastructure, institutional identities, established funding, political risk and public anxiety.

Nor does a crisis necessarily remove those forces.

We tend to imagine that a crisis creates its own mandate for transformation – I’ve long been guilty of imagining that! But crises do not arrive carrying an agreed interpretation.

A youth justice crisis can be taken as evidence that we should reduce reliance on detention and turn our attention to the interventions we know that stop reoffending. But sadly, as you see in my own State of Queensland, it can somehow just as readily produce calls for more detention capacity.

Aged care waiting lists can be understood as evidence that Australia must prevent avoidable demand, or simply as proof that we need more residential aged care beds.

A crisis opens a window. It does not decide what comes through it.

It seems that unless a credible alternative has political support, institutional capability and a compelling public explanation, the existing system is likely to absorb the criticism and survive. It may even use the crisis to justify expanding the model that produced it.

This is where Mazzucato’s warning about state capability becomes particularly relevant.

Government cannot shape a system it no longer properly understands. Nor can it write, negotiate and enforce serious public-interest conditions if its own knowledge and capability have been hollowed out – when in fact these are the very factors that are the predicate to change.

A common-good approach cannot therefore amount to telling government to become bolder. Government must also have the capability to learn, design, negotiate and act.

It must be able to draw knowledge from outside government without merely extracting that knowledge during consultation and then returning to make the essential decisions alone.

We must begin with the outcome, understand the system producing the present result, and identify whose knowledge and authority will be needed to change it. Funding, incentives, institutional relationships and accountability can then be designed around that purpose.

That is what a theory of change contributes. It forces us to explain what must actually change for a different outcome to become possible, rather than beginning with what the existing system finds easiest to deliver.

Until this point, I have been talking largely about what our failure to do this means for policy outcomes.

But there is another cost, and it may ultimately be the more serious one.

Repeatedly promising change without delivering it, teaches our voting public something about whether government can be trusted at all.

Citizens do not encounter democracy primarily as a theory. They encounter it through the institutions that make decisions affecting their lives.

They learn what government means from the way an agency treats them, whether promised support arrives and whether speaking up changes anything.

They observe whether institutions can acknowledge that an approach has failed and then alter it.

Policy-feedback research shows that policies do not simply emerge from politics. They also shape how citizens understand politics.

Repeated disappointment in particular programs can gradually diminish the broader confidence people hold in public institutions.

Across a range of research areas, we find that when government says it has listened, “learned the lessons” and accepted the need for transformation, but people experience little material change, the consequence is not confined to that reform.

People begin to doubt whether participation matters or whether evidence has any meaningful influence over decisions. Eventually, they may doubt whether public institutions are capable of learning at all.

A poorly designed system does not only fail to deliver its stated service.

It can teach people that government is distant, unresponsive or hostile.

And as you might have expected me to say next, there is also a wider political implication.

Declining trust increases pressure for highly visible responses that convey control.

Governments become more sensitive to the politics of reassurance. Symbolic policies begin to displace slower and less visible measures that may be more likely to change outcomes.

When those symbolic responses fail, distrust deepens and the demand for another display of control grows stronger.

Youth Justice is the perfect example of this. Declining confidence can create demand for punitive action. Punitive action may fail to reduce offending and may further damage trust in the communities most affected. That failure then produces another demand for government to appear tougher.

Ineffective policy and political distrust begin to reinforce one another.

That leads to a question larger than the one with which I began.

If we repeatedly know better and repeatedly fail to do better, what does that teach citizens about democracy itself?

That is why the gap between evidence and action matters so much.

It doesn’t only waste money or weaken services. It diminishes confidence in our collective capacity to solve shared problems.

Across all of the examples I described, the pattern is remarkably similar:

• We identify the problem.

• We collect the evidence.

• We commission the inquiry.

• We accept the recommendations.

• We promise to do better.

And yet somehow, far less changes than we would expect.

If citizens repeatedly see governments acknowledge failure, commission inquiries, accept the evidence and promise reform, yet little changes, eventually they stop asking whether a particular policy works.

They begin asking whether government itself can learn.

And that is why this matters.

Because what is ultimately at stake is not merely the effectiveness of public policy.

It is the credibility of democratic government itself.