Why Martin Parkinson wants a second generation of behavioural economics

By Martin Parkinson

June 25, 2018

Behavioural economics began small — assisting people to keep their appointments — but it’s about to get a lot wider. The head of the Australian Public Service, Dr Martin Parkinson, today called on public servants to embrace randomised control trials, question assumptions, and make humans the centre of their next idea.

Towards the end of 2010, Time magazine headlined on its front cover its 50 best inventions of that year.

In the magazine’s words, the list featured the “biggest and coolest breakthroughs in science technology and the arts.”

At the very top of this list, coming in at number one, was an invention called the NeoNurture.

The NeoNurture was an incubator. It was aimed at saving the lives of babies in countries that might not have ready access to modern medical devices

The goal was clear: to build a simple, cheap machine that could be used in developing countries to keep premature babies warm in those precious first few days of their faltering lives.

The “genius” of the machine – to use Time’s description – was that it was made from old car parts.

The headlights would provide heat; a repurposed dashboard fan would circulate air; a door-chime and signal-light assembly became an alarm system to alert caregivers when something went wrong with the heating. The device could even be powered from a motorcycle battery.

The parts were to be relatively easy and low-cost to source, and easy to maintain – an important consideration in the sorts of countries it would be used in.

It sounds fantastic: a potentially simple solution to a very serious problem. There are, after all, no problems of a greater magnitude than those concerning life and death.

How many premature babies do you think have since used the NeoNurture since 2010? The answer is: one.

Only one baby ever used the incubator. And that was the baby who modelled it for the Time magazine photoshoot.

The NeoNurture failed. But why did it fail?

At this point you’re probably expecting me to say that the inventors forgot to put humans at the centre of their design.

But that’s not quite right.

One of the inventors said he and his research team conducted months of user-research, speaking with people on the ground, in the hospitals.

They listened to what people said they wanted, paying close attention to the people who would use their machine, and they designed a machine intended to benefit poor families, rural doctors, overloaded nurses, repair technicians.

In the inventor’s words: “we thought we had all our bases covered.”

So what went wrong?

The inventors didn’t account for all their stakeholders. They failed to recognise the humans who would play a big role in getting the project off the ground and into hospitals in the first place: the humans involved in manufacturing, financing, distribution, and regulation.

They actually needed to account not just for the humans who would use the machine, but also the humans who would choose and pay for that machine.

There were also some concerns about how the machine looked – and whether parents would really want to put their fragile new baby into a machine cobbled together from car parts.

The good news is that the inventors recognised their mistakes and applied those to the development of the Firefly incubator. In 2016, these new units were used in 23 developing countries.

“Solutions to some of our really big problems can lie in the smallest, simplest solutions … but finding these solutions requires correctly identifying the right end-goal.”

This story tells us a few things.

It tells us something about the importance of thinking broadly and inclusively about all of the stakeholders.

It tells us about the importance of identifying just who that human is when taking a human-centered approach.

And it tells us that failure can sometimes be our greatest teacher.

And I also think it’s illustrative of some of the contradictory forces at play for those of us charged with trying to make the world a better place: that solutions to some of our really big problems can lie in the smallest, simplest solutions … but finding these solutions requires correctly identifying the right end-goal.

Why is this important? What problems are we facing?

These sorts of considerations are more important now than ever. Like all countries, Australia faces some significant public policy challenges.

Take the area of health – which I note is a focus for one of the breakout sessions this afternoon.

We know that, overall, health outcomes are improving.

Australian males can expect to live to just over 80 years of age, and Australian females just over 84 years. This is five to 10 years more than other OECD countries, where, for example, Polish men live to around 73 and Mexican women to almost 78 on average.

However, the number of Australians who are overweight and obese has risen to well over half the population. People are having to manage modern illnesses like Type 2 diabetes – particularly in our Indigenous communities.

We’re healthier but we’re at greater risk from more complex illnesses than in the past.

We’re also getting a more detailed and complicated picture about some health interventions, and we’re grappling with the repercussions of those.

For example, the over-prescription of antibiotics has led to increasingly drug-resistant superbugs. And we have seen recent advice suggesting arthroscopic knee surgeries may not produce any better results than avoiding surgery altogether.

There are also significant challenges with our environment.

Australia produced 64 million tonnes of waste in 2014–15 – an amount that is almost impossible to wrap your head around.

We’re going to need new and better incentives and solutions to use less, recycle more, and turn our waste into energy.

There are of course many other challenges, all serious and important  – our ageing population; housing affordability; the scourge of domestic violence; energy security; congestion in our cities; infrastructure in our regional and rural areas.

We need to tackle those challenges while navigating a world that has changed, and is changing, at breakneck speed.

Ten years ago there were six billion connected devices online. Today there are 15 billion.We’re seeing huge advances in bioengineering, artificial intelligence, robotics, and block chain.

This pace of change is being referred to as the ‘fourth industrial revolution’ and we need to buckle up for a wide ride which will present enormous opportunities and challenges for our economy and our society.

As I’ve said before, these technological changes may help drive economic growth and new jobs, which would be a fantastic outcome.

But they might also lead to some jobs being made ‘technologically redundant’. Some full-time jobs may become one-off jobs in the gig economy; some workers may see their wage growth flatline.

A lot of these changes are giving us more life choices, but the flipside is that they may lead us to choose the wrong thing. The wealth of information we have available is creating a scarcity of attention to devote to it.

The disruption we’re facing is also reflected in the international strategic order, in political debate and in our societies and communities.

Australians are also expecting more, and rightly so. One of the corollaries of a more successful – and richer – society is that people demand more from their governments and the services they traditionally supply.

How can policy makers use behavioural economics to address modern complexities?

Not even the most enthusiastic and dedicated champion of behavioural economics would seriously suggest that it is the panacea for the medley of modern challenges I’ve just outlined.

But for those of us making policy, I believe that it offers us some suggestions on a better way to do things, a way for us to anchor the hurricane, to consider an alternative approach.

Behavioural economics reminds us that people don’t always behave in ways we expect. It reminds us to question our assumptions on how well we know what people want or think, how they engage and make decisions, what shapes and drives their daily interactions.

It teaches us to ensure we don’t just consider our own perspectives, but those of a broad and diverse group around us.
“We need to make sure that our services actually suit the person who receives them, instead of the organisation delivering that service.”

As I said in an address to the APS last year, being in the public service means we need to get to know the public we serve – in all the depth and breadth that encompasses.

Our work needs to reflect the fact that behaviours, expectations and challenges look very different if you live in regional and remote Australia than if you live in a city.

It needs to account for the experiences of minority groups, for the perspectives of businesses large and small, and for households of big families or for those living alone.

We need to make sure that our services actually suit the person who receives them, instead of the organisation delivering that service.

Behavioural economics in the public service

For these reasons, I’m proud to have BETA – the Behavioural Economics Team of the Australian Government – in the Department of the Prime Minister and Cabinet.

BETA is working across the federal public service, working collaboratively to make sure humans are at the centre of our thinking.

One benefit in particular that I see is in the use of randomised control trials or RCTs.

There’s nothing new about RCTs. The medical profession has been using them for years. In a clinical setting they’re a way to test, as definitively as can be known, what works and what doesn’t.

And yet in the context of public policy, testing what works is not routine. Not only can these methods lead to better evaluation, but they are also frequently cheaper.

Proper evaluation ensures we do something different because of what we learnt (or didn’t learn) about human behaviour.

It helps us understand how people will respond so we can build those responses into policy from the beginning.

But we have a long way to go.

The public service needs to get better at improving citizens’ experience of government – and this imperative forms part of the Terms of Reference of the Independent Review of the Australian Public Service announced last month by the Prime Minister and the Minister Assisting the Prime Minister for the Public Service.

We need to become better aware of our own biases and limitations.

We need to be working across paradigms. Reaching out to the private sector, to academia, to a local community, wherever the ideas are.

They might be found by talking to patients and doctors at a GP clinic in the outback, or by scrutinising a huge dataset.

There are also great opportunities to combine behavioural economics with some of the most exciting technological advances. For example, machine learning applied to behavioural economics problems offers the potential to produce personalised defaults.

These could have applications across a whole range of government services.  This field has the potential to reduce the problems that are sometimes associated with one-size-fits-all defaults.


The number of people here at BX 2018 confirms the growing interest and depth in this field.

With Richard Thaler and Daniel Kahneman now both recognised with Nobel prizes in economics (along with several others who could also claim to be behavioural economists), it’s easy to forget how far the field has come.

Behavioural economics began by assisting people to keep their doctor’s appointments, pay their bills, and save more for retirement.

But the scope is widening to include getting employees back to work quicker, helping nurses track patients’ vital signs and managing peak traffic demand in major cities.

And it needs to get wider still.

I note with interest that tomorrow afternoon’s Great Debate will be arguing the toss on whether “There are no more ‘easy wins’ in applied behavioural insights”. I’m sure it will be passionately argued.

It’s a good question.

How can we support the various behavioural economics units in Australia, and those around the world, to continue to look for new wins?

Is there a ‘second generation’ of behavioural economics we can harness to guide our approach to tackling the tough challenges we face in our hospitals, our homes, our oceans, our schools, and in our remote communities?

At the end of last year, I put out a challenge to those of you in the APS to think of your big idea. After all, as policymakers, ideas are our stock-in-trade. We develop them, deliver them, refine them.

We need to ensure our ideas are not simply part of a Departmental echo-chamber.

I invite you to consider the new ways – or the new frontiers – where we might be able to apply behavioural economics to make the world a little bit better, easier, cleaner, simpler, or faster.

I hope it leads you to big ideas which are simple, yet sophisticated, and have people – the right people – at their heart.

This address was made at the BX 2018 conference in Sydney on June 26. This adaptation was first published at pmc.gov.au.

Want to ask the experts?

Several of the international speakers at BX 2018 will be heading to Canberra later this week for a follow-up discussion.

BETA/PM&C is partnering with the Institute for Public Administration Australia for Behavioural Insights: Global Perspectives, on Thursday. The event will feature David Halpern along with OECD, Canadian and French panellists.

About the author
Inline Feedbacks
View all comments

The essential resource for effective
public sector professionals