Skill up

pattern

Blog: New Frontiers in behavioural insights

Sam Hannah-Rankin - Director Public Sector Innovation

It was a great privilege to be at the Behavioural Exchange 2018 (BX2018) conference and listen to the great thinking and work that is underway in behavioural insights internationally.

There were some really consistent themes - particularly around the importance of focusing on a problem and thinking systemically.

A common theme throughout has been looking at how we can use behavioural insights systemically to achieve a policy outcome.

This is close to the heart of what we're doing in Victoria.

Our work in behavioural insights is situated in the much broader context of public sector reform: moving the people, systems and accountability of government to deliver better outcome for citizens.

It means that we're looking for opportunities to make enduring change - and working on these both on-the-ground and from within the government system.

There are four aspects to this that I'd like to draw out, because I'd like to think that these ways of working will be part of behavioural insights’ new frontiers.

1. We want to use behavioural insights systemically to achieve policy outcomes

Government works in complex multi-sectoral areas - consider family violence and the multitude of areas (health, housing, justice, emergency services, etc) working both reactively and proactively to address, combat and eradicate this issue. When we're working in these spaces, we need to identify the areas of greatest opportunity for impact. We focus on problems, and then look at what we can do to improve the situation.  

The challenge is that these opportunities are not restricted to any single discipline.   

In this context, behavioural insights is one of many policy tools available to support change. Several people at BX2018 talked about having a whole toolbox of different approaches, where you can choose the right tool for the situation. 

The exciting new frontier is that the potential is even more than just picking the right tool from the toolbox - it’s about actually integrating approaches across discipline silos to achieve the best outcome.

We're lucky in Victoria because our Behavioural Insights Unit shares the public sector reform space with branches that specialise in design and innovation, data and analytics, policy and process, and outcomes and evidence - and the great opportunity is where this all comes together.  

We're learning how to bring these different disciplines together, but it requires a lot of flexibility and trust and active collaboration - and I know other governments are working on this as well.

So one new frontier we'd like to see is behavioural insights being integrated with other disciplines, looking at how we can bring the strengths of behavioural economics and psychology to bear in coordination with design, agile development, data analytics, change management and delivery.

2. We need to apply behavioural insights to policy design

There’s increasing thinking about how we can apply what we know about human behaviour earlier in the policy design process.

All programs and services sit within a broader policy architecture. Thinking about human behaviour when we design policy has the potential for much greater impact than service optimisation alone.

This means using behavioural insights at multiple points in the policy cycle - to understand the problem and design the policy architecture, as well as to optimise the services that ultimately sit within this framework.

But traditionally the use of behavioural insights has been driven by experiments. Behavioural insights has been heavily shaped by its use of Randomised Controlled Trials (RCT) to experiment with and determine the best approach to policy delivery.

Experimentation and quantitative analysis are immensely valuable contributions to government policy-making, and play a critical role in optimising service delivery.

The challenge is that experiments are scoped tightly by necessity, with limited variables in structured environments. Policy development and a lot of service delivery happens at a much larger and more complex level, so we need to be aware that behavioural insights means much more than experimentation alone.

Without intervention, policy making will continue to happen based on traditional approaches - including a reliance on economic models that we know don't represent how people actually behave.

The opportunity in behavioural insights is about continuing our focus on trials as an important way of testing and informing our work.  

And we also need to bring behavioural perspectives to bear in policy development.

We can't always experiment to find out what works - but we should always be trying to bring the best contemporary thinking into policy development.

This will allow behavioural insights to be used to tackle increasingly complex areas, where the messiness of life may not allow us to isolate the neat variables that randomised control trials rely on.

3. Turning behavioural insights on ourselves - applying behavioural insights to internal processes and decision-making.

I hope that the next frontiers of behavioural insights application will involve applying behavioural insights to how we work, in addition to what we deliver.

On day one of BX2018 we heard about the World Bank's work to explore how our biases - as public policy professionals - play out in practice.

The consideration of ourselves as choice architects (designing how options are presented) in the public, private, not for profit - or even academic sectors! - will be an increasingly important application of behavioural insights.

We are all human, we are all inherently biased; and there are many opportunities for us to improve our decision-making.

Some of these changes are about our internal processes - small interventions around how we structure meetings and engagement.

4. So much of behavioural insights’ origins have developed as a reaction to traditional neo-classical economic models

But modern economics has shifted, and taken a lot of psychology and new thinking on board, so the distinction isn’t as clear as it was.  

In the meantime we're seeing new disciplines emerge that could broaden and inform the concept of behavioural insights.

We need to do more to get policy-making out of the office and out with the people who deliver and receive policy. Because this is where the actual policy is made - not in the papers, but in the street level decision-making where the rubber hits the road.

There's also the opportunity for us all - policy-makers, academics, experimenters - trying to move away from thinking about ourselves as scientists working on people, and more about how we can work with people to uncover and understand the situation as it is.

This has always been part of behavioural insights methodology, but it’s sometimes been less emphasised than other parts of the process. Understanding the context in which behaviour happens on the ground is critically important.

This new approach to policy design and more agile and iterative development of thinking and implementation requires more interaction between us all.   

The publication cycle for peer reviewed findings is long and challenging, and the divide between sectors is deep, so we need to find new ways of sharing findings and learning from each other on an ongoing basis.

From a Victorian perspective, we're working to make the new frontiers of behavioural insights ones that are:

  • permeable to new disciplines
  • cognisant of our own limitations as human actors in the system
  • impactful in delivering positive change

We are all learning in this space, and I look forward to working to uncover new models and ways of working across the sectors - to bring the full rigour and science and boundary-pushing potential of behavioural insights into policy-making.