About behavioural insights

What are behavioural insights?

• Behavioural insights concern how people perceive things, how they decide, and how they behave.
• Behavioural insights are generated by empirical evidence from the behavioural sciences, which study human behaviour in an attempt to identify the factors that affect our behaviour.
• Behavioural factors include, for instance, our emotions, the behaviour of those around us, and our perceptions of the information that is available to us.
• Behavioural insights also help to anticipate people’s reactions to policy interventions. These interventions can be behavioural, such as changes in the information, context and situation. They can also be conventional, such as financial incentives or bans.

How do we gather behavioural evidence?

We use the following methods, alone or in combination

• Literature reviews provide a systematic overview of the available scientific evidence. They should be the first step before any other empirical work. Literature reviews help establish the value of a behavioural approach, warranting or dismissing it.
• Qualitative research uses focus groups, interviews and some sort of participant observation. This type of research is used often in preliminary phases, to better understand the underlying behavioural factors.
• Surveys ask people about their views, perceptions, beliefs, and intended courses of action. Surveys are useful in behavioural studies or by themselves. Their downside is that they capture what people say they do, not what they actually do.
• Experiments observe how people behave in more or less controlled situations. Participants are randomly allocated to a treatment or control group. The treatment group is exposed to an intervention designed to change their behaviour; the control group is not. Both groups' behaviour is then measured and compared. If there is a difference, it will be due to the intervention. Experiments can be conducted online, in a laboratory, or in real life (field experiments).

Why are behavioural insights relevant for EU policy?

• Behavioural insights have the potential to contribute to all phases of the EU policy cycle (see figure below).
• In practice, however, behavioural insights mainly contribute to the impact assessment process. This process consists in gathering and analysing evidence about the likely impacts of a planned policy.
• In the EU Better Regulation guidelines, we suggest a 4-step “DO IT” approach for applying behavioural insights to the impact assessment process:

1. Define the behavioural element
   If there is a behavioural element to a policy problem, the first step is to precisely define the relevant behaviour. What is it and how does it relate to the policy problem? Do we need to change behaviour or understand it better? Take the example of improving information in a product label. Is the aim to make the information clearer or to attract more attention?
    
2. Observe the behaviour and try to understand it
   Before addressing a behaviour, we need to observe it and try to understand what lies behind it (i.e. identify the behavioural factors). For this, we usually rely on a literature review or qualitative empirical analysis. How do people think, act and feel in relation to the policy problem? Is there some ‘ideal’ behaviour which people are not showing in the first place? Why would this be?
    
3. Identify policy options to address this behaviour
   Behavioural insights can help identify policy options to tackle the behaviour that lies at the core of the policy problem. These options can be 'soft' behavioural interventions. An example is changing the content and presentation of a label on food or appliances. But policy options can also be more traditional, like a regulation. An example is charging for or banning unsustainable plastic bags that consumers use because it is easier than remembering to bring reusable bags.
    
4. Test the effectiveness of these policy options
   Applying behavioural insights to policy-making is more than bringing a better understanding of behaviour. It is also about empirically testing in advance how different policy changes could affect actual behaviour. The tests take place in a controlled environment to see which policy options could be effective. Behavioural experiments are the usual method for testing policy options.