Behavioural Science & Policy-Making in the EU

• Notes on lectures and articles about Behavioural Science, BPP, and more...

Baggio et al. (2021). The evolution of behaviourally informed policy-making in the EU

• Behavioural science (BI) is increasingly relevant in policymaking globally.
• Traditional policy models assume rational individual decisions, but behavioural research indicates decisions are influenced by heuristics, biases, and environmental factors, and this should be considered in policy design.
• Behavioural science improves policy effectiveness by helping to understand how people make decisions, designing nudges, and enhancing policy outcomes.
• BI is used in healthcare, finance, environmental policy, and consumer behaviour with examples including automatic pension enrolment, tax reminder designs, and cigarette package warning labels.

Evolution of Behavioural Science in Policy-Making

• BI in policymaking has evolved in three phases: Exploratory, Development, and Consolidation.

Exploratory Phase (Early 2000s – 2011)

• Initial experiments consisted of policy interventions, like default option adjustments and improved product warning labels.
• In 2009, Cass Sunstein was appointed head of the Office of Information and Regulatory Affairs (OIRA) in the U.S., to promote BI-based policies.
• The UK established the Behavioural Insights Team (BIT), to lead efforts in applying BI to public policy.

Development Phase (2012-2015)

• BI was applied more systematically through experiments in health policies, taxation, and consumer protection.
• Governments began testing behavioural interventions using Randomized Controlled Trials (RCTs).

Consolidation Phase (2016-Present)

• Behavioural insights became an integral part of policymaking.
• International organizations, such as the World Bank and the United Nations, incorporated BI into their policy frameworks.
• Many countries developed dedicated BI teams to support evidence-based decision-making.

Behavioural Science in the Policy Cycle

• It improves policymaking by influencing decision-making during these stages: Problem Identification, Behavioural Research, Designing Behavioural Interventions, and Policy Testing.
• Problem Identification involves understanding behavioural factors contributing to policy issues.
• Behavioural Research involves collecting data through experiments, surveys, and literature reviews.
• Designing Behavioural Interventions involves the creation of nudges or regulatory policies informed by behavioural science.
• Policy Testing includes conducting RCTs and other experiments to measure effectiveness before full implementation.
• Nudging uses BI by modifying choice architecture to encourage better decisions without restricting freedom, such as automatic pension enrolment, simplified tax letters, and health warnings.

Challenges and Future Perspectives of Behavioural Science in Policy

• Challenges include late integration in policy processes, over-reliance on nudging, and ethical concerns.
• The future of BI policymaking includes enhanced collaboration, use of new technologies and big data, and the development of ethical frameworks.

Ewert, B. (2019). Moving beyond the obsession with nudging individual behaviour Towards a broader understanding

• Behavioural Public Policy (BPP) is a key concept in policymaking, initially associated with nudging, but should go beyond nudging to consider broader, collective applications.
• This paper highlights the need to shift focus from individual nudging to a more pluralistic and multi-purpose understanding of BPP
• The idea is to use behavioural insights throughout the policy process and not as an isolated technique.
• Nudging, is based on modifying choice architecture to influence behaviour without restricting options and examples include default pension enrolment and simplified tax forms.
• Nudging faces criticism regarding complexity for social issues, reliance on RCTs, ethical manipulation, along with an ideological focus on individual responsibility instead of systemic problems.

Expanding Behavioural Public Policy Beyond Nudging

• BPP should evolve beyond nudging by considering the influence of policymakers.
• Traditional BPP focuses on citizens and consumers, but insights can improve decision-making and government efficiency.
• This includes debiasing recruitment processes with improved risk management in policymaking.
• The focus should be to target Organizational and Collective Behaviour by modifying individual choices and shaping organizational cultures.
• This approach applies to government agencies, businesses, and public institutions, for systemic change
• BPP should integrate with Traditional Policy Tools and complement regulations, financial incentives, and structural interventions is key
• Addressing childhood obesity requires both school meal design nudges and structural changes that improve food access.

The Future of Behavioural Public Policy

• BPP should be applied through the end to end policy process and be embedded in any policy development.
• Qualitative methods like ethnographic research should be combined with quantitative methods like RCTs, and interviews
• Ethical guidelines that promote transparency, informed consent, and public participation for behavioural policymaking should be put into place
• BPP is not a tool to shift responsibility to individuals but to be integrated into multi-level governance strategies.
• The approach to BPP requires it influence organisations, policymakers, and public institutions.
• Regulations, incentives, and public engagement should uses behavioural insights

Schlüter et al. (2017). A framework for mapping and comparing behavioural theories in models

• Social-ecological systems (SES) models are used to study interactions between humans and their environment influencing resource management and policy decisions.
• Better integration of behavioural theories is needed as decision making is influenced by social norms heuristics, learning, emotions, and risk perception.
• The MoHuB framework helps researchers compare behavioural theories, identify decision-making elements, and improve representations of human behaviour.

Challenges of Integrating Behavioural Theories into Models

• There is a diversity of theories each focues on decision making making
• Incomplete explanations where some theories emphasize individual choices, while others emphasize social interactions.
• Many models oversimplify human behaviour by using irrational economic assumptions

The MoHuB Framework

• The MoHuB integrates different behavioral decision theory through decomposing into elements
• Context includes the social and environmental setting
• State is the individual attributes needs, values

Key Behavioural Theories in SES Modelling

• There are six behavioural theories relevant to SES models

Rational Choice Theory (Homo Economicus)

• One key theory is to assume self interested utility with perfect knowledge
• Decision making is the key optimization step to picking the best option in Economic and environmental policy models.

Bounded Rationality

• This theory leverages decision making through satisficing and leans on heuristics
• Useful in agent based SES models

Theory of Planned Behavior (TPB)

• This includes how Decision-making is influenced by attitudes, social norms, and perceived control over actions
• In decision making the process includes forming intentions, which may lead to action.
• This is found to predict environmental behaviour (e.g., recycling, conservation efforts).

Descriptive Norm Theory

• People observe others' behaviour and tend to conform.
• Explains behaviour such as energy conservation or voting.

Prospect Theory

• People weigh potential losses more heavily than gains when making decisions.
• A biased evaluation drives this in climate adaptation and resource use.

Applications of the Framework in SES Modelling

• The MoHuB can compare and integrate theories in natural resource management.
• It is allows modelers to map existing theories onto decision-making processes, identify gaps, combine theories, and improve policy predictions.
• SES can better predict responses to new policies.
• Models include aspects like bounded rationality, social norms, learning, and risk perception that enable more relevant policy.

Uher, J. (2016) What is Behaviour And (when) is Language Behaviour A Metatheoretical Definition

• This paper address what qualifies as behavior across scientific disciplines, arguing that a approach is needed to distinguish it from physiological processes.
• To systematically define what behaviour is requires consideration

The Problem of Defining Behaviour

• Defines several challenges like a Lack of consensus as Different disciplines define behaviours differently.
• Different researchers rely on an intuitive understanding instead of building up off scientific definitions
• Sometimes defining what behaviour is gets confused with Physiological (e.g., sweating, digestion) along with psychological processes (e.g., thinking, emotions).
• Definitions should be open to non humans

Different Definitions of Behaviour (Across Disciplines)

• There is no single definition which makes it hard to differentiate behavior from physiological or psychological phenonmenia

Challenges in Defining Behaviour

• There are anthropocentic Biases such as overemphasized human traits that exclude non-human behaviours.

Key Characteristics of Behaviour

• Uher identifies defining properties to help what distringuish behavior from other processes.
• External Observability is key with the only exception being ultrasonic
• A key criteria is Functionalilty as Behavior a process that serves a function
• The meaning of the behavior happens in the Present moment and can only be understood via the specific context

The Metatheoretical Definition of Behaviour

• Behaviour consists of external changes or activities mediated by other stimuli in the present moment - Behaviour must be observable, with the intent of communicating or doing - Communication and interaction happens presently as it pertains to time

Implications for Research and Modelling

• Understanding what qualifies as behaviour impacts multiple scientific fields by improving the rigor
• As models of human and animal behaviour need to clearly specify what is being measured (e.g., observable actions vs. inferred mental states).