Learning Normative Lessons From Behavioural Public Policy for Well-Being Public Policy

Behavioural Public Policy (BPP) was a milestone in the history of applying psychological science in governance. While psychology has always been influential—consider the chequered history of asylums, for example—BPP was the first time it was applied in a general way across the gamut of policy domains. 

Well-Being Public Policy (WPP) is emerging as a second such general application. Yet it has met with criticisms reminiscent of those directed at BPP. This déjà vu suggests that psychological science tends to trip on a set of ethical issues when moving from the laboratory to policy. To help, our new papers distils a normative guide for psychologists seeking to translate their wisdom into policy.

One reason for the success of BPP was that it was grounded in an ethical and political theory—libertarian paternalism—that provided a normative foundation for the transfer of behavioural insights from psychology to public policy. Libertarian paternalism has 4 pillars:

Effectiveness posits that BPPs need to be based on strong empirical evidence, both for the underlying psychological mechanism and the policy applications, and offer ‘bang for buck’ or cost-effectiveness.

Transparency insists that for BPPs to be ethical policymakers must be frank or even explicit about the psychological mechanisms they employ and the behavioural change they seek to induce.

Autonomy is where the ‘libertarianism’ really comes in. Libertarian paternalism insists that BPPs maintain the availability of all choices. For example, cigarettes are obscured rather than removed from shops.

The welfare pillar holds that in steering citizens towards ‘better’ choices, psychological science must let citizens themselves determine what ‘better’ means. This mitigates strong forms of paternalism and technocracy.

Libertarian paternalism is not uncontroversial. We discuss a long list of critiques in our paper. But it has undeniably succeeded in promoting the adoption of psychological science in public policy. A challenge for WPP is that libertarian paternalism does not map well onto many of the interventions it seeks to implement.

We define WPP in terms of interventions that intend to improve people’s mental states, such as happiness and life satisfaction, but also feelings of meaningfulness, social support, competence, engagement, and other themes that preponderate in positive psychology. We include in WPP the emerging notion of well-being budgeting, which is where government funds are allocated according to where they will produce the largest improvements in people’s mental states.

WPP has historically been quite strong on transparency, with WPP usually announced amid much fanfare. On autonomy, effectiveness, and especially welfare, WPP is more problematic.

Advocates of WPP argue that well-being metrics, especially those concerning subjective assessments, such as life satisfaction scale questions, put citizens in charge of policy priorities, ensuring autonomy. But while there is potential for this noble goal, WPP as currently practiced diverges from it. Once people’s subjective well-being (SWB) is elicited, variables chosen by experts are assessed in terms of their contribution to SWB. Citizens are not asked directly what improves their SWB as this might contaminate their responses with various cognitive biases. This is a sensible research practice, but it is not democratic, nor does it empower citizens to autonomously guide policy choices.

While the science of well-being is maturing rapidly, debates continue to rage about such fundamental questions as “what is well-being?” and whether common measures effectively assess the construct. Similarly, while experimental designs are common in the psychological literature, very few WPPs have been assessed in this high-quality way. And much of the sociological-scale research on well-being relies on correlations that explain little of the variance in SWB across individuals, groups, and geographies. WPP thus has a way to go before it can claim ‘effectiveness’.

‘Welfare’ is especially tricky. The prevailing economic approach to policy analysis defines well-being in terms of people’s ability to satisfy their preferences. This aligns neatly with liberal-democratic political ideals, as in libertarian paternalism, because it empowers citizens. Shifting the definition of well-being to mental states raises the possibility that governments will promote things like happiness in ways that run counter to citizens’ preferences. Some advocates of WPP endorse this view, arguing that citizens often have “irrational” or self-destructive preferences, and indeed there is evidence for this claim. But given the potential for technocracy here we must be cautious.

It is worth noting that economists and political theorists have for a long time raised objections to the use of mental state accounts of well-being in public policy. A critical issue is that someone in objectively bad circumstances can nonetheless have decent SWB. Life satisfaction researchers, for example, have noted the phenomenon of happy peasants and miserable millionaires. If budgeting aims to maximise SWB, then it might direct policy resources to the well-off—this seems a perverse outcome.

Our ability to adapt to bad circumstances raises additional quandaries for WPP. Notably, it may be more straightforward for government to help people adapt than to address the root cause of their distress. For example, it might be cheaper and easier to provide anxiety therapy than to address climate change or assure affordable housing. Government seems let off the hook from its obligations here. 

In the spirit of constructive criticism, our paper presents a framework for legitimising WPPs, namely ‘Boosts’. Boosting was developed as an alternative to ‘nudging’ in the context of BPP, but it can be used to legitimate a range of applications of psychological science in public policy. Boosting has three principles:

Capacity: boosts should be educative and sustainable over the long term.

Empowerment: recipients should develop an enhanced understanding of their psychology, the ways it manifests across different circumstances, and how to harness it for flourishing.

Participation: boosts are not only transparent but require the conscious cooperation of the citizens being boosted.

One example of a boost is the “men in sheds” program, which was designed to tackle feelings of isolation and incompetence among the elderly and unemployed.

The program builds capacity in the form of psychological skills including social coping mechanisms. It is empowering because these skills are transferable across contexts, such as from the shed to the household. And it is inherently participatory as it involves getting people together to voluntarily build items useful to the community, like repairing scoreboards for the local football field. 

Boosting is stricter than libertarian paternalism and admits fewer policies, but this ensures that boosts have a strong ethical foundation. WPPs can certainly be justified by other means, but thinking in terms of boosts increases the likelihood that early WPPs will uncontroversial, building momentum for more ambitious ventures

This article was originally published here on the blog of the Bennett Institute for Public Policy at the University of Cambridge. It is based on a longer academic article: 

Fabian, M. and Pykett, J. (2021). Be Happy: Navigating Normative Issues in Behavioural and Well-Being Public Policy. Online First in Perspectives on Psychological Science.