Behavioural Science, Randomized Evaluations and the Transformation Of Public Policy: The Case of the UK Government


Chapter by Peter John: “Behaviour change policy conveys powerful image: groups of psychologists and scientists, maybe wearing white coats, messing with the minds of citizens, doing experiments on them without their consent, and seeking to manipulate their behaviours. Huddled in an office in the heart of Whitehall, or maybe working out of a windowless room in the White House, behavioural scientists are redesigning the messages and regulations that governments make, operating very far from the public’s view. The unsuspecting citizen becomes something akin to the subjects of science fiction novels, such as Huxley’s Brave New World or Zamyatin’s We. The emotional response to these developments is to cry out for a more humanistic form of public policy, a more participatory form of governance, and to base public policy on the common sense and good judgements of citizens and their elected representatives.

Of course, such an account is a massive stereotype, but something of this viewpoint has emerged as a backdrop to critical academic work on the use of behavioural science in government in what is described as the rise of the psychological state (Jones et al 2013a b), which might be seen to represent a step-change in use of psychological and other form of behavioural research to design public policies. Such a claim speaks more generally to the use of scientific ideas by government since the eighteenth century, which has been subject to a considerable amount of theoretical work in recent years, drawing on the work of Foucault, and which has developed into explorations of the theory and practice of governmentality (see Jones et al 2013:182-188).

With behaviour change, the ‘central concern has been to critically evaluate the broader ethical concerns of behavioural governance, which includes tracing geo-historical contingencies of knowledge mobilized in the legitimation of the behavior change agenda itself’ (190). This line of work presents a subtle set of arguments and claims that an empirical account, such as that presented in this chapter, cannot⎯nor should⎯challenge. Nonetheless, it is instructive to find out more about the phenomenon under study and to understand how the uses of behavioural ideas and randomized evaluations are limited and structured by the institutions and actors in the political process, which are following political and organizational ends. Of particular interest is the incremental and patchy nature of the diffusion of ideas, and how the use of behavioural sciences meshes with existing standard operating procedures and routines of bureaucracies. That said, behavioural sciences can make progress within the fragmented and decentralized policy process, and has the power to create innovations in public policies, often helped by articulate advocates of such measures.

The path of ideas in public policy is usually slow, one of gradual diffusion and small changes in operating assumptions, and this route is likely for the use of behavioural sciences. The implication of this line of argument is that agency as well as structure plays an important role in the adoption and diffusion of the ideas from the behavioural sciences. It implies a more limited and less uniform use of ideas and evidence than implied by the critical writers in this field, but one where public argument and debate play a central role….(More)”