Nudge fudge or crystal clear? Investigating nudges and how they work

Recently there has been growing interest in the use of “nudges” as a policy tool and their potential to solve some of society’s biggest problems, such as obesity or climate change. Thaler and Sunstein in their influential book Nudge provide a broad but specific definition of “nudge”, with the essence of nudging being changing the “choice architecture” or environment of people to influence behavior in ways that promote human welfare and protect freedom of choice. Nudges have a focus on how to influence behavior through engaging the “automatic system” of the human mind, and in ways that people are not always consciously aware. A major aim of the thesis is to examine what nudges are and how they work. Our approach is to examine the “mechanisms of change” that underlie the process from nudge to changed behavior. We categorise nudges as working through “core” mechanisms of change if they directly enters the “automatic system”, harness social influence or appeal to our emotions. We categorise nudges that work through “peripheral” mechanisms of change as those that appeal more to the “reflective system” to influence behaviour change. We show how these mechanisms vary between nudges. In order to find out if nudges actually work and are cost effective, we selected a nudge for special investigation, namely prompted stair use interventions. We referred to systematic reviews and carried out our own small scale secondary evaluation of these interventions. We argued that it is not possible to make categorical statements as to whether this nudge works due to methodological weaknesses in these studies, particularly the lack of control group designs and lack of cost efficiency analyses. We argue the effective implementation of nudges may rely on traditional governance interventions, such as legal restrictions and information provision. Nudge has led to political initiatives in the form of the establishment of departments or committees in the UK, US and Norway to bring the “nudge” concept into policy making. We gathered empirical material, mainly through qualitative interviews, from members of a committee set up to advise and/or implement nudges in Lillehammer, Norway in order to compare to what extent we find similar issues related to what nudges are and how they work in this committee and in the academic literature. Similarities include awareness of nudge as a broad concept, an emphasis on nudges working at a ‘semi-conscious’ level as well as the importance of cost-effectiveness, generalizability and the value of control groups when evaluating nudges. A key difference was the limited reference to the ethical debate about nudges among committee members, in contrast to academic authors, perhaps because the former view nudges as small-scale interventions that aim to help people make better choices and do not raise serious ethical considerations. In conclusion we question whether “nudges” can in fact be defined as “libertarian” and “paternalistic”, we call for public debates about the use of nudges and involvement in overcoming some of the ethical dilemmas nudges raise (particularly for those that work mainly through the automatic system) and enhance legitimacy. Finally we argue for more robust evaluation studies to build evidence base for this relatively new intervention