Are Default Nudges Deemed Fairer When They Are More Transparent? People's Judgments Depend on the Circumstances of the Evaluation

Abstract

To avoid concerns of manipulation, nudges should be transparent to the people affected by the intervention. Whether increasing the transparency of a nudge also leads to more favorable perceptions of the nudge is however not certain, and may depend on the circumstances of the evaluation. Across three preregistered experiments (N = 1915), we study how increased transparency affects the perceived fairness of a default nudge, in joint vs. separate, and description- vs. experience-based evaluations. We find that transparency increases perceived fairness of the nudge in a joint comparison, when the relative benefits of transparency are easy to see. However, in a real choice-context, with nothing to compare against, transparency instead decreases perceived fairness. Efforts to make nudges more ethical may thus ironically make choice architects perceived as less ethical. Additionally, we find that the transparent default nudge still successfully affects behavior, that different default-settings communicate different perceived intentions of the choice architect, and that participants consistently favor opt-in defaults over opt-out defaults nudges – regardless of their level of transparency.

Nudging for impact: beyond the immediate effectiveness of nudge interventions promoting healthy food intake

“…Although nudges' effectiveness was originally thought to depend upon the nudgee being oblivious to their use (bovens, 2009), our findings show that nudges can still affect behaviour even when implemented by the nudgee themselves. These results are in line with recent findings that show that openness about the goal and mechanisms of nudging does not negate its effectiveness (bruns et al, 2018;cheung et al, 2019;Kroese et al, 2016;Loewenstein et al, 2015;Michaelsen et al, 2020;Michaelsen et al, 2021a;Paunov et al, 2019a;Paunov et al, 2019b;steffel et al, 2016).…”

“…Initially, it was thought that nudge transparency would be detrimental to their effectiveness (Hansen & Jespersen, 2013;Smith et al, 2013). Or, in the often cited words of Bovens (2009), nudges would work 2 Although our definition of nudge transparency is similar to the dominant view of nudge transparency in the literature (Michaelsen et al, 2020), we do like to point out that, for some, nudge transparency refers to those instances in which the intention and means through which behaviour change is pursued can reasonably be known to the one being nudged (Hansen & Jespersen, 2013).…”

“…It was even thought that resistance or reactance would result in counter effects (Brehm, 1989). Fortunately, empirical evidence is now accumulating that shows that transparency does not, in fact, hinder nudge effectiveness (Bang et al, 2020;Bruns et al, 2018;Cheung et al, 2019;Kroese et al, 2016;Loewenstein et al, 2015;Michaelsen et al, 2020;Michaelsen et al, 2021a;Steffel et al, 2016). Transparent nudges have even been found to be more effective than non-transparent nudges (e.g.…”

“…For a long time, however, it was thought that nudging 'works best in the dark' and that transparency would even hinder nudge effectiveness through compensatory reactance (bovens, 2009). However, empirical evidence suggests that this concern is not valid as transparency manipulations do not hamper, or even increase, the effectiveness of nudges (bruns et al, 2018;cheung et al, 2019;Kroese et al, 2016;Loewenstein et al, 2015;Michaelsen et al, 2020;Michaelsen et al, 2021a;Paunov et al, 2019a;Paunov et al, 2019b;Steffel et al, 2016). As making nudges transparent does not negatively impact effectiveness, and mitigates some of the ethical concerns surrounding the use of nudging (but see Wachner et al, 2020 for a study showing that ethical concerns related to nudging may not be supported by the evidence), it would seem that there are no reasons to avoid their use.…”

“…being unable to recognize the influence of the nudge used to be thought of as a key ingredient for nudges' effectiveness (bovens, 2009). However, more recent studies examining the effect of explaining the aim of the nudge show that such nudge transparency does not compromise, and can even enhance, its effectiveness (bruns et al, 2018;cheung et al, 2019;Kroese et al, 2016;Loewenstein et al, 2015;Michaelsen et al, 2020;Michaelsen et al, 2021a;Paunov et al, 2019a;Paunov et al, 2019b;steffel et al, 2016). Awareness of the nudge when self-nudging may thus not compromise its effectiveness.…”