Deviant or Wrong? The Effects of Norm Information on the Efficacy of Punishment

Abstract

A stream of research examining the effect of punishment on conformity indicates that punishment can backfire and lead to suboptimal social outcomes. We examine whether this effect is due to a lack of perceived legitimacy of rule enforcement, enabling agents to justify selfish behavior. We address the question of punishment legitimacy by shedding light upon the importance of social norms and their interplay with punishment. Often people are presented with incomplete norm information: either about what most others do (empirical) or what most others deem appropriate (normative). We show that neither punishment nor empirical/normative information in isolation result in prosocial behavior. In turn, we find that prosociality is significantly increased when normative information and punishment are combined, but only when compliance is relatively cheap. When compliance is more expensive, we find that the combination of punishment and empirical information about others' conformity can have detrimental effects on prosocial behavior. We attribute this outcome to the differential ability to distort one's own beliefs about applicable norms. Our results have important implications for researchers and practitioners alike.

Requiem for a Nudge: Framing effects in nudging honesty

“…Specifically, recent work -particularly in the field -has revealed that reminding people about descriptive and injunctive norms can lead to a positive behavioral response, especially when implemented with care and regard for the cultural and institutional circumstances (e.g., Hallsworth et al, 2017;Hernandez et al, 2017;Bott et al, 2019, and see also discussion in Bicchieri and Dimant, 2019). Importantly, the literature has also produced conflicting results in that not all interventions are equally successful, as some interventions have been observed to yield no effect or even a backfiring effect (e.g., Blumenthal et al, 2001;Fellner et al, 2013;Bicchieri et al, 2019c).…”

“…While we find that simple norm-nudges can be unsuccessful at shifting norms in environments in which a behavioral norm is already in place (cheating is normatively not acceptable, as our results from the Krupka and Weber (2013) et al, 2019b), which in turn may undermine the effectiveness of norm-nudges. The extent to which the effectiveness of norm-interventions is mediated by the common perception of an existing norm is subject to scientific debate (see, e.g., Bicchieri et al, 2019c) and requires further research. In addition, while our results can speak to the effectiveness of such interventions when the risk of detection for deviant behavior is absent by design (in our case implemented through the mind game as per Jiang, 2013), the results may or may not be comparable when risk of detection is present (as is often the case in the standard dice paradigm of Fischbacher and Föllmi-Heusi, 2013).…”

“…There, personalized messages -and sometimes physical letters -are being sent with the intent to achieve behavior change in the forms of increasing tax compliance, charitable giving, primary school enrollment, and student learning, or in the form of decreasing energy consumption, student absenteeism from school, credit card debt, and even the spread of HIV infections, among other examples (Reinikka and Svensson, 2005;Dupas, 2011;Luttmer and Singhal, 2014;Boyer et al, 2016;Brandon et al, 2017;Rogers and Feller, 2018;Bursztyn et al, 2019). In existing literature, such information-only interventions indicate mixed results, with some studies demonstrating success (Hallsworth et al, 2017;Hernandez et al, 2017;Bott et al, 2019) while others either fail to detect a significant effect (Blumenthal et al, 2001;Fellner et al, 2013;Castro and Scartascini, 2015;Kettle et al, 2017;Cranor et al, 2018;Dunning et al, 2019) or indicate that interventions may backfire (John and Blume, 2018;Bicchieri et al, 2019c).…”

“…To ensure the truthfulness of all norm-nudges, the information was based on the behavior and beliefs of other uninvolved participants from a previous study run at the University of Pennsylvania. Such methodology is commonly adopted in experimental research on social norms (e.g.,Goldstein et al, 2008;Bicchieri and Xiao, 2009;Krupka and Weber, 2013;Bicchieri et al, 2019c). The truthful results were that the majority of the participants did not lie and that the majority did not approve of lying.6 Particular emphasis was put on whether participants were able to recall correctly what they were told previous minorities / majorities said or did.…”

Requiem for a Nudge: Framing Effects in Nudging Honesty

“…There, personalized messages -and sometimes physical letters -are being sent with the intent to achieve behavior change in the forms of increasing tax compliance, charitable giving, primary school enrollment, and student learning, or in the form of decreasing energy consumption, student absenteeism from school, credit card debt, and even the spread of HIV infections, among other examples (Reinikka and Svensson, 2005;Dupas, 2011;Luttmer and Singhal, 2014;Boyer et al, 2016;Brandon et al, 2017;Rogers and Feller, 2018;Bursztyn et al, 2019). In existing literature, such information-only interventions indicate mixed results, with some studies demonstrating success (Hallsworth et al, 2017;Hernandez et al, 2017;Bott et al, 2019) while others either fail to detect a significant effect (Blumenthal et al, 2001;Fellner et al, 2013;Castro and Scartascini, 2015;Kettle et al, 2017;Cranor et al, 2018;Dunning et al, 2019) or indicate that interventions may backfire (John and Blume, 2018;Bicchieri et al, 2019c).…”