Back to History

Fields Changed

Registration

Field Before After
Trial Status in_development completed
Abstract The overall aim of the experiment is to test the impact of self-and social image concerns on the likelihood that firms participate in a voluntary carbon offsetting program as an example for a public good. Though public good provision of firms can be observed empirically (e.g. CSR measures), little is known about the different motivations and the mechanics behind it. We contribute to the literature by implementing an experimental setting that provides the sender of a delivery the opportunity to offset the carbon emissions of this service by paying a price premium, while we vary the opportunity to signal this pro-environmental behavior to the recipient of the delivery. As an extension we consider to introduce price variations in the second phase of the experiment conditional on the acceptance of our field partner. The overall aim of the experiment is to test the impact of self-and social image concerns on the likelihood that firms participate in a voluntary carbon offsetting program as an example for a public good. Though public good provision of firms can be observed empirically (e.g. CSR measures), little is known about the different motivations and the mechanics behind it. We contribute to the literature by implementing an experimental setting that provides the sender of a delivery the opportunity to offset the carbon emissions of this service by paying a price premium, while we vary the opportunity to signal this pro-environmental behavior to the recipient of the delivery. In a second phase, we extend the field experimental setting by introducing a 1:1 matching scheme to learn more about the firms’ evaluation of the signal. The matching doubles the amount of emissions offset while the price for the participant stays the same, if she decides to offset with a sticker attached to her delivery.
Last Published December 19, 2018 05:15 AM February 28, 2019 05:01 AM
Intervention (Public) From the beginning of the intervention onwards customers ordering deliveries via the webshop of a courier service provider are randomly assigned to one of the three different treatments in which they remain during the intervention. Observations of the treatments are therefore made in parallel. The intervention is planned to last approximately 3 months. Based on previous orders we expect about 300 orders per month. So that a total of 900 observations (with equal shares of about 300 for each treatment) will be envisaged. From the beginning of the intervention onwards customers ordering deliveries via the webshop of a courier service provider are randomly assigned to one of the three different treatments in which they remain during the intervention. Observations of the treatments are therefore made in parallel. The intervention is planned to last at least 3 months. Based on previous orders we expect about 300 orders per month. So that a total of 900 observations (with equal shares of about 300 for each treatment) will be envisaged. The three treatments vary the option to signal a carbon compensation decision (no signal, mandatory or voluntary signal). In a second ("matching") phase we offer to double the amount of emissions that are offset. This 1:1 matching is contingent upon signalling.
Primary Outcomes (End Points) Proportion of orders with carbon offseting. Proportion of orders with carbon offseting in total for the three treatments and share of signaled offsetting decisions among the 'voluntary signal' treatment.
Experimental Design (Public) In our experimental setting, we aim at testing the impact of self- and social image concerns on firm behavior. Decision behavior will be examined by exposing individuals to different instruments in a randomized control trial. The focus of the experiment are business clients. We implement treatments in a carbon offsetting context that differ in the opportunity to signal the pro-environmental behavior. In the baseline scenario (T1) participants can decide whether they want to offset the carbon emissions resulting from their delivery but without any possibility to directly signal pro-environmental behavior to the recipient of the delivery. In a second treatment (T2) we introduce a sticker which is automatically attached to the address label of the parcel to signal pro-environmental behavior. In the third treatment (F3), the sticker option is optional and has to be actively chosen by the client. In our experimental setting, we aim at testing the impact of self- and social image concerns on firm behavior. Decision behavior will be examined by exposing individuals to different instruments in a randomized control trial. The focus of the experiment are business clients. We implement treatments in a carbon offsetting context that differ in the opportunity to signal the pro-environmental behavior. In the baseline scenario (T1) participants can decide whether they want to offset the carbon emissions resulting from their delivery but without any possibility to directly signal pro-environmental behavior to the recipient of the delivery. In a second treatment (T2) we introduce a sticker which is automatically attached to the address label of the parcel to signal pro-environmental behavior. In the third treatment (F3), the sticker option is optional and has to be actively chosen by the client. In a second phase we add a 1:1 matching scheme. This adresses those participants that initially decided against the offsetting or against the signal. It offers them to double the amount of emissions offset in case they decide to purchase the carbon compensation and signal it.
Randomization Method Clients are assigned to the treatments by a random mechanism based on their customer IDs. Clients are assigned to the three treatments by a computer based random number generator mechanism based on their unique customer IDs.
Planned Number of Clusters We observe orders in the webshop of a courier service provider over a duration of 3 month. We observe orders in the webshop of a courier service provider over a duration of 3 month. [An extension of this minimum duration allowed us to conduct the field experiment over a total of 9 month until the end of December 2019.]
Planned Number of Observations About 900 orders in the webshop. About 900 orders in the webshop. [We exceeded this minimum requirement by extending the total duration of data collection.]
Sample size (or number of clusters) by treatment arms approximately 300 orders per treatment. approximately 300 orders per treatment. [By extending the total duration we were able to extend the samplesize.]
Additional Keyword(s) Public good provision, corporate behavior, social norm Public good provision, corporate social behavior, social norms, information disclosure
Back to top