Experimental Design
We conduct a survey experiment among approximately 7500 participants representative of the German population. Prior to our experiment, we collect information about participants’ household size, homeowner status, the size of their dwelling, energy consumption for heating purposes, their household’s vehicle fleet and annual vehicle usage. We briefly inform participants about carbon pricing in the building and transport sectors in Germany and ask them to rate how closely they follow information about carbon pricing on a five-point Likert scale.
We randomly assign participants to one of four separate experiments, each providing personalized information about their household’s carbon pricing costs in a specific sector at a given carbon price. Specifically, we implement one experiment for the building sector and one for the transport sector at a price of 100 €/t of CO2, and likewise one experiment for each sector at a price of 200 €/t of CO2. Within each experiment, respondents are randomly assigned to a treatment group that receives personalized cost information or to a control group that does not receive any additional information. Consequently, we have four segments, each including a treatment and a control group of roughly 940 participants.
The experiments begin by asking participants to estimate their household’s additional annual costs of carbon pricing at a given price level and to express their uncertainty regarding their estimation on a five-point Likert scale. 50% of participants are then randomly assigned to receive personalized quantitative information on their household’s actual costs from carbon pricing. Additional heating costs are calculated using the previously collected information on household's annual energy consumption. Additional transportation expenses are inferred from the household's distance traveled and the average fuel consumption of their vehicle fleet. The remaining 50% of participants act as a pure control group and receive no further information. The experiment concludes by eliciting participants’ willingness to undertake carbon cost-reducing measures, their demand for additional information about reducing carbon costs from the consumer advice center, their acceptance of carbon pricing, their trust in the provided information, and their posterior cost estimate. Respondents who indicate no intention to act are further asked to provide an explanation for their decision in an open-ended question.