This paper reports the results of a field experiment investigating how attributes of carbon footprint information affect consumer choice in a large dining facility. Our hypotheses and research methods were pre-registered via the Journal of Accounting Research’s registration-based editorial process. Manipulating the measurement units and visualizations of carbon footprint infor-mation on food labels, we quantify effects on consumers’ food choices. Treated consumers choose less carbon-intensive dishes, reducing their food-related carbon footprint by up to 9.2 %, depending on the treatment. Effects are strongest for carbon footprint information expressed in monetary units (‘environmental costs’) and color-coded in the familiar traffic-light scheme. A post-experimental survey shows that these effects obtain although respondents, on average, self-report low concern for the environmental footprint of their meal choices. Our study contributes to the accounting literature by using an information processing framework to shed light on the information usage and decision-making processes of an increasingly important user group of accounting information: consumers.