Can warning labels communicating the environmental impact of meat reduce meat consumption? Evidence from two multiple treatment reversal experiments in college dining halls

Meat consumption has an adverse impact on both human and planetary health. To date, very few studies have examined the effectiveness of interventions tackling the overconsumption of meat in field settings. The present research addresses this gap by examining the impact of gain-framed labelling interventions communicating the adverse environmental consequences of meat consumption, using a multiple treatment reversal design across two university college dining halls over a period of five weeks. In College A the intervention weeks consisted of text-only or text-and-image labels communicating the adverse environmental consequences of meat consumption, and in College B patrons were exposed to either environmental or health labels (gain-framed; combining images and text). In total 13,869 (6,577 in College A and 7,292 in College B) meals (dishes) were analysed over the period of interest. Beta-binomial regressions found no statistically significant impact of the intervention periods compared to baseline on meat consumption in both College A and College B. The number of meal type options emerged as the only consistent predictor of meat consumption across models and across both colleges: meat consumption decreased with an increase in non-meat meal options. A post-study survey (College A: n = 88; College B: n = 53) revealed that patrons in both dining halls perceived environmental labels bearing both text and images as more informative and influential at changing behaviour compared to the other labelling interventions, although this did not translate into a change in behaviour. We discuss the implications of these findings for research, policy, and practice.