Tastes better than expected: Post-intervention effects of a vegetarian month in the student canteen

Interventions to reduce meat consumption are often implemented for short durations, raising questions about their long-term effects. We analyze how a one-month intervention that removed meat meals from a university canteen menu influenced consumer behavior after the intervention. Using transaction data from over 270,000 purchases by more than 4,500 guests, we analyze difference-in-difference effects in an intent-to-treat framework, classifying guests as treated or control based on the canteen they primarily frequent pre-intervention. Treated guests remained 4% less likely to choose a meat option in the two months following the intervention, while their likelihood of visiting the canteen remained unchanged. Guest consumption patterns and additional survey data provide empirical evidence that post-intervention effects are driven mainly by taste discovery, and less by habit formation or a change in social norms. This is a joint work with Charlotte Klatt.