Please keep ordering! A natural field experiment assessing a carbon label introduction

We conduct a natural field experiment investigating the efficacy of environmental information provision while
considering its relationships with individual consumers’ habitual behaviour. A carbon label was introduced to
the lowest-emission dish for each food category on the menu of a full-service restaurant; its efficacy was assessed
by distinguishing its immediate impact on orders placed by the restaurant’s occasional and regular customers as
well as the impact over time for repeated orders. We collected 1,737 customer orders – of which 1,200 were
placed by 99 regular customers – taking advantage of an electronic ordering system that identified customers
through a unique number. Independently of customer type, we find no immediate effect of the carbon label after
its introduction. However, the probability of ordering an environmentally friendly dish increases significantly
with repeated exposure and additional orders, albeit with a progressively diminishing effect. We discuss the
importance of the repetition effect when assessing a new label, including implications for research and policy.