We investigate how the introduction of an Active Choice requirement influences subject proclivity to contribute to an impure public good in one time and repeated interactions. In a large-scale field experiment, we analyze more than 10000 contribution decisions to a carbon offsetting program in the context of online ticket sales for long-distance buses. We find that the simple requirement of an Active Choice – which circumvents the ethical issues posed by an opt-out design – not only increases participation rates by almost 50% in a first booking decision, but also boosts participation in subsequent bookings. At the same time, the introduction of Active Choice does not induce a substantial decline in returning customer rates. Our data support the theoretical assumption that anticipated guilt is a causal mechanism by which Active Choice induces higher contribution rates, as the opportunity for “choice avoidance” that is inherent to opt-in settings may help subjects circumvent feelings of guilt that would otherwise result from explicit free-riding.


voluntary carbon offsets, randomized field experiment, default setting, choice architecture