Corrupting cooperation and how anti-corruption strategies may backfire

Understanding how humans sustain cooperation in large, anonymous societies remains a central question of both theoretical and practical importance. In the laboratory, experimental behavioural research using tools like public goods games suggests that cooperation can be sustained by institutional punishment—analogous to governments, police forces and other institutions that sanction free-riders on behalf of individuals in large societies1,2,3. In the real world, however, corruption can undermine the effectiveness of these institutions4,5,6,7,8. Levels of corruption correlate with institutional, economic and cultural factors, but the causal directions of these relationships are difficult to determine5,6,8–10. Here, we experimentally model corruption by introducing the possibility of bribery. We investigate the effect of structural factors (a leader’s punitive power and economic potential), anti-corruption strategies (transparency and leader investment in the public good) and cultural background. The results reveal that (1) corruption possibilities cause a large (25%) decrease in public good provisioning, (2) empowering leaders decreases cooperative contributions (in direct opposition to typical institutional punishment results), (3) growing up in a more corrupt society predicts more acceptance of bribes and (4) anti-corruption strategies are effective under some conditions, but can further decrease public good provisioning when leaders are weak and the economic potential is poor. These results suggest that a more nuanced approach to corruption is needed and that proposed panaceas, such as transparency, may actually be harmful in some contexts.