Why are COVID-19 mitigation strategies successful in reducing infections in some cases but not in others? Existing studies of individual policies tend to neglect the many interaction effects that arise when multiple policies are enacted simultaneously. Particularly, if a socially undesirable behavior has a close (and equally problematic) substitute, then a prohibition of that behavior will simply cause people to switch to the substitute, resulting in no effect on infections. However, joint policies that prohibit both the targeted behavior and the substitute will create a positive interaction effect, which closes the loophole. Respectively, behaviors that are complements (rather than substitutes) can be discouraged by prohibiting one behavior because this discourages complementary behaviors as well.
We provide a new argument for why COVID-19 policies can fail and why the evaluation of such policies may be incorrect: policies are effective only when they reduce infections as a general equilibrium, accounting not only for the targeted behavior but also for interaction effects.
We illustrate our arguments by applying insights from traditional and behavioral law and economics to several examples. Thereby, we highlight regulators’ challenge when facing interaction effects and factors such as social norms and time preferences.