Institutions are an important means for fostering prosocial behaviors, but in many contexts their scope is limited and they govern only a subset of all socially desirable acts. We use a laboratory experiment to study how the presence and nature of an institution that enforces prosocial behavior in one domain affects behavior in another domain and whether it also alters prosocial preferences and beliefs about others' behavior. Groups play two identical public good games. We vary whether, for only one game, there is an institution enforcing cooperation and vary also whether the institution is imposed exogenously or arises endogenously through voting. Our results show that the presence of an institution in one game generally enhances cooperation in the other game thus documenting a positive spillover effect. These spillover effects are economically substantial amounting up to 30 to 40 percent of the direct effect of institutions. When the institution is determined endogenously spillover effects get stronger over time, whereas they do not show a trend when it is imposed exogenously. Additional treatments indicate that the main driver of this result is not the endogeneity but the temporal trend of the implemented institution. We also find that institutions of either type enhance prosocial preferences and beliefs about others' prosocial behavior, even toward strangers, suggesting that both factors are drivers of the observed spillover effects.