All communities have common resources that are vulnerable to selfish motives. The current paper explores this challenge in the specific case of the urban commons, defined as the public spaces and scenery of city neighborhoods. A theoretical model differentiates between individual incentives and social incentives for caring for the commons. The quality of a commons is defined as the level of physical (e.g., loose garbage) and social (e.g., public disturbances) disorder. A first study compared levels of disorder across the census (...) block groups of a single city; the second compared the disorder generated by individual addresses in two neighborhoods. Each study found that homeownership, an individual incentive, was the main predictor of disorder. Owner-occupied parcels generated less disorder than their renter-occupied neighbors, but both parcel types produced less disorder in a neighborhood with greater homeownership. The results emphasize the need for considering both individual and social incentives for group-beneficial behaviors. (shrink)