Standard game theoretic models predict, based on subgame perfection, that public goods will not be provided even if agents are allowed to monitor free riders at some cost. Further, because punishment is not credible in these environments, this prediction is invariant to the size of groups. However, there is now substantial evidence that people are reciprocally motivated and will punish free riders, regardless of the material costs of doing so. To examine the implications of reciprocally-minded agents, we simulate an environment populated with the behavioral strategies often seen in the experimental lab and use the simulation to develop hypotheses that are more specific about why group size should matter when sanctions are allowed. We then test these hypotheses experimentally using the voluntary contribution mechanism. We examine whether the effect of group members or if information about other group members is what is important. We find large groups provide public goods at levels no less than small groups because punishment does not fall in large groups. However, hindrances to monitoring do reduce the provision of the public good.