The theoretical and experimental literature on crowding-out usually assumes that taxes and voluntary contributions are used to fund the “same” public good. Based on this assumption, theoretical models of both pure and impure altruism predict a result of crowding-out that is unable to explain an outcome of crowding-in. This experimental study examines the crowding-out effect when both publicly and privately provided public goods are present. It is found that when both publicly and privately provided public goods are equally valuable to subjects, or when the former is more valuable than the latter, the traditional results of crowding-out occur. However, crowding-in is present if the privately provided public good is relatively more valuable than the publicly provided public good.