Abstract: How to attract participation and elicit contributions in charity-based online crowd-funding has remained a challenging problem for a long time. Leveraging on corporate or individual funds to set up a matching fund mechanism has been widely adopted as a popular approach to motivate lenders’ donations. This study examines how the matching fund mechanism affects individual lenders’ contribution, theoretically and empirically. On the one hand, the classical tenet of public economics claims that each individual relies on peers to contribute more and enjoy the project. Hence, an outsider’s contribution via the matching fund mechanism should decrease individual lenders’ contribution due to the “free-riding” problem. On the other hand, the “strategic complements across time” effect in the framework of dynamic provision of public goods results in more contributions from individual lenders. Our theoretical model illustrates that which of the two opposing effects dominates hinges crucially on the intrinsic quality of a project. When the endogenous completion time of a project is short, it has high intrinsic quality and getting matched is more likely to crowd-in individual lenders’ contribution. Otherwise, for a project with low intrinsic quality, getting matched may crowd-out individual lenders’ contribution. Using real-time data from Kiva.org, a well-known charity-based crowd-funding platform, we find that the empirical results provide supportive evidence for the theoretical predictions.