How should entrepreneur's design a crowdfunding game to maximize the chances of successfully producing a public good? In a standard crowdfunding contract, potential contributors are refunded their contributions if an aggregate contribution threshold is not reached. In a refund-bonus contract, potential contributors are refunded their contributions plus a bonus if this threshold is not reached. Tabarrok (1998) and Zubrickas (2014) in theory and Cason, Tabarrok and Zubrickas (2021) in a laboratory setting show that refund bonuses can increase entrepreneur profits and the probability of successfully producing a public good. Earlier papers, however, did not endogenize the entrepreneur’s choice of contract. Refund bonuses are paid only when fundraising fails, thus entrepreneurs who choose to offer refund bonuses are signaling high-quality projects. Yet the more successful the signal, the greater the incentive of entrepreneurs with low-quality projects to offer refund bonuses. Hence, full separation is not possible.
We analyze the refund bonus signaling game in theory and test the predictions of the model in a laboratory experiment. Consistent with a hybrid (signaling) equilibrium, high-quality projects are successfully funded more frequently when entrepreneurs can choose whether to offer refund bonuses. Entrepreneurs choose to offer bonuses more often when project quality is high than low, and investors invest more frequently when bonuses are offered. Relative to equilibrium point predictions, however, entrepreneurs do not offer bonuses as frequently as they should, and investors invest more often than predicted. We conclude with a discussion of how refund bonuses can best be designed for crowdfunding in practice.