We develop a dynamic search and matching model to estimate the impact of venture capital (VC) contract terms on start-up outcomes, and the split of value between entrepreneur and investor, in the presence of endogenous selection. Using a new, large data set of rst nancing rounds of start-up companies, we find an internally optimal equity split between VC and entrepreneur that maximizes the probability of success, consistent with standard double moral hazard theories. However, in virtually all deals, VCs use their bargaining power to receive more equity than is value-maximizing for the start-up. In most cases, participation rights and investor board representation reduce company value, while shifting more value to the investors. Pay-to-play has the opposite effect. Conditioning on the entrepreneur’s quality, high quality investors receive more investor friendly terms in equilibrium. But due to the positive impact of VC quality, a better VC still benefits the start-up and the entrepreneur, though not as much as theoretically possible. We conclude with counterfactual exercises that eliminate certain terms,which benefits entrepreneurs but can decrease the number of deals in the market.