Many theories address how information technology (IT) affects the number of suppliers and supply chain governance. However, their predictions are at times contradictory and there is relatively little empirical evidence with which to evaluate them. We therefore develop an integrated, multiperiod model of the optimal number of suppliers that combines search and coordination theory, transaction cost economics, and incomplete contracts theory, and we assess our theoretical predictions using a large new data set on the global IT sourcing decisions of 1,355 firms in 12 countries. Our empirical results support three key predictions about trust, IT, and supply base size. First, investments in coordination IT, which reduce search and coordination costs, are correlated with using more suppliers, while use of vendor-specific IT is associated with fewer suppliers. Second, repeated relationships and trust play a major role in supply chain governance. As firms work with fewer suppliers, they also engage in more repeated relationships. At the same time asset specificity and the need to induce relationship-specific investments are correlated not only with fewer suppliers, but also with a larger fraction of repeated relationships. Third, supply chain governance differs in human capital-intensive and physical capital-intensive industries. The correspondence between asset specificity and repetition is strong in physical capital-intensive firms and not significant in human capital-intensive firms, while the correspondence between fewer suppliers and more repeated relationships is strong in human capital-intensive firms but not significant in physical capital-intensive firms. This corroborates the differential implications of human and physical capital for bargaining power, contractual risk, and trust in buyer–supplier relationships. The online appendix is available at https://doi.org/10.1287/mnsc.2016.2631 . This paper was accepted by Chris Forman, information systems.