Early experimental studies showed negative results regarding the potential of Internet use to enhance social capital. Since then, a significant number of studies have reported positive effects. However, the findings are still inconclusive due to the diversity of Internet use, conceptual ambiguity of social capital, lack of cross-cultural studies, and methodological limitations regarding causal inference. In this paper, we use data from the Asian Barometer Survey and statistical causal inference to examine whether using social media to connect with other people increases social capital in four East Asian countries and territories. The effects of social media use on participation in voluntary associations and general trust were mixed at best, with no clear and consistent positive effects. In contrast, social media use fairly consistently increased network size in our data. The implications of these findings are discussed.