Online Social Stock Picking: An Empirical Examination

Abstract

This study examines the online market for social stock investing where individuals follow other investors in the community for their investment decisions. Using data from one of the largest social stock-picking marketplaces, this study explores whether and how various aspects of an individual’s social networks affect his or her investment performance. Using social network methodologies and theory of social learning, this study explores the outcomes of such networks and herding behavior on investors’ performance. In particular, this study examines how different types of social learning—direct and indirect— impact investment performance. The results show that the network structure of a user’s network (indirect social learning) has an impact on performance; in particular, following people with structural holes contributes to improved investment performance. This study also analyzes two different modes of herding (direct social learning), and finds that while information-based herding leads to improved performance, reputation-based herding negatively affects performance. This study extends the theoretical implications regarding the role of different types of social learning in the investment realm. Platform providers should thus consider the various types of information provided to investors in the marketplace to facilitate social learning behavior that improves investment outcomes. Investors should also distinguish the type of social learning that improves their own investment outcomes.