Abstract
Investment in start-ups is highly uncertain. Therefore, a rational basis can be found for collective formation through syndicated investment, and the members of such syndication formations can be regarded as a ‘mutually supportive group’. However, a possible side effect is that it might create groups with similar traits, which may lead to a loss of diversity and the homogenisation of information within the group. In short, although group formation may positively impact the performance of start-ups investors, information homogenisation may have a negative impact. Therefore, this study sought to elucidate the actual situation of mutually supportive groups through syndicated investment in Japan and examine differences in the performance of investors depending on their position in a group. Using data on 4287 investors, we prepared a graphical representation of the investor network based on an adjacency matrix and identified cut-points and bi-components in the network. The relationship between cut-points and investment performance was then examined, and the investment performance across investor position types was compared using Kruskal–Wallis and Dunn–Bonferroni tests. The analysis confirmed the existence of one large bi-component in the network, with numerous smaller bi-components and isolated points around it. This suggests that the network through syndicated investment in Japan is structured such that similar information is easily propagated and the network is prone to information homogenisation. Moreover, it is suggested that to improve investment performance it is not sufficient to simply form a group, but that it is important to belong to multiple groups and play a role in the inter-group exchange.
