Growing Stars: A Laboratory Analysis of Network Formation

The acquisition and dispersion of information, a critical aspect of economic decisions, can occur through a network of agents (Jackson, 2009). Empirical and theoretical findings suggest that an efficient information dispersion network takes the form of a star: small numbers of agents gather information and distribute it to a large group. Controlled tests of this theory, however, have typically found little evidence of star network emergence. An exception is Goeree et al (2009), which reports reliable star network formation in an environment that includes ex ante heterogeneous agents. While heterogeneity may explain network formation sometimes, it seems to play a smaller role in other cases (Feick and Price, 1987; Conley and Udry, 2010). In this paper we investigate whether specific institutional conditions promote star network formation with ex ante homogeneous agents. We find that investment limits and the “right-offirst-refusal,” both of which are institutions that stabilize decision making, have a surprising ability to promote star network formation. Further, using a cluster analysis that allows us to draw inferences about individuals’ behavioral rules, we find that these effective institutions encourage individual rationality as well as positive habits. We argue that these decision rules emerge due to the stabilizing features of the institutions and this stability facilitates improved network behaviors and outcomes. JEL classification: D85, D02, C92