Identifying network ties from panel data: theory and an application to tax competition

Abstract

Social interactions determine many economic behaviors, but information on social ties
\ndoes not exist in most publicly available and widely used datasets. We present results on the
\nidenti…cation of social networks from observational panel data that contains no information on
\nsocial ties between agents. In the context of a canonical social interactions model, we provide
\nsu¢cient conditions under which the social interactions matrix, endogenous and exogenous
\nsocial e¤ect parameters are all globally identi…ed. While this result is relevant across di¤erent
\nestimation strategies, we then describe how high-dimensional estimation techniques can be
\nused to estimate the interactions model based on the Adaptive Elastic Net GMM method. We
\nemploy the method to study tax competition across US states. We …nd the identi…ed social
\ninteractions matrix implies tax competition di¤ers markedly from the common assumption
\nof competition between geographically neighboring states, providing further insights for the
\nlong-standing debate on the relative roles of factor mobility and yardstick competition in
\ndriving tax setting behavior across states. Most broadly, our identi…cation and application
\nshow the analysis of social interactions can be extended to economic realms where no network
\ndata exists. JEL Codes: C31, D85, H71.