This paper presents a methodology to draw causal inference in a non-experimental setting subject to network interference. Specifically, we develop a generalized propensity score-based estimator that allows us to estimate both direct and spillover effects of a continuous treatment, which spreads through a network with weighted and directed edges. To showcase this methodology, we investigate whether and how spillover effects shape the optimal level of producers’ support in agricultural markets. Our results show that, in this context, neglecting interference may lead to a downward bias when assessing policy effectiveness.
Download paper (PDF)