Government purchases reloaded: Informational insufficiency and heterogeneity in fiscal VARs

Using a large Bayesian VAR, we approximate the flow of information received by economic agents to investigate the effects of changes to government purchases. We document robust evidence that informational insufficiency in conventional models explains inconsistent results across samples and commonly employed identifications in recursive Structural VARs and Expectational VARs. Furthermore, we report heterogeneous effects of components of government purchases. While aggregate government purchases do not appear to produce strong stimulative effects with output multiplier around 0.7, government investment components have multipliers well above unity. State and local consumption, which captures investment in education and health, elicits a strong response.