We examine the link between public spending in health and health outcomes by leveraging differential exposure to a health spending reform prompted by Brazil's 29th Constitutional Amendment, which mandated municipalities to spend at least 15% of their budget on health. We map dynamic effects on health care spending, inputs, access, outputs and outcomes. For municipalities initially spending below the 15% threshold, we find (a) large increases in health spending specifically, driven by administrative spending, infrastructure investment, and human resources; (b) a resulting greater supply of personnel, primary care coverage, and municipal hospitals; and (c) reductions in infant mortality rates, in particular for deaths during the neonatal period. While we find substantial cost increases and lower mortality elasticities compared with previous correlational parameters, benefits still exceed costs provided any VSL greater than US$764 thousand. Our results contribute to the literature by providing one of the first well-identified causal parameters of the relationship between public spending in health and health outcomes, by documenting the links in the chain connecting government health expenditure to health outcomes, and by considering spillovers across space and sectors.