Identification and Estimation of Casual Mechanisms and Net Effects of a Treatment

An important goal in the analysis of the causal effect of a treatment on an outcome is to understand the mechanisms through which the treatment causally works. In the economics literature, however, there seems to be no available framework to estimate the relative importance of different causal mechanisms of a treatment. We fill this void by precisely defining a causal mechanism effect of a treatment and the causal effect of a treatment net of that mechanism using the potential outcomes framework, and by considering their identification and estimation. The definition of our parameters results in an intuitive decomposition of the total effect of a treatment that is useful for policy purposes. We offer conditions under which these causal effects can be estimated for the case of a randomly assigned treatment and when selection into the treatment is random conditional on a set of covariates. We close with two empirical applications that illustrate the concepts and methods introduced in this paper.