Aggregating Distributional Treatment Effects: A Bayesian Hierarchical Analysis

Researchers who study policy interventions often estimate sets of treatment effects to characterize the impact across the entire distribution of household outcomes. Yet there is currently no methodology to formally aggregate the evidence on sets of effects across setting when the generalizability of the results is not known.

Rachel Meager developed methods to aggregate evidence on distributional treatment effects from multiple studies conducted in different settings, and applied them to the microcredit literature. These methods, developed using a Bayesian hierarchical framework, were used to generate new insights about the impacts of expanding access to microcredit. Considering results from seven randomized experiments, the study examined six household outcomes: business profit, business revenues, business expenditures, consumption, consumer durables spending, and temptation goods spending. The study concluded that there is strong evidence that microcredit typically does not lead to worse outcomes at the group level, but there is no generalizable evidence on whether it improves group outcomes.