Spotlight on researcher decisions: Infrastructure evaluation, instrumental variables, and specification screening

This paper revisits the instrumental variable (IV) approach in Lipscomb et al. (2013, 2021, LMB) to study the impacts of electrification. We first make corrections to the construction of the dataset, including the modelled IV. Revised estimates on main outcomes and mechanisms are statistically insignificant, with substantially lower effect sizes. We second develop a framework that accounts for weak IVs and discourages specification screening. Applying it to LMB, we find that most theoretically justified specifications yield insignificant results. The proposed framework is transferable to other IV applications to reduce potential bias stemming from researcher's or replicator's discretion.