Learning What to Learn: Experimental Design when Combining Experimental with Observational Evidence

TL;DR

This paper presents a framework for designing experiments that optimally combine experimental and observational evidence under budget constraints, accounting for potential biases and improving policy decision-making.

Contribution

It introduces a unified, minimax-regret-based approach for experimental design that explicitly considers observational bias without requiring bias bounds.

Findings

Framework effectively balances bias and variance in experimental design.

Application to cash-transfer experiments improves estimation of general equilibrium effects.

Optimized site selection enhances microfinance intervention outcomes.

Abstract

Experiments deliver credible treatment-effect estimates but, because they are costly, are often restricted to specific sites, small populations, or particular mechanisms. A common practice across several fields is therefore to combine experimental estimates with reduced-form or structural external (observational) evidence to answer broader policy questions such as those involving general equilibrium effects or external validity. We develop a unified framework for the design of experiments when combined with external evidence, i.e., choosing which experiment(s) to run and how to allocate sample size under arbitrary budget constraints. Because observational evidence may suffer bias unknown ex-ante, we evaluate designs using a minimax proportional-regret criterion that compares any candidate design to an oracle that knows the observational study bias and jointly chooses the design and estimator. This yields a transparent bias-variance trade-off that does not require the researcher to specify a bias bound and relies only on information already needed for conventional power calculations. We illustrate the framework by (i) designing cash-transfer experiments aimed at estimating general equilibrium effects and (ii) optimizing site selection for microfinance interventions.