Kirill Borusyak, Peter Hull, Evan Munro
arXiv 21 Jul 2026 · Econometrics
arXiv:2607.18601 · PDF · Extracted main text
We study the optimal design and analysis of experiments for estimating spillover effects. Assuming a known (e.g., linear) exposure mapping, we characterize the treatment-assignment distribution and regression-based estimator that minimize worst-case asymptotic variance against a broad class of distributions of unobservables. The design problem yields an intuitive solution in which the planner trades off spillover signal strength against diffusion of spillover variation. The analysis problem yields a simple recentered instrumental variable estimator to best leverage this variation. This framework produces natural solutions in several benchmark cases - such as clustered exposure - and suggests computationally tractable approximations for general networks, including bipartite settings. We illustrate these new tools in semi-synthetic experiments based on two applications from development economics. Our approach yields large standard error reductions in both experiments, increasing effective sample sizes by 50-100% or more.
appendix boundary found by appendix_command · 47% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Borusyak, Kirill and Hull, Peter (2023) Non-Random Exposure to Exogenous Shocks self | 0.928 | 5 | 3 | 80% |
| 2 | Miguel, Edward and Kremer, Michael (2004) Worms: Identifying impacts on education and health in the presence of treatment externalities | 0.924 | 19 | 4 | 79% |
| 3 | Cai, Jing and de Janvry, Alain and Sadoulet, Elisabeth (2015) Social Networks and the Decision to Insure | 0.914 | 17 | 4 | 76% |
| 4 | Borusyak, Kirill and Hull, Peter (2026) Optimal Formula Instruments self | 0.855 | 8 | 3 | 62% |
| 5 | Thiyageswaran, Vydhourie and Kokot, Alex and Brennan, Jennifer and M… (2026) Optimal Design under Interference, Homophily, and Robustness Trade-offs | 0.737 | 3 | 2 | 100% |
| 6 | Baird, Sarah and Bohren, J Aislinn and McIntosh, Craig and Özler, Berk (2018) Optimal Design of Experiments in the Presence of Interference | 0.644 | 2 | 2 | 100% |
| 7 | Borusyak, Kirill and Cáceres Bravo, Mauricio and Hull, Peter (2025) Estimating Demand with Recentered Instruments self | 0.644 | 2 | 2 | 100% |
| 8 | Viviano, Davide (2026) Experimental Design under Network Interference | 0.644 | 2 | 2 | 100% |
| 9 | Goemans, Michel X. and Williamson, David P (1995) Improved approximation algorithms for maximum cut and satisfiability problems using semidefinite programming | 0.585 | 3 | 1 | 100% |
| 10 | Harshaw, Christopher and Sävje, Fredrik and Eisenstat, David and Mir… (2023) Design and Analysis of Bipartite Experiments under a Linear Exposure-Response Model | 0.585 | 3 | 1 | 100% |
Showing the top 10 of 34 scored citations.