Quantitative Economics
Journal Of The Econometric Society
Edited by: Stéphane Bonhomme • Print ISSN: 1759-7323 • Online ISSN: 1759-7331
Quantitative Economics: Jul, 2024, Volume 15, Issue 3
Testing Firm Conduct
https://doi.org/10.3982/QE2319
p. 571-606
Marco Duarte, Lorenzo Magnolfi, Mikkel Sølvsten, Christopher Sullivan
Evaluating policy in imperfectly competitive markets requires understanding firm behavior. While researchers test conduct via model selection and assessment, we present the advantages of Rivers and Vuong (2002) (RV) model selection under misspecification. However, degeneracy of RV invalidates inference. With a novel definition of weak instruments for testing, we connect degeneracy to instrument strength, derive weak instrument properties of RV, and provide a diagnostic for weak instruments by extending the framework of Stock and Yogo (2005) to model selection. We test vertical conduct (Villas‐Boas (2007)) using common instrument sets. Some are weak, providing no power. Strong instruments support manufacturers setting retail prices.
Full Content: Print
Supplemental Material
Supplement to "Testing Firm Conduct"
Marco Duarte, Lorenzo Magnolfi, Mikkel Sølvsten, and Christopher Sullivan
This supplement contains material not found within the manuscript.
Supplement to "Testing Firm Conduct"
Marco Duarte, Lorenzo Magnolfi, Mikkel Sølvsten, and Christopher Sullivan
The replication package for this paper is available at https://doi.org/10.5281/zenodo.11106460. The authors were granted an exemption to publish parts of their data because either access to these data is restricted or the authors do not have the right to republish them. However, the authors included in the package, on top of the codes and the parts of the data that are not subject to the exemption, a simulated or synthetic dataset that allows running the codes. The Journal checked the data and the codes for their ability to generate all tables and figures in the paper and approved online appendices. Whenever the available data allowed, the Journal also checked for their ability to reproduce the results. However, the synthetic/simulated data are not designed to produce the same results.
