Antitrust & Competition Policy Blog

Editor: D. Daniel Sokol
University of Florida
Levin College of Law

Monday, August 3, 2020

Endogeneity Problems in Empirical Analyses of Antitrust and Competition

Endogeneity Problems in Empirical Analyses of Antitrust and Competition

Martino De Stefano

Charles River Associates

Ilton Soares

Charles River Associates (CRA)

Tanguy Brachet

Charles River Associates (CRA)

Date Written: May 9, 2020


In empirical analysis, endogeneity is at the root of the refrain that “correlation is not causation” in that its presence can undermine the reliable estimation of true causal effects. Because valid antitrust examination must be based on true causal relationships, a basic understanding of endogeneity is of great importance for antitrust practitioners. In most settings, the observed associations between market outcomes of interest and their hypothesized determinants are contaminated by endogeneity in that they reflect two kinds of relationships: non-causal correlation (i.e., the variables move together without a causal relationship between them) and true causation (i.e., changes in one variable cause changes in another, all else being equal). While discussions of endogeneity can quickly get technical, the purpose of this chapter is to make it approachable for non-economists and practitioners who do not have a background in econometrics by introducing the concept, its consequences, and possible solutions through a series of examples relevant to antitrust in non-technical and intuitive terms.

| Permalink


Post a comment