Philip G. Wright, directed acyclic graphs, and instrumental variables

TL;DR

This paper reviews Philip Wright's pioneering work on causal inference, including structural models, identification, and empirical methods, highlighting its relevance and influence on modern econometrics and causal analysis.

Contribution

It reinterprets Wright's foundational ideas within a modern framework, emphasizing their significance and connection to current causal inference methods.

Findings

Wright introduced a structural equation model for supply and demand.

He established identification of elasticities using directed acyclic graphs.

Developed empirical methods for demand estimation with weather instruments.

Abstract

Wright (1928) deals with demand and supply of oils and butter. In Appendix B of this book, Philip Wright made several fundamental contributions to causal inference. He introduced a structural equation model of supply and demand, established the identification of supply and demand elasticities via the method of moments and directed acyclical graphs, developed empirical methods for estimating demand elasticities using weather conditions as instruments, and proposed methods for counterfactual analysis of the welfare effect of imposing tariffs and taxes. Moreover, he took all of these methods to data. These ideas were far ahead, and much more profound than, any contemporary theoretical and empirical developments on causal inference in statistics or econometrics. This editorial aims to present P. Wright's work in a more modern framework, in a lecture note format that can be useful for teaching and linking to contemporary research.