Price theory says that the most important effects of policy and technological change are often found beyond their first point of contact. This appears opposed to econometric methods that rule out spillovers of one person’s treatment on another’s outcomes. This paper uses the industry model from price theory to represent the statistical concepts of treatments and controls. When treated and control observations are in the same market, the controls are indirectly affected by the treatment. Moreover, even the effect of the treatment on the treated reveals only part of the consequence for the treated of treating the entire market, which is often the parameter of interest. Marshall’s Laws of Derived Demand provide a guide for empirical work: precise price-theoretic interpretations of the direct and spillover effects of a treatment, the quantitative relationships between them, and how they correspond to the scale and substitution effects emphasized in price theory.

More on this topic

BFI Working Paper·Oct 23, 2024

Dual Reductions and the First-Order Approach for Informationally Robust Mechanism Design

Benjamin Brooks and Songzi Du
Topics: Uncategorized
BFI Working Paper·Oct 1, 2024

Fear and Dreams: Understanding the Non-Institutional Sources of Leader Strategy

Maria Angélica Bautista, Juan Sebastián Galán, James Robinson, Rafael F. Torres, and Ragnar Torvik
Topics: Uncategorized
BFI Working Paper·Sep 24, 2024

On the Identifying Power of Generalized Monotonicity for Average Treatment Effects

Yuehao Bai, Shunzhuang Huang, Sarah Moon, Azeem Shaikh, and Edward J. Vytlacil
Topics: Uncategorized