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.