The regulation of repugnant goods and the limits of the Alchian–Allen effect

We argue in this article that constant-quality models like the one used to derive the Alchian–Allen effect are of very limited use in the analysis of the regulation of repugnant goods (morally contested goods ranging from human organs to sex work, pornography, and surrogate motherhood). The Alchian–Allen effect has often been applied in the literature on the regulation of repugnant goods, for instance to suggest that a common unit cost imposed on alcoholic drinks will lead the consumer to substitute less potent alcoholic drinks for more potent alcoholic drinks. Various authors have suggested that this effect is so broadly relevant that we should speak of an iron law of prohibition: "the harder the enforcement, the harder the drugs." We demonstrate that the Alchian–Allen effect has a narrow domain of validity, does not always suggest that regulation has adverse effects, and its predicted effects depend strongly on the goods selected for analysis. Moreover, real-world regulations do not impose a common unit cost. Therefore, the Alchian–Allen effect is not helpful to analyze the regulation of repugnant goods. We further highlight the relevance of Barzel's work on the incompleteness of regulation, and the importance of quality adjustments in response to regulation.