Walras (1897) defined a natural monopoly as an industry where monopoly is the efficient market structure and suggested, following Smith (1776), to price the prod- uct of the firm by balancing its budget. This led to the Ramsey (1927) and Boiteux (1956) theory of optimal pricing under a budget constraint.
After some price cap regulation attempts in the nineteenth century, the prac- tice of regulation was rate of return regulation, which ensures prices covering costs inclusive of a (higher than the market) cost of capital. This led to the Averch and Johnson (1962) over-capitalization result, which was largely overemphasized.
In 1979, Loeb and Magat finally put the regulation literature in the frame- work of the principal-agent literature with adverse selection by stressing the lack of information of the regulator. They proposed to use a Groves dominant strategy mechanism, which solves the problem of asymmetric information at no cost when there is no social cost in transfers from the regulator to the firm.
Baron and Myerson (1982) transformed the problem into a second-best prob- lem by weighting the firm’s profit with a smaller weight than consumers’ surplus in the social welfare function maximized by the regulator. In this scenario, opti- mal regulation entails a distortion from the first-best (pricing higher than marginal cost) to decrease the information rent of the regulated firm. Laffont and Tirole (1986) used a utilitarian social welfare function with the same weight for profit and consumers’ surplus, but introduced a social cost for public funds (due to distortive taxation), which also creates a rent-efficiency trade-off. Their model features both adverse selection and moral hazard, but the ex post observability of cost (commonly used in regulation) makes it technically an adverse selection model.14 This model was developed in Laffont and Tirole (1993) along many dimensions (dynamics, renegotiation, auctions, political economy, etc.).
Source: Laffont Jean-Jacques, Martimort David (2002), The Theory of Incentives: The Principal-Agent Model, Princeton University Press.