We are grateful to Martin J. Bailey, William Baumol, Michelle White, and anonymous referees for comments on an earlier draft of this paper, and to the National Science Foundation and the Sloan Foundation for their support of this work.

This paper points out how a system of marketable discount coupons applied to the sale of water by an agency can satisfy both redistributional and revenue constraints without compromise to economic efficiency, in effect 'creating marketable multipart pricing. This procedure achieves marginal cost pricing for each user while decreasing the average consumer's expenditures on water. Efficiency would obtain both in allocation among users and in overall usage. No information on consumer demand for water is requ!red. A useful by-product is market revelation of the marginal value of incremental water usage, thereby also shedding light on the value of incremental water projects.Policymakers "must find a way to make water rights secure and transferable." (T. L. Anderson and D. R. Leal, 1989, p. 432) "We want to raise prices but not kill the consumer. We don't want to bring him to his knees !" (L. Abalkin as quoted by P. Gumbel, 1990, p. A14) 1. INTRODUCTION Redistributional impacts and state constitutions often prevent m.unicipalities from using water as a source of net revenues. The regulting average cost pricing of water usage is particularly inefficient in areas with water supply problems {high marginal. replacement costs), and leads to excessive use of existing water supplies, shortages, and costly supplemental water projects. This paper proposes a solution to the trade-off between efficiency and equity in the pricing of water. Specifically, this paper points out an alternative to exhorting public officials to "bite the bullet" and drastically raise the price per unit of publicly provided water [e.g., Feldman, 1981; Brown and McGuire, 1984; Kucewicz, 199•1] or forgo most potential efficiency gains for the sake of political acceptability [e.g., Dandy et al., 1984].The paper proposes a "discount coupon system" (DCS), which is a technique to price water efficiently at the margin for each consumer while at the same time meet independently set revenue goals (such as zero revenues in excess of costs). The administering agency requires no information regarding consumer demand; information on the cost of

Entitlements to marketable water rate options (coupons) have been shown capable of addressing the issue of equity in water pricing, thus allowing efficient marginal cost pricing. This article proposes a "behind the scenes" variation that eliminates learning or transaction costs on the part of the consumer. The article proceeds to explore possibilities to adjust water rates so as to enhance revenues along with efficiency while using the companion instrument of transferable rate entitlements to address equity concerns. This combination opens up possibilities for privatization or deregulation of municipal water utilities.Innovative pricing schemes ... should be the order of business in the design of water service rates. —Martin and Wilder (1992, 101) Municipal water rates have traditionally served both an allocative and distributive function. Because one policy instrument must compromise its goals when serving two masters, economists and others have long searched for the optimal trade-off between efficiency and equity in municipal water pricing (e.g., Martin and Wilder 1992;Brajer and Martin 1989;Dandy, McBean, and Hutchinson 1984). Holding municipal water revenues constant, Collinge (1992) has pointed out that the trade-off between efficiency and equity can be avoided through supplementing water rates with a second policy instrument called discount coupons. That idea involves using entitlements to marketable price options (coupons) as the instrument targeting equity in the incidence of water rates. This frees policymakers to address efficiency as a separate issue.

No abstract

No abstract

A combination of fees and rebates—“feebates”—offers the economic efficiency of marginal cost pricing but keeps revenues equal to costs. Municipal water rates are expected to satisfy three objectives: efficiency, revenue neutrality to the utility, and distributional equity. Unfortunately, adjusting rates to efficiently achieve use and conservation targets would ordinarily generate excessive revenues. Rather than mold one tool to the service of three masters, this article suggests combining three separate tools. The first sets the water rate to cover the utility's costs. The second assigns customers allotments to water use. The third either charges a fee for use that exceeds the customer's allotment or hands out rebates for consumption below that allotment. The fees pay for the rebates—thus the term “feebate.” The outcome is (1) revenues to the utility just sufficient to cover costs, (2) efficient water consumption by municipal water customers without conservation mandates, and (3) revenue effects that can be spread fairly across various categories of customers.

Marginal-cost pricing of municipal water in water-scarce regions is both an efficient way to achieve conservation goals and an efficient revenue source for municipalities. Nevertheless, marginal-cost water pricing is quite unusual and often prohibited, probably because water revenues are perceived to have an unfairly regressive incidence. To examine the validity of this perception, Census data from San Antonio, Texas, is used to estimate the income elasticity of demand for water there. This result and similar estimates elsewhere suggest that the regressivity of water revenues compares favorably with published regressivity estimates for traditional municipal revenue sources. Moreover, a form of two-part water pricing is shown capable of increasing progressivity without inefficiency. Revenue under this pricing regime is flexible, which allows authorities to meet a range of preset revenue targets, trading off neither efficiency nor fairness.

Confusion surrounding the appropriateness of long-run considerations in effluent regulation has arisen in the literature and recently carried over into textbooks. We use a factor input model under oligopsony to show that, when firms can influence the level of marginal damages, a linear pollution tax does not satisfy the long-run entry-exit condition. Previous results to the contrary are shown to depend on restrictive assumptions. Efficient policy design requires a lump-sum refund or any one of various non-linear pricing schemes. JEL Classification: Q2, L1Les impôts linéaires à la Pigou et la taille optimale d'une industrie polluante. Il y a beaucoup de confusion dans les débats qui entourent la réglementation des effluents, et son caractère plus ou moins approprié à long terme, tant dans la littérature spécialisée que dans les manuels. Les auteurs tentent d'éliminer cette confusion en analysant les émissions à l'aide d'un modèle standard de demande d'intrant. Quand les entreprises peuvent influencer la valeur présente des dommages marginaux, un impôt linéaire sur la pollution n'entraîne pas nécessairement les décisions appropriées d'entrée et de sortie. On peut corriger le problème à l'aide d'un remboursement forfaitaire ou de l'une ou l'autre des formes de tarification non-linéaire.