The International Yearbook of Environmental And Resource Economics 2005 2006: A Survey of Current Issues (New Horizons in Environmental Economics)

The International Yearbook of Environmental and Resource Economics 2005/2006

NEW HORIZONS IN ENVIRONMENTAL ECONOMICS Series Editors: Wallace E. Oates, Professor of Economics, University of Maryland, USA and Henk Folmer, Professor of General Economics, Wageningen University and Professor of Environmental Economics, Tilburg University, The Netherlands This important series is designed to make a significant contribution to the development of the principles and practices of environmental economics. It includes both theoretical and empirical work. International in scope, it addresses issues of current and future concern in both East and West and in developed and developing countries. The main purpose of the series is to create a forum for the publication of high quality work and to show how economic analysis can make a contribution to understanding and resolving the environmental problems confronting the world in the twenty-first century. Recent titles in the series include: Environmental Management and the Competitiveness of Nature-Based Tourism Destinations Twan Huybers and Jeff Bennett The International Yearbook of Environmental and Resource Economics 2003/2004 A Survey of Current Issues Edited by Henk Folmer and Tom Tietenberg The Economics of Hydroelectric Power Brian K. Edwards Does Environmental Policy Work? The Theory and Practice of Outcomes Assessment Edited by David E. Ervin, James R. Kahn and Marie Leigh Livingston The International Yearbook of Environmental and Resource Economics 2004/2005 A Survey of Current Issues Edited by Tom Tietenberg and Henk Folmer Voluntary Approaches in Climate Policy Edited by Andrea Baranzini and Philippe Thalmann Welfare Measurement in Imperfect Markets A Growth Theoretical Approach Thomas Aronsson, Karl-Gustaf Löfgren and Kenneth Backlund Econometrics Informing Natural Resources Management Selected Empirical Analyses Phoebe Koundouri The Theory of Environmental Agreements and Taxes CO2 Policy Performance in Comparative Perspective Martin Enevoldsen Modelling the Costs of Environmental Policy A Dynamic Applied General Equilibrium Assessment Rob B. Dellink The International Yearbook of Environmental and Resource Economics 2005/2006 A Survey of Current Issues Edited by Henk Folmer and Tom Tietenberg

The International Yearbook of Environmental and Resource Economics 2005/2006 A Survey of Current Issues Edited by

Henk Folmer Professor of General Economics, Wageningen University, The Netherlands and Professor of Environmental Economics, Tilburg University, The Netherlands

Tom Tietenberg Mitchell Family Professor of Economics, Colby College, US NEW HORIZONS IN ENVIRONMENTAL ECONOMICS

Edward Elgar Cheltenham, UK • Northampton, MA, USA

© Henk Folmer, Tom Tietenberg 2005 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited Glensanda House Montpellier Parade Cheltenham Glos GL50 1UA UK Edward Elgar Publishing, Inc. 136 West Street Suite 202 Northampton Massachusetts 01060 USA

A catalogue record for this book is available from the British Library

ISSN 460 7352 ISBN 1 84542 206 6 (cased) Printed and bound in Great Britain by MPG Books Ltd, Bodmin, Cornwall

Contents List of figures List of tables Contributors Preface Editorial board 1

vi vii viii ix x

Issues in water pricing reforms: from getting correct prices to setting appropriate institutions Ariel Dinar and R. Maria Saleth

2

Spatial environmental policy Jacqueline Geoghegan and Wayne B. Gray

3

Environmental equity and the siting of hazardous waste facilities in OECD countries: evidence and policies James T. Hamilton

1 52

97

4

Strategies to conserve biodiversity Stephen Polasky

157

5

Corporate sustainability Stefan Schaltegger and Roger Burritt

185

6

The double-dividend hypothesis of environmental taxes: a survey Ronnie Schöb

223

Valuing environmental changes in the presence of risk: an update and discussion of some empirical issues W. Douglass Shaw, Mary Riddel and Paul M. Jakus

280

7

312

Index

v

List of figures 1.1 Average price for irrigation water by per capita water availability 1.2 Average price for residential water by per capita water availability 5.1 Corporate sustainability challenges 5.2 Relationship of different corporate sustainability dimensions to each other 5.3 Potential relations between corporate environmental and/or social performance and economic success 6.1 Pigovian tax 6.2 Labor tax system versus green tax system 6.3 Energy taxation and competitive labor markets 6.4 Marginal green tax reforms 6.5 The one-period model of exhaustible resource consumption 6.6 A two-period model of exhaustible resource consumption 7.1 Graham’s expected utility locus for any pair of payments, PS and PH

vi

13 15 189 195 197 225 248 259 262 265 266 286

List of tables 1.1 1.2 1.3 3.1 3.2 3.3

Methods for pricing of irrigation water Comparison of key variables of various pricing methods Stakeholder matrix for major country price reform case studies Evidence on the distribution of hazardous waste facilities Determinants of exposure to hazardous waste facilities Conclusions of Chapter 3

vii

6 17 23 141 143 144

Contributors Roger Burritt, The Australian National University, Australia Ariel Dinar, The World Bank, Washington, DC, USA Jacqueline Geoghegan, Clark University, USA Wayne B. Gray, Clark University, USA James T. Hamilton, Duke University, USA Paul M. Jakus, Utah State University, USA Stephen Polasky, University of Minnesota, USA Mary Riddel, University of Nevada, Las Vegas, USA R. Maria Saleth, International Water Management Institute, Colombo, Sri Lanka Stefan Schaltegger, University of Lueneburg, Germany Ronnie Schöb, Otto-von-Guericke-University, Magdeburg and CESifo, Munich, Germany W. Douglass Shaw, Texas A&M University, USA

viii

Preface As a discipline, Environmental and Resource Economics has undergone a rapid evolution over the past three decades. Originally the literature focused on valuing environmental resources and on the design of policy instruments to correct externalities and to provide for the optimal exploitation of resources. The relatively narrow focus of the field and the limited number of contributors made the task of keeping up with the literature relatively simple. More recently, Environmental and Resource Economics has broadened its focus by making connections with many other subdisciplines in economics as well as the natural and physical sciences. It has also attracted a much larger group of contributors. Thus the literature is exploding in terms of the number of topics addressed, the number of methodological approaches being applied and the sheer number of articles being written. Coupled with the high degree of specialization that characterizes modern academic life, this proliferation of topics and methodologies makes it impossible for anyone, even those who specialize in Environmental and Resource Economics, to keep up with the developments in the field. The International Yearbook of Environmental and Resource Economics: A Survey of Current Issues was designed to fill this niche. The Yearbook publishes state-of-the-art papers by top specialists in their fields who have made substantial contributions to the area which they are surveying. Authors are invited by the editors, in consultation with members of the editorial board. Each chapter is critically reviewed by the editors and by experts in the field. The editors would like to thank Wallace Oates for his help in getting the project started. We also owe a special debt of gratitude to Ed Barbier, Lans Bovenberg, Tom Crocker, Shelby Gerking, Larry Goulder, Charles Howe, Madhu Khanna, Katherine Kiel, Kees van Kooten, Ray Palmquist and Mary Riddell for their assistance in editing this volume. Henk Folmer Tom Tietenberg

ix

Editorial board EDITORS Henk Folmer, Wageningen University and Tilburg University, The Netherlands Tom Tietenberg, Colby College, USA

EDITORIAL BOARD Kenneth Arrow, Stanford University, USA Scott Barrett, Johns Hopkins University, USA Nancy Bockstael, University of Maryland, USA Peter Bohm, Stockholm University, Sweden Lans Bovenberg, Tilburg University, The Netherlands Trond Björndal, University of Portsmouth, UK Carlo Carraro, University of Venice, Italy Partha Dasgupta, University of Cambridge, UK Ariel Dinar, The World Bank, USA Shelby Gerking, University of Central Florida, USA Lawrence Goulder, Stanford University, USA Eiji Hosoda, Keio University, Japan Per-Olov Johansson, Stockholm School of Economics, Sweden Bengt Kriström, Swedish University of Agricultural Sciences, Sweden Karl-Gustav Löfgren, University of Umeå, Sweden Juan-Pablo Montero, Catholic University of Chile, Chile Adolf Mkenda, University of Dar es Salaam, Tanzania Wallace Oates, University of Maryland, USA Charles Perrings, York University, UK Alan Randall, The Ohio State University, USA Michael Rauscher, Rostock University, Germany Kathleen Segerson, University of Connecticut, USA Bernard Sinclair-Desgagné, HEC Montreal, Canada V. Kerry Smith, Duke University, USA Robert Solow, MIT, USA Alistair Ulph, University of Southampton, UK Michael Young, CSIRO land and water, Australia x

1. Issues in water pricing reforms: from getting correct prices to setting appropriate institutions Ariel Dinar and R. Maria Saleth* INTRODUCTION Although water misuse and scarcity are two issues that many countries have been living with for many years, it was not until 1992 that the world had faced the declaration, known as the 1992 Fourth Dublin Principle,1 which was soon supplemented by what became known as the 1992 First Rio Principle.2 While the Dublin Principle views water as an economic good, the Rio Principle suggests, implicitly, that water is a social good and humans are entitled to, at least, a certain level (in terms of quantity and quality) of it, especially under the responsibility of their respective governments. The Dublin and the Rio Principles reflect, in fact the contrasting beliefs of water experts and policymakers and demonstrate, to a larger extent, the polarization that could and actually does take place in the global dialogue on managing our planet’s scarce water resources. Such views and beliefs also affect the ever-growing debate over pricing or charging for water by experts representing various disciplines.3 We could start our discussion by describing the growing scarcity of water in many parts of the world. We could emphasize the especially severe situation in developing countries, but we prefer to refer the reader to existing literature on such future scenarios and spatial distribution of the problem (e.g., Seckler et al., 1998; Gleick, 2002). We would rather start by simply referring to the simple ratio for each country of available water resources per capita,4 which is one of the indicators for relative water availability driven by population growth (Falkenmark, 1984). Per capita availability dropped by more than one half in the past 50 years. Driven mainly by population growth rates, per capita water availability in the next 50 years will continue to drop. The per capita water availability in 2050 is expected to reach a level of 10–20 per cent of that in 1955, independently of the initial ‘water wealth’ of countries or their level of development (Population 1

2

The international yearbook of environmental and resource economics

Action International, 1995). The available amount of water is being used globally for irrigation (70–90 per cent), in residential uses (5–10 per cent), in industrial uses (5–10 per cent), depending on the country (World Bank, 2003b). Because of its lion share in water use, irrigation water will also deserve more attention in this chapter. The quantity dimension of water scarcity is further exacerbated by increased water pollution from economic activity, leading to water quality that could, in many locations, exclude the resource from being used for certain economic and human activities. Although not documented in the literature on global level impact, water pollution is one of the main limiting factors of water use in many developed as well as developing countries. Policymakers are searching for and debating on ways to close the gap between soaring usage of water and its limited availability. Water pricing is considered as one of many policy interventions to mitigate both quantity and quality dimensions of water scarcity and enhance efficient water use.5 It is one of the most important policy instruments for integrating supply augmentation with demand management so that an efficient allocation and use of the already uitilized resources provide the economic and financial justification for the development of additional supplies from both conventional and unconventional sources. Pricing of water has two key roles, (1) a financial role as a mechanism for recovering the investment and operation and maintenance (O&M) cost of the water system, and (2) an economic role of signaling the scarcity value and opportunity cost of water to guide allocation decisions both within and across water sub-sectors. Under certain conditions, pricing of water could also promote equity objectives. There are big differences among sectors such as irrigation, urban, hydropower and environment. These differences stem from three causes: (1) The first cause is the nature of use of the water by the sector; for example, irrigation and urban use permanently remove the water (consumptive use) from the system (rivers, lakes, or aquifers) and in most cases deteriorate its quality by loading it with pollutants (chemicals). Hydropower removes the water temporarily (nonconsumptive use) from the system and may affect its quality. And environment uses the water also in a non-consumptive way and in many cases improves its quality (if not too deteriorated in the first place). (2) The second cause is the ability to identify individual users; for example, while the use by hydropower is transparent, it is hard to identify individual users and monitor their usage when appropriate infrastructure doesn’t exist in irrigation or urban uses. Also, in the case of environment, assigning quantities to uses has still not been established. (3) The third cause is the value assigned to the water in the sector; for example, the two extreme values are

Issues in water pricing reforms

3

urban and irrigation, with urban leading in the value its uses assign to water and irrigation closing the list with relatively lower values obtained from use of the water. Environment and hydropower are within this range (e.g., Gibbons, 1986; Ward and Michelsen, 2002; see also a detailed discussion later on). In addition to sectoral differences, we must also recognize, especially in the irrigation sector, the impacts that physical locations have on these three aspects. The locational effect is especially important when natural drainage, that removes excess applied water, does not exist and thus sub-surface water table starts to rise and create logging and salinity problems in down slope locations (Dinar and Zilberman, 1991). A major difficulty arises because the pollution is a non-point source one, so identification of polluters is becoming complex. Water pricing policies in most countries fail to perform most likely due to a faulty approach and inappropriate institutions that have their roots in a complicated political economy environment. While development agencies, such as the World Bank, maintained the policy that cost recovery should be sufficient to pay both operation and maintenance and for fair return on capital investment, implementation on the ground is unsatisfactory for most of the cases. Use of prices for rationing scarce water use is almost non-existent. Examples where pricing for water rationing has been attempted include: Israel (both in the irrigation and urban sectors (Yaron, 1997), United Kingdom (Scott and Eakins, 2002), and Broadview water district in California (Wichelns, 1991). In the latter example, block rate pricing was implemented and is still in existence, in order to reduce water quality related problems. A common problem in such cases is that the intent is to ration water but the design of the rate structures tends to be a compromise of political pressure by interest groups and thus, falls short of its intent. The most extreme example can be found in Scott (2003), where a water rate structure that was agreed upon in one administration for rationing water in Ireland was abolished after the elections. As can be seen in Figures 1.1 and 1.2, actual water prices in the irrigation and residential sectors (the same holds for other sectors) do not reflect the level of water scarcity in the said country. Apart from allocation and efficiency implications, nominal water charges and poor cost recovery also risk the necessary maintenance level of existing water infrastructure as well as the additional investments in future water development projects. We would like to familiarize the reader with the ongoing debate about the applicability of water pricing under real world conditions. For example, a recent World Bank Water Resources Management Strategy (World Bank, 2003a) acknowledges the past difficulties in collecting payments for water and proposes the use of water rights as a first step towards a more efficient

4

The international yearbook of environmental and resource economics

use of the resource.6 Having rights assigned to users has been suggested by Coase (1960) as a prerequisite for efficient use of a common pool resource, such as water. In the case of water, this principle was implemented and assessed by Ciriacy-Wantrup (1967, 1969) and by Clark (1967), who also reported the complexities involved in its application – especially the difficulties in handling rights associated with return flows. The fact that efficient water pricing schemes are rare, if not completely absent, even in economically advanced countries with extreme water scarcity levels, provides sufficient evidence for the persistence of a vast gap between the development of pricing theory and its practical application. In view of its tremendous welfare implications, this literature–policy gap is obviously a serious challenge to both theoreticians and policymakers. Why is there this gap? Does the innate nature of the resource limit its practical treatment within standard theoretical principles? Is it due to the inherent conflicts among the multiple roles (cost recovery, allocative, and distributional) that water pricing has to play? Are there any limitations on the theoretical front, especially the inability of economic theory to capture non-economic concerns and actual experiences? Do these theoretical limitations constrain practical implementation of pricing principles? Can the theory–policy gap be bridged by integrating some recent developments in related fields of political economy and institutional economics with the water pricing theory? In this chapter, we will address water pricing reforms by synthesizing three sets of literature: water pricing theory, political economy theory of reforms, and institutional economics principles of institutional design. Such a synthesis could indeed link theory development with practical application, especially by moving the focus from getting prices right to setting appropriate institutions. The chapter will start by addressing the notion of price with its different dimensions. We will continue by discussing whether water is an economic or social good. The answer to that question affects to a large extent the pricing approach to be implemented. The pricing literature will then provide the technical basis for setting prices that are necessary for an efficient and equitable utilization of the resource. This literature will be amended by a variety of case study literature that demonstrates attempts to apply water pricing mechanisms under a range of conditions in the irrigation and residential sectors of various countries. The chapter then will review both political economy literature and institutional economics literature related to water pricing. The political economy literature will focus on both the constraints and opportunities for the implementation of water pricing policies. And, the institutional economics literature will focus on the linkage between institutional change and economic performance in the particular context of water pricing reforms. The chapter will conclude

Issues in water pricing reforms

5

by identifying outstanding gaps and by indicating suggested areas for further research in all three sets of literature.

THE NOTION OF WATER PRICE Water is not a well-defined ‘commodity’. Beginning with what is called bulk water and ending with what is increasingly becoming common – bottled water, one can realize that there are several purposes for water use, each of which is associated with a different commodity and thus should also be distinguished by a different price. The discussion of water price in this section does not address the notion of the value of water that will be discussed in the next section. A very eye opening analysis of the various ‘agents’ along the chain of water production is provided by Massaruto (2004), who distinguishes among: water owners (government), operators of bulk-water facilities, the retail sector, comprised of the city water agencies and the irrigation water supply agencies, and the end users of all sectors. Each of these ‘agents’ observes a different price. In our chapter, we focus mainly on the end users. Such a focus enables us to concentrate on the economically and practically most relevant issues to be addressed here. Using a simple example of two uses only, urban and irrigation, both use the same source, but their needs in terms of timing, flow and quality make the water they will consume different. Urban users need more or less a constant flow all year round at a different quality standard than does irrigation. Irrigators need the water mainly during the irrigation period, in relatively big bulks. The quality vector of irrigation water includes much less components than the one used for urban consumers. As a result, water for urban uses necessitates additional processing and different conveyance infrastructure than those for irrigation, resulting in different cost components to affect the price users are charged. In addition to differences in water products end consumers face, that affect the price of water, there are also differences in the type of costs that regulators include in the calculus of water pricing (charging). These costs are a function of the time horizon – long- versus short-run – that regulators consider for setting prices. As Tsur and Dinar (1997) assert, first-best efficiency ignores implementation costs or incomplete information (such as when water is unmetered); second-best efficiency accounts also for these factors. Short-run efficiency ignores fixed cost (e.g., infrastructure). A more detailed explanation, by pricing method (in irrigation) is provided in Table 1.1. A general presentation of these issues can be found in Massaruto (2004) for the irrigation sector, and in Bauman et al. (1998) for the urban sector.

6

The international yearbook of environmental and resource economics

Table 1.1

Methods for pricing of irrigation water

Method

Operation principles

Volumetric (single rate)

Irrigators pay per volume of water consumed. Variations of the volumetric approach include (1) indirect calculation based on measurement of flow time (as from a reservoir) or time of uncertain flow (proportions of a flow of a river), and (2) a charge for a given minimal volume to be paid for even if not consumed.

Volumetric (tiered)

Irrigators pay per volume according to the volume consumed. This is a multirate volumetric method, in which water rates vary as the amount of water consumed exceeds certain threshold values. Number of tiers could be greater than two.

Output

Irrigators pay a water fee for each unit of the output they produce.

Input

Irrigators pay a water fee for each unit of a certain input used.

Per unit area

Irrigators are charged per irrigated area, depending on the kind and extent of crop irrigated, irrigation method, the season of the year, etc. Charges may be higher if there are storage works (investment) than for diversions directly from streams. Pumped water is usually charged higher than gravity water. Farmers are required to pay, in some cases, the per area charges also for their non-irrigated land.

Two-part tariff

Involves charging irrigators a constant marginal price per unit of water purchased (volumetric marginal cost pricing) and a fixed annual (or admission) charge for the right to purchase the water. The admission charge is the same for all farmers. This pricing method has been advocated, and practiced, in situations where a public utility produces with marginal cost below average cost and must cover total costs (variable and fixed).13

Betterment levy

Water fees are charged per unit area, based on the increase in land value accruing from the provision of irrigation.

Water market

Irrigators sell/buy/rent water rights for an agreeable price for use at present or in the future.

Source: Based on Tsur and Dinar (1997).

Issues in water pricing reforms

7

THE VALUE OF WATER: A SOCIAL AND AN ECONOMIC GOOD Since water needs to be everyone’s business (Prince of Orange, in World Water Commission, 2000), one can find the cross-current of multidisciplinary influence on the national and global dialogues on water pricing policies and their implementation. We start this review by addressing the literature on the emerging related questions of whether water is a social or an economic good and what is the economic value of water. The two early attempts to assess the economic value of water are Lichty and Anderson (1985) and Gibbons (1986) who proposed analytical tools to deal with determining the value of in various sectors, using mainly ‘wellbehaved’ water demand and supply functions. Gibbons acknowledges the difficulties arising from different nature of water use and measurement problems in the municipal, irrigation, industry, recreation, navigation, and hydropower sectors. Lichty and Anderson came up with some alternative assessments in the case of measurement problems. Young and Haveman (1985) review the literature on economics of water resources, including also work attempting to assess sectoral values associated with investment in water projects. Young and Haveman review mainly literature that applies cost benefit analysis for many sectors, and in the case of agriculture, also the ‘residual approach’.7 Detailed sectoral and locational studies are less common, as actual measurement of values is difficult. Some studies pertaining to irrigation water include Madariaga and McConnell (1983) who use surveys of agricultural production and water use, as well as use by competing sectors to assess the value of water in irrigated agriculture. Following the 1992 Dublin and Rio debates that were mentioned earlier, and with the increased attention to the state of water on earth, several authors have initiated the discussion on whether water is an economic or social good. Those who ended up concluding that it is an economic good followed with an assessment how it should be managed as such. Briscoe (1996, 1997) develops a conceptual framework to treat water as an economic good, and applies it to various sectors to discuss its use value in each of the sectors, including two case studies from Germany and Chile. Rogers et al. (2002) develop also a framework to treat water as an economic good. They take another route, however, and use that framework to advocate for the use of pricing in order to promote efficiency, equity and resource sustainability for various uses. In another, rather detailed and sector specific work in the Rio Grande Basin, Ward and Michelsen (2002) apply the concept of the economic value of water to the irrigated agriculture sector. The additional aspects that are introduced here are the inclusion of extreme

8

The international yearbook of environmental and resource economics

water supply situations and interactions between irrigated agriculture and other sectors. Use of policy instruments, such as pricing, for affecting the efficient use of the resource, is demonstrated. All three works on the economic value of water also address the conceptual differences among the price, cost, and value of water. Rogers et al. (2002), in particular, argue for the need for full cost recovery, an issue that is still in debate, as could be seen in Abu-Zeid (2001) who introduces the notion of water as a social good8 and advocates for the special attention needed for the irrigation sector in developing countries, by addressing implicitly the ‘willingness to pay’ and the ‘ability to pay’ concepts. A parallel paper by Massaruto (2002) reviews experiences with cost recovery in irrigation in the European context and provides a detailed analysis of the terms ‘economic cost’ of water and ‘recovery of the cost’. An interesting approach that incorporates additional sets of water values is suggested by Just et al. (1999), using the example of Israel. The authors suggest that food self-sufficiency, ideological settlement goals, equity, and several other long-run considerations may rule over short-run economic efficiency in the Israeli water industry. They then incorporate new peace opportunities and regional water related projects into the price equation to have complementary effects on equilibrium quantities and prices.

WATER PRICING: THEORY AND PRACTICE Pricing of water is suggested in order to send signals of its economic value to users. Prices could lead individuals to behave efficiently in their allocation and use of water resources in the light of the total net benefit generated from use of existing water volumes and given existing management practices and technologies. This assumes profit maximizing decision users, symmetric information, no transaction costs, and competitive market structure. The theoretical foundation of pricing water is similar to that of other products. For a detailed analysis of pricing in the irrigation water sector see Tsur et al. (2004). For a detailed analysis of the residential water sector see for example Howe and Linaweaver (1967), Brown and Sibley (1986), Hanemann (1998). The economics of water pricing assumes a well-behaved water demand function.9 When the price of water is set to equal its marginal value a profit seeking user will demand a given quantity that is linked to that price, and can be predicted by the regulator who sets prices.10 In the case of residential consumers maximization is performed over utility rather than over production profit. Methodologies to estimate demand for irrigation water are available in the literature (Bontemps and Couture, 2002; Tsur et al.,

Issues in water pricing reforms

9

2004). A more practical set of best management practices is provided in Hydrosphere (1997, 1998). In the absence of taxes, subsidies and other distortions (to be addressed in the sections on the political economy and institutions), a price leading to allocation that maximizes total net benefits is called first-best or Pareto efficient. Pareto efficiency can be regarded as ‘short-run efficient’ if the pricing optimization problem involves variable (i.e., short-run) costs, and a ‘long-run efficient’ if fixed (long-run) costs are included. When prices are set under distortions, such as information asymmetry, institutional limitations, or political constraints, the price leading to that allocation is termed second-best efficient (Mascollel et al., 1995; Tsur and Dinar, 1997). Highlights from Tsur et al. (2004) provide several guidelines as to what one could expect from various water pricing schemes for irrigation water. Several of them are provided below: ●

●

●

●

●

●

●

Marginal-cost pricing achieves efficient water allocation in that it maximizes the joint surplus of water users – farmers – and water suppliers. When water derived from sources of different quality, for example, fresh, saline, or reclaimed water, has different effects on crop yield, each water quality is treated as a separate input and must be priced separately. The demand for each water type depends on the available supply as well as demand for other water types. In light of demand for different types of water, prices should be determined simultaneously for all types of irrigation water. When water is priced volumetrically, efficiency requires that the price of water reflect the marginal cost of water supply, disregarding water allocation between crops, that is, water price should not change across crops. Under per area pricing, changing the per hectare water fee across crops can be used to improve efficiency by affecting farmers’ crop choices. Block-rate pricing can be used to transfer wealth between water suppliers and farmers, while retaining efficiency, that is, a maximal joint surplus of farmers and water suppliers. Water prices have limited effects on income distribution within the farming sector and are therefore a poor means of addressing income distribution goals. Average-cost pricing guarantees a balanced water supply budget, but entails a loss of efficiency, because it decreases the joint welfare of farmers and water suppliers. Moreover, the farmers carry the burden of the welfare loss.

10

The international yearbook of environmental and resource economics ●

●

●

The cost of implementing a pricing method is an integral part of the cost of water supply and should be included in the variable cost (VC) – hence also in the marginal cost (MC) – and fixed cost (FC) of water supply. Macroeconomic analysis of reform must be linked to the microeconomic analysis of water market reforms at the farm or perimeter level. Economic reforms outside of agriculture affect the productivity of water in irrigated agriculture. Linking macroeconomic reform and water policy reform can increase not only the national productivity of water in agriculture but also the welfare of rural households by more that macroeconomic reform alone.11

An important aspect of water pricing, that has been investigated mainly in regard to pricing of domestic water is ‘the willingness to pay’ for water (e.g., Whittington, Lauria and Mu, 1991; Whittington and Choe, 1992; The World Bank Water Demand Research Team, 1993; Anand, 2001a; Merrett, 2002a and 2002b). In developing countries, the literature also considered the ‘ability to pay’ for water by low-income populations.12 In a series of works on cost effectiveness and socioeconomic impacts of water conservation rate structures in the United States, Beecher et al. (2001), Chesnutt et al. (1997), and Pekelney et al. (1996) come up with a set of guidelines suggested for the design and implementation of recommended practices for urban sector water rates. Beecher et al. (2001) also address the affordability vs. conservation of the impact of the proposed rates. They, in particular, address social issues related to the elderly population, and ethnicity. In essence, the work by Beecher et al. (2001) serves as an example of how water is treated as a social good, addressing the special needs and ability to pay of the weakest segments of society – elderly and new immigrants. Pricing Methods Irrigation water The literature dealing with use of irrigation water pricing (e.g., Rhodes and Sampath, 1988; Cummings and Nercissiantz, 1989; Sampath, 1992; Small and Carruthers, 1992; Shah; 1993; Tsur and Dinar, 1995), displays a variety of methods, depending on natural, political, and economic background. Pricing irrigation water consists of volumetric, non-volumetric, and market based pricing methods. Volumetric mechanisms are based on measuring or assessing the actual amounts of water consumed. Non-volumetric methods

Issues in water pricing reforms

11

are based on measurement of surrogate measures such as output, input, area or land values. It is likely that non-volumetric methods face inadequate information concerning actual consumption volumes. This is also a concern with volumetric methods, although to a lesser extent. The recently exercised market based mechanisms for allocating and setting prices for irrigation water (e.g., Easter et al., 1998) necessitate well-defined water rights to determine the irrigation water price. There are also combinations of methods that are thoroughly reviewed by Johansson et al. (2002) and Bosworth (2002). Tsur and Dinar (1995, 1997) review the various irrigation pricing methods in practice around the world. In principle, there are two distinct groups of methods: (1) administrative and (2) market based. The administrative methods can also be grouped into volumetric and non-volumetric ones (as was indicated earlier), and combination of the two. Table 1.1 provides the list of the methods and a short explanation of their characteristics. A comparison of the implementation aspects of the various pricing methods is provided in Table 1.2. A priori objectives of tiered pricing schemes mean to address efficiency and equity issues among users/consumers. However, Tsur and Dinar (1997) assert that irrigation water pricing schemes that do not involve quantity quotas cannot be used in policies aimed at affecting income inequality. This includes the volumetric, output, input, tiered, and per hectare pricing methods, among others. To affect income inequality, a water pricing method should include certain forms of water quantity restrictions. A similar observation holds also in the case of urban water supply pricing schemes that are discussed below. As was found in many empirical works (e.g., Saleth and Dinar, 2001), the important factor in addressing equity issues is not the price of the first tier, but its length, namely, the quantity. A study of the urban sector of Chile and Colombia by Gómez-Lobo and Contreras (2003) supports the quantity limit thesis that was mentioned above in the context of the irrigation sector. Two subsidy programs associated with lower monthly tariffs to poor residents were implemented. The Chile program targets individual families based on their annual income level, while the Colombia program targets poor families based on their residential location. In both cases the tariff for the first tier was subsidized and the quantity of water was restricted. Both subsidy schemes were defined as efficient in terms of targeting the right population, rationing water, and managing and balancing the water supply agency’s budget. See Figure 1.1 for a depiction of the average price for irrigation water by per capita availability.

12

The international yearbook of environmental and resource economics

Urban water Among urban water pricing methods, the most commonly observed are fixed rates, two-part tariff, and tiered pricing (Brown and Sibley, 1986; OECD, 1987; Bonbright et al., 1988; Garfield and Lovejoy, 1994; Espey, Espey and Shaw, 1997; Baumann et al., 1998; Renzetti, 2002). Besides betterment levies, taxes are also imposed on water abstraction and environmental damage (Malik, Larson and Ribaudo, 1994; Krinner and Lallana, 1999). While the methods discussed above are widely used, there are also instances where water prices are determined by spontaneous/informal or established water markets where water is bought by either municipalities or farmers from farmers (Howe, Schurmeier and Shaw, 1984; Rosegrant and Binswanger, 1994; Hearne and Easter, 1997; Easter, Rosegrant and Dinar, 1998 and 1999). Contrary to what many believe, water markets (either based on established water rights or based on informal and de facto water rights) are gaining popularity among users. The established market price does usually reflect the economic value of the traded water. Demand for establishing markets for water is more recognized in countries where supportive institutions exist. In such cases (e.g., McKay and Bjornlund, 2002), a great deal of sophistication can be achieved (e.g., addressing social justice, sustainability), by addressing some prerequisites, such as assigning water rights to existing users (and especially the poor), and accounting for the rights of the rivers to certain levels of flow, or allowing ‘friends of the rivers’ to purchase water rights to keep the flow, on behalf of rivers. Water markets have been operating among irrigators and between irrigators and urban water suppliers. Impact of water market transactions on farmers’ decisions can be found in Arriaza et al. (2002) for the case of Southern Spain, and in Hearne and Easter (1997) for the case of Chile. In situations where the legal and institutional system is not sufficiently developed, informal markets are more common in setting water prices (Easter et al., 1998; Kloezen, 1998; Kumar and Singh, 2001). Generally speaking, impact of the water market on irrigators could be divided into short- and long-term effects. In the short-run, farmers as sellers can fallow parts of their land and use the water more efficiently on the rest of their farm. In the longer run they could use the proceeds to invest in improved irrigation systems, or change their cropping pattern and adjust it to the remaining water quota and thus even be able to farm the same amount of land and even be financially better off (Dinar and Letey, 1991; Hearne and Easter, 1997). In some cases, as observed in Chile, very small farmers may have to sell or desert all their assets and migrate to the neighbouring cities or become labourers on other farms (Bauer, 1998). Impact of water pricing on both welfare and conservation is the subject

Issues in water pricing reforms

13

1.40

$/cubic meter

1.20 1.00 0.80 0.60 0.40 0.20 0.00 0 5000 10000 15000 Available water per capita [cubic meter per year] Source: Based on Dinar (2000b).

Figure 1.1 Average price for irrigation water by per capita water availability of numerous works, especially in the residential water sector. We will mention a few recent works only. In extreme water-scarce countries water pricing has special implications for both conservation and welfare. It is therefore, interesting to observe the similar conclusions obtained in the case of Cyprus, Tunisia, and Abu Dhabi. Hajispyrou et al. (2002) find that uniform marginal cost pricing in Cyprus will reduce distortions and increase the efficiency of the water allocation system but on the other hand it will also be biased towards improving the welfare of the better-off households. In the case of Tunisia, Zekri and Dinar (2003) compare both the efficiency and welfare aspects of public and private water supply alternatives in the rural sector and find that although the welfare effects of the private supply systems are far more noticeable, both private and public supply systems are far from being efficient. Abu Qdais and Nassay (2001) assessed the impact of a change in sea-water desalinization pricing method in Abu Dhabi City, where a flat rate was replaced by a volumetric method with a significant price increase of 300 per cent. Sixty-six per cent of the sampled consumers reduced their consumption by 5–85 per cent, which had an immediate effect on the proceeds of the supplier, with short- and long-term consequences for consumers. In the case of irrigation water, efficient water pricing schemes do not necessarily lead to overall water conservation. As suggested by Dinar and

14

The international yearbook of environmental and resource economics

Zilberman (1991) and Dinar and Mody (2004) for the case of irrigation, while profit maximizing farmers react by shifting to less water-intensive crops or opt out of irrigated crops altogether and switch to rain fed agriculture, this need not always be the case (Massarutto, 2002). In situations where opportunities to increase the productivity of water exist, farmers may actually modify their management practices, such as investing in more water-efficient technologies, and/or switching to more market oriented, high-value but also more water intensive crops. In this way, where feasible, sufficiently higher prices may cause an increase in total demand for water.14 Similar results were obtained by Massaruto (2003), where it was shown clearly that using water more efficiently does not necessarily have the effect of reducing water demand: to the contrary, high-value crops entail usually higher water requirements and lower elasticity. Productivity of water might become greater, but pressure on water resources is not likely to vary too much. A more efficient water use might well be also a less sustainable one. See Figure 1.2 for a depiction of the average price of residential water by per capita water availability. Available Evidence on Pricing Practices With a solid background of pricing theory and methods, sectoral water rates and the underlying water pricing methods observed in various countries are presented (e.g., Dinar and Subramanian, 1997 and 1998; Jones, 1998; OECD, 1999; Savedoff and Spiller, 1999; Ahmad, 2000; Dinar, 2000a and 2000b). Evidences are used to show how the prevalent water rates even in advanced countries reflect neither the true supply cost nor the scarcity value of water. With information on various sectors in each country surveyed in the above mentioned studies, it is also apparent that the most common approach to pricing/charging for water in all sectors is the flat rate per household per month, or the per hectare charge in the agricultural sector. The effectiveness of the same pricing method and pricing structure shows considerable variation across regions/countries, suggesting the effects of interplay between water resources and their surrounding physical, socioeconomic and institutional conditions. With that evidence, the stage is now set to introduce the political economy and institutional considerations into the discussion. Since pricing of water and water services is addressed in the literature mostly on a sectoral basis, the review will also take a sectoral perspective, concentrating especially on the irrigation and urban sectors that account together for over 90 per cent of water used in most circumstances. Existing reviews of pricing literature in the context of irrigation (e.g., Young and

Issues in water pricing reforms

15

Haveman, 1985; Easter, 1992; Sampath, 1992; Moore, 1999; Johansson, Tsur, Roe, Doukkali and Dinar, 2002) and urban water pricing (e.g., Hirshleifer, DeHaven and Milliman, 1960; Howe and Linaweaver, 1967; Hall and Hanemann, 1996; Espey, Espey and Shaw, 1997; Saleth and Dinar, 2001; Renzetti, 2002) provide a basis for further discussion. There are strings of literature that are related to the way the prices are determined with respect to the social and environmental costs and benefits of water development and use (Baumol, 1972; Young and Haveman, 1985; Tietenberg, 1988; Baumol and Oates, 1995). And there is another set of literature relying on contingent valuation techniques to assesss willingness to pay by various consumer groups (e.g., Whittington, 1992; Griffin, Briscoe, Singh, Ramasubban and Bhatia, 1995). In a comparative study of water pricing effectiveness, Cummings and Nercissiantz (1992) in Mexico and South Western United States, and Griffin and Perry (1990) point out the advantages and disadvantages associated with pricing policy. Another work that attempts to address the consequences of federal water policy reform, including surcharges, allocations and institutional reforms describes the impacts of the 1992 Central Valley Project Improvement Act (CVPIA) in California, which addresses agriculture, 8.00 7.00

$/cubic meter

6.00 5.00 4.00 3.00 2.00 1.00 0.00 2500 5000 7500 10000 12500 15000 0 Available water per capita [cubic meter per year] Source: Based on Dinar (2000b).

Figure 1.2 Average price for residential water by per capita water availability

16

The international yearbook of environmental and resource economics

urban and environment (Weinberg, 2002). The work concludes that price reforms that are based on full-cost prices do little to motivate behavioral changes, but will be effective in raising funds to finance environmental restoration policies. Weinberg (2002) reaches a similar finding to Moore (1999) in that tiered water prices as a policy tool will be effective only when prices are raised above the shadow values of the constrained water resources. This finding applies also to single-rate pricing policy. The rationale behind it is that users will not change their behavior unless the price exceeds the value they assign to the last unit of water that is reflected in the calculated shadow value. In evaluating and comparing experiences with the various methods for pricing water, one could focus on several variables, such as ease of implementation, type of efficiency to be achieved and its time horizon, its ability to affect the demand for water, and the ability to control pollution. Water pricing policies are correlated with energy pricing policies, especially in situations where substitution between water sources (e.g., surface and ground water) is possible. Works that address the intersectoral impact between energy and water pricing include Shah (1993) for the case of irrigation water in Gujarat, India, and Hansen (1996) for the case of urban water supply in Copenhagen, Denmark. Renzetti (2002) and Pint (1999) address institutional issues that create distortions or incentives for pricing to operate appropriately. Pint (1996) uses the case of the prolonged California drought to demonstrate the impact of external shocks on the performance of various conservation measures, especially increased charging rates. A comparison across the various irrigation water pricing methods is provided in Table 1.2. Below we will provide several highlights for each major category of methods, namely volumetric, nonvolumetric and water markets. Volumetric methods Implementation costs associated with volumetric pricing are relatively high and include: price setting, monitoring volumes, and fee collection. Although many think that volumetric measurement is possible only by installing expensive meters, for reasonably constant water flows in the supply system (canal), implicit volumetric pricing is possible by charging according to delivery time, which is much less expensive to monitor (Small and Carruthers, 1991; Bandaragoda, 1998a). An example of a temporal block-pricing method where price varies by crop and season due to unstable flows is found in Maharashtra, India, (Easter et al., 1997). Alongside simplistic volumetric pricing systems one can find also sophisticated monitoring technology with tiered pricing based on a multirate volumetric

17

Tsur et al. (2004).

Second-best Second-best None First-best First-best First-best

Less complicated Less complicated Easy Relatively complicated Relatively complicated Difficult without pre-established institutions

Source:

First-best

Complicated

Volumetric (single rate) Output Input Per area Tiered Two part Water market

Possible efficiency achieved

Implementation

Short-run Short-run n/a Short-run Long-run Short-run/ Long-run

Short-run

Time horizon of efficiency

Comparison of key variables of various pricing methods

Pricing scheme

Table 1.2

Relatively easy Relatively easy Hard Easy Relatively easy Depends on market type

Easy

Ability to affect demand

Small/none Small/none Small/none Moderate Moderate Depends on market type

Small/none

Effect on intrasectoral income distribution

Difficult Difficult Difficult Relatively easy Relatively easy Easy

Easy

Adoptability of water quality conditions

18

The international yearbook of environmental and resource economics

method and two-part tariff pricing based on volumetric marginal-cost pricing plus a fixed admission charge, such as in California (Rao, 1988) and in Israel (Yaron, 1997). Boland and Whittington (2000) review cases of increased block tariffs in the urban sector in developing countries. Non-volumetric methods In the many situations in which volumetric pricing is infeasible, non-volumetric pricing is introduced. Several such pricing methods are common for irrigation: output pricing, input pricing, area pricing, and betterment levy pricing. In the urban sector, residential water charges are based on monthly fees per family, depending on the number of household members and their age. Under output pricing methods, a water fee is charged for each unit of output produced by the user. Except for per area pricing, all the other methods are very complicated in terms of monitoring and are mostly common among small and closed water user associations, where control is much simpler to implement. Input pricing is less complicated to control, depending on the type of input selected for use as the pricing control. Usually it is a fertilizer that can be monitored and also be used for pollution control.15 Area pricing should not be perceived as having low sophistication. The two famous allocation/pricing methods, the Warabandi and Shejpali methods16 in Punjab, Pakistan, and Haryana, India (Narain, 2003), are relatively complex systems that combine elements of volumetric pricing with area pricing (Bandaragoda, 1998a; Perry and Narayanamurthy, 1998). Other methods such as betterment levy pricing are described in Young (1996), Easter (1980), and Easter and Welsch (1986). Market based methods The literature on experience with water markets includes a variety of works, some of which are empirical and some descriptive (Vaux and Howitt, 1984; Rosegrant and Binswanger, 1994; Hearne and Easter, 1997; Tsur, 1997; Wilson, 1997; Easter and Feder, 1998; Easter et al., 1998a and 1998b; Thobani, 1998; Mariño and Kemper, 1999; Pigram, 1999). Water markets are often established informally in cases of scarcity and lack of appropriate institutions and water right allocations (Renfro and Sparling, 1986; Shah, 1993) and when governments fail to respond to rapidly changing water demands (Thobani, 1998). Examples of such informal markets include India (Shah and Zilberman, 1991 and 1995; Saleth, 1996), Pakistan (Bandaragoda, 1998), and Mexico (Thobani, 1998).

Issues in water pricing reforms

19

PRICING EFFECTIVENESS AND COST RECOVERY ALTERNATIVES A vocal debate exists regarding the effectiveness and relevance of pricing, especially as it is related to irrigation water. With all the deficiencies of the volumetric water pricing methods that have been described earlier, some argue that a distinction between pricing of and charging for water has to be made. Critics of the volumetric water pricing assert that a minimal response by users is rather the common phenomenon. In trying to understand the economic rational for such low response, Moore and Dinar (1995) and Moore (1999) investigate the behavioral reasoning that could affect irrigators’ behavior in light of a water price increase. Moore and Dinar (1995) find for the case of California that farmers will respond differently to regulations, depending on the binding resource. If land is relatively scarcer than water, then water quotas will be more effective in sending the water scarcity signal rather than prices. A similar finding, with more general implications is suggested also by the Finkelstein and Kislev (1997). Moore (1999) goes the extra mile to check what could be the price increase that will affect the demand for water by the Bureau of Reclamation districts’ farmers in the Western United States. Not surprisingly, as baseline prices are unrealistically low, the answer is that in most districts the increase in the price of water that will affect farmers’ decisions (that is equate the shadow price of their water supply) is so big that it is politically infeasible to obtain. A recently published survey (Bosworth et al., 2002) draws lessons from the literature and concludes that since prices have to be increased dramatically in order to affect the demand by farmers, and much higher than the cost of service provision, the likelihood of that happening is nil. Thus, to be practical and effective, emphasis should be put on cost recovery to allow at least a proper operation and maintenance of the irrigation system. However, going by all practical experience with cost recovery in developing (Bosworth et al., 2002) as well as developed countries (Massarutto, 2002), cost recovery schemes are also malfunctioning and do not provide the expected results. Massaruto extends his analysis of the role of cost recovery for long-run sustainability of irrigation systems in light of the new European Water Directive (Massaruto, 2004). Using an accounting approach Massaruto offers alternative arrangements for allocation of the cost of irrigation water provision, asserting that as an instrument aimed at the efficient allocation of water resources among users and as an instrument of environmental policy, water pricing can be (moderately) effective but it will hardly be sufficient, calling upon additional tools and cost allocation arrangements. But a more in depth discussion on the

20

The international yearbook of environmental and resource economics

various aspects of the political economy of water pricing is provided in the next section.

POLITICAL ECONOMY OF WATER PRICING REFORMS While pricing theory in general and market approaches to pricing reform in particular are supported by a well-defined behavioral theory based on rationality, full information, and frictionless transaction, they often fail in practice, mainly due to their unrealistic assumptions. For instance, the implementation of water pricing reforms proves difficult because of many distortion constraints (e.g., Easter, Becker and Tsur, 1997; Finkelstein and Kislev, 1997; Thobani, 1997; Spulber and Sabbaghi, 1998; Boland and Whittington, 2000; Renzetti, 2000). How these constraints allow only second-best solutions is shown in the particular context of irrigation evaluated both in partial and general equilibrium settings (Berck, Robinson and Goldman, 1991; Goldin and Roland-Holst, 1995; Diao and Roe, 2000). The difficulties for implementation of water pricing reforms are explained in the literature essentially in terms of the physical, technical, and economic characteristics of water. These characteristics include the public good and open-access nature of water and the attendant phenomena of scale diseconomies, short-term non-renewability, and uncertainty (e.g., Hardin, 1968; Dasgupta and Heal, 1979; Provencher, 1995; Easter, Becker and Tsur, 1997; Vermillion, 1997). That is, in most cases measurement of quantities consumed is impossible and thus free riders are affecting the performance and creating diseconomies of the infrastructure, that by itself is inflexible in responding to needed changes that result from external impacts of high and low supply of water. Besides the problems due to the innate character of water, the implementation costs of pricing mechanisms are also influenced by exogenous factors represented by the physical, institutional, and political environment (e.g., Sampath, 1992; Rosegrant and Binswanger, 1994; Tsur and Dinar, 1997; Renzetti, 2000). Similarly, there are also equally serious design and enforcement implications of incomplete and asymmetric information (e.g., Bos and Walters, 1990; Laffont and Tirole, 1993; Smith and Tsur, 1997; Tsur, 2000), that can create both additional transaction cost for the regulator and inefficiency in the allocation of the resource. It is these factors that explain the variations in water pricing method and structure, cost recovery, allocative functions, environmental and social objectives, and implementation speed and coverage of water reform as observed across different physical, cultural, economic, political, and institutional settings.

Issues in water pricing reforms

21

From the perspective of explaining differential pattern of reform both within and across countries, it is unrealistic to use either the single-player model or social-economic or political aggregates (e.g., Shubik, 1985) or ‘social planner’ models (e.g., Alesina, 1996). Since political pressure can enhance or distort the reform process, political coalitions and power structure (e.g., Olson, 1982; North, 1990b; Eggertsson, 1996) have to be explicitly incorporated into the formal model of water pricing reform. Political dimensions of sectoral reforms could be traced as early as the stage where the reform was originally considered, and as late as the post-reform stage. A framework for explaining difficulties in sector reforms has to take into account the political dimensions of the economy (White, 1990; Krueger, 1992; John Williamson, 1994; Haggard and Webb, 1996). Components may include: reasons for reform (John Williamson and Haggard, 1994), some of which could be understood differently by the parties involved; existing institutions (Olson, 1982; Ostrom, 1987; Bromley, 1989a and b) that could create implementation problems; power system (Israel, 1987; Stalling and Brock, 1993; John Williamson, 1994; Haggard, Lafay and Morrisson, 1995; Eggertsson, 1996; Dinar, 1998; Rausser, 2000), such as castes and feudalism that could raise opposition; electoral system (McCann and Zilberman, 2000) that can block some of the parties from utilizing power granted by the reform; and compensation mechanisms to address negative impacts (e.g., Krueger, 1992; John Williamson, 1994; Haggard and Webb, 1996), that could miss groups or underestimate their level of being impacted. A simple political economy framework to handle implementation of water pricing reforms and explain the possible difficulties that arise from the nature of the existing interests, powers and institutions is suggested in Dinar (2003). Reforms in the water sector are described as pairs of prices and quantities that face different interest groups. Components of the framework include: identification of the affected groups and the reforms, groups’ influence on the reforms, groups’ transaction cost of influencing the reforms’ level of achievement, and the level of achievement of the reforms. A handful of case studies are described, using the framework suggested in Dinar (2003). The case studies cover five countries, both from the developed (Australia) and developing (Brazil, Mexico, Pakistan, Yemen) worlds. The concluding statements from the various cases verify the importance of interaction between the factors mentioned above and the outcome of the reform. The common findings are that desired outcomes are more likely with established clear reform objectives; with prior information on political parameters that lead to the backing of an effective coalition; with a consultative, incentive based and transparent process; with proper sequencing in case of a multistage reform; and with clear linkage between the reform

22

The international yearbook of environmental and resource economics

in the water sector and other macroeconomic reforms that have components with links to the water sector. For example, price reforms in the energy sector should be linked to water pricing reforms. Trade reforms, especially of agricultural crops may impact the agricultural sector and this impact could be enhanced if they are linked to water pricing reforms that lead to similar objectives (Diao and Roe, 2004; Tsur et al., 2004). Other aspects of such linkages between water and other sectoral reforms are discussed in the Institutional Economics section late on. A summary of the political economy structure in recent pricing reforms in various countries is presented in Table 1.3 to demonstrate the complexity of dealing with water pricing reform implementation. Corruption, Influence and Initiation of Pricing Reforms Corruption and group pressure during the stage of water pricing reform design and implementation in the case of Pakistan’s water and drainage sector is a well-studied case with implications for water pricing policies elsewhere (Kijne, 2001; Rinaudo, 2002; Dinar et al., 2004). The conclusions from the three studies, though addressing different levels of stakeholders, suggest that pricing reforms will not always succeed even if they are well designed and implemented. There will always be the hidden issue that creates the tension and obstacles in the implementation. So, there are very few successful pricing reforms, from the point of view of having all the components implemented. Pricing reform that is an acceptable compromise could be preferred in most cases. Unfortunately, not too many examples exist. In a separate global study, Dinar and Subramanian (1998) developed and applied a framework for a Water Pricing Progress Index. Their hypothesis is that countries with a high level of gross national product per capita, high level of water scarcity, and high level of budget deficit will be more likely to implement an aggressive water pricing policy. Of course, the impacts of these variables on the final result reflect the relative impact of each variable, as reflected for example in the case of Canada and Botswana. Canada, a water-abundant country, has made little progress on reform, while Botswana, also with abundant water supplies, has made significant reforms. In a more political economy oriented work, Dinar (1998) and Dinar (2003) address implementation issues associated with pricing reforms. Dinar (1998) analyses the information needs of water pricing reforms and the various obstacles standing in their way and possible bypasses. The policy interventions analysed are: pollution taxes, tradable pollution permits, input taxes, output taxes, subsidies on technologies, production

23

National/State Government/ President

The Council of Australian Governments, State Governments

Flemish Government

Federal Government, State Governments, President

President, Economic Cabinet, Ministers

Country

Australia

Belgium

Brazil

Honduras

Congress, Political Parties

Politicians, Congress

Political parties/ Parliament

Municipality

Municipalities, State Governments

Local Governments

SANAA (Water Agency), CNSSP (Tariff Regulator)

River Basin Authorities, National Water Agency

Water industry, Water utilities

Water agencies

Table 1.3 Stakeholder matrix for major country price reform case studies

Union workers of Water Agencies, users, households not connected to the system, industry

Low-income citizens

Taxpayers, farmers

Individuals

World Bank, IDB

World Bank, IDB

Int’l development agencies

Int’l firms

Environmental NGOs, National Competition Commission, Productivity Commission

Others

24

Dinar (2003).

Government, Imam

Yemen

Source:

Government

Government

Morocco

Senegal

President, Federal Government

Mexico

Prime Minister, Provincial Governments, President

National/State Government/ President

Country

Pakistan

(continued)

Table 1.3

Political Parties, Parliament Members

Parliament

Congress

Tribal leaders

Municipality

Regional Administration

Water agencies

SONES, SDE, Municipal Water Company

WAPDA

ORMVAS

CNA

Local Governments

State Governments, Municipalities

Political parties/ Parliament

Influential farmers, qat growers

Gardeners

Large farm associations, small farm associations

Large and small farmers

Water user associations

Individuals

World Bank, EC, and other donors

World Bank

World Bank, ADB

World Bank

World Bank

Int’l development agencies

Food and nutrition NGOs

Food processing industries

Others

Issues in water pricing reforms

25

quotas, input regulations, water pricing, and tradable water rights. Policy measures associated with precise monitoring (pollution related) are also information-sensitive, and therefore, likely to cost more. Some policy measures (input related) are subject to asymmetric information, and therefore may be associated with enforcement irregularities. Both cases demonstrate the need for prioritizing criteria for information development. Dinar (2003), via a conceptual framework and anecdotal evidence attempts to explain obstacles and successes by combining physical-economic (such as existing infrastructure of the water system) and political (e.g., power structure in the country) considerations that affect the implementation costs of pricing reforms.

INSTITUTIONAL ECONOMICS OF WATER PRICING REFORMS As shown earlier, the economic theory of pricing based on the neoclassical approach sheds light on the theoretical principles and practical problems in designing water rates that would ensure cost recovery and achieve efficiency and equity in the use of water in various settings. The political economy approach provides more insights than the standard neoclassical approach, especially on the role of non-economic aspects during the process of implementing water pricing reforms. But, both approaches fail to account for the role that certain important aspects – both endogenous and exogenous to the process of pricing reform – play in enhancing the cost recovery, allocation, and equity roles of water pricing policies. In order to highlight the nature and the role of these aspects, it is necessary to take an institutional perspective of water pricing. That is, instead of the usual approach of viewing water pricing as an economic and financial instrument, we will view it as an important institutional configuration. In this section, we will outline the rationale and basis for taking such an institutional approach to water pricing and indicate the theoretical and practical advantages of such an approach, especially in understanding the constraints and opportunities for undertaking politically and technically feasible water pricing reforms. Water Pricing Methods as Institutional Configurations An institutional approach to pricing per se, especially in a generic context, is nothing new. For instance, a few studies on administered pricing (e.g., Ware and Means, 1936; Tool, 1991) did take an institutional approach to pricing in a general context. As the neoclassical approach fails to provide a

26

The international yearbook of environmental and resource economics

general theory of price determination, these studies underline the role of institutional factors in the emergence of discretionary and markup price setting behavior that characterizes a vast segment of the industrial and commercial sectors in the modern capitalistic system. Price setting is both normal and has useful economic functions such as the reduction of uncertainty and the creation and adoption of new technologies (Galbraith, 1967: 190).17 Thus, discretionary or administered prices are one of the institutional mechanisms that the society has created to counter the problem caused by the absence of market determined prices in a wide range of economic contexts. Price setting can also be viewed as part of conventions and habitual patterns.18 Given the institutional character of pricing, as opposed to its market character, what is central is the institutional configuration through which price setting occurs and gets perpetuated (Tool, 1991: 24). The core of the institutional configuration that makes price setting possible and unavoidable in the modern capitalist system is the legal, policy, and organizational arrangements underlying the megacorporations with dispersed ownership and concentrated control (Veblen, 1904; Berle and Means, 1932). Unfortunately, the rich insights from the literature taking an institutional approach to pricing have not been extended to other contexts such as water pricing. As a result, not only is a major opportunity for theoretical development missed but also some extensions and refinements possible to such an approach are unexplored.19 Although a comprehensive exploration of the institutional approach to water pricing is conspicuous by its absence, a few studies have taken such an approach in a rudimentary form. For instance, Saleth (1996) argued that even with higher levels of water rates, the economic role of water pricing cannot be expected unless the institutional arrangements such as the legal system of water quotas and the organizational basis of user associations are in place. Nagaraj (1999) has explained the differential performance of the water pricing and cost recovery systems in India and France essentially in terms of the presence or absence of important organizational arrangements. Beyond these attempts, there is a lamentable dearth of attention to the theoretical and practical implications of taking an institutional perspective of water pricing in the existing literature. The logic of our institutional perspective on water pricing lies in the fact that the levels and methods of water pricing can be interpreted as forms of ‘rules’ to determine the payment for and use of the resource in different contexts. These rules do not exist in isolation but are structurally and operationally linked with other water related legal, policy, and organizational or administrative rules or mechanisms. In a recent study, Saleth and Dinar (2004b) have demonstrated such linkages in terms of the ‘Institutional

Issues in water pricing reforms

27

Ecology’ principle20 using the ‘Institutional Decomposition and Analysis’ (IDA) framework that decomposes water institutions in terms of individual ‘rules’ and ‘rule configurations’ to trace and highlight their linkages. This approach, which is relevant to all institutions and their components, is helpful in understanding water pricing as an institutional configuration. The nested and embedded nature of water pricing means that there are upstream, downstream, and lateral linkages between water pricing and other institutional components. It is in view of these linkages that water pricing reform does not end just with changing the level and method of water rates but encompasses concurrent changes in all related legal, policy, and organizational components. Let us see below what these concurrent changes are and how they can be promoted effectively whilst minimizing social and political costs. Once we consider water pricing as a key part of water institutions, then it is straightforward to view different methods of water pricing as different institutional configurations involving various combinations of legal, policy, and organizational aspects. For the purpose of illustrating how different methods of water pricing (Table 1.1) involve various institutional configurations, let us consider just the two broad categories, i.e., volumetric and non-volumetric methods of water pricing.21 Let us note first that the institutional and technical requirements of non-volumetric methods are relatively simple as compared to those of the volumetric methods. It has already been noted, for instance, that non-volumetric methods such as the area based water rates are easier to administer as they have a lower information and monitoring requirement. In contrast, volumetric methods require both accurate information on changing water values as well as strong enforcement and monitoring arrangements. Such differential informational and institutional requirements of the two categories of pricing methods need to be elaborated further. The choice of water pricing methods cannot be made in an economic, institutional and infrastructural vacuum. Since the water allocation pattern under the area based method reflects the prevailing land tenure, the legal institution of land rights, the physical location of land parcels, and a centralized bureaucracy with a thin spatial spread are enough to determine water allocation and collect water rates.22 Unfortunately, this method is unavoidable under conditions involving large surface irrigation systems with traditional water conveyance systems, large number of tiny landholders, and a centralized bureaucracy often lacking technical information, enforcement mechanisms, and qualified personnel, especially at its lower echelons. The volumetric method, on the other hand, has high technical and institutional requirements. For instance, it requires the infrastructure of a modern water conveyance system conducive to measuring water

28

The international yearbook of environmental and resource economics

deliveries, a legalized or locally sanctioned system of water rights or water quotas to determine individual allocation, and a decentralized organization for monitoring, enforcement, and fee collection. While land tenure institutions may have an effect in determining initial water quotas, the addition of the institutional component of a water rights system can ensure economic incentives and minimize allocation distortions both at the use and exchange stages. The other institutional requirement of a decentralized and use based organization can be met either with autonomous local entities (e.g., irrigation districts and private water companies) or with water user organizations.23 Although we have illustrated the institutional differences by considering only two water pricing methods, similar differences evident among other methods listed in Table 1.2 can also be easily understood. With the interpretation of water pricing as a major component of water institutions, it is possible now to distinguish its institutional arrangements (or governing structure) from its institutional environment (or governance framework) (Oliver Williamson, 1985 and 1993; North, 1990a). What we have described as the institutional configuration of water pricing is actually the institutional arrangement underlying respective water pricing methods. As we can see, the institutional arrangements relate to the endogenous linkages that different water pricing methods have with other legal, policy, organizational, infrastructural, and technical aspects of water institutions. In contrast, the institutional environment captures the linkages that different water pricing methods have with factors exogenous to their institutional arrangements. These exogenous factors cover the overall physical, social, religious, economic and political factors affecting water pricing arrangements in particular and water institution and the water sector in general. They include constitution, land tenure, population settlements, quality of civil administration, and other sectoral policies ranging from agricultural and trade policies to food self-sufficiency and national security. Besides, macroeconomic reform, political changes including administrative decentralization, bilateral and multilateral agreements, and even natural factors such as droughts can also form part of the institutional environment of water pricing.24 From our illustration of the institutional differences among the pricing methods, it is clear that the institutional arrangements vary among the pricing methods, whereas the institutional environment facing them remains, more or less, common to all. Nevertheless, it is important to recognize that changes in the institutional environment will have differential implications for different pricing methods. It is in view of these differential impacts and variations in institutional configurations that the transaction costs associated with different pricing methods vary considerably.

Issues in water pricing reforms

29

Transactions Costs of Water Pricing Reform Transaction costs in the context of water pricing reform relate to the spectrum costs involved in changing the level and method of pricing water. This includes not only the monetary costs of creating the necessary legal, policy, and organizational conditions essential for the effectiveness of the preferred level and method of water pricing but also the real costs in terms of time loss and political risks. As we will see a little later, there are also costs in terms of the redundancy of institutional and organizational arrangements related to the previous pricing methods. For a better understanding of the transaction costs of water pricing reform in particular and water institutional reform in general, it is essential to clarify a number of ambiguities prevalent in current literature. Let us first identify and explain these ambiguities and then provide practical instances and indicate their implications for water pricing reform. First, the transaction costs of alternative pricing methods are usually considered to be exogenously determined and, therefore, are to be taken as given. This perspective clearly misses the fact that the transaction costs can be minimized with suitable institutional design and implementation principles that exploit well the linkages within institutional arrangements and synergy from the institutional environment. For instance, transaction costs of reforms can be minimized by adopting appropriate prioritization, sequencing, and packaging of institutional components as well as with a proper choice of reform timing, spacing, and scale (e.g., White, 1990; John Williamson, 1994; Savedoff and Spiller, 1999; Dinar, 2000b; Saleth and Dinar, 2004a). Second, the transaction costs have to be reckoned not just in economic terms but equally also in real and political terms. For this purpose, the generalized form of the institutional transaction cost approach (Guttman, 1982; North, 1990b; Milgrom and Roberts, 1992; Dixit, 1996; Saleth and Dinar, 2004a and 2004b) is needed. Since this approach generalizes the original transaction-cost approach of Coase (1937) and Oliver Williamson (1985 and 1993) so as to allow the transaction-cost implications of the role of non-economic factors, it provides a unified framework for addressing both the political economy and institutional economics aspects of water pricing reform. Third, it is very important to consider the time dimensions of transaction costs, as they evince vast differences in terms of their short-run and long-run implications. For instance, the transaction costs of switching from the institutional configuration underlying the area based method of pricing irrigation water to that behind the volumetric method will be higher when reckoned from a short-run and static perspective as compared to the same

30

The international yearbook of environmental and resource economics

in the long-run and dynamic context. Since the institutional configuration underlying the volumetric pricing method enables volumetric allocation, it provides the scope for the emergence of next-level institutions such as water markets and water banks. Thus, from a larger and long-term perspective, its transaction costs tend to decline because the transaction costs of creating the volumetric allocation reduce the transaction costs of creating water markets and water banks. This particular case also demonstrates both the linkages between the transaction costs of earlier and subsequent institutional reform components as well as the scope for scale economies in institutional reforms. And, finally, a switch from one pricing method to another, as effected by a pricing reform, can not only create or modify existing institutional or technical arrangements but can also replace or reduce the role of the bureaucratic mechanisms that were involved in the earlier pricing and allocation regime. This fact points to an aspect that is often overlooked in the reckoning of institutional transaction costs, i.e., the costs of administrative and bureaucratic redundancy. That is, as we shift from, say, area based methods to volumetric methods in pricing irrigation water, besides the additional costs in creating the institutional and technical conditions for volumetric allocation, there are also indirect costs due to the redundancy or underutilization of organizational mechanisms and procedures relied upon under the earlier method of water pricing. Since these costs are substantial and assume the form of dead-weight loss, they often delay or constrain the political decisions on pricing reforms. Implied in this redundancy issue is also the role of path-dependency effects created by the network externalities associated with the replaced pricing method. Most of these factors are not captured either by an economic or political economy approach to water pricing. This is why an institutional approach that reckons the full transaction-cost implications of both the institutional and political factors within a common framework is necessary to explain institutional reform in general and water pricing reform in particular. How to Promote Water-Pricing Reform? The economic theory of pricing does help us in designing the prices that could achieve the cost recovery, efficiency, and equity objectives. Although such prices assume certain institutional requirements, the neoclassical approach has an inherent limitation in explicitly incorporating these institutional prerequisites within its analytical framework. As per the original transaction-cost theory, it is also naively presumed that when the social benefits of the alternative pricing scheme are higher than the transaction costs, there will be an economic incentive for adopting the scheme. As the

Issues in water pricing reforms

31

political economy approach reveals, such incentive is not automatic in view of the role of interest group politics, political bargaining, and rent seeking considerations. Similarly, there is also an incomplete reckoning of relevant transaction costs because the transaction-cost implications of political risks, path dependency aspects, and institutional linkages are often underestimated. On the other hand, the political economy approach, though it sheds light on the political and economic process through which a given reform program emerges, ignores, however, the fact that the political economy calculus is not independent of the reform process and institutional change. Thus, the political economy calculus of groups does not remain stable but can change through the course of the reform. Similarly, the political economy approach is also not accounting for the role of institutional linkages. For instance, it is not clear how the transaction-cost minimizing role of institutional linkages or the transaction-cost enhancing role of institutional/organizational redundancy can be handled within the contexts of interest group politics, political bargaining, or rent seeking behavior. Since neither a purely economic nor a political economy approach to pricing reform is adequate, there is need to supplement them with the institutional approach. The rationale for institutional approach is clear. The institutional approach, unlike the neoclassical approach, actually endogenizes the institutional configurations while making the choice on the level and method of pricing within a pricing reform program. Similarly, the institutional approach, unlike the political economy approach, goes beyond simple rent seeking and interest group politics by highlighting the role of internal linkages among institutional aspects as well as by bringing various economic, institutional, and political factors within the institutional transaction-cost framework. In this sense, the institutional approach is more comprehensive in bringing all relevant economic, political, and institutional factors within a common analytical framework. It is on this logic that we will argue below that an institutional approach can enable us to derive strategies for promoting water pricing reforms that are economically desirable, institutionally feasible, and politically acceptable. As we will show, these strategies are, in fact, developed by exploiting the endogenous features of and the exogenous influences on the institutional configurations underlying alternative water pricing methods. Before dealing with the reform strategies to be derived from an institutional approach, let us first recognize the fact that some of the features of generic institutions apply equally to water pricing institutions. These institutional features include the nestedness and embeddedness (North, 1990a; Ostrom, Gardner and Walker, 1994; Boyer and Hollingsworth, 1997), path dependency (North, 1990a; David, 1994; Coriat and Dosi,

32

The international yearbook of environmental and resource economics

1998), synergy effects from institutional linkages and institutional environment (Boyer and Hollingsworth, 1997; Saleth and Dinar, 2004a and 2004b), and scale economy benefits from institutional linkages (North, 1990a; Saleth and Dinar, 2004a and 2004b). The nested and embedded feature means that various components within the institutional configuration underlying different water pricing methods are interrelated with considerable implications for transaction costs. For instance, volumetric methods are critically linked with the water rights or water quota system, accuracy and efficiency of water delivery networks, and a decentralized and use based form of organizational arrangements. Similarly, the operation of water markets depends on volumetric methods and their institutional configurations. These upstream and downstream linkages among institutional components remain a source of synergetic effects, which can enhance the mutual strength and performance of these institutional components. The path-dependency features imply that there is a particular sequencing among the institutional components. For instance, volumetric allocation with a water quota system should precede the emergence of water markets, but not the other way around. Similarly, the scale economy feature suggests both the scope for long-run decline in transaction costs as well as the need for institutional packaging or combining related institutional components within a given reform program. For instance, as we have argued already, although the transaction costs of establishing a volumetric allocation system are high from a short-run perspective, in the long run such costs can decline, as this system also promotes the additional institutions such as water markets and water banks with no or less additional costs. This scale economy effect in institutional reform also suggests that there are fundamental linkages between the transaction costs of earlier reforms (e.g., canal modernization conducive to introducing a volumetric allocation system, irrigation turnover to user organizations, and the introduction of a water quota system) and those of the subsequent reforms (i.e., water markets, local level conflict resolution arrangements, and water withdrawal restrictions such as those under the Cap program in the Murray-Darling Basin of Australia). The synergy effects not only exist among water pricing institutions but also flow from the exogenous influences emerging from their institutional environment. Since the institutional environment of water pricing institutions captures the effects of exogenous and non-economic aspects ranging from the physical setting to cultural contexts (Douglas, 1986; Grief, 1989; Ostrom, 1990; Ostrom, Gardner and Walker, 1994; Schmitter, 1997), its synergetic or discordant effects can also be beneficially utilized to provide the context and thrust for water pricing reforms. For instance, macroeconomic reforms, international or bilateral economic, trade, and water agreements, and extreme events such

Issues in water pricing reforms

33

as droughts can provide an appropriate context for undertaking water pricing reforms in particular and institutional reforms in general. Such factors show that although political economy factors are considered as constraints for institutional change, they also provide reform opportunities, as the political economy context associated with these factors can reduce the political transaction costs. Since water pricing reform is nothing but a special case of institutional change, the endogenous institutional features and the effects of the exogenous factors can be exploited to develop a number of institutional design and implementation principles that can overcome both the political economy and institutional constraints on water pricing reform. As we have noted already, these principles include institutional prioritization, sequencing, and packaging as well as reform scale, timing, spacing, and pacing. Although some of the principles are also recognized in the political economy literature, they are interpreted here essentially in terms of their implications for institutional transaction cost (Ostrom, Gardner and Walker, 1994; Ruttan, 1999; Saleth and Dinar, 2004a). For instance, whereas the political economy approach looks at institutional packaging as a means for creating pro-reform coalitions (e.g., White, 1990; Haggard and Webb, 1996), the institutional approach will underline its role in exploiting institutional synergy and minimizing, thereby, the overall transaction cost of institutional reform (Saleth and Dinar, 2004a). Similarly, the supplydemand perspective of institutional change (e.g., Ruttan and Hyami, 1978; Feeny, 1993) can also be interpreted as a form of institutional prioritization and sequencing as well as reform spacing. That is, as long as institutional prioritization is guided, inter alia, by the extent of reform pressure that different institutional components can generate on their upstream and downstream counterparts, those components are supplied at suitable temporal and institutional intervals to generate an endogenous demand for the creation of other institutional components linked with them. Let us now consider the implications of these issues from the particular perspective of the developing countries. Water Pricing Reforms in Developing Countries The effectiveness and practical use of the institutional design and implementation principles are illustrated with empirical, anecdotal, and case study based evidences for the general case of water institutional reforms undertaken recently in several countries (e.g., Saleth and Dinar, 2000, 2004a and 2004b). Similar evidences are also available for the particular context of water pricing reforms (Dinar and Subramanian, 1998; Dinar, 2000a and 2000b). Here, we will provide additional instances to illustrate

34

The international yearbook of environmental and resource economics

how these principles can be used to advance water pricing reform, especially in developing countries. The clear linkages between water rights and pricing policies are well-recognized in the literature (e.g., Randall, 1981; Bromley, 2000). While water prices are associated with some element of subsidy, they are relatively close to their economic level mostly in advanced countries with land tenure patterns, water rights systems, water delivery networks, and local organizational arrangements conducive to volumetric allocation. Since water rights and volumetric allocation add the muchneeded exclusivity and rivalry properties to water, the legal, organizational, and technical conditions underlying such allocation systems contribute to the commoditization of water. The economic character ensured by these physical attributes of water ensures an efficient pricing and allocation. Obviously, the institutional and technical conditions for a volumetric pricing system cannot be created overnight in most developing countries. In these cases, there should be a long-term strategy for switching from the present inefficient pricing system to an improved system based on volumetric allocation. It should be gradual by first rationalizing the levels of water rates to achieve the immediate objective of full cost recovery (operation and maintenance costs plus capital costs of irrigation investment) and then going for economic pricing (scarcity value of water). Even for moving into this second stage of pricing reform, prioritization should be on the creation of the necessary institutional and technical conditions for volumetric allocation such as system modernization to improve delivery networks and management decentralization to create the organizational basis for local water allocation and cost recovery. This kind of sequencing is likely to minimize the overall long-term transaction costs as well as generate the demand for further institutional changes. System improvement and management decentralization can also be packaged within pricing reform as has happened in countries such as Mexico, India, and China where the irrigation management transfer occurred along with system improvement and water rate revisions. Although such improvements are not enough to allow volumetric allocation to individual users, they can facilitate initially the adoption of group based volumetric allocation. Moreover, the multitiered water user associations developed throughout the irrigation systems can certainly lay down the organizational basis for the eventual emergence of an individual water allocation and volumetric pricing system. Research evidences from several countries support the fact that the involvement of user associations has led to a significant improvement both in cost recovery and system maintenance (e.g., Johnson, 1997; Vermillion, 1997). Other forms of reform packaging in the context of water pricing include the practice of combining a water rate hike with improvement in

Issues in water pricing reforms

35

water services. The obvious rationale for this reform aspect is that users are willing to pay higher prices when there is a tangible improvement in service quality (e.g., Whittington, Laurai and Mu, 1991; World Bank Water Demand Research Team, 1993). While this practice is more feasible in urban supply contexts, the general principle can certainly be extended with suitable adjustments to irrigation sectors in developing countries. From a general perspective though, irrigation water rates can be increased with least opposition provided there are corresponding increases in the agricultural prices or improvement in the provision of complementary farm inputs such as credit, fertilizer, and extension services. In countries such as Mexico and Sri Lanka, there are efforts to link water user associations with input supply and marketing arrangements with a view to strengthening the interface between water and agricultural institutions and improving thereby the economic viability and long-term sustainability of the institutional reform in the irrigation sector. The other most important form of institutional packaging observed especially in the urban sector involves the combining of pricing reforms with privatization or corporatization of urban water supply systems. This practice is observed widely in many countries, especially those in Latin America (Spulber and Sabbghi, 1998; Savedoff and Spiller, 1999). Instances of the adoption of implementation principles such as reform timing include the use of macroeconomic reforms (e.g., India, Mexico, and China), international treaties such as those of the World Trade Organization (e.g., many countries) and the Water Directive of the European Community (e.g., most European countries), and natural events such as droughts (e.g., India and California) for providing the right political context for initiating water pricing reforms in particular and institutional reform in general (Saleth and Dinar, 2000). When reform is timed in such an environment characterized by economic, political, and resource related compulsions, the institutional transaction costs, as reckoned from a social perspective, will be lower due to a lower political opposition and hence, lower political transaction costs.25 In countries such as South Africa, there is a regular drought cycle that can be used to time and space various institutional reforms including pricing reforms (Backeberg, 2003). Similarly, instances of the use of reform scale include the gradual vs big-bang approaches. The choice between these approaches is based on considerations such as the sectoral context, administrative will and implementation capacity, and the time taken for political opposition to materialize. The scale aspect also involves the issue of designing pricing reform as a part of a larger reform program for the whole sector or country. The case of South Africa illustrates how water rights and water pricing reform have been a key part of water sector reform and how the latter has been a key component of the countrywide

36

The international yearbook of environmental and resource economics

economic and political reconstruction programs (Backeberg, 2003). In such cases, the transaction costs of the smaller program will be lower due to the scale economy effects in the institutional transaction costs of the larger program. Implications for Theory and Policy Theoretical research can be close to practical policy and vice versa, provided that political economy and institutional economic aspects are integrated with the economic and technical aspects of water pricing. Since most of the social and political aspects form part of the institutional environment, the political economy approach can deal mostly with the exogenous effects. In contrast, the institutional economics approach, especially in terms of the institutional transaction-cost framework, can deal both with the exogenous effects (e.g., economic crisis, reforms in other sectors) as well as the endogenous aspects of water pricing reform (e.g., transactioncost implications of institutional linkages, path-dependency features, and reform design aspects). In this sense, the institutional economics approach ensures that the choice of the level and method of water pricing is based on all relevant economic, political, and institutional factors. As such, the institutional economics approach provides a more general framework for explaining water pricing reform not just in terms of its success or failure but also in terms of the extent of its success and failure in the sense of how good or how bad it is compared with the original objectives (Crisp and Kelly, 1999). Given the seriousness of the informational, technical, political, and institutional constraints, water pricing reform is likely to involve the creation of institutionally diverse pricing arrangements involving public, private, and collective initiatives. Diversity of pricing levels and methods will coexist with institutional pluralism as do the first-best situation with the second and third-best situations. A realistic reform strategy involves a gradual and incremental approach with an appropriate sequencing of reform components, water sub-sectors, and geographic units, focusing first on setting right the cost recovery role and gradually broadening the scope to set the right institutional arrangements necessary for performing the allocating and distributional roles of water pricing. While full cost recovery water pricing has to be achieved in most, if not all, water subsectors or regions, economic pricing of water can be achieved in settings characterized by volume based allocation such as urban areas and irrigation schemes with controlled water deliveries. In the case of rural water supplies, especially in developing regions, the implementation of full-cost pricing has to be accompanied by other group-specific programs such as ‘water stamps or vouchers’ to insulate the poor and vulnerable groups.

Issues in water pricing reforms

37

GAPS IN THE LITERATURE AND FUTURE WORK NEEDED There are several specific theoretical and technical gaps in the pricing literature, as well in the literature on institutional economics of pricing reforms. These include: pricing under imperfect information or asymmetry of information (Tsur, 2000); water pricing implications of water markets (Howe et al., 1986; Young, 1986; Hall and Hanemann, 1996; Kloezen, 1998); equity and poverty (Dasgupta and Heal, 1979; Dasgupta, 1993a and 1993b; Gómez-Lobo and Contreras, 2003); non-conclusive nature of average and marginal cost pricing (Saleth and Dinar, 2001) and short-run and long-run pricing (Dasgupta and Heal, 1979); and intersectoral and micro-macro linkages in water pricing (e.g., Berck, Robinson and Goldman, 1991; Goldin and Ronald-Holst, 1995; Diao and Roe, 2000; Tsur et al., 2004b). The latter is becoming an increasingly important issue in policy analysis as many recognize the important role of water in the economy and its links to multisector issues. Use of computerized general equilibrium (CGE) models for this purpose has been demonstrated to be useful, but surely, additional work is needed to account for more intersectoral and interregional externalities that may result from use of water. While the need to focus on the technical aspects of water pricing is more or less defined in terms of needs and possible direction, the literature is much more in despair vis-à-vis the political and institutional economy of water pricing. Furthermore, most works to date are in the form of case studies. What is missing in our opinion is a direction of work that develops analytical models to deal with the interest groups and their relative power (e.g., Peltzman, 1976; Becker, 1983). Research should focus on several theoretical issues, including (a) defining and measuring the extent of water reforms to allow better comparison among reform attempts; (b) defining and measuring the achievement of reform objectives to allow a comparison among completed reforms; and (c) defining status quo conditions and their impact on reform implementation, such as institutional setup, power structure, and physical conditions, to allow better packaging and adjustment of reform interventions to initial conditions in the country. Additional developments needed in the field of the political economy of water pricing reforms include the development of game-theory based quantitative modeling of power structures and the physical conditions and quantitative approaches to group-specific impact of price reform (e.g., Crisp and Kelly, 1999). With the overcasting shadow of climate change, attention should be given to aspects of stochastic water supply patterns that could have devastating impacts on the financial structure of water supply companies, as well as on stability of irrigation water user associations. As most water pricing

38

The international yearbook of environmental and resource economics

arrangements are based on deterministic supply (rainfall) patterns they fail to address situations where the amount of available water falls below the volume that allows the operational costs of the supplier to be covered. Incorporating stochastic water supply patterns into pricing reforms could be an important contribution. But the most important concern continues to be the literature–policy gap, which, as outlined in this chapter, is very serious because it blocks both the development of water pricing theory and its practical application. While this chapter argues that this gap can be bridged by integrating the three sets of literature (pricing theory, political economy of reform, and institutional economics) that are developing, more or less, in isolation from each other, further works – both theoretical and empirical (case studies) – are needed to refine and extend the institutional transactioncost approach to water pricing reform. Analytical and empirical studies that could integrate these three approaches have the greatest potential to both advance and integrate the theory and policy in the realm of water pricing reform in particular and water institutional reform in general.

NOTES * 1. 2. 3.

4.

5. 6.

The views expressed in this chapter are those of the authors and should not be attributed to the World Bank. ‘Principle 4: Water has an economic value in all its competing uses and should be recognized as an economic good.’ ‘Principle 1: Human beings are at the centre of concerns for sustainable development. They are entitled to a healthy and productive life in harmony with nature.’ The difference between charging and pricing lies in the objectives policymakers try to achieve. While ‘pricing’ basically refers to a mechanism for signaling the economic (scarcity) value of the resource to users so that they allocate and use the resource efficiently, ‘charging’ refers to a mechanism for recovering the financial costs of providing the water with the aim of balancing the budget. A more detailed discussion on this is reserved for a latter section of this chapter. Water availability per capita is the ratio between the amount of water from all sources that can be developed and the population of the country. This is a crude measure of water availability. It does not take into account quality (that is a big factor in making water usable for various purposes), nor the development cost that sometimes makes water resources prohibitively expensive to develop and thus they remain not available. In addition, this index does not take into account regional differences of availability within a country. However, this index is widely used for providing a first comparison of cross-country water scarcity. Water use efficiency can be measured in several ways, but we will use here the economic definition of Pareto efficiency. For further reading (focused on irrigation), please see Burt et al. (1997), Wichelns (1999) and World Bank (2003, Annex 1). ‘Many countries face multiple concerns regarding the growing scarcity of water, the associated conflicts among users, and ways of transferring water from low-value to highvalue uses. It has often been stated that having users pay “the full cost of water” would solve these problems. Experience has shown that the situation is considerably more complex and nuanced, and that it is not enough to just extol the virtues of pricing. . . . [A] different approach . . . [proposes] economic principles such as ensuring that users

Issues in water pricing reforms

7.

8.

9. 10.

11. 12. 13. 14. 15. 16.

17.

18.

39

take financial and resource costs into account when using water, are very important. And, . . . solutions [that] to be tailored to specific, widely varying natural, cultural, economic and political circumstances, in which the art of reform is the art of the possible.’ The ‘Residual Approach’ (e.g., Young, 1996) is a method for assigning a value to a factor of production (input), especially where the market value (or price) for that input does not exist. Water is such input in the production process of agricultural goods. After accounting for all costs of purchasing other inputs, the difference between the revenue of selling the product and the calculated cost of production – the residual – is assigned to the input for which no market price exists. Dividing this residual by the volume of water used provides a value per unit of water. The debate on whether what is an economic good or a social good will probably continue to be on the agenda of various groups and disciplines associated with water policy. The main argument in the basis of water being an economic good is that water is a scarce resource and as such any decision of allocation of water leads to alternative cost for society and thus has to be taken into account. On the other side of the spectrum, those who argue that water is a social good depart from the point that the costs and benefits associated with changes in the allocation of water should be equitably distributed among affected segments of society. Since such groups do not believe that economic tools can take care of equitable distribution, they propose the introduction of various safety nets such as ‘lifeline’ quantities to be allocated to poor users at no cost. A similar consideration was actually introduced in the recent water reform in South Africa in both irrigation and rural water supply sectors (Tsur et al., 2004), and in the urban water supply in Chile (Rivera, 1996). The literature on water demand functions is by itself a subject for a separate review. It will not be discussed here. For several lead literature reviews see for example Saleth and Dinar (2000) and Espey and Espey (1999). The latter requirement necessitates an appropriate infrastructure that allows the regulator to monitor the individual usage (e.g., water meters). Another possibility (in irrigation) is to use proxies for water quantity by correlating it to better-observed factors of production (e.g., use of other inputs that are correlated with water such as fertilizers, or fuel used for operation of water pumps, or calculation of water quantity using measures of water flow and time of flow). See discussion in the Political Economy and the Institutional Economics sections later on. It should be mentioned that the latter considerations have been raised also in regard to irrigation water, but the literature has been approaching the subject mainly from an equity point of view (Anand, 2001b; Lampietti et al., 2001). This will allow the utility to operate under stochastic demand situation, leaving the customers with the responsibility to pay at least the fixed charge independent on the amount of water consumed. This is called ‘the expansion effect’ of water-efficient technology adoption (Dinar and Zilberman, 1991). Fertilizers are purchased in the market or provided by government agencies where records can be kept, while water usage is much more difficult to monitor. The Warabandi method is a rotational system that allocates water by time in a rotational manner to irrigators. The Shejpali method is a demand based system where farmers submit their requests for water at the beginning of the season, based of land type and crop mix. For further explanation of the features of the two systems see Narain (2003). This point is made much clearer by Kregel (1980: 40) as: ‘The information required for rational decision making does not exist; the market mechanism cannot provide it. But, just as nature abhors a vacuum, the economic system abhors uncertainty. The system reacts to the absence of the information the market cannot provide by creating uncertainty-reducing institutions: wage contracts, debt contracts, supply agreements, administered prices [and] trading agreements.’ Shackle (1972: 227) provides the rationale for this as follows: ‘Prices that stood at particular levels for some time acquire thereby some sanction and authority. They are the

40

19.

20.

21.

22. 23.

24.

25.

The international yearbook of environmental and resource economics “right” and even the “just” prices. But, also they are the prices which the society has adapted as its ways and habits, they are prices which mutually cohere in an established frame of social life.’ For instance, the studies discussed above assume that the institutional configurations (e.g., megacoporations) underlying the price setting behavior are here to stay and hence, cannot be changed. However, in other contexts such as the water sector, these configurations can be changed through carefully designed and implemented institutional reforms. This principle extends the ‘ecosystem’ concept to institutional systems so as to analytically show (a) the linkages and synergies among institutions across institutional domains (law, policy, and organization), resource spheres (water, land, and environment), and geographic scales (national, basin, system, and outlet) and (b) the nested and embedded character of institutions within the social, economic, political, and resource systems. We can also note that with certain additional conditions, the categories of volumetric and non-volumetric methods can also correspond to market based and bureaucratically administered water pricing methods. For instance, with the legal and technical conditions of water rights (or, water quotas) and their transferability and the emergence of voluntary water exchanges among users, the volumetric method can evolve into a market based method. Similarly, with the creation of an independent water pricing body within water administration and the periodic revision in water rates, non-volumetric methods can become administered pricing methods. We can note that the administered pricing of water will also be interpreted as a case of a substitute for missing institution (Adelman, Taft-Morris, Fetini and Golan-Hardy, 1992). In view of its underlying institutional configuration, the area based method distorts the economic incentives for water use efficiency and also gets distorted by economically and politically influential water users (Rinaudo, 2000). It is necessary to note that the larger role of decentralized organizational mechanisms under volumetric methods does not mean the centralized and macro-level organizations are unimportant, as the latter are needed for maintaining the legal and regulatory apparatus as well as for providing the technical and informational support. Since the institutional economic concepts relate mainly to institutional arrangement and the political economy concepts relate mostly to institutional environment, the former will be interpreted as endogenous and the latter will be interpreted as exogenous factors affecting water pricing reform. In fact, factors exogenous to the water sector have provided the immediate prompt for the water institutional reforms undertaken in recent years by most countries including the developed countries such as the US (California) and Australia (Saleth and Dinar, 2000). This fact shows how important can be the choice of time for initiating reforms.

REFERENCES Abu Qdais, H.A. and H.I. Al Nassay (2001), ‘Effect of pricing policy on water conservation: a case study’, Water Policy, 3, 207–14. Abu-Zeid, M. (2001), ‘Water pricing in irrigated agriculture’, Water Resources Development, 17(4), 527–38. Adelman, Irma, Cynthia Taft Morris, Habib Fetini and Elise Golan-Hardy (1992), ‘Institutional change, economic development, and the environment’, Ambio, 21(1), 106–10. Ahmad, Mahmood (2000), ‘Water pricing and market in the Near East: policy issues and options’, Water Policy, 2, 229–42. Alesina, Alberto (1996), ‘Political models for macroeconomic policy and fiscal

Issues in water pricing reforms

41

reforms’, in Stephen Haggard and Steven B. Webb (eds), Voting for Reform, New York: Oxford University Press, pp. 37–60. Anand, P.B. (2001a), Water ‘Scarcity’ in Chennai, India: 39, Helsinki: United Nations University, World Institute for Development Economics Research. Anand, P.B. (2001b), Consumer Preference for Water Supply: 37, Helsinki: United Nations University, World Institute for Development Economics Research. Arriaza, Manuel, Jose A. Gomez-Limon and Martin Upton (2002), ‘Local water markets for irrigation in southern Spain: multi-criteria approach’, The Australian Journal of Agricultural and Resource Economics, 46(1), 1–23. Backeberg, Gerhard (2003), ‘Progress in institutional reforms in the water Sector of South Africa’, paper presented in the workshop Water Reforms, Institutions’ Performance, Allocation, Pricing, and Resource Accounting, 25th Congress of the International Association of Agricultural Economists, Durban, South Africa, 16 August 2003. Bandaragoda, D.J. (1998), ‘Design and practice of water allocation rules: lessons from Warabandi in Pakistan’s Punjab’, research report no. 17, Colombo, Sri Lanka, International Irrigation Management Institute. Baumann, D., J. Boland and M.W. Hanemann (1998), Urban Water Demand Management and Planning, New York: McGraw-Hill. Baumol, William J. (1972), ‘On taxation and the control of externalities’, American Economic Review, 54, 307–21. Baumol, William J. and Wallace E. Oates (1995), The Theory of Environmental Policy, New York: Cambridge University Press. Becker, Gary S. (1983), ‘A theory of competition among pressure groups for political influence’, Quarterly Journal of Economics, 98, 371–400. Beecher, Janice A., Thomas W. Chesnutt and David M. Pekelney (2001), Socioeconomic Impacts of Water Conservation, Denver, CO: AWWA Research Foundation and American Water Works Association. Berck, Peter, Sherman Robinson and George Goldman (1991), ‘The use of computable general equilibrium models to assess water policies’, in Ariel Dinar and David Zilberman (eds), The Economics and Management of Water and Drainage in Agriculture, Boston, MA: Kluwer Academic Press. Berle, Adolf A. and Gardiner C. Means (1932), The Modern Corporation and Private Property, New York: Commerce Clearing House. Boland, John J. and Dale Whittington (2000), ‘The political economy of water tariff design in developing countries: increasing block tariffs vs. uniform price with rebate designs’, in Ariel Dinar (ed.), The Political Economy of Water Pricing Reforms, New York: Oxford University Press. Bonbright, James C., Albert L. Danielsen and David R. Kamerschen (1988), Principles of Public Utility Rates, Arlington, VA: Public Utilities Reports, Inc. Bontemps, Christophe and Stephane Couture (2002), ‘Irrigation water demand for the decision maker’, Environment and Development Economics, 7, 643–57. Bos M.G. and W. Walters (1990), ‘Water charges and irrigation efficiencies’, Irrigation and Drainage Systems, 4(2), 267–78. Bosworth, B., G. Cornish, C. Perry and F. van Steenbergen (2002), Water Charging in Irrigated Agriculture, Walingford: HR Walingford. Boyer, Robert and J. Rogers Hollingsworth (1997), ‘From national embeddedness to spatial and institutional nestedness’, in J. Rogers Hollingsworth and Robert Boyer (eds), Contemporary Capitalism: The Embeddedness of Institutions, Cambridge: Cambridge University Press, pp. 433–77.

42

The international yearbook of environmental and resource economics

Briscoe, J. (1996), Water as an Economic Good: The Idea and what it Means in Practice, Cairo: International Commission on Irrigation and Drainage. Briscoe, John (1997), ‘Managing water as an economic good: rules for reformers’, Water Supply, 15(4), 153–72. Bromley, Dan W. (1989a), Economic Interests and Institutions: The Conceptual Foundations of Public Policy, New York: Basil Blackwell. Bromley, Dan W. (1989b), ‘Institutional change and economic efficiency’, Journal of Economic Issues, 23(3), 735–59. Bromley, Dan W. (2000), ‘Property regimes and pricing regimes in water resource management’, in Ariel Dinar (ed.), The Political Economy of Water Pricing Reforms, New York: Oxford University Press. Brown, Stephen J. and David S. Sibley (1986), The Theory of Public Utility Pricing, New York: Cambridge University Press. Burt, C.M., A.J. Clemmens, T.S. Strelkoff, K.H. Solomon, R.D. Bliesner, L.A. Hardy, T.A. Howell and D.E. Eisenhouer, (1997), ‘Irrigation performance measures: efficiency and uniformity’, Journal of Irrigation and Drainage Engineering, 123(6), 423–42. Chesnutt, Thomas W., Janice A. Beecher, Patrick C. Mann, Don M. Clark, W. Michael Hanemann and George A. Raftelis (1997), Designing, Evaluating, and Implementing Conservation Rate Structures, Sacramento, CA: California Urban Water Conservation Council. Ciriacy-Wantrup, S.V. (1967), ‘Water economics: relations to law and policy’, in R.E. Clark (ed.), Water and Water Rights, vol I, Indianapolis, IN: Allen Smith Co. Ciriacy-Wantrup, S.V. (1969), ‘Natural resources in economic growth: the role of institutions and policies’, American Economic Review, 51, 1314–24. Clark, Robert E. (ed.) (1967), Water and Water Rights: A Treatise on the Law of Waters and Related Problems, vol I–VII, Indianapolis, IN: Allen Smith Co. Coase, Ronald H. (1937), ‘The nature of the firm’, Economica, 4(2), 386–405. Coase, Ronald H. (1960), ‘The problem of social cost’, Journal of Law and Economics, 3(1), 1–44. Coriat, Benjamin and Giovanni Dosi (1998), ‘The institutional embeddedness of economic change: an appraisal of the “evolutionary” and “regulationist” research programmes’, in Klaus Nielsen and Bjorn Johnson (eds), Institutions and Economic Change, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 3–32. Crisp, Brian F. and Michael J. Kelly (1999), ‘The socioeconomic impacts of structural adjustments’, International Studies Quarterly, 43(3), 533–52. Cummings, Ronald and V. Nercissiantz (1992), ‘The use of water pricing as a means for enhancing water use efficiency in irrigation: case studies in Mexico and the United States’, Natural Resources Journal, 32(3), 731–55. Dasgupta, Partha (1993a), An Inquiry into the Well Being and Destitutions, Oxford: Oxford University Press. Dasgupta, Partha (1993b), ‘Poverty, resources, and fertility. The household as a reproductive partnership’, in Mohan Munasinghe (ed.), Environmental Economics and Natural Resource Management in Developing Countries, Committee of International Development Institutions on the Environment, Washington, DC: World Bank. Dasgupta, Partha S. and G.M. Heal (1979), Economic Theory of Exhaustible Resources, New York: Cambridge University Press. David, Paul (1994), ‘Why are institutions the “carriers of history”? Path depend-

Issues in water pricing reforms

43

ence and the evolution of conventions, organizations and institutions’, Structural Change and Economic Dynamics, 5, 205–20. Diao Xinsen and Terry Roe (2000), ‘The win-win effect of joint water market and trade reform on interest groups in irrigated agriculture in Morocco’, in Ariel Dinar (ed.), The Political Economy of Water Pricing Reforms, New York: Oxford University Press. Dinar, Ariel (1998), ‘Water policy reforms: information needs and implementation obstacles’, Water Policy, 1(4), 367–82. Dinar, Ariel (ed.) (2000a), The Political Economy of Water Pricing Reforms, New York: Oxford University Press. Dinar, Ariel (2000b), ‘Political economy of pricing policy reforms’, in Ariel Dinar (ed.), The Political Economy of Water Pricing Reforms, New York: Oxford University Press. Dinar, Ariel (2003), ‘The political economy context of water pricing reforms’, in Phoebe Koundouri, P. Pashardes, Thimoty Swanson, and Anastasios Xepapadeas (eds), Economics of Water Management in Developing Countries: Problems, Principles and Policies, Cheltenham, UK and Northampton, MA: Edward Elgar. Dinar, Ariel, Trichur Balakrishnan and Joseph Wambia (2004), ‘Politics of institutional reforms in the water and drainage sector of Pakistan’, Environment and Development Economics, 9(3), June 2004. Dinar, A. and J. Letey (1991), ‘Agricultural water marketing, allocative efficiency and drainage reduction’, Journal of Environmental Economics and Management, 20(2), 210–23. Dinar, Ariel and Jyothsna Mody (2004), ‘Irrigation water management policies: pricing and allocation principles and implementation experiences’, Natural Resources Forum, May 2004. Dinar, Ariel, Aaron Ratner and Dan Yaron (1992), ‘Evaluating cooperative game theory in water resources’, Theory and Decision, 32, 1–20. Dinar, Ariel and Ashok Subramanian (1997), ‘Water pricing experiences – an international perspective’, in World Bank technical paper no 386, Washington, DC. Dinar, Ariel and Ashok Subramanian (1998), ‘Policy implications from water pricing experiences in various countries’, Water Policy, 12, 239–50. Dinar, Ariel and David Zilberman (1991), ‘The economics of resource-conservation, pollution-reduction technology selection: the case of irrigation water’, Resources and Energy, 13, 323–48. Dinar, Ariel and David Zilberman (1991), The Economics and Management of Water and Drainage in Agriculture, Boston, MA: Kluwer Academic Publishers. Dixit, Avinash (1996), The Making of Economic Policy: A Transaction Cost Politics Perspective, Cambridge, MA: MIT Press. Douglas, Mary (1986), How Institutions Think, New York: Syracuse University Press. Easter, K. William (1980), ‘Capturing the economic surplus created by irrigation’, University of Minnesota Department of Agriculture and Applied Economics, staff paper P80–14, St. Paul. Easter, K. William and D.E. Welsch (1986), ‘Implementing irrigation projects: operational and institutional problems’, in K. William Easter (ed.), Irrigation Investment, Technology, and Management Strategies for Development, Boulder, CO: Westview Press. Easter, K. William (1992), ‘Intersectoral water allocation’, in Guy L. Moigne, Shawki Barghouti, Gershon Feder, G. Garbus and Mei Xie (eds), Country Experiences

44

The international yearbook of environmental and resource economics

with Water Resources Management, Economic, Institutional, Technical and Environmental Issues, World Bank technical paper no. 175, Washington, DC. Easter, K. William, Nir Becker and Yacov Tsur (1997), ‘Economic mechanisms for managing water resources: pricing, permits, and markets’, in Asit K. Biswas (ed.), Water Resources: Environmental Planning, Management and Development, New York: McGraw-Hill. Easter, K. William and G. Feder (1998), ‘Water institutions, incentives, and markets’, in D.D. Parker and Y. Tsur (eds), Decentralization and Coordination of Water Resource Management, Boston, MA: Kluwer Academic. Easter, K. William, Mark W. Rosegrant and Ariel Dinar (eds) (1998a), Markets for Water Potential and Performance, Boston, MA: Kluwer Academic Press. Easter, K. William, Mark W. Rosegrant and A. Dinar (1998b), ‘The future of water markets: a realistic perspective’, in K. William Easter, M.W. Rosegrant and A. Dinar (eds), Markets for Water: Potential and Performance, Boston, MA: Kluwer Academic. Easter, K. William, Mark W. Rosegrant and Ariel Dinar (1999), ‘Formal and informal markets for water: institutions performance and constraints’, The World Bank Research Observer, 14(1), 99–116. Eggertsson, Thrainn (1996), ‘A note on the economics of institutions’, in Lee J. Alston, Thrainn Eggertsson and Douglass C. North (eds), Empirical Studies in Institutional Change, Cambridge: Cambridge University Press, pp. 6–21. Espey, M., J. Espey and W.D. Shaw (1997), ‘Price elasticity of residential demand for water: a meta-analysis’, Water Resources Research, 33(6), 1369–74. Falkenmark, Malin (1984), ‘New ecological approach to water cycle: ticket to the future’, Ambio, 13(2), 152–60. Feeny, David (1993), ‘The demand for and supply of institutional arrangements’, in Vincent Ostrom, David Feeny and Hartmut Picht (eds), Rethinking Institutional Analysis and Development: Issues, Alternatives, and Choices, San Francisco, CA: Institute for Contemporary Studies Press, pp. 159–209. Finkelstein, Israel and Yoav Kislev (1997), ‘Prices versus quantities: the political perspective’, Journal of Political Economy, 105(1), 83–100. Galbraith, John Kenneth (1967), The New Industrial State, Boston, MA: Houghton Mifflin. Garfield, Paul J. and Wallace F. Lovejoy (1964), Public Utility Economics, Englewood Cliffs, NJ: Prentice-Hall. Gibbons, Diana C. (1986), The Economic Value of Water, Washington, DC: Resources for the Future. Gleick, Peter (2002), The World’s Water 2002–2003: The Biennial Report on Freshwater Resources, Washington, DC: Island Press. Goldin, Ian and D. Roland-Holst (1995), ‘Economic policies for sustainable resource use in Morocco’, in Ian Goldin and L. Alan Winters (eds), The Economics of Sustainable Development, Cambridge: Cambridge University Press. Gómez-Lobo, Andrés and Dante Contreras (2003), ‘Water subsidy policies: a comparison of the Chilean and Colombian schemes’, World Bank Economic Review, December. Grief, Avner (1989), ‘Reputations and coalitions in medieval trade: evidence on the Maghribi traders’, Journal of Economic History, 49, 857–82. Griffin, Charles C., John Briscoe, Banwar Singh, Radhika Ramasubban and Ramesh Bhatia (1995), ‘Contingent valuation and actual behavior: predicting

Issues in water pricing reforms

45

connections to new water systems in the state of Kerala, India’, World Bank Economic Review, 9(3), 373–95. Griffin, Ronald and G. Perry (1990), ‘Volumetric pricing of agricultural water supplies: a case study’, Water Resources Research, 21(7), 944–50. Guttman, J.M. (1982), ‘Can political entrepreneurs solve the free rider problem?’, Journal of Economic Behaviour and Organisation, 3, 1–10. Haggard, Stephen, Jean-Dominique Lafay and Christian Morrisson (1995), The Political Feasibility of Adjustment in Developing Countries, Paris: Development Centre of the Organisation for Economic Co-operation and Development. Haggard, Stephen and Steven B. Webb (eds) (1996), Voting for Reform, New York: Oxford University Press. Hajispyrou, Soteroula, Phoebe Koundouri and Panos Pashardes (2002), ‘Household demand and welfare: implications of water pricing in Cyprus’, Environment and Development Economics, 7, 659–85. Hall, Darwin (ed.) (1996), Marginal Cost Rate Design and Wholesale Water Markets, vol 1, Greenwich, CT: JAI Press. Hall, Darwin and Hanemann Michael (1996), ‘Urban water rate design based on marginal cost’, in Darwin Hall (ed.), Marginal Cost Rate Design and Wholesale Water Markets, Advances in the Economics of Environmental Resources, vol. 1, Greenwich, CT: JAI Press. Hanemann, Michael W. (1998), ‘Determination of urban water use’, in D. Baumann, J. Boland and M.W. Hanemann (eds), Urban Water Demand Management and Planning, New York: McGraw-Hill. Hansen, L.G. (1996), ‘Water and energy price impact on residential water demand in Copenhagen’, Land Economics, 72(1), 66–79. Hardin, Garrett (1968), ‘The tragedy of the commons’, Science, 168, 1243–8. Hearne, Robert and K. William Easter (1997), ‘The economics and financial gains from water markets in Chile’, Agricultural Economics, 15, 187–99. Hirshleifer, Jack, James C. DeHaven and Jerome W. Milliman (1960), Water Supply: Economics, Technology, and Policy, Chicago: University of Chicago Press. Howe, Charles W. and F.P. Linaweaver (1967), ‘The impact of price on residential water demand and its relation to system design and price structure’, Water Resources Research, 3(1), 13–31. Howe, Charles W., D.R. Schurmeier and W.D. Shaw (1984), ‘Innovative approaches to water allocation: the potential for water markets’, Water Resources Research, 22(4), 439–45. Hydrosphere (1997), Incentive Pricing Handbook for Agricultural Water Districts: 53, Washington, DC: US Dept of Interior, Bureau of Reclamation. Hydrosphere (1998), Incentive Pricing Best Management Practice for Agricultural Irrigation Districts: 27, Washington, DC: US Dept of Interior, Bureau of Reclamation. Israel, Arturo (1987), Institutional Development Incentives to Performance, Baltimore, MD: John Hopkins University Press. Johansson, Robert C., Yacov Tsur, Terry L. Roe, Rachid M. Doukkali and Ariel Dinar (2002), ‘Pricing irrigation water: a review of theory and practice’, Water Policy, 4, 173–99. Johnson, Sam H. (1997), ‘Irrigation management transfer: decentralizing public irrigation in Mexico’, Water International, 22(3), 159–67. Jones, Tom (1998), ‘Recent developments in the pricing of water service in OECD countries’, Water Policy, 1, 637–51.

46

The international yearbook of environmental and resource economics

Just, Richard E., Netanyahu Sinaia and John K. Horowitz (1999), ‘Water pricing and water allocation in Israel’, Policy Reform, 3, 97–119. Kijne, Jacob W. (2001), ‘Lessons learned from the changes from supply to demand management in irrigated agriculture: a case study from Pakistan’, Water Policy, 3, 109–23. Kloezen, W.H. (1998), ‘Water markets between Mexican water user associations’, Water Policy, 1, 437–55. Kregel, Jan A. (1980), ‘Markets and institutions as features of a capitalistic production system’, Journal of Post Keynsian Economics, 3(1), 32–48. Krinner, C. and C. Lallana (1999), Sustainable Water Use in Europe, Part 1, Environmental Assessment report no. 1, Brussels: European Environment Agency. Krueger, Anne (1992), The Political Economy of Agricultural Pricing Policy, vol. 5: A Synthesis of the Political Economy in Developing Countries, Baltimore, MD: Johns Hopkins University Press. Kumar, Dinesh M. and O.P. Singh (2001), ‘Market instruments for demand management in the face of scarcity and overuse of water in Gujarat, Western India’, Water Policy, 3, 387–403. Laffont, J.J. and J. Tirole (1993), A Theory of Incentives in Procurement and Regulation, Cambridge, MA: MIT Press. Lampietti, Julian A., Anthony A. Kolb, Sumila Gulyani and Vahram Avenesyam (2001), Utility Pricing and the Poor: Lessons from Armenia, Washington, DC: World Bank. Lichty, R. and C. Anderson (1985), ‘Assessing the value of water: some alternatives’, Regional Science Perspectives, 15(2), 39–51. Madariaga, B. and K. McConnell (1983), ‘Value of irrigation water in the Middle Atlantic States: an econometric approach’, Southern Journal of Agricultural Economics, 16(2), 91–8. Malik, S.A, B.A. Larson and M. Ribaudo (1994), ‘Economic incentives for agricultural non-point source pollution control’, Water Resource Bulletin, 30, 471–80. Mariño, M. and K.E. Kemper (1999), ‘Institutional frameworks in successful water markets’, technical paper 427, Washington, DC: World Bank. Mascollel, A., M.D. Whinston and J.R. Green (1995), Microeconomic Theory, New York: Oxford University Press. Massaruto, Antonio (2002), ‘The full-cost recovery of irrigation: rationale, methodology, European experience’, in Irrigation Water Policies: Micro and Macro Considerations, Agadir, Morocco, www.worldbank.org/agadirconference. Massaruto, Antonio (2003), ‘Water pricing and irrigation water demand: economic efficiency vs. environmental sustainability’, European Environment, 13, 100–19. Massaruto, Antonio (2004), ‘Water pricing: a basic tool for a sustainable water policy?’, in E. Cabrera and R. Cobacho (eds), Challenges for New Water Policies in the XXI Century, Amsterdam: Balkema. McCann, Richard J. and David Zilberman (2000), ‘Governance rules and management in California’s agricultural water districts’, in Ariel Dinar (ed.), The Political Economy of Water Pricing Reforms, New York: Oxford University Press. McKay, Jennifer and Henning Bjornlund (2002), ‘Recent Australian market mechanisms as a component of an environmental policy that can make choices between sustainability and social justice’, Social Justice Research, 14(4), 387–403.

Issues in water pricing reforms

47

Merrett, Stephen (2002a), ‘Deconstructing households’ willingness-to-pay for water in low-income countries’, Water Policy, 4, 157–72. Merrett, Stephen (2002b), ‘Behavioral studies of the domestic demand for water services in Africa’, Water Policy, 4, 69–81. Milgrom, Paul and J. Roberts (1992), Economic Organizations and Management, Englewood Cliffs, NJ: Prentice-Hall. Moore, Michael and Ariel Dinar (1995), ‘Water and land as quantity-rationed inputs in California agriculture: empirical tests and water policy implications’, Land Economics, 71(4), 445–61. Moore, Michael R. (1999), ‘Estimating irrigators’ ability to pay for reclamation water’, Land Economics, 75(4), 562–80. Narain, Vishal (2003), Institutions, Technology and Water Control, Wagenigen University Water Resources Series, New Delhi: Orient Longman. North, Douglass C. (1990a), Institutions, Institutional Change, and Economic Performance, Cambridge: Cambridge University Press. North, Douglass C. (1990b), ‘A transaction cost theory of politics’, Journal of Theoretical Politics, 2(4), 355–67. OECD (Organisation for Economic Co-operation and Development) (1999), The Price of Water: Trends in OECD Countries, Paris: OECD. OECD (Organisation for Economic Co-operation and Development) (1987), Pricing of Water Services, Paris: OECD. Olson, Mancur (1982), The Rise and Decline of Nations: Economic Growth, Stagflation, and Social Rigidities, Hew Haven, CT: Yale University Press. Ostrom, Elinor, Roy Gardner and James Walker (1994), Rules, Games and Common Pool Resources, Ann Arbor, MI: University of Michigan Press. Ostrom, Elinor (1987), ‘Institutional arrangements for resolving the commons dilemma: some contending approaches’, in B.J. McCay and J.M. Acheson (eds), The Question of the Commons: The Culture and Ecology of Communal Resources, Tucson, AZ: University of Arizona Press, pp. 250–65. Ostrom, Elinor (1990), Governing the Commons: The Evolution of Institutions for Collective Action, Cambridge: Cambridge University Press. Pekelney, David M., Thomas W. Chesnutt and W. Michael Hanemann (1996), Guidelines to Conduct Cost-effectiveness Analysis of Best Management Practices for Urban Water Conservation, Sacramento, CA: California Urban Water Conservation Council. Peltzman, Sam (1976), ‘Toward a more general theory of regulation’, Journal of Law and Economics, 19(2), 211–40. Perry, C.J. and S.G. Narayanamurthy (1998), ‘Farmer response to rationed and uncertain irrigation supplies, International Irrigation Management Institute research report no. 24, Colombo, Sri Lanka. Pigram, J.J. (1999), Tradable Water Rights: The Australian Experience, Taiwan Institute for Economic Research working paper, Taipei: Centre for Water Policy Research. Pint, E. (1999), ‘Household responses to increased water rates during the California drought’, Land Economics, 75(2), 246–66. Population Action International (1995), ‘Sustaining the water: an update’, Washington, DC: Population Action International. Provencher, W. (1995), ‘Structural estimation of the stochastic dynamic decision problems of resource users: an application to the timber harvest decision’, Journal of Environmental Economics and Management, 29(3), 321–38.

48

The international yearbook of environmental and resource economics

Randall, Alan (1981), ‘Property entitlement and pricing policies for a maturing water economy’, Australian Journal of Agricultural Economics, 25(3) December, 195–220. Rao, P.K. (1988), ‘Planning and financing water resource development in the United States: a review and policy perspective’, American Journal of Economics and Sociology, 47(1), 81–96. Rausser, Gordon C. (2000), ‘Collective choice in water resource systems’, in Ariel Dinar (ed.), The Political Economy of Water Pricing Reforms, New York: Oxford University Press. Renfro, R.Z.H. and E.W. Sparling (1986), ‘Private tubewell and canal water trade on Pakistan Punjab watercourses’, in K.W. Easter (ed.), Irrigation Investment, Technology and Management Strategies for Development, Boulder, CO: Westview Press. Renzetti, Steven (1999), ‘Municipal water supply and sewage treatment: costs, prices and distortions’, Canadian Journal of Economics, 32(2), 688–704. Renzetti, Steven (2000), ‘An empirical perspective on water pricing reforms’, in Ariel Dinar (ed.), The Political Economy of Water Pricing Reforms, New York: Oxford University Press. Renzetti, Steven (2002), The Economics of Water Demands, Norwell, MA: Kluwer Academic Publishers. Rhodes, George and R.K. Sampath (1988), ‘Efficiency, equity and cost recovery implications of water pricing and allocation schemes in developing countries’, Canadian Journal of Agricultural Economics, 36, 103–17. Rinaudo, Jean-Daniel (2002), ‘Corruption and allocation of water: the case of public irrigation in Pakistan’, Water Policy, 4, 405–22. Rivera, Daniel (1996), Private Sector Participation in the Water Supply and Wastewater Sector: Lessons from Six Developing Countries, Washington, DC: World Bank. Rogers, P., Radhika de Silva and Ramesh Bhatia (2002), ‘Water is an economic good: how to use prices to promote equity, efficiency, and sustainability’, Water Policy, 4, 1–17. Rosegrant, Mark W. and Hans Binswanger (1994), ‘Markets in tradable water rights: potential for efficiency gains in developing country water resource allocation’, World Development, 22(1), November, 1613–25. Ruttan, Vernon W. (1999), ‘Induced institutional innovation’, paper presented in the Conference on Induced Technology Change and the Environment, International Institute for Applied Systems Analysis, Laxenberg, Austria. Ruttan, Vernon W. and Y. Hayami (1984), ‘Toward a theory of induced institutional innovation’, Journal of Development Studies, 20, July, 203–23. Saleth, R.M. (1996), Water Institutions in India: Economics, Law, and Policy, New Delhi: Commonwealth Publishers. Saleth, R.M. and Ariel Dinar (2000), ‘Institutional changes in global water sector: trends, patterns, and implications’, Water Policy, 2(3), 175–99. Saleth, R.M. and Ariel Dinar (2001), ‘Preconditions for market solution to urban water scarcity: empirical results from Hyderabad City, India’, Water Resources Research, 37(1), 119–31. Saleth, R.M. and Ariel Dinar (2004a), The Institutional Economics of Water: A Cross-Country Analysis of Institutions and Performance, Cheltenham, UK and Northampton, MA: Edward Elgar. Saleth, R.M. and Ariel Dinar (2004b), ‘Institutional linkages, transaction costs, and water institutional reforms: analytical approaches and anecdotal evidences’,

Issues in water pricing reforms

49

paper presented for discussion at the Environmental Economics Task Force Meeting, Columbia University, New York, 10–11 October, 2003. Sampath, R.K. (1992), ‘Issues in irrigation water pricing in developing countries’, World Development, 20(7), 967–77. Savedoff, William and Pablo Spiller (1999), Spilled Water: Institutional Commitment in the Provision of Water Services, Washington, DC: Inter-American Development Bank. Scott, Susan (2003), Abolition of Domestic Water Charges in Ireland, accessed at www.worldbank.org/grouponei. Scott, Susan and John Eakins (2002), ‘Household income effects and implementation options’, in Daniel McCoy and Susan Scott (eds), Green and Bear It? Implementing Market-based Policies for Ireland’s Environment, Dublin: Economic and Social Research Institute. Seckler, David, Upali Amarasinghe, David Molden, Radhika de Silva and Randolph Parker (1998), World Water Demand and Supply, 1990–2025: Scenarios and Issues, International Water Management Institute research report no. 19, Colombo, Sri Lanka. Shackle, George G.L. (1972), Epistemics and Economics: A Critique of Economic Doctrines, Cambridge: Cambridge University Press. Shah, F. and D. Zilberman (1995), ‘Political economy of the transition from water rights to water markets’, in A. Dinar and E. Loehman (eds), Water Quantity/Quality Management and Conflict Resolution, Westport, CT: Praeger. Shah, T. (1993), Groundwater Markets and Irrigation Development: Political Economy and Practical Policy, Bombay: Oxford University Press. Shubik, Martin (1985), ‘What is game theory trying to accomplish? Comment’, in Kenneth J. Arrow and S. Honkapohja (eds), Frontiers of Economics, New York: Basil Blackwell, pp. 88–97. Small, L.E. and I. Carruthers (1991), Farmer Financed Irrigation: The Economics of Reform, Cambridge: Cambridge University Press. Smith, Rodney B.W. and Yacov Tsur (1997), ‘Asymmetric information and pricing of natural resources: the case of unmetered water’, Land Economics, 73(3), August: 392–403. Spulber, Nicolas and Asghar Sabbaghi (1998), Economics of Water Resources: From Regulation to Privatization, 2nd edn, Boston, MA: Kluwer Academic Publishers. Stallings, Barbara and Philip Brock (1993), ‘The political economy of economic adjustment: Chile 1973–1990’, in Robert, H. Bates and Anne O. Krueger (eds), Political and Economic Interactions in Economic Policy Reform, Oxford: Blackwell, pp. 78–122. The World Bank Water Demand Research Team (1993), ‘The demand for water in rural areas: determinants and policy implications’, The World Bank Research Observer, 8(1), 47–70. Thobani, M. (1997), ‘Formal water markets: why, when, and how to introduce tradable water rights’, The World Bank Research Observer, 12(2), 161–79. Thobani, M. (1998), ‘Meeting water needs in developing countries: resolving issues in establishing tradable water rights’, in K.W. Easter, M.W. Rosegrant and A. Dinar (eds), Markets for Water: Potential and Performance, Boston, MA: Kluwer Academic. Tietenberg, T. (1988), Environmental and Natural Resource Economics, 2nd edn, Boston, MA: Scott, Foresman.

50

The international yearbook of environmental and resource economics

Tool, Marc R. (1991), ‘Contributions to an institutional theory of price determination’, in Geoffrey M. Godgson and Ernesto Screpanti (eds), Rethinking Economics: Markets, Technology, and Economic Evolution, Aldershot, UK and Brookfield, US: Edward Elgar, pp. 19–39. Tsur, Y. (2000), ‘Water regulation via pricing’, in Ariel Dinar (ed.), The Political Economy of Water Pricing Reforms, New York: Oxford University Press. Tsur, Y. and A. Dinar (1995), ‘Efficiency and equity considerations in pricing and allocating irrigation water’, World Bank policy research working paper 1460, Washington, DC. Tsur, Yacov and Ariel Dinar (1997), ‘The relative efficiency and implementation costs of alternative methods for pricing irrigation water’, World Bank Economic Review, 11(2), 243–62. Tsur, Yacov, Ariel Dinar, Rachid Doukkali and Terry Roe (2004a), ‘Irrigation water pricing: policy implications based on international comparison’, Environment and Development Economics, in Press. Tsur, Yacov, Terry Roe, Rachid Doukkali and Ariel Dinar (2004b), Pricing Irrigation Water: Principles and Cases from Developing Countries, Washington, DC: Resources for the Future Press. Tsur, Yacov (1997), ‘The economics of conjunctive ground and surface water irrigation systems: basic principles and empirical evidence from Southern California’, in D.D. Parker and Y. Tsur (eds), Decentralization and Coordination of Water Resource Management, Boston, MA: Kluwer Academic. United Nations Conference on Sustainable Development (1992), ‘The Dublin statement on water and sustainable development’, accessed at www.wmo.ch/web/ homs/documents/english/icwedece.html#p 4. United Nations Environment Program (1992), ‘Rio declaration on environment and development’, accessed at www.unep.org/unep/rio.htm. Vaux, H.J. and R.E. Howitt (1984), ‘Managing water scarcity: an evaluation of inter-regional transfers’, Water Resources Research, 20, 785–92. Veblen, Thorstein B. (1904), The Theory of Business Enterprise, New York: Charles Scribner’s. Vermillion, Douglas L. (1997), Impact of Irrigation Management Transfer: A Review of Evidence, Colombo: IWMI. Ward, Frank A. and Ari Michelsen (2002), ‘The economic value of water in agriculture: concepts and policy applications’, Water Policy, 4, 423–46. Ware, Caroline F. and Gardiner C. Means (1936), The Modern Economy in Action, New York: Harcourt, Brace. White, Louise G. (1990), Implementing Policy Reforms in LDCs: A Strategy of Designing and Effecting Change, Boulder, CO: Lynne Rienner. Whittington, Dale (1992), ‘Possible adverse effects of increasing block water tariffs in developing countries’, Economic Development and Cultural Change, 40(1), October, 75–87. Whittington, Dale and K Choe (1992), Economic Benefits Available from the Provision of Improved Potable Water Supplies, Washington, DC: US Agency for International Development. Whittington, Dale, Don Lauria and X. Mu (1991), ‘A study of water vending and willingness to pay for water in Onitsha, Nigeria’, World Development, 19(2–3), 179–98. Wichelns, Dennis (1991), ‘Increasing block-rate prices for irrigation water motivate drain water reduction’, in Ariel Dinar and David Zilberman (eds), The Economics

Issues in water pricing reforms

51

and Management of Water and Drainage in Agriculture, Dordrecht: Kluwer Academic. Wichelns, Dennis (1999), ‘Economic efficiency and irrigation water policy with an example from Egypt’, International Journal of Water Resources Development, 15(4), 543–60. Williamson, John (ed.) (1994), The Political Economy of Policy Reform, Washington, DC: Institute for International Economics. Williamson, John and Stephan Haggard (1994), ‘The political conditions for economic reform’, in John Williamson (ed.), The Political Economy of Policy Reform, Washington, DC: Institute for International Economics. Williamson, Oliver E. (1985), The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting, New York: Free Press. Williamson, Oliver E. (1993), ‘Transaction cost economics and organizations theory’, Industrial and Corporate Change, 2(2), 107–56. Wilson, P.N. (1997), ‘Economic discovery in federally supported irrigation districts: a tribute to William E. Martin and friends’, Journal of Agricultural and Resource Economics, 22(1), 61–77. Winpenny, J. (1994), Managing Water as an Economic Resource, London: Routledge. World Bank (2003a), ‘Prospects for irrigated agriculture. Whether irrigated area and irrigation water must increase to meet food needs in the future: validation of global irrigation-water-demand projections by FAO, IFPRI, and IWMI’, report no. 26029, June. The World Bank, Agriculture and Rural Development Department. World Bank (2003b), World Development Indicators 2003, Washington, DC: World Bank. World Bank (2003c), Water Resources Management Strategy, Washington, DC: World Bank. World Bank Water Demand Research Team (1993), ‘The demand for water in rural areas: determinants and policy implications’, The World Bank Research Observer, 8(1), January, 47–70. World Water Commission (2000), World Water Vision Commission Report – A Water Secure World, World Water Council, London: Thanet Press. Yaron, D. (1997a), ‘Israel’, in A. Dinar and A. Subramanian (eds), ‘Water pricing experiences: an international perspective’, technical paper 386, Washington, DC: World Bank. Young, Robert A. (1996), ‘Measuring economic benefits for water investment and policies’, technical paper 338, Washington, DC: World Bank. Young, Robert A. and R. Haveman (1985), ‘Economics of water resources: a survey’, in Allen V. Kneese and James L. Sweeny (eds), Handbook of Resource and Energy Economics, vol. II, Amsterdam: Elsevier. Zekri, Slim and Ariel Dinar (2003), ‘Welfare consequences of water supply alternatives in rural Tunisia’, Agricultural Economics, 28(1), 1–12.

2.

Spatial environmental policy Jacqueline Geoghegan and Wayne B. Gray

INTRODUCTION Many, if not most, natural resources are inherently spatial in location, quality or both. Consequently, human interactions with those natural resources are also spatial. Therefore, the benefits and costs of human actions will also vary over space, resulting in the need for policy instruments such as taxes, subsidies, tradable permits and other regulations that explicitly recognize this spatial dependence. However, while some subdisciplines of economics such as urban and regional economics have spatial relationships as a central focus of analysis, much of natural resource and environmental economics has been non-spatial. For economic processes that are inherently spatial, ignoring the spatial dimension in analysis is analogous to analysing a dynamic process without knowing the chronological order of events (Irwin and Geoghegan, 2001). Indeed, this dearth of spatial analysis in natural resource and environmental economics was raised by Deacon and colleagues a few years ago in a discussion of research opportunities: ‘The spatial dimension of resource use may turn out to be as important as the exhaustively studied temporal dimension in many contexts. Curiously, the profession is only now beginning to move in this direction’. (Deacon et al., 1998, p. 393). The recognition of the importance of space in economics arguably began with Hotelling (1929), in a model of consumers and producers distributed along a linear market. This model spawned an entire literature in industrial organization on product differentiation and spatial competition. Spatial analysis was subsequently advanced by the further development of models in urban and regional economics in the 1960s. One of the cornerstones of urban economics is the monocentric city model, the structure and insights of which are the basis for many of the location-based spatial models reviewed in this chapter. Briefly, the monocentric city model is an equilibrium model of urban spatial structure, where the distribution of land uses on a featureless plain around a central business district is the result of an equilibrium between a declining land price gradient and increasing transportation costs. This model results in concentric rings around the city made 52

Spatial environmental policy

53

up of different land uses, with agricultural and forest lands as the residual land use beyond the developed land uses. For the most part, the only heterogeneity in the landscape is the distance to the central business district or some amenity or disamenity that is implicitly tied to this distance. For an excellent overview of these spatial urban economic land use models see Anas, Arnott, and Small (1998). Two previous Yearbook chapters have treated different aspects of land use in depth that are of particular interest to natural resource and environmental economists (Bockstael and Irwin, 2000 and Hardie, Parks and van Kooten, 2004), while Nijkamp (1999) and Gerking and List (2001) focus on the intersection between urban and regional economics and natural resource and environmental economics. Research on environmental regulation has examined its effectiveness in reducing pollution and increasing compliance, as well as the determinants of regulatory stringency. The observed variation in stringency has a strong spatial component, though most of the research focuses on differences across jurisdictions. An optimistic view of such variation can be taken from Tiebout’s (1956) model of local public good provision in which different localities compete by offering different combinations of taxes and public goods. Local jurisdictions could balance reductions in local pollution against the costs of abating that pollution – though this depends on the benefits and costs of that regulation being borne locally. Oates and Schwab (1988) identify some distortions that can arise in these tradeoffs when jurisdictions are not homogeneous, and Markusen et al. (1995) show that imperfect competition and increasing returns can result in either a ‘race to the bottom’ or a ‘race to the top’ in stringency, with inefficiently low (or high) stringency levels. See the previous Yearbook chapter by Rauscher (2000) for a detailed review of these models. A greater role for spatial analysis can arise in models incorporating detailed measures of costs and benefits to help explain local variation in stringency across polluters within a jurisdiction. The topic of ‘economic geography’ has recently re-emerged in economics, as manifested in the new Journal of Economic Geography and most notably begun with Krugman (1995, 1998) and continued with Fujita, Krugman and Venables (1999) and Fujita and Thisse (2002). These authors have argued that this resurgence of spatial analysis in mainstream economics is due to the recent development of modeling techniques in increasing returns and monopolistic competition that enable the study of the formation and structure of urban centers, the causes of industrial clustering, and the persistence of hierarchical systems of cities. Therefore, instead of taking the location of the central business district as given, these models endogenously determine these locations and the emergence of the location of new economic activity.

54

The international yearbook of environmental and resource economics

It is important to note however, that economic models often constrain ‘space’ to affect the system under consideration in a very simplistic way. For example, space is reduced to distance in a transportation cost model of location, or differences between wage levels at different locations in a model of migration. In many situations, space matters in terms of complex spatial processes and must then be modeled as such at the appropriate scale of analysis. In order to explain and predict these spatial processes, theoretical and empirical models must be developed to address where, when, and why these processes happen (Bockstael, 1996). In order to do this, improved spatial data, increased theoretical understandings of human behavior in space, and new methods to use these data and test these hypotheses are needed (Irwin and Geoghegan, 2001). When using spatial data, two related issues must be considered: how to use the data ‘creatively’ and how to use the data ‘correctly’. The former refers to developing ways of visualizing data to assist with the analysis as well as creating variables from spatial data that can be used in a model; the latter refers to issues of spatial econometrics (Irwin and Geoghegan, 2001). While many studies use data that are based on location, such as a study of manufacturing plants that derives its measure of regulatory stringency based on the state in which the plant is located, a truly ‘spatial’ study should contain a ‘creative’ spatially explicit model of behavior or the interaction over space of the natural and human environments, or the explicit analysis of the spatial pattern of outcomes, or some combinations of these. The second modeling issue, spatial econometrics, deals with the methodological concerns that follow from explicit consideration of spatial effects in econometric models (Anselin, 1988). Such effects may take the form of spatial dependence, in which the values of observations in space are functionally related, or spatial heterogeneity, in which model parameters are not stable across locations. An example of the latter is the varying parameters model in the hedonic housing model found in Geoghegan, Wainger and Bockstael (1997), where the estimated coefficients on explanatory variables are allowed to vary over space as a function of the observation’s distance from the central business district. While spatial econometric issues will not be a focus of this chapter, attention will be drawn to specific spatial econometric specifications to help make a point. See Anselin (2002) for overviews of spatial econometric issues. Since the publication of Deacon et al. in 1998, research in environmental and natural resource economics explicitly incorporating a spatial dimension has grown considerably. Evidence of this increase in research interest includes: the ‘Learning Workshop on Spatial Data Analysis’ at the American Agricultural Economics Association annual meeting 2001, which led to a special edition of the journal Agricultural Economics in 2002;

Spatial environmental policy

55

quite a few sessions during the past few years at the AAEA and ASSA meetings with some sort of explicitly spatial focus; and most recently, the Association of Environmental and Resource Economics workshop in 2003 on ‘Spatial Theory, Modeling, and Econometrics in Environmental and Resource Economics’. In addition, a book dedicated to the use of GIS in natural resource and environmental economics applications was recently published (Bateman, Lovett and Brainard, 2003). Much of this growth in research has been driven or facilitated by the availability of spatially explicit social science data. In addition to this increase in spatial data (remotely sensed data, such as satellite data, as well as other georeferenced data), the advances in geographical information systems (GIS) and global positioning systems (GPS) availability and ease of use have increased the interest in spatial issues among economists. GIS software has a myriad of uses, including data visualization, organization, and integration across sources, as well as variable creation. Often spatial data for a study will come from a wide variety of sources as well as different scales, such as: satellite data on land use, digital soil maps, tax assessment data on property values and structural characteristics, census data of different kinds, road networks, and school district boundaries. An important attribute of GIS is that it allows the linking of these data via their location in space to permit the assignment of the value of these data to the appropriate unit of observation in an analysis. GPS technology, which gives the coordinates of any location, allows a researcher to create a spatial database from ‘scratch’, albeit a time-consuming process. For example, in a household survey of peasant farmer land use choices in southern Mexico, which included a GPS mapping of agricultural fields, each observation required a day of labor by the enumerator to generate it (Vance and Geoghegan, 2002). Finally, as spatial data and spatial data software applications have become more available, so too have the statistical tools to use these data appropriately. For an introduction to spatial data and these technologies that have been used in environmental and natural resource economics, see Bateman et al. (2002), Nelson and Geoghegan (2002) and Bell and Irwin (2002). However, the benefits of spatial analyses also entail some costs. Spatial data technology such as GIS and GPS has become much more user friendly over the past decade, but learning new software packages such as ArcInfo or IDRISI still entails startup costs. New statistical methods have been developed to address issues that arise with the use of spatial data, but while some of these have been programmed into existing statistical software, there is still a paucity of readily available routines. Also, while more and more spatially explicit datasets of interest to economists are becoming available, they often appear in formats that are not immediately usable for

56

The international yearbook of environmental and resource economics

economic analysis and require considerable preparatory manipulation. Although developing any new database requires some effort, using the visualization tools in GIS software to ‘see’ data problems can lead researchers to spend more time on data cleaning. Finally, the traditional graduate training in environmental and natural resource economics does not regularly include a background in spatial economic theories, such as those used in urban and regional economics or in the ‘new economic geography’ on increasing returns or social interactions across space. This adds further startup costs for the prospective researcher to acquire a familiarity with the relevant literature. In addition, a recurring theme throughout this chapter is that a spatially explicit approach to modeling environmental and natural resource issues can require an interdisciplinary approach, where space is explicitly considered in both the human and natural systems. Environmental and natural resource economics has a long tradition of interdisciplinary research: hydrological models are added to agricultural production models to understand the transport of water pollutants, atmospheric science models are added to property value models to quantify the effects of air pollutants, population biology dynamics are key ingredients in fisheries models. Understanding these models of natural systems generates another startup cost for researchers, but one that is necessary to develop credible models incorporating a spatial component. We now turn to the focus of the chapter: the burgeoning field of spatial natural resource and environmental economics, where the spatial aspect of human behavior or the natural environment makes a crucial difference in the analysis and policy response to the problem. We begin with a discussion of some of the initial spatial analyses in environmental economic policy in the area of market based incentives: where the incorporation of a pollution externality that is heterogeneous across spatial locations into the design of these instruments, creating spatially targeted policy instruments, leads to greater efficiency. Following this, we discuss in more detail the causes of, and responses to, specific location based environmental externalities. With increased spatial specificity, models are able to produce more accurate estimates of the consequences of different policies, such as the impact on firm location decisions or the distributional effects across locations and populations. Subsequently, we review the literature on the different benefit estimation methodologies that have recently incorporated more spatially explicit approaches to improve environmental valuation estimates, and conclude our review with a discussion of natural resource management issues, where spatial specificity has contributed to policy options such as environmental refuges for certain species. In each area the degree of ‘spatialness’ varies, from models where the heterogeneity of space is

Spatial environmental policy

57

simply incorporated as location-specific characteristics, to economic models linked with sophisticated natural science spatial diffusion or interaction models, to models that explicitly incorporate spatial human behavior and empirical models that specifically model the spatial nature of the data through spatial econometric techniques.

LOCATION BASED ENVIRONMENTAL EXTERNALITIES AND REGULATORY RESPONSES For many spatial environmental externalities, the location of an economic activity on a particular parcel of land is the cause of the environmental externality, be it industrial, agricultural or residential land use. Traditionally, in air quality issues, the dichotomy is between stationary and mobile source pollution, while in water quality issues it is between point and non-point source pollution. Stationary and point sources of pollution are usually large individual sources and have been subject to greater environmental regulation than the smaller and more diffuse mobile or non-point sources of pollution. The importance of spatial specificity in analysis can depend on the nature of the pollution under consideration as well as other environmental characteristics. For example, the location of a carbon dioxide source does not affect its contribution to global warming, so spatial targeting for sources would not be necessary, while the impacts of the sources of the precursors to ground-level ozone depend upon local environmental conditions, so potential efficiency gains exist from spatially targeted policies. However, even for perfectly mixing pollutants like carbon dioxide, their ultimate environmental consequences on human welfare will likely be distributed heterogeneously, requiring spatial explicitness for the analysis of those impacts. We begin with a preliminary introduction to spatially explicit market based incentives designed to address these environmental externalities. However, most environmental regulation is not market based, but command-and-control: regulators set standards for emissions, conduct inspections, and punish violations. In this area, researchers have examined how effective regulation is in affecting environmental performance, measured either by emissions or by compliance. They have also examined the determinants of regulatory stringency, which can include spatially differentiated benefits from pollution abatement as well as political pressures on the regulators from different interest groups. We examine the role that location plays in these analyses, and then consider research that seeks to explain a firm’s decision on where to locate its plants. We should note at this point that most of this research has not used spatially explicit

58

The international yearbook of environmental and resource economics

modeling, in the sense of econometric models that incorporate spatial dependence across locations, so the primary role of space in these analyses is to define some explanatory variables which are then entered in a standard econometric model. Following this is a discussion of non-point source pollution, where most of the spatially explicit economic research has focused on agricultural land use. Subsequently, we investigate the potential positive externalities associated with agricultural land use and government policy responses to both these negative and positive externalities. Mobile sources are not addressed here, as a number of existing works on the linkages between spatial transportation systems and the environment can be found within Proost and Braden (1998) and Roson and Small (1998), as well as the previous Yearbook chapter by Bohm and Hesselborn (1999). Market Based Incentives: Emission Taxes and Tradable Permits The literature on market based incentives contains many examples of the tradeoffs that exist between the benefits and costs associated with greater spatial specificity. While increased spatial specificity can lead to large potential gains as in the above examples, transaction costs in the form of information and monitoring required to implement a spatial policy could be high, so that a second-best approach with less spatial specificity could be desirable. Another approach to developing market based incentive policies could be to use information gained from a spatially explicit study and to apply those results elsewhere in a benefits transfer framework. An early theoretical paper that investigated the use of pollution taxes to internalize an environmental externality in a spatially differentiated framework is found in Hochman, Pines and Zilberman (1977). In this paper, an upstream agricultural firm generates pollution in the production of a crop that either decreases welfare in a downstream urban center or increases the production costs of downstream firms. The von Thunen-like models analyse the effect of the externality on the spatial pattern of land use and rent. In the model that only looks at the impact on welfare in the urban center, the optimal per unit pollution tax is a function of the distance between the location of the source of the pollution and the urban center, as the pollution in this model is assumed to have a natural rate of absorption along the pathway from source to sink. As a result, the tax decreases the further the upstream polluting firm is from the urban center. In the second model, which assumes no natural rate of absorption but focuses exclusively on the pollution externality’s affect on downstream producers, the pollution from sources that are further upstream will affect more downstream producers, so the resulting optimal tax is now an increasing

Spatial environmental policy

59

function of the distance between source and sinks. The implication of these two different sets of modeling assumptions on land use is that in the first model, an optimal tax can change the traditional von Thunen pattern of the intensity of land use, while the second model results in the typical monotonically decreasing rent gradient from the urban center. More recent papers that address issues concerning optimal pollution taxes in a spatial framework include Xepapadeas (1992) and Uimonen (2001). The former paper, in addition to addressing spatial aspects of pollution, also includes dynamic and stochastic aspects to derive the optimal tax in each spatial region, while the latter paper uses a spatial general equilibrium framework to derive a two-part optimal tax and emissions rights policy. Two analytical papers that bring together the firm location literature and the market based incentives literature in a spatial context are found in White and Wittman (1982) and Krumm and Wellisch (1995). In the first paper, optimal taxes are considered where an additional policy tool is the pattern of the location of polluting firms and their associated pollutees. For example, either the polluters or the pollutees can separate themselves from each other to avoid the externality entirely. The authors considered both short-run and long-run efficiency conditions, where the optimal tax gives incentives to polluting firms in fixed locations to reduce emissions to the optimal level, while the latter also includes incentives for optimal relocation to the polluter or pollutee. They show that for the short run a tax solely on polluters is efficient, but does not lead to the long-run efficient spatial pattern of polluter and pollutee locations, while an incentive scheme for both can lead to the long-run efficient outcome. In the second paper, firms choose optimal locations among different jurisdictions, where each jurisdiction sets overall allowable emissions in its region. If firms are mobile, then the optimal market based incentive must result in the efficient use of resources in each jurisdiction when firms chose their locations. The main result of the paper shows that market based incentives can generate incentives for both the optimal location and production decision for each firm if the revenues generated by the policy are redistributed to immobile residents. The large literature on the spatial effects associated with marketable permits is considered in Tietenberg (1995). The different types of tradable permits are reviewed, starting with the optimal design of a system that takes into account the location of each source and receptor. In this firstbest system, each source of pollution could have separate permits for each pollution receptor, which can lead to a large, complicated portfolio of tradable permits for each source. The other extreme is a system with no restrictions on trades and all permits trading equally – implicitly assuming away the spatial component of the model. Though simpler to implement,

60

The international yearbook of environmental and resource economics

a free-trade system may result in an inefficient distribution of pollution if there really are heterogeneous impacts of pollution from different sources. A number of intermediate, ‘second-best’ alternatives have been suggested in the literature, such as only allowing trading in geographically specified zones, assuming that pollution from nearby sources has similar impacts. An important spatial policy issue involved with tradable permits is the possibility of ‘hot spots’, where trading results in an unacceptable concentration of a pollutant in a particular local area. Note that a hot spot can be described as a special case of heterogeneous impacts, with the marginal cost of additional pollution being much higher at that point than it is elsewhere. Tietenberg argues that issues concerning hot spots and the relative efficiency of second-best tradable permits versus a more traditional command-and-control approach are empirical questions and cites a number of empirical studies to show that when spatial considerations are important, then the actual local conditions matter a great deal in this determination. The sulfur dioxide trading system established under the 1990 Clean Air Act Amendments for US coal fired power plants has no spatial restrictions on trading, despite its being a non-uniformly mixed pollutant. Concerns about hot spots are addressed by regulatory tiering, as plants in nonattainment counties are not permitted to buy permits in order to expand their emissions. Shadbegian et al. (2004) use a detailed model of the transport of air pollutants to quantify differences in the marginal benefits of pollution abatement from different plants. They find that permit buyers tend to have higher-impact emissions than permit sellers, suggesting the need for some variation in permit prices across sources, based on those heterogeneous impacts rather than on the presence of hot spots. Location-specific Differences in Regulation of Point Sources While monitoring and enforcement of environmental policies were covered in a previous Yearbook chapter (Cohen, 1999) the question asked here is more focused: how does location affect the regulation of point sources under traditional command-and-control regulation? First, the location of a plant helps determine the type and stringency of environmental regulations to which it is subject. Under a federal system, such as in the United States, a federal agency sets the overall rules, but substantial scope remains for variations in stringency across state agencies (in a few cases local agencies may also become involved). The location of a plant affects which agency is responsible for its regulation. Some research simply takes the enforcement activity directed towards the plant as given: numbers of inspections, other enforcement actions, or penalties. Other research

Spatial environmental policy

61

examines the determinants of the enforcement activity themselves, which may include other characteristics of the area in which the plant is located. For example, Gray and Deily (1996) use county unemployment rates and plant size relative to the local labor market to help predict the air pollution enforcement directed at steel mills. They also use the overall state-level inspection rate to help instrument for the inspections at a particular plant. One location-specific regulatory characteristic commonly used in US research on air pollution regulation has been attainment status, defined for several key air pollutants in each county since the 1970s. When a county is not in attainment, new plants in that county are subject to stricter regulation. Therefore attainment status provides a well-defined measure of differences in regulatory stringency across counties, providing thousands of observations for cross-sectional analyses. A county’s attainment status can change, making it usable for panel data analyses, though the infrequency of such changes may affect the amount of explanatory variation available (most counties maintain the same attainment status for a decade or more, though with over 3000 counties in the US there is usually some time series variation for researchers to work with). Many researchers have found that attainment status affects pollution emissions. Looking at county-level data for total suspended particulates, Chay, Dobkin and Greenstone (2003) find that non-attainment counties had large and significant reductions in particulate concentrations after they were designated as not in attainment. Henderson (1996) finds reductions in ozone concentrations in counties that are in non-attainment for ozone. Using plant-level data, Shadbegian and Gray (2003) find that particulate emissions are significantly lower at pulp and paper mills located in nonattainment counties, even after controlling for the air pollution abatement expenditures at those plants. Becker (2001) and Becker and Henderson (1999) document significantly higher pollution abatement costs, and higher overall production costs, for polluting plants located in non-attainment counties, which is consistent with those plants putting more effort into abatement due to the stricter regulations. Several other measures of regulatory stringency have been developed, based on the political jurisdiction in which the plant is located, typically at the state level. Congressional voting records compiled by the League of Conservation Voters have been used to measure pro-environmental political support in a state, with the advantage of providing variation over time as well as across states. Measures of state-level environmental regulations and legislation, government spending on environmental agencies, business spending on pollution abatement, and membership in conservation organizations, may also help capture the intensity of a state’s political support for stricter environmental regulations.

62

The international yearbook of environmental and resource economics

Gray and Shadbegian (1998) find that the production technology used at paper mills in a state is influenced by the state’s regulatory stringency. New paper mills in states with more pro-environment Congressional voting tend not to incorporate highly polluting pulping techniques. Different pulping techniques result in different amounts of air and water pollution: plants in states with stricter air pollution regulations are less likely to use the most air-pollution-intensive pulping technique, while plants in states with stricter water pollution regulations are less likely to use the most waterpollution-intensive pulping technique. Yet another measure is developed by Berman and Bui (2001), who use detailed information on specific stringent local air pollution regulations directed at stationary sources in the Los Angeles area to explain the sharp rise in pollution abatement costs at oil refineries there. Border Effects of Environmental Regulation Not all location-specific influences on a plant’s environmental performance are limited to identifying the political jurisdiction responsible for regulating a plant. A plant located near a border may send most of its pollution into the next jurisdiction. Such pollution is truly an externality, with no incentive to clean it up, even from the viewpoint of the local regulatory agency – assuming that only people in the same jurisdiction can provide political support for regulators (an alternative perspective is that people in the neighboring jurisdiction are free riders, benefiting from any pollution cleanup that occurs without contributing towards its cost). If the border is between states, a federal regulatory agency can insist that state agencies take steps to reduce transboundary pollution; if the border is international, some connection between the countries is needed to provide the neighboring country with some leverage to compel more stringent regulation. This could be an external connection, such as trade flows between the countries, or transboundary pollution flowing in the other direction – although the tendency of pollution to flow downwind or downstream reduces the probability of mutual incentives for pollution reduction, except where the river itself forms the boundary between jurisdictions. Helland and Whitford (2003) find that facilities located in counties on state borders release more toxic chemicals into the air and water than facilities in non-border counties. Sigman (2002) looks at water quality within rivers, finding worse quality upstream of international borders, where much of the damage would be borne by people downstream in the other country. She also finds that these border effects are smaller when both countries are in the European Union, suggesting that closer political ties tend to reduce these border effects. Gray and Shadbegian (2004b) find

Spatial environmental policy

63

border effects for air pollution across US states, with significantly higher emissions at paper mills whose emissions primarily affect people in neighboring states. This effect is substantially reduced when the neighboring state is pro-environmental politically (as measured by Congressional voting records). These latter two studies show the importance of going beyond a simple ‘on-off’ characterization of border effects. By modeling the factors that affect the connections between jurisdictions, we can predict the extent to which regulators will impose stringent regulations on transboundary pollution, depending on whether the adjacent jurisdiction is willing and able to compel such stringency. Environmental Justice and Political Activism Another way in which a plant’s location can affect the regulatory stringency it faces is through the characteristics of its neighborhood, specifically the demographic composition and political activity of the surrounding population. Since a plant’s pollution will tend to have its greatest impact on its nearest neighbors, those neighbors are likely to be in the forefront of efforts to lobby regulators for more stringent standards. If so, regulators should pay less attention to plants surrounded by people who are unorganized, politically inactive, and with no political clout, so those plants are likely to face less stringent regulation. The availability of spatially detailed data on demographics and political activity led to many studies in this area. The growing literature in ‘Environmental Justice’ focuses on demographic composition, looking for evidence that poor and minority communities bear a disproportionate share of the costs of pollution exposure, presumably because either the regulators or the polluters are less concerned about the effects of pollution on those communities. Kreisel et al. (1996) find that minorities are not disproportionately exposed to toxic releases, but observe some evidence that the poor are disproportionately exposed. Arora and Cason (1999) find that race is a significant positive determinant of toxic releases in non-urban areas of the South, but not elsewhere. One concern with many of these studies is that the demographics of an area might be endogenous: if pollution lowers housing prices and attracts a disproportionately poor population, that might give the appearance of environmental ‘injustice’. To address this issue, studies by Been and Gupta (1997) and Wolverton (2002) consider the relationship between the siting of polluting plants and neighborhood demographics at the time of siting. Been and Gupta find that commercial hazardous waste treatment storage and disposal facilities were sited no more often in disproportionately African American neighborhoods and less often in poor neighborhoods (though more often in Hispanic areas). Wolverton finds that race does not

64

The international yearbook of environmental and resource economics

matter and that poor neighborhoods attract disproportionately fewer polluting plants. In a study using a direct measure of regulatory behavior, Jenkins, Maguire, and Morgan (2002) show that poor/minority communities receive lower ‘host’ fees for the siting of landfills while richer communities receive higher ‘host’ fees. Other researchers have focused on the political activity of the neighboring population. These studies use voter turnout to measure political influence and pro-environment voting to measure environmental activism. Hamilton (1993, 1995) uses data at the ‘zip-code neighborhood’ level to analyse the capacity expansion or contraction decisions of commercial hazardous waste facilities, finding that capacity expansions are negatively correlated with voter turnout. Viscusi and Hamilton (1999) find that Superfund sites in counties with greater voter turnout and in pro-environmental states have stricter environmental cleanup targets for cancer risk. Sigman (2001) considers the length of time it takes EPA to process Superfund sites, finding that community influence (measured by voter turnout and median income) is an important factor affecting EPA’s bureaucratic priorities. Gray and Shadbegian (2004b) examine the interactions between voter turnout and environmental support (measured by conservation membership), finding that greater political activity is only associated with lower pollution where the population is strongly pro-environment. The wide range of studies in this area makes it difficult to reach a unanimous verdict on the importance of neighborhood demographic characteristics and political activity as determinants of environmental outcomes (regulatory stringency or pollution levels). The overall evidence for political activity influencing these outcomes does seem to be stronger than the evidence for demographic characteristics. Within the different demographic characteristics, there seems to be little evidence for environmental inequity related to race, either for the black or the Hispanic community. There is a bit of evidence that poor communities face greater amounts of pollution (and perhaps less attention from regulators), but efforts to measure this are hampered by the difficulty, noted above, of controlling for the possibility that poor people choose to live in low-rent neighborhoods, where they face a variety of undesirable characteristics, including pollution. Spatial Inequalities and Diffusion Models Most studies of regulatory stringency and environmental outcomes focus on the consequences for average pollution levels. Millimet and Slottje (2002) instead focus on measuring the inequality in pollution exposures for people in certain demographic groups, specifically minorities. They

Spatial environmental policy

65

note evidence that the health effects of pollution may be non-linear, so that groups exposed to especially high concentrations might have greater marginal benefits from pollution reductions. Their calculations, based on a proportional increase in regulatory stringency across all states, show that greater stringency could increase the inequality of exposures. The increase in inequality could result in negative net health effects, even when there is an overall reduction in pollution. They use these calculations to argue that federally set standards targeting high-pollution areas are needed to reduce inequality in exposure among the population. Gray and Shadbegian (2004b) add another spatial dimension to analyses of regulatory stringency at 409 US pulp and paper mills, using spatially explicit models to quantify the marginal benefits of reducing the air and water pollution from each plant. For air pollution, an air dispersion model is applied to information about the pollution source (stack height and pollution emissions) and local meteorological data (mixing height, wind directions and speeds), resulting in a measure of the impact of another ton of emissions (PM10 and SO2) on average ambient particulate exposure in a wide circle around the plant. For water pollution, an experimental version of the EPA’s National Water Pollution Control Assessment Model (NWPCAM) is used to measure the impact of changes in each plant’s discharges of water pollutants (BOD and TSS) on water quality at regular intervals down all relevant watersheds, using a continuous measure of water quality rather than relying on the traditional quality distinctions of fishable, boatable, and swimmable. These impacts on air and water quality are translated into a measure of the marginal dollar benefit from reducing emissions at each plant, a measure of the marginal impact of pollution from that plant. Plants whose pollution has a larger marginal impact have significantly lower emissions, for both air and water pollutants. These measures of marginal pollution impact for each plant are then interacted with the characteristics of the population surrounding the plant, measured in terms of their demographic characteristics and state of residency. This analysis is looking for indirect evidence that regulators ‘weight’ people differently based on those characteristics. If regulators only value people who reside in their own state, benefits going to people in other states will get zero coefficients; if regulators consider everyone equally, in-state and out-of-state benefits will get the same coefficient. The study finds significant differences in the impact of benefits going to different groups. For example, health benefits from reduced SO2 emissions for people living outside the state seem to count only about one-third as much as comparable benefits for people living inside the state. Similarly, benefits from reduced BOD discharges for poor people seem to be given only 40 per cent of the weight given to the non-poor.

66

The international yearbook of environmental and resource economics

Gray and Shadbegian (2004b) also address a methodological point, mentioned earlier under Environmental Justice. The demographic characteristics of the population around a polluter may be endogenous, if sufficient time has passed since the plant began operations. This could lead to bias in the estimated coefficients on the demographic variables. Been and Gupta (1997) and Wolverton (2002) avoided this bias by measuring the demographic characteristics of the neighborhood before the plant opened, but many industrial polluters (including the pulp and paper mills considered in Gray and Shadbegian) are too old for presiting data to be available. Gray and Shadbegian develop a spatially lagged instrumental variables method, using the population characteristics in a large ‘ring’ around the plant (e.g. people living more than five miles but less than 20 miles from the plant) as instruments for the same characteristics in the immediate neighborhood (e.g. less than five miles from the plant). Thus the instrumental variable is calculated on a ‘donut shaped’ area around the plant, with the circular neighborhood nearest the plant deleted from the calculation. The results show that the demographics in the surrounding area provide strong instruments, and that the estimated effects of demographic variables aren’t greatly affected by using the instrumental approach (suggesting that endogeneity is not too serious a problem in this case). Environmental Influences on Plant Location Much of the research in environmental economics on polluters’ decisions deals with plant location: do firms consider differences in environmental regulation when choosing where to site a new plant, or when allocating their production across existing plants. Empirical models of plant location have evolved over time, based in large part on the data available to the researchers. The estimated impacts of regulation on plant location are greatly affected by the methodology and data employed. Jeppesen and Folmer (2001) characterize the important differences, and Jeppesen et al. (2002) provide a meta-analysis of the results from different studies. Most of the studies of plant location examine the number of new plant births in an area, though some studies have examined plant closings, shifts in production, and plant relocations. As with the studies mentioned earlier, the most common measure of regulatory stringency is county attainment status for air pollution, though measures of pollution abatement spending and state political support for environmental regulation have also been used. Recent work in this area includes Becker and Henderson (2000), who used plant-level Census data to identify new plant births in several polluting industries, then related them to county non-attainment status for ozone.

Spatial environmental policy

67

They find large and significant reductions, on the order of 40–50 per cent, in the number of new plants opening in non-attainment counties. These reductions appeared first in the 1970s for larger plants, but spread to smaller plants in later years. They conclude that the regulatory pressures from the Clear Air Act tend to lead to the ‘graying’ of the industrial landscape. A similar conclusion is reached by Greenstone (2002), who finds significant reductions in economic activity of polluting industries in non-attainment counties. List et al. (2000, 2003) find similar evidence using plant location data from New York State: fewer plants open in non-attainment areas, and plants are more likely to relocate into attainment areas. International flows of capital have been studied in a similar manner, given data on flows of foreign direct investment (FDI) into the US. List (2001) finds fewer new foreign owned plants in California counties that are in non-attainment. Keller and Levinson (2002) use aggregate FDI inflows at the state level, and find that states with higher pollution abatement costs attract less FDI. Xing and Kolstad (2002) examine FDI flows from the US, using SO2 emissions per capita to measure the stringency in the target country. They find that countries with less stringent regulation tend to attract more FDI flows in heavily polluting industries, but not in less polluting industries. On the other hand, Eskeland and Harrison (2003) use data for foreign investment flows into Mexico, Venezuela, Morocco and Cote d’Ivoire, and find little evidence that these capital flows are driven by foreign abatement costs. In addition, foreign owned plants are found to use less energy than domestically owned plants, which suggests that they are generating less pollution, and might be improving the environmental situation in the host country. Summarizing the results of these studies on plant location is difficult. This is partly due to differences in research methodology, and partly due to differences in emphasis. Some studies take pains to note that the impact of the environmental stringency measure (while statistically significant) isn’t all that large, while others focus on statistical significance. Abatement costs are only a few per cent of production costs, even in highly pollutionintensive industries, so we might be surprised to see a large effect of regulatory stringency on plant location, if we expected such differences across jurisdictions to be swamped by differences in taxes or in wages. One possible explanation for the studies which find large and significant effects might lie with less tangible costs of regulation, such as delays in obtaining environmental permits before a new large polluting plant can open. Gathering location-specific data on permit delays may help resolve this puzzle. Decker (2003) uses permit data to help solve a different puzzle: why firms seem to put substantial effort into compliance with ‘voluntary’ programs, even in the absence of enforcement pressures from regulators. Decker shows that

68

The international yearbook of environmental and resource economics

firms which have performed well on voluntary compliance issues (such as reducing toxic releases) tend to receive their environmental permits for new plants more quickly. The high costs of delay for these large investments may provide sufficient leverage to induce firms to put extra efforts into compliance with regulations and reductions in emissions. Non-point Source Negative Externalities: Agricultural Land Use While the important policy considerations for non-point source (NPS) pollution have much in common with the point source pollution discussed above, the policy concerns are exacerbated for non-point sources. Examples of this increased difficulty for non-point sources include: establishing the responsible source of the pollution, instrument choice in regulation, monitoring and enforcement of regulation, and the large heterogeneity in the attributes of the different sources. For an introduction to these issues see the previous Yearbook chapter by Shortle and Abler (1997) as well as Dosi and Tomasi (1994). It appears that the entire extent of the economic literature on non-point source pollution focuses on agricultural land use, perhaps because the linkages between agricultural practices and downstream water quality and other environmental damages are well recognized. In the United States, the main government policy in response to this has been the Conservation Reserve Program (CRP) and the newer Conservation Reserve Enhancement Program (CREP) of the US Department of Agriculture. A large portion of the spatial economic research on non-point source pollution focuses on these programs. The goal of these programs is to target environmentally sensitive agricultural land for retirement, to lead to an improvement in water quality through the reduction in non-point source pollution, as well as to enhance wildlife habitat. However, while agricultural land use as a cause of non-point source pollution is a major area of concern, additional causes of NPS pollution do not appear to have received much, if any attention in the economic literature. These include chemical use on golf courses and residential lawns, nutrient loading from septic fields in areas without sewer service, sedimentation from altered hydrological regimes in urbanized areas, sediment discharge from the act of development, which disturbs soils, and toxics from petroleum products runoff on impervious surfaces. For residential land use, government policies that could be used to reduce NPS include: spatially differentiated growth controls, zoning regulations, and transferable development rights. While these regulations were not designed specifically with NPS in mind, they do have the potential to reduce these sources of NPS. While a large literature exists on the effects of these different government policies on residential land use decisions (see the previous Yearbook chapter

Spatial environmental policy

69

by Bockstael and Irwin (2000) for an overview), we are not aware of any studies that specifically link these policies to the ultimate effect on environmental externalities. Similar to the research in habitat preservation described below, much of the research on the CRP and CREP has focused on targeting funds to achieve the maximum environmental benefit, given a fixed budget. Early research (e.g. Ribaudo, 1989) focused on using the program to maximize water quality benefits, while later research included additional environmental benefits. In addition, as research progressed, so too has the level of spatial sophistication. For example, Babcock et al. (1996, 1997), show that the tradeoffs between targeting land based on a least-cost land approach versus an environmental benefits approach depend upon the spatial distribution of those costs and benefits. Other spatial analysis considered multiple management options for reducing non-source point pollution such as sediment containment (Braden et al., 1989). In these models, substantial differences in the optimal policy are found as a function of spatial locations, once again suggesting the importance of spatially targeting policies for non-point source pollution reduction. Subsequent research has continued to investigate the cost effectiveness of targeting funds to particular agricultural lands to reduce non-point source pollution, with increased spatial specificity. Carpentier, Bosch and Batie (1998) focus on the compliance and transaction costs associated with a reduction in nitrogen loadings in a case study of dairy farms in the Lower Susquehanna watershed. They compare a uniform standard applied across all farms to a perfectly targeted performance standard that is based on farm heterogeneity, using individual farm survey data which include information on farm practices and nitrogen runoff estimates. The results suggest large costs savings from the targeted standard that is based on the spatial data. Similarly, in Qiu and Prato (1999) different spatially explicit policy instruments for reducing agricultural non-point source pollution are compared, with an application to a watershed in Missouri. Theoretical advances include Goetz and Zilberman (2000), in which optimal control is used first to solve for the optimal tax, varying over both space and time, which results in the optimal level of site-specific agricultural production and subsequently to solve for the socially optimal intertemporal allocation of agricultural activity across space. This previous research assumed that the ‘transport coefficients’, that is, the relationship between agricultural land use practices and environmental damages, were exogenous, so the next step in this literature was to make these endogenous (Lintner and Weersink, 1999; Khanna et al., 2003; Yang et al., 2003). While Lintner and Weersink (1999) simplify the problem by assuming that all parcels are identical, Khanna et al. (2003) allow different

70

The international yearbook of environmental and resource economics

land management practices on contiguous parcels in the calculation of the transport coefficient for each parcel. Once again, results show the efficiency gains allowed through greater spatial specificity of models. Positive Externalities: Conservation Easements While the previous subsection focused on the role of agriculture in creating negative externalities via non-point source pollution, agricultural land use can also be the source of positive environmental externalities, for example, through the provision of habitat and open space amenities. Beside the CRP and CREP programs discussed above that were designed to reduce the negative externalities associated with agricultural land use, state and local governments in the United States also have assorted conservation programs to keep land in agricultural use to preserve these positive externalities, especially as development pressures mount for converting agricultural land to residential uses. One particular policy that has received attention is the purchase of development rights or conservation easements on agricultural land, where the land owner is paid to retain the parcel in agricultural use. In order to implement such a policy the value of the development right must be calculated and information must be known about the difference in value of an individual parcel in future residential use versus continued agricultural use. In the brief introduction to the von Thunen-style spatial models of land use and land value models above, agricultural land was considered the ‘residual’ use which simply existed beyond the developed fringe. However, additional factors explain agricultural land values. A more Ricardian approach would also consider the quality attributes of the land, such as productivity and other measures of the returns to agricultural use. For example, an agricultural land value model that incorporates both spatial and land quality features can be found in Boisvert, Todd, and Regmi (1997). Subsequent research on spatial models of agricultural land values has included environmental amenities in addition to agricultural productivity measures using GIS data (Bastian, McLeod, Germino et al., 2002). A measure of future residential land development pressures on current agricultural land values is incorporated in Plantinga and Miller (2001) in order to estimate the value of the development right on an agricultural parcel. An econometric model is specified that includes net returns to agricultural use, income levels, population pressures, and commuting times to nearby metropolitan areas. This estimated model, in conjunction with a spatial interpolation algorithm, is used to generate a surface of spatially explicit development rights values for a region in New York State. Spatially explicitness and the use of GIS are important in the calculation of

Spatial environmental policy

71

explanatory variables, the estimation of development rights, as well as the presentation of results. Looking at a different aspect of conservation easements on agricultural land, Geoghegan, Lynch and Bucholtz (2003) test the hypothesis that preserved agricultural land increases the value of nearby residential parcels, using a hedonic model for three counties in Maryland. They included as regressors the per cent of land in different land uses, including preserved and unpreserved agricultural land surrounding each residential parcel. The results suggested that agricultural preservation could be partially selffinancing by generating enough increase in residential land values to increase land tax revenues in two out of the three counties.

BENEFIT ESTIMATION METHODOLOGIES Among the three main benefit estimation methodologies – hedonic pricing models, travel cost, and contingent valuation – spatial relationships have always played an important role. Indeed, the raison d’etre for the travel cost approach is explicitly spatial: we estimate individuals’ demands for recreational resources based on the distances they are willing to travel to access those resources. In the hedonic pricing literature of housing values, many of the research questions involve location based amenities or disamenities, where often the effect measured is the distance from the house to the amenity in question. Contingent valuation is a much broader approach to benefit estimation. Therefore, the incorporation of space into a contingent valuation study depends upon the specific research question at hand. The continued growth in the use of GIS applications in benefit estimation could potentially improve these methodologies well beyond the mere provision of more accurate distance measures through GIS. Travel Cost As the measurements of distance and travel time are crucial for this method (Bockstael (1995) provides a general introduction to the travel cost approach), Bateman and colleagues focus on the use of GIS to measure distance and travel times from a residential location to a recreational site as well as its substitutes in a series of papers based in the United Kingdom. A description of these GIS techniques is found in Bateman, Garrod, Brainard et al. (1996). In a comparison of simple straight-line distances versus GIS calculated distances via the road network, Bateman, Brainard, Garrod et al. (1999) show that the former approach can bias consumer

72

The international yearbook of environmental and resource economics

surplus measures up to 20 per cent compared to the latter approach. They also demonstrate the usefulness of GIS to perform benefits transfer of recreational benefits from a travel cost study for woodlands in Wales (Bateman, Lovett and Brainard, 1999) to estimate the demand for other potential forest sites throughout that country, an analysis only possible within a GIS framework. This line of research is further refined for woodlands in England in Brainard, Lovett, and Bateman (1999) and Brainard, Bateman and Lovett (2001). Hedonic Property Models The lack of research on the links between residential land use and nonpoint source pollution was noted above. However, a great deal of research on the effects of both positive and negative environmental externalities related to residential land use can be found in the hedonics literature. Some of this is reviewed in the previous Yearbook chapter by Bockstael and Irwin (2000) where they focus on the role of landscape amenities in hedonic models. For a very extensive review of the hedonic pricing literature on the effect of environmental externalities on residential prices, see Boyle and Kiel (2001), and an introduction to hedonic property models and their use in litigation can be found in the previous Yearbook chapter by Palmquist and Smith (2003). In many of these papers, ‘space’ is often reduced to a unidimensional measure of distance to the amenity or disamenity of interest, or distance to the nearest city. The importance of externalities in land use might not only be a function of distances to other points, but also involve other, more complex spatial relationships and concepts that can be modeled using spatial data and GIS. For example, the pattern of landscape features and land uses that surround a parcel of land are likely to have a major influence on its value and use, as individuals value the pattern of land uses surrounding their homes. A second example is that the view from a parcel of both manmade and natural features could affect residential land values. We discuss these two types of applications to demonstrate some of the additional uses of GIS tools beyond calculating the distances between features of interest. In order to test the effect of land use patterns on residential prices, Geoghegan, Wainger and Bockstael (1997) create spatial indices of land use fragmentation and diversity, borrowing from the landscape ecology literature. These indices were calculated for each residential land parcel at different scales in a model of residential land values for a region in Maryland. These variables were found to be statistically significant in the different empirical specifications of the model. This line of analysis on the effects of different types and sizes of open space at different spatial scales

Spatial environmental policy

73

in Maryland is continued in Irwin and Bockstael (2001), Irwin and Bockstael (2002), Irwin (2002), Geoghegan (2002), and Geoghegan, Lynch and Bucholtz (2003), while Acharya and Bennett (2001) use similar methods for a county in Connecticut, and Smith, Poulos and Kim (2001) focus on a county in North Carolina. In all of these models, the use of spatial data and GIS has improved models of land values and therefore the policy implications that can be drawn from them. Examples of the use of GIS to create variables to capture visibility attributes from a site can be found in a series of papers by Lake and colleagues (Lake, Lovett and Bateman et al., 1998; Lake, Bateman and Day et al., 2000; Lake, Lovett and Bateman et al., 2000) for Glasgow, Scotland. Paterson and Boyle (2002) use this approach for an area of Connecticut. Both sets of authors argue that, while the distance between a residential location and other points can affect residential land values, the topography of the area will also affect values, as both positive and negative externalities can be associated with the view from a location. In the former set of papers, the authors create a GIS based ‘viewshed’ variable for each observation. First, they use a Digital Elevation Model (DEM) to generate a digital topographic map of an area, in order to create elevation data for each observation. Second, they perform a survey of buildings to record the height of each building. Finally, with a land use map of the region, they combine the height data with the land use data to create a measure of the different types of land use visible from each observation up to 500 meters. These variables were then included with other typical hedonic variables in a model, where the viewshed variables were included with different weights to allow for diminishing effects over space. While most of the estimated coefficients in Lake, Lovett and Bateman et al. (2000) were statistically insignificant, those results could possibly be attributed to a misspecified model. In Paterson and Boyle (2002), similar visibility variables are constructed from the per cent of each land use type visible within one kilometer of a parcel, using a DEM and GIS land use data; the paper focuses on the econometric specification needed to use these data in a hedonic model. They use different specifications of these variables in the hedonic model to test if these viewshed variables affect residential land prices and if their exclusion biases the other estimated coefficients in the model. In addition, they investigate spatial econometric issues and find spatial autocorrelation in the residuals. As a result, they then test different specifications of spatial autoregressive models. Contrary to the Lake et al. results, these authors find the visibility variables are statistically significant and furthermore, the omission of these variables can lead to biased and inefficient estimates of the coefficients of the other environmental variables in the hedonic equation.

74

The international yearbook of environmental and resource economics

Contingent Valuation More varied research questions have begun to be asked in the contingent valuation literature. In a study of the values of preventing salt water intrusion into a freshwater ecosystem in Eastern England, Bateman and colleagues (Bateman et al., 2000) use GIS to calculate the distance from the address of each mail survey respondent to the wetland in question, to test if individual willingness-to-pay (WTP) measures were a function of this distance. These results were then compared to an approach that assumed a uniform spatial distribution of WTP across the assumed relevant population up to a certain arbitrarily defined distance with zero WTP after this distance. The spatially explicit approach yielded an aggregate value of WTP of one-sixth of the size of the spatially undifferentiated approach. The authors conclude that using the spatial information available in the data was a modeling improvement, by eliminating the need for arbitrarily defined spatial limits of the affected population. While this paper used GIS as a tool for creating a distance variable from an existing CV study, some of the later research by these authors focused on using some of the more advanced tools that can be found in some GIS packages to create threedimensional visualizations of different landscapes that could be used in survey instruments for CV or other stated preference approaches (Lovett et al., 2001; Appleton et al., 2002). While this previous research focused on using GIS as a tool in improving stated preference surveys, Johnston, Swallow and Bauer (2002) investigate how individuals react to different spatial features in a stated preference survey instrument of residential development for an area of Rhode Island. Subsequently, they examine the impacts of these spatial features on welfare estimation. The survey instrument included both spatial (e.g. shape and fragmentation of a development) and non-spatial (e.g. taxes) features, with the spatial attributes presented in map form. Similar to the hedonic study by Paterson and Boyle (2002) discussed previously, the results here show that the spatial aspects of the survey can influence the estimated marginal values of WTP for the non-spatial attributes. This conclusion has policy implications for stated preference surveys that implicitly assume separability between spatial and non-spatial attributes, as the estimated coefficients can suffer from omitted variable bias.

NATURAL RESOURCE MANAGEMENT While natural resource economics has a long history, with Hotelling (1931) as the foundation for non-renewable resources and Faustmann (1849) for

Spatial environmental policy

75

renewable resources, research on the spatial aspect of natural resource management issues in the economics literature has increased substantially in recent years. Differences exist between the physical aspects of nonrenewable and renewable resources, as well as the institutions associated with different types of resources, such as private property rights or common pool resources, but the most important aspects for spatial modeling are whether the resource is fixed in space or not, and whether spatial interactions could occur between separate locations of the resource. For example, non-renewable resources, discussed next, have no opportunity for physical interactions, but the locations of these resources can affect management decisions over space, while biological resources, either fixed in space, such as timber resources or mobile, such as fisheries, discussed subsequently, do have the potential for some degree of spatial interaction, leading to exciting advances in spatial modeling. Once again, these papers demonstrate the importance of correctly modeling the spatial interactions of the natural system under consideration as well as its relationship with human decision makers. Non-renewable Resources: Energy The classic result for non-renewable resource theory of Hotelling (1931) is that along an optimal extraction path, the price of the resource at any point in time is the sum of marginal extraction costs and marginal user cost (scarcity rents) and that the marginal user cost rises at the interest rate. A theoretical paper that brings together this Hotelling result with the Hotelling spatial model (1929) referred to in the introduction is Kolstad (1994). The model in this paper addresses the implication for the Hotelling result if deposits of the energy resources are separated in space as well as from consumers. The results of the model show that separation in space creates a price differential between otherwise homogenous resources and that the marginal user costs of the separated deposits are jointly determined, linking them through time and space. Forestry: Timber Management and Deforestation The original emphasis of timber management models was to maximize the value of the timber products deriving from a stand. These stands were assumed to be spatially homogeneous, with no spatial interactions between stands. These models assumed that the only benefit from the forest was the value of the timber, with no offsite impacts from the growing forest or from timber harvesting. In a pair of papers (Swallow and Wear, 1993; Swallow, Talukdar and Wear, 1997), these assumptions are relaxed. In these papers

76

The international yearbook of environmental and resource economics

the authors argue that ignoring interactions between stands can lead to faulty management and policy prescriptions, as the spatial relationships that lead to many ecosystem functions are ignored, such as the relationship between forest fragmentation and wildlife populations. By including these spatially explicit ecological interdependencies, Swallow and colleagues show that the optimal forestry management scheme can differ substantially from a non-spatial modeling approach: different harvesting periods for the different stands, different total benefits of recreation, and different forage availability to wildlife. Koskela and Ollikainen (2001) extend this framework by considering how these spatial interdependencies can change over time and deriving new optimal rotation lengths that depend on recreational access to amenity externalities. The results of these papers have important policy implications. For example, this approach can be used by public land managers who are managing the forest for multiple use and have to take into consideration non-timber outputs, such as wildlife, that are located on adjacent land that is not under their jurisdiction. In a series of econometric studies of timber yields in Wales (Bateman and Lovett, 1998, 2000a) a creative application of a GIS algorithm is used to spatially extrapolate existing data on explanatory variables to cover the entire country. As more observations were created, which included greater variability in the values of the observations, the authors conclude that their estimated results are more detailed as well as more robust. Using this improved estimated model, they then predict yields on lands not currently in forestry use. This approach is subsequently used to simulate the effect of replanting land in forestry on the net flux of carbon from tree growth, timber products and soil changes (Bateman and Lovett, 2000b). Tropical forest management issues have an important global policy context, owing to the linkages between tropical deforestation and the loss of biodiversity, as well as the decrease in carbon sequestration potential and those effects on global warming. The major uses of tropical forests include timber activities as in the temperate forest context, but to a greater degree than exists in temperate forests. Tropical forests are also used for gathering of fuelwood, agroforestry, and land clearing for cultivation. In this context, Albers (1996) develops a theoretical spatial model that incorporates the different potential land uses and their spatial interaction across plots. Tropical forest fragmentation and its spatial pattern have important effects on biodiversity and the ability of the forest to regenerate. The simulation model optimizes the pattern of land uses under different spatial and management scenarios, and investigates how these characteristics affect forest fragmentation and the associated ecological effects of the different scenarios. Albers concludes that interactions between land use patterns and forest recovery generate flexibility in spatial planning that can generate

Spatial environmental policy

77

income for land managers in the short run, while maintaining the option for large preserved areas in the long run. In a subsequent theoretical model, Robinson, Williams and Albers (2002) develop a spatial optimization model of non-timber forest product extraction by local peasant farmers from a buffer zone surrounding a forest reserve. The farmers differ according to market access and their opportunity cost of labor, important considerations in a developing country context (see also Vance and Geoghegan, 2004). The implication of this heterogeneity in how farmers respond to different government policies on the resulting pattern of resource extraction is striking. For example, villages that are exactly alike except for their market access costs will have different extraction patterns, so that even non-spatial government policies can have spatial effects that must also be taken into consideration when designing spatial policies like buffer zones for conservation. One of the most active areas of spatially explicit economic modeling is in the area of tropical deforestation. This is likely due to the availability of time series satellite data on land use that gives detailed spatial observations on tropical deforestation. In an earlier Yearbook, the general topic of tropical deforestation was covered (Van Kooten, Sedjo and Bulte, 1999). Kaimowitz and Angelsen (1998) provide an excellent review of deforestation models and other models of forest use in developing countries through the mid-1980s. A recent issue of Land Economics (vol. 77, no. 2, 2001) presents several papers on this topic as well. Finally, a specific introduction to spatially explicit models of tropical deforestation can be found in Nelson and Geoghegan (2002). While this literature has been covered in these previous reviews, some of the path breaking work in this literature for different developing countries can be found in the following papers for these specific countries: Belize (Chomitz and Gray, 1996), northern Mexico (Nelson and Hellerstein, 1997), Brazil (Pfaff, 1999; Mertens et al., 2002), Thailand (Cropper, Griffiths, and Mani, 1999; Cropper, Puri, and Griffiths, 2001), Panama (Nelson, Harris, and Stone, 2001), Vietnam (Muller and Zeller, 2002), Honduras (Munroe, Southworth and Tucker, 2002), and southern Mexico (Turner, Geoghegan and Foster, 2004). Water: Ground and Surface Management While some of the large literature on the spatial relationships between water quality and land use was discussed above in the subsection on nonpoint source pollution, this subsection focuses on models of water project management, specifically on models to analyse the loss of water from conveyance through the system. For a general introduction to the economics of water use, see the previous Yearbook chapter by Becker, Zeitouni and

78

The international yearbook of environmental and resource economics

Zilberman (2000). In a series of papers by Chakravorty and colleagues (Chakravorty, Hochman, and Zilberman, 1995; Umetsu and Chakravorty, 1998; Chakravorty and Umetsu, 2003) models are developed that investigate the implications of spatially explicit considerations for management of water projects. In the first paper, a spatial model of a water project is developed, with a centrally regulated utility supplying water to firms along a canal. These firms differ only by their location along the canal, and water is lost at a constant rate along the distribution system as well as on-farm. At each location, the optimal price of water, optimal investment by the utility in the distribution system, and each individual firm’s investment in irrigation technology is determined. The implications of this model across two scenarios are examined: one with uniform pricing and one with water trading between firms. Their results show that in a spatial model, policies that improve water conveyance will likely have negative distributional effects on upstream users that would have to be compensated for in some way. The later papers include both surface and ground water, where return flow is included, that is, some of the water that is ‘lost’ upstream becomes available to downstream users. If this return flow replenishes the water table, then the optimal solution is spatial specialization: upstream farmers use canal water and downstream farmers pump groundwater. In both modeling frameworks, the spatial distribution of benefits and costs of different modeling scenarios become apparent, which would not have been the case in a non-spatial framework. Fisheries: Fish Patches and Marine Reserves Historically, the primary focus of fisheries management models has been the common property feature of most fisheries. Often, owing to mismanagement, common property leads to open access problems resulting in overfishing and depletion of stocks. The policy responses to address these problems have included reduction in fishing season and the development of tradable permits. See the previous Yearbook chapter by Bjorndal and Munro (1998) for an introduction to the economics of fisheries management. In a series of papers by Sanchirico and Wilen (1999, 2001a, 2001b, 2002) the authors argue that these traditional models assume a spatially homogenous distribution of the resource and therefore the policy implications concerning the associated human behavior will likely be incorrect. They develop models that include the spatial configuration of the resource, focusing on the heterogeneous distribution of the fish populations and the interconnections among and between patches of fish. The economic variables also vary over time and space and spatially explicit fishing effort is determined endogenously. The main result of these models is that by

Spatial environmental policy

79

incorporating different forms of these dynamic spatial relationships, such important biological concepts as ‘source’ or ‘sink’ of the living resource can depend on both biological and socioeconomic parameters. Therefore, spatially differentiated management techniques, such as refuges or rotating harvesting zones that do not consider the interaction of human behavior and the resource over space could lead to suboptimal policies. These previous models are theoretical, and while useful for developing conceptual optimal spatially explicit fishery management policies, Smith (2002) argues that they do not go far enough in that they do not address the institutional aspects of any particular policy. Therefore, he develops two empirical models to predict how the magnitude and spatial distribution of fishing effort respond to changes in biological, economic, and oceanographic conditions. These models simultaneously incorporate both the participation and location choice for a commercial fishery, using data for California sea urchins on fishery management, weather, and opportunity costs. The first is based on aggregate monthly data and the second is a disaggregate approach based on data on individual fishermen. While the focus of this paper is on comparing the aggregate and disaggregate econometric approaches, it is the site-specificity of the data that allows the simulation of the effects of the spatial policies. This empirical work is continued in Smith and Wilen (2003, 2004), combining econometric models of harvester behavior with a spatially explicit and dynamic bioeconomic model of the fishery to simulate the effects of spatial closures on the fishery. Their results suggest that the conclusions from biological models might be overly optimistic in their predictions of the effects of spatial closures, as the dispersal of harvester behavior in response to the closure can mitigate some of the predicted benefits. Flora and Fauna: Management of Species Similar to these fishery models are spatial terrestrial flora and fauna models that analyse potential management schemes of these resources. In a series of papers by Bhat and colleagues (Huffaker, Bhat, and Lenhart, 1992; Bhat, Huffaker and Lenhart, 1993; Bhat, Huffaker and Lenhart, 1996), a spatial model of the dispersal behavior of wildlife, in this case beaver populations, is linked with an economic model of forestry management. An optimization model of minimizing the costs of timber damage from the beavers and trapping costs is developed. As beaver populations move over space, management of the beavers on one property can have unintended spillovers on contiguous properties. By taking these spatial interactions into account, the optimal beaver control strategy is derived for each location and time period, leading to a more efficient management

80

The international yearbook of environmental and resource economics

scheme than would be possible without taking into account the spatial dynamics of the system. In a spatial bioeconomic benefit-cost analysis of a biological agent to control an invasive species, Nordblom and co-authors (Nordblom, et al., 2002) analyse the use of the crown weevil to control an invasive weed in southern Australia. Given a model of the natural reproduction and geographical spread of the insect, the authors simulate its spatial and temporally specific location trajectories. While previous analyses have attempted to perform cost-benefit analysis of biological controls, this study is the first to calculate the geographical areas where, as well as how often, it would be economically efficient to release further populations of the insect, leading to decreased estimated costs in weed control. Biodiversity: Habitat Conservation and Reserves One important approach to conserving biodiversity and other environmental goods is to preserve specific parcels of land where values for those goods are high. The economic literature in this area has largely focused on how to efficiently allocate conservation funds, as in the applications of the Conservation Reserve Program discussed previously. Similarly, biological research on biodiversity preservation has focused on choosing sites that include the maximum number of species subject to a constraint on the number of sites chosen. However, this approach implicitly assumes the opportunity cost of land is constant across biodiversity sites (Polasky, Camm and Garber-Yonts, 2001). A preliminary US-wide study of a costeffective approach to selecting reserves, using county level data on agricultural land values and incidence of endangered species, can be found in Ando et al. (1998). A similar, but more detailed approach for Oregon with a finer geographic scale, a broader array of species, and assessed land values for all non-urban land, can be found in Polasky, Camm and GarberYonts (2001), where the benefits gained from preserving one parcel depend upon the spatial distribution of the other parcels that are preserved. Results for both papers demonstrate how the spatial distribution of sites chosen and total costs differ under the traditional biological approach compared to the cost-effective approach and require an integrated analysis of both biological and economic data. As many species require large contiguous patches of habitat, which can often encompass private land with multiple land owners, Smith and Shogren (2002) and Parkhurst et al. (2002) focus on designing a mechanism for private landowners to voluntarily retire land that borders on other land retired from use. In this incentive scheme, landowners are paid an ‘agglomeration bonus’ as part of their compensation from the government for these

Spatial environmental policy

81

spatially specific land retirements. While the game-theoretic results of the latter paper show that multiple Nash equilibria exist with the agglomeration bonus, laboratory experiments suggest that the no-bonus outcome always creates a fragmented landscape, while the bonus outcome results in the first-best contiguous habitat landscape. Agriculture: Precision Farming and Carbon Sequestration Agricultural land use can be a source of both positive and negative environmental externalities as discussed above. In this subsection, the use of agricultural land as a resource for the growing of crops and the sequestration of carbon to reduce global warming is highlighted. Many factors can vary spatially across an agricultural field: soil characteristics, water availability, nutrient levels, and pest populations. The goal of precision agriculture is to maximize this yield and to be more cost-effective in the application of inputs, using GPS and GIS technologies (described above), as well as yield monitoring sensors and variable rate application (VRA) equipment for input application. Precision agriculture, by also potentially reducing chemical input use, has the potential for environmental benefits, through the reduction in leaching and fertilizer runoff. Once again, this spatial approach to a natural resource use issue requires an interdisciplinary approach, bringing together economics, agronomy, soil science, and hydrology (Weiss, 1996). As observed rates of the adoption of precision farming technologies are fairly low, most of the current research in this area has focused on farm profitability to explain this low rate. Using an option value approach, Khanna, Isik and Winter-Nelson (2000) explain some of this low rate by showing that the spatial variability in the distribution in soil characteristics can delay the optimal timing of investment in precision agricultural technology for at least three years. For a recent review of other issues and this literature, see Bullock, Lowenberg-DeBoer and Swinton (2002). While most of the studies reviewed in Bullock, Lowenberg-DeBoer and Swinton (2002) are US based, an interesting application to traditional agriculture using ‘low-tech’ precision agriculture in Niger is found in Florax, Voortman and Brouwer (2002). The paper focuses on applying spatial econometric techniques to estimate yield functions for millet and the regression results show how dramatic differences between spatial and non-spatial specifications can be, with the spatial models yielding more accurate prescriptions for fertilizer application and yield predictions. Agricultural land, in addition to being an input into the production of food and fiber, is also a sink for atmospheric carbon, an issue of policy concern, given global warming. One potential policy option to reduce atmospheric carbon is to provide farmers with incentives to adopt new

82

The international yearbook of environmental and resource economics

management techniques that increase carbon sequestration in their soils. While this topic was covered in a previous Yearbook chapter (Antle and McCarl, 2003), the discussion here focuses on a spatially explicit, spatially disaggregate, field based simulation model to compare spatially homogenous per acre contracts to spatially heterogeneous per ton contracts to farmers (Antle et al., 2003). Land is assumed to be heterogeneous in both biophysical (e.g. soil carbon) and economic (e.g. farmgate prices) characteristics. The model is interdisciplinary, using both a crop ecosystem model that estimates changes in soil carbon, and an econometric model to quantify the responses of farmers to the different contracts. The results of the case study, based on the Northern Plains of the US, suggest that although the measurement costs to implement spatially homogenous contracts are lower, their overall cost per ton of carbon sequestered can be up to five times higher than the overall cost of spatially heterogeneous contracts, and the relative inefficiency increases with increased spatial heterogeneity of land.

CONCLUSIONS AND SUGGESTIONS FOR FUTURE RESEARCH In this chapter, we have attempted to make the case for the value of taking a spatially explicit approach in environmental and natural resource economics and the contribution this approach can make to policy analysis, while noting some of the costs of doing so. In some cases, a spatially explicit approach simply makes the policy options under consideration more efficient; in others, the outcomes can be more surprising, such as when non-spatial policies can affect spatial outcomes as was found in Robinson, Williams and Albers (2002). As we hope we have demonstrated, both theoretical and methodological advances in economic models and applications have increased in this area in the recent past, but this is only the beginning. Many advances in this literature are possible and many of the researchers cited in this chapter as well as others have such research currently underway. Still, much of the current literature has focused on the heterogeneous characteristics of space: economic actors or environmental resources have different attributes associated with their locations in space. More could be done theoretically, for example, in the realm of the spatial interaction among agents, as is done in Irwin and Bockstael (2002), where they develop a model of urban spatial structure that is a result of spatial interdependences among spatially distributed economic agents. Many researchers are making use of spatial data and GIS, using these to determine assorted geographical distances,

Spatial environmental policy

83

creating buffers, and using digital elevation models to generate variables on slope and elevation, but the power of GIS software is expanding rapidly and more creative uses of spatial data are possible. While spatial econometrics has advanced significantly in recent years, it is still a relatively new subfield of econometrics, with much more work to be done. For example, the development of spatial econometric methods to handle limited dependent variable models is still in its infancy (see Fleming (2004) for an overview). Finally, other ‘types’ of space besides geographical space could be considered in environmental and natural resource economic models as appropriate. For example, in Akerloff (1997) the concept of ‘social distance’, which is a function of social class, is developed in a model of decisions involving education and fertility choices. In looking at the literature related to stationary sources of pollution, we have seen little or no research that takes an aggressively spatial approach to the analysis of environmental regulation. Nearly all papers simply use the plant’s location to provide information about the regulatory pressures the plant will be facing, without allowing for interactions across plants in their decisions. The analysis of environmental justice issues has brought a somewhat greater spatial focus, if only because detailed demographic data at the county or even block group data are being connected to the plant’s location to define the ‘nearby’ population. Gray and Shadbegian (2004b) use spatially delineated models of air and water pollution flows to give quantitative measure of the marginal benefits from pollution reductions at specific mills; Shadbegian et al. (2004) use data on the impacts of sulfur dioxide emissions from each coal fired power plant on air quality in each county of the US. These data connecting pollution from source to recipient can be combined with detailed demographic data to measure the impacts of emissions on different demographic groups and on different regions of the country. One possible area for ‘creative’ use of spatial data could come from combining similarly sophisticated models of the spatial diffusion of pollution with research on possibly non-linear effects of pollution. Such data could show the amount of pollution flowing from each source to each neighborhood (or person) affected by the pollution. This could be used to expand Millimet and Slottje’s (2002) analyses to see whether changes in the distribution of pollution across plants could seriously affect the calculation of the marginal benefits from pollution reduction at an individual source. Such non-linearities would require a full-fledged spatial analysis of pollution flows from nearby plants to see whether regulators might need to take the emissions from surrounding plants into account when setting permit levels, at least in relatively crowded urban areas. One preliminary example of a fully spatial analysis in the area of environmental regulation of stationary sources is provided by Gray and

84

The international yearbook of environmental and resource economics

Shadbegian (2004a). Each plant in their model decides whether to comply with regulations and how much pollution to emit, based on its own regulatory experience as well as regulatory activity at ‘nearby’ plants. They find evidence that inspections at nearby plants matter for compliance, and find spatial dependence across nearby plants for compliance status. Their model explains part of the observed spatial dependence in compliance, indicating the presence of additional spatially related unobservable factors that are also influencing compliance. Another possible avenue of research involves getting new measures of regulatory stringency, based not only on differences in overall stringency but also identifying situations where regulations would be stricter for firm A in jurisdiction X than for firm B in jurisdiction Y. Berman and Bui (2001) have done some detailed measures for the especially stringent local air pollution regulations affecting plants near Los Angeles, and have compared those plants to plants in Texas and in other parts of the US as well as comparing them to other plants in the Los Angeles area which face different local regulations. A simpler and more comprehensive test of this sort is used in both Becker (2001) and Greenstone (2002), where the different air pollutants coming from plants in each industry are identified, and then interacted on a county-level basis with the attainment status for that pollutant in the county. Becker finds significantly higher air pollution abatement costs, and Greenstone finds significantly less economic activity, when a plant that is a heavy emitter of a pollutant is in a county which is in non-attainment for that pollutant. Previously we suggested that more powerful analyses are possible, using spatial data and GIS in ways that go beyond spatial data management and the creation of additional spatial variables. An exciting area of new spatial environmental economic research is the attempt to establish causality through the use of natural experiments (Rosenzweig and Wolpin, 2000): either truly natural experiments such as the effects on housing prices of the occurrence of a nearby hurricane (Hallstrom and Smith, 2004; Carbone, Hallstrom and Smith, 2004) or manmade policy experiments, such as the effects of spatially delineated growth control regulations on suburban development (Geoghegan and Bockstael, 2003). Causation is notoriously difficult to establish using statistical analysis, as one never observes the outcome of both treatment and control on the same unit of observation at the same moment in time. In addition, regression analysis cannot guarantee independence between the varying characteristic of the observational units and the treatment status. This is especially true in spatial analyses, where explanatory variables can spatially vary in a systematic way, causing correlations between variables without any causal relationship between them. Other recent studies have used the impact of economic slowdowns

Spatial environmental policy

85

on emissions as a natural experiment (on the reasonable assumption that recessions are not deliberately started in order to reduce pollution). Kahn (1999) looks at the decline of the Rust Belt in the US to see which cities had improvements in environmental quality; Chay and Greenstone (2003) show that recession induced reductions in air emissions led to reductions in infant mortality; Kahn (2003) examines the improvements in air quality in Eastern European cities after the decline of communism. Natural experiments are a way to overcome the inability to observe the counterfactual by creating some scheme in which to match each observation that has received the treatment with a non-treatment observation. One method of matching observations is based on the propensity score approach (Rosenbaum and Rubin, 1983), where the probability of receiving the treatment is estimated, followed by a calculation of the propensity scores (i.e. predicted probabilities) for all observations. A matching procedure is then implemented, such as nearest-neighbor, where each treatment observation is paired with a control observation with the closest propensity score. Any difference in responses within pairs to the treatment is therefore a measure of the treatment effect. This method has been widely used in the biomedical literature; within economics it has mostly been applied in labor economics. An interesting environmental application is found in List et al. (2003), investigating the effects of air quality (attainment status) on new plant locations in counties in New York State. Being in non-attainment is the treatment, so for each county in non-attainment, a comparable county in attainment was selected, with the ‘comparable’ county chosen as the one with the closest value on the propensity score. They find much larger results for the propensity score analysis than they do for ordinary regressions, and conclude that traditional analyses grossly underestimate the impact of the stringency of regulation. Greenstone (2004) also uses a propensity score analysis, but finds that little of the overall improvement in SO2 concentrations can be attributed to counties being in non-attainment status. Ongoing research in this area includes the investigation in Margolis, Osgood and List (2004) of the effects of the ‘critical habitat’ designation of the Endangered Species Act on habitat destruction. Their preliminary results suggest that landowners are preemptively destroying habitat on their land parcels to avoid the potential of future designation. Another matching approach uses the regression discontinuity design, where a treatment effect is identified by exploiting a discontinuity in the treatment, with similar observations on either side of the discontinuity matched and then tested as in the propensity score approach. For example, air pollution attainment status category for an area could be held fixed and then its ambient air quality controlled for separately (e.g. comparing two

86

The international yearbook of environmental and resource economics

non-attainment counties, one with much worse air quality, or two counties with relatively similar air quality, one just barely in attainment and the other just barely out of attainment) which might help distinguish between the procedural stringency due to attainment status and general pressures to reduce emission levels due to poor air quality. This technique is used by Chay and Greenstone (2004) to show that reductions in air pollution (driven by a county being in non-attainment status) are associated with significant increases in housing prices. Many other applications of spatial analysis to environmental policy should be possible, as long as environmental regulations are a function of some geographic boundaries, so that policies can change abruptly at these boundaries, causing observations on either side of the boundary to be in a different treatment status. Gray and Shadbegian (2004a) find that inspections at nearby plants affect compliance only when the nearby plant is in the same state. Inspections at plants equally close but in a neighboring state do not have such an effect, so state borders matter, consistent with the fact that much of environmental regulation is done by state regulators. In ongoing research related to the Maryland land use studies discussed above, Geoghegan and Bockstael (2003) exploit a spatial discontinuity to estimate the effects of government regulations on growth controls for new residential development. The treatment effect and matching strategy used are based on the observation that the ‘value’ of each new development moratorium changes abruptly at its boundary, where other variables are either changing continuously or remaining constant. This discontinuity offers an identification strategy for statistical analysis, as well as controlling for omitted variables by effectively shifting the model of each location, with all parcels within 500 meters of a moratorium boundary identified and each group of parcels that share a common border assigned the same dummy variable. The preliminary results suggest that these moratoria are more effective at slowing down larger developments over smaller ones, where larger is defined as a subdivision of a parcel with the potential for 100 or more housing lots. While the previous paper uses 500 meters as a cutoff distance for an initial matching strategy, other ongoing research that combines a regression discontinuity design with propensity scores can be found in Poulos and Smith (2002). In this example, the authors investigate the loss in property values in a neighborhood bisected by a new highway, using a repeat sales hedonic analysis to determine the potential compensation due these landowners as a result of the road building policy. They make the case that the specification of the control group has been mostly undervalued in the hedonic literature, but that a careful specification of both the treatment group and the control group is necessary for such policy analyses. Therefore, instead of simply relying on an ad hoc cutoff distance for matching, they use propensity scores

Spatial environmental policy

87

to evaluate the selection of the geographic boundary for the control group. Their results suggest that the new road reduced real property values by up to 20 per cent. In closing, we optimistically forecast that research in the area of spatial environmental and natural resource economics will expand, with theoretical and methodological advances, while the costs of spatial analysis that we noted in the introduction will decrease, as further cross-fertilization between subfields of economics continues and more of this literature infiltrates into graduate programs, generating a new generation of graduate students better able to tackle these problems.

REFERENCES Acharya G. and L. Bennett (2001), ‘Valuing open space and land-use patterns in urban watersheds’, Journal of Real Estate Finance and Economics, 22(2/3), 221–37. Akerlof, G.A. (1997), ‘Social distance and social decisions’, Econometrica, 65(5), 1005–27. Albers, H.J. (1996), ‘Modeling ecological constraints on tropical forest management: spatial interdependence, irreversibility and uncertainty’, Journal of Environmental Economics and Management, 30: 73–94. Anas, A., R. Arnott and K.A. Small (1998), ‘Urban spatial structure’, Journal of Economic Literature, 36, 1426–65. Ando, A., J. Camm, S. Polasky and A. Solow (1998), ‘Species distributions, land values and efficient conservation’, Science, 279, 2126–8. Antle, J., S. Capalbo, S. Mooney, E. Elliott and K. Paustian (2003), ‘Spatial heterogeneity, contract design, and the efficiency of carbon sequestration policies for agriculture’, Journal of Environmental Economics and Management, 46, 231–50. Antle, J.M. and B.A. McCarl (2003), ‘The economics of carbon sequestration in agricultural soils’, in H. Folmer and T. Tietenberg (eds), The International Yearbook of Environmental and Resource Economics 2002/2003, Cheltenham, UK and Northampton, MA: Edward Elgar. Anselin, L. (2002), ‘Under the hood: issues in the specification and interpretation of spatial regression models’, Agricultural Economics, 27, 247–67. Appleton, K.J., A.A. Lovett, G. Sunnenberg and T.D. Dockerty (2002), ‘Rural landscape visualisation from GIS databases: a comparison of approaches, options and problems’, Computers, Environment and Urban Systems, 26, 141–62. Arora, S. and T.N. Cason (1999), ‘Do community characteristics influence environmental outcomes? Evidence from the Toxic Release Inventory’, Southern Economic Journal, 65, 691–716. Babcock, B.A., P.G. Lakshminarayan, J. Wu and D. Zilberman (1996), ‘The economics of a public fund for environmental amenities: a study of CRP contracts’, American Journal of Agricultural Economics, 78, 961–71. Babcock, B.A., P.G. Lakshminarayan, J. Wu and D. Zilberman (1997), ‘Targeting tools for the purchase of environmental amenities’, Land Economics, 73(1), 325–39.

88

The international yearbook of environmental and resource economics

Barcena-Ruiz, J.C. and M.B. Garzon (2003), ‘Strategic environmental standards, wage incomes and the location of polluting firms’, Environmental and Resource Economics, 24, 121–39. Bastian, C.T., D.M. McLeod, M.J. Germino, W.A. Reiners and B.J. Blasko (2002), ‘Environmental amenities and agricultural land values: a hedonic model using geographical information systems data’, Ecological Economics, 40, 337–49. Bateman, I.J., J.S. Brainard, G.D. Garrod and A.A. Lovett (1999), ‘The impact of journey origin specification and other measurement assumptions upon individual travel cost estimates of consumer surplus: a geographical information systems analysis’, Regional Environmental Change, 1(1), 24–30. Bateman, I.J., G.D. Garrod, J.S. Brainard and A.A. Lovett (1996), ‘Measurement, valuation and estimation issues in the travel cost method: a geographical information systems approach’, Journal of Agricultural Economics, 47(2), 191–205. Bateman, I.J., I.H. Langford, N. Nishikawa and I. Lake (2000), ‘The Axford debate revisited: a case study illustrating different approaches to the aggregation of benefits data’, Journal of Environmental Planning and Management, 43(2), 291–302. Bateman, I.J. and A.A. Lovett (1998), ‘Using geographical information systems (GIS) and large area database to predict yield class: a study of Sitka spruce in Wales’, Forestry, 71(2), 147–68. Bateman, I.J. and A.A. Lovett (2000a), ‘Modelling and mapping timber values using geographical information systems’, Quarterly Journal of Forestry, 94(2), 127–38. Bateman, I.J. and A.A. Lovett (2000b), ‘Modelling and valuing carbon sequestration in softwood and hardwood tress, timber products and forest soils’, Journal of Environmental Management, 60(4), 310–23. Bateman, I.J., A.A. Lovett and J.S. Brainard (1999), ‘Developing a methodology for benefit transfers using geographical information systems: modelling demand for woodland recreation’, Regional Studies, 33(3), 191–205. Bateman, I.J., A.A. Lovett and J.S. Brainard (2003), Applied Environmental Economics: A GIS Approach to Cost-Benefit Analysis, Cambridge: Cambridge University Press. Becker, N., N. Zeitouni and D. Zilberman (2000), ‘Issues in the economics of water use’, in T. Tietenberg and H. Folmer (eds), The International Yearbook of Environmental and Resource Economics 2000/2001, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 55–99. Becker, R. (2001), ‘Air pollution abatement costs under the Clean Air Act: evidence from the PACE survey’, Center for Economic Studies working paper 01-12. Becker, R. and J.V. Henderson (1999), ‘Costs of Air Quality Regulation’, NBER working paper 7038. Becker, R. and J.V. Henderson (2000), ‘Effects of air quality regulation on polluting industries’, Journal of Political Economy, 108, 379–421. Been, V. and F. Gupta (1997), ‘Coming to a nuisance or going to a barrios? A longitudinal analysis of environmental justice claims’, Ecology Law Quarterly, 24, 1–56. Bell, K.P. and E.G. Irwin (2002), ‘Spatially explicit micro-level modeling of land use change at the rural-urban interface’, Agricultural Economics, 27, 217–32. Berman, E. and L. Bui (2001), ‘Environmental regulation and productivity: evidence from oil refineries’, Review of Economics and Statistics, 83, 498–510.

Spatial environmental policy

89

Bhat, M.G., R.G. Huffaker and S.M. Lenhart (1993), ‘Controlling forest damage by dispersive beaver populations: centralized optimal management strategy’, Ecological Applications, 3(3), 518–30. Bhat, M.G., R.G. Huffaker and S.M. Lenhart (1996), ‘Controlling transboundary wildlife damage: modeling under alternative management scenarios’, Ecological Modeling, 92, 215–24. Bjorndal, T. and G.R. Munro (1998), ‘The economics of fisheries management: a survey’, in T. Tietenberg and H. Folmer (eds), The International Yearbook of Environmental and Resource Economics 1998/1999, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 153–88. Bockstael, N.E. (1995), ‘Travel cost models’, in D.W. Bromley (ed.), Handbook of Environmental Economics, Oxford: Blackwell. Bockstael, N.E. (1996), ‘Modeling economics and ecology: the importance of a spatial perspective’, American Journal of Agricultural Economics, 78(5), 1168–80. Bockstael, N.E. and E.G. Irwin (2000), ‘Economics and the land use-environment link’, in T. Tietenberg and H. Folmer (eds), The International Yearbook of Environmental and Resource Economics 2000/2001, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 1–54. Bohm, P. and P.-O. Hesselborn (1999), ‘Transportation and environmental policy’, in H. Folmer and T. Tietenberg (eds), The International Yearbook of Environmental and Resource Economics 1999/2000, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 1–43. Boisvert, R.N., T.M. Schmit and A. Regmi (1997), ‘Spatial, productivity, and environmental determinants of farmland values’, American Journal of Agricultural Economics, 79(5), 1657–64. Boyle, M.A. and K.A. Kiel (2001), ‘A survey of house price hedonic studies of the impact of environmental externalities’, Journal of Real Estate Literature, 9(2), 117–44. Braden, J.B., G.V. Johnson, A. Bouzaher and D. Miltz (1989), ‘Optimal spatial management of agricultural pollution’, American Journal of Agricultural Economics, 61, 404–13. Brainard, J.S., I.J. Bateman and A.A. Lovett (2001), ‘Modelling demand for recreation in English woodlands’, Forestry, 74(5), 423–38. Brainard, J.S., A.A. Lovett and I.J. Bateman (1999), ‘Integrating geographical information systems into travel cost analysis and benefit transfer’, International Journal of Geographical Information Systems, 13(3), 227–46. Bullock, D.S., J. Lowenberg-DeBoer and S.M. Swinton (2002), ‘Adding value to spatially managed inputs by understanding site-specific yield response’, Agricultural Economics, 27, 233–45. Carbone, J.C., D.G. Hallstrom and V.K. Smith (2004), ‘Can natural experiments measure behavioral responses to environmental risks?’, manuscript. Carpentier, C.L., D.L. Bosch and S.S. Batie (1998), ‘Using spatial information to reduce costs of controlling agricultural nonpoint source pollution’, Agricultural and Resources Review, 27, 72–84. Chakravorty, U., E. Hochman and D. Zilberman (1995), ‘A spatial model of optimal water conveyance’, Journal of Environmental Economics and Management, 29, 25–41. Chakravorty, U. and C. Umetsu (2003), ‘Basinwide water management: a spatial model’, Journal of Environmental Economics and Management, 45, 1–23.

90

The international yearbook of environmental and resource economics

Chay, K. and M. Greenstone (2003), ‘The impact of air pollution on infant mortality: evidence from geographic variation in pollution shocks induced by a recession’, Quarterly Journal of Economics, 118, 1121–67. Chay, K. and M. Greenstone (2004), ‘Does air quality matter? Evidence from the housing market’, working paper. Chay, K., C. Dobkin and M. Greenstone (2003), ‘The Clean Air Act of 1970 and adult mortality’, Journal of Risk and Uncertainty, 27, 279–300. Cohen, M.A. (1999), ‘Monitoring and enforcement of environmental policy’, in H. Folmer and T. Tietenberg (eds), The International Yearbook of Environmental and Resource Economics 1999/2000, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 44–106. Deacon, R.T., D.S. Brookshire, A.C. Fisher, A.V. Kneese, C.D. Kolstad, D. Scrogin, V.K. Smith, M. Ward, J. Wilen (1998), ‘Research trends and opportunities in environmental and natural resource economics’, Environmental and Resource Economics, 11(3–4), 383–97. Decker, C. (2003), ‘Corporate environmentalism and environmental statutory permitting’, Journal of Law and Economics, 46, 103–29. Deily M.E. and W.B. Gray (1991), ‘Enforcement of pollution regulations in a declining industry’, Journal of Environmental Economics and Management, 21, 260–74. Dosi, C. and T. Tomasi (eds) (1994), Nonpoint Source Pollution Regulation: Issues and Analysis, Dordrecht: Kluwer Academic. Eskeland, G.S. and A.E. Harrison (2003), ‘Moving to greener pastures? Multinationals and the pollution haven hypothesis’, Journal of Development Economics, 70, 1–23. Faustmann, M. (1849), ‘Berechnung des Werthes welschen Waldboden sowie noch nicht haubare Holzbestände für die Waldwirtschaft besitzen’, Allegemeine Forst und Jagd Zeitung, 25, 441–55. Translated into English in 1995 as ‘Calculation of the value which forest land and immature stands possess for forestry’, Journal of Forest Economics, 1(1), 7–44. Fleming, M. (2004), ‘Techniques for estimating spatially dependent discrete choice models’, in L. Anselin, R. Florax and S. Rey (eds), Advances in Spatial Econometrics, Methodology, Tools and Applications, Berlin: Springer, pp. 145–68. Fujita, M., P. Krugman, and A.J. Venables (1999), The Spatial Economy: Cities, Regions and International Trade, Cambridge, MA: MIT Press. Fujita, M. and J.-F. Thisse (2002), Economics of Agglomeration: Cities, Industrial Location and Regional Growth, Cambridge: Cambridge University Press. Geoghegan, J. (2002), ‘The value of open spaces in residential land use’, Land Use Policy, 19(1), 91–8. Geoghegan, J. and N.E. Bockstael (2003), ‘Testing for the effect of growth control measures using quasi-experimental design’, paper presented at the Association of Environmental and Resource Economists Workshop on Spatial Theory, Modeling, and Econometrics in Environmental and Resource Economics, Madison, WI, June 2003. Geoghegan, J., L. Lynch and S. Bucholtz (2003), ‘Capitalization of open spaces into housing values and the residential property tax revenue impacts of agricultural easement programs’, Agricultural and Resource Economics Review, 32(1), 33–45. Geoghegan, J., L. Wainger and N.E. Bockstael (1997), ‘Spatial landscape indices in a hedonic framework: an ecological economics analysis using GIS’, Ecological Economics, 23(3), 251–64.

Spatial environmental policy

91

Gerking, S. and J. List (2001), ‘Spatial economic aspects of the environment and environmental policy’, in H. Folmer et al. (eds), Frontiers of Environmental Economics, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 57–84. Goetz, R.U. and D. Zilberman (2000), ‘The dynamics of spatial pollution: the case of phophorus runoff from agricultural land’, Journal of Economic Dynamics and Control, 24, 143–63. Gray, W.B. (1997), ‘Manufacturing plant location: does state pollution regulation matter?’, NBER working paper no. 5880. Gray, W.B. and M.E. Deily (1996), ‘Compliance and enforcement: air pollution regulation in the U.S. steel industry’, Journal of Environmental Economics and Management, 31, 96–111. Gray, W.B. and R.J. Shadbegian (1998), ‘Environmental regulation, investment timing, and technology choice’, Journal of Industrial Economics, 46, 235–56. Gray, W.B. and R.J. Shadbegian (2004a), ‘The environmental performance of polluting plants: a spatial econometric approach’, presented at January 2004 annual meetings of ASSA in San Diego; also presented at June 2003 AERE Summer Workshop in Madison, WI. Gray, W.B. and R.J. Shadbegian (2004b), ‘ “Optimal” pollution abatement – whose benefits matter and how much?’, Journal of Environmental Economics and Management, 39. Greenstone, M. (2002), ‘The impacts of environmental regulations on industrial activity: evidence from the 1970 and 1977 Clean Air Act amendments and the census of manufactures’, Journal of Political Economy, 110, 1175–1219. Greenstone, M. (2004), ‘Did the Clean Air Act cause the remarkable decline in sulfur dioxide concentrations?’, Journal of Environmental Economics and Management, 47, 585–611. Hallstrom, D.G. and V.K. Smith (2004), ‘Market responses to hurricanes: using nature’s “natural experiments” to evaluate risk preferences’, manuscript. Hamilton, J. (1993), ‘Politics and social costs: estimating the impact of collective action on hazardous waste facilities’, Rand Journal of Economics, 24, 101–25. Hamilton, J. (1995), ‘Testing for environmental racism: prejudice, profits, political power?’, Journal of Policy Analysis and Management, 14, 107–32. Hardie, I.W., P.J. Parks and G.C van Kooten (2004), ‘Land use decisions and policy at the intensive and extensive margin’, in T. Tietenberg and H. Folmer (eds), The International Yearbook of Environmental and Resource Economics 2004/2005, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 101–38. Harrison Jr., D. (2003), ‘Tradable permits for air quality’, in H. Folmer and T. Tietenberg (eds), The International Yearbook of Environmental and Resource Economics 2002/2003, Cheltenham, UK and Northampton, MA: Edward Elgar. Helland, E. and A.B. Whitford (2003), ‘Pollution incidence and political jurisdiction: evidence from the TRI’, Journal of Environmental Economics and Management, 46(3), 403–24. Henderson, J.V. (1996), ‘Effects of air quality regulation’, American Economic Review, 86, 789–813. Hochman, E, D. Pines and D. Zilberman (1977), ‘The effects of pollution taxation on the pattern of resource allocation: the downstream diffusion case’, Quarterly Journal of Economics, 91, 626–38. Hotelling, H. (1929), ‘Stability in competition’, The Economic Journal, 39, 41–57.

92

The international yearbook of environmental and resource economics

Hotelling, H. (1931), ‘The economics of exhaustible resources’, Journal of Political Economy, 39(2), 137–75. Huffaker, R.G., M.G. Bhat and S.M. Lenhart (1992), ‘Optimal trapping strategies for diffusing nuisance-beaver populations’, Natural Resource Modeling, 6(1), 71–98. Irwin, E.G. (2002), ‘The effects of open space on residential property values’, Land Economics, 78, 465–80. Irwin, E.G. and N.E. Bockstael (2001), ‘The problem of identifying land use spillovers: measuring the effects of open space on residential property values’, American Journal of Agricultural Economics, 83(3), 698–704. Irwin, E.G. and N.E. Bockstael (2002), ‘Interacting agents, spatial externalities, and the endogenous evolution of residential land use pattern’, Journal of Economic Geography, 2(1), 31–54. Irwin, E.G. and J. Geoghegan (2001), ‘Theory, data, methods: developing spatiallyexplicit economic models of land use change’, Agriculture, Ecosystems, and Environment, 84, 7–24. Jenkins, R., K. Maguire and C. Morgan (2002), ‘Host community compensation and municipal solid waste landfills’, US Environmental Protection Agency, National Center for Environmental Economics working paper 2002–04. Jeppesen, T. and H. Folmer (2001), ‘The confusing relationship between environmental policy and location behaviour of firms: a methodological review of selected case studies’, Annals of Regional Science, 35, 523–46. Jeppesen, T., J.A. List and H. Folmer (2002), ‘Environmental regulations and new plant location decisions: evidence from a meta-analysis’, Journal of Regional Science, 42, 19–49. Johnston, R.J., S.K. Swallow and D.M. Bauer (2002), ‘Spatial factors and stated preference values for public goods’, Land Economics, 78(4), 481–500. Kahn, M. (1999), ‘The silver lining of Rust Belt manufacturing decline’, Journal of Urban Economics, 46, 360–76. Kahn, M. (2003), ‘New evidence on Eastern Europe’s pollution progress’, Topics in Economic Analysis and Policy, 3. Kaimowitz, D. and A. Angelsen (1998), Economic Models of Tropical Deforestation: A Review, Bogor: Center for International Forestry Research. Keller, W. and A. Levinson (2002), ‘Pollution abatement costs and foreign direct investment inflows to US states’, Review of Economics and Statistics, 84, 691–703. Khanna, M., M. Isik and A. Winter-Nelson (2000), ‘Investment in site-specific crop management under uncertainty: implications for nitrogen pollution control and environmental policy’, Agricultural Economics, 24, 9–21. Khanna, M., W. Yang, R. Farnsworth and H. Onal (2003), ‘Cost-effective targeting of land retirement to improve water quality with endogenous sediment deposition coefficients’, American Journal of Agricultural Economics, 85(3), 538–53. Kolstad, C. (1994), ‘Hotelling rents in Hotelling space: product differentiation in exhaustible resource markets’, Journal of Environmental Economics and Management, 26, 163–80. Koskela, E. and M. Ollikainen (2001), ‘Optimal private and public harvesting under spatial and temporal interdependence’, Forest Science, 47(4), 484–96. Kreisel, W., T.J. Centner and A.G. Keeler (1996), ‘Neighborhood exposure to toxic releases: are there racial inequities?’, Growth and Change, 27, 479–99. Krugman, P. (1995), Development, Geography, and Economic Theory, Cambridge, MA: MIT Press.

Spatial environmental policy

93

Krugman, P. (1998), ‘Space: the final frontier’, Journal of Economic Perspectives, 12(2), 161–74. Krumm, R.D. and D. Wellisch (1995), ‘On the efficiency of environmental instruments in a spatial economy’, Environmental and Resource Economics, 6, 87–98. Lake, I.R., I.J. Bateman, B.H. Day and A.A. Lovett (2000), ‘Improving land compensation procedures via GIS and hedonic pricing’, Environment and Planning, 18, 681–96. Lake, I.R., A.A. Lovett, I.J. Bateman and B. Day (2000), ‘Using GIS and large-scale digital data to implement hedonic pricing studies’, International Journal of Geographical Information Systems, 14(6), 521–41. Lake, I.R., A.A. Lovett, I.J. Bateman and I.H. Langford (1998), ‘Modelling environmental influences on property prices in an urban environment’, Computers, Environment and Urban Systems, 22(2), 121–36. Lintner, A.M. and A. Weersink (1999), ‘Endogenous transport coefficients: implications for improving water quality from multicontaminants in an agricultural watershed’, Environmental and Resource Economics, 14, 269–96. List, J.A. (2001), ‘US county-level determinants of inbound FDI: evidence from a two-step modified count data model’, International Journal of Industrial Organization, 19, 953–73. List, J.A. and W.W. McHone (2000), ‘Measuring the effects of air quality regulations on “dirty” firm births: evidence from the neo- and mature-regulatory periods’, Papers in Regional Science, 79, 177–90. List, J.A., W.W. McHone and D.L. Millimet (2003), ‘Effects of air quality regulation on the destination choice of relocating plants’, Oxford Economic Papers, 55, 657–78. List, J.A., D.L. Millimet, P.G. Fredriksson and W.W. McHone (2003), ‘Effects of environmental regulations on manufacturing plant births: evidence from a propensity score matching estimator’, The Review of Economics and Statistics, 85(4), 944–52. Lovett, A.A., J.R. Kennaway, G. Sunnenberg, R.N. Cobb, P.M. Dolman, T. O’Riordan and D.B. Arnold (2001), ‘Visualising sustainable agricultural landscape’, in P. Fisher and D. Unwin (eds), Virtual Reality in Geography, London: Taylor and Francis, pp. 102–30. Magat, W.A. and W.K. Viscusi (1990), ‘Effectiveness of the EPA’s regulatory enforcement: the case of industrial effluent standards’, Journal of Law and Economics, 33, 331–60. Margolis, M., D.E. Osgood and J.A. List (2004), ‘Measuring the preemption of regulatory takings in the US Endangered Species Act: evidence from a natural experiment’, Department of Agricultural and Resource Economics, University of Maryland working paper, College Park. Markusen, J., E. Morey and N. Olewiler (1995), ‘Competition in regional environmental policies when plant locations are endogenous’, Journal of Public Economics, 56, 55–77. Mertens, B., R. Poccard-Chapuis, M.-G. Piketty, A.-E. Lacques and A. Venturieri (2002), ‘Crossing spatial analyses and livestock economics to understand deforestation processes in the Brazilian Amazon: the case of Sao Felix do Xingu in South Para’, Agricultural Economics, 27, 269–94. Millimet, D.L. and D. Slottje (2002), ‘Environmental compliance costs and the distribution of emissions in the US’, Journal of Regional Science, 42, 87–105.

94

The international yearbook of environmental and resource economics

Muller, D. and M. Zeller (2002), ‘Land use dynamics in the central highlands of Vietnam: a spatial modeling combining village survey data with satellite imagery interpretation’, Agricultural Economics, 27, 333–54. Munroe, D.K., J. Southworth and C.M. Tucker (2002), ‘The dynamics of landcover change in Western Honduras: exploring spatial and temporal complexity’, Agricultural Economics, 27, 355–69. Nelson, G.C. and J. Geoghegan (2002), ‘Modeling deforestation and land use change: sparse data environment’, Agricultural Economics, 27(3), 201–16. Nijkamp, P. (1999), ‘Environment and regional economics’, in J.C.J.M Van der Bergh (ed.), Handbook of Environmental and Resource Economics 1996/97, Cheltenham, UK and Northampton, MA: Edward Elgar. Nordblom, T.L., M.J. Smith, A. Swirepik, A.W. Sheppard and D.T. Briese (2002), ‘Spatial economics of biological control: investing in new releases of insects for earlier limitation of Paterson’s Curse in Australia’, Agricultural Economics, 27, 403–24. Oates, W. and R. Schwab (1988), ‘Economic competition among jurisdictions: efficiency-enhancing or distribution-inducing?’, Journal of Public Economics, 64, 416–24. Palmquist, R.B. and V.K. Smith (2003), ‘The use of hedonic property value techniques for policy and litigation’, in H. Folmer and T. Tietenberg (eds), The International Yearbook of Environmental and Resource Economics 2002/2003, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 115–64. Parkhurst, G.M., J.F. Shogren, C. Bastian, P. Kviv, J. Donner and R.B.W. Smith (2002), ‘Agglomeration bonus: an incentive mechanism to reunite fragmented habitat for biodiversity conservation’, Ecological Economics, 41(2), 305–28. Paterson, R.W. and K.J. Boyle (2002), ‘Out of sight, out of mind? Using GIS to incorporate visibility in hedonic property value models’, Land Economics, 78(3), 417–25. Plantinga A.J. and D.J. Miller (2001), ‘Agricultural land values and the value of rights to future land development’, Land Economics, 77(1), 56–67. Polasky, S., J.D. Camm and B. Garber-Yonts (2001), ‘Selecting biological reserves cost-effectively: an application to terrestrial vertebrate conservation in Oregon’, Land Economics, 77(1), 68–78. Poulos, C. and V.K. Smith (2002), ‘Transparency and takings: applying an RD design to measure compensation’, manuscript. Proost, S. and J.B. Braden (eds), (1998), Climate Change, Transport and Environmental Policy: Empirical Applications in a Federal System, New Horizons in Environmental Economics series, Cheltenham, UK and Northampton, MA: Edward Elgar. Qui, Z. and T. Prato (1999), ‘Accounting for spatial characteristics of watersheds in evaluation water pollution abatement policies’, Journal of Agricultural and Applied Economics, 31(1), 161–75. Rauscher, M. (1997), International Trade, Factor Movements, and the Environment, Oxford: Oxford University Press. Rauscher, M. (2000), ‘Interjurisdictional competition and environmental policy’, in T. Tietenberg and H. Folmer (eds), The International Yearbook of Environmental and Resource Economics 2000/2001, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 197–230. Ribaudo, M.O. (1989), ‘Targeting the conservation reserve program to maximize water quality benefits’, Land Economics, 65(4), 320–33.

Spatial environmental policy

95

Robinson, E.J.Z., J.C. Williams and H.J. Albers (2002), ‘The influence of markets and policy on spatial patterns of non-timber forest product extraction’, Land Economics, 78(2), 260–71. Rosenbaum, P.R. and D.B. Rubin (1983), ‘The central role of the propensity score in observational studies for causal effects’, Biometrika, 70(1), 41–55. Rosenzweig, M.R. and K.I. Wolpin (2000), ‘Natural “natural experiments” in economics’, Journal of Economic Literature, 38(4), 827–74. Roson, R. and K.A. Small (eds) (1998), Environment and Transport in Economic Modelling, Fondazione Eni Enrico Mattei Series on Economics, Energy and Environment, vol. 10, Dordrecht, Boston, MA and London: Kluwer Academic. Sanchirico, J.N. and J.E. Wilen (1999), ‘Bioeconomics of spatial exploitation in a patchy environment’, Journal of Environmental Economics and Management, 37, 129–50. Sanchirico, J.N. and J.E. Wilen (2001a), ‘Dynamics of spatial exploitation: a metapopulation approach’, Natural Resource Modeling, 14(3), 391–418. Sanchirico, J.N. and J.E. Wilen (2001b), ‘A bioeconomic model of marine reserve creation’, Journal of Environmental Economics and Management, 42, 257–76. Sanchirico, J.N. and J.E. Wilen (2002), ‘The impacts of marine reserves on limitedentry fisheries’, Natural Resource Modeling, 15(3), 380–400. Shadbegian, R.J. and W.B. Gray (2003), ‘What determines environmental performance at paper mills? The roles of abatement spending, regulation and efficiency’, Topics in Economic Analysis and Policy, 3(1), Article 15. Shadbegian, R.J., W.B. Gray and C. Morgan (2004), ‘The 1990 Clean Air Act amendments: who got cleaner air – and who paid for it?’, presented at AERE Summer Workshop in June, NBER Summer Institute in July. Shortle J.S. and D.G. Abler (1997), ‘Nonpoint pollution’, in H. Folmer and T. Tietenberg (eds), The International Yearbook of Environmental and Resource Economics 1997/1998, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 114–55. Sigman, H. (2001), ‘The pace of progress at superfund sites: policy goals and interest group influence’, Journal of Law and Economics, 44, 315–44. Sigman, H. (2002), ‘International spillovers and water quality in rivers: do countries free ride?’, American Economic Review, 92, 1152–9. Smith, M.D. (2002), ‘Two econometric approaches for predicting the spatial behavior of renewable resource harvesters’, Land Economics, 78(4), 522–38. Smith, M.D. and J.E. Wilen (2003), ‘Economic impacts of marine reserves: the importance of spatial behavior’, Journal of Environmental Economics and Management, 46, 183–206. Smith, M.D. and J.E. Wilen (2004), ‘Marine reserves with endogenous ports: Empirical bioeconomics of the Californian sea urchin fishery, Marine Resource Economics, 19(1): 85–112. Smith, R.B.W. and J.F. Shogren (2002), ‘Voluntary incentive design for endangered species protection’, Journal of Environmental Economics and Management, 43, 169–87. Smith, V.K., C. Poulos and H. Kim (2001), ‘Treating open space as an urban amenity’, Resource and Energy Economics, 515, 1–23. Swallow, S.K., P. Talukdar and D.N. Wear (1997), ‘Spatial and temporal specialization in forest ecosystem management under sole ownership’, American Journal of Agricultural Economics, 79(2), 311–26. Swallow, S.K. and D.N. Wear (1993), ‘Spatial interactions in multiple-use forestry

96

The international yearbook of environmental and resource economics

and substitution effects for the single stand’, Journal of Environmental Economics and Management, 25, 103–20. Tiebout, C.M. (1956), ‘A pure theory of local expenditures’, Journal of Political Economy, 64, 416–24. Tietenberg, T. (1995), ‘Tradeable permits for pollution control when emission location matters: what have we learned?’, Environmental and Resource Economics, 5, 95–113. Turner, B.L., J. Geoghegan and D.R. Foster (2004), Integrated Land-Change Science and Tropical Deforestation in the Southern Yucatan: Final Frontiers, Oxford: Oxford Geographical and Environmental Studies, Clarendon Press of Oxford University Press. Uimonen, S. (2001), ‘The insufficiency of pigouvian taxes in a spatial general equilibrium model’, Annals of Regional Science, 35(2), 283–98. Umetsu, C. and U. Chakravorty (1998), ‘Water conveyance, return flows, and technology choice’, Agricultural Economics, 19, 181–91. Vance, C. and J. Geoghegan (2002), ‘Temporal and spatial modeling of tropical deforestation: a survival analysis linking satellite and household survey data’, Agricultural Economics, 27(3), 317–32. Vance, C. and J. Geoghegan (2004), ‘Modeling the determinants of semi-subsistent and commercial land-uses in an agricultural frontier of southern Mexico: a switching regression approach’, International Regional Science Review, 27(3), 326–47. Van Kooten, G.C., R.A. Sedjo and E.H. Bulte (1999), ‘Tropical deforestation: issues and policies’, in H. Folmer and T. Tietenberg (eds), The International Yearbook of Environmental and Resource Economics 1999/2000, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 198–249. Viscusi W.K. and J.T. Hamilton (1999), ‘Are risk regulators rational? Evidence from hazardous waste cleanup decisions’, American Economic Review, 89, 1010–27. Weiss, M.D. (1996), ‘Precision farming and spatial economic analysis: research challenges and opportunities’, American Journal of Agricultural Economics, 78, 1275–80. White, M.J. and D. Wittman (1982), ‘Pollution taxes and optimal spatial location’, Economica, 49, 297–311. Wolverton, A. (2002), ‘Race does not matter: an examination of a polluting plant’s location decision’, US Environmental Protection Agency, National Center for Environmental Economics, mimeo. Xepapadeas, A. (1992), ‘Optimal taxes for pollution regulation: dynamic, spatial and stochastic characteristics’, Natural Resource Modeling, 6(2), 139–70. Xing, Y. and C.D. Kolstad (2002), ‘Do lax environmental regulations attract foreign investment?’, Environmental and Resource Economics, 21, 1–22. Yang, W., M. Khanna, R. Farnsworth and H. Onal (2003), ‘Integrating economic, environmental and GIS modeling to target cost effective land retirement in multiple watersheds’, Ecological Economics, 46(2), 249–67.

3. Environmental equity and the siting of hazardous waste facilities in OECD countries: evidence and policies1 James T. Hamilton INTRODUCTION Measuring environmental equity entails as many challenges as defining it. Economics increasingly is being used to explain and evaluate the distribution of environmental quality across socioeconomic groups (see Pearce, 2002). This chapter looks at a particular type of environmental hazard, the siting of hazardous waste facilities, from the perspective of environmental equity. Conclusions about the distribution of risks from hazardous waste facilities depend in part on how these hazards are defined. Studies of facility siting, operation, and cleanup indicate that the greatest hazards appear to be distributed in some countries as if the environment were a normal good. Risks are greater for those with lower incomes. During the 1980s and early 1990s, many of the policies dealing with hazardous waste focused on how to site new facilities and how to clean up older plants. The explicit incorporation of environmental equity concerns came in later policies. This means that while efforts to focus attention on the distribution of risks by income class have recently succeeded in generating new policies in some countries, it is too early to determine the actual efficacy of these policies. This chapter explores the environmental equity debate through the prism of hazardous waste facility siting. Section 1 reviews the nature of the data available, the methodologies of analysis used, and the comparability of studies within and across OECD countries. Section 2 reviews and discusses the studies of hazardous waste facilities and focuses in particular on the distribution of potential risks by demographic group, including different income groups. Section 3 discusses the determinants of disparities in exposure. Section 4 reviews the policy actions taken to address the disparities in the distribution of exposure to environmental impacts from hazardous waste facilities. Though the majority of the studies analysed in each section 97

98

The international yearbook of environmental and resource economics

focus on the United States, the available research published in English from other OECD countries is included in each part of the analysis. Some conclusions are drawn.

1

DATA AND METHODS IN HAZARDOUS WASTE STUDIES

Assessing how risks arising from hazardous waste facilities vary by demographic groups involves defining risks, wastes, facilities, and demographic groups. The definition of a ‘facility’ offers numerous options: plants that generate hazardous waste, facilities that treat, store, or dispose (TSDs) of hazardous waste, or even sites now abandoned that once generated or managed these materials. The operation of the facility can be judged in the siting stage (e.g., who will be exposed to new risks?), during its operation (e.g., which facilities violate environmental regulations), and during remediation (e.g., how has environmental contamination been handled?). Risks can be characterized in a number of ways. Some studies use a simple indicator variable approach, where a facility either does or does not handle specific wastes, is or is not in violation of rules, or contains or does not contain a particular type of environmental contamination. Risks are also proxied by function of facility, so that plants are categorized by whether they generate and manage their own hazardous waste or whether they receive shipments from other facilities and process the waste for a fee. In countries with detailed data on waste management, plants are often grouped by the amount of hazardous waste stored or treated. More sophisticated assessments describe risks by tracking amounts of waste released to the environment, such as air emissions or underground injections. Databases that track reported and/or detected chemical spills at a facility are another source of risk information. The advent of geographic information systems (GIS) technology has allowed risk assessments to be conducted at the facility level for some types of hazardous waste sites. This involves a number of judgments, starting with the radius of externalities generated by a plant. Modeling how far out risks extend requires assumptions about dispersion of air emissions or the likelihood of groundwater contamination and migration. Calculations of risks posed by groundwater contamination entail assumptions about ingestion, chemical toxicity, and the population around a facility. Some analyses focus on determining the lifetime excess cancer risks arising from exposure to a particular chemical for an individual. Though these estimates are generally made with a standard set of assumptions, more sophisticated analyses do use Monte Carlo analysis to generate a range of risks arising

Environmental equity and hazardous waste

99

from exposure to a given chemical. Multiplying individual risk levels by likely populations exposed offers a way to characterize risks by estimating the likely number of cancer cases arising from the presence of a site. Noncancer health effects are harder to quantify since an analyst often lacks slope factors that allow one to translate a given level of exposure into an estimated probability of experiencing a non-cancer effect. These noncancer risks are more often expressed in terms of the degree to which exposure exceeds the level of exposure generally associated with no adverse health effects. Defining the radius of risks around a plant is most often an exercise in modeling. Exposure routes are estimated and health effects are calculated by modeling. An alternative methodology used in some studies of hazardous waste sites draws on epidemiology. The health of residents living around a site is monitored and calculations are made to determine whether higher than expected levels of disease are noted. The multiple sources of risks make it difficult to isolate the separate effect of a plant’s operation on residents, however. If residents close to a plant have lower incomes, for example, higher rates of illness may come from poor diet, inadequate health care, and exposure to toxics from a facility. The turnover in residents may also make it difficult to detect effects through epidemiology, since there may be a long lag time between the exposure and the onset of cancer. Another drawback of using a single radius to assess risks is that this will often ignore the transportation risks that arise in a community if wastes are transported to a facility by truck or rail. Defining risks by plant location also misses another set of stakeholders, those individuals who have an existence or bequest motive to value the environment surrounding a plant. Individuals may care about natural resource damages in an area even if they never visit it. People distant from a plant may also have a willingness to pay value that they attach to living in a just society. These existence values may become important in siting conflicts where residents may be willing to trade off risks for jobs but others outside the community may wish to block the siting because of perceived injustice. Studies of environmental equity also entail decisions about what reference points to use in analysing exposure and what demographic categories to use in comparing risks. Researchers may focus on a single site or small group of facilities and analyse how demographics change as distance from the hazard increases. This type of approach uses the residents within a given radius as the population to study. Other studies take the city as the geographic unit to analyse and explore how neighborhoods within a city that contain hazardous waste sites differ from those without. Some analyses view the country as a whole as the potential site of hazardous plants and compare how geographic units such as counties, cities, census tracts, or even zip-code areas

100

The international yearbook of environmental and resource economics

compare for those with and without facilities. The demographic categories that researchers use to examine variations in exposure include race, income, education, and age. Population density is also considered in comparisons of areas with and without hazardous waste facilities. A snapshot of facility locations across a country can reveal what factors are correlated with risks, but more information is required to tell a causation story about how exposures arise. For example, if one observes a hazardous waste facility located in a neighborhood with low-income residents, several scenarios could explain the association of income and pollution. A plant might locate in a given neighborhood because of factor price considerations. The plant’s negative externalities could lower housing prices, and lowincome residents might move in because of their willingness (due to constrained budgets) to trade off environmental quality for housing costs. Or a facility might target its location in a low-income neighborhood because anticipated political opposition might be lower. To see the relation between pollutants and people, some studies analyse census data for a number of years to see the change over time in neighborhood demographics and facility location. Other works use plans about the expansion of hazardous waste facilities to determine what types of areas are targeted by plants. Researchers also analyse how the current composition of an area affects the enforcement of environmental regulations at facilities and the cleanup of hazardous waste sites. These approaches, which go beyond a simple snapshot of exposure, allow researchers to explore the degree to which disparities in environmental quality are driven by differences in race, income, political power, or education. A final way to analyse the disposition of hazardous waste is to examine the flow of hazardous materials across borders. This analysis takes the country as the unit of observation and explores how trade in waste varies with differences in income levels and environmental policies across states. The literature reviewed in this chapter focuses on the analysis of environmental equity within a given country. Most of the detailed studies on hazardous waste facility siting and operation are conducted using data from the United States. For each section of the chapter, however, I compare results from the United States with those studies published in English that examine the operation of plants in other North American countries, Europe, and Asia. Studies of environmental equity in the US have evolved rapidly with the decline in computing prices, increase in data availability, and growth in the sophistication of spatial analysis software. Research in the US in the early 1990s on hazardous waste facilities focused on the state or county as the unit of geographic analysis. Investigators next explored how risks varied by zip-code area or census data tract. The ease of GIS analysis in the

Environmental equity and hazardous waste

101

mid-1990s made explicit modeling of risks possible at plants and hazardous waste sites. In the next section I will review the environmental equity literature on hazardous waste facilities using the following strategy. I will not attempt to summarize every article on disparities in exposure to hazardous waste, since detailed research summaries are available in Bryant and Mohai (1992), Bullard (1996), Mohai (1996), Foreman (1998) and Bowen (2002). Instead I will discuss in detail representative results from a subset of the most sophisticated analyses. I will divide the results based on the geographic unit of analysis adopted in the report, the degree to which the research focuses on exposure versus causation, and whether the analysis uses data from the United States versus other OECD countries. Overall this review of the literature indicates that in general low-income residents face higher risks from hazardous waste facilities.

2 2.1

LITERATURE REVIEW ON THE DISTRIBUTION OF HAZARDOUS WASTE FACILITIES Exposures within the United States

The US Environmental Protection Agency (EPA) generally defines hazardous waste as waste that is ignitable, corrosive, reactive, or toxic. Facilities that generate hazardous waste or manage hazardous waste through treatment, storage, or disposal (TSDs) register with the EPA under the Resource Conservation and Recovery Act (RCRA) and often supply regulators with quantity data through shipping forms or regular surveys. Early research on environmental equity used information from the EPA’s hazardous waste program to examine how potential risks are distributed across demographic groups. In 1987 the Commission on Racial Justice issued a report that found that when communities with commercial hazardous waste facilities were compared with their surrounding county, the community with the facility had a higher minority percentage, lower household income, more sites contaminated by previous exposure to hazardous waste, lower house values, and higher levels of waste generated per person. The study sparked a debate still continuing in the United States over equity and the environment, a debate which involves concepts such as environmental racism, environmental equity, and environmental justice. In this subsection I review four types of studies generated by this debate: national studies of commercial TSDs, reports that focus primarily on hazardous waste sites in a given city, research that uses company self-reported pollution figures from the US Toxics Release Inventory, and information on environmental cleanups in the US Superfund program. Table 3.1

102

The international yearbook of environmental and resource economics

contains a summary of studies dealing with the distribution of exposure to hazardous waste facilities. Anderton et al. (1994a, 1994b) use the census tract, a county subdivision they indicate averages about 4000 individuals, as the unit of observation in their national study of hazardous waste distribution. The authors first compare the 408 census tracts in 1980 with commercial hazardous waste facilities (i.e., privately owned plants that receive waste from other firms) to the 31 595 other census tracts. They find no statistically significant differences in the mean percentage of black residents in the tracts with TSDs (14.54 black %) versus the other tracts (15.20%). They did find statistically significant differences between the TSD tracts and others for the mean percentage of Hispanic residents (9.41% vs 7.74%), the mean percentage of families below the poverty line (14.50% vs 13.94%), mean value of housing stock ($47 120 vs $58 352), and mean percentage employed in precision occupations (38.60% vs 30.61%). When they restrict the analysis to tracts in the largest 25 metropolitan areas in the US, they find when comparing the 150 census tracts with commercial TSDs with the other 17 406 tracts that those with TSDs had statistically significant differences in means for percentage black (12.23% in tracts with TSDs vs 16.43% without), percentage Hispanic (13.88% vs 10.05%), percentage employed in precision occupations (37.08% vs 28.95%), and mean value of housing stock ($55 980 vs $65 764). There was no statistically significant difference in the mean percentage of families below the poverty line (12.46% vs 13.53%). The authors also compared the 408 tracts with commercials TSDs to the immediately surrounding tracts, defined as 4239 tracts where at least 50 per cent of the area is within a 2.5 mile radius of the TSD tract. They find statistically significant differences in the mean percentage black (14.54% for TSDs vs 25.70% other), percentage of families below the poverty line (14.50% vs 19.48%), and percentage employed in precision occupations (38.60% vs 35.41%). There were no statistically significant differences in mean percentage Hispanic (9.41% vs 10.79%) or mean value of housing stock ($47 120 vs $45 754). Anderton et al. go on to combine the tracts with commercial TSDs with the nearby tracts to form a new set of TSD and nearby tracts (N
4647) to compare to the other tracts (27 356) in the nation. Here they find that the TSD and nearby tracts had statistically different means in terms of percentage black residents (24.72% vs 13.57%), Hispanic residents (10.67% vs 7.27%), percentage of families below the poverty line (19.04% vs 13.08%), percentage employed in precision occupations (35.69% vs 29.87%), and mean value of housing stock ($45 876 vs $60 291). One’s assessment of environmental equity thus depends in part on how far one believes the negative externalities generated by plants extend. If the harms extend to

Environmental equity and hazardous waste

103

nearby tracts, then those minority and low-income residents appear to bear higher risks than residents living in other areas. If the harms extend primarily to the census tract with the TSD, then poor families and Hispanic residents appear to be more exposed. Relative to nearby tracts those that contain the actual TSD have lower percentages of black residents or poor families. A consistent pattern is the association of employment in industrial facilities (denoted by percentage employed in precision occupations) with the presence of commercial TSDs. Anderton et al. note that their work is intended to show patterns in potential exposure to risks from TSDs rather than test theories of causation about why the associations they observe may arise. Atlas (2001a) makes several contributions to the analysis of national TSD locations. He reviews the evidence on the actual risks posed by current hazardous waste TSDs and concludes (p. 952) that: There is no evidence that TSDFs pose, much less have produced, meaningful harm to surrounding populations. The strict regulations under which they operate, the types and quantities of substances that they manage, the minimal potential exposure paths from them to people, and their compliance records all make the risks that they pose pale in comparison to other environmentally regulated facilities, such as those with air emissions.

He puts the waste handled by TSDs in perspective by noting that of 1.1 trillion pounds of hazardous waste generated in the US in 1995, less than 2 per cent (21 billion pounds) were transported offsite for management. The focus on commercial hazardous waste facilities thus can miss the 98 per cent of waste that is managed onsite by generators. The EPA estimated that in 1997 there were between 700 000 and 950 000 generators of hazardous waste and 2025 TSDs. Atlas focuses his analysis on those TSDs that received waste from other facilities and accounted for at least 0.2 per cent of the managed hazardous wastes tracked in surveys by the EPA. This results in a set of 97 TSDs in 1991, 104 in 1993, 101 in 1995, and 108 in 1997. Using GIS technology and 1990 census data, Atlas determined there were 65 736 individuals living within a radius of 0.5 miles of the TSDs he examined in 1991 and 1 690 505 within a two-mile radius. For 1997, there were 71 079 individuals living within 0.5 miles of the TSDs in the sample and 1 494 231 within two miles. Looking at the mean of the percentage of minorities living within a given ring, Atlas found mean minority percentages of 27.0 for the 0.5 ring and 28.4 for the two-mile ring in 1991 and 23.8 for the 0.5 ring and 26.1 for the two-mile ring in 1997. These means are generally higher than the 24.2 minority percentage in the national population in the 1990 census. For low-income populations (defined as those with incomes less than 150 per cent of the poverty level), in 1991 the mean low

104

The international yearbook of environmental and resource economics

income population percentage for the 0.5 ring was 23.8 per cent and 26.1 per cent for the two-mile ring; for 1997 it was 29.6 per cent for the 0.5 ring and 30.4 per cent for the two-mile ring. These mean percentages are higher than the 21.7 per cent figure for the national low-income population. By looking at the mean percentages, one sees some evidence that TSDs operate in neighborhoods with higher percentages of minority and low-income residents. If you weight the means by population, a stronger association between race, income, and exposure appears. Of the total populations within a given ring, for 1991 at the 0.5 mile radius 29.0 per cent were minorities and 44.1 per cent minorities at the two-mile ring. In 1997, the figures were 23.4 per cent minority population at the 0.5 ring and 41.4 per cent at the two-mile ring. In terms of income, 25.3 per cent of the population living within 0.5 miles of TSDs in 1991 were low-income and 27.9 per cent within the two-mile ring. In 1997, 25.9 per cent of the population within 0.5 miles were low income residents and 30.4 per cent for the two-mile ring. Atlas notes that although on an aggregate basis minority and low-income residents have a greater likelihood of living near a TSD, this is because of the presence of a small subset of TSDs in heavily populated areas that have a high percentage of minority or low-income residents. He finds that half of the total minority population living within a given radius of a TSD are concentrated at between 2 per cent and 7 per cent of all TSDs. Risks from a TSD may vary depending on the amount of waste managed at the facility. Atlas generally finds a negative correlation between the minority population percentage in a ring and the amount of hazardous waste managed at the TSD (i.e., the higher the minority population percentage, the lower the amount of hazardous waste managed at the facility). This negative correlation also held for the percentage of low-income population and the amount of waste managed. The expansion plans of hazardous waste facilities provide another way to look at potential exposure by demographic group. Hamilton (1995) uses a 1987 EPA national survey of TSD capacity plans and matches facility decisions with census data on the zip-code neighborhood surrounding a plant. Of 207 zip codes with commercial hazardous waste facilities, net positive expansions in capacity were planned in 84 areas versus no net expansion in 123 zip codes. The difference of means tests comparing the areas targeted for expansion to the other zip codes were statistically significant for a number of demographic variables. The mean percentage of families in poverty was higher in zips targeted for expansion (14% vs 11%), the average of the median household incomes was lower ($15 750 vs $17 060), mean non-white population percentage was higher (25% vs 18%), average zip code population lower (18 700 vs 24 000), and mean voter turnout in the county (a proxy for collective action potential) was lower (51.8 vs 54.8).

Environmental equity and hazardous waste

105

Commercial hazardous waste facilities were thus planning expansions in areas with lower populations, more minorities, poorer populations, and less politically active individuals. Many authors analysing environmental equity have examined the populations surrounding hazardous waste sites slated for cleanup in the EPA’s Superfund program (see Zimmerman, 1993; Hird, 1993, 1994; Anderton et al., 1997). Hamilton and Viscusi (1999) use 1990 census data to determine that overall 50.9 million people live within four miles of the 1173 sites they examine on the Superfund’s National Priorities List. Overall, minorities account for 28.9 per cent of residents living within 4 miles of the sites, 35.1 per cent of onsite populations, and 24.2 per cent of the US population. The percentage of most non-white populations and minority groups as a whole declines as one moves farther from the sites, indicating that these groups bear more of the potential exposures from Superfund sites. Viewed in terms of probability of living within one mile of a site, minorities had a 0.05 probability compared to 0.03 for whites. At the four-mile range, a minority resident had a 0.24 probability of living in this area compared to a 0.20 probability for white residents. The probability of living within four miles was particularly high for Asians (0.31) and Hispanics (0.29), while the probability for blacks (0.21) is close to that for US residents as a whole (0.20). Hamilton and Viscusi show the dangers of focusing on a single measure of environmental equity in assessing the distribution of exposure. For example, they find that the average white population percentage at Superfund sites is 85.6 per cent, which is larger than the national white percentage of 80.3 per cent. This comparison, however, is not weighted by population and misses the fact that sites with higher minority percentages tend to be more populous. Less than one third of the sites (347 out of 1173) account for 89 per cent of the minority residents living within one mile of the NPL sites. Sites with 0 to 10 per cent minority populations in the onemile ring around a site have a mean population of 3966, while sites with 40 to 50 per cent minorities in the one-mile ring had a mean population of 22 396. Thus on a site basis Superfund problems are concentrated in neighborhoods with lower minority population percentages than the national minority population percentage. The location of some sites in highly populous minority neighborhoods, however, means that the overall set of residents surrounding Superfund sites are more likely to be minorities than one would predict based on their national population percentages. Looking at site-level (i.e., unweighted by population) means, Hamilton and Viscusi find that site-level mean household incomes are lower at the onemile ring around Superfund sites ($36 930) and the four-mile ring ($37 690) than the mean household income for the nation as a whole ($38 450). Note that mean household income steadily increases as one moves from one-mile

106

The international yearbook of environmental and resource economics

to four-mile to ten-mile rings. At 61 per cent of the sites, the mean household income is lower in the 0–1-mile ring than in the 1–4-mile ring. The increase in income levels with distance from the site is consistent with the prediction that the environment is a normal economic good. The site-level mean house values for residents living within one mile ($98 590) or within four miles ($103 900) were lower house values than the US mean ($112 660). Such differences would be consistent with the location of NPL sites in industrial working class neighborhoods. The ring trend is generally consistent with the theory that the negative externalities associated with the sites will drive down housing values. At 62 per cent of the sites, the mean housing value is lower for the 0–1-mile ring than for the 1–4-mile ring. For populations living within one mile, the percentage of residents with less than a high school education (25.5%) is higher than the national figure (24.8%) and the percentage of residents with higher education levels (16.5%) is lower than the national figure (20.3%). If one weights the results by population or household, the ring trends generally remain evident. As distance from a site increases, the mean household income for the populations potentially exposed increases, the mean housing values increase, and the percentage of highly educated residents increases. However, on a population weighted basis, residents within four or ten miles of Superfund sites have higher mean household incomes and greater housing values than those for the United States as a whole. Such income differences may arise because of the high concentration of sites in urban areas, where both incomes and housing values are higher. For a subset of 150 Superfund sites, Hamilton and Viscusi conducted risk assessments to determine the potential cancer and non-cancer risk arising over a 30-year period. They estimate that there would be 731 expected cancer cases arising from contamination at these sites. The breakdown by demographic group of the percentage of the 731 estimated cancer cases was minorities 43 per cent, whites 68 per cent, other race 9 per cent, Hispanic 22 per cent, black 4 per cent, Asian 18 per cent, and American Indian 1 per cent. If the site (i.e., the Westinghouse site in Sunnyvale, California) with the largest number (652) of cancers is dropped from the analysis, however, the results are reversed. Minorities would account for 16 per cent of the remaining cancer cases, while whites (including Hispanic whites) would account for 87 per cent of the expected cases. The conclusion that minorities bear a disproportionate share of the expected cancers must be tempered by the fact that this result is driven primarily by one extremely hazardous site. The EPA’s risk assessments at NPL sites focus on individual lifetime excess cancer risks arising from contamination rather than the expected number of cancer cases. In terms of population weighted mean maximum individual cancer risks at sites, minorities face higher risks than

Environmental equity and hazardous waste

107

white populations surrounding the 150 NPL sites in the sample. Minority populations within four miles of the sites face a mean risk of 0.142 versus 0.125 for the white population. The magnitude and distribution of the risk exposure again depends to a great extent on the Westinghouse site. If this extremely hazardous site in California is dropped from the analysis, the gap between mean risks faced by minorities (0.108) and whites (0.102) nearly disappears. These calculated risk levels are high in part because of the EPA’s requirement that analyses use conservative parameter values for variables such as ingestion rate or exposure duration in the calculation of individual risks. In the EPA’s site-level risk assessment, the agency distinguishes between current risks and future risks, which are hypothetical risks that could arise if land use changed or if the likelihood of contamination changed through a mechanism such as the migration of a groundwater plume. Data on both potential exposures and estimated individual cancer levels indicate that minorities may be more likely to be exposed to current risks from Superfund sites. At sites where minorities account for more than 20 per cent of the population within one mile, the mean of the maximum current cumulative risks is 0.013, while the mean for sites where minority population percentages are 20 per cent or lower is 0.0022 (t
1.7). Minorities make up a higher proportion of the population at sites where EPA survey data indicate current residential use. At the 165 sites where the EPA data indicate current residential land use, minorities constitute 45 per cent of the population living within a quarter mile. At the sites (N
343) where there is no current use (e.g., residential, industrial, commercial), minorities constitute 22 per cent of the populations living within a quarter mile. The approach by Hamilton and Viscusi shows how multiple indicators can be used to assess national environmental equity outcomes when significant amounts of data are available (note that the EPA budgets over $1 million to study contamination and remediation options at each Superfund site). In terms of the estimated risks at Superfund sites, minority groups are disproportionately exposed. There is some evidence that minority groups account for a larger fraction of the estimated cancers than their national population percentage, evidence that the population weighted mean maximum cancer risk for minorities is higher than that of whites, and strong evidence that minorities bear larger current risks arising from present land uses at sites. Though this review focuses on the siting of hazardous waste facilities, the fact that much hazardous waste is managed onsite by industrial facilities and the overlap between hazardous and toxic chemicals make research conducted on toxic emissions from plants relevant. Brooks and Sethi (1997) use information from the EPA’s Toxics Release Inventory, which contains annual self-reported figures by plants on their toxic releases and transfers.

108

The international yearbook of environmental and resource economics

They construct an air pollution index at the zip-code level that takes into account TRI emissions in and around the zip code and the toxicity of the chemicals released. Using 1990 census data for US zip codes, Brooks and Sethi find that minorities, renters, individuals with fewer years of schooling, and people with incomes below the poverty line are more highly exposed to toxic air emissions from TRI facilities. Sadd et al. (1999) use GIS technology to study TRI air releases in southern California. They find that census tracts in the metropolitan Los Angeles area that contain a facility releasing air emissions tracked in the TRI had many statistically significant differences from other Los Angeles census tracts. The TRI tracts had higher minority percentages, higher percentages of Latino residents, lower per capita incomes, lower household incomes, a higher percentage of industrial land, a higher percentage of the population employed in manufacturing, and lower housing values. Chakraborty (2001) uses data on the amount and toxicity of hazardous chemicals stored at plants in a given area (Hillsborough County, Florida) to model the dangers arising from acute events such as the accidental releases of toxic chemicals. The study found a positive and statistically significant association between the degree of potential exposure to chemical accidents and the proportion of non-white residents and residents below the poverty line. These studies are typical of the growing environmental equity literature that uses TRI data (Cutter, Holm, and Clark, 1996; Ringquist, 1997; Hockman and Morris, 1998; Arora and Cason, 1999; Daniels and Friedman, 1999) and/or tries to devise more direct indicators of risk exposure from air toxics (Graham et al., 1999; Morello-Frosch, Pastor and Sadd, 2001). While recent studies often use GIS technology to link exposures with populations in one-mile rings around facilities or sites, Millimet and Slottje (2000) demonstrate the usefulness of broad assessments of equity. They develop environmental Gini coefficients to measure inequality across US states in per capita releases of different types of pollution. They find that states with relatively high proportions of women, minorities, and children are (p. 25) ‘over-represented in the upper tail of the per capita pollution distribution’ and point out that environmental policies that do not take this into account may end up increasing measures of environmental inequity. 2.2

Exposures in other OECD Countries

Detailed analysis of exposure to hazardous waste risks within a country requires data on waste facility location, quantities of waste handled, and the demographics of the communities surrounding hazardous sites. The building blocks of this analysis are available in some OECD countries. A 1998 report by the OECD, for example, presents estimates of hazardous

Environmental equity and hazardous waste

109

waste generation, export, and import in the early 1990s. McDougall and Fonteyne (1999) examine waste management data from 11 European cities and find that quantitative comparisons were difficult because of variations in the definitions of waste. Prokop, Schamann, and Edelgaard (2000) survey the management of contaminated sites in 18 Western European countries, determine that 13 had started a systematic process to identify potentially contaminated industrial or waste disposal sites, and (while noting the wide variation in data quality) present estimates of potentially contaminated sites for most of these countries. Page (1997) describes the cleanup programs for contaminated sites in the Netherlands, United Kingdom, and Central and Eastern Europe. Christiansen and MunckKampmann (2000) also note the difficulties of comparing hazardous waste generation data across OECD countries in Europe. A report by the Commission for Environmental Cooperation (1999) notes the problems associated with tracking the transborder shipments of hazardous waste between Canada, the United States, and Mexico. Connor (1992) notes the problems associated with tracking the disposal of hazardous waste by Mexican facilities on the United States border. The Canadian Institute for Environmental Law and Policy (2000) uses information from Ontario’s Hazardous Waste Manifest tracking system to provide a detailed description of the generation and shipment waste in the province. Overall these studies indicate that data on hazardous waste generation or contamination at sites may be available in some OECD countries, but differences in definition of hazardous waste and lack of consistent reporting would make it difficult to compare environmental equity across countries in terms of hazardous waste exposure. As the use of Pollutant Release and Transfer Registers (PRTRs) that record which facilities release particular types of pollution spreads across countries, more detailed environmental justice analyses will become available in the future. Harjula (2003) notes that countries with operating PRTRs include Australia, Canada, Ireland, Korea, Japan, Netherlands, Norway, Mexico, Slovak Republic, United Kingdom, and the United States. For more on PRTRs, see Johnson (2001). There are analyses that focus on the calculation of risks at particular types of sites within a given country. Openshaw (1982) presents estimates of populations exposed and expected thyroid cancers around a set of nuclear plants in the UK. Walker and Pratts (2000) offer estimates of the number of residents exposed to major industrial accident hazards for a set of industrial facilities in Britain. Ragaini (1997) describes how site-level assessments can be conducted at contaminated waste sites in Central and Eastern Europe. Dolk et al. (1998) examine data from registries of congenital anomalies in five countries (Belgium, Denmark, France, Italy, and the UK) to analyse at 21 hazardous waste landfills the impact of proximity

110

The international yearbook of environmental and resource economics

to potential contamination and birth defects. The authors conclude that living within 3 km of a landfill was associated with an increased chance of congenital anomalies (after controlling for socioeconomic status), that the risk decreases for residents more distant from a site, and that more research is needed to determine whether the association is caused by contamination at the sites. The most complete studies on environmental equity from OECD countries other than the United States are from the United Kingdom and Canada (but see also the work by Kruize and Bouwman (2003) that analyses environmental justice outcomes in the Rijnmond region of the Netherlands using GIS technology). A 1999 report for Friends of the Earth used information on postcode location of industrial plants registered between 1992 and 1996 under the Integrated Pollution Control program and household income distribution by postcode. The authors conclude (p. 1): All across England and Wales the poorest families (reporting average household income below 5,000 [GBP]) are twice as likely to have a polluting factory close by than those with average household incomes over 60 000 [GBP]. . . . Over ninety percent of London’s most polluting factories are located in communities of below average income. London is just the most extreme example. A similar pattern is found throughout England and Wales. Overall, almost two-thirds of the most polluting industrial facilities are to be found in areas of below average income.

A 2001 report by Friends of the Earth examines the distribution of the 156 plants in England emitting more than 1000 kilogrammes of carcinogens in 1999. Using the government’s Index of Multiple Deprivation that ranks wards by indicators such as health, income, education, employment, housing, and access to services, the authors found that the polluting facilities were primarily located in the most deprived wards. They note (p. 1) that of the 11 400 tonnes of carcinogenic chemicals emitted in 1999 by the factories in the study, ‘66 per cent of carcinogen emissions are in the most deprived 10 per cent of wards; 82 per cent of carcinogen emissions are in the most deprived 20 per cent of wards; only 8 per cent of carcinogen emissions are in the least deprived 50 per cent of wards’. Three studies focus on the exposure to particular air pollutants by demographic group. Brainard et al. (2002) use modeled emissions from vehicles and measured emissions from monitoring sites to estimate exposures to carbon monoxide (CO) and nitrogen dioxide (NO2) in Birmingham, England. Using GIS technology and 1991 census data at the enumeration district level, the authors conclude that (p. 707) ‘both ethnicity and poverty are associated with pollutant emissions in Birmingham, with the highest emissions being recorded for populations with the highest proportions of minority ethnic groups and impoverished residents’. Note that census

Environmental equity and hazardous waste

111

forms did not provide an explicit question about income, so the authors use questions about occupation of the household head (e.g., works in a professional and managerial position versus a partly or unskilled position). McLeod et al. (2000) used monitoring data on sulphur dioxide, nitrogen dioxide, and fine particulates to estimate exposures at the local authority district level in England and Wales in 1994. They find in regression analysis that pollution decreases as their social class index increases. Once they control for population density, however, they find that (p. 82) ‘the concentrations of all three air pollutants are higher in higher social class areas’. Pye et al. (2001) use data on air pollution, GIS technology, and demographic data at the ward level to study four areas: Greater Belfast in Northern Ireland, Cardiff City in Wales, and Greater London and Birmingham City in England. They conclude that (p. iii): Greater London, Birmingham City District and Greater Belfast appear to show a positive correlation between air pollution and social deprivation, with higher pollutant concentrations of NO2 and PM10 found in areas exhibiting higher levels of deprivation. Cardiff City Council does not appear to show any significant relationship between air pollution and social deprivation.

Canada’s National Pollutant Release Inventory, which contains selfreported facility emissions similar to these collected in the US. Toxics Release Inventory program, has generated research on environmental equity. Jerrett, Eyles, Cole and Reader (1997) aggregate 1993 facility air, water, and land emissions to the county level in Ontario and model the county emissions total as a function of four county characteristics: median income per household, average dwelling value, total population, and manufacturing employment. They found that the coefficient on median income per household is positive and statistically significant (as are the population and manufacturing variables) and that the housing variable is negative and statistically significant. They note that the positive relation between income and pollution may arise if high wages are part of compensation for pollution exposure, and note (p. 1793) that their use of interaction terms suggests that housing value is a ‘more important explanator of the location of pollution emissions than income’. Harrison and Antweiler (2002) examine at the facility level onsite releases (i.e., air, water, land, and underground injection) and offsite transfers. Modeling the level and changes across time in releases and transfers as a function of plant and community characteristics, they (p. 22) ‘generally do not find significant impacts of community income on either the current releases or changes in releases over time’. They measure average community income based on census figures for the enumeration districts within a 50 km2 area around each plant.

112

3

3.1

The international yearbook of environmental and resource economics

LITERATURE REVIEW OF THE DETERMINANTS OF EXPOSURE TO HAZARDOUS WASTE FACILITIES Influences on Siting and Exposure in the United States

Studies that link pollution data with demographic information provide snapshots of who is exposed to potential risks from hazardous waste facilities at a given point in time. Research that focuses on the current location of plants and people, however, cannot determine causation. Because the externalities generated by facilities change the landscape and perceptions of an area, the actual operation of hazardous waste plants may change the desirability of living in an area, affect housing values, and lead to shifts in population characteristics. Residents present at the time of siting may not receive any compensation for risks arising from a facility, for example, while those who move in at a future time may pay lower housing prices because of the drop in environmental quality. The key to isolating what determines exposures to risk is to gather information on the demographics of a community when the decisions of interest are made. In this section I review a number of different approaches used in environmental research in the US to examine what causes exposures to risk to vary across demographic groups. These approaches include analysing what communities are targeted by firms when they plan to expand hazardous waste capacity, how regulators respond to communities as they clean up hazardous waste sites, the impact of neighborhood characteristics on the reduction of carcinogenic air emissions by facilities, the change in area demographics over time as plants locate, the response of housing prices to changes at waste sites, and the reported reactions of individuals when they are queried about siting hazardous facilities. Table 3.2 contains a summary of studies dealing with the determinants of exposure. Hamilton (1993, 1995) examines the expansion plans for 1987–92 submitted by commercial hazardous waste facilities to the EPA. The study of planned changes has the advantage of being prospective, so that the effects of the proposed expansion of the facilities are unlikely to be reflected in changes in neighborhood demographics. The work tests three theories (i.e., pure discrimination, Coase Theorem, and collective action) of why race may be associated with the location of hazardous waste facilities. In the pure discrimination model, owners of waste facilities may trade off profits for prejudice and gain utility from the exposing minority communities to potential risks. According to standard interpretations of the Coase Theorem (1960), a polluting plant such as a hazardous waste facility may locate where it does the least damage, ceteris paribus, because this

Environmental equity and hazardous waste

113

is where compensation is the least. The firm takes into account the physical and demographic characteristics of the surrounding neighborhood that influence the ‘cost’ of its externalities: the number of people affected; incomes; property values; and residents’ willingness to pay for environmental amenities. To the extent that low incomes and education are related to low willingness to pay for the environment and low expected damages in liability cases, and these variables are associated with race, profit maximizing firms may choose to locate in minority areas because compensation demands and expected liabilities from operation are lower there. In the actual process of siting facilities, compensation demands are typically voiced through the political process. Firms will care about the expressed opposition to siting, which depends on a combination of political activity and willingness to pay. If collective action is required to lead a firm to internalize its externalities, then differences in political participation may help explain why minority neighborhoods would be less costly locations for polluting firms. Of the 205 zip codes with commercial hazardous waste facilities operating in 1986, Hamilton finds 83 had net planned expansions in processing capacity. He uses a logit model to predict where firms will decide to expand which includes community demographic variables and market variables relating to processing capacity surplus in the county and state importation and generation of hazardous waste. The results demonstrate that firms care about compensation and political involvement. Consistent with the collective action theory, voter turnout in the county associated with the zip code is negative and statistically significant. The higher the county voter turnout, the less likely that a zip code neighborhood will be targeted for additional capacity. The Coasean compensation variables generally have the expected sign. The number of people in the zip code and percentage of renters are both statistically significant. The more people in the zip code (a factor in compensation demands and liability calculations), the less likely a firm is to expand in the area. The higher the percentage of renters, the more likely firms are to expand capacity, in part because compensation may be lower where residents have fewer sunk costs associated with living in a particular area. The higher the average house price or percentage of adults with a high school education, the less likely the area would be chosen as a site for expansion (though these effects are not statistically significant). The higher the income in the zip code, the greater the probability of expansion. This result, which is statistically significant in one of two expansion specifications, may be due to the fact that areas with expanding waste capacity are areas with expanding industry and higher incomes. The non-white population figure is not statistically significant. Though zip codes with planned expansions do have a higher non-white population percentage, once one

114

The international yearbook of environmental and resource economics

controls for other community characteristics race is not a predictor of where firms target expansions. Decisions by firms to reduce their toxic emissions offer another avenue to examine how differences in risks arise across demographic groups. Hamilton (1999) examines at the facility level the change in air releases between 1988 and 1991 of 16 carcinogens. For a set of 2788 plants tracked in the Toxics Release Inventory, he uses GIS technology to calculate the expected cancer risks arising around a plant and the nature of the community bearing these risks. He finds that controlling for the level of air pollution emitted in 1988, a facility with a higher expected number of deaths due to the release of the carcinogen had greater reductions in emissions between 1988 and 1991. In other words, the most hazardous plants in terms of human carcinogenic risks reduced their emissions more. As voter turnout in the area surrounding a facility increased, emissions declined. This indicates that for a given level of pollution, facilities may be more likely to engage in reductions if they believe that the affected parties are likely to engage in collective action to force firms to internalize the cost of their pollution. The impact of collective action is evident even after one controls for other socioeconomic measures, such as median household income, percentage of college graduates, or percentage vote for the Republican presidential candidate in 1988. None of the community variables other than voting was consistently statistically significant. Median household income and percentage of the zip code population that was black were not statistically significant. While plants do take into account the nature of who bears the risks of their contaminants, it is the likelihood that residents will engage in collective action and thereby force plants to consider the costs of their pollutants that affects plant decisionmaking. A growing literature (see Zimmerman, 1993; Hird, 1994; Gupta, Van Houtven and Cropper, 1996; Hamilton and Viscusi, 1999; Atlas, 2001) examines how the EPA responds to hazardous waste sites depending on the nature of the surrounding community. Viscusi and Hamilton (1999) provide a detailed analysis using risk assessments conducted at a sample of 150 Superfund sites. They analyse in regression analyses the target risk levels regulators choose to allow to remain after cleanups and the cost per cancer case avoided implied by the remediation decisions. Looking first at the risk pathway targets chosen, they divide up the risks into two sets – those where the original risk posed by a given pathway of exposure to a chemical was greater than or equal to 104 (i.e., high-risk pathways) and those where the unremediated risk was less than 104 (i.e., low-risk pathways). The higher the voter turnout in an area, ceteris paribus, the more stringent the target risk chosen to be allowed to remain after remediation when the original risks are low. When risks are high, political activity has

Environmental equity and hazardous waste

115

no effect on cleanup standards. It is only when risks are low that political activity matters. The higher the average income level in the one-mile ring around a site, the less stringent risk target will be chosen. This result may be because regulators believe wealthier residents are less likely to be exposed as assumed in risk assessments (e.g., groundwater exposures assume wellwater consumption, while wealthier residents may be connected to public systems). A higher minority percentage in the area leads to the selection of a more stringent risk target. This could be evidence that regulators were concerned with environmental equity or might believe that calculated risks were more likely to arise in minority communities (e.g., if minorities were more likely to consume contaminated groundwater). In terms of the cost per cancer case avoided at Superfund sites, Viscusi and Hamilton find that variables such as the minority population percentage within a one-mile ring of a site or the mean income of residents within one mile had no impact on the cleanup expenditures chosen by the EPA. The higher the voter turnout in the county, however, the greater the cost per cancer averted implied by the EPA’s cleanup decision. Cleanups at hazardous waste sites appear in part to follow a ‘fire alarm’ process (McCubbins and Schwarz, 1984), where regulators respond to the likelihood that residents will complain about the nature of site remediation. Another explanation for demographic variations in exposure to hazardous waste facilities is that facilities may generate negative externalities that lower surrounding housing prices. Lower housing prices in turn may attract lower-income residents, whose budget constraints limit their ability to pay for a clean and safe environment. Economists have used a variety of hedonic methods to see how housing prices react to the presence of ‘locally undesirable land uses’ (see Nelson, Genereux and Genereux, 1992; Kiel and McClain, 1995; Hite, 2000). Farber (1998) summarizes the results of 25 studies on the impact of undesirable facilities on housing prices. He finds for ten studies of the impact of hazardous waste or Superfund sites that housing prices do increase with distance from these sites. For three studies that estimated the housing price impacts after a site had been placed on the Superfund’s NPL list, Farber finds a consistent effect that housing prices increased approximately $3500 (1993$) per mile from the site. Gayer, Hamilton and Viscusi (2000) find that residents’ willingness to pay to avoid risks actually declines after the release of remedial investigation studies at Superfund sites, suggesting that the information in the EPA studies lowers the perceived risks at sites. Gayer (2000) uses estimates of cancer risks at a set of Superfund sites and analyses what housing prices imply about the marginal valuation of risk reduction in different neighborhoods. He finds that the price–risk tradeoff is greater for households in areas with residents with higher education and residents with higher incomes. The price–risk

116

The international yearbook of environmental and resource economics

tradeoff implied in housing sale data is lower in neighborhoods with a higher proportion of non-white residents. Research about the current patterns of exposure to hazardous waste facilities have generated significant debate and controversy (see Been, 1995; Mohai, 1995; Yandle and Burton, 1996; Boer et al., 1997; Liu, 1997; Boyce et al., 1999). One of the most direct ways to analyse exposure causation is to examine the nature of communities at the time of facility siting. Been (1994) examined four hazardous waste landfills studied in a 1983 General Accounting Office environmental report and concluded ‘at the time the facilities were sited . . . the host communities were home to a considerably larger percentage of African Americans and were somewhat poorer than other communities within the host states. The analysis therefore suggests that the siting process had a disproportionate effect on the poor and people of color.’ In examining ten landfills and incinerators first studied by Robert Bullard (1983, 1990), Been found that when they were originally sited five of the ten facilities were in areas with higher percentages of minority residents than the surrounding county and that three out of ten were in areas with higher poverty rates. By 1990 the neighborhoods had changed so that nine out of ten had greater than average proportions of minorities and seven of ten had relatively higher poverty rates. Been and Gupta (1997) conducted a national study of 544 communities that in 1994 contained active commercial hazardous waste TSDs. Using census data at the tract level for 1970, 1980, and 1990, they examined the nature of neighborhoods at the time sitings occurred and the changes in demographics for these areas across time. They found (p. 9): no substantial evidence that the facilities that began operating between 1970 and 1990 were sited in areas that were disproportionately African American. Nor did we find any evidence that these facilities were sited in areas with high concentrations of the poor; indeed, the evidence indicates that poverty is negatively correlated with sitings. We did find evidence that the facilities were sited in areas that were disproportionately Hispanic at the time of siting. The analysis produced little evidence that the siting of a facility was followed by substantial changes in a neighborhood’s socioeconomic status or racial and ethnic composition.

Environmental equity studies focused on causation have also begun to look at historical relationships between facilities and neighborhoods within a given city. Baden and Coursey (2002) examine the distribution of hazardous facilities and waste sites across time in Chicago. By matching facility siting dates with community demographic data (often at the census tract level), they determine that (p. 87): past waste-generating activities tended to be in less populous, lower income areas with good access to highways and waterways. Present waste sites tend to

Environmental equity and hazardous waste

117

be located in less populous, wealthier neighborhoods, with convenient access to transportation infrastructure. There is no good evidence that African Americans of any income class are more likely to live in areas with more concentrated waste sites in the city of Chicago, or that they have been targeted to be disproportionately exposed to more hazardous waste. Several of the 1990 regressions found that the percentage Hispanic in a community was significant in describing the presence of a site in, or near, a community.

Lambert and Boerner (1997) examine changes over time in the city of St Louis for census tracts with hazardous facilities or waste sites. They find that between 1970 and 1990 the percentage of residents below the poverty line and percentage of minority residents increased disproportionately in neighborhoods with TSDs, hazardous waste sites, and non-hazardous landfills and incinerators. They determine that mean real family incomes fell in these tracts (while the mean for St Louis as a whole was rising) and that median real housing values increased at a less rapid pace around the hazardous sites. Pastor, Sadd and Hipp (2001) examine the historical siting patterns of high-capacity TSDs in Los Angeles by matching siting dates with census tract data from 1970 through 1990. They find more evidence to support disproportionate exposure of minorities at the time of siting than a later move-in of minorities once a TSD is located in an area. A final way to analyse why hazardous waste facilities are distributed in current patterns is to examine the results of surveys done about siting, risk perception, and compensation (see Bord and O’Connor, 1992; Groothius and Miller, 1994; Rogers, 1997; Halstead, Whitcomb and Hamilton, 1999). A significant literature exists on how individuals in surveys differ in their expressed support for the siting of hypothetical or actual NIMBY (NotIn-My-Backyard) facilities. Summarizing the role of potential compensation in facility siting, Oberholzer-Gee and Kunreuther (1999) note that compensation appears to increase support for siting of facilities such as prisons or airports but has little impact on prospective acceptance of projects perceived as having high risks such as incinerators or nuclear waste repositories; they note that for projects perceived as very risky compensation can in some cases reduce support for siting if it is perceived as a bribe. Smith and Kunreuther (1999) estimated in a study of compensation and mitigation for four hypothetical facilities that there is a core of respondents who view a siting as unacceptable even under a wide range of compensation/mitigation. They estimated this core of opponents to be 11.7 per cent for a prison siting, 13.4 per cent for a municipal landfill, 26.9 per cent for a hazardous waste incinerator, and 40.2 per cent for a disposal repository for high-level radioactive waste. Mitchell and Carson (1986) report that to reach a cumulative percentage of 50 per cent of respondents accepting a facility a large factory or coal plant would have to be five miles

118

The international yearbook of environmental and resource economics

from residents but a hazardous waste facility would not reach this level of acceptance until it was nearly 50 miles from residents. Mansfield, Van Houtven and Huber (2001) find that those who are more likely to oppose nuisance facilities are also more likely to vote or participate in other forms of collective action, which suggests that political siting processes may engender more participation from opponents. Focusing on the role of trust in siting, Groothuis and Miller (1997) find that younger respondents and those with lower incomes express more distrust of waste disposal firms and the government, that people who distrust the media, government, and business express a higher estimate of the risks of hazardous waste disposal facilities, and that distrust affects the willingness to accept a facility siting. Swallow et al. (1994) show the importance of using contingent-valuation surveys to estimate different willingness to pay measures in siting disputes for different demographic groups (e.g., based on age or income). Much of the research on siting attitudes comes from surveys relating to the siting of a high-level nuclear waste repository in Nevada (see Kunreuther and Easterling, 1990 and 1996; Easterling, 1992; Dunlap, Kraft, and Rosa, 1993). This research shows residents more willing to support a facility the lower the perceived risks to future generations, the better the mitigation efforts made to limit risks from a facility, and the more residents view the siting as the best policy outcome. 3.2

Influences on Siting and Exposure in other OECD Countries

Most of the literature that explains siting patterns for hazardous waste facilities in OECD countries other than the US focuses on individual-level survey data or case studies of particular siting mechanisms. A series of papers focus on survey interviews done in Switzerland in 1993 a week before a referendum on nuclear waste repositories (see Frey and Oberholzer-Gee, 1996, 1997; Frey, Oberholzer-Gee and Eichenberger, 1996). The researchers find the expected results that willingness to have a nuclear waste repository located in a resident’s community declined as perceived risks or negative economic impacts were larger. The authors determine, however, that compensation offers reduce the willingness to accept a nuclear waste site. They report (1997, p. 749), ‘While 50.8 per cent of the respondents agreed to accept the nuclear waste repository without compensation, the level of acceptance dropped to 24.6 per cent when compensation was offered.’ They determine that compensation reduces acceptance not because it signals more risk but because it crowds out a feeling of civic duty. Once monetary rewards are introduced, respondents are less likely to view siting acceptance as related to civic duty and less likely to accept the facility. Respondents viewed compensation as bribes to be rejected (though note that rejecting

Environmental equity and hazardous waste

119

compensation in votes or surveys is relatively costless, compared to rejected compensation in actual market settings). If a proposed siting process was seen as fair by respondents, they were more likely to accept the facility. Research by Renn, Webler and Kastenholz (1996) on siting of a Swiss landfill showed via focus groups that notions of fair siting involved considerations of the amount of waste generated in an area, whether an area was already exposed to a hazardous facility, and the need to prevent unacceptable levels of risk. A different assessment emerges in research on the role of compensation in siting in Japan. Lesbirel (1998) examines the siting of energy plants in Japan and finds that compensation packages facilitated the siting of these facilities. Lesbirel notes that the Ministry of International Trade and Industry (MITI) has set up structures to facilitate bargaining between utilities and community interests, that adjacent areas to locations are involved in negotiations, that risk mitigation strategies are pursued, and that powerful fishing cooperatives are able to strike bargains with prospective plants. Broadbent (1998) examines the general impact of protest on environmental politics in Japan and finds that national party politics, bureaucratic constraints, the influence of big business interests, and (at times) local protest influenced the location of polluting plants and landfills. Kleinhesselink and Rosa (1994) use survey evidence from college students in the United States and Japan to demonstrate the similarities in risk perception across the two countries. In both countries perceptions of the risks from particular sources are governed by characteristics such as the degree to which the risk is seen as involuntary or infrequent. They note that the Japanese respondents (p. 116) ‘rated the technological risks associated with nuclear power as older and risks of which they had individual or scientific knowledge, whereas US students rated these as newer and risks for which they had significantly less individual or scientific knowledge’. Studies of waste siting in Canada stress the role of participation. Fischer (1995) discusses the successful siting of a hazardous waste treatment plant in Alberta and ascribes the final approval to the high level of public participation designed into the process, including an early local plebiscite on accepting the siting, the regional government’s provision of funds for the local community to hire experts to help analyse and discuss the plant’s impacts, the government’s provision of funds once the plant was sited to compensate for infrastructure costs and retain more experts, and the formation of a local committee to advise the facility on community concerns about its operation and review monitoring data. Huitema (1998) reviews studies of hazardous waste disposal facilities in Canada and the US and concludes that voluntary approaches mixed with compensation may be the most effective, though he notes that few new facilities have actually been

120

The international yearbook of environmental and resource economics

sited overall, that capacity is underutilized at the sites, that communities outside the host community may be left out of the process, and that it is difficult to generate high levels of sustained public participation. Baxter, Eyles and Elliott (1999) analyse landfill siting in Canada and note that environmental suitability and community control played much larger roles in the siting procedures than equity considerations, that considerations of fair procedure were often crowded out, and that the importance attached to public participation meant that some (p. 520) ‘residents were successful in their opposition to the siting process in part because they had the financial and educational means to mount sophisticated opposition’. Explanations for siting difficulties are also the topic of research in many European countries. Linnerooth-Bayer and Fitzgerald (1996) found in a survey of 111 residents of Lower Austria that there are wide variations in what constitutes fairness in siting, with support for hierarchical, market, and lottery approaches among some segments of the population. They note (p. 6): When the respondents were asked if they would disregard all features of the host communities (whether they are already burdened by other industrial hazards, whether they are poor and vulnerable, whether they have benefitted from industrial production, etc.) if experts reported that the proposed site was technically superior or posed the lowest overall risk to the public, 53% of the respondents answered positively (and 70% of the industrial experts). This shows remarkable deference to expert authority and an acceptance of Austrian hierarchical political procedures . . . .

Schneider and Renn (1999) describe how a structured discussion process involving ten groups of citizens was used to elicit rankings of sites for waste facilities in the Northern Black Forest Region of Germany, though the political process ultimately did not draw upon the reports written. Gaussier (2001) analyses the actual location of garbage dumps in the ProvenceAlpes-Cote d’Azur region of France and demonstrates the influence of transportation costs and NIMBY forces on the spatial distribution of dumps. Sjoberg, Viklund and Truedsson (1999) describe how debates in Sweden at the municipal level over the desirability of initiating a feasibility study in a given municipality for siting a national high-level nuclear waste repository often focus on the benefits and costs of the repository rather than of the study. In terms of jobs brought by a future repository, they note (p. 5): . . . unemployment, while certainly unpleasant, is not economically disastrous in Sweden. The social welfare benefits are generous. . . . The jobs promised at a repository are in a rather distant future and may not be, in themselves, very attractive to the young people of today who would be the ones to get them.

Environmental equity and hazardous waste

121

Vari (1996) conducted 24 interviews with individuals involved in a disputed siting in Hungary of a low-level radioactive waste facility and found that perceptions of fairness depended in part on questions about whether those who generate waste bear its risks and a desire to avoid targeting of those who are economically or socially disadvantaged. Snary (2002) finds that the public participation process involved in the potential siting of a waste-to-energy incinerator in Portsmouth, Hampshire, UK was flawed in part because participants were directed to focus on technical questions rather than broader waste management issues. Coenen (1998) describes how the process of participation in ‘green planning’ at the national and province level in the Netherlands can lead to improved decisionmaking about pollution because of communication and learning sparked by the planning process.

4

SITING POLICIES AND ENVIRONMENTAL EQUITY

4.1 General Siting Policies for Hazardous Waste Facilities in the United States A. Siting in theory The difficulties associated with siting commercial hazardous waste TSDs in the US have in part generated large literatures on theories of siting noxious facilities (Sullivan, 1990, 1992; Gregory et al., 1991; Swallow, Opaluch and Weaver, 1992; O’Sullivan, 1993; Ingberman, 1995; Fredriksson, 1998; Quah and Tan, 1998; Lejano and Davos, 2002; Minehart and Neeman, 2002; Waehrer, 2002) and on lessons learned from attempts to locate NIMBY projects (O’Hare and Sanderson, 1992; Gerrard, 1994, 1997; Wheeler, 1994; Miranda, Miller and Jacobs, 2000; Richman, 2001). Understanding the potential role for equity concerns to play in siting policies entails an understanding of how siting practices and the siting literature evolved in the US and the limitations evident in siting procedures. The ‘Coase theorem’ offers a framework for understanding both the details of state siting laws and the design of academic siting models. Coase focused on the relationship between property rights and externalities in his seminal 1960 article ‘The Problem of Social Cost’ (often referred to as Coase II, to distinguish it from an earlier influential 1937 piece, ‘The Nature of the Firm’ (hereafter referred to as Coase I)). He noted that in a world of zero transaction costs, property rights would be fully defined, contracts could be costlessly negotiated and enforced, and trades would be easily consummated so that resources would flow to their highest-valued

122

The international yearbook of environmental and resource economics

use. In this model, the definition of property rights leads to a market for pollution. If individuals in a community enjoyed the right to be free from pollution, then a firm such as a hazardous waste facility that generated externalities would consider the impact of its location on communities. The firm would end up locating, ceteris paribus, where its environmental damage would be the least because that is where the compensation for pollution it would have to pay affected communities would be the lowest. The market for pollution rights, where either individuals enjoy the right to be free from pollution and must be compensated by firms or firms possess the right to pollute and must be paid by affected neighbors to restrict their pollution, creates a market for the location of polluting facilities. Hanemann’s work (1991) on the possible divergence between willingness to pay and willingness to accept values when there are large endowment effects and few substitutes (for a good such as safety) may mean, however, that even with well-defined property rights gridlocks may develop in siting negotiations. State laws in the US adopted during the 1980s that dealt with siting hazardous waste facilities were often viewed as establishing markets for locating these facilities. Though the statutes varied in format, the basic provisions involved defining the specific property rights of facility operators, affected communities, and state and local governments. The statutes were interpreted as facilitating negotiations in a world where transaction costs did matter. Explicit compensation mechanisms were established, so that a locality targeted for a facility would be able to extract some compensation for hosting a facility. Technical assistance grants were provided to communities so that they could conduct their own studies to determine risks posed by a facility. Public hearings were designed to educate potential neighbors, who would then negotiate based on information about particular operations at the proposed facility. Some states even considered the spillover effects on communities close to the community with the facility, so that some compensation would be paid to areas which did not enjoy the larger compensation package which came with hosting a facility but did bear some additional risk because of the transportation of waste through their area. Academic models of the siting process generally recommended making the market elements of siting mechanisms even more explicit. Many proposals for breaking the NIMBY gridlock in siting unwanted facilities involved auctions in which communities submitted bids to receive compensation for hosting hazardous waste facilities. Though the formats of the proposed auctions differed, economists and political scientists who made these recommendations generally focused on the siting problem as an exercise in demand revelation, i.e., in determining the demand for environmental amenities across communities. The community was generally the unit of observation, though sometimes individual preferences were also

Environmental equity and hazardous waste

123

aggregated within each community through procedures such as referenda on siting compensation. The models stressed that if a community’s rights to compensation were well-defined and firms had to pay compensation through a system such as an auction for the right to locate in an particular community, then a firm could end up choosing to locate where citizens place a lower value on the environmental risks posed by the facility’s operation. In the models, clearer property rights and an auction system would lead to the Coasean solution. Despite the attention devoted by legislators and scholars to resolving locational conflicts, few commercial hazardous waste facilities were sited under the newly revised laws. Models and statutes that view the location of a hazardous waste facility as an exercise in auctioning a facility among communities focus on one strand of the law and economics literature (Coase II) while ignoring other insights from this literature. Specifically, lessons from at least four separate literatures are useful in understanding the current regime of US siting laws: Coase’s early work on the firm versus the market (Coase I); the theories of collective action; evidence on the psychology and political economy of risk perception; and the positive political theory of institutions. These disparate literatures in law, economics, and political science offer explanations of why previous understandings of siting statutes and previous versions of siting models are inadequate and, in part, unlikely to succeed as long as they are based primarily on establishing a ‘market’ for locally noxious facilities. One key to understanding the design of the current siting literature is to see how the evolution of siting statutes parallels the discussion of institutional design in Coase I. In ‘The Nature of the Firm’, Coase determined that whether a decision would be made within a firm or through the marketplace depended on the relative transaction costs of using the particular mode of operation. If the centralized decisionmaking power that characterizes a firm can reach a decision more easily because of the transaction costs of market exchange, then the production step will take place within the firm. The evolution of decisionmaking power about siting hazardous waste facilities shows a shift from market to firm, a shift associated with a rise in transaction costs. In the era of less intense scrutiny of hazardous wastes, firms that treated hazardous substances enjoyed the right to locate freely within the constraints imposed by local zoning ordinances. As local opposition to such facilities increased with heightened perceptions of hazards, however, the operation of the ‘market’ for location broke down as protests, zoning battles, and litigation slowed the siting of hazardous waste facilities. These siting battles led state legislatures in the US to clarify and redefine property rights involving facility location. Some of the states adopted provisions which did attempt to establish a market in siting

124

The international yearbook of environmental and resource economics

through better definition of the rights of the parties involved, a solution in the spirit of Coase II. But some legislatures opted instead to create a decisionmaking process similar to that of a firm to site LULUs (locally undesirable land uses) such as hazardous waste facilities. Ultimate authority to site a facility was given to the state, or the state was given the right to override local attempts to block a facility. At times this use of centralized decisionmaking was supplemented by a process that would elicit preferences within communities about proposed siting, much like firms may attempt to use market-like mechanisms internally to establish the appropriate transfer prices in production decisions. Whether the state actually possesses sufficient centralized decisionmaking power to overrule a locality and place a facility where it is not wanted is an empirical question. The important point in understanding the legislation that emerged to deal with the siting gridlock, however, is that it was not necessarily trying to establish a market in well-defined pollution rights across communities (Coase II). State laws are better understood as designed to make a decision (where to locate a facility) given a set of transaction costs that may vary by state, with some states choosing a system that resembles a market and some states choosing a mechanism that resembles a firm in its reliance on centralized authority to select a site. Theories of collective action offer another set of qualifications to current siting statutes and models, which generally treat ‘the community’ as the relevant unit of observation in siting disputes. In legislation, compensation schemes are often based on negotiations between the elected officials of a locality and developers. In academic models, auctions are conducted where a ‘community’ names its compensation fee for accepting a facility, though the details of arriving at such a figure are often ignored or are seen as a preference revelation problem. Yet Mancur Olson (1965) pointed out long ago the dangers of equating group interests with group action. Current models assume that compensation demands expressed by communities vary only because of the individuals’ differences in valuing environmental risks and willingness to pay for the environment. Often, however, individuals will vary in the degree that they are able to overcome free rider problems and engage in collective action. If communities vary in the degree that they engage in the collective action necessary to force a locating firm to pay compensation, then facilities which generate externalities will choose to locate where the expressed compensation demands are the least (ceteris paribus). Further complications arise when one considers the principal–agent relationships in models where elected representatives announce the compensation figures. Depending on the strength of monitoring by the electorate and the state of local political competition, the compensation figures announced by elected representatives may or may not relate to the preferences of affected

Environmental equity and hazardous waste

125

constituents. Compensation may flow directly to representatives, to representatives’ favored constituents, or to those put at risk by the facility. The ‘community’ affected by a facility is often viewed as those residents whose activities are physically affected by the potential operation of the facility, for it is their ‘use value’ of the environment that is threatened. Individuals also have existence values and bequest motives over the environment, however, which means that affected parties who may be active in siting disputes also include those people who place a value on knowing that the environment is undisturbed by the risks of such a facility and those who wish to transfer an environment to future generations that does not entail the risks posed by such facilities. (For analysis of how bequest values for the siting of nuclear waste storage can be estimated, see Riddel and Shaw (2003).) Thus the theories of collective action indicate the importance of starting the modeling process and statutory design with individuals rather than ‘the community’ as the basic unit of observation. The literature on risk perception provides a third set of qualifications to the market models of locating hazardous waste facilities, which generally are based on an expected utility framework where individuals’ preferences over facility siting are driven by expected values for outcomes calculated from the available information about a facility’s risk. Many of the siting statutes provide for public education programs to inform citizens about the risks associated with particular technologies. State officials often view part of the NIMBY gridlock as arising from a tendency of environmentalists’ and citizens’ perceived risks to outrun the actual risks posed by facility operations. This ignores, however, the fact that most citizens will remain rationally ignorant of hazardous waste policy and that risk regulation policies are more likely to be driven by perceived risks than actual risks. Noll and Krier (1990) have assembled the evidence from cognitive psychology about behavior that indicates risk perceptions may diverge from the expected utility model: individuals reason by relating situations to previous experiences (representativeness heuristic); valuations are determined by how a choice is presented (framing effect); estimates of probabilities are influenced by whether a similar event readily comes to mind (availability heuristic); people act as if they believe small-probability events are more likely than their own beliefs would suggest; and people have preferences over how probabilities arise. The implications for siting a hazardous waste facility are that public reactions to a facility will not simply be based on the risk analyses presented for an individual facility’s technology. Acceptance of the facility will in part be path-dependent, in that previous experiences with similar technologies will drive assessments of newly proposed facilities. Assessments of the dangers posed may depend on general world views of participants (e.g., their interpretations of past experiences with the

126

The international yearbook of environmental and resource economics

ability of market incentives and regulatory scrutiny to affect corporate health and safety decisions). Participants in siting battles will also react to how probabilities are generated, e.g., what process ultimately results in the expected outcomes at particular facilities. The positive political theory of institutions also provides insights for understanding the operation of siting statutes and the design of siting models. Discussion of the decisionmaking procedures used to select a site often portray the design of the procedure itself as a question of how to incorporate scientific criteria into the process so that the ‘best’ site is chosen. Positive political theory, however, implies that just as legislators have preferences over the outcomes of siting battles they will have explicit preferences over the types of institutional designs likely to lead to particular outcomes. Legislators who wish for a facility to be sited may attempt to delegate the decision to a separate board and raise the transaction costs of overturning the board’s decision, an exercise in blame shifting (to the delegated decisionmakers) and in commitment strategy (the costs of overturning a decision may make future legislative intervention less likely). Though it is often assumed that legislators make decisions about siting procedures behind a veil of ignorance that hides whose districts may be likely candidates, some legislators will have general preferences over whether any facility is sited and thus preferences over particular features of the siting procedures. Those concerned with discouraging siting may include substantial public participation requirements that provide activists with the ability to use litigation to halt siting activity, while those concerned with facilitating location may provide the state siting authority with the ability to override local objections to a facility. Separate from the outcomes likely with a particular decisionmaking process, individuals may also have preferences over such elements as whether a process is fair, open, and incorporates public participation. Preferences over procedure allow opponents of facilities to combine multiple issues into opposition to a facility, so that the debates are framed both in terms of risks posed by technology and the fairness of procedures used to narrow down possible sites. Though discussions of siting laws often proceed as if the goal were to site a facility, the positive theory of institutions offers warnings on how preferences over outcomes can map backwards into preferences over the design of institutions and how people may value the details of institutions because they relate to broader notions of democracy, participation, and equity. B. Siting and compensation Academic articles and models written during the 1980s and 1990s concentrated on breaking siting impasses through compensation mechanisms aimed at reducing local opposition. These models typically used the

Environmental equity and hazardous waste

127

community as the unit of observation and proposed siting mechanisms that involved communities submitting compensation bids in an auction for siting a facility. The community with the lowest compensation figure announced or submitted would be the host to the facility. The details of the auctions varied, including sealed-bid auctions in which communities submitted a compensation figure, a reverse Dutch auction in which the first community to accept a declared compensation package would ‘win’ the facility, and public referenda in which citizens would vote on particular compensation packages. A common thread throughout these models is the focus on siting battles as a demand revelation problem, e.g., how can the facility operator or state government determine which community places the lowest value on the potential environmental risks and would demand the least compensation for hosting a hazardous waste facility. These siting models differed in whether they emphasized strategic factors in securing approval for a facility, equity considerations in the distributional impact, or efficiency in matching a plant with an area which demanded the lowest compensation for its externalities. Works by O’Hare and others (1977, 1983) emphasized how compensation mechanisms could contribute to the probability of successfully siting a facility with concentrated costs and diffuse benefits. These models focused not on questions of efficiency but on how to overcome local opposition. Methods for securing local support for a noxious facility included a community referendum on a single proposed compensation offer from a developer, a vote on different bids, or the use of the existing political structure such as the town government to negotiate on behalf of the affected community. These models also stress the equity considerations of compensating residents whose health and safety are threatened by the operation of a facility or who experience economic losses such as property value declines because of externalities generated by the locating firm. Compensation thus becomes a method of sharing the benefits generated by the facility with those experiencing the costs. Sullivan (1990) has explored the interaction between equity and efficiency considerations in the design of compensation mechanisms. He notes that if victims are in part compensated for their exposure to pollution then they may be less likely to take averting actions, and explores the conditions under which these schemes may make residents less likely to locate away from polluting facilities. A third type of compensation model emphasizes the use of an auction process as an instrument to reveal where the facility’s externalities will result in lower disamenities. Models by Kunreuther and co-authors (1986, 1987) approach the NIMBY process as a demand revelation process in which the challenge is to structure an auction that will cause communities to reveal truthfully the compensation they would demand for receiving a facility.

128

The international yearbook of environmental and resource economics

These models are in the tradition of public finance mechanisms that attempt to elicit true preferences for public goods. An auction process is designed in which a community submits the payment it would demand to be a host site, the community with the lowest bid wins, and other communities pay it compensation based on their submitted bids. The community is the unit of observation, although the authors point out that if one views individuals’ preferences in each community as quasilinear in site value and income then one can view the community’s bids as aggregations of residents’ willingness to accept compensation figures. Though the auction process is typically described as a sealed-bidding process, Inhaber (1992) proposes a ‘reverse Dutch’ auction in which the government would announce a figure that would increase over time that would be paid to compensate a host community. The first community to accept the proffered compensation package would end up with the facility. Discussion of these siting statutes and models often centers on the creation of a ‘market’ for the location of hazardous waste facilities. The notion is that if property rights are well-defined then the operation of the market will lead a firm to locate where its risks involve the lowest compensation to affected neighbors. Some siting articles explicitly cite Coase II as evidence that if property rights can be specified better than the siting gridlock can be eased. Mitchell and Carson (1986) state that the often ambiguous nature of property rights in siting disputes results in protracted disputes; they recommend that a community be given the explicit right to refuse a proposed siting through a referendum process in which residents would vote on a proposed compensation package. They point out that if each individual in the community had the right to block sitings then the transaction costs of negotiating with the developer of a facility could be prohibitive. Bacow and Milkey (1982) point to the compensation negotiations provided for in the Massachusetts siting statute as evidence that Coasean transactions can be facilitated through the explicit arrangement of property rights. They say that the Massachusetts process where developers negotiate compensation with potentially affected communities is ‘more efficient because developers must consider the full social cost when choosing where to site a facility’. The evolution and details of the state siting laws, however, are also consistent with Coase’s earlier article, ‘The Nature of the Firm’. In that work Coase points out that whether a particular production decision is made within the firm or across a market will depend on the transaction costs of using one method of organization versus another. When costs of negotiation and contract enforcement in a market are high relative to the costs of internal production, then a good will be produced within a firm. Numerous factors associated with negotiating compensation contracts between a

Environmental equity and hazardous waste

129

community and a developer make such agreements costly to arrive at: the problem of how elected officials can bind their successors to honor particular agreements; the possibility that disgruntled residents within the community or environmental groups outside the community may attempt to use legal means to delay construction even after community officials have reached an agreement; the difficulty of specifying exact payments dependent on environmental outcomes, especially since a firm may be bankrupted by adverse outcomes; and the danger that explicit compensation contracts will be interpreted as bribes by residents. The key advantage of using a firmlike structure to make the production decision is that the use of centralized decisionmaking authority reduces the problems of breach and holdup that are associated with incomplete contracts in a market. C. Siting in practice Facilities that process hazardous waste attracted growing attention during the 1980s from environmentalists, state legislators, and academics. Incidents of chemical contamination such as the highly publicized unraveling of a dump site in Love Canal, New York and the abandonment of an entire town because of dioxin exposure in Times Beach, Missouri raised public awareness of dangers posed by hazardous waste and increased scrutiny of facilities dealing with such waste. Legislation such as the Resource Conservation and Recovery Act of 1976, which established cradle-to-grave monitoring of hazardous wastes, and the Comprehensive Environmental Response, Compensation, and Liability Act of 1980, which established the Superfund ex post liability system for cleaning up waste sites, marked the beginning of an era of more stringent regulation. Attempts to site new facilities to treat, store, or dispose of hazardous wastes (TSDs), however, were often halted by public opposition. Public hearings during the permitting process for these facilities became a forum for debate over potential adverse impacts: groundwater contamination from accidental releases; airborne contamination from spills and incineration of wastes; noise and odors from plant operation and traffic; and threats to neighborhood property values. A survey of state hazardous waste officials in 1987 noted that nearly half of the commercial facilities that were rejected were stopped by public opposition (Mason, 1989). The gridlock in siting led to continual revision of state laws governing location of such locally undesirable land uses (LULU). By 1988, 41 states had enacted specific laws dealing with the siting of hazardous waste facilities (National Governors’ Association, 1989). During the same period, legal scholars, economists, and political scientists produced models aimed at breaking the siting impasse and easing the location of facilities that entailed concentrated costs and dispersed benefits (see Morell and Magorian, 1982; O’Hare, Bacow and Sanderson, 1983).

130

The international yearbook of environmental and resource economics

In 1977 three states in the US had statutes that dealt with the siting of hazardous waste facilities; by 1988, 41 states had passed such laws. The evolution of these laws demonstrate how more resources are devoted to structuring property rights as the returns to well-specified rights in a given area increase. While plants such as commercial hazardous waste facilities have always generated nuisances such as odors and truck traffic, vehement opposition to such facilities did not coalesce around most proposed sitings until the later part of the 1970s and the early 1980s. Few new facilities that treated hazardous waste generated at another site (e.g., commercial facilities) were sited in the 1980s, according to the permit data for the facilities in an EPA survey of hazardous waste TSDs (Hamilton, 1993). Demand for offsite waste treatment increased as more wastes were declared hazardous under federal legislation and companies that treated waste onsite were subject to strict regulatory scrutiny. Yet public opposition to commercial facilities led to siting gridlocks. Environmentalists used local zoning power to halt attempted sitings across the country, buttressed by additional litigation over whether proposed facilities violated environmental statutes. Against this background of public protest, state legislatures passed laws specifically detailing the procedures governing the siting of hazardous waste facilities. These laws spelled out with clarity the property rights of developers, localities, and state government in terms of the power to initiate sitings, compensate affected localities, and finalize the location of a particular facility. This marked a shift in the previous siting method that was often referred to as the ‘Decide-Announce-Defend’ model in which developers simply attempted to place the facility in a given locality without an extended public selection process. The changes in siting regimes for hazardous waste facilities are consistent with Coase’s insight about the relative transaction costs of different institutional arrangements. In an era when the perceived risks of waste handling were low, firms that operated hazardous waste facilities were free to site their facilities within the normal constraints on the market for industrial location imposed by zoning. As public opposition mounted, however, negotiations over siting became increasingly protracted and expensive. The transaction costs of siting battles became a weapon used by environmentalists to raise the costs of location to firms. State legislatures responded to a perceived need to increase available treatment capacity by drafting laws that explicitly dealt with siting of hazardous waste TSDs. Though the laws across states were similar in that they attempted to specify the particular property rights of developers, localities, and state government, they varied in the degree that they left the location decision to institutions that resembled a market versus a firm (National Governors’ Association, 1989). States that adopted a market model typically left the initiation of siting to a

Environmental equity and hazardous waste

131

private developer, specified a process of explicit negotiation between the developer of the facility and the targeted community, and provided for compensation mechanisms that transferred payments to affected localities. In a 1988 National Governors’ Association survey of state siting regimes, at least 13 of the states had laws that mandated the provision of compensation to host communities. Compensation took various forms, including a tax on gross receipts at the facility whose proceeds went to the locality, per ton tipping fees that went to the community, license fees, or general compensation packages that included money and in-kind contributions of goods and services. Twenty-two states had programs that entailed direct negotiations between developers and communities affected by proposed facilities. Sixteen states provided technical assistance grants to allow communities selected as potential sites to develop the expertise and information necessary to participate in negotiations over siting. Other states relied on a siting process that resembled a firm’s decisionmaking process in reliance on centralized decisionmaking power to ultimately determine the location of the facility. In 11 states, the state itself had authority to initiate the siting process for a new hazardous waste facility. Fourteen states gave their state governments preemption power, where the state essentially preempts the delegated zoning power of localities in this case and simply makes the siting decision. In this process, the state assumes zoning power and excludes the targeted community from the decisionmaking process. The NGA study found that 22 states provided state government with override power, so that a state agency or board could override a local decision that attempted to block a siting. The state would not initiate the siting, but the threat of using state power remained in the background so that it could be used to resolve siting disputes. Note that even where the firm model was adopted, often provisions for negotiation with a community and compensation were included in the siting statute. Local representatives from affected communities may be provided with membership on state siting boards. Such procedures are similar to mechanisms within a firm that help establish ‘transfer prices’ for internal transactions. Though the location decision has been moved within a central decisionmaking authority (in this case a state agency or board with ultimate decisionmaking power), there is still a need to elicit information on how communities vary in the value they place on environmental amenities. Statutes that provide a state agency or board with preemption or override power are attempts to solve the NIMBY gridlock by creating a firmlike structure in which centralized decisionmaking power makes the resource allocation decision, i.e., where the facility should be located. It remains an empirical question, however, whether such decisionmaking power can be effectively exercised. For even if the state nominally has the

132

The international yearbook of environmental and resource economics

right to locate a facility in an area where opposition runs deep, local governments may possess sufficient power to frustrate the developer and the state so that the attempt is ultimately unsuccessful. Local governments determined to block a siting have used police powers to slow traffic to sites to a standstill, disrupted roads leading into facilities for long-term construction, and established zoning requirements that generate further litigation. Environmentalists have also been successful in appealing to state legislatures to remove sites selected by agencies from consideration once they are targeted. In at least one state, the backlash against a siting process directed by a state commission caused the legislature to halt all funding for the commission. The leverage of local governments and possibility for appeal back to state legislatures thus weaken the actual exercise of ‘centralized’ decisionmaking power in states with preemption and override provisions. Morell and Magorian (1982) captured the essence of this operation of siting mechanisms in their book title Siting Hazardous Waste Facilities: Local Opposition and the Myth of Preemption. Bacow and Milkey (1982) have also pointed out that a state’s use of preemption power may simply lead facility opponents to turn to litigation to slow construction of a facility. No one configuration of property rights has become dominant across states or proved successful in facilitating the siting of commercial hazardous waste facilities. Most of the models of the siting process emphasize the use of market mechanisms such as auctions to solve the problem of determining where facilities will elicit the lowest demand for compensation. Discussions of the siting laws that have clarified property rights in this area often assume that the definition of property rights is the same thing as the establishment of a market for location of facilities, a market where firms internalize their externalities in the manner described in Coase II, ‘The Problem of Social Cost’. Yet a review of the siting statutes reveals that the procedures employed resemble both the arm’s length transactions of a market and the centralized decisionmaking power of the firm. Coase still provides insight into the structure of these laws, but it is the insight from Coase I that relative transaction costs determine the institutional framework used for making a decision that explains the evolution of siting laws. To see how siting progressed after the debate in the academic literature and state legislatures over effective procedures, Ibitayo and Pijawka (1999) conducted a national survey of state environmental agencies to analyse the siting of hazardous waste facilities over the previous decade. Of the 42 states responding, they found five states had successfully sited hazardous waste facilities, 12 states had mounted efforts that did not work, 13 had ongoing siting processes, and 12 had not attempted to site such facilities.

Environmental equity and hazardous waste

133

Contrasting the procedures in the five states with successful sitings with the efforts of other states, they conclude (p. 387): The success of any strategy depends on the extent to which the strategy deals with issues such as public trust, early and continuous public involvement in the siting process, public education on hazardous waste, empowerment of host communities, and the incorporation of citizens’ concerns into risk mitigation plans.

4.2 General Siting Policies for Hazardous Waste Facilities in other OECD Countries The concentrated costs and generally diffuse benefits of facilities that treat and dispose of hazardous waste have made their sitings controversial and relatively rare recently in most OECD countries. The literature on siting policies generally focuses on case studies within a given country and assessments by the authors of what factors led to siting success or failure in a particular case. The successful sitings of hazardous waste facilities in the Canadian provinces of Alberta and Manitoba have generated much research (Rabe, 1994; Castle and Munton, 1996; Richards, 1996). The voluntary siting process used there involved the stages of an expression of local interest by a community, hosting of open houses to discuss the facility with residents, creation of local citizen advisory committees funded by the proponent to investigate the process and plant, a referendum on the proposal, and a negotiation stage where facility operation issues and local input are discussed. This gives a community multiple points at which to veto a project. In terms of the socioeconomic characteristics of the two towns (Swan Hills and Montcalm) that accepted facilities, Castle and Munton note (p. 78): The evidence to support the argument that the Alberta or Manitoba volunteer siting process singled out poor or otherwise disadvantaged communities to host hazardous waste treatment facilities seems . . . lacking. . . . The town of Swan Hills had an average household income of $44 023 in 1986, significantly higher than the Alberta norm ($36 796). Unemployment in Swan Hills in 1986 was 7%, one of the lowest in the province. Montcalm is much poorer than Swan Hills, but the community is not economically disadvantaged compared to other rural areas in Manitoba.

The jobs associated with the plants were a considerable factor in public acceptance. In the case of Swan Hills, a town of approximately 2000 residents, a ‘1991 report concluded that the facility provides more than ninety full-time jobs in Swan Hills that contribute $2.7 million a year to the local economy in salaries and makes an overall impact of $6 million on the

134

The international yearbook of environmental and resource economics

economy each year’(Rabe, 1994, p. 76). Additionally, proponents in Alberta stressed that the waste would come primarily from within the province so that the siting was seen as solving a local problem rather than creating a outlet for waste from across the country. Additional research on siting waste facilities (Lawrence, 1996) and toxic disposal facilities (Ristoratore, 1987; Richards, 1996) in Canada analyses the use of voluntary process, the role of compensation or jobs in generating support, and the importance of ensuring a minimum level of safety in the operation of facilities. Research on hazardous waste facility siting in European countries reveals wide variety in public participation and reaction. Linnerooth-Bayer (1997) notes that survey evidence from Austria indicates multiple views on fairness in siting, with some stressing support for hierarchical processes based on expert opinion, others favoring market mechanisms, a segment objecting on distributive justice grounds to siting in economically or socially disadvantaged communities, and others believing in spreading the burdens of waste management broadly across communities. Lidskog (1993) emphasizes the role that economic benefits played in leading a local government to host a disposal facility in Sweden. Seeliger (1996) examines four case studies of siting waste facilities in Germany and stresses the large role of state governments in initiating the siting process and the experimentation with public consultation in more recent policies. Describing the lack of opposition in one neighborhood hosting an incinerator, he notes (p. 241): The GSB facility is located not in the town of Ebenhausen itself but a subsection called Ebenhausen-Werk . . . Ebenhausen-Werk is about a kilometer from the town proper, and home to a working-class population of 400–500, of low social status. The proportion of foreign and migrant workers living in Ebenhausen-Werk is also said to be high . . . . It is possible that this social setting is not conducive to the emergence of a local environmental initiative.

Analysing lessons from siting disputes in Europe overall, Kunreuther, Linnerooth-Bayer and Fitzgerald (1996) describe variations in siting policies by two main characteristics, the degree to which the process is open versus closed to public participation and the degree to which decisionmaking authority rests with the local community versus the national/state government or developer. They note examples of efforts taken to increase public support for proposed facilities, such as the case in Austria where local citizens were involved with the selection of experts chosen to assess risks from a facility and use of substantial compensation (i.e., a promise of 2.5 million Swiss francs for 25 years for a town of 500 households) in the siting of an underground nuclear waste repository in Switzerland. Dente, Fareri and Ligteringen (1998) review successful sitings of waste facilities in France, Hungary, Italy, the Netherlands, Slovenia, Spain, and

Environmental equity and hazardous waste

135

Switzerland. Their findings emphasize that many factors related to equity are involved in the siting of facilities in these European countries. In areas with higher unemployment rates, residents appear more likely to accept facilities because of the employment opportunities provided. In successful sitings those trying to locate the plants are more likely to succeed if they can transform the debate from a zero sum game in which one area gains and another loses to a situation where compensation in some form comes along with the risks of a facility. The distance the waste travels appears to matter to residents, for if the waste is seen as local then disposal is interpreted as more equitable. Fewer communities appear willing to accept waste generated far away. Siting attempts overall worked more easily in industrial areas for at least three reasons: the waste is seen as local; residents are more likely to work in plants generating the waste; and residents are used to the risks posed by industrial pollution. A final equity consideration emerges in the design of the siting process, with procedures involving more public participation being seen as more equitable. Huitema (2002) develops in depth three case studies of hazardous waste siting attempts in each of three different countries: Canada, the Netherlands, and the United Kingdom. In terms of procedure, he finds all three countries provide an increasing role for citizen participation, though this does not overwhelm the prominent roles played by experts and elected officials. Environmental impact assessments are now common elements of siting in each country. Overall the market plays a greater role in sitings in the UK and Canada and a lesser role in the Netherlands. Huitema notes that there are distinct procedural differences across the three countries, differences which may help explain outcomes in sitings. In the Netherlands siting disputes are much more likely to spill over into the courts. In the UK the inquiry system (a quasi-judicial investigation procedure) generates substantial dialogue about siting issues, though Huitema finds overall that regulations in the UK are likely to limit the ability of government to block the actions of private developers. Looking across the three countries he concludes that debates about siting are more likely to involve notions of managerialism (e.g., importance of experts) and conservative pluralism (e.g., right of private parties to pursue development) than ideas of distributive equity. Research on Japan notes the role of rewards used in the siting process. Shaw (1996) describes how laws in Japan smoothed the siting of power plants by providing communities with compensation without having to prove damages. Electricity is taxed, and the revenues are redistributed to local communities in the form of expenditures on public facilities such as roads, schools, or sewage systems. Ohkawara (1999) discusses compensation in the case of siting of nuclear power plants. Munton (1996) reviews the siting of hazardous waste facilities in Japan and notes problems similar

136

The international yearbook of environmental and resource economics

to that found in the US: NIMBY opposition to construction, declining capacity, and illegal dumping. Attempts to overcome opposition to siting included a policy of offering sidepayments to communities, which would receive public works projects from the national government in return for hosting a treatment facility. Probst and Beierle (1999) stress how facility operation is embedded in a larger regulatory structure affecting hazardous waste generation and management. They examine hazardous waste management systems in Germany, Denmark, United States, Canada, Malaysia, Hong Kong, Thailand, and Indonesia. Their results stress the time needed to establish a comprehensive system to regulate waste, the need to create a culture of compliance, and the expense of constructing hazardous waste treatment facilities. To encourage the development of treatment capacity, the countries studied had tried a range of ownership for the facilities (e.g., public, private, mixed). Initial subsidies from some governments encouraged generators to use the facilities, a step in creating compliance with hazardous waste management. A substantial legal and policy literature exists on international trade and hazardous waste (see Engfer et al., 1991; Walsh, 1992; Murphy, 1994; Pinzon, 1994; Kummer, 1995; Marbug, 1995; Bradford, 1997; Sundram, 1997; OECD, 1998; Park, 1998; Belenky, 1999; Verchick, 1999; O’Neill, 2000; Waugh, 2000). Part of this literature focuses on notions of fairness in the export of waste and risks across borders. Lofstedt (1996) analyses the literal spillover of risks across borders in the case of a nuclear power plant in Sweden sited close to the border with Denmark. Explicit consideration of the impact of siting on risk exposure by income or ethnicity is less prevalent in much of the European literature, though Johnson (2001) notes that analysis of such disparities is growing as more detailed information on pollution incidence becomes available in OECD countries. A European Commission March 1997 Council Directive on environmental impact assessment does specifically encourage the analysis of how waste disposal installations dealing with hazardous waste will affect the environment. The European Commission (2001) has also encouraged greater use of public participation mechanisms in environmental decisionmaking. While the literature on international trade and hazardous waste is too extensive to summarize here, it does highlight a distinct set of issues: the potential conflicts among policies that promote free trade and those that allow a country to restrict waste imports in the name of environmental protection; the tradeoffs involved in different stages of development between income, pollution, and growth; the connection between the preferences of constituents and the preferences pursued by government officials, which may vary dramatically by government type; the relative lack of environmental data in developing countries; and the substantial disparities in

Environmental equity and hazardous waste

137

potential valuation of environmental outcomes that arise when analysts use willingness to pay as a measure of value and the incomes of those modeled varies widely because they reside in different countries. 4.3 Incorporating Environmental Equity Considerations into Siting – US Perspective In 1994 President Clinton issued Executive Order 12898, ‘Federal Actions to Address Environmental Justice in Minority Populations and Low-Income Populations’, which required federal agencies to create environmental justice strategies and examine whether their policies have disproportionate impacts on minority or low-income populations. In response to this order, the EPA began to develop guidance documents to incorporate environmental equity considerations into government decisionmaking. The EPA’s Administrator, Christine Todd Whitman, noted in 2001 that the agency’s program includes: (a) Conducting our programs, policies, and activities that substantially affect human health and the environment in a manner that ensures the fair treatment of all people, including minority populations and/or low-income populations; (b) Ensuring equal enforcement of protective environmental laws for all people, including minority populations and/or low-income populations; (c) Ensuring greater public participation in the Agency’s development and implementation of environmental regulations and policies; and (d) Improving research and data collection for Agency programs relating to the health of, and the environment of all people, including minority populations and/or low-income populations.

The Agency’s efforts to define and implement environmental equity have attracted significant legal research (see Georges, 1999; Johnson, 1999; Mank, 1999; Foster, 2002; Yang, 2002). Yet the number of permitting or enforcement actions specifically taken on environmental justice concerns is so small that to date no statistical investigations have analysed the Agency’s implementation. In the most famous case to date, the transaction costs generated by the EPA’s environmental justice investigation of the Shintech plant in part led the company to shift the plant’s proposed location from a predominantly African American community to a white community (see Lambert, 2000). The EPA’s consideration of environmental justice concerns in permitting facilities (see EPA, 2000) has generated controversy among industry participants concerned about the freedom to locate facilities, officials in economically depressed areas who want to attract plants to stimulate the economy, and activists concerned that the EPA’s consideration of disparate impacts will not alter the location and operation of polluting facilities. The greatest lessons to date for considering equity in the

138

The international yearbook of environmental and resource economics

siting of hazardous waste facilities may lie in examining two sets of environmental equity guidance instructions adopted by the EPA. Under the National Environmental Policy Act (NEPA), EPA must often create an environmental impact statement (EIS) or environmental assessment (EA) of agency actions. In 1998 the EPA issued its guidance documents for incorporating environmental justice considerations into developing an EIS or EA. The guidance encourages analysts to look at affected populations using detailed census data; take into account how differences in diet might result in different exposure; look for cumulative effects (e.g., what other sources of pollution are already in the area); examine local health outcomes data; be cognizant of local literacy rates, especially when communicating complex risk information; consider occupational exposures; and determine whether community representatives are involved in local decisionmaking. The guidance encourages analysts to use GIS technology to analyse potential exposures. Mitigation measures are also to be incorporated into the analysis, including monitoring of emissions, encouraging participation of affected communities, reducing pollutants to reduce cumulative exposures, and requiring mitigation of pollution as part of a permitting process. The operations of many environmental regulatory programs are delegated to the states by the EPA, so that state environmental agencies will end up writing permits needed for the construction and operation of polluting facilities. The most controversial EPA environmental equity rules concern the guidance documents issued that define when a permit may be challenged under Title VI of the 1964 Civil Rights Act, which relates to discrimination involving disparate impacts by entities that receive federal funding (such as funding for state environmental programs). These rules have attracted great attention among legal scholars (see Lazarus and Tai, 1999; Lyle, 2000; Mank, 2000, 2001; Guana, 2001; Cody, 2002; Santiago, 2002). The guidance documents set forth what types of evidence the EPA would expect when parties file a civil rights complaint with the EPA about a state or local permitting decision. The guidance states that for evidence of disparate impacts the agency expects detailed information on the exposed population, which may include census data on racial composition in an area, GIS analysis of exposure to pollutants, information on reported chemical releases, monitoring data, and health outcomes information. The guidance notes that the EPA’s Office of Civil Rights (OCR) is unlikely to find an adverse health impact when cumulative risks of cancer calculated in the analysis are less than 106 and much more likely in cases where the cancer risk is 104 or higher; the higher the non-cancer hazard index for a chemical is above 1 the more likely the OCR is to find an adverse health impact. The rules indicate that in determining whether a plant generates a disparate impact that the comparison population can be the general surrounding

Environmental equity and hazardous waste

139

population (e.g., city or state) or the non-affected segment of the local population. The guidance documents suggest that a denial of a permit solely on environment justice grounds is unlikely. Rather, if the EPA finds a violation the agency will focus on mitigating the impact of a facility, such as reducing permissible pollution levels or requiring stricter monitoring.

CONCLUSIONS Statistical studies in the United States suggest that low-income and minority populations are often exposed to greater risks arising from the siting and operation of hazardous waste facilities. Less detailed information exists on the exposure of residents by income to hazardous waste risks in other OECD countries, though the evidence suggests that in some countries disparate exposures may exist by income (in part because of the draw of jobs and compensation programs in siting procedures). Concerns about environmental equity may involve dissatisfaction with disparate exposure or dismay at how these differences arise across demographic groups. Thinking about policies to address environmental equity (see Table 3.3) requires definitions of equity, an assortment of policy tools that take into account how disparate impacts arise, and a recognition of the prospect that market dynamics may make equity policies hard to implement in the long run. Besides the general links between income, minority populations, and risk, the statistical studies to date also hold a number of specific results: The conclusions about environmental equity one can draw often depend on the geographic unit of observation used in the research, which at times depends more on data costs than the expected radius of externalities generated by a facility. Studies frequently use the presence/absence of a plant as an indicator of hazards, though more sophisticated risk or exposure assessments should be used in the future. In the US research results often depend on how population groups are defined (e.g., minority? African American? Hispanic? Asian American?). Research to date focuses more on describing exposure than testing theories of causation, in part because of the difficulty of assembling data on risk and demographics over time. Researchers may also vary in the standards they use in judging outcomes, ranging from analysts who focus primarily on efficiency and willingness to pay to others who focus primarily on equity and fairness. To see the difficulties in doing research on equity, consider first how to define a fair siting process. Been (1993, p. 1008) notes at least seven ways that fair siting is defined: even distribution of NIMBY sites across neighborhoods; compensation to those affected by residents in neighborhoods that do not host facilities; progressive siting, where areas with more income bear

140

The international yearbook of environmental and resource economics

more sitings or pay more in damages; equal vetoes by communities to bid in excluding facilities; siting where those who benefit pay the cost; no intentional discrimination against minorities; and a process that shows ‘equal concern and respect’ for all communities. Hampton (1999) adds the complication that concepts of fairness reside in both procedures and outcomes, so fair policies may involve meeting standards of participation, information provision, trust in the decisionmaking process, and expression of public values. The toolbox of policy options to promote equity in siting is large. Site permitting could involve an analysis of whether a plant involves a disparate impact on low-income or minority communities, with impact defined by the radius of externalities and risks modeled using GIS technology and information on emissions or adverse health outcomes. If differences in collective action give rise to disparities, authorities may take steps to increase involvement such as technical assistance grants to communities involved in siting, information provision, or attentive enforcement of rules in areas less likely to call upon regulators for help. The process of negotiation in siting and permitting may give rise to mitigation efforts, which may include reductions in emissions or increases in monitoring for ambient exposures. Compensation for siting may come in numerous forms, such as the jobs provided by plants, tax payments that reduce tax burdens in a community, or expenditures on public goods such as roads or schools. If disparate impact arises from racism in housing markets or job market outcomes that limit incomes, then attempts to address discrimination in these markets will provide minority residents with greater freedom to avoid environmental hazards. Many questions remain open to further research in this area. Hedonic price studies and risk assessment analyses could explore the exact radius of negative externalities generated by different types of facilities. The mapping of cumulative environmental impacts from polluting plants with GIS technology would improve the picture of environmental risks borne by different demographic groups. While the current literature often describes risk in terms of the presence or absence of a facility in the neighborhood, future researchers may be more likely to use GIS and risk modeling to describe how the cancer risks for an individual or the expected number of cancer cases in a community rise with the operation of a plant. The development of Pollutant Release and Transfer Registers (PRTRs) in many countries will allow analysts to explore how political institutions may affect the distribution of pollution and risk across groups. The operation of the PRTRs in turn will generate a literature on how information provision operates across countries (see Hamilton (2004) for analysis of one information provision program in the US, the Toxics Release Inventory program). Going inside the plants will allow researchers to explore the risks that workers are exposed to at hazardous waste TSDs, which may be higher than those

Environmental equity and hazardous waste

141

modeled for residents in the surrounding community. Research may also focus on the factors which determine what TSDs generate more negative externalities in terms of current flawed operating procedures and future leaking of chemicals into the groundwater or soil. Siting studies may also focus more on the impact of existing polluting facilities on the decisions of other plants to locate in an area, the differences in siting decisions between public facilities (e.g., military bases) and private facilities, and how variations in the political participation of particular demographic groups (e.g., immigrants) may affect their power in siting disputes. At least three problems may arise with equity policies. In the short run some compensation may offset damages for current residents when a facility is sited. Yet in the long run a facility may generate externalities that end up lowering property values and attracting more low-income residents to an area. These low-income individuals may be willing because of their constrained budgets to accept a larger risk for a lower housing price, though this offends the values of individuals who do not believe environmental quality should be traded like a normal good. A related point is that in some areas poor residents may be willing to accept a facility because of the jobs provided, which again may generate dissatisfaction among those who have preferences about the distribution of risk across demographic groups. If plants are channeled outside of current industrial areas because of equity concerns, this raises the likelihood that new environments will be diminished. In devising environmental equity policies, there are few easy and obvious choices. Selection of programs to pursue environmental equity involve potential tradeoffs between equity and efficiency, across demographic groups, and among values relating to procedures, outcomes, and self-determination. Table 3.1

Evidence on the distribution of hazardous waste facilities Sample of US studies

Anderton et al. (1994a, b) Some 408 census tracts (c. 4000 residents each) with commercial TSDs had slightly higher percentage poverty than other tracts in nation in 1980. Within top 25 cities, tracts with commercial TSDs do not have statistically significant difference in percentage poverty compared to other tracts in the areas. If one compares 408 tracts with 4239 tracts within 2.5 miles of facility, tracts closer to plant have higher percentage workers in precision occupations and lower percentage of families in poverty. If one compares the location and nearby tracts (4647) with other national tracts, the tracts affected by TSDs have higher percentage of families below poverty (19% vs 13%) and lower housing prices ($45 876 vs $60 291). Conclusion: Radius? Reference group?

142

The international yearbook of environmental and resource economics

Table 3.1

(continued)

Atlas (2001) In 1997, 108 commercial TSDs (in sample if accounted for at least 0.2% managed wastes). For 0.5 mile ring around facility, mean low-income population percentage (150% of poverty level) was 29.6 per cent and 30.4 per cent for twomile ring; nationwide 21.7 per cent. On a population weighted basis: 25.9 per cent of residents within 0.5 mile and 30.4 per cent for two-mile ring were low-income populations. Hamilton (1995) Of 207 zip codes with commercial hazardous waste facilities in 1987, 84 areas had plans for expansion in capacity. Mean percentage of families in poverty higher in zips targeted for expansion (14% vs 11%) and average of median household income was lower ($15 750 vs $17 060). Expansion planned in areas with lower populations, more minorities, poorer populations, and less politically active individuals. Hamilton and Viscusi (1999) For 1173 hazardous waste sites being cleaned up under the Superfund program, site-level mean household incomes lower at the one-mile ring ($36 930) and four-mile ring ($37 690) than national average ($38 450 at 1990 census). At 61 per cent of sites, mean household income lower in 0–1 ring than 1–4-mile ring. Site-level mean house values for residents within one mile ($98 590) or within four miles ($103 900) were lower than the US mean ($112 660). Also calculated individual cancer risks, expected cancer cases, current land use at 150 sites. Found some evidence that minority groups account for a larger fraction of the estimated cancers than their national population, evidence that population weighted mean maximum individual cancer risk higher for minorities, and strong evidence that minorities bear larger current risks arising from present land uses at sites. Studies in other OECD countries Friends of the Earth (1999) Examines postcode location of industrial plants registered under Integrated Pollution Control program and household income distribution by postcode. Found: All across England and Wales the poorest families (reporting average household income below 5000 [GBP]) are twice as likely to have a polluting factory close by than those with average household incomes over 60 000 [GBP]. . . . Overall, almost two-thirds of the most polluting industrial facilities are to be found in areas of below average income.

Friends of the Earth (2001) Analyses distribution of 156 plants in England emitting more than 1000 kilogrammes of carcinogens in 1999. Found ‘66 per cent of carcinogen emissions

Environmental equity and hazardous waste

Table 3.1

143

(continued)

are in the most deprived 10 per cent of wards; 82 per cent of carcinogen emissions are in the most deprived 20 per cent of wards; only 8 per cent of carcinogen emissions are in the least deprived 50 per cent of wards’. Jerrett, Eyles, Cole and Reader (1997) Examines Canada’s National Pollutant Release Inventory. Model county aggregate pollution as function four variables: median income per household, average dwelling value, total population, manufacturing employment. All statistically significant, with income, population, and manufacturing positive and housing negative. Harrison and Antweiler (2002) Using Canadian NPRI at facility level,‘generally do not find significant impacts of community income on either the current releases or changes in releases over time’.

Table 3.2

Determinants of exposure to hazardous waste facilities Sample of US studies

Hamilton (1993, 1995) In expansion plans of commercial hazardous waste facilities, zip-code areas targeted for expansion had lower voting rates, fewer people, and higher percentage of renters. Hamilton (1999) Uses Toxics Release Inventory data to analyse reduction in air carcinogen emissions between 1988 and 1991. Plants reduced emissions more the greater the expected cancers generated by the facility and the higher the voting rate around the plant, a proxy for collective action. Median household income and minority percentage in the zip code were not statistically significant. Viscusi and Hamilton (1999) At cleanup of hazardous waste sites, when cancer risks are low more stringent cleanups are chosen if surrounding residents are more politically active. The higher the average income in the one-mile ring around a site, less stringent cleanup chosen (though this may be because wealthier residents can take more preventive measures on their own). Higher voter turnout at a site, greater cost per cancer case averted implied in EPA cleanup. Income level of residents had no impact on cleanup expenditure levels. Been and Gupta (1997) Examines census tract data for 1970, 1980, 1990. Found that when TSDs originally sited they were not located in areas with high concentrations of the poor or African Americans. Locations did have a disproportionate share of Hispanics.

144

The international yearbook of environmental and resource economics

Table 3.2

(continued)

Lambert and Boerner (1997), Pastor, Sadd and Hipp (2001) Mixed evidence on whether TSDs sited in poor areas originally or whether poor residents moved to areas after plants were located. Sample studies in other OECD countries: surveys and case studies Frey and Oberholzer-Gee (1996, 1997), Frey, Oberholzer-Gee and Eichenberger (1996) Survey interviews in Switzerland in 1993 before referendum on nuclear waste repositories. Willingness to have repository located in a resident’s community declined as perceived risks or negative economic impacts were larger. Compensation offers in the survey reduced willingness to accept the nuclear waste site. Compensation offers can crowd out a feeling of civic duty. Lesbirel (1998) Compensation facilitates siting of energy plants in Japan. Fisher (1995) Examines successful siting of hazardous waste treatment plant in Alberta, Canada. Ascribes approval to early local plebiscite on accepting siting, regional government’s provision of funds for local community to hire experts to analyse plant impacts, government’s provision of compensation for infrastructure costs and more experts, and formation of local committee to monitor plant operation. Numerous case studies in Europe Emphasize role of compensation, unemployment, public participation in explaining success/failure of siting.

Table 3.3

Conclusions of Chapter 3

How can environmental equity be incorporated in siting policies? Analysis of impact; technical assistance grants to communities to analyse facilities; mitigation efforts, including reductions in pollution or increases in monitoring; compensation mechanisms (e.g., payments for community infrastructure). Example: In the US, Executive Order requires agencies to incorporate environmental justice considerations (e.g., exposure by race and income) into their policies. EPA guidance documents on when permit may be challenged on civil rights grounds ●

For evidence on disparate exposure, expect detailed information. Examples would include GIS analysis of census population, potential routes of

Environmental equity and hazardous waste

Table 3.2

● ● ●

145

(continued)

exposure, monitoring data on cumulative impacts, and health outcomes information. Agency unlikely to find adverse health impact if cumulative risks of cancer calculated are less than 106, more likely if 104. Comparison population: general population, or non-affected local population. Guidance suggests unlikely EPA would deny permit solely on environmental justice grounds. Rather, agency would focus on reducing pollution levels or requiring stricter monitoring.

Challenges to incorporating equity in siting ●

●

●

●

●

Residents may accept risks in return for jobs provided by facilities. Policies that steer facilities away from poor/minority areas may discourage development/employment. Even if local residents who are poor are willing to accept facility, those outside the area may have existence/bequest motives of justice that hold this tradeoff is unfair. Whose preferences matter more? How incorporate existence values (contingent valuation)? Related point – procedures have intrinsic and instrumental fairness values. If compensation is involved in siting, where does it flow? How can you avoid principal–agent problems between representatives and residents in compensation negotiations? Will environmental equity policies force more sitings into green fields? Where is marginal damage of plant greater, already polluted area or new environment? What are the long-term impacts on the poor of cleaning up an area? Rise in housing prices that causes them to relocate?

These are the types of concerns that make debates over environmental equity difficult to resolve.

NOTE 1. This chapter is based on a paper prepared for the OECD Environment Directorate March 2003 Paris Workshop, which resulted in the volume The Distribution of Environmental Benefits and Costs: Economics and Policy Issues, OECD, forthcoming, Cheltenham, UK and Northampton, USA: Edward Elgar.

146

The international yearbook of environmental and resource economics

REFERENCES Anderton, Douglas L., Andy B. Anderson, John Michael Oakes and Michael R. Fraser (1994), ‘Environmental equity: the demographics of dumping’, Demography, 31(2), 229–48. Anderton, Douglas L., Andy B. Anderson, Peter H. Rossi, John Michael Oakes, Michael R. Fraser, Eleanor W. Weber and Edward J. Calabrese (1994), ‘Hazardous waste facilities: “environmental equity” issues in metropolitan areas’, Evaluation Review, 18(2), 123–40. Anderton, Douglas L., John Michael Oakes and Karla L. Egan (1997), ‘Environmental equity in superfund: demographics of the discovery and prioritization of abandoned toxic waste sites’, Evaluation Review, 21(1), 3–26. Arora, Seema and Timothy N. Cason (1999), ‘Do community characteristics influence environmental outcomes? Evidence from the toxics release inventory’, Southern Economic Journal, 65(4), 691–716. Atlas, Mark K. (2001a), ‘Safe and sorry: risk, environmental equity, and hazardous waste management facilities’, Risk Analysis, 21(5), 939–54. — (2001b), ‘Rush to judgment: an empirical analysis of environmental equity in US Environmental Protection Agency actions’, Law and Society Review, 35, 633–82. Bacow, Lawrence S. and James R. Milkey (1982), ‘Overcoming local opposition to hazardous-waste facilities: the Massachusetts approach’, Harvard Environmental Law Review, 6, 265–305. Baden, Brett M. and Don L. Coursey (2002), ‘The locality of waste sites within the City of Chicago: a demographic, social, and economic analysis’, Resource and Energy Economics, 24, 53–93. Baxter, Jamie W., John D. Eyles and Susan J. Elliott (1999), ‘From siting principles to siting practices: a case study of discord among trust equity, and community participation’, Journal of Environmental Planning and Management, 42(4), 501–25. Been, Vicki (1993), ‘What’s fairness got to do with it? Environmental justice and the siting of locally undesirable land uses’, Cornell Law Review, 78, 1001–85. — (1994), ‘Locally undesirable land uses in minority neighborhoods: disproportionate siting or market dynamics?’, Yale Law Journal, 103(6), 1383–1422. — (1995), ‘Analyzing evidence of environmental justice’, Journal of Land Use and Environmental Law, 11(1), 1–36. Been, Vicki and Francis Gupta (1997), ‘Coming to the nuisance or going to the barrios? A longitudinal analysis of environmental justice claims’, Ecology Law Quarterly, 24, 1–56. Belenky, Lisa T. (1999), ‘Cradle to border: US hazardous waste export regulations and international law’, Berkeley Journal of International Law, 17, 95–137. Boer, Tom, J. Manuel Pator, Jr., James L. Sadd and Lori D. Snyder (1997), ‘Is there environmental racism: the demographics of hazardous waste in Los Angeles County’, Social Science Quarterly, 78(4), 793–810. Bord, Richard J. and Robert E. O’Connor (1992), ‘Determinants of risk perceptions of a hazardous waste site’, Risk Analysis, 12(3), 411–29. Bowen, William (2002), ‘An analytical review of environmental justice research: what do we really know?’, Environmental Management, 29(1), 3–15. Boyse, James K., Andrew R. Klemer, Paul H. Templet and Cleve E. Willis (1999), ‘Power distribution, the environment, and public health: a state-level analysis’, Ecological Economics, 29, 127–40. Bradford, Mark (1998), ‘The United States, China, and the Basel convention on the

Environmental equity and hazardous waste

147

transboundary movements of hazardous wastes and their disposal’, Fordham Environmental Law Journal, 8, 305–49. Brainard, Julii S., Andrew P. Jones, Ian J. Bateman, Andrew A. Lovett and Peter J. Fallon (2002), ‘Modeling environmental equity: access to air quality in Birmingham, England’, Environment and Planning A, 34, 695–716. Broadbent, Jeffrey (1998), Environmental Politics in Japan: Networks of Power and Protest, New York: Cambridge University Press. Brooks, Nancy and Rejiv Sethi (1997), ‘The distribution of pollution: community characteristics and exposure to air toxics’, Journal of Environmental Economics and Management, 32, 233–50. Bryant, Bunyan and Paul Mohai (eds) (1992), Race and the Incidence of Environmental Hazards: A Time for Discourse, Boulder, CO: Westview. Bullard, Robert D. (1983), ‘Solid waste sites and the black Houston community’, Social Inquiry, 53, 273–88. — (1990), Dumping in Dixie: Race, Class and Environmental Quality, Boulder, CO: Westview. — (1996), ‘Environmental justice: it’s more than waste facility siting’, Social Science Quarterly, 77(3), 493–9. Canadian Institute for Environmental Law and Policy (2000), Ontario: Open for Toxics. Castle, Geoffrey and Don Munton (1996), ‘Voluntary siting of hazardous waste facilities in western Canada’, in Don Munton (ed.), Hazardous Waste Siting and Democratic Choice, Washington, DC: Georgetown University Press. Chakraborty, Jayajit (2001), ‘Acute exposure to extremely hazardous substances: an analysis of environmental equity’, Risk Analysis, 21(5), 883–95. Christiansen, Kim Michael and Birgit Munck-Kampmann (2000), ‘Waste: annual topic update 1999’, European Environment Agency topic report 2/2000. Coase, Ronald (1937), ‘The nature of the firm’, Economica, 4, 386–405. — (1960), ‘The problem of social cost’, Journal of Law and Economics, 3, 1–44. Cody, Brendan (2002), ‘South Camden citizens in action: siting decisions, disparate impact discrimination, and section 1983’, Ecology Law Quarterly, 29, 231–62. Coenen, Frans H.J.M. (1998), ‘Participation in strategic green planning in the Netherlands’, in Frans Coenen, Dave Huitema and Laurence J. O’Toole, Jr. (eds), Participation and the Quality of Environmental Decision Making, Dordrecht: Kluwer Academic. Commission for Environmental Cooperation (1999), Tracking and Enforcement of Transborder Hazardous Waste Shipments in North America. Commission for Racial Justice, United Church of Christ (UCC) (1987), Toxic Wastes and Race in the United States: A National Report on the Racial and Socioeconomic Characteristics of Communities With Hazardous Waste Sites, New York: United Church of Christ. Connor, Michael (1992), ‘Maquiladoras and the border environment: prospects for moving from agreements to solutions’, Colorado Journal of International Environmental Law and Policy, 3, 683–710. Cutter, Susan L., Danika Holm and Lloyd Clark (1996), ‘The role of geographic scale in monitoring environmental justice’, Risk Analysis, 16(4), 517–26. Daniels, Glynis and Samantha Friedman (1999), ‘Spatial inequality in the distribution of industrial toxic releases: evidence from the 1990 TRI’, Social Science Quarterly, 80(2), 244–62.

148

The international yearbook of environmental and resource economics

Dente, Bruno, Paolo Fareri and Josee Ligteringen (eds) (1998), The Waste and the Backyard: The Creation of Waste Facilities in Six European Countries, Dordrecht: Kluwer Academic. Dolk, H., M. Vrijheid, B. Armstrong, L. Abramsky, F. Bianci, E. Garne, V. Nelen, E. Robert, J.E.S. Scott, D. Stone and R. Tenconi (1998), ‘Risk of congenital anomalies near hazardous-waste landfill sites in Europe: the EUROHAZCON study’, The Lancet, 352, 423–7. Dunlap, Riley E., Michael E. Kraft and Eugene A. Rosa (1993), Public Reactions to Nuclear Waste, Durham, NC: Duke University Press. Easterling, Douglas (1992), ‘Fair rules for siting a high-level nuclear waste repository’, Journal of Policy Analysis and Management, 11(3), 442–75. Engfer, Victoria L., Gilbert A. Partida, Thomas C. Vernon, Alejandro Toulet and David A. Renas (1991), ‘By-products of prosperity: transborder hazardous waste issues confronting the maquiladora industry’, San Diego Law Review, 28, 819–51. Environmental Protection Agency (1998), Final Guidance for Incorporating Environmental Justice Concerns in EPA’s NEPA Compliance Analyses. Environmental Protection Agency, Office of Environmental Justice (2000), Environmental Justice in the Permitting Process: A Report from the National Environmental Justice Advisory Council’s Public Meeting on Environmental Permitting, Arlington, Virginia, November 30–December 2, 1999, Washington, DC: US EPA. European Commission (1997), Council Directive 97/11/EC, Amending Directive 85/337/EC of 27 June 1985 on the Assessment of the Effects of Certain Public and Private Projects on the Environment. — (2001), Proposal for a Directive of the European Parliament and of the Council, Providing for Public Participation in Respect of the Drawing Up of Certain Plans and Programmes Relating to the Environment and Amending Council Directives, 85/337/EEC and 96/61/EC. Farber, Stephen (1996), ‘Undesirable facilities and property values: a summary of empirical studies’, Ecological Economics, 24, 1–14. Feldman, Elliot J. and Michael A. Goldberg (eds) (1987), Land Rites and Wrongs, Cambridge, MA: Lincoln Institute on Land Policy. Fischer, Frank (1995), ‘Hazardous waste policy, community movements and the politics of NIMBY: participatory risk assessment in the USA and Canada’, in Frank Fisher and Michael Black (eds), Greening Environmental Policy: The Politics of a Sustainable Future, New York: St.Martin’s. Foreman, Christopher H. (1998), The Promise and Peril of Environmental Justice, Washington, DC: Brookings Institution. Foster, Sheila (2002), ‘Environmental justice in an era of devolved collaboration’, Harvard Environmental Law Review, 26, 459–98. Fredriksson, Per G. (2000), ‘The siting of hazardous waste facilities in federal systems’, Environmental and Resource Economics, 15, 75–87. Frey, Bruno S. and Felix Oberholzer-Gee (1996), ‘Fair siting procedures: an empirical analysis of their importance and characteristics’, Journal of Policy Analysis and Management, 15(3), 353–76. — (1997), ‘The cost of price incentives: an empirical analysis of motivation crowding-out’, American Economic Review, 87(4), 746–55. Frey, Bruno S., Felix Oberholzer-Gee and Reiner Eichenberger (1996), ‘The old lady visits your backyard: a tale of morals and markets’, The Journal of Political Economy, 104(6), 1297–1313.

Environmental equity and hazardous waste

149

Friends of the Earth (2001), Pollution and Poverty: Breaking the Link. Gaussier, Nathalie (2001), ‘The spatial foundations of obnoxious goods location: the garbage dumps case’, Regional Studies, 35(7), 625–36. Gayer, Ted (2000), ‘Neighborhood demographics and the distribution of hazardous waste risks: an instrumental variables estimation’, Journal of Regulatory Economics, 17(2), 131–55. Gayer, Ted, James T. Hamilton and Kip W. Viscusi (2000), ‘Private values of risk tradeoffs at superfund sites: housing market evidence on learning about risk’, Review of Economics and Statistics, 82(3), 439–51. Georges, Elizabeth (1999), ‘If I had a hammer: why permitting challenges do not fit in the fight for environmental justice’, Fordham Environmental Law Journal, 10, 347–57. Gerrard, Michael B. (1994), Whose Backyard, Whose Risk: Fear and Fairness in Toxic and Nuclear Waste Siting, Cambridge, MA: MIT Press. — (1997), ‘Territoriality, risk perception and counterproductive legal structures: the case of waste facility siting’, Environmental Law, 27, 1017–33. Graham, John D., Nancy Dean Beaulier, Dana Sussman, March Sadowitz and Yi-Ching Li (1999), ‘Who lives near coke plants and oil refineries? An exploration of the environmental inequity hypothesis’, Risk Analysis, 9(2), 171–86. Gregory, Robin, Howard Kunreuther, Dough Easterling and Ken Richards (1991), ‘Incentives policies to site hazardous waste facilities’, Risk Analysis, 11(4), 667–75. Groothuis, Peter A. and Gail Miller (1994), ‘Locating hazardous waste facilities: the influence of NIMBY beliefs’, American Journal of Economics and Sociology, 53(3), 335–46. — (1997), ‘The role of social distrust in risk-benefit analysis: a study of the siting of a hazardous waste disposal facility’, Journal of Risk and Uncertainty, 15, 241–57. Guana, Eileen (2001), ‘EPA at 30: Fairness in Environmental Protection’, Environmental Law Reporter, 31, 10528. Gupta, Shreekant, George VanHoutven and Maureen L. Cropper (1996), ‘Paying for permanence: an economic analysis of EPA’s cleanup decisions at superfund sites’, Rand Journal of Economics, 27(3), 563–82. Halstead, John M., Joanna L. Whitcomb and Lawrence C. Hamilton (1999), ‘Economic insights into the siting problem: an application of the expected utility model’, Agricultural and Resource Economics Review, 28(1), 65–75. Hamilton, James T. (1993), ‘Politics and social costs: estimating the impact of collective action on hazardous waste facilities’, RAND Journal of Economics, 24(1), 101–25. — (1995), ‘Testing for environmental racism: prejudice, profits, political power?’, Journal of Policy Analysis and Management, 14(1), 107–32. — (1999), ‘Exercising property rights to pollute: do cancer risks and politics affect plant emission reductions?’, Journal of Risk and Uncertainty, 18(2), 105–24. — (2004), ‘Regulation through revelation: the origin and impacts of the toxics release inventory program’, book manuscript. Hamilton, James T. and Kip W. Viscusi (1999), Calculating Risks? The Spatial and Political Dimensions of Hazardous Waste Policy, Cambridge, MA: MIT Press. Hampton, Greg (1999), ‘Environmental equity and public participation’, Policy Sciences, 32, 163–74. Hanemann, Michael W. (1991), ‘Willingness to pay and willingness to accept: how much can they differ?’, American Economic Review, 81(3), 635–47.

150

The international yearbook of environmental and resource economics

Harjula, Henrik (2003), e-mail from Henrik Harjula (Principal Administrator, Environment Directorate, OECD) to the author, 11 March. Harrison, Kathryn and Werner Antweiler (2002), ‘Incentives for pollution abatement: regulation, regulatory threats and non-governmental pressures’, working paper, University of British Columbia, Vancouver. Hird, John A. (1993), ‘Environmental policy and equity: the case of superfund’, Journal of Policy Analysis and Management, 12(2), 323–43. — (1994), Superfund: The Political Economy of Environmental Risk, Baltimore, MD: Johns Hopkins University Press. Hite, Diane (2000), ‘A random utility model of environmental equity’, Growth and Change, 31, 40–58. Hockman, Elaine M. and Charles M. Morris (1998), ‘Progress towards environmental justice: a five-year perspective of toxicity, race and poverty in Michigan, 1990–1995’, Journal of Environmental Planning and Management, 41(2), 157–76. Huitema, Dave (1998), ‘Hazardous decisions: the siting of hazardous waste disposal facilities in Canada and the United States’, in Frans Coenen, Dave Huitema and Laurence J. O’Toole, Jr. (eds), Participation and the Quality of Environmental Decision Making, Dordrecht: Kluwer Academic. Huitema, Dave (2002), Hazardous Decisions: Hazardous Waste Siting in the UK, The Netherlands and Canada. Institutions and Discourses, Dordrecht: Kluwer Academic. Ibitayo, O.O. and K.D. Pijawka (1999), ‘Reversing NIMBY: an assessment of state strategies for siting hazardous-waste facilities’, Environment and Planning C: Government and Policy, 17, 379–89. Ingberman, Daniel E. (1995), ‘Siting noxious facilities: are markets efficient?’, Journal of Environmental Economics and Management, 29, S20–S33. Inhaber, Herbert (1992), ‘Of LULU, NIMBY, and NIMTOO’, Public Interest, 107, 52–64. Jerrett, M., J. Eyles, D. Cole and S. Reader (1997), ‘Environmental equity in Canada: an empirical investigation into the income distribution of pollution in Ontario’, Environment and Planning A, 29, 1777–1800. Johnson, Stephen M. (1999), ‘Economics v. equity: “Do market-based environmental reforms exacerbate environmental injustice?” ’, Washington and Lee Law Review, 56, 111–66. — (2001), ‘Economics v. equity II: the European experience’, Washington and Lee Law Review, 58, 417–86. Kiel, Katherine A. and Katherine T. McClain (1995), ‘House prices during siting decision stages: the case of an incinerator from rumor through operation’, Journal of Environmental Economics and Management, 28, 241–55. Kleindorfer, Paul R., Howard C. Kunreuther and David S. Hong (eds) (1996), Energy, Environment, and the Economy: Asian Perspectives, Cheltenham, UK and Northampton, MA: Edward Elgar. Kleinhesselink, Randall and Eugene A. Rosa (1994), ‘Nuclear trees in a forest of hazards: a comparison of risk perceptions between American and Japanese university students’, in Thomas C. Lowinger and George W. Hinman (eds), Nuclear Power at the Crossroads: Challenges and Prospects for the Twenty-First Century, Boulder, CO: International Research Center for Energy and Economic Development. Kruize, Hanneke and Arno Bouwman (2003), ‘The distribution of benefits and

Environmental equity and hazardous waste

151

costs of environmental policies: a case study on the distribution of environmental impacts in the Rijnmond region, the Netherlands’, paper prepared for OECD Environment Directorate Workshop, March. Kummer, Katharina (1995), International Management of Hazardous Wastes: The Basel Convention and Related Legal Rules, Oxford: Clarendon Press. Kunreuther, Howard and Douglas Easterling (1990), ‘Are risk-benefit tradeoffs possible in siting hazardous facilities?’, The American Economic Review, 80(2), 252–6. — (1996), ‘The role of compensation in siting hazardous facilities’, Journal of Policy Analysis and Management, 15(4), 601–22. Kunreuther, Howard and Paul R. Kleindorfer (1986), ‘A sealed-bid auction mechanism for siting noxious facilities’, American Economic Review, 76, 295–9. Kunreuther, Howard, Paul R. Kleindorfer, Peter J. Knez and Rudi Yaksick (1987), ‘A compensation mechanism for siting noxious facilities: theory and experimental design’, Journal of Environmental Economics and Management, 14, 371–83. Kunreuther, Howard, Joanne Linnerooth-Bayer and Kevin Fitzgerald (1996), ‘Siting hazardous waste facilities: lessons from Europe and America’, in Paul R. Kleindorfer, Howard C. Kunreuther and David S. Hong (eds), Energy, Environment, and the Economy: Asian Perspectives, Cheltenham, UK and Northampton, MA: Edward Elgar. Lambert, Thomas (2000), ‘EPA’s “Revised Guidance” for implementing Title VI: environmental justice on faulty legal footing’, Washington University Center for the Study of American Business policy brief 206. Lambert, Thomas and Christopher Boerner (1997), ‘Environmental inequity: economic causes, economic solutions’, Yale Journal on Regulation, 14, 195–234. Lawrence, D.P. (1996), ‘Approaches and methods of siting locally unwanted waste facilities’, Journal of Environmental Planning and Management, 39(2), 165–87. Lazarus, Richard and Stephanie Tai (1999), ‘Integrating environmental justice into EPA permitting authority’, Ecology Law Quarterly, 26, 617–68. Lejano, Paul P. and Climis A Davos (2002), ‘Fair share: siting noxious facilities as a risk distribution game under nontransferable utility’, Journal of Environmental Economics and Management, 43, 251–66. Lesbirel, S. Hayden (1998), NIMBY Politics in Japan: Energy Siting and the Management of Environmental Conflict, Ithaca, NY: Cornell University Press. Lidskog, R. (1993), ‘Whose environment? Which perspective? A critical approach to hazardous waste management in Sweden’, Environment and Planning A, 25, 571–88. Linnerooth-Bayer, Joanne (1997), ‘Siting hazardous waste facilities: issues of fairness’, in R.C. Ragaini (ed.), Risk Management Strategies Applied to Environmental Cleanup in Central and Eastern Europe, Singapore: World Scientific. Linnerooth-Bayer, Joanne and Kevin B. Fitzgerald (1995), ‘Conflicting views on fair siting processes: evidence from Austria and the US’, Risk: Health, Safety and Environment, 7, 119–34. Linnerooth-Bayer, Joanne and Ragnar E. Lofstedt (1995), ‘Fairness and siting: introduction to a symposium’, Risk: Health, Safety and Environment, 7, 95–8. Liu, Feng (1997), ‘Dynamics and causation of environmental equity, locally unwanted land uses, and neighborhood changes’, Environmental Management, 21(5), 643–56.

152

The international yearbook of environmental and resource economics

Lofstedt, Ragnar E. (1995), ‘Fairness across borders: the Barseback Nuclear Power Plant’, Risk: Health, Safety, and Environment, 7, 135–44. Lowinger, Thomas and George Hinman (1994), Nuclear Power at the Crossroads: Challenges and Prospects for the Twenty-First Century, Boulder, CO: International Research Center for Energy and Economic Development. Lyle, June (2000), ‘Reactions to EPA’s interim guidance: the growing battle for control over environmental justice decisionmaking’, Indiana Law Journal, 75, 687–709. Mank, Bradford (1999), ‘Is there a private cause of action under EPA’s Title VI regulations? The need to empower environmental justice plaintiffs’, Columbia Journal of Environmental Law, 24, 1–60. — (2000), ‘The draft Title VI recipient and revised investigation guidance: too much discretion for the EPA and a more difficult standard for complainants?’, Environmental Law Reporter, 30, 11144. — (2001), ‘Proving an environmental justice case: determining an appropriate Comparison Population’, Virginia Environmental Law Journal, 20, 365–430. Mansfield, Carol, George Van Houtven and Joel Huber (2001), ‘The efficiency of political mechanisms for siting nuisance facilities: are opponents more likely to participate than supporters?’, Journal of Real Estate Finance and Economics, 22, 141–61. Marbug, Hugh (1995), ‘Hazardous waste exportation: the global manifestation of environmental racism’, Vanderbilt Journal of Transnational Law, 28, 251–97. Mason, Stephen G. Jr. (1989), ‘Closure and rejection of waste facilities: what effect has public pressure’, Hazardous Material Control, July/August, 54–8. McCubbins, Mathew and Thomas Schwarz (1984), ‘Congressional oversight overlooked: police patrols versus fire alarms’, American Journal of Political Science, 28, 165–79. McDougall, Forbes and Jacques Fonteyne (1999), ‘Towards an integrated approach to waste management: the lessons learned from case studies of European waste management systems’, International Directory of Solid Waste Management. McLaren, Duncan, Olivier Cottray, Mary Taylor, Susan Pipes and Simon Bullock (1999), The Geographic Relation Between Household Income and Polluting Factories, Friends of the Earth. McLeod, H., I.H. Langford, A.P. Jones, R.J. Day, I. Lorenzoni and I.J. Bateman (2000), ‘The relationship between socio-economic indicators and air pollution in England and Wales: implications for environmental justice’, Regional Environmental Change, 1(2), 78–85. Millimet, Daniel and Daniel Slottje (1999), ‘The distribution of pollution in the United States: an environmental Gini approach’, Southern Methodist University working paper, Dallas, TX. Minehart, Deborah and Zvika Neeman (2002), ‘Effective siting of waste treatment facilities’, Journal of Environmental Economics and Management, 43, 303–24. Miranda, Marie Lynn, James N. Miller and Timothy L. Jacobs (2000), ‘Talking trash about landfills: using quantitative scoring schemes in landfill siting processes’, Journal of Policy Analysis and Management, 19(1), 3–22. Mitchell, Robert Cameron and Richard T. Carson (1986), ‘Property rights, protest, and the siting of hazardous waste facilities’, American Economic Review, 76(2), 285–90. Mohai, Paul (1995), ‘The demographics of dumping revisited: examining the impact of alternate methodologies in environmental justice research’, Virginia Environmental Law Journal, 14, 615–52.

Environmental equity and hazardous waste

153

— (1996), ‘Environmental justice of analytic justice? Reexamining historical hazardous waste landfill siting patterns in metropolitan Texas’, Social Science Quarterly, 77(3), 500–7. Morell, David and Christopher Magorian (1982), Siting Hazardous Waste Facilities: Local Opposition and the Myth of Preemption, Cambridge, MA: Ballinger. Morello-Frosh, Rachel, Manuel Pastor and James Sadd (2001), ‘Environmental justice and southern California’s “RISKSCAPE”: the distribution of air toxics exposures and health risks among diverse communities’, Urban Affairs Review, 36(4), 551–78. Munton, Don (1996), ‘Siting hazardous waste facilities, Japanese style’, in Don Munton (ed.), Hazardous Waste Siting and Democratic Choice, Washington, DC: Georgetown University Press. Murphy, Sean D. (1994), ‘Prospective liability regimes for the transboundary movement of hazardous wastes’, American Journal of International Law, 88(1), 24–75. National Governors’ Association (1989), Siting New Treatment and Disposal Facilities, Washington, DC: NGA. Nelson, Arthur C., John Genereux and Michelle Genereux (1992), ‘Price effects of landfills on house values’, Land Economics, 68(4), 359–64. Noll, Roger G. and James E. Krier (1990), ‘Some implications of cognitive psychology for risk regulation’, Journal of Legal Studies, 19, 747–79. Oberholzer-Gee, Felix and Howard Kunreuther (1999), ‘Social pressure in siting conflicts: a case study of siting a radioactive waste repository in Pennsylvania’, paper presented at the International Workshop on Challenges and Issues in Facility Siting, Institute of Economics, Academia Sinica, Taipei, Taiwan. O’Hare, Michael (1977), ‘ “Not on my block you don’t”: facility siting and the strategic importance of compensation’, Public Policy, 25, 407–58. O’Hare, Michael, Lawrence Bacow and Debra Sanderson (1983), Facility Siting and Public Opposition, New York: Van Nostrand Reinhold. O’Hare, Michael and Debra Sanderson (1993), ‘Facility siting and compensation: lessons from the Massachusetts experience’, Journal of Policy Analysis and Management, 12(2), 364–76. Ohkawara, Toru (1999), ‘Nuclear power plant siting issues in Japan: relationships between utilities and host communities’, paper presented at the International Workshop on Challenges and Issues in Facility Siting, Institute of Economics, Academia Sinica, Taipei, Taiwan. Olson, Mancur (1965), The Logic of Collective Action, Cambridge, MA: Harvard University Press. O’Neill, Kate (2000), Waste Trading Among Rich Nations: Building a New Theory of Environmental Regulation, Cambridge, MA: MIT Press. Openshaw, S. (1982), ‘The siting of nuclear power stations and public safety in the UK’, Regional Studies, 1693, 183–96. Organisation for Economic Co-operation and Development (1998), Trade Measures in the Basel Convention on the Control of Transboundary Movements of Hazardous Wastes and Their Disposal, Paris: OECD. O’Sullivan, Arthur (1993), ‘Voluntary auctions for noxious facilities: incentives to participate and the efficiency of siting decisions’, Journal of Environmental Economics and Management, 25, S12–S26. Page,William G. (1997), Contaminated Sites and Environmental Cleanup: International Approaches to Prevention, Remediation, and Reuse, San Diego, CA: Academic Press.

154

The international yearbook of environmental and resource economics

Park, Rozelia S. (1998), ‘An examination of international environmental racism through the lens of transboundary movement of hazardous wastes’, Indiana Journal of Global Legal Studies, 5, 659–709. Pastor, Manuel Jr., Jim Sadd and John Hipp (2001), ‘Which came first? Toxic facilities, minority move-in, and environmental justice’, Journal of Urban Affairs, 23(1), 1–21. Pearce, David (2002), ‘The distribution of environmental benefits and costs: conceptual framework and literature survey’, paper prepared for the OECD Environment Directorate Workshop on the Distribution of Benefits and Costs of Environmental Policies, March 2003. Pinzon, Lillian M. (1994), ‘Criminalization of the transboundary movement of hazardous waste and the effect on corporations’, DePaul Business Law Journal, 7, 173–221. Probst, Katherine N. and Thomas C. Beierle (1999), The Evolution of Hazardous Waste Programs: Lessons from Eight Countries, Washington, DC: Resources for the Future. Prokop, Gundula, Martin Schamann and Irene Edelgaard (2000), Management of Contaminated Sites in Western Europe, European Environment Agency report no. 13/1999. Pye, Steve, John Stedman, Martin Adams and Katie King (2001), Further Analysis of NO2 and PM10 Air Pollution and Social Deprivation, AEA Technology for the Department for Environment, Food, and Rural Affairs, The National Assembly for Wales, and Department of the Environment in Northern Ireland. Quah, E. and K.C. Tan (1998), ‘The siting problem of NIMBY facilities: cost-benefit analysis and auction mechanisms’, Environment and Planning C: Government and Policy, 16, 255–64. Rabe, Barry G. (1994), Beyond NIMBY: Hazardous Waste Siting in Canada and the United States, Washington, DC: Brookings Institution. Ragaini, R.C. (ed.) (1997), Risk Management Strategies Applied to Environmental Cleanup in Central and Eastern Europe, Singapore: World Scientific. Renn, Ortwin, Thomas Webler and Hans Kastenholz (1996), ‘Procedural and substantive fairness in landfill siting: a Swiss case study’, Risk: Health, Safety and Environment, 7, 145–68. Richards, Alice (1996), ‘Using co-management to build community support for waste facilities’, in Don Munton (ed.), Hazardous Waste Siting and Democratic Choice, Washington, DC: Georgetown University Press. Richman, Barak D. (2001), ‘Mandating negotiations to solve the NIMBY problem: a creative regulatory response’, UCLA Journal of Environmental Law and Policy, 20, 223–36. Riddel, Mary and W. Douglas Shaw (2003), ‘Option wealth and bequest values: the value of protecting future generations from health risks of nuclear waste storage’, Land Economics, 79(4), 537–48. Ringquist, Evan J. (1997), ‘Equity and the distribution of environmental risk: the case of TRI facilities’, Social Science Quarterly, 78(4), 811–29. Ristoratore, Mario (1987), ‘Siting toxic waste disposal facilities: best and worst cases in North America’, in Elliot J. Feldman and Michael A. Goldberg (eds), Land Rites and Wrongs, Boston, MA: Lincoln Institute of Land Policy. Rogers, George O. (1997), ‘Dynamic risk perception in two communities: risk events and changes in perceived risk’, Journal of Environmental Planning and Management, 40(1), 59–79.

Environmental equity and hazardous waste

155

Sadd, James L., Manuel Pastor, Jr., Thomas J. Boer and Lori D. Snyder (1999), ‘ “Every breath you take. . .”: the demographics of toxic air releases in Southern California’, Economic Development Quarterly, 13(2), 107–23. Santiago, Tessa Mayer (2002), ‘An ounce of preemption is worth a pound of cure: state preemption of local siting authority as a means for achieving environmental equity’, Virginia Environmental Law Journal, 21, 71–113. Schneider, Elke and Ortwin Renn (1999), ‘Fairness in public participation: German experiences with a structured public participation process in regional waste management planning’, paper presented at the International Workshop on Challenges and Issues in Facility Siting, Institute of Economics, Academia Sinica, Taipei, Taiwan. Seeliger, Robert (1996), ‘Siting hazardous waste incinerators in Germany: from political imposition to public involvement?’, in Don Munton (ed.), Hazardous Waste Siting and Democratic Choice, Washington, DC: Georgetown University Press. Shaw, Daigee (1996), ‘An economic framework for analysing facility siting policies in Taiwan and Japan’, in Paul R. Kleindorfer, Howard C. Kunreuther and David S. Hong (eds), Energy, Environment, and the Economy: Asian Perspectives, Cheltenham, UK and Northampton, MA: Edward Elgar. Sjoberg, Lennart, Mattias Viklun and Jana Truedsson (1999), ‘Attitudes and opposition in siting a high level nuclear waste repository’, paper presented at the International Workshop on Challenges and Issues in Facility Siting, Institute of Economics, Academia Sinica, Taipei, Taiwan. Smith, Hank Jenkins and Howard Kunreuther (1999), ‘Mitigation and compensation as policy tools for siting potentially hazardous facilities: evidence from field survey data’, paper presented at the International Workshop on Challenges and Issues in Facility Siting, Institute of Economics, Academia Sinica, Taipei, Taiwan. Snary, Christopher (1999), ‘Risk communication and the waste-to-energy incinerator environmental impact assessment process: a UK case study of public involvement’, Journal of Environmental Planning and Management, 45(2), 267–83. Sullivan, Arthur M. (1990), ‘Victim compensation revisited: efficiency versus equity in the siting of noxious facilities’, Journal of Public Economics, 41, 211–25. — (1992), ‘Siting noxious facilities: a siting lottery with victim compensation’, Journal of Urban Economics, 31, 360–74. Sundram, Muthu. S. (1997), ‘Basel Convention on Transboundary Movement of Hazardous Wastes: total ban amendment’, Pace International Law Review, 9, 1–56. Swallow, Stephen, James Opaluch and Thomas F. Weaver (1992), ‘Siting noxious facilities: an approach that integrates technical, economic, and political considerations’, Land Economics, 68(3), 283–302. Swallow, Stephen K., Thomas Weaver, James J. Opaluch and Thomas Michelman (1994), ‘Heterogeneous preferences and aggregation in environmental policy analysis: a landfill siting case’, American Journal of Agricultural Economics, 76, 431–43. Vari, Anna (1999), ‘Public perceptions about equity and fairness: siting low-level radioactive waste disposal facilities in the US and Hungary’, paper presented at the International Workshop on Challenges and Issues in Facility Siting, Institute of Economics, Academia Sinica, Taipei, Taiwan. Verchick, Robert (1999), ‘Critical space theory: keeping local geography in American and European environmental law’, Tulane Law Review, 73, 739–86.

156

The international yearbook of environmental and resource economics

Viscusi, Kip W. and James T. Hamilton (1999), ‘Are risk regulators rational? Evidence from hazardous waste cleanup decisions’, American Economic Review, 89(4), 1010–27. Waehrer, Keith (2003), ‘Hazardous facility siting when cost information is private: an application of multidimensional mechanism design’, Journal of Public Economic Theory, 5(4), 605–22. Walker, Gordon, John Mooney and Derek Pratts (2000), ‘The people and the hazard: the spatial context of major accident hazard management in Britain’, Applied Geography, 20,119–35. Walsh, Maureen (1992), ‘The global trade in hazardous wastes: domestic and international attempts to cope with a growing crisis in waste management’, Catholic University Law Review, 42, 103–40. Waugh, Theodore (2000), ‘Where do we go from here: legal controls and future strategies for addressing the transportation of hazardous wastes across international borders’, Fordham Environmental Law Journal, 11, 477–544. Wheeler, Michael (1994), ‘Negotiating NIMBYs: learning from the failure of the Massachusetts siting law’, Yale Journal on Regulation, 11, 241–91. Whitman, Christine Todd (2001), ‘EPA’s commitment to environmental justice’, 9 August 2001, Environmental Protection Agency internal memo, accessed at www.epa.gov. Yandle, Tracy and Dudley Burton (1996), ‘Reexamining environmental justice: a statistical analysis of historical hazardous waste landfill siting patterns in metropolitan Texas’, Social Science Quarterly, 77(3), 477–527. Yang, Tseming (2002), ‘Melding civil rights and environmentalism: finding environmental justice’s place in environmental regulation’, Harvard Environmental Law Review, 26, 1–32. Zimmerman, Rae (1993), ‘Social equity and environmental risk’, Risk Analysis, 13(6), 649–66.

4.

Strategies to conserve biodiversity Stephen Polasky

1

INTRODUCTION

For many biologists the loss of biodiversity is perhaps the single most important environmental issue at the beginning of the 21st century (see for example, Myers, 1979; Wilson, 1992, 2002; Levin, 1999; Pimm, 2001; McKee, 2003). A 1998 survey of 400 biological scientists found that ‘the rapid disappearance of species was ranked as one of the planet’s gravest environmental worries, surpassing pollution, global warming . . .’ and other threats (Warrick, 1998). Several studies estimate that current rates of extinction are several orders of magnitude above the average extinction rate through geologic time (Lawton and May, 1995; NRC, 1995; Pimm et al., 1995). Loss of tropical forests where a large portion of global biodiversity resides is of particular concern. What is known about present and past rates of extinction, and estimates of future extinction rates, however, is far from conclusive. Projections of how many species are likely to go extinct over the coming century often use species-area curve relationships, which predict the total number of species as a function of the total size of the area that species could inhabit (MacArthur and Wilson, 1967). Predictions of large-scale extinction come from combining the species-area curve relationship with projections of current and future habitat loss. Using this approach Wilson (1999) estimated that 27 000 species a year are likely to go extinct. Skeptics claim that biologists have vastly overestimated the loss of biodiversity (e.g., Lomborg, 2001). Fewer than 1000 species have been documented as having gone extinct since 1500 (IUCN, 2003). Estimating extinction rates is complicated by the fact that we don’t really know how many species exist and often have little record of their passing. There are approximately 1.7 million named species while most estimates of the total number of species on earth are in the range from 5 to 15 million (May, 1988, 1990; Wilson, 1988; Humphries et al., 1995). Like other large environmental changes (e.g., climate change), just how much damage human actions have already caused, or will cause if current trends continue, will not be known until the process is already well

157

158

The international yearbook of environmental and resource economics

along. Of course, by then the die may be cast and the loss of a large amount of biodiversity may be irreversible. Though species numbers and extinction rates grab much of the attention, biodiversity is a much broader concept than just the number of species and the loss of biodiversity is more than simply the number of extinctions. As Noss (1990) states: ‘Conservation biologists now recognize the biodiversity issue as involving more than just species diversity or endangered species. The issue is grounded in concern about biological impoverishment at multiple levels of organization.’ Biodiversity conservation involves everything from conserving genetic variability within a population, to different populations within a species, to assemblages of species within ecosystems, to ecosystem processes, and a diverse array of ecosystems. A number of arguments for the importance of conserving biodiversity have been advanced. Many species generate direct use benefits to humans for food, clothing, pharmaceuticals and other products. Non-consumptive use values, such as wildlife viewing and ecotourism, are also significant. At higher levels of organization, ecosystems perform a number of valuable services including nutrient cycling, waste recycling, water purification, and climate regulation (Daily, 1997). Option values for conserving species exist even for species with no known current use value (Fisher and Hanemann, 1986). For example, new pharmaceuticals or other products may be found through bioprospecting (Principle, 1989; Weitzman, 1992; Polasky et al., 1993; Polasky and Solow, 1995; Simpson et al., 1996; Rausser and Small, 2000). Further, species or other components of biodiversity may have existence value (Brookshire et al., 1983; Bishop and Welsh, 1992) or intrinsic value as distinct from instrumental/utilitarian value (Norton, 1987; Ehrenfeld, 1988; Rolston, 1994). Threats to biodiversity arise from a wide range of human actions. At the top of the list of threats is habitat loss and fragmentation (Wilson, 2002; Wilcove et al., 1998). Estimates of species loss derived from species-area relationships noted above are driven solely by habitat loss. Invasive species are another major threat, particularly on island ecosystems where native species are not well adapted to compete with newcomers. Humans also overharvest some species. There is evidence of large-scale declines in many fish species and changes in the composition of some marine ecosystems due to overfishing (e.g., Jackson et al., 2001; Myers and Worm, 2003). Other threats include pollution, particularly water pollution in aquatic ecosystems, and climate change (Parmesan and Yohe, 2003; Root et al., 2003; Pounds and Puschendorf, 2004; Thomas et al., 2004). The effects of climate change are particularly important in conjunction with habitat loss and fragmentation that may prevent the movement of species to new potentially suitable habitat as climate changes.

Strategies to conserve biodiversity

159

Conserving biodiversity will require reducing the threats from human activity. Restrictions on land use, introductions of non-native species, harvesting, pollution and emissions of greenhouse gases, all may impose substantial costs on at least some portion of society. As in other areas of environmental economics, an evaluation of conservation strategies requires assessments of difficult tradeoffs about whether the benefits of conservation exceed the costs. For biodiversity conservation, obtaining quantitative estimates of the value of conservation is particularly problematic. What is the option value for preserving genetic material? What is the existence value of preserving an endangered species? How valuable are various ecosystems services? Though some useful evidence exists, we are far from definitive answers on these and scores of other questions related to the value of biodiversity conservation. Rather than attempting to compare the costs and benefits of conserving biodiversity, this chapter focuses on the analysis of the efficiency of conservation strategies. What conservation strategy will obtain a conservation objective, such as maximizing the number of species conserved, at least cost? Or, equivalently, what conservation strategy will maximize a conservation objective given limited resources? Such analysis is important for comparing the cost-effectiveness of alternative strategies while largely sidestepping the difficult issue of the valuation of biodiversity. There is a direct analogy here with environmental policy that affects human health. A full cost-benefit analysis requires estimates of the value of human life, which is a controversial and unsettled area (much like the valuation of biodiversity). Even without an answer to the value of human life though, cost-effectiveness analysis shows where additional expenditures can have the greatest impact in terms of lives saved per dollar. Such analysis can be quite useful for making intelligent policy decisions. Of course, at some point, the difficult question of whether further expenditures are justified requires judgments about acceptable tradeoffs between expenditures and lives saved (or biodiversity conserved). In the next section, two general approaches to cost-effectiveness analysis are described, one by Weitzman (1998) that focuses on a species-by-species approach, and one by Polasky and Solow (1999) that allows for analysis of a habitat approach. In discussing the details of cost-effective conservation strategy, it is useful to break the analysis into two component parts. The first part involves an analysis of efficient conservation plans assuming a planner with access to all available information and the ability to fully implement plans. Literature on efficient conservation planning is reviewed in Section 3. The second part involves an analysis of decentralized decisionmaking, asymmetric information, and other implementation issues, which may preclude attaining an efficient outcome. Implementation and policy issues are

160

The international yearbook of environmental and resource economics

discussed in Section 4. The analysis in Sections 3 and 4 will focus on habitat conservation strategies because habitat loss is the dominant threat to biodiversity conservation and because issues related to habitat loss involves novel analyses that are not commonplace in economics. On the other hand, analysis of harvesting and pollution issues is very well developed within resource and environmental economics with little new to add here. Though interest has increased dramatically over the past couple of years, there is not yet an extensive literature within economics on invasive species (see Polasky et al., 2005a for a recent review). For the sake of brevity and focus, only terrestrial conservation will be discussed in this chapter. The rapidly growing literature on conservation in marine systems and marine reserves will not be covered here (see Botsford et al., 2001; Sanchirico and Wilen, 2001, 2002; Polasky et al., 2005a; and various articles in a special issue of Ecological Applications in February 2003 for discussion of marine conservation issues).

2

GENERAL FRAMEWORKS FOR COST-EFFECTIVE CONSERVATION

‘The preservation of biodiversity is plagued by the absence of a workable cost-effectiveness framework . . .’ (Weitzman, 1998, p. 1279) Public and private groups with a mandate to conserve biodiversity have limited resources relative to what is necessary to accomplish the task. These groups must set priorities and make difficult choices. In essence, conservation groups face the classic economic problem of allocating scarce resources. For this reason economic analysis should play a more active role in biodiversity conservation. This section reviews two cost-effectiveness frameworks with which to analyse biodiversity conservation plans: Weitzman’s ‘Noah’s Ark Problem,’ and Polasky and Solow’s ‘Conservation with Scarce Resources’. A.

Weitzman: ‘The Noah’s Ark Problem’

In ‘The Noah’s Ark Problem,’ Weitzman sets out to define a cost-effective approach to conservation that is both intuitive and rigorously derived from basic principles. Weitzman models a resource constrained Noah who doesn’t have an ark big enough to fit all species. The conservation problem for the space constrained Noah is to choose species survival probabilities to maximize expected utility from species conservation subject to a budget constraint. Weitzman assumes that the cost of increasing survival probabilities is a linear function. He further assumes that utility consists of the

Strategies to conserve biodiversity

161

direct value from the existence of the species, and the ‘distinctiveness’ value, which measures the difference between a species and its closest genetic neighbor. The distinctiveness value captures the degree to which each species adds unique genetic information to the set of surviving species. Given this setup, Weitzman proves that the optimal conservation policy is an ‘extreme policy’ in which each species is either conserved to the maximum degree possible or not conserved at all (with the possible exception of a single fractionally conserved species). This result occurs because the objective function is convex and the constraint is linear in probabilities. The expected direct value from existence is simply the sum of direct utility value for each species times its survival probability, which is linear in probabilities. The distinctiveness value is convex in probabilities. The intuition for this result can be most easily seen from an example with two related species. Suppose the distinctiveness value of conserving only one of two species is the same regardless of which of the species is conserved, V
V(1)
V(2). Let the distinctiveness value of conserving both species be V(1,2). Because of the relatedness of the species, adding the second species does not increase the distinctiveness value by as much as adding the first species so that: V(1,2) V(1)V(2)
2V. Now consider a problem in which the survival probabilities cannot exceed one: P1 P2 1. Assuming the constraint is binding, with P1
p and P2
1p, the expected distinctiveness payoff is equal to: P1P2V(1,2)P1(1P2)V(1)(1P1)P2V(2)
p(1p)V(1,2)p2V(1p)2V
p(1p)[V(1,2)2V]V Since V(1,2) 2V0, this expression is maximized by setting p
0 or p
1. In words, it is better to conserve one species for sure rather than having a chance of conserving both species with an equal chance of conserving no species. Compared to conserving a single species, the loss from having no species exceeds the gain from adding a second species. Under Weitzman’s approach, which species should be conserved is determined by a simple ranking criterion: Ri
(Di  Ui )

∆Pi Ci

where Ri is the ranking criterion score for species i, Di is the distinctiveness value added by species i, Ui is the direct existence value of species i, Pi is the change in the probability of survival of species i, and Ci is the cost of increasing the survival probability of species i. The conservation budget should be allocated to conserve the highest ranking species.

162

The international yearbook of environmental and resource economics

Weitzman’s conservation allocation rule is exceedingly simple and intuitive. If the real conservation problem were the same as the constrained ‘Noah’s Ark Problem’ there would be little left to discuss, with the exception of how large a budget should be given to Noah. There are at least two reasons that real conservation problems are not the same as the ‘Noah’s Ark Problem’. First and foremost, conserving species typically requires conserving the habitat of the species. Many species live in the same habitat so that one cannot isolate the effect of a strategy on the survival probability of a single species. In economic terms, habitats exhibit ‘joint production’ providing increased survival probabilities for a number of species simultaneously. Second, the costs of increasing survival probabilities are unlikely to be linear. It is reasonable to expect that there may be some critical range size for which the marginal change in species survival probabilities is high per unit of additional area, and beyond which the marginal change falls. B. Polasky and Solow: ‘Conserving Biological Diversity with Scarce Resources’ Polasky and Solow describe a simple general approach to conserving biodiversity under a budget constraint. The conservation problem is to choose affordable conservation strategy s from the set of potential conservation strategies S to maximize expected biodiversity conserved: Max


D(x)P (x)

x
X

s

s.t. C(s)  B where x is a particular outcome from the set of possible outcomes, X, D(x) is the biodiversity measure of outcome x, Ps(x) is the probability of outcome x under conservation strategy s, C(s) is the cost of strategy s, and B is the size of the conservation budget. Formally, this problem is virtually the same as maximizing expected utility under a budget constraint, with D(x) playing the role of the utility function. What makes this problem different from a standard economic expected utility maximization problem comes from defining the measure of biodiversity, D(x), and translating a conservation strategy into a probability distribution over possible biological outcomes, Ps(x). There are a range of potential biodiversity measures that could be used, including species richness (the number of species), measures of species diversity that includes a premium for taxonomic distinctiveness (Vane-Wright et al., 1991; Faith, 1992, 1994; Weitzman, 1992; Solow et al., 1993; Solow and Polasky, 1994), measures of relative abundances of species (see Magurran, 1988, 2004), or

Strategies to conserve biodiversity

163

measures of ecosystem properties such as productivity (Naeem et al., 1994, 1995; Tilman et al., 1996; Hector et al., 1999), stability (Tilman and Downing, 1994; McGrady-Steed et al., 1997; Naaem and Li, 1997), resilience (Hollings, 1973; Perrings et al., 1995; Carpenter et al., 1999; Walker et al., 1999; Scheffer et al., 2001) or the value of ecosystem services (Costanza et al., 1997; Daily, 1997; Daily et al., 2000). There are also different approaches for determining how conservation strategies impact on the probability of various potential outcomes occurring depending in part on what objective is used for D(x). Understanding the likelihood of outcome x occurring under strategy s, Ps(x), requires understanding biological cause-and-effect relationships in relation to management actions. In other words, it requires integrating biological knowledge into an economic decisionmaking framework. There is a long tradition of this kind of integration in bioeconomic models of optimal harvesting (e.g., Clark, 1990) but this integration is in its infancy for modeling conservation of habitat or invasive species controls. Biologists are not often used to thinking in terms of marginal analysis useful in analysing policy alternatives. Economists often ignore or simplify biological relationships in economic models. Integration of biological relationships with decisionmaking approaches from economics will ultimately make such approaches of greater value. Integrated bioeconomic models of conservation strategy may turn out to be a rich avenue of research, but such models are unlikely to give neat analytic solutions as in Weitzman (1998). Even when a measure of species diversity is used, as in Weitzman (1998), the conservation strategy is rarely specific to a single species. As stressed in the introduction, the key threats to species conservation are loss of habitat and invasive species. Conserving habitat, or protecting an ecosystem from invasion, typically provides protection for multiple species within an ecosystem (joint production). Joint production along with spatial and dynamic relationships make conservation problems complex. While the Noah’s Ark rule works for a space constrained Noah, modern Noahs that seek to conserve biodiversity through habitat protection or invasive species control have a tougher challenge in determining cost-effective conservation strategies. This challenge is taken up in the next section.

3

COST-EFFECTIVE CONSERVATION STRATEGIES

This section reviews work from conservation biology and economics on strategies to conserve biodiversity via habitat preservation and control of invasive species. As compared to the general frameworks of Section 2, this

164

The international yearbook of environmental and resource economics

section is more specific about the details of conservation strategies. Of the two issues, habitat conservation and invasive species control, far more literature to date has been directed toward habitat conservation. The bulk of this section will review habitat conservation strategies. The small literature on controlling invasive species, which has been starting to grow more rapidly within the past couple of years, will be covered at the end of the section. A.

Habitat Conservation: the Reserve Site Selection Problem

The widespread conversion of land to human dominated uses in many areas has left small fragmented islands of more natural habitat capable of supporting a wide range of biodiversity. Vast areas of temperate forests and grasslands have been converted to farmlands, pasture and urban development. While some temperate areas have shown slower rates of conversion in recent years or even regrowth of forests, habitat destruction in tropical developing countries has continued unabated. In response to the loss of habitat, conservation biologists have proposed establishing a system of formal protected areas to preserve key remnants of remaining natural habitat. It is estimated that protected areas now cover 11.5 per cent of land globally (Chape et al., 2003). Yet, recent studies show that current protected areas are inadequate to conserve all of biodiversity (Rodrigues et al., 2004). With an expanding human population and unmet human needs, particularly in tropical developing countries, there will be limits on how much land will be devoted to meet conservation objectives. What lands should be set aside as nature reserves to conserve biodiversity given the other pressing demands on land use is a classic economic problem. In fact, this is an excellent problem with which to explain basic economic concepts such as opportunity cost or the optimal allocating of scarce resources under a budget constraint to biologists who may be unfamiliar (or skeptical) about the relevance of economic tools. There is a large literature, written mostly by conservation biologists though economists have become more active in recent years, on setting priorities for habitat conservation. A simple yet instructive approach to systematic conservation planning is the ‘reserve site selection’ problem. In the standard formulation of this problem, a conservation planner chooses sites to include in a conservation reserve network to represent the maximal number of species within the reserve network, subject to a constraint on the total number of sites that can be included: m

Max


yi i
1

(1)

Strategies to conserve biodiversity

165

m

s.t.

xj  yi



(2)

j Ni n


xj  k,

(3)

j
1

where yi is an indicator variable for species i survival (yi
1 if species i survives and yi
0 as species i goes extinct) for all i
I, where I is the set of all species; xj as an indicator variable for whether site j is selected (xj
1 if site j is selected in the reserve network and xj
0 if site j is not selected) for all j
J, where J is the set of all potential reserve sites; Ni is the set of sites in which species i occurs; and k is the number of sites that may be included in the reserve network. This is an integer programming problem that is called the ‘maximal coverage problem’ in operations research (Church and Revelle, 1974; Underhill, 1994; Camm et al., 1996). Even reasonably largesized problems with hundreds of sites can be solved using branch-andbound algorithms (Church et al., 1996; Csuti et al., 1997; Pressey et al., 1997; Ando et al., 1998). In some respects, the reserve site selection problem resembles the Noah’s Ark problem, with the collection of reserves sites constituting the conservation network playing the role of the ark. There is, however, one important difference between the two problems. In the Noah’s Ark problem each species is chosen individually. In the reserve site selection problem, sites that contain numerous species are chosen. Because of the potentially complicated pattern of overlap in species there isn’t a simple method for finding an optimal solution. In choosing sites, complementary sites that add different species than are represented in other selected sites are more valuable than sites with higher species richness but more overlap (Pressey et al., 1993). By solving the reserve site selection problem for different levels of the constraint on the number of sites that can be included in the reserve network one can trace out an accumulation curve showing the number of species that can be represented for various sized conservation networks. Csuti et al. (1997) solved for an accumulation curve for terrestrial vertebrates in Oregon. The accumulation curve is initially quite steep as numerous species co-occur in the same biologically rich sites, but declines quickly after the first few sites. Over 90 per cent of all terrestrial vertebrate species in Oregon were included in a reserve network of five sites and over 95 per cent were included in ten sites. In total 23 sites were needed to include all species, with the last sites adding only one or two species as a time. The reserve site selection problem can be translated into a budget constrained rather than site constrained problem by incorporating the cost of including site j in the reserve network. Doing so, the constraint in

166

The international yearbook of environmental and resource economics

equation (3) can be rewritten as:  cj xj B, where cj is the cost of including site j in the reserve network and B is the total conservation budget (Ando et al., 1998; Polasky et al., 2001). In this case, the accumulation curve becomes a cost curve showing the minimum total cost necessary to represent a given number of species in the network. Ando et al. (1998) used data on average value of agricultural land and endangered species by county to solve for both a budget constrained and a site constrained reserve network. By choosing sites that have a high species represented per dollar ratio rather than the biologically rich sites, the budget constrained solution resulted in the same number of endangered species represented in selected sites at onethird to one-half the cost of the site constrained approach. Under the budget constrained approach, sites with lower land costs in the interior mountain states are included more often whereas the site constrained solution includes more sites with high land costs such as coastal Southern California. Both Ando et al. (1998) and Polasky et al. (2001) found low costs for conserving the majority of species with steeply rising costs as solutions approach complete representation. At a global scale, Balmford et al. (2003) used cost-effectiveness analysis to find that large efficiency gains could be made by redistributing conservation efforts toward tropical developing countries where the costs of protection are low and the benefits of protection are high. They found a ‘gross mismatch’ between beneficial conservation projects that are concentrated in tropical developing countries and current conservation spending that is heavily skewed toward temperate developed countries. They find that cost differences among sites range over several orders of magnitude, which is greater than the variance of biological benefits. Balmford et al. (2003) conclude that failure to incorporate costs into conservation planning will result in missed opportunities for greater conservation efficiency. This study highlights important differences in costs and benefits of conservation across countries. However, the study was limited in the type of cost data it used, focusing on management costs while not incorporating land purchase or opportunity costs that are likely to make up a large fraction of conservation costs. More work analysing cost-effective conservation strategies at an international scale would be highly beneficial. In almost all applications, information about costs or benefits of conservation is incomplete. Several studies have analysed reserve site selection when information about species ranges is incomplete. Polasky et al. (2000) use heuristic methods to maximize expected species representation in a reserve network given only probabilistic information about species ranges. Camm et al. (2002) solved the expected species representation problem using linear approximations to achieve a solution arbitrarily close to the optimal solution using linear programming techniques. Overall, solutions

Strategies to conserve biodiversity

167

to expected species representation problem tend to be similar to solutions assuming presence/absence data. One difference between the approaches, however, is that two nearby sites can be included under the probabilistic approach, where doing so increases the probability of representation for some set of species, but not under the presence/absence approach. Incorporating uncertainty opens several important dimensions for research including species persistence probabilities, threats of habitat conversion and stochastic events, some of which are discussed in the next subsection. B. Modeling Land Use and Species Persistence: Beyond Reserves and Beyond Representation Ultimately what is of importance is the long-term survival of biodiversity. Just because species are currently represented inside a reserve network does not necessarily guarantee their persistence over time. Reserve sites may be too small to sustain a viable population. There may be drought, disease or other stochastic events that wipe out local populations. Conditions in reserves may become less hospitable to species over time due to natural succession or climate change. Similarly, species outside of the reserve system will not necessarily perish. Many species can tolerate some level of human activity and habitat disturbance. Some species thrive in human dominated landscapes, though these species are not necessarily species that humans prefer (e.g., rats and pigeons). Further, not all land outside of reserves will be heavily impacted by human activities (at least in the near term). For these reasons, conservation analysis must progress beyond mere representation (as in the reserve site selection approach) and address likely persistence. Another fact pushing analysis beyond consideration of reserves is the fact that the vast majority of land lies outside of protected areas. For conservation of biodiversity to be successful roughly 90 per cent of land that lies outside of formal protected areas must contribute to conservation goals. Many analysts have pointed out the need to move beyond reserves and analyse the likely conservation outcomes as a function of what is happening on the entire landscape (e.g., Franklin, 1993; Miller, 1996; Reid, 1996; Wear et al., 1996; Daily et al., 2001; Polasky et al., 2005b; Rosenzweig, 2003). Miller (1996, p. 425) summed up a landscape approach as follows: ‘biodiversity will be retained to the extent that whole regions are managed cooperatively among protected areas, farmers, foresters and other neighboring land users.’ The most well-developed approach to modeling species persistence is population viability analysis (Soule, 1987; Boyce, 1992; Beissinger and McCullough, 2002). Population viability analysis incorporates demographic, genetic and environmental stochasticity to predict the likelihood

168

The international yearbook of environmental and resource economics

of survival of a species with a given static population. This analysis can be combined with a landscape analysis that provides the distribution of habitat. Several economic studies have combined spatially explicit biological models with human land use decisions to find tradeoffs between species persistence and the value of economic production activities. In particular, a number of papers have traced out production possibility frontiers for the value of timber production and species persistence for a single or small number of forest dwelling species (e.g., Montgomery et al., 1994; Haight, 1995; Hof and Bevers, 1998; Marshall et al., 2000; Rohweder et al., 2000; Calkins et al., 2002; Nalle et al., 2004). Montgomery et al. (1994) combined a population biology model for the spotted owl with economic models of the value of timber harvest to estimate a marginal cost curve for increasing owl survival probabilities. The results showed that marginal costs of increasing owl survival were low for survival probabilities below 90 per cent but increased sharply for survival probabilities above 90 per cent. Land use decisions simultaneously affect a large set of species so conservation planning would ideally move beyond a species-by-species approach. Several papers have expanded the landscape level analysis to include a large number of species (Montgomery et al., 1999; Lichtenstein and Montgomery, 2002; Polasky et al., 2005b). Doing so necessitates a change in approach because detail intensive population biology modeling becomes unwieldy with a large set of species. Montgomery et al. (1999) used the percentage of habitat conserved under various land use decisions in Monroe County, Pennsylvania to construct probabilities of survival for 147 bird species that currently inhabit the county. Polasky et al. (2005b) used a spatially explicit model of the consequences of alternative land use decisions on the persistence of various species and the value of agricultural and forestry production, based on conditions in the Willamette Basin in Oregon. In their analysis of efficient land use patterns, they found that a large fraction of species conservation could be obtained at low cost. For example, they found it was possible to obtain 96 per cent of the maximum species persistence value while also obtaining 93 per cent of the maximum commodity production value. Trying to increase either objective from this point, however, resulted in dramatic reductions in the value of the other objective. In comparison, running a reserve site selection analysis that assumes no biological value for lands outside of reserves shows both lower scores and more continuous tradeoffs between biological and economic objectives. Virtually all of the landscape analyses done to date have been static. This is perhaps not surprising since these analyses require integrating spatially explicit biological and economics models. Adding dynamics on top of this is a daunting task. However, there are important dynamic elements to

Strategies to conserve biodiversity

169

conservation that cannot be ignored. Costello and Polasky (2004) analyse a dynamic reserve site selection problem in which each site currently unprotected has a probability of being developed during that period. A conservation agency would like to protect as many sites as possible as early as possible but faces constraints on when funds are available. The model is solved using stochastic dynamic programming, which limits the size of problems that can be handled. However, a heuristic solution that involves selecting sites that combine high biological value added per unit cost plus face a high development threat performed quite well for a set of small-scale problems (whether this remains true for large-scale problems is unclear). This result provides some support for the ‘hotspots strategy’ that gives high priority to conserving places of high biodiversity or high endemism and facing immanent threats (e.g., Myers, 1988; Mittermeier et al., 1998; Myers et al., 2000). The hotspot approach was recently criticized by Kareiva and Marvier (2003), though some of the criticisms, such as ignoring cost and issues of complementarity among sites, have been addressed in the literature discussed above. In general, unforeseen changes originating either from the economic side (development activity or changes in relative prices), from the biological side (biological invasions, disease outbreaks) or from changes in the physical environment (hydrology, climate change), mean that once-and-for-all conservation decisionmaking is inappropriate. Conservation decisionmaking should adapt to changing conditions and be forward looking, trying to anticipate what may lie ahead. This latter point is especially important with irreversible outcomes, such as extinction. The path breaking work by Arrow and Fisher (1974) highlights the value of avoiding irreversible outcomes prior to the resolution of uncertainty, finding that there is an option value to maintaining flexibility. Recent work by economists and ecologists has also emphasized the importance of preserving the resilience of ecosystems and of avoiding potentially damaging and difficult-to-reverse shifts between alternative ecosystem states (e.g., Perrings et al., 1995; Carpenter et al., 1999; Walker et al., 1999; Dasgupta and Mäler, 2003). Solving for optimal solutions in spatially explicit integrated biological and economic models, with either dynamics or uncertainty, can be quite difficult. While optimal solutions are a useful benchmark they are not a prerequisite for analysis to be useful to decisionmakers. In fact, given how suboptimal most land use and conservation policy is at present, most well-grounded analysis can only help to improve the situation. One quite useful approach is to blend the type of analysis discussed in this section with predictions of likely land use that will result under various policy scenarios. The next section takes up issues concerning policy and implementation issues related to conserving biodiversity.

170

4

The international yearbook of environmental and resource economics

CONSERVATION POLICY AND IMPLEMENTATION ISSUES

The previous section focused on the question of optimal conservation plans. This section focuses on the question of how conservation plans might be implemented. The fundamental problem raised by biodiversity conservation is the mismatch between the scale at which benefits accrue and the scale at which land use decisions are typically made. Conservation often generates widespread benefits including global public goods such as species existence value or carbon sequestration. But most decisions affecting habitat are local land use decisions made by individual landowners, small communities or local governments. Because local decisionmakers do not receive the full benefits of conservation they will not typically have adequate incentives to conserve. Like most issues in environmental economics the question is how to internalize the externalities (in this case the positive externalities from widespread conservation benefits). To some extent, there may be ways that local decisionmakers can capture at least some of the benefits of conservation. The first part of this section addresses the degree to which biodiversity conservation can pay for itself. To the extent that local decisionmakers receive adequate returns from conserving biodiversity, there is no need for explicit conservation policy at a higher scale of governance (national or international). However, to the degree that conservation benefits cannot be internalized directly, there is a role for explicit policies to promote conservation. The second part of the section will look at conservation policies at the national level, in particular the US Endangered Species Act. The final part of this section will look at conservation policies at the international level, in particular the Convention on International Trade in Endangered Species and the Convention on Biological Diversity. A.

Marketing Biodiversity: Can Biodiversity Conservation Pay for Itself?

Though some of the benefits from conserving biodiversity generate global public goods, other conservation benefits generate private goods that may be sold in markets potentially generating returns to local landowners. Given the problems that arise with conservation policy, to be discussed below, making ‘conservation pay’ thereby generating direct incentives to landowners and other local decisionmakers to conserve is an attractive option (Heal, 2000; Daily and Ellison, 2002; Pagiola et al., 2002). In fact, some conservation biologists and economists have made the point that conservation will likely occur only to the extent that it is used and generates returns, i.e., ‘use it or lose it’ (Jantzen, 1992; van Kooten and Bulte, 2000).

Strategies to conserve biodiversity

171

Perhaps the most straightforward way in which species conservation can generate returns to landowners and local communities is through ecotourism. Nature based tourism is one of the fastest growing segments of the overall tourism industry, which generated estimated revenues of $463 billion in 2001 (World Tourism Organization, 2002). Areas with unique resources or charismatic megafauna (e.g. Krueger National Park, Serengeti National Park, Yellowstone National Park) have the potential to generate large amounts of revenue. Examples of successful development of nature based tourism include Costa Rica, where about 1 million tourists spent approximately $1 billion in 2000 (Daily and Ellison, 2002, p. 178), and South Africa, which generated over $2 billion in nature based tourism revenue in 2000 (World Tourism Organization, 2002). Other examples of ecotourism are analysed and summarized in Maille and Mendelsohn (1993), Alyward et al. (1996), Wunder (2000) and Lindberg (2001). Ecotourism, however, raises its own set of problems. As ecotourism becomes more successful it brings more tourists and more economic activity to the area, thereby increasing the danger of damaging the very ecosystems that provide the attraction (Liu et al., 2001). A second major issue with ecotourism is that of who captures rents that might be generated from ecotourism. Large revenue figures, as quoted in the previous paragraph, do not say anything about the size of rents created from ecotourism because costs are not included. Presuming there are positive rents once costs are subtracted from revenues, what share of rents goes to international companies, national governments, and local communities? An important issue, in terms of both equity and providing the right set of incentives, is to insure that local communities receive an adequate return from ecotourism. Receiving an adequate return is especially important in cases where wildlife damages crops or livestock. Much of the push for community based conservation was the recognition that local communities will not have enough incentive to conserve unless they are given access to ecotourism revenues or other benefits generated by conservation (Barbier, 1992; Wells and Brandon, 1992; Western and Wright, 1994). Giving local communities more control over resources and a larger slice of the revenue stream, however, requires that national governments and companies cede some control over resources and revenues to local communities. Issues of power and control over resources are major stumbling blocks that often prevent adequate sharing with local communities. Bioprospecting is another potential way in which biodiversity conservation may generate market rewards that could provide incentives for conservation. Bioprospecting is the systematic search for useful genetic material from plant or animal species for development of valuable pharmaceuticals or other products. An agreement in 1991 between Merck

172

The international yearbook of environmental and resource economics

and Costa Rica’s Instituto Nacional de Biodiversidad (INBio) provided $1 million to INBio. Despite the initial excitement, bioprospecting has failed to provide much if any spur to conservation. There have been no other major deals signed since the Merck–INBio deal. Questions have been raised as well about how much difference the deal has made in slowing deforestation or other forms of habitat loss in Costa Rica. During the 1990s, Costa Rica had one of the highest rates of deforestation in Latin America with average annual losses of forest area exceeding 3 per cent (World Resources Institute et al., 2000). Whether there are likely to be significant rents generated by bioprospecting, and how to share these rents are open questions. Simpson et al. (1996) showed that economic returns from bioprospecting are likely to be quite small, far too small to generate adequate incentives for conservation (though a different view is given in Rausser and Small, 2000). If there are significant rents from bioprospecting, it is unclear how those rents should be allocated between local communities and countries in which the biological resources are located and companies or countries supplying intellectual discoveries that turn these biological resources into valuable products. The rent allocation issue has been the subject of a heated debate between developing countries and developed countries (particularly the US). There is a strongly held feeling by some in developing biodiversityrich countries that the application of the intellectual property rights under the aegis of the World Trade Organization’s Agreement on Trade Related Aspects of Intellectual Property Rights (TRIPs) would result in global corporations profiting from biological resources and traditional knowledge without giving local communities their fair share. This has led some to refer to bioprospecting as ‘biopiracy’. On the other side, the US failed to ratify the Convention on Biological Diversity largely because it felt there was not adequate protection of intellectual property rights the way the Convention was drafted. In economic terms, if adequate returns are not given to the host community or country supplying the biological resources, there will be insufficient incentives to conserve habitat on the ground. On the other hand, if adequate returns are not given to the company supplying the innovation, there will be insufficient incentives to develop new products. On the latter issue, several analysts have pointed out the often large gap between the private returns to a firm commercializing a new product and the social returns from such an innovation (Mendelsohn and Balick, 1995; Koo and Wright, 1999). Other than vertically integrating the suppliers of biological resources with the suppliers of intellectual resources, there does not appear to be an easy solution to the incentives problem nor to the bitter political dispute over what is a fair sharing of the rents.

Strategies to conserve biodiversity

B.

173

National Conservation Policy

In some fortuitous circumstances, conservation may generate sufficient returns to private parties to make conservation pay. Biodiversity conservation, however, creates many goods and services that generate benefits that accrue far beyond local decisionmakers, including global public goods. For this reason, there is a clear rationale for conservation policy by governments as well as actions by non-governmental conservation organizations. At the national level, there are a number of policies related to conservation. One of the most important, and controversial, is the US Endangered Species Act (ESA). The ESA has proved to be a powerful tool for conservationists and a magnet for criticism for groups promoting private property rights and deregulation (Brown and Shogren, 1998). The two central provisions of the ESA are contained in Section 7, which prohibits federal government actions that cause ‘jeopardy’ (i.e., risk of extinction) to listed species, and Section 9, which prohibits public and private parties from ‘taking’ listed species. ‘Taking’ includes causing harm to species, where harm includes adverse habitat modification from otherwise legal land uses. The ESA has caused changes in timber harvesting plans in the Pacific Northwest to protect the spotted owl and the Southeast to protect the red cockaded woodpecker, in residential and commercial development plans in Southern California and elsewhere. One criticism of the ESA among economists is that it fails to create positive incentives to conserve, and worse, may create perverse incentives that actually result in less protection for listed species (Innes et al., 1998). Because a landowner may face land use restrictions and is not guaranteed compensation for lost value, there is an incentive to prevent listed species from becoming established or preventing their discovery (Mann and Plummer, 1995, Polasky and Doremus, 1998). There also may be incentives to race to develop in order to beat the imposition of the ESA (Innes, 1997). A second criticism of the ESA is that it does not weigh costs and benefits of actions but is an absolute prohibition against harming listed species. Several initiatives were begun in the Clinton Administration in an attempt to make the ESA more flexible and less onerous on private landowners. Landowners subject to ESA prohibitions were encouraged to file Habitat Conservation Plans (HCP). Landowners with an approved HCP would be guaranteed ‘no surprises’, so that costs of further prohibitions to protect the listed species covered by the HCP would be the government’s responsibility not the landowner’s, and ‘safe harbors’, so that a landowner that improved habitat for a listed species would not be penalized. More fundamental reforms in conservation policy call for more market based regulatory approaches and more voluntary (less proscriptive)

174

The international yearbook of environmental and resource economics

approaches. One such proposal is to institute a system of transferable development rights (Field and Conrad, 1975; Mills, 1980; Panayotou, 1994; Merrifield, 1996; Renard, 1999; Thorsnes and Simons, 1999; Weber and Adamowicz, 2002). Such transferable development rights (TDRs) would operate much like a marketable pollution permits or individual transferable quotas (ITQs) in fisheries management. As with marketable pollution permits and ITQs, TDRs face questions about the efficient number of permits to issue as well as questions about how to allocate the permits. A problem that is arguably more severe with TDRs than with marketable pollution permits or ITQs is figuring out what constitutes an equivalent trade. For conservation purposes, the spatial pattern, extent and quality of habitat matter. Allowing trades on an equal area basis is not likely to be a sensible policy. Instead, trading ratios should be established based on the relative contribution of particular land parcels to conservation goals. However, the contribution of particular parcels is not constant but in general depends upon the overall pattern of land use in a region. How to design a reasonably effective yet workable TDR scheme given land heterogeneity and interdependent values is an open question. Another approach to conservation is to use direct payments for conservation. Proponents of this approach argue that the direct approach is the most efficient means of promoting conservation (Ferraro and Kiss, 2002; Ferraro and Simpson, 2002). Ferraro and coauthors argue that by directly targeting and paying for conservation, this method can deliver more in terms of conservation per dollar spent than indirect schemes that like integrated conservation and development projects. Direct payment schemes also have the advantage of being voluntary rather than coercive. Costa Rica has instituted a system of payments for ecosystem services: mitigation of greenhouse gas emissions, watershed protection, biodiversity conservation, and scenic beauty. Many developed countries have some form of ‘green payments’ in their agricultural policies that pay farmers who adopt environmentally friendly management practices or land uses (OECD, 2001). Smith and Shogren (2002) outline a voluntary payments scheme for biodiversity conservation similar to the US Conservation Reserve Program, in which farmers receive payments to retire land from active production. Many local communities in the US have recently passed bond issues to raise money for the purchase of open space. In 2000, 174 out of 210 open space bond issues were passed, raising $7.5 billion for the purchase of open space (The Trust for Public Lands). A number of non-governmental organizations, such as the Nature Conservancy, raise substantial amounts of money used to purchase important habitat for conservation. Many policies affect conservation outcomes besides those that are expressly about conservation. Any policy that affects land use decisions,

Strategies to conserve biodiversity

175

including agricultural and forestry policy, the placement of infrastructure and local zoning ordinances potentially has impacts upon conservation. Often these policies promote activity that is harmful to conservation. Coordinating policies and removing subsidies for activities harmful to the conservation of biodiversity remain high on the wish list of reforms by conservationists in most countries. C.

International Conservation Policy

Some of the benefits of biodiversity conservation may spill beyond national boundaries, and indeed may be global in scale, such as existence value for species. Adequately dealing with such benefits requires international policy and international institutions. At the international level, there are two major conventions related to biodiversity conservation, the Convention on International Trade in Endangered Species (CITES) and the Convention on Biological Diversity (CBD), as well as other programs and policies. CITES has provisions that allow prohibitions on international trade in endangered species (for species listed under its Appendix I) and regulation of international trade (for species listed under its Appendix II). Because it is focused on trade, CITES deals primarily with high-profile species that are harvested either for food, medicines, pets or trophies. CITES does not directly address issues of habitat loss and fragmentation or control of invasive species. The most high-profile and controversial action taken under CITES was the ban on trade in ivory that began in 1989. Prior to the ban, rampant poaching of African elephants had caused an approximate 50 per cent decline in elephant populations (Barbier et al., 1990). Poaching and population decline was a severe problem in East Africa, while Southern African countries (Botswana, Malawi, Namibia, South Africa, Zimbabwe) had relatively healthy elephant populations. The Southern African countries opposed the ban arguing that selling ivory provided a large financial reason for conserving elephants and the resources to prevent poaching. Many economists predicted that the ban would be ineffective, driving trade into the black market where high prices would provide large incentives to keep poaching (Barbier and Swanson, 1990). In practice, the ivory ban has been largely a success story. Poaching of elephants declined and elephant populations in East Africa recovered. The ban appears to have been successful because it acted not only to restrict supply but also to reduce demand. The major demand for ivory is largely for display purposes. Banning ivory caused a large decline in this demand because most would-be consumers do not wish to purchase and display an illegal item. Van Kooten and Bulte (2000) provide a useful summary of economic arguments about the ivory ban.

176

The international yearbook of environmental and resource economics

Trade bans on products from other endangered species have not proven to be as effective as the ivory ban. Black market trade in many items banned under CITES is alive and well (Webster, 1997). The work of Brown and Layton (2001) on the rhinoceros provides one economic argument for why CITES may be ineffective for curtailing poaching of some species. Demand for rhino is fueled by use of ground rhino horn in traditional Asian medicine. Medicinal demand is unlikely to decline even with the imposition of a trade ban. The case of rhino horn is more likely to resemble the case of illegal drugs, which is a large and thriving industry despite being illegal, than it is to resemble the case for ivory. The Convention on Biological Diversity was one of two high-profile conventions discussed at the Earth Summit in Rio de Janeiro in 1992, the other being the Convention on Climate Change. The goals of the CBD are to conserve biodiversity, sustainably use biodiversity, and equitably share the benefits from use of genetic resources. This last goal has been at the center of debates over sharing rents on bioprospecting as discussed above. The CBD provides guidance to national governments on biodiversity issues, however it is up to national governments themselves to take action. The CBD itself has no enforcement power. Unlike the Climate Change Convention, which spawned a set of ongoing international negotiations to address emissions of greenhouse gases, the CBD has spawned limited interest and little action since its inception. There is at present a striking negative correlation between the global distribution of economic wealth and the global distribution of biological wealth. As Balmford et al. (2003) point out, most of the cost-effective targets for conservation occur in developing countries while most of the conservation resources are in developed countries. Therefore, an effective and efficient conservation strategy may require large income transfers from economically rich but relatively biodiversity-poor temperate countries to biodiversity-rich but economically poor tropical countries. There are a number of mechanisms under which this might occur. One approach is to arrange for direct conservation payments through bilateral agreements or through multilateral institutions like the Global Environmental Fund. A second approach is to regulate international trade, either through CITES, or through other trade agreements. A third approach is to try to foster international markets for the sustainable use of biodiversity. This can be done either by finding marketable products that rely on the sustainable use of biodiversity (as discussed above in subsection A), or through the creation of environmental markets via TDRs, carbon sequestration credits, or other means. See Barbier (2000) for a summary of international policy approaches to address biodiversity conservation.

Strategies to conserve biodiversity

5

177

SUMMARY

Great strides have been made over the past decade in combining economic and biological analysis in integrated models of biodiversity conservation. However, there remain many unanswered questions and much more work remains to be done. For cost-effectiveness analysis, we need better models of how strategies affect long-term persistence of elements of biodiversity and the production of ecosystem services. This will require advances in ecology and economics and the links between the two fields. Existing approaches generally do not do an adequate job of incorporating dynamics and uncertainty. Going beyond cost-effectiveness analysis, economists and others will need to be able provide reliable evidence of the value of conserving biodiversity to be able to justify expenditures on conservation. As mentioned in the introduction, the valuation of biodiversity presents some difficult challenges. Another important area where further work is necessary is how to provide the right set of incentives for conservation to landowners and other decisionmakers. Designing incentive schemes that incorporate heterogeneity and spatial relationships but can be administered in a simple manner can be demanding. Designing and implementing incentive schemes are particularly demanding in developing countries where there is the additional handicap of limited existing institutions. Additional work on conservation policy and implementation issues is particularly important in developing countries, which contain a large share of biodiversity, have rapidly growing populations, and have urgent needs for economic development. Given the rapid pace of change and the enormity of the threats to biodiversity, there is a pressing need for research that can provide insights and information useful for conservation.

REFERENCES Ando, A., J. Camm, S. Polasky and A. Solow (1998), ‘Species distributions, land values, and efficient conservation’, Science, 279, 2126–8. Aylward, B., K. Allen, J. Echeverria and J. Tosi (1996), ‘Sustainable ecotourism in Costa Rica: the Monteverde Cloud Forest Preserve’, Biodiversity and Conservation, 5(3), 315–44. Arrow, K. and A. Fisher (1974), ‘Environmental preservation, uncertainty and irreversibility’, Quarterly Journal of Economics, 88, 312–9. Balmford, A., K. Gaston, S. Blyth, A. James and V. Kapos (2003), ‘Global variation in terrestrial conservation costs, conservation benefits, and unmet conservation needs’, Proceedings of the National Academy of Sciences, 100, 1046–50. Barbier, E.B. (1992), ‘Community-based development in Africa’, in T.M. Swanson and E.B. Barbier (eds), Economics for the Wilds: Wildlife, Diversity and Development, Washington, DC: Island Press, pp. 103–35.

178

The international yearbook of environmental and resource economics

Barbier, E.B. (2000), ‘Biodiversity, trade and international agreements’, Journal of Economic Studies, 27(1/2), 55–74. Barbier, E.B., J. Burgess, T. Swanson and D. Pearce (1990), Elephants, Economics and Ivory, London: Earthscan Publications. Barbier, E.B. and T. Swanson (1990), ‘Ivory: the case against the ban’, New Scientist, 17 November, 52–4. Beissinger, S.R. and D.R. McCullough (2002), Population Viability Analysis, Chicago: University of Chicago Press. Bishop, R.C. and M.P. Welsh (1992), ‘Existence value in benefit-cost analysis and damage assessment’, Land Economics, 69, 405–17. Botsford, L., A. Hastings and S. Gaines (2001), ‘Dependence of Sustainability on the configuration of marine reserves and larval dispersal distance’, Ecology Letters, 4, 144–50. Boyce, M.S. (1992), ‘Population viability analysis’, Review of Ecology and Systematics, 23, 481–506. Brookshire, D.S., L.S. Eubanks and A. Randall (1983), ‘Estimating option prices and existence value for wildlife resources’, Land Economics, 59(1), 1–15. Brown, G.M., Jr. and D. Layton (2001), ‘A market solution for preserving the black rhinoceros’, in J. Shogren and J. Tschirhart (eds), Endangered Species Protection in the United States: Biological Needs, Political Realities, Economic Choices, Cambridge: Cambridge University Press, pp. 32–50. Brown, G.M., Jr. and J.F. Shogren (1998), ‘Economics of the Endangered Species Act’, Journal of Economic Perspectives, 12(3), 3–20. Calkin, D., C.A. Montgomery, N.H. Schumaker, S. Polasky, J.L. Arthur and D.J. Nalle (2002), ‘Developing a production possibility set of wildlife species persistence and timber harvest value using simulated annealing’, Canadian Journal of Forest Research, 32(8), 1329–42. Camm, J., S. Polasky, A. Solow and B. Csuti (1996), ‘A note on optimization algorithms for reserve site selection’, Biological Conservation, 78, 353–55. Camm, J.D., S.K. Norman, S. Polasky and A.R. Solow (2002), ‘Nature reserve selection to maximize expected species coverage’, Operations Research, 50(6), 946–55. Carpenter, S.R., D. Ludwig and W.A. Brock (1999), ‘Management and eutrophication for lakes subject to potentially irreversible change’, Ecological Applications, 9, 751–71. Chape, S., L. Fish, P. Fox and M. Spaulding (2003), United Nations List of Protected Areas, Gland, Switzerland: IUCN/UNEP. Church, R. and C. ReVelle (1974), ‘The maximal coverage location problem’, Papers of the Regional Science Association, 32, 101–18. Church, R.L., D.M. Stoms and F.W. Davis (1996), ‘Reserve selection as a maximal coverage problem’, Biological Conservation, 76(2), 105–12. Clark, C. (1990), Mathematical Bioeconomics: The Optimal Use of Renewable Resources, 2nd edn, New York: John Wiley. Costanza, R., R. d’Arge, R. de Groot, S. Farber, M. Grasso, B. Hannon, K. Limburg, S. Naeem, R.V. O’Neill, J. Pareulo, R.G. Raskin, P. Sutton and M. van den Belt (1997), ‘The value of the world’s ecosystem services and natural capital’, Nature, 387, 253–60. Costello, C. and S. Polasky (2004), ‘Dynamic reserve site selection’, Resource and Energy Economics, 26(2), 157–74. Csuti, B., S. Polasky, P.H. Williams, R.L. Pressey, J.D. Camm, M. Kershaw, A.R. Kiester, B. Downs, R. Hamilton, M. Huso and K. Sahr (1997), ‘A compari-

Strategies to conserve biodiversity

179

son of reserve selection algorithms using data on terrestrial vertebrates in Oregon’, Biological Conservation, 80, 83–97. Daily, G.C. (ed.) (1997), Nature’s Services: Societal Dependence on Natural Ecosystems, Washington, DC: Island Press. Daily, G.C., P.R. Ehrlich and G.A. Sanchez-Azofeifa (2001), ‘Countryside biogeography: use of human dominated habitats by the avifauna of southern Costa Rica’, Ecological Applications, 11, 1–13. Daily, G.C. and K. Ellison (2002), The New Economy of Nature: The Quest to Make Conservation Profitable, Washington, DC: Island Press. Daily, G.C., T. Söderqvist, S. Aniyar, K. Arrow, P. Dasgupta, P.R. Ehrlich, C. Folke, A.M. Jansson, B.-O. Jansson, N. Kautsky, S. Levin, J. Lubchenco, K.-G. Mäler, D. Simpson, D. Starrett, D. Tilman and B. Walker (2000), ‘The value of nature and the nature of value’, Science, 289, 395–6. Dasgupta, P. and K.-G. Mäler (2003), The Economics of Non-Convex Systems, Dordrecht: Kluwer. Ehrenfeld, D. (1988), ‘Why put a value on biodiversity?’, in E.O. Wilson (ed.), BioDiversity, Washington, DC: National Academy Press. Faith, D.P. (1992), ‘Conservation evaluation and phylogenetic diversity’, Biological Conservation, 61, 1–10. Faith, D.P. (1994), ‘Phylogenetic pattern and the quantification of organismal biodiversity’, Philosophical Transactions of the Royal Society of London B, 345, 45–58. Fernandez, L. and L. Karp (1998), ‘Restoring wetlands through wetlands mitigation banks’, Environmental and Resource Economics, 13(3), 323–44. Ferraro, P.J. and A. Kiss (2002), ‘Direct payments to conserve biodiversity’, Science, 298, 1718–19. Ferraro, P.J. and R.D. Simpson (2002), ‘The cost-effectiveness of conservation payments’, Land Economics, 78(3), 339–53. Field, B. and J. Conrad (1975), ‘Economic issues in programs of transferable development rights’, Land Economics, 51, 331–40. Fisher, A.C. and W.M. Hanneman (1986), ‘Option values and the extinction of species’, Advances in Applied Microeconomics, 4, 169–90. Franklin, J. (1993), ‘Preserving biodiversity: species, ecosystems, or landscapes?’, Ecological Applications, 3(2), 202–5. Haight, R.G. (1995), ‘Comparing extinction risk and economic cost in wildlife conservation planning’, Ecological Applications, 5(3), 767–75. Heal, G.M. (2000), Nature and the Marketplace: Capturing the Value of Ecosystem Services, Washington, DC: Island Press. Hector, A., B. Schmid, C. Beierkuhnlein, M.C. Caldeira, M. Diemer, P.G. Dimitrakopoulos, J.A. Finn, H. Freitas, P.S. Giller, J. Good, R. Harris, P. Högberg, K. Huss-Danell, J. Joshi, A. Jumpponen, C. Körner, P.W. Leadley, M. Loreau, A. Minns, C.P.H. Mulder, G. O’Donovan, S.J. Otway, J.S. Pereira, A. Prinz, D.J. Read, M. Scherer-Lorenzen, E.-D. Schulze, A.-S.D. Siamantziouras, E.M. Spehn, A.C. Terry, A.Y. Troumbis, F.I. Woodward, S. Yachi and J.H. Lawton (1999), ‘Plant diversity and productivity experiments in European grasslands’, Science, 286, 1123–7. Hof, J. and M. Bevers (1998), Spatial Optimization for Managed Ecosystems, New York: Columbia University Press. Holling, C.S. (1973), ‘Resilience and stability of ecological systems’, Annual Review of Ecology and Systematics, 4, 1–23.

180

The international yearbook of environmental and resource economics

Humphries, C.J., P.H. Williams and R.I. Vane-Wright (1995), ‘Measuring biodiversity value for conservation’, Annual Review of Ecology and Systematics, 26, 93–111. Innes, R. (1997), ‘Takings, compensation and equal treatment for owners of developed and undeveloped property’, Journal of Law and Economics, 40(2), 403–32. Innes, R., S. Polasky and J. Tschirhart (1998), ‘Takings, compensation and endangered species protection on private land’, Journal of Economic Perspectives, 12(3), 35–52. International Union for Conservation of Nature and Natural Resources (IUCN) (2003), 2003 IUCN Red List of Threatened Species, Gland, Switzerland: IUCN. Jackson, J., M. Kirby, W. Berger, K. Bjorndal, L. Botsford, B. Bourque, R. Bradbury, R. Cooke, Jon Erlandson, J. Estes, T. Hughes, S. Kidwell, C. Lange, H. Lenihan, J. Pandolfi, C. Peterson, R. Steneck, M. Tegner and R. Warner (2001), ‘Historical overfishing and the recent collapse of coastal ecosystems’, Science, 293, 629–38. Jantzen, D. (1992), ‘A south–north perspective on science in the management, use, and economic development of biodiversity’, in O.T. Sandlund, K. Hindar and A.H.D. Brown (eds), Conservation of Biodiversity for Sustainable Development. Oslo: Scandanavian University Press. Kareiva, P. and M. Marvier (2003), ‘Preserving biodiversity coldspots’, American Scientist, 91(4), 344. Kohn, R.E. (1994), ‘Alternative property rights to wetlands externalities’, Ecological Economics, 10(1), 61–8. Koo, B. and B. Wright (1999), ‘The role of biodiversity products as incentives for conserving biological diversity: some instructive examples’, The Science of the Total Environment, 240, 21–30. Lawton, J.H. and R.M. May (eds) (1995), Extinction Rates, Oxford: Oxford University Press. Levin, S. (1999), Fragile Dominion: Complexity and the Commons, Reading, MA: Perseus Books. Lichtenstein, M.E., and C.A. Montgomery (2003), ‘Biodiversity and timber in the Coast Range of Oregon: inside the production possibility frontier’, Land Economics, 79(1), 56–73. Lindberg, K. (2001), ‘Economic impacts’, in D. Weaver (ed.), The Encyclopedia of Ecotourism, London: CAB International, pp. 363–77. Liu, J., M. Linderman, Z. Ouyang, L. An, J. Yang and H. Zhang (2001), ‘Ecological degradation in protected areas: The case of Wolong Nature Reserve for giant pandas’, Science, 292, 98–101. Lomborg, B. (2001), The Skeptical Environmentalist: Measuring the Real State of the World, Cambridge: Cambridge University Press. Loomis, J.B. and D.S. White (1996), ‘Economic benefits of rare and endangered species: summary and meta-analysis’, Ecological Economics, 18, 197–206. MacArthur, R. and E.O. Wilson (1967), Island Biogeography, Princeton, NJ: Princeton University Press. Magurran, A.E. (1988), Ecological Diversity and Its Measurement, Princeton, NJ: Princeton University Press. Magurran, A.E. (2004), Measuring Biological Diversity, Malden, MA: Blackwell. Maille, P. and R. Mendelsohn (1993), ‘Valuing ecotourism in Madagascar’, Journal of Environmental Management, 39, 213–18. Mann, C.C. and M.L. Plummer (1995), Noah’s Choice: The Future of Endangered Species, New York: Alfred A. Knopf.

Strategies to conserve biodiversity

181

Marshall, E., F. Homans and R. Haight (2000), ‘Exploring strategies for improving the cost effectiveness of endangered species management’, Land Economics, 76(3), 462–73. May, R.M. (1988), ‘How many species are there on earth?’, Science, 241, 1441–9. May, R.M. (1990), ‘How many species?’, Philosophical Transactions of the Royal Society of London, Biological Series, 330, 283–304. McGrady-Steed, J., P.M. Harris and P.J. Morin (1997), ‘Biodiversity regulates ecosystem predictability’, Nature, 390, 162–5. McKee, J.K. (2003), Sparing Nature: The Conflict Between Human Population Growth and Earth’s Biodiversity, New Brunswick, NJ: Rutgers University Press. Mendelsohn, R. and M.J. Balick (1995), ‘The value of undiscovered pharmaceuticals in tropical forests’, Economic Botany, 49(2), 223–8. Merrifield, J. (1996), ‘A market approach to conserving biodiversity’, Ecological Economics, 16, 217–26. Miller, K.R. (1996), ‘Conserving biodiversity in managed landscapes’, in R.C. Szaro and D.W. Johnston (eds), Biodiversity in Managed Landscapes: Theory and Practice, Oxford: Oxford University Press, pp. 425–41. Mills, D. (1980), ‘Transferable development rights markets’, Journal of Urban Economics, 7, 63–74. Mittermeier, R., N. Myers, J. Thomsen, G. da Fonseca and S. Olivieri (1998), ‘Biodiversity hotspots and major tropical wilderness areas: approaches to setting conservation priorities’, Conservation Biology, 12, 516–20. Montgomery, C.A., G.M. Brown, Jr. and D.M. Adams (1994), ‘The marginal cost of species preservation: the northern spotted owl’, Journal of Environmental Economics and Management, 26(2), 111–28. Montgomery, C.A., R.A. Pollak, K. Freemark and D. White (1999), ‘Pricing biodiversity’, Journal of Environmental Economics and Management, 38(1), 1–19. Myers, N. (1980), The Sinking Ark: A New Look at the Problem of Disappearing Species, New York: Pergamon Press. Myers, N. (1988), ‘Threatened biotas: “hot spots” in tropical forests’, Environmentalist, 8, 187–208. Myers, N., R. Mittermeier, C. Mittermeier, G. da Fonseca and J. Kent (2000), ‘Biodiversity hotspots for conservation priorities’, Nature, 403, 853–8. Myers, R.A. and B. Worm (2003), ‘Rapid worldwide depletion of predatory fish communities’, Nature, 423, 280–3. Naaem, S. and S. Li (1997), ‘Biodiversity enhances ecosystem reliability’, Nature, 390, 507–9. Naaem, S., L.J. Thompson, S.P. Lawler, J.H. Lawton and R.M. Woodfin (1994), ‘Declining biodiversity can alter the performance of ecosystems’, Nature, 368, 734–6. Naaem, S., L.J. Thompson, S.P. Lawler, J.H. Lawton and R.M. Woodfin (1995), ‘Empirical evidence that declining species diversity may alter the performance of terrestrial ecosystems’, Transactions of the Royal Society of London B, 347, 249–62. Nalle, D.J., C.A. Montgomery, J.L. Arthur, S. Polasky and N.H. Schumaker (2004), ‘Modeling joint production of wildlife and timber’, Journal of Environmental Economics and Management. National Research Council (NRC) (1995), Science and the Endangered Species Act, Washington DC: National Academy Press. Norton, B. (1987), Why Preserve Natural Variety?, Princeton, NJ: Princeton University Press.

182

The international yearbook of environmental and resource economics

Noss, R. (1990), ‘Indicators for monitoring biodiversity: a hierarchical approach’, Conservation Biology, 4(4), 355–64. Organisation for Economic Co-operation and Development (OECD) (2001), Agricultural Policies in OECD Countries: Monitoring and Evaluation 2000, Paris: OECD. Pagiola, S., J. Bishop and N. Landell-Mills (2002), Selling Forest Environmental Services: Market-Based Mechanisms for Conservation, London: Earthscan Publications. Panayotou, T. (1994), ‘Conservation of biodiversity and economic development: the concept of transferable development rights’, Environmental and Resource Economics, 4, 91–110. Parmesan, C. and G. Yohe (2003), ‘A globally coherent fingerprint of climate change impacts across natural systems’, Nature, 421, 37–42. Perrings, C., K.-G. Mäler, C. Folke, C.S. Hollings and B.-O. Jansson (1995), ‘Introduction: framing the problem of biodiversity loss’, in C. Perrings, K.G. Mäler, C. Folke, C.S. Holling and B.-O. Jansson (eds), Biodiversity Loss: Economic and Ecological Issues, Cambridge: Cambridge University Press, pp. 1–17. Pimm, S.L., G.J. Russell, J.L. Gittleman and T.M. Brooks (1995), ‘The future of biodiversity’, Science, 269, 347–50. Pimm, S.L. (2001), The World According to Pimm: A Scientist Audits the Earth, New York: McGraw-Hill. Polasky, S., J.D. Camm and B. Garber-Yonts (2001), ‘Selecting biological reserves cost-effectively: an application to terrestrial vertebrate conservation in Oregon’, Land Economics, 77(1), 68–78. Polasky, S., J.D. Camm, A.R. Solow, B. Csuti, D. White and R. Ding (2000), ‘Choosing reserve networks with incomplete species information’, Biological Conservation, 94, 1–10. Polasky, S, C. Costello and A. Solow (2005a), ‘The economics of biodiversity conservation’, in J. Vincents and K.-G. Mäler (eds), The Handbook of Environmental Economics, Malden, MA: Blackwell. Polasky, S. and H. Doremus (1998), ‘When the truth hurts: endangered species policy on private land with imperfect information’, Journal of Environmental Economics and Management, 35(1), 22–47. Polasky, S., E. Nelson, E. Lonsdorf, P. Fackler and A. Starfield (2005b), ‘Conserving species on a working landscape: land use with biological and economic objectives’, Ecological Applications (in press). Polasky, S. and A. Solow (1995), ‘On the value of a collection of species’, Journal of Environmental Economics and Management, 29, 298–303. Polasky, S. and A. Solow (1999), ‘Conserving biological diversity with scarce resources’, in J. Klopatek and R. Gardner (eds), Landscape Ecological Analysis: Issues and Applications, New York: Springer. Polasky, S., A. Solow and J. Broadus (1993), ‘Searching for uncertain benefits and the conservation of biological diversity’, Environmental and Resource Economics, 3(2), 171–81. Pounds, J.A. and R. Puschendorf (2004), ‘Clouded futures’, Nature, 427, 107–8. Pressey, R.L., C.J. Humphries, C.R. Margules, R.I. Vane-Wright and P.H. Williams (1993), ‘Beyond opportunism: key principles for systematic reserve selection’, Trends in Ecology and Evolution, 8(4), 124–8. Pressey, R.L., H.P. Possingham and J.R. Day (1997), ‘Effectiveness of alternative heuristic algorithms for identifying minimum requirements for conservation reserves’, Biological Conservation, 80, 207–19.

Strategies to conserve biodiversity

183

Principe, P. (1989), The Economic Valuation of Biodiversity Among Medicinal Plants, Paris: OECD. Rausser, G. and A. Small (2000), ‘Valuing research leads: bioprospecting and the conservation of genetic resources’, Journal of Political Economy, 108(1), 173–206. Reid, W.V. (1996), ‘Beyond protected areas: changing perceptions of ecological management objectives’, in R.C. Szaro and D.W. Johnston (eds), Biodiversity in Managed Landscapes: Theory and Practice, Oxford: Oxford University Press, pp. 442–53. Renard, V. (1999), ‘Application of tradable permits to land-use management’, in Implementing Domestic Tradable Permits for Environmental Protection, Paris: Organisation for Economic Co-operation and Development. Rodrigues, A.S.L., S.J. Andelman, M.I. Bakarr, L. Boitani, T.M. Brooks, R.M. Cowling, L.D.C. Fishpool, G.A.B. da Fonseca, K.J. Gaston, M. Hoffman, J.S. Long, P.A. Maquet, J.D. Pilgrim, R.L. Pressey, J. Schipper, W. Sechrest, S.N. Stuart, L.G. Underhill, R.W. Waller, M.E.J. Watts and X. Yan (2004), ‘Effectiveness of the global protected area network in representing species diversity’, Nature, 428, 640–3. Rohweder, M.R., C.W. McKetta and R. Riggs (2000), ‘Economic and biological compatibility of timber and wildlife production: an illustrative use of production possibilities frontier’, Wildlife Society Bulletin, 28(2), 435–47. Rolston, H., III (1994), Conserving Natural Value, New York: Columbia University Press. Root, T.L., J.T. Price, K.R. Hall, S.H. Schneider, C. Rosenzweig and J.A. Pounds (2003), ‘Fingerprints of global warming on wild animals and plants’, Nature, 421, 57–60. Rosenzweig, M.L. (2003), Win-Win Ecology: How the Earth’s Species Can Survive in the Midst of Human Enterprise, Oxford: Oxford University Press. Sanchirico, J. and J. Wilen (2002), ‘The impacts of marine reserves on limited-entry fisheries’, Natural Resource Modeling, 15(3), 291–310. Sanchirico, J. and J. Wilen (2001), ‘A bioeconomic model of marine reserve creation’, Journal of Environmental Economics and Management, 42, 257–76. Scheffer, M., S. Carpenter, J. Foley, C. Folke and B. Walker (2001), ‘Stochastic events can trigger large state shifts in ecosystems with reduced resilience’, Nature, 413, 591–6. Shabman, L., K. Stephenson and P. Scodari (1998), ‘Wetland credit sales as a strategy for achieving no-net-loss: the limitations of regulatory conditions’, Wetlands, 18(3), 471–81. Simpson, D., R. Sedjo and J. Reid (1996), ‘Valuing biodiversity for use in pharmaceutical research’, Journal of Political Economy, 104(1), 163–85. Smith, R.B.W. and J. Shogren (2002), ‘Voluntary incentives design for endangered species protection’, Journal of Environmental Economics and Management, 43(2), 169–87. Solow, A., S. Polasky and J. Broadus (1993), ‘On the measurement of biological diversity’, Journal of Environmental Economics and Management, 24(1), 60–8. Solow, A. and S. Polasky (1994), ‘Measuring biological diversity’, Environmental and Ecological Statistics, 1(2), 95–107. Soule, M.E. (1987), Viability Populations for Conservation, Cambridge: Cambridge University Press. Thomas, C.D., A. Cameron, R.E. Green, M. Bakkenes, L.J. Beaumont, Y.C. Collingham, B.F.N. Erasmus, M.F. de Siqueira, A. Grainger, L. Hannah,

184

The international yearbook of environmental and resource economics

L. Hughes, B. Huntley, A.S. van Jaarsveld, G.F. Midgley, L. Miles, M.A. OrtegaHuerta, A.T. Peterson, O.L. Phillips and S.E. Williams (2004), ‘Extinction risk from climate change’, Nature, 427, 145–8. Thorsnes, P. and G.P.W. Simons (1999), ‘Letting the market preserve land: the case for a market driven transfer of development rights program’, Contemporary Economic Policy, 17(2), 256–66. Tilman, D. and J.A. Downing (1994), ‘Biodiversity and stability in grasslands’, Nature, 367, 363–5. Tilman, D., D. Wedin and J. Knops (1996), ‘Productivity and sustainability influenced by biodiversity in grassland ecosystems’, Nature, 379, 718–20. Underhill, L.G. (1994), ‘Optimal and suboptimal reserve selection algorithms’, Biological Conservation, 70, 85–7. Van Kooten, G.C. and E.H. Bulte (2000), The Economics of Nature: Managing Biological Assets, Malden, MA: Blackwell. Vane-Wright, R.I., C.J. Humphries and P.H. Williams (1991), ‘What to protect? – Systematics and the agony of choice’, Biological Conservation, 55, 235–54. Walker, B., A. Kinzig and J. Langridge (1999), ‘Plant attribute diversity, resilience, and ecosystem function: the nature and significance of dominant and minor species’, Ecosystems, 2(2), 95–113. Warrick, J. (1998), ‘Mass extinction underway, majority of biologists say’, Washington Post, 21 April. Wear, D.N., M.G. Turner and R.O. Flamm (1996), ‘Ecosystem management with multiple owners: landscape dynamics in a Southern Appalachian watershed’, Ecological Applications, 6(4), 1173–88. Weber, M. and W. Adamowicz (2002), ‘Tradable land-use rights for cumulative environmental effects management’, Canadian Public Policy, 28(4), 581–95. Webster, D. (1997), ‘The looting and smuggling and fencing and hoarding of impossibly precious, feathered and scaly wild things’, The New York Times Magazine, 16 February. Weitzman, M.L. (1992), ‘On diversity’, Quarterly Journal of Economics, 107(2), 363–405. Weitzman, M.L. (1998), ‘The Noah’s Ark problem’, Econometrica, 66, 1279–98. Wells, M. and K.E. Brandon (1992), People and Parks: Linking Protected Area Management With Local Communities, Washington, DC: The World Bank. Western, D. and R.M. Wright (eds) (1994), Natural Connections: Perspectives in Community-based Conservation, Washington, DC: Island Press. Wilcove, D., D. Rothstein, J. Dubow, A. Phillips and E. Losos (1998), ‘Quantifying threats to imperiled species in the United States’, Bioscience, 48, 607–15. Wilson, E.O. (1999), The Diversity of Life, New York: W.W. Norton. Wilson, E.O. (2002), The Future of Life, New York: Vintage Books. World Resources Institute, United Nations Development Programme, United Nations Environment Programme, and World Bank (2000), World Resources 2000–2001, Amsterdam: Elsevier. World Tourism Organization (2002), Compendium of Tourism Statistics. Wunder, S. (2000), ‘Ecotourism and economic incentives – an empirical approach’, Ecological Economics, 32, 465–79.

5.

Corporate sustainability Stefan Schaltegger and Roger Burritt

1

INTRODUCTION

This chapter provides an overview of and discussion about current approaches to corporate sustainability. Considerable interest has been generated in the notion of corporate sustainability for a number of reasons. Corporations provide a practical, direct point of contact for the implementation of government policy. Many large corporations control more resources than many sovereign nations. Managements of corporations that seek to gain a competitive advantage are beginning to appreciate the necessity for promoting corporate sustainability initiatives as a way of differentiating themselves from competitors as well as of reducing costs of undertaking business and risks associated with operations. For example, since the concept of ecologically sustainable development appeared (Commission for the Future, 1987) and the related ‘precautionary principle’ was introduced (Commonwealth of Australia, 1990, p. 9), environmental risk has become a growing concern (Schaltegger et al., 2003, pp. 195–203). There is now greater focus on the management of environmental risks through voluntary means. A further factor promoting corporate interest in sustainability is that when problems occur, such as severe or persistent corporate impacts on the environment, communication with stakeholders is an important way of trying to minimize damage before or after the event. A distinction should be made between sustainability and sustainable development. The former is taken here to represent the goal or end point of the process of sustainable development. The term ‘corporate sustainability’ links the general approach to sustainability with sustainability at the corporate level. Section 2 examines the question as to what motivates managers to address sustainability issues within the corporate milieu. Section 3 briefly introduces some of the main aspects of the general concept of sustainability and its links with the corporate level. The fourth section examines the characteristics of and challenges for corporate sustainability in greater depth and considers what is currently included in corporate sustainability. Section 5 discusses the directions corporate

185

186

The international yearbook of environmental and resource economics

sustainability should take. When provided with strong enough reasons management is challenged to aim for corporate sustainability and, thus, needs compelling competitive strategies and management tools to guide implementation. This leads into the discussion of corporate strategies for sustainability. In Section 6 the empirical literature examining the measurement of corporate sustainability and economic performance is reviewed. The chapter concludes (Section 7) with a discussion about the future direction of corporate sustainability and the outlook for further research.

2

FROM THE GENERAL SUSTAINABILITY CONCEPT TO CORPORATE SUSTAINABILITY

Sustainability has become a much cited and debated objective at the political as well as the corporate level. As a global vision sustainability embodies hopes for a peaceful society with social equity and justice and economic prosperity in a clean, natural environment. For more than a century a range of scholars has addressed various aspects of sustainability and sustainable development (Keating, 1993). For a recent comprehensive economics perspective on the notion of sustainability in general see Pezzey and Toman (2002, pp. 165–233). For specific corporate accounting perspectives on sustainable development see Bebbington (2001) and Milne (1996). Whereas economic theory has focused for a long time on the relation and effects of economic growth on income distribution and social equity, the link between economic growth and environmental impacts grew in interest mainly in the 1980s and 1990s (see e.g. Arrow et al., 1995; Daly, 1996; Ekins, 2000). With the path breaking Brundtland report promoting sustainable development in 1987 (UNWCED, 1987) sustainable development has escaped from the academic world and entered the political arena in the form of a policy set to be adopted. The emphasis of the Brundtland definition of sustainable development is on the consideration of future generations and intergenerational justice. With the involvement of more groups in discussions about sustainability the focus has shifted for the last two decades towards a three-dimensional approach considering economic, environmental and social aspects of economic development and their interrelationships. With this step the approach started to embrace the previous extensive work of economists, sociologists and ecologists. While academic papers about sustainability were initially accompanied by considerable political debate and non-governmental organization (NGO) activity, with the Brundtland report it was only the UNCED ‘Earth Charter’ Conference in Rio de Janeiro in 1992 and the book Changing Course launched on that occasion by Stephan Schmidheiny and the

Corporate sustainability

187

Business Council for Sustainable Development (now World Business Council for Sustainable Development; Schmidheiny and BCSD, 1992) which brought widespread attention to the notions of sustainability and sustainable development into the boardrooms of companies. A sustainable society is one that has reached sustainability, at a given point of time, through the process of ecologically and economically sustainable development (Schaltegger and Burritt, 2000, p. 46). In this chapter reference to the notion of sustainability includes reference to the concept of sustainable development. In spite of this increasing level of attention in the political arena and within corporations the literature on sustainability and sustainable development remains rather confusing to those concerned about operationalizing the concept. Only two years after the Brundtland report appeared Pearce et al. (1989, pp. 173–85) cited 24 different definitions of sustainable development. Hayer (1995, p. 1) identified at least 40 definitions up to 1991 and Kastenholz et al. (1996, p. 1) assembled over 60. Although sustainable development has probably been defined in over one hundred different ways by now (Steurer, 2001, p. 537) some main points of consensus seem to prevail such as the aim to create a development towards the best possible quality of life for an indefinite period of time which can be extended to the whole globe. Given this universal objective it is not surprising that the term has become a popular ‘catch-all’ phrase. ‘Sustainable development is a neo-renaissance idea that covers the whole of human endeavour and planetary survival. Who could possibly oppose it?’ (O’Riordan and Voisey, 1997, p. 4). Being accepted by many people sustainable development is being put into operation and supported with different instruments by various groups. As a result, a number of core debates on sustainable development can be distinguished (e.g. Steurer, 2001). Perhaps the main theoretical debate in economics has been about strong and weak sustainability (Pearce et al., 1989, p. 34; Reid, 1995, pp. 192ff.; Daly, 1996, pp. 110–14; Gutés, 1996, pp. 147f.; Neumayer, 1999, pp. 1–7, 22–9). Weak sustainable development or economic growth requires that the total amount of a society’s natural and human made (or manufactured) assets must not be reduced, but it allows for substitution between environmental, economic and social assets. This enables, in theory, all types of capital and the services and welfare generated by them to be expressed in the same monetary unit for comparison (Ekins et al., 2003, p. 168). Strong sustainability on the other hand does not permit such substitution effects. In particular, it considers that critical natural capital should not be diminished by human activity (Ekins, 2003, p. 167). The ‘constant capital rule’ aims at a ‘steady-state economy’ (Daly, 1991, pp. 35–40; Daly, 2000,

188

The international yearbook of environmental and resource economics

pp. 15f.), the protection of life-support services and critical assets. Very strong and very weak forms of sustainability have also been defined to represent no substitutability and complete substitutability (Turner, 1993, pp. 9–15) but from a pragmatic, corporate perspective these extremes are of less interest. The bipolar debate has somewhat been overtaken by the approach of ‘balanced or sensible sustainability’ (Bartelmus, 1997, p. 329; Hueting and Reijnders, 1998, p. 140) which focuses on the central aspects of the mainstream sustainability paradigm (balanced sustainability) which includes ‘non declining human welfare over time’ (Pearce, 1991, p. 1), the consideration of the three – social, environmental and economic – dimensions and harmonization of goals as well as intragenerational and intergenerational justice. Furthermore, most scholars agree that a more active role for public environmental policy is necessary (for example through the development of an appropriate regulatory mix, including environmental taxation, see Ayres and Braithwaite (1992); Gunningham et al. (1998)), that the limits of growth can be extended with technological development and that the implementation of sustainability requires measurement and the use of sustainability indicators. However, as government is not immune to rent seeking and because government interventions can also be a source of inefficiency and sustainability problems (e.g. Tietenberg, 1992, pp. 62ff.), voluntary corporate sustainability measures have become an issue to management. Entry of the debate into the corporate world has also lead to a widespread agreement that sustainable development of society and the economy requires sustainable development of corporations because of their pervasive nature in the transformation process (see Stone, 1975). Any kind of sustainable development requires a sustainable development of the economy and of its corporations. Changing from the macro to the business perspective raises the question of what are the main characteristics of and challenges for corporate sustainability.

3

CHARACTERISTICS OF AND CHALLENGES FOR CORPORATE SUSTAINABILITY

Corporate Sustainability is a Heuristic Multi-criteria Approach The Rio conference made a major contribution to the fact that environmental protection in business practice crossed the psychological threshold from being addressed as a technological problem to becoming a business challenge and opportunity. Since then environmental management has become established as an important management function in

189

Corporate sustainability

countless corporations around the world. The World Business Council for Sustainable Development (WBCSD) is probably the most visible business organization that has popularized eco-efficiency as one key specific aspect of corporate sustainability. In the past decade the term ‘corporate sustainability’ has gained considerable interest amongst corporate practitioners and has also been examined in the academic literature (e.g. Gladwin et al., 1995a; Gladwin et al., 1995b; Hart, 1997; BMU and BDI, 2002; Dyllick and Hockerts, 2002; Schaltegger and Dyllick, 2003). However, as the vision of corporate sustainability is currently not welldefined it remains a broad approach that includes various characteristics, in particular relating to the contextual integration of economic, environmental and social aspects. It may seem astonishing to realize that the bestknown aspect of corporate sustainability is the heuristic, multi-criteria triple bottom line perspective which aims to integrate economic, social and environmental aspects of business management (Elkington, 1998). This differs from the macro and political levels where the orientation towards future and present needs as formulated in the Brundtland report has dominated for much longer. Figure 5.1 illustrates the three pillar approach of corporate sustainability. Economic effectiveness

Economic

3

Eco-efficiency

4

6

Socio-efficiency

Integration Ecological

Social 5

1

Figure 5.1

Eco-effectiveness

Eco-justice

2

Corporate sustainability challenges

Socio-effectiveness

190

The international yearbook of environmental and resource economics

Contextual Challenges for Corporate Sustainability Economic effectiveness, that is achieving the best possible economic result, is the classic entrepreneurial and management task, which is also of relevance in the context of sustainable development. However, as it is the subject of conventional business management it is usually not specifically addressed as a task of and challenge for corporate sustainability. As a result, the following contextual corporate sustainability challenges remain to be dealt with by corporate sustainability management (BMU and BDI, 2003; Schaltegger et al., 2003 and Figure 5.1): ●

The ecological challenge to increase ecological effectiveness, or ecoeffectiveness (① in Figure 5.1) of business activities: all human activities influence the ecosystem. Some influences have irreversible effects and are considered to be of major relevance to the survival and existence of an intact natural environment. The central environmental problems include the greenhouse effect, the destruction of the ozone layer, acidification and overnitrification of soil and water, declining biodiversity, photochemical smog, toxicological burdens harmful to humans and the environment, desertification, etc. The excessive overall environmental burdens in many fields such as CO2 emissions therefore confront businesses with the challenge of making substantial reductions in the absolute scale of the environmental impacts of their production processes, products, investments, etc. (e.g. Braungart and McDonough, 2002). For various applications like product life cycle assessment (LCA) aggregate indicators of eco-effectiveness are commonly proposed. An impressive number of assessment approaches have been proposed for this purpose, among which the method of the Centre for Milieukunde at the University of Leiden (Heijungs et al., 1992) and the ecological footprint (Wackernagel and Rees, 1996) have maybe received most international attention. Because of difficulties in arriving at a commonly accepted integrative measure of environmental impact added ecoeffectiveness is mostly expressed in terms of specific indicators such as CO2 emissions or CO2 equivalents (as one indicator of the CML method). Other authors propose to consider the mere total quantity of material mass involved in a product life cycle (e.g. Schmidt-Bleek, 1994). The criterion for assessing how successfully a company meets the ecological challenge is ecological effectiveness (also ecoeffectiveness or environmental effectiveness). Ecological effectiveness measures the absolute environmental performance (e.g. tonnes of CO2 emissions reduced last period) and is a general description of the

Corporate sustainability

●

●

191

extent to which the targeted objective of minimizing environmental impacts has actually been achieved (see Freimann, 2000; Dyllick and Hockerts, 2002). The social challenge to increase social effectiveness, or socioeffectiveness (② in Figure 5.1) of the company: the social challenge related to corporate sustainability consists in ensuring the existence and success of the enterprise while at the same time taking account of the diversity of social, cultural and individual social demands. This is related to safeguarding the social acceptance of the enterprise and the legitimation of its business activities. When dealing with a great variety of social factors such as interregional and intertemporal equality of rights, fairness, equity of needs and performance, it has to be borne in mind that these can never be completely satisfied as human desires may be unlimited. Thus management is challenged to set priorities in a dialogue or multilogue with principal stakeholders. Socio-effectiveness is the criterion that indicates how well a company has reduced the absolute level of negative social impacts, relative to expectation, and the extent to which it gives rise to important positive social impacts and benefits. The economic challenge to environmental and social management aimed at improving eco-efficiency (③ in Figure 5.1) and socioefficiency (④ in Figure 5.1): whereas the traditional economic challenge consists of creating corporate and shareholder value and increasing the company’s profitability, the economic sustainability challenge is concerned with undertaking effective environmental management and social management in as economic a way as possible. Because profit orientated businesses operating in a competitive setting are established and run primarily for economic purposes, environmental protection and social commitment are always confronted with the challenge of increasing value, making a contribution to profitability or at least minimizing costs. The traditional criterion to achieve economic success is efficiency. Efficiency is a relative measure of performance. The economic interpretation of efficiency is based on monetary performance data and is expressed in profitability indicators (e.g. return on investment, return on equity, value added, etc.). In the context of the goal of corporate sustainable development, the monetary efficiency interpretation is supplemented with ecological and social aspects. In addition to economic efficiency two types of efficiency are of special importance: eco-efficiency as economic-ecological efficiency and socio-efficiency as economicsocial efficiency. Eco-efficiency (also called E2-efficiency) is defined

192

●

The international yearbook of environmental and resource economics

as the ratio of an economic (monetary) to a physical (ecological) measure (Schaltegger and Sturm, 1990, pp. 279ff.; Schmidheiny and BCSD, 1992). It can be defined as the ratio of value added to environmental impact added per unit, where environmental impact added is equivalent to the sum of all environmental impacts generated directly or indirectly by a product or activity. Examples of eco-efficiency measures are value added (in $ or Euro)/per tonne of emitted CO2, contribution margin of a product (in $ or Euro)/ contribution to greenhouse effect (in CO2 equivalents), etc. Various publications provide examples of possible ratios between economic and environmental performance (e.g. ‘factor four’ by von Weizsäcker et al., 1997 or ‘factor ten’ by Schmidt-Bleek, 1994) and case collections of companies applying and promoting eco-efficiency (e.g. Schmidt-Bleek, 1994; von Weizsäcker et al., 1997; Hawken et al., 1999). Furthermore, various company websites illustrate the application of eco-efficiency analysis (e.g. www.basf.com.de). On the same lines as eco-efficiency, socio-efficiency (also societal efficiency) can be defined as the ratio of value added to social impact added, where social impact added represents the sum of all negative social impacts originating from a company, product, process or activity. Examples of socio-efficiency yardsticks are value added (in $ or Euro)/staff accidents (number), or value added (in $ or Euro)/absence due to illness (days). The integration challenge (⑤ of Figure 5.1) bringing together the first three challenges and integrating environmental and social management in conventional economically orientated business management: the three challenges of sustainable management as described above can be met by means of systematic efforts to act in an ecoeffective and socially effective as well as in an eco-efficient and socioefficient manner. However, the real and biggest challenge corporate sustainability management faces is the integration challenge. This integration challenge derives from combining and simultaneously satisfying the three objectives described above. Contextual integration of the ‘three pillars’ of sustainable development requires simultaneous attention to and improvement of the four challenges of ecological effectiveness, social effectiveness, eco-efficiency and socioefficiency.

Further Challenges for Corporate Sustainability Apart from the four contextual issues corporate sustainability embraces further aspects dealing with time, participation, methodological integration

Corporate sustainability

193

into core business methods and processes and the role of the corporation in society: ●

●

●

Orientation towards the future: Orientation towards the future has always been a core business management issue, such as in management accounting with tools like investment appraisal and budgeting, or in the assessment of the company’s economic value by financial analysts and investors. With environmental management, consideration of future impacts of emissions and other environmental impacts has been added to the set of management responsibilities. However, explicit consideration of future generations and non-economic stakeholders has only been addressed in the business literature more recently (see e.g. Dyllick and Hockerts, 2003; Schaltegger et al., 2003). Several companies, such as the carpet supplier Interface, have also addressed this issue: ‘Sustainability is a fabric that must be woven through all aspects of our lives, constantly challenging us to apply new metrics and new solutions in the daily decisions that we make. It translates into the choices we make today and how those choices will affect future generations long after we are gone’ (Interface, 2003, www.interfacesustainability.com/whatis.html). It is of interest to note that Interface, in its move towards sustainability in the future, has had to choose between different ecological options. For example, while it has developed fibres made from renewable rather than synthetic based resources for use in its carpet products, environmental groups have questioned the use of such fibres because they cannot be certified as non-genetically modified. ‘The current commercial infrastructure for corn processing is such that there is no economically viable way to separate GM corn from non-GM corn during the milling process’ (www.interfacesustainability.com/biobased.html). Sustainable development of the company as an organizational development process: Sustainable development of a corporation requires organisational development and organizational learning processes. If this view is taken to its extreme, corporate sustainability cannot reflect a state which management may strive for but will always have to be a state of constant change, progress and development. Corporate sustainability will always remain a moving target of organizational development. Nevertheless, for reasons of clarity it is helpful to distinguish between the target state of corporate sustainability and the process of sustainable development for a company striving towards corporate sustainability. Integration into core business: The integration of environmental and social activities in core business processes and with other management

194

●

●

The international yearbook of environmental and resource economics

tools has been addressed as one aspect of the integration challenge for corporate sustainability (e.g. BMU and BDI, 2003). In practice, environmental and social management are mostly established in parallel with conventional management systems. This can lead to inefficient business solutions where, for example, environmental management corrects problems through ‘end-of-pipe developments’ which impede attempts to find innovative products and other sustainability orientated process based innovations. Thus one of the core challenges for corporate sustainability management is the integration of environmental and social issues and management activities into the core business management processes and systems. Participation and integration of stakeholders: When transposing Brundtland’s commonly accepted notion of sustainable development, corporate sustainable development can be seen accordingly as meeting the needs of a corporation’s direct and indirect stakeholders without compromising its ability to meet the needs of future stakeholders as well (e.g. Dyllick and Hockerts, 2002). To achieve this goal, corporations have to maintain and increase their economic, social and environmental capital base while actively contributing to sustainability in the political sphere. Corporate sustainability thus includes the vision of participation in processes for analysing sustainability problems, for finding solutions to these problems and in decision and implementation processes. Contribution towards sustainable development of the economy and society: As a consequence of the broad approach and its various contextual aspects, corporate sustainability is not limited to the corporate organization itself but directs attention towards social embeddedness of the corporation and the influence it has on its social environment. In the more recent marketing and entrepreneur literature corporate sustainability is therefore seen as an approach that is not limited to niche markets and market related business activities (e.g. Schaper, 2003). Instead, corporate sustainability requires the adoption of sustainability as a high-priority business goal as well as recognition of its considerable potential impact on mass markets and society. Sustainability managers can thus be seen as actors who of necessity have to involve themselves in the development of market frameworks for internalization of external effects of business and who, through lobbying and other means, increase public awareness of the need for sustainability (e.g. Dyllick et al., 1997). Thus, corporate sustainability management, through the adoption of a more encompassing view, is seen as a business approach designed to shape the environmental, social and economic effects of a company in a way that,

195

Corporate sustainability

firstly, results in sustainable development of the company and, secondly, provides an important contribution towards sustainable development of the economy and society (e.g. Schaltegger et al., 2003). Corporate sustainability can be seen as the result of management attempts to tackle sustainability challenges. How Do the Characteristics of Corporate Sustainability Relate to Each Other? The fact that corporate sustainability is a heuristic multi-criteria approach raises the question of how the different characteristics relate to each other. One attempt to conceptualize these links is the three-dimensional sustainability portfolio which relates three axes for economic, environmental and social performance to each other (Figure 5.2). Better known is the eco-efficiency portfolio which is the equivalent two-dimensional matrix combining environmental and economic issues (for a more detailed discussion of the eco-efficiency portfolio see Ilinitch and Schaltegger, 1995; Schaltegger and Burritt, 2000). The portfolio can be applied at the level of a product, a firm or a nation but for the corporate level concern is mostly with eco-efficiency of individual companies.

Economic performance A

B

0

C Eco-efficiency line

Social performance Environmental performance Figure 5.2 Relationship of different corporate sustainability dimensions to each other (Schaltegger and Burritt, 2000)

196

The international yearbook of environmental and resource economics

In Figure 5.2, economic performance is measured on the vertical axis while environmental performance is measured on the horizontal axis and the social performance dimension of sustainable development is notionally included as a third measure on the z-axis. The current position, concentrating on economic and environmental performance, is defined as point 0, observable in the middle of Figure 5.1. The eco-efficiency line represents the eco-efficiency frontier as a border between eco-efficient and inefficient development. Movements to the right of the diagonal ‘eco-efficiency line’ indicate that the ratio between economic and environmental performance has improved, as has economic-ecological efficiency. However, sustainable development is characterized by movements towards the upper right quadrant of the diagram as well as an improvement in the social dimension (arrow A plus a movement towards better social performance). ‘Strong’ ecologically sustainable development does not allow an increase in environmental impacts, i.e. movements to the left of the vertical line through point 0. The same argument applies to ‘strong’ economically sustainable development, which must be above the horizontal line through point 0. Weak sustainability allows development above the eco-efficiency line, with substitutability between natural and economic capital being permitted. Finally, a movement indicated by an arrow in Figure 5.1 represents the time dimension of sustainable corporate development. In summary, the development can distinguish between, for example, a strong (arrow A plus a movement towards better social performance) and a weak improvement in corporate sustainability (arrows B and C). To focus on a strong improvement in corporate sustainability can be seen as a mutually beneficial ‘no-regrets strategy’ for corporate environmental protection and social involvement because measures taken result in an improvement in economic, environmental and social performance. Weak improvements in corporate sustainability imply either an improving economic situation that is traded off against lower environmental and/or social performance, or an improving environmental and/or social performance that is traded off against a lower economic level of performance, etc. An improvement in corporate sustainability is normally termed a ‘triple win’ situation for economic, environmental and social performance and is promoted by government and business and accepted by non-governmental organizations (NGOs) involved in partnerships with government or industry. Yet, the integration of environmental and social issues into business thinking goes beyond a focus on ‘win-win’ and may improve eco-efficiency or socio-efficiency without necessarily enhancing sustainability (similar to Reinhardt, 1999). Although the sustainability portfolio serves as a good visualization of some dimensions or characteristics of corporate sustainability, it also hides other dimensions such as orientation towards the future

197

Corporate sustainability

and intergenerational equity (which could either be considered as part of the social dimension or as part of the time aspect represented by an arrow), or the extent to which the company contributes towards sustainable development of the economy and society as a whole, participation and the integration of stakeholders into management processes. The sustainability portfolio furthermore does not address the question of functional dependence, that is how performances in different dimensions of corporate sustainability influence each other. How Do the Dimensions of Corporate Sustainability Influence Each Other? A model that attempts to describe the functions of and influences between environmental and/or social performance and economic performance is shown in Figure 5.3 (Schaltegger and Synnestvedt, 2002). This model also provides a framework to structure empirical research on corporate sustainability.

Economic success A

ES* ES0

B E

F

D

C

0 ESP*

ESP1 ESP0 Voluntary corporate environmental and/or social activities

ES = Economic success, ESP = Environmental and/or social performance

Figure 5.3 Potential relations between corporate environmental and/or social performance and economic success (similar to Schaltegger and Synnestvedt, 2002)

198

The international yearbook of environmental and resource economics

Two ‘schools of thought’ have emerged about the link between environmental and/or social performance and economic performance (movements in the direction of the axes in Figure 5.2). Some feel that voluntary adoption of corporate environmental protection and/or social engagement conflicts with the core business objective of increasing economic success (e.g. Friedman, 1970; Walley and Whitehead, 1994; Cohen et al., 1995). This sometimes assumed purely negative relation is shown in Figure 5.3 by line ES0-E-F-D. Beginning at a certain level of economic success (e.g. ES0 as the starting point of voluntary environmental and social activities after complying with regulations), every environmental protection activity or social engagement (moving to the right in Figure 5.3) will reduce economic performance. The negative marginal impact on economic success can be expected to increase as these voluntary non-economic activities are increased. Below point D in Figure 5.3, with an economic success of 0 and the amount of environmental protection ESP0, the company becomes unprofitable. Others (for example Porter, 1991; Porter and van der Linde, 1995a, 1995b; Porter and Esty, 1998; see also Schmidheiny and BCSD, 1992), believe that environmental protection and social engagement practised by a company is not only economically sustainable but that it even has a beneficial effect on its economic performance (e.g. movements from ES0 into the half-circle ES0-A-B, whereas point A represents the optimum position and, thus, ES* in Figure 5.3 is achieved with the amount EP* of environmental protection). However, realistically nobody would believe that an indefinite amount of environmental and social activities would continue to improve economic performance. Hence, net marginal benefits from social and environmental activities will decrease (picking the ‘low hanging fruit’ first) and sooner or later increased environmental and social efforts will represent net costs (after point A in Figure 5.3). Most industries with a certain level of public exposure will consider economic incentives (e.g. to reduce risk, create reputational gains, higher sales margins, etc.) to undertake some degree of voluntary environmental and social activities when there is both an economic and environmental gain (e.g. Arora and Cason, 1996; Khanna and Damon, 1999; Khanna, 2001; Khanna and Anton, 2002). Nevertheless, there may be several reasons for the different views on the relationship between environmental and social performance and economic success. The perception of a pure tradeoff may stem from a feeling that in competitive industries there are economic disincentives towards voluntary corporate environmental and social engagement. In particular, ignorance of available opportunities may be encouraged by the perception of high information costs of identifying them – something which corporate sustainability accounting academics seek to address (Schaltegger and Burritt, 2000).

Corporate sustainability

199

In contrast, voluntary involvement in improving corporate environmental and social performance is guided by economic incentives such as: anticipation of tighter regulation, favourably influencing the social legitimacy of the corporation (Deegan and Rankin, 1996), and reducing perceptions of corporate risk thereby lowering the cost of capital. The managerial challenge includes: (1) identifying, developing and ranking those corporate environmental and social activities which promise high potential economic reward, choosing the optimal level of environmental and social performance which results in the highest economic success (ES*), and choosing the best ranked environmental and social activities, and (2) obtaining that level of environmental and social performance at the lowest possible cost in order to realize the maximum of economic success. It can thus be assumed that the suggested functions not only represent a perception of the link between environmental and/or social performance and economic performance but that they represent qualitative differences in managerial decisions as well as the position a company has already achieved (e.g. if the company is already at ES* the function is descending). One answer to the question of when it pays to be green and social is, thus, that it depends on the kind of environmental and social management which is established in the company. The upper curve represents the situation where voluntary environmental and social management is institutionalized and how well management takes the specific situation of the company into account. This is accompanied by an awareness of both cost-efficiency and market gains in the context of corporate environmental and social activities. In contrast, the lower curve represents management perceptions characterized by costly reactions to problems, such as adoption of end-ofpipe-technologies and solving social disputes once they have occurred. Because a company on the upper curve in Figure 5.3 manages its environmental and social issues in a better economic way than a company on the lower curve, the slope A-B-C is flatter than E-F-D because of the lower marginal costs (a) of environmental and social engagement for the second company. The curve might furthermore be influenced by other, external and internal factors such as the introduction of new regulations, industryspecific circumstances facing the company, technologies, the size of the company, etc. (e.g. see Gabel and Sinclair-Desgagné’s (1993) discussion of organizational failure). In reality the functions described may not be as smooth as shown in Figure 5.3. Fixed costs of environmental protection and social arrangements can cause ‘steps’ in the cost function. The same may occur for revenues, for example because of sudden shifts in demand when a threshold value for environmental and social performance is exceeded (e.g. because of an improved image, certification or product labels, etc.).

200

The international yearbook of environmental and resource economics

Several factors may lead to a shift of the curve to the right (dotted curve in Figure 5.3). The development of new technologies and organizational arrangements reduces the marginal costs of environmental and social activities over time; regulatory changes and changes in consumer preferences increase the market gains of good environmental and social performance, etc. The model proposes a wide range of possible economic outcomes from voluntary environmental and social engagement between curve ES0-E-F-D and curve ES0-A-B-C in Figure 5.3. In any empirical study the population of firms may be somewhere in the space between the two curves. It is thus not surprising that empirical studies – which are always based on some specific environmental or social indicator or indicator set – lead to very different results (Cohen et al., 1995; Hart and Ahuja, 1996; Adams, 1997; Cordeiro and Sarkis, 1997; Konar and Cohen, 1997, 2001; Day, 1998; Reinhardt, 1999a, 1999b; Dowell et al., 2000; King and Lenox, 2000, 2001, 2002). Picking different samples from the wide range of companies spread between the curves may lead to conflicting results about the relationship between environmental and social effort and economic performance unless the samples are very large. Furthermore, the curves may vary both over time as well as between countries and industries that provide the different sets of companies being analysed. The model suggests several closely related conclusions. First, environmental and social performance can vary for a given level of economic success. Point B in Figure 5.3 reflects the same economic success as point ES0. The difference is that one level of economic success reflects environmental and social ignorance (or inaction) whereas the other level represents a higher degree of environmental and social responsibility. Second, the economic effect of corporate environmental protection can vary for a given environmental and social performance level. For instance at ESP* economic success can vary between A and E, where A represents the situation where the potential economic gains from environmental and social improvement are fully realized, whereas E represent a situation where inefficient environmental and social management incurs a loss of economic performance as compared to the initial level ES0. Third, the correlation between economic and environmental or social performance, or in other words, the question when it pays voluntarily to exceed minimum environmental and social standards, does not only depend on external company variables such as compliance with regulation. The company’s position between ES0-E-F-D and ES0-A-B-C is not given or fully determined externally but can be influenced by management. The company’s position thus also depends substantially on management perceptions of the benefits from voluntary engagement with environmental and social aspects of the business. Environmental and social management

Corporate sustainability

201

becomes an additional explanatory variable of economic success. As a result the question what kind of environmental and social strategies and management represent good corporate sustainability management becomes crucial. So far corporate sustainability, its main characteristics and the relations and influences between them have been outlined. Allowing for the possibility that there is potential for corporations to increase economic performance through their voluntary environmental and social activities, the model outlined provides a framework of the fundamental economic reasons why management engages with corporate sustainability. However, the question may be raised as to what the detailed motives are for managers to implement corporate sustainability management.

4

WHY IS CORPORATE SUSTAINABILITY AN ISSUE FOR MANAGEMENT?

Economic and business literature as well as discussions with managers who are involved with corporate sustainability reveal that there is not just one reason, or one main reason, why managers address sustainability in the corporate context. Instead, reasons are interrelated. This section provides a short overview of the main reasons which range from pure business orientated reasons to those that are macro-orientated political and ethical reasons: ●

●

Legal compliance: Regulations in most countries require the consideration of some environmental and social aspects of business activity. Wider social, environmental and sustainability issues are more frequently being addressed by regulations and politicians thereby forcing managers to consider sustainability management. However, legal compliance is not sufficient to explain why management is dealing with sustainability as the strong increase of various voluntary approaches to environmental protection shows (e.g. Segerson and Na, 2000). Increasing competitiveness: The hypothesis that strict environmental regulations do not inevitably hinder competitive advantage against foreign rivals, but that they can enhance competitiveness, has been brought into discussion by Porter (1991, p. 96). The Porter hypothesis may hold true in three main cases (Gabel and Sinclair-Desgagné, 1999, pp. 93f.). First, strict environmental regulations can create a market for companies specialized in offering products and services to protect the environment. These companies will be more competitive through an early mover advantage, experience and knowledge

202

●

●

The international yearbook of environmental and resource economics

created, if other nations adopt the regulations and market demand is generated abroad. Second, relative competitiveness of some companies can be enhanced by the fact that other, e.g. foreign, companies are slower or less efficient in adapting their products and production technologies to new market conditions because they are not forced to do so early enough in their home market. Third, companies may not have identified ‘low hanging fruit’ so far. Regulations or the development of new management information methods (see below) may make them realize the potential economic gains from environmental and social management. The basic argument of the Porter hypothesis can be interpreted in a wider sense to indicate that the company which deals with corporate sustainability early enough – perhaps as a result of specific demand or regulations or because of management decisions based on some other reasoning as discussed below – will profit from its experience and built up know-how when demand or regulations change the setting for national competitors or in international markets. Managing business risk: Business risks stemming from contaminated land, leaking landfills, health issues at the workplace, child labour or discrimination, etc. are reasons for adopting risk orientated sustainability management. Business risks may be related to legal risks such as (contingent) environmental or social liabilities, technical risks such as leakages in production facilities, social risks resulting from stakeholder reactions to corporate environmental impacts, natural risks such as floods, or to general economic risks. Furthermore, the reduction of insurable risks can result in reduced costs. On 24 March 1989 the oil tanker Exxon Valdez ran aground in Prince William Sound in Alaska. The oil spills caused reactions from the stock market and resulted in cleaning costs and juristitial claims of over 15 billion US$ (e.g. Economist, 1994a, 1994b, 1994c; Vaughan, 1994; Schaltegger and Burritt, 2000, p. 181). Sustainability management is often designed as risk management which should prevent situations like with Exxon Valdez. Personal risk and reputation of managers: Due diligence checks to protect senior management and the board from risk of liability are linked with the management of business risks. Engaging with sustainability, however, may also provide an attractive platform for managers to position themselves in media and among peers and friends as well as to signal responsibility and to create trust among employees. From a public choice perspective the personal incentives of managers to protect the environment may be an important and neglected driver of corporate sustainability activities.

Corporate sustainability ●

203

Cost reduction: Environmental management and environmental accounting literature argues that incomplete information resulting from historically developed information management methods, that have not been adapted for the increasing economic relevance of environmental and social issues, leaves management ignorant about win-win solutions. An increasing body of more recent environmental accounting literature (e.g. Bennett et al., 2001; Schaltegger and Burritt, 2000) deals with new methods of environmental management accounting and shows that with changed costing methods production costs and product related costs can be reduced through the dematerialization of production processes. The basic idea is to calculate the costs of environmental impacts caused by excess material and energy use. Thus the cost-efficient reduction of material throughput and energy use in production and material use for products results in increased eco-efficiency as a consequence of lower resource consumption. This reduces costs of purchase, storage, amortization of facilities, etc. as well as the environmental impacts related to material flows. These accounting approaches are called activity based material flow cost accounting, zero loss accounting, etc. Many companies have experienced that the application of environmental accounting methods can help to identify cost reduction potentials. Cadbury Schweppes Pty Ltd, for instance, realized on the basis of its cost potential analysis in its plant in Ringwood, Victoria, annual cost savings of $780,000 with cleaner production measures (Environment Australia, 1999).

●

●

Integration of parallel activities of shared service units: Multinational companies facing a large number of social and environmental regulations and demands have for years built up shared service units through various departments and groups that address sustainability issues such as waste water, environmental liabilities, air pollution control, workers’ health and safety, gender issues, product development, environmental reporting, environmental costs, etc. A coordinated effort to establish corporate sustainability management can thus be motivated by the attempt to reduce costs of parallel and uncoordinated activities (e.g. Figge et al., 2002). Business opportunities: Sustainability issues can be a driver for competitive advantage, not only by lowering production costs as a result of more material and energy-efficient production processes, but also through corporate and product differentiation in the market (Dyllick, 1999). Business opportunities for sustainability related product differentiation can be driven by higher sales revenue or by

204

●

●

●

●

The international yearbook of environmental and resource economics

increased contribution margins caused by a greater willingness by customers to pay. Examples can be found in some European food markets (e.g. Switzerland, Sweden, Germany) and in niche markets of the textiles industry. Certifications and labels can play a role (e.g. Karl and Orwat, 2000). Reference point for innovations: Dealing with sustainability issues can be used as a reference point for the creation of business innovations. Approaching business processes and products from all perspectives on sustainability provides managers with a framework to change their focus. This in turn may help to identify innovation potential. Such business innovations can have an environmental technology character such as wind power generators, a technical character like powder coating and solvent-free painting, a market differentiation character such as natural textiles or they can have a more organizational character such as car sharing businesses. Increasing shareholder value: The question of how environmental management can increase the shareholder value of a company has been discussed under the concept of ‘environmental shareholder value’ (Schaltegger and Figge, 1997, 2000; WBCSD, 1997) and has been tested empirically for German industry (Wagner, 2002). The basic logic of this approach is – in contrast with the argument about increased competitiveness – that environmental and sustainability management can influence the value drivers of shareholder value, that is, investments in fixed and current assets, profitability, sales revenue, duration of the value increase and costs of capital. These value drivers are linked with each other through the net present value related formula (Rappaport, 1987; Copeland et al., 1993). Depending on how the company manages its environmental issues these value drivers will influence behaviour in different ways. Improving corporate reputation and brand value: Reputation is one of the most important non-material (or intangible) assets of a company, has a long-term character and represents the sum of all perceptions of corporate stakeholders. These perceptions are a result of facts of corporate activities and products and of perceived facts. The value of reputation is mostly recognized in mergers and acquisitions as so called ‘goodwill’ (e.g. Shenkar and Yuchtman-Yaar, 1997). Corporate reputation is measured and published regularly using different indexes (e.g. Fombrun, 1996). Among the main measurement components is social and environmental responsibility which reflects the activities of management to improve corporate sustainability. Maintaining legitimacy and ‘social license to operate’: Stakeholders can influence corporate success in more or less direct ways and in

Corporate sustainability

●

●

●

205

both directions through boycotts, by imposing or enforcing regulations, by promoting market trends, etc. One reason to address corporate sustainability is to secure social acceptance and legitimacy of operations and to increase the motivation of employees, suppliers, customers, the local community, public administrations, etc. This argument about securing social acceptance and legitimacy of operations and products is often summarized in corporate practice as maintaining the ‘social license to operate’ by behaving as a ‘good corporate citizen’. The response to social needs and demands of stakeholders may, however, also be driven by ethical considerations. Promotion of self-regulation and influence of the future direction of regulation: Companies with a large market share, industry associations or a group of companies may be enticed to organize the management of sustainability issues themselves instead of facing social pressure and politically enforced regulations which may cause higher costs than necessary (e.g. Gunningham et al., 1998; Lévêque and Nadaï, 2000; Cansier, 2001; Burritt, 2002). To deal with corporate sustainability and to show corporate responsibility can be used to signal to society and politicians that there is no need for further regulation. On the other hand, some leaders in sustainability and in environmental technology innovations may be interested in promoting specific new regulations and in shaping future regulations by demonstrating that companies could provide greater contributions to sustainable development. Role of corporations as drivers of economic and social development: Management, public administrations and governments may recognize the role of corporations as important drivers of economic and social progress. This may be a motivation for public–private partnerships but also for industry associations and some progressive companies to engage with corporate sustainable development. The view of companies as social actors can be linked with the promotion of self-regulation and the attempt to reduce external control by regulators but it may also accord with the vision of shaping markets and creating business opportunities. However, the goal of exerting influence on society and the economy may also be seen in the light of attempts to increase corporate and personal power and the status of management. Moral commitment of managers and individual employees: In addition to the business reasons cited above, intrinsic, personal motivations of business leaders who believe that attempting to achieve corporate sustainability is ‘the right thing to do’ can be a strong driver for corporate sustainability. This phenomenon whereby personal values

206

The international yearbook of environmental and resource economics

of company leaders have a large influence on corporate activities can be seen very often among entrepreneurs and is discussed in the sustainability context under notions of environmental entrepreneurship, ecopreneurship or sustainable entrepreneurship (Schaper, 2003). Sustainable entrepreneurship is not restricted to, but is more frequent among, family businesses and small- and medium-sized companies that are owned by the managers. Managers who are motivated to deal with corporate sustainability – no matter what their reasons for doing so – have to find pragmatic approaches to implementation. To be effective corporate sustainability has to be linked with corporate strategies, that can be broken down within the organization and operationalized by management.

5

STRATEGIES FOR CORPORATE SUSTAINABILITY

A general framework of corporate management distinguishes normative, strategic and operational levels of management approaches and processes. Given that the vision of corporate sustainability has a strongly normative character the next step is to translate it into corporate and business strategies. In the sustainability context three central strategies can be distinguished: general sustainability strategies, competitive strategies and risk strategies. As a first general step towards breaking down the vision of sustainable development three ‘sustainability strategies’ have been proposed. The first two are directly relevant to corporations, whereas the last strategy would work indirectly through consumer pressures in the marketplace: ●

●

The first is labelled the efficiency strategy sometimes even called the ‘efficiency revolution’. The idea is to improve the ratio between inputs used to produce a desired output thereby becoming more efficient by using fewer materials and less energy. This approach has been encapsulated in the term ‘eco-efficiency’ (Schaltegger and Sturm, 1990) and promoted strongly by the World Business Council for Sustainable Development. A different angle on sustainability is taken with the consistency strategy (Huber, 1995, pp. 39–42) which aims at changing the quality of material and energy flows by replacing toxic, persistent and other non-natural substances with substances that have been part of natural ecological material cycles for centuries. For example, one element of this strategy is to replace toxic with carbon based materials. This can

Corporate sustainability

●

207

reduce health and environmental impacts if the lower relative environmental impact is not overcompensated by higher quantities of carbon dioxide. The sufficiency strategy (Huber, 1995, pp. 39–42) addresses sustainability by asking if consumers really need a certain product or service for their well-being. The sufficiency strategy addresses the needs of people and whether certain functions can be served with fewer products or shared product use, thereby curtailing unnecessary use of resources. It is a strategy driven by consumer demand, but, it can also be addressed by companies for instance in a simpler design of products.

These basic sustainability strategies have an environmental bias, even though social and economic aspects are involved. They are often used as a reference point for sustainable product and business innovations. However, from an economic and business perspective more integral are environmental and sustainability orientated approaches to corporate competitive strategies. Given that sustainability issues are not just limited to market processes but that they also exert a strong influence on political actions, media and social developments, competitive strategies in the sustainability context go beyond corporate activities in the market. Thus, corporate competitive strategies that explicitly refer to sustainability may have a broader societal orientation as well as a market orientation. Furthermore, a distinction can be made between defensive and offensive strategies, the latter being proactive. The combination of these orientations provides four possible sustainability strategies with reference to competitiveness (analogous to Dyllick, 1999): ●

●

The sustainability market buffering strategy is based on defensive corporate behaviour and is orientated towards environmental impacts and societal influences that threaten existing businesses. The goal of the company is to influence political and social developments through public relations activities and lobbying in order to reduce the legal and social pressure for more social and environmental corporate action. As with market development strategies buffering strategies require a stakeholder analysis identifying particular interests of the stakeholders, their influence on the company and its market as well as the management of stakeholder relationships. Sustainability accounting information can be used to manage and legitimize these relationships with stakeholders (Deegan, 2002). The sustainability cost strategy is directed towards the market and internal company processes. This type of competitive strategy takes

208

●

●

The international yearbook of environmental and resource economics

environmental and social demands and pressures as given and tries to meet them as efficiently as possible. Furthermore, this strategy focuses on approaches to reduce costs of neglected environmental and social engagement (Schaltegger and Burritt, 2002). The sustainability differentiation strategy is based on proactive behaviour and is orientated towards the market. Its objective is to create economically rewarding innovations on the basis of meeting social and environmental goals of potential and current customers. The success of this strategy depends very much on its innovation potential, the specific knowledge of management about sustainability issues and the company’s public reputation (Schaltegger et al., 2003). The sustainability market development strategy has a social and proactive character as opposed to the buffering strategy which is defensive in kind. With a market development strategy the company attempts to shape preconditions to create and enlarge areas where sustainable products are more competitive in the market. It focuses on changing regulations, political programmes and public perceptions influencing the market in order to create more favourable market conditions for selling social and environmental goods. Market development strategies thus attempt to improve the market conditions for sustainable products which is contrary to the goal of buffering strategies trying to preserve current market conditions.

With improved knowledge and skills relating to corporate sustainability, management is enabled to identify more opportunities for creating business success by addressing environmental and social issues. This may motivate management to change its competitive strategy from market buffering to differentiation and market development. The cost strategy becomes an underlying precondition to secure a competitive cost structure. Securing competitiveness requires management activities and information relating to many interrelated spheres of business influence such as political, sociocultural, legal, economic, ecological, and technological (Schaltegger et al., 2003). For instance, competitive strategies with a societal orientation address legitimacy, the legal status of operations and freedom of action in the specific political context. Within the framework of a specific sustainability strategy a number of companies are currently coming to grips with the task of the internal operationalization of corporate sustainability or partial aspects of it. For example, a number of corporate legitimacy strategies have been identified – for maintaining, gaining and extending, repairing and defending corporate

Corporate sustainability

209

legitimacy (Lindblom, 1994). The main focus of activities in research and practice seems now to be on the integration of the three contextual challenges in management concepts, the metrics of establishing indicators for measuring certain aspects of corporate sustainability, mainly the ecoefficiency aspects, and of establishing further operative tools to implement corporate sustainability. As discussed above corporate sustainability management represents a pluralistic approach in the flux of development. Nevertheless, economists and managers are interested in answering the question as to whether the concept or parts of corporate sustainability have produced empirically significant results in practice. To this end, the next section examines empirical research on corporate sustainability.

6

EMPIRICAL APPROACHES TO MEASURING CORPORATE SUSTAINABILITY

So far there have been no significant attempts to operationalize the measurement of the wider interpretation of corporate sustainability in all its major facets. The focus is mainly on two or sometimes all the three contextual dimensions of sustainability, considering social, environmental and economic issues. If considered to be an integrative approach, moves towards corporate sustainability cannot be measured by simply adding up results of eco-efficiency and socio-efficiency but need to be operationalized through a single, integrative measure. One approach to a combined or integrative measure of corporate economic, environmental and social performance is to choose a common denominator such as money for all three dimensions rather than trying to create a new measure of sustainability (e.g. Figge and Hahn, 2002). Such an approach can be useful for decisions based on highly aggregated information. From a pragmatic point of view, however, there is no doubt that using various indicators (e.g. Callens and Tyteca, 1999) and looking at the links between environmental and economic goals and/or between social and economic goals is preferred in practice because it can be handled better. Given the problems of operationalizing social performance it is not surprising that most research focuses on links between environmental and economic performance measures (e.g. Tyteca, 1996, 1997; Wagner et al., 2001). In the relevant literature and corporate practice, approaches creating operational eco-efficiency measures, such as CO2 per unit of output, are common and have been accorded strategic as well as operational management relevance.

210

The international yearbook of environmental and resource economics

Apart from the question of how corporate sustainability could be measured and what measurement might be useful an investigation of the link between environmental or social performance and economic performance requires greater thought about the characteristics of the link, the influence of the type and quality of environmental and social management on economic performance and how this could be modeled. The empirical statistical and econometric studies can be classified broadly into three groups (Jaffe et al., 1995; Day, 1998; Wagner et al., 2001): event studies assessing market responses after a positive or negative environmental event (e.g. Cormier et al., 1993; Hamilton, 1995; Klassen and McLaughlin, 1996; White, 1996; Jones and Rubin, 1999), model portfolios of environmental leaders and laggards (e.g. Cohen et al., 1995), and multiple regression analysis assessing the influence of different factors among environmental aspects (Hart and Ahuja, 1996; Johnson, 1996; White, 1996). No specific relationship has prevailed in empirical studies so far (Wagner et al., 2001; Schaltegger and Synnestvedt, 2002). Apart from different questions asked or measures taken (Wehrmeyer and Tyteca, 1998) and methodological issues (Jaffe et al., 1995), such as the lack of statistical data, its low quality, or the fact that environmental and sustainability data are often only available for short time periods (e.g. Konar and Cohen, 1997), the model discussed in Section 3 shows that no automatic or given relationship can be expected between social and environmental performance on one hand and economic performance on the other. It is thus not surprising that many empirical studies conclude that no systematic relationship exists between environmental and economic performance (e.g. Adams, 1997; Cordeiro and Sarkis, 1997; Day, 1998). At a given level of environmental or social performance different levels of economic performance are possible depending on the kind and quality of sustainability management the company conducts, and vice versa; with a given level of economic performance different levels of social and environmental performance can be achieved depending on how efficiently resources are used. The proposed framework furthermore implies that empirical studies should not just correlate two data sets representing environmental or social performance and economic performance, but rather investigate the effect of different environmental and social management approaches on economic performance. Two research approaches to investigate the kind of sustainability management that results in an improvement of both environmental and social as well as economic performance are possible (Schaltegger and Synnestvedt, 2002). A first research strategy is to draw the focus away from the typical large sample statistical research approach towards more in-depth case studies (see Burritt, 2004). The case study based approach tries to test the

Corporate sustainability

211

practical relevance of theoretically determined factors driving the economic effects of corporate sustainability management by analysing specific companies. Complex processes and links may sometimes be investigated better in case studies as they provide greater insight into the causal mechanisms. However, case studies, most often based on small samples with high individual data quality and company-specific information, may result in widely different answers. The second strategy, statistically based research, attempts to estimate the ex-post economic impact of good sustainability management on the basis of more or less large data sets (e.g. Wagner, 2003). Initially, the economic effects of different sustainability management concepts applied to comparable companies identified as having the same level of environmental or social performance could be compared. Use of this strategy requires that groups of environmental management approaches (e.g. market orientated or process orientated environmental management approaches, etc.) and the relevant factors driving the economic effects of corporate environmental management (which may again differ according to the environmental management approach chosen) be distinguished. Measures of these different approaches and factors must then be obtained and the respective data sets compiled in order to carry out appropriate statistical analysis. The fact that other factors, such as market conditions, the regime of general business regulations, available technology, the development of the world economy, etc., influence economic success, underlines the necessity to control for other variables by isolating their effects in the analysis. Then the environmental and economic effects of a given environmental management concept applied in comparable companies could be investigated. Companies with the same environmental management concept (e.g. environmental management accounting, or ISO 14001) but different levels of environmental and economic performance could be compared (e.g. Wagner, 2003). The results of this analysis could be used to compare the actual (realized) effects of different management approaches. This kind of analysis provides information about the best observed practical result of each management approach. This requires that different sustainability management concepts be characterized and distinguished very clearly. Only then could comparisons be made on an empirical basis. Testing these effects requires statistical approaches that handle more than one dependent variable. Relevant statistical methods would include structural equation modeling or regression analysis applied on equation systems. As a consequence the availability of data sets and, most of all, of data on environmental and social performance, are an important prerequisite for empirical analysis of corporate sustainability. Such data sets are

212

The international yearbook of environmental and resource economics

being compiled by external company sustainability rating organizations and the sustainability research groups of financial service providers (e.g. Sustainability Asset Management for the Dow Jones Sustainability Index, the Sarasin Sustainability Investment Group, or Reputation Measurement Pty Ltd). Furthermore, it can be expected that a more standardized set of indicators will be published by companies with the increasing application of the internationally recognized sustainability reporting guideline of the Global Reporting Initiative (GRI). A different path of empirical research deals with the question of what are the drivers and stakeholders which make management deal with corporate sustainability. An uncountable number of publications address this issue but very little empirical work has been conducted. A useful source in this respect is the European Business Environmental Barometer, a survey based on a questionnaire to managers which has been conducted every two to three years since 1993 it changes the number and combination of countries and research institutions involved (see e.g. Ytterhus et al., 1994; Belz and Strannegård, 1997; Wagner and Schaltegger, 2002). One main result of all surveys is that national and international regulators seem to exert more influence on corporations to deal with sustainability than customers, investors or NGOs (e.g. Ytterhus et al., 1994, pp. 24–6; Belz and Strannegård, 1997, pp. 147–52; Wagner and Schaltegger, 2002, pp. 5–8). This is also supported by an empirical analysis for the US chemical and petroleum industry (Hoffmann, 1996). However, some differences exist between nations and the importance of market actors ‘pulling’ management towards more sustainability is increasing over time in comparison with the ‘push’ factors of regulation. In the first surveys, the actual operational environmental measures taken in corporate practice used to reflect cost driven activities focusing on cost reductions rather than compliance orientated activities (e.g. Belz and Strannegård, 1997, pp. 154–5). A more recent analysis for Germany indicates some changes towards more market orientated corporate environmental activities such as eco-marketing and the development of green products (Wagner and Schaltegger, 2002, pp. 15–17). The fact that ever more financial institutions and rating agencies are considering sustainability in their financial analysis and in their products (e.g. sustainability funds) as well as the strong growth of funds which are claimed to be managed under sustainability criteria may be an indication that investors are drawing more attention to sustainability issues. As a conclusion the drivers seem to be compliance and costs rather than market opportunities though indication suggests that market issues might gain in importance. Empirical analysis and international comparison of these factors is a field with many open research opportunities.

Corporate sustainability

7

213

OUTLOOK AND DISCUSSION OF THE FUTURE DIRECTION OF CORPORATE SUSTAINABILITY

Given the many challenges discussed for corporate sustainability it is not a surprise that different aspects are emphasized by different authors, corporations and stakeholders. The main directions for discussion about the direction of the corporate sustainability debate are the claim that corporate sustainability should go beyond the business case, that it should address issues beyond the environmental aspects and that corporate sustainability should develop its role as a change driver in the economy and society as a whole: ●

●

●

Beyond the business case? In contrast with corporate practice various academics in business administration claim that corporate sustainability is more than good business management considering environmental and social issues. For example Dyllick and Hockerts (2002) distinguish three cases for corporate sustainability: the business case, the natural case and the societal case. They argue that corporate sustainability should also embrace approaches where environmental goals (natural case) or social goals (societal case) have priority over economic goals (eg. line C in Figure 5.2). Although these approaches provide valuable points of reference they lack the necessary business reality in order to make them economically sustainable, although they do reinstate the notion of management of public goods. Nevertheless, from a precautionary point of view they still have economic merit and can become relevant with changing market conditions, customer demands, etc. as well as for non-profit organizations. Beyond greening: A major development of the corporate sustainability debate emphasizes that sustainability management is more than environmental management and should explicitly deal with the social dimension (e.g. Gladwin et al., 1995). Issues of corporate social responsibility, corporate citizenship, stakeholder dialogue and participation, partnerships, networks and the contribution of the company to civil society are addressed. Beyond the company and its stakeholder network? One claim is that corporate sustainability should not be limited to the company’s organization development and the consideration of the direct stakeholder network but should in addition think about the possible contribution of the company to sustainable development in general and develop approaches to strengthen the role of company management as a sustainable development actor in the economy and the society.

214 ●

The international yearbook of environmental and resource economics

Operational implementation: One main challenge in management research as well as in corporate practice is how to implement corporate sustainability most successfully. As a consequence the growing body of literature dealing with this issue is evolving fast. The next decade may show which concepts and tools of corporate sustainability will prevail in the contest.

Results of this contemporary review of corporate sustainability reveal that everything remains questionable, or contestable. This will disappoint if the aim is to discover a well-established body of literature that reveals indisputable empirical evidence about the links between corporations and their combined or integrated economic, environmental and social performance. Such evidence remains in the process of discovery. Taking stock of the current situation it should be reiterated that: ●

●

●

●

The actions of corporate managers, in the name of their corporations, have a pervasive influence on economic, environmental and social well-being and, in such a context, corporate sustainability is likely to remain a critical sub-goal of general sustainability (Section 2). Corporate sustainability faces a range of challenges, the most important of which is the need for development of integrative tools and measurements that reflect interrelatedness, interactions, and multicriteria based decisionmaking within the corporate sustainability management milieu (Section 3). The challenges for corporate sustainability include development of a mental map for business managers that takes economic thinking beyond the economic, sociological thought beyond the social, and ecological thinking beyond ecology. Static, marginalist equilibrium analysis (see Figure 5.2) provides an initial conceptual model for understanding potential mixes of economic, social and environmental activities of sustainable corporations, but does not necessarily capture the richness of management motivations for addressing corporate sustainability; for example, intrinsic rewards from power and status, the social license to operate and ethical commitment to ‘do the right thing’ in terms of the sustainability goals of politicians (Section 4). Pragmatic solutions are needed. As a richer understanding of the possible set of business and personal motives for managers to engage in corporate sustainable development processes becomes apparent through fieldwork and case studies, potential defensive and offensive strategies relating to efficiency, consistency and sufficiency have been generated to help guide

Corporate sustainability

●

215

management towards the development of operational forms of corporate sustainability (Section 5). At this time there is little systematic empirical evidence available to demonstrate the success of measurement tools and indicators being developed to assess corporate sustainability performance (Section 6).

Corporate sustainability management is a fast evolving management discipline and many companies are developing and applying new tools and organizational approaches to deal with environmental and social issues. Given that the profitability orientated corporate form and environmental and social goals in society and politics are likely to continue, corporate sustainability remains an important issue for the future. As a consequence, research based on a range of methodologies that continue to help discover and understand the relationships between different sustainability characteristics of corporations is essential.

REFERENCES Adams, R. (1997), ‘Linking financial and environmental performance’, Environmental Accounting and Auditing Reporter, 10(2), 4–7. Arora, S. and T. Cason (1996), ‘Why do firms volunteer to exceed environmental regulations? Understanding participation in EPA’s 33/50 program’, Land Economics, 72(4), 413–32. Ayres, I. and J. Braithwaite (1992), Responsive Regulation: Transcending the Deregulation Debate, New York: Oxford University Press. Bammert, M. and R. Haßler (2001), ‘Environmental performance: sectoral ratings and environmental protection in the home electronics industry’, Greener Management International, 33, 25–30. Bartelmus, P. (1994), Environment, Growth and Development: The Concepts and Strategies of Sustainability, London: Routledge. Bartelmus, P. (1997), ‘Whither economics? From optimality to sustainability?’, Environment and Development Economics, 2, 323–45. Bebbington, J. (2001), ‘Sustainable development: a review of the international development, business and accounting literature’, Accounting Forum, 25(2), 128–57. Beder, S. (1997), Global Spin. The Corporate Assault on Environmentalism, Melbourne: Scribe Publications. Belz, F. and L. Strannegård (eds) (1997), International Business Environmental Barometer 1997, Oslo: Cappelen Akademisk. BMU and BDI (The Federal Ministry for the Environment, Nature Conservation and Nuclear Safety and Federation of German Industries), (eds) (2003), Sustainability Management in Business Enterprises. Concepts and Instruments for Sustainable Organisation Development, S. Schaltegger, C. Herzig, O. Kleiber and J. Mueller (authors), Berlin/Lueneburg: BMU/CSM. Braungart, M. and W. McDonough (2002), From Cradle to Grave. Remaking the Way We Make Things, New York: North Point Press.

216

The international yearbook of environmental and resource economics

Brundtland, G. (1991), ‘Sustainable development. The challenges ahead’, The European Journal of Development Research, 3(1), 32–41. Burritt, R.L. (2002), ‘Voluntary agreements: effectiveness analysis – tools, guidelines and checklist’, in P. ten Brink, (ed.), Voluntary Environmental Agreements: Process, Practice and Future Use, Sheffield: Greenleaf Publishing. Burritt, R.L. (2004) ‘Environmental management accounting: roadblocks on the way to the green and pleasant land’, Business Strategy and the Environment, 13(1), 13–32. Callens, I. and D. Tyteca (1999), ‘Towards indicators of sustainable development for firms: a productive efficiency perspective’, Ecological Economics, 28, 41–53. Cansier, D. (2001), ‘Free riding and voluntary agreement in the area of environmental protection’, Journal of Environment Policy and Law, 2, 209–39. Checkland, P. (1999), Systems Thinking, Systems Practice: Includes a 30-year Retrospective, Chichester: Wiley. Cohen, M., S. Fenn and J. Naimon (1995), Environmental and Financial Performance. Are They Related?, Nashville, TN: Vanderbilt University. Commission for the Future (1987), Our Common Future, The World Commission on Environment and Development, Melbourne: Oxford University Press. Commonwealth of Australia (1990), Ecologically Sustainable Development. A Commonwealth Discussion Paper, Canberra: Australian Government Printing Service. Cordeiro, J. and J. Sarkis (1997), ‘Environmental proactivism and firm performance: evidence from security analyst earnings forecasts’, Business Strategy and the Environment, 2, 135–55. Cormier, D., M. Magnan and B. Morard (1993), ‘The impact of corporate pollution on market valuation: some empirical evidence’, Ecological Economics, 8, 135–55. Daly, H. (1991), ‘Elements of environmental macroeconomics’, in R. Costanza, (ed.), Ecological Economics. The Science and Management of Sustainability, New York: Columbia University Press. Daly, H. (1996), The Economy Beyond Growth: Economics of Sustainable Development, Salzburg: Pustet. Daly, H. (2000), ‘Uneconomic growth: Some remarks on the economic theory and practice of globalization’, Political Ecology, 66, 15–19. Dasputa, S., B. Laplante and H. Wang (2001), ‘Inspections, pollution prices, and environmental performance: evidence from China’, Ecological Economics, 36, 487–98. Davis, D. and D. Gartside (2001), ‘Challenges for economic policy in sustainable management of marine natural resources’, Ecological Economics, 36, 223–36. Day, R. (1998), Beyond Eco-Efficiency: Sustainability as a Driver for Innovation, Washington, DC: World Resources Institute. Deegan, C. (2002) ‘The legitimising effect of social and environmental disclosures – a theoretical foundation’, Accounting, Auditing and Accountability Journal, 15(3), 282–311. Deegan, C. and M. Rankin (1996), ‘Do Australian companies report environmental news objectively? An analysis of environmental disclosures by firms prosecuted successfully by the Environmental Protection Authority’, Accounting, Auditing and Accountability Journal, 9(2), 50–67. Denning, K. and K. Shastri (2000), ‘Environmental performance and corporate behavior’, Journal of Economic and Social Research, 2(1), 13–38.

Corporate sustainability

217

Dowell, G., S. Hart and B. Yeung (2000), ‘Do corporate global environmental standards create or destroy market value?’, Management Science, 46(2), 1059–74. Dyllick, T. (1999), ‘Environment and competitiveness of companies’, in D. Hitchens, J. Clausen and K. Fichter (eds), International Environmental Management Benchmarks, Berlin: Springer, pp. 55–69. Dyllick, T. and K. Hockerts (2002), ‘Beyond the business case for corporate sustainability’, Business Strategy and the Environment, 11(4), 130–41. Dyllick, T., F. Belz and U. Schneidewind (1997), Ecology and Competitiveness, Munich: Hanser. Economist (1994a), ‘After Valdez’, The Economist, 18 June, p. 20. Economist (1994b), ‘Corporate liability. UnExxonerated’, The Economist, 20 August, pp. 57–9. Economist (1994c), ‘Insurers Get that Sinking Feeling’, The Economist, December, p. 9. Ekins, P., S. Simon, L. Deutsch, C. Folke and R. De Groot (2003), ‘A Framework for the Practical Application of the Concepts of Critical Natural Capital and Strong Sustainability’, Ecological Economics, 44, 165–85. Elkington, J. (1998), Cannibals with Forks: The Triple Bottom Line of 21st Century Business, Gabriola Island, BC: New Society Publishers. Endres, A. (2000), Environmental Economics, Stuttgart: Kohlhammer. Endres, A. and I. Querner (2000), Economics of Natural Resources, Stuttgart: Kohlhammer. Environment Australia (1999), ‘Cleaner production in confectionery manufacturing. Cadbury Schweppes Pty Ltd’, cleaner production case studies, Environment Australia. Figge, F. and T. Hahn (2002), ‘Sustainable value added. Measuring corporate sustainable performance beyond eco-efficiency’, CSM research paper, Lueneburg. Figge, F., T. Hahn, S. Schaltegger and M. Wagner (2002), ‘The sustainability balanced scorecard. Linking sustainability management to business strategy’, Business Strategy and the Environment, 11(5), 269–84. Fombrun, C. (1996), Reputation. Realizing Value from the Corporate Image, Cambridge, MA: Harvard Business School Press. Fombrun, C. and M. Shanley (1990), ‘What’s in a name? Reputation building and corporate strategy’, Academy of Management Journal, 2, 233–58. Freeman, E.R. (1984), Strategic Management: A Stakeholder Approach, Boston, MA: Pitman. Freimann, J. (2000), ‘Environmental management and information systems’, in H. Folmer and H. Landis Gabel (eds), Principles of Environmental and Resource Economics, Cheltenham, UK and Northampton, MA: Edward Elgar. Friedman, M. (1970), ‘The social responsibility of business is to increase its profits’, in G.D. Chryssides and J.K. Kaler (eds), An Introduction to Business Ethics, London: Chapman. Gabel, L. (1995), ‘Environmental management as a competitive strategy. The case of CFCs’, in H. Folmer, L. Gabel and H. Opschoor (eds), Principles of Environmental and Resource Economics. A Guide for Students and Decision Makers, Aldershot, UK and Brookfield, US: Edward Elgar, pp. 328–46. Gabel, L. and B. Sinclair-Desgagné (1993), ‘Managerial incentives and environmental compliance’, Journal of Environmental Economics and Management, 24, 229–40. Gabel, L. and B. Sinclair-Desgagné (1995), ‘Corporate responses to environmental concerns’, in H. Folmer, L. Gabel and H. Opschoor (eds), Principles of

218

The international yearbook of environmental and resource economics

Environmental and Resource Economics. A Guide for Students and Decision Makers, Aldershot, UK and Brookfield, US: Edward Elgar, pp. 347–61. Gabel, L. and B. Sinclair-Desgagné (1999), ‘The firm, its routines and the environment’, in T. Tietenberg and H. Folmer (eds), International Yearbook of Environmental and Resource Economics 1998/1999, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 89–118. Gladwin, T., J. Kennelly and T. Krause (1995a), ‘Shifting paradigms for sustainable development. Implications for management theory and research’, Academy of Management Review, 20(4), 874–907. Gladwin, T., J. Kennelly and T. Krause (1995b), ‘Beyond eco-efficiency. Towards socially sustainable business’, Sustainable Development, 3, 35–43. Gunningham, N., P. Grabosky and B. Sinclair-Desgagné (1998), Smart Regulation: Designing Environmental Policy, Oxford: Clarendon Press. Gutés, M. (1996), ‘The concept of weak sustainability’, Ecological Economics, 17(3), 147–56. Hamilton, J. (1995), ‘Pollution as news. Media and stock market reactions to the toxic release inventory data’, Journal of Environmental Economics and Management, 28, 98–113. Hart, S. (1997), ‘Beyond greening. Strategies for a sustainable world’, Harvard Business Review, 75, 66–76. Hawken, P., L. Lovins and A. Lovins (1999), Natural Capitalism. Creating the Next Industrial Revolution, Boston, MA: Little, Brown. Heijungs, R., J. Guinée, G. Huppes, R. Lankreijer, H. Udo de Haes and A. Sleeswijk (1992), Environmental Life Cycle Assessment of Products: Guide and Backgrounds, Leiden: Centre for Milieukunde, University of Leiden. Hoffmann, A. (1996), ‘Trends in corporate environmentalism. The chemical and petroleum industries, 1960–1993’, Society and Natural Resources, 9, 47–64. Hoffmann, V. and V. Radke (2000), ‘Indications of sustainable development: A critical analysis of the London School approach’, Journal of Environmental Policy and Law, 2, 145–63. Huber, J. (1995), ‘Sustainable development through sufficiency, efficiency and consistency’, in P. Fritz, J. Huber and H. Levi (eds), Sustainability in Natural and Social Science Perspectives, Stuttgart: Hirzel, pp. 31–46. Hueting, R. (1996), ‘Three persistent myths in the environmental debate’, Ecological Economics, 18, 81–8. Hueting, R. and L. Reijnders (1998), ‘Sustainability is an objective concept’, Ecological Economics, 27(2), 139–47. Jaffe, A., S. Peterson, P. Portney and R. Stavins (1995), ‘Environmental regulation and the competitiveness of US manufacturing: what does the evidence tell us?’, Journal of Economic Literature, 33, 132–63. Jones, K. and P. Rubin (1999), Effects of Harmful Environmental Events on Reputations of Firms, Atlanta, GA: Emory University. Johnson, S. (1996), ‘Environmental performance evaluation. Prioritising environmental performance objectives’, Corporate Environmental Strategy, Autumn, 17–28. Karl, H. and C. Orwat (2000), ‘Economic aspects of environmental labelling’, in H. Folmer and T. Tietenberg (eds), The International Yearbook of Environmental and Resource Economics 1999/2000, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 107–70. Khanna, M. (2001), ‘Economic analysis of non-mandatory approaches to environmental protection’, Journal of Economic Surveys, 15(3), 291–324.

Corporate sustainability

219

Khanna, M. and W. Anton (2002), ‘Corporate environmental management. Regulatory and market-based pressures’, Land Economics, 78(4), 539–58. Khanna, M. and L. Damon (1999), ‘EPA’s voluntary 33/50 program. Impact on toxic releases and economic performance of firms’, Journal of Environmental Economics and Management, 37(1), 1–25. King, A. and M. Lenox (2000), ‘Industry self-regulation without sanctions. The chemical industry’s responsible care program’, Academy of Management Journal, 43(1), 698–716. King, A. and M. Lenox (2001), ‘Does it really pay to be green? Accounting for strategy selection in the relationship between environmental and financial performance’, Journal of Industrial Ecology, 5, 105–16. King, A. and M. Lenox (2002), ‘Exploring the locus of profitable pollution reduction’, Management Science, 48, 289–99. Klassen, R. and C. McLaughlin (1996), ‘The impact of environmental management on firm performance’, Management Science, 42(8), 1199–214. Konar, S. and M. Cohen (1997), Does the Market Value Environmental Performance?, Nashville, TN: Vanderbilt University. Konar, S. and M. Cohen (2001), ‘Does the market value environmental performance?’, Review of Economics and Statistics, 83, 281–9. Korhonen, J. (2002), ‘The dominant economics paradigm and corporate social responsibility’, Corporate Social Responsibility and Environmental Management, 9, 67–81. Lévêque, F. and A. Nadaï (2000), ‘A firm’s involvement in the decision making process’, in H. Folmer and H.L. Gabel (eds), Principles of Environmental and Resource Economics. A Guide for Students and Decision Makers, 2nd edn, Cheltenham, UK and Northampton, MA: Edward Elgar. Lindblom, C.K (1994), ‘The implications of organizational legitimacy for corporate social performance and disclosure’, paper presented at the Critical Perspectives on Accounting Conference, New York. Milne, M.J. (1996), ‘On sustainability, the environment and management accounting’, Management Accounting Research, 7(1), 135–61. Neumayer, E. (1999), ‘Is economic growth the environment’s best friend?’, Zeitschrift für Umweltpolitik und Umweltrecht, 2, 161–76. Pearce, D., A. Markandya and E. Barbier (1989/1991), Blueprint for a Green Economy, London: Earthscan. Pearce, D. (1991), Blueprint 2. Greening the World Economy, London: Earthscan. Pezzey, J.C.V. and M.A. Toman (2002), ‘Progress and problems in the economics of sustainability’, in T. Tietenberg and H. Folmer (eds), International Yearbook of Environmental and Resource Economics 2002/3, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 265–332. Porter, M. (1987), ‘From competitive advantage to corporate strategy’, Harvard Business Review, May/June, 43–59. Porter, M. (1990), The Competitive Advantage of Nations, London: Macmillan. Porter, M. (1991), ‘America’s green strategy’, Scientific American, April, p. 96. Porter, M. (1999), Competitive Strategy, New York: Free Press. Porter, M. and D. Esty (1998), ‘Industrial ecology and competitiveness. Strategic implications for the firm’, Journal of Industrial Ecology, 2(1), 35–43. Porter, M. and C. van der Linde (1995a), ‘Green and competitive. Ending the stalemate’, Harvard Business Review, September/October, 120–34. Porter, M. and C. van der Linde (1995b), ‘Toward a new conception of the

220

The international yearbook of environmental and resource economics

environment–competitiveness relationship’, Journal of Economic Perspectives, Fall. Reid, D. (1995), Sustainable Development: An Introductory Guide, London: Earthscan. Reinhardt, F. (1999a), ‘Bringing the environment down to earth’, Harvard Business Review, 77, 83–93. Reinhardt, F. (1999b), ‘Market failure and the environmental policies of firms’, Journal of Industrial Ecology, 3, 9–21. Resnick, J. (2002), ‘A matter of reputation. A new survey of pharma companies evaluates leading and lagging players and identifies potential stars’, Pharmaceutical Executive, June, 1–7. Roberts, P.W. and G.R. Dowling (2002), ‘Corporate reputation and sustained superior financial performance’, Strategic Management Journal, 23, 1077–93. Rose, C. and S. Thomsen (2004), ‘The impact of corporate reputation on performance: some Danish evidence’, European Management Journal, 22(2), 201–10. SA/NZS (1999), SA/NZS 4360 Risk Management, Strathfield, Australia: Standards Australia and Standards New Zealand, April. SA/SNZ (2000), Environmental Risk Management: Principles and Process HB 203–2000, Canberra: Standards Australia and Standards New Zealand. Schaltegger, S. (2003), ‘A framework for ecopreneurship. Leading pioneers and environmental managers to ecopreneurship’, Greener Management International, theme issue on environmental entrepreneurship, M. Schaper (ed.), 38, 45–58. Schaltegger, S. and R. Burritt (2000), Contemporary Environmental Accounting, Sheffield: Greenleaf. Schaltegger, S. and R. Burritt (2002), ‘Environmental management accounting and the opportunity cost of neglecting environmental protection’, in M. Bennett, J. Bouma and T. Walters (eds), Environmental Management Accounting: Informational and Institutional Developments, Dordrecht/Boston/London: Kluwer Academic. Schaltegger, S., R. Burritt and H. Petersen (2003), Corporate Environmental Management. Striving for Sustainability, Sheffield: Greenleaf. Schaltegger, S. and T. Dyllick (2002), Managing Sustainability with the Balanced Scorecard, Wiesbaden: Gabler. Schaltegger, S. and H. Petersen (2001), Ecopreneurship: Concept and Typology, Lüneburg/Lucerne: Rio-Managementforum. Schaltegger, S. and A. Sturm (1990), ‘Environmental Rationality’, Die Unternehmung, 4, 273–90. Schaltegger, S. and M. Wagner (eds) (2004), Sustainability Performance and Competitiveness, Sheffield: Greenleaf. Schaper, M. (2003), ‘Environmental entrepreneurship’, Greener Management International, theme issue on environmental entrepreneurship, M. Schaper (ed.), 38. Scheelhase, J. (1999), ‘More sustainable economics and economic growth. Does this match?’, Journal of Apllied Environmental Research, 12(1–4), 553–62. Scherpereel, C., K. van Koppen and F. Heering (2001), ‘Selecting environmental performance indicators: the case of Numico’, Greener Management International, 33, 97–114. Schmidheiny, S. with the Business Council for Sustainable Development (1992), Changing Course, Cambridge, MA: MIT Press.

Corporate sustainability

221

Schmidt-Bleek, F. (1994), How Much Environment Does Man Need? MIPS: The Measure for Ecological Economics, Basel: Birkhäuser. Segerson, K. and L. Na (2000), ‘Voluntary approaches to environmental protection’, in T. Tietenberg and H. Folmer (eds), International Yearbook of Environmental and Resource Economics 1999/2000, Cheltenham, UK and Northampton, MA: Edward Elgar, pp. 273–306. Shenkar, O. and E. Yuchtman-Yaar (1997), ‘Reputation, image, prestige and goodwill. An interdisciplinary approach to organizational standing’, Human Relations, 50(11), 1361–82. Shrivastava, R., T. McInish, R. Wood and A. Capriaio (1997), ‘The value of corporate reputation: evidence from the equity markets’, Corporate Reputation Review, 1(2), 62–7. Stead, J. and E. Stead (2000), ‘Eco-enterprise strategy: standing for sustainability’, Journal of Business Ethics, 24(4), 313–29. Steurer, R. (2001), ‘Paradigms of sustainability’, Journal of Environmental Policy and Law, 4, 537–67. Stone, C. (1975), Where the Law Ends: The Social Control of Corporate Behavior, San Francisco, CA: Harper & Row. Taistra, G. (2001), ‘The Porter Hypothesis for environmental policy’, Journal of Environmental Policy and Law, 2, 241–62. Tietenberg, T. (1998a), Environmental and Natural Resource Economics, Reading, MA: Addison Wesley. Tietenberg, T. (1998b), ‘Disclosure strategies for pollution control’, Environmental and Resource Economics, 11, 587–602. Turner, R.K. (1993), Sustainable Environmental Economics and Management: Principles and Practice, Chichester: Wiley. Tyteca, D. (1996), ‘On the measurement of the environmental performance of firms. A literature review and a production efficiency perspective’, Journal of Environmental Management, 46, 281–308. Tyteca, D. (1997), ‘Linear programming models for the measurement of environmental performance of firms. Concepts and empirical results’, Journal of Productivity Analysis, 8, 183–97. UNWCED (1987), Our Common Future (the Brundtland Report), Geneva: United Nations World Commission on Environment and Development and the Commission for the Future. Vaughan, S. (1994), Greening Financial Markets, Geneva: International Academy of the Environment. Villinger, A., R. Wüstenhagen and A. Meyer (2000), Beyond the Eco-Niche, Basel: Birkhäuser. Von Weizsäcker, E., A. Lovins and L. Lovins (1997), Factor Four. Doubling Wealth, Halving Resource Use, London: Earthscan. Wackernagel, M. and W. Rees (1996), Our Ecological Footprint; Reducing Human Impact on the Earth, Gabriola Island, BC: New Society Publishers. Wagner, M. (2003), ‘The influence of ISO 14001 and EMAN certification on environmental and economic performance of the firm: an empirical analysis’, in M. Bennett, P. Rikhardsson and S. Schaltegger (eds), Environmental Management Accounting. Purpose and Progress, Dordrecht: Kluwer, pp. 367–86. Wagner, M. (2002), How Does it Pay to be Green? An Analysis of the Relationship Between Environmental and Economic Performance at the Firm Level and the Influence of Corporate Environmental Strategy Choice, Marburg: Tectum.

222

The international yearbook of environmental and resource economics

Wagner, M. and S. Schaltegger (2002), Environmental Management in German Companies: The Current State of Practice, Lüneburg: CSM. Wagner, M., S. Schaltegger and W. Wehrmeyer (2001), ‘The relationship between the environmental and economic performance of firms’, Greener Management International, 34, 95–108. Walley, N. and B. Whitehead (1994), ‘It’s not easy being green’, Harvard Business Review, May/June, 46–52. WBCSD (World Business Council for Sustainable Development) (1997), Environmental Performance and Shareholder Value, Geneva: WBCSD. Wehrmeyer, W. and D. Tyteca (1998), ‘Measuring environmental performance for industry’, International Journal of Sustainable Development and World Ecology, (5), 111–24. Welford, R. (1998), ‘Corporate environmental management, technology and sustainable development’, Business Strategy and the Environment, 7(1), 1–12. Werner, F. and M. Bammert (2001), ‘Environmental performance: sectoral ratings and environmental protection in the food and consumer goods industry’, Greener Management International, 34, 27–33. White, M. (1996), Corporate Environmental Performance and Shareholder Value, Charlottesville, VA: University of Virginia Press. Wüstenhagen, R. (1998), ‘Greening Goliaths versus Multiplying Davids. Paths of a coevolution of ecological mass markets and sustainable niches’, St Gallen: IWÖDiskussionsbeitrag no. 61. Ytterhus, B., T. Synnestvedt, A. Hildebrandt and S. Aasebø (1994), The Norwegian Environmental Business Barometer 1993, Oslo: Handelshøyskilen BI.

Internet Addresses: www.climatechangesolutions.com/english/sme/stories/interface.htm www.eco-web.com www.educationguardian.co.uk/higher/news/story/0,9830,931782,00.html www.harc.edu/harc/Content/OurWork/ShowTheme.asrx/37 www.interfacesustainability.com/ www.sustainability-management.de www.sustainability-performance.org/index.php www.unece.org/env/epr www.wsu.edu/~susdev/Gutes 96.html www.wbcsd.org This chapter has profited greatly from comments of Henk Folmer, Tom Tietenberg and an anonymous reviewer.

6. The double-dividend hypothesis of environmental taxes: a survey Ronnie Schöb* 1

INTRODUCTION

In the early 1970s, environmental awareness grew and environmental protection started to climb up the political agenda. Right from the beginning of the greening of politics, the idea of taxing polluting activities dating back to Pigou (1920) has been taken up in the political discussion. It was widely accepted that environmental taxes are an efficient instrument to protect the environment, superior to the classical environmental policy instruments of command and control. The enthusiasm for environmental taxes gained momentum with the double-dividend hypothesis. Tax revenues from environmental or green taxes can be used to cut other taxes. This can reap a second dividend as it reduces the distortion due to other taxes. The weak form of this hypothesis indicates that tax revenues from a revenue-neutral green tax reform can be used to cut other distorting taxes thus lowering the non-environmental efficiency cost of the green tax reform. The strong form of the double dividend asserts that a green tax reform not only improves the environment but also increases non-environmental welfare, i.e. it reduces the total deadweight loss of the tax system. If the latter holds, a green tax reform would be a so-called ‘no-regret’ option: even if the environmental benefits are in doubt, an environmental tax reform may be desirable (Bovenberg, 1999, p. 421).1 The weak form of the double-dividend hypothesis is widely accepted among economists. One implication of this form is that if there is a general consensus about an environmental target, green tax reforms are preferable to other environmental tax instruments that are cost-efficient in regulating the environment but do not raise public revenues. Generating public revenues by reaching an environmental target allows reducing tax induced distortions. The question as to whether the strong form holds, however, depends heavily on the structure of the economy. If the government has optimized 223

224

The international yearbook of environmental and resource economics

the tax system without taking into account environmental considerations, a green tax reform fails to increase non-environmental welfare, i.e. reap a second dividend, in economies with fully flexible labor markets. It may, however, be successful in economies suffering from involuntary unemployment. This survey focuses on this distinction in reviewing the literature on the double-dividend hypothesis and its recent extensions. The next section provides a brief sketch of the classical concept of environmental taxation. Then a model is presented that allows us to restate the main results of the double-dividend literature derived in the 1990s and discuss two important extensions made in the recent literature. The first extension focuses on an apparently technical point that turns out to be of importance if it comes to sound policy recommendations: if environmental problems are severe, individuals protect themselves from the consequences of pollution by e.g. buying defensive goods or they may reduce labor supply because of health problems. In both cases, it is necessary to relax the standard assumption of separability between pollution and consumption of commodities and leisure. As the ongoing political debate about the introduction of green taxes demonstrates, distributional considerations cannot be separated from efficiency considerations. Section 2 therefore also studies how optimal tax formulae have to be adjusted, taking equity considerations into account. The main purpose of the third section is then to point out the importance of labor market institutions for the determination of optimal environmental taxation or – adopting a more moderate approach – for welfare improving green tax reform. While the analysis of Section 2 assumes perfect labor markets and thus may be a good approximation for the US economy, it certainly fails to provide an appropriate framework for analysing green tax reforms in European countries. Section 3 therefore considers the double-dividend hypothesis for labor markets where wages are negotiated between employers and trade unions. The analysis of this section will show under which conditions environmental taxes on polluting inputs in production and on polluting consumption goods reap a second dividend in the form of an employment dividend and will discuss the welfare implications. The aim of this section is to point out the differences in the tax incidence for countries with perfect labor markets and countries facing unionized labor markets. Section 4 turns to the international aspects of environmental taxation and analyses under what circumstances countries should introduce environmental taxes unilaterally or when they should try to harmonize environmental taxes. These questions will be addressed by first looking at the competitiveness of an economy but it will become apparent that they are

225

Double-dividend hypothesis of environmental taxes

closely connected to the question of whether a strong double dividend exists. A further question is about the double dividend countries can reap at the cost of other countries when they coordinate environmental policy. As environmental taxes are inevitably taxes on the use of exhaustible resources, they affect the international distribution of resource rents. Depending on the time path of an environmental tax, the extraction rate of natural resources varies and hence the time path of pollution. The design of environmental taxes in the long run may thus not only affect the intertemporal allocation but will have severe consequences for the international distribution of wealth as they affect the distribution of resource rents. A final section concludes.

2

THE DOUBLE-DIVIDEND HYPOTHESIS: RECENT EXTENSIONS

In the classical contribution about environmental taxation, Pigou (1920) has shown that an optimal tax on emissions has to be set equal to the marginal environmental damage (MED). Such a ‘Pigovian tax’ can ensure that polluters bear the whole marginal social costs of their consumption of polluting goods. The concept of Pigovian taxation can be seen from Figure 6.1, which can be found in every textbook on environmental economics. MCsoc MC MB

F MB(x) D

MCpriv tp

A

E

B

xd* Figure 6.1

Pigovian tax

C

xd0

MCpriv

xd

226

The international yearbook of environmental and resource economics

In Figure 6.1 we consider the case of a polluting consumption good xd. MB(xd ) describes the marginal benefit of consumption, MCpriv the private marginal costs, and MCsoc the social marginal costs, respectively. Without environmental regulation, the competitive market outcome leads to an equalization of private marginal costs and private marginal benefits. The market equilibrium is xd0. The welfare loss in the equilibrium is equal to the area CDF as the marginal social costs MCsoc exceed the marginal benefit of consumption MB for all units consumed in excess of x*d. Piecemeal extension of the dirty good consumption from zero to x*d however, increases welfare. Pareto optimality is achieved where the marginal social costs MCsoc, which consists of marginal private costs MCpriv and the marginal external costs MED, equal the marginal benefit (see e.g. Baumol and Oates, 1988). The Pigovian tax tp can sustain the first-best Pareto-efficient outcome and allows the government to raise tax revenues equal to the area shaded in grey. These tax revenues become important in the presence of distortionary taxes as they may be used to reduce the excess burden of other taxes. According to the weak form of the double-dividend hypothesis, environmental taxes can not only improve the quality of the environment but also reduce the distortions of existing taxes on e.g. labor and capital income. This idea was first mentioned by Tullock (1967) and has been supported by partial equilibrium models in the 1980s, developed by Nichols (1984), Terkla (1984) and Lee and Misiolek (1986). Based on the seminal paper by Sandmo (1975), however, several papers, e.g. Bovenberg and de Mooij (1994), Bovenberg and van der Ploeg (1994a, b, c) and Goulder (1995) questioned this view by looking at a somewhat different definition of a second dividend. According to their interpretation, a positive second dividend in its strong form only exists if the total deadweight loss of the tax system, i.e. the non-environmental gross cost including the excess burden of the environmental tax, declines. Increasing a narrow based green tax and reducing a broad based tax like a tax on labor income will typically increase the gross distortion of the tax system. Bovenberg and de Mooij (1994) conclude ‘that environmental taxes typically exacerbate, rather than alleviate, pre-existing distortions – even if revenues are employed to cut pre-existing distortionary taxes’ (p. 1085). Hence, the second dividend is negative and the double-dividend hypothesis in its strong definition fails. In the following subsection 2.1, we will present a model that allows us both to replicate the standard results of the double-dividend literature and to show the validity of both interpretations (Section 2.2). The model is set up in a more general way in order to later on incorporate two important extensions recently made in the literature. Most of the standard results have been derived under the assumption that there is separability between

Double-dividend hypothesis of environmental taxes

227

consumption and the environment. FitzRoy (1996) points out convincingly that only if environmental problems are important do we observe people protecting themselves from the consequences of pollution. If this is the case, separability between consumption and environmental quality is too strong an assumption and it is necessary to explicitly take account of the interaction between pollution and consumption. As will be shown in Section 2.3, lower environmental taxes increase pollution and induce a higher level of the consumption of taxed defensive goods and therefore lower the welfare loss from taxation. Secondly, the double-dividend literature has widely ignored redistributional objectives in determining optimal environmental taxation2 although equity considerations may change the structure of optimal taxes significantly. For instance, if the government employs a linear income tax, optimal taxation requires that differentiated commodity taxes should be used to supplement a linear income tax as a redistributive device (see Atkinson and Stiglitz, 1980). In an economy with externalities, distributional considerations may affect the optimal environmental tax in two ways. As Smith’s (1992) review of some empirical studies indicates, environmental taxes may have a regressive nature because low-income persons largely consume some environmentally harmful goods. In this case, the presence of redistributional objectives might lower the level of taxes on environmentally harmful commodities. Secondly, distributional considerations also influence the valuation of environmental damage. While the physical incidence of pollution is typically higher in the low-income groups, e.g. due to badly situated housing, well-off people tend to put a higher value on environmental quality (Harrison, 1994). Section 2.4 incorporates redistributional objectives in the standard model. The individual willingness to pay for environmental quality, summed up to derive the environmental damage, has to be weighted by the social weight given to the individuals in the social welfare function. The stronger society’s inequality aversion, the more heavily weighted the valuation of the poor and, ceteris paribus, environmental taxes should be larger the more pollution affects the poor. Following Pirttilä and Schöb (1999), this section derives the many-person Ramsey tax rule by allowing for environmental externalities, which arise from the consumption of an environmentally harmful good, and discusses how environmental externalities influence the condition for the optimal tax structure. 2.1

The Model

We consider a closed economy with H households with identical preferences but different income earning abilities. There are two private consumption

228

The international yearbook of environmental and resource economics

goods c and d, a public good G and labor
. The private consumption good c is clean, i.e. its consumption has no external effect, whereas the private good d is dirty, i.e. its consumption creates negative external effects that cause the environmental quality E to deteriorate. The quantities demanded or supplied by household h are denoted by xih, i
c, d,
, the aggregate quantities of the consumption goods are denoted by Xc and Xd, respectively. There is a linear technology for the production of the private goods and the public good, with labor being the only input. Assuming perfect competition, we can choose units for all goods such that all producer prices are equal to 1. As labor productivity differs between households, we denote the marginal productivity of each household’s labor by p
h. For the normalization chosen, p
h also represents the wage rate for household h. The production possibilities are described by


p
h x
h
Xc  Xd  G.

(2.1)

h

The government provides the public good G and grants each household a uniform lump-sum subsidy T (which might be negative). To finance its expenditures for a given amount of the public good, the government can levy taxes on the private commodities and labor supply. The government’s budget constraint is therefore given by G  HT
t



p
h x h
 tc Xc  td Xd ,

(2.2)

h

where tc and td denote the commodity taxes on the clean good and the dirty good, respectively, and t
denotes the labor tax rate. As all private demands are homogeneous of degree zero in consumer prices, we can normalize one tax rate to zero. In what follows we will make use of different normalizations in order to derive and compare the standard results from the doubledividend literature. Environmental quality E deteriorates due to polluting production or consumption. As the main emphasis of this chapter is on the interaction of optimal environmental taxes with other forms of taxation, we restrict our analysis to the case of environmental externalities that are proportional to the quantity of a polluting commodity produced or consumed. The environmental quality is thus a decreasing function of the aggregate quantity of the dirty good Xd produced and consumed, i.e. E
e(Xd ), e
de dXd  0.

(2.3)

The preferences of household h with respect to both the clean and dirty commodity, leisure x h0, the public good G, and the environmental quality E,

Double-dividend hypothesis of environmental taxes

229

can be represented by a twice continuously differentiable, strictly quasiconcave utility function U h
u(x h0, x hc, x hd, G, E ),

(2.4)

with ui 0, i
0, c, d, G, E, denoting the marginal utility of good i. The time endowment is normalized to one, hence x h0  x
h
1. The budget constraint of the household is given by (1  tc)x ch  (1  td )x dh
(1  t
)ph
x h
 T.

(2.5)

As households differ in their earning abilities, represented by differences in the wage rate, households will also differ in their consumption patterns. When consuming the dirty good, the single household does not take account of the negative effect of its consumption on environmental quality. The benevolent government maximizes social welfare, represented by a Bergson-Samuelson welfare function W
W(v1(t
, tc, td, T, G, E ), v2(t
, tc, td, T, G, E ), . . ., vh(t
, tc, td, T, G, E )),

(2.6)

subject to its budget constraint (2.2). The term vh refers to the indirect utility function of household h. The government can influence private utility, and hence social welfare by (i) varying the lump-sum transfer, (ii) imposing commodity and labor taxes in general and (iii) determining the environmental quality E by imposing a particular environmental tax on the dirty good.3 The Lagrangean of the government’s maximization problem is therefore 
W(v1, v2, . . ., vh )  [t
X
 tc Xc  td Xd  G  HT ].

(2.7)

Denoting the private marginal valuation of income (the Lagrange multiplier of the individual household’s optimization problem) by h, and using Roy’s identity, the first-order conditions are as follows (using the notation X


h p h
x h
) : 
 t
ˇ

ˇ

W

W vh E


vh h ph
x h

vh E t
ˇ

h



ˇ

h

ˇ

 X





i

,c,d

ti

ˇ

ˇ

ˇ

ˇ

ˇ



Xi Xi E  t
0, t
i

,c,d i E t
ˇ

ˇ




ˇ

ˇ

ˇ

ˇ

(2.8)

230

The international yearbook of environmental and resource economics


 tc




ˇ

h

ˇ

W h h xc  vh



ˇ

ˇ


h

ˇ

ˇ

ˇ

ˇ

ˇ

ˇ

ˇ

ˇ

X

X E

(2.9)

i

,c,d




ˇ

h

ˇ



W v h E vh E tc

ti i 
ti i 
0,
E tc i

,c,d tc

W h h xd  v h

 Xd  
T

h

ˇ

 Xc  
 td




ˇ

W vh E v h E td ˇ

ˇ

ˇ

ˇ

ˇ

ˇ


ti tdi 
ti Ei td
0, X

X E

i

,c,d

W

(2.10)

i

,c,d

W v h E


vh h 
vh E T ˇ

h

ˇ

h

ˇ



ˇ

ˇ

ˇ

ˇ

ˇ


ti Ti 
ti Ei T
0. X

 H

i

,c,d

X E

(2.11)

i

,c,d

The derivative of E with respect to a parameter Z, Z
t
,tc,td,T, can be calculated by total differentiation of equation (2.3): dE
dZ ˇ

ˇ

e

x h


Zd ˇ

h

1  e

ˇ

x dh


E ˇ

h

 e

x h


Zd , ˇ

h

(2.12)

ˇ

ˇ

where (0) denotes the environmental feedback effect. The environmental feedback effect takes account of the fact that the quality of the environment may influence the demand for the dirty good. If a cleaner environment increases the consumption of the dirty good,  becomes smaller than unity. Congestion fees e.g. will reduce traffic jams during the rush hour. Less traffic, however, will encourage more traffic.4 2.2

Optimal Environmental Tax without Distributional Considerations

In the following subsection we assume homogenous households and separability between private consumption and the environment E, and private consumption and the public good G, respectively. Thus, all marginal rates of substitution between private goods are independent of E and G. There are H identical households whose preferences are represented by the same indirect utility function. The welfare function simplifies to W
Hv(t
, tc, td ,T, G, E ).

(2.6a)

Double-dividend hypothesis of environmental taxes

231

Welfare is maximized with respect to the government’s budget constraint (normalizing p
1) G  HT
H(t
x
 tcxc  td xd ).

(2.2a)

The first-order conditions (2.8) to (2.11) have to be adjusted accordingly. If the government has unlimited access to lump-sum taxes T0, the conditions
and t

tc
0 establish a first-best solution where the government sets the environmental tax equal to the marginal environmental damage (using
uc): td
tP  

HuE e . uc

(2.13)

In a first-best world the government will set the optimal environmental tax equal to the Pigovian tax and will not apply any other distorting taxes. There is thus no scope for a second dividend of the environmental tax. A labour tax system If lump-sum taxation is not available, the government has to rely on distortionary taxes to raise revenues. The second-best solution can be derived from the equation system (2.8) and (2.10) or the equation system (2.9) and (2.10), depending on the normalization chosen. Following Schöb (1997) we describe the normalization tc
0 as a labor tax system. For this normalization and – following Bovenberg and de Mooij (1994) – the assumptions that the utility function is (i) separable between environmental quality, public good, leisure and consumption goods and (ii) homothetic in consumption goods, it is optimal to have a labor tax but no commodity tax in the absence of environmental externalities. In the presence of external effects, however, there will be an environmental tax in addition to the labor tax (see Appendix 1 to this chapter for all relevant calculations for this section): td
tP .

(2.14)

For an upward sloping labor supply curve, Bovenberg and de Mooij (1994) show that the second-best optimal environmental tax is lower than the first-best Pigovian tax. The intuition behind this result is that increasing a narrow based green tax and reducing a broad based tax like a tax on labor income will – under these conditions – increase the gross distortion of the tax system. To see this, consider the whole consumption bundle and its consumption price index. It is obvious that a reduction in the labor tax and a revenue-neutral increase of td will not affect the real after-tax wage,

232

The international yearbook of environmental and resource economics

if the household does not alter the composition of its consumption basket. However, if it substitutes the clean good for the dirty good, the government’s revenue will fall. Revenue-neutrality requires an increase in taxes, which implies that the consumer price index will increase at a higher rate than the net-of-tax wage. As a consequence the real after-tax wage actually falls. Labor supply falls and welfare decreases. Since the Pigovian tax completely internalizes the marginal environmental damage, the only effect of a marginal increase of the Pigovian tax is a higher marginal cost of public funds, i.e. a negative second dividend in its strong definition occurs. A commodity tax system Things look different, however, if we normalize the net wage rate to unity, i.e. t

0. Using the same assumptions made above, this would yield a commodity tax system with equiproportional tax rates in the absence of environmental externalities. In the presence of external effects, however, the tax on the dirty tax must be adjusted (see Appendix 1): td


 R t  t ,

d P

(2.15)

where tR d denotes the Ramsey tax component, which relies on the efficiency of the tax system only. From equation (2.15) it is no longer clear whether the tax on the dirty good lies above or below the Pigovian tax, even if the marginal utility of the public good exceeds marginal utility of the clean good as before. The two alternative optimal tax formulae (2.14) and (2.15) are the essence of an apparently ongoing controversy which has emerged in the literature about the magnitude of the second-best optimal tax on a polluting good: it seems to be unclear whether in second-best situations, characterized by distortionary taxes, optimal taxes on polluting goods should be higher or lower than the first-best Pigovian tax associated with the same allocation. As this analysis, which followed the analysis of Schöb (1997) (also see Fullerton, 1997), has shown, the difference in the results concerning the optimal tax rate on a polluting good is due to different normalizations of tax rates which lead to different definitions of what the tax on a polluting good actually is. The controversy can be settled by looking at a secondbest internalization tax.5 In the presence of externalities, Pareto efficiency requires the equality of social and private marginal welfare of consuming a dirty good. In a first-best world, characterized by the feasibility of lumpsum taxes, this can be achieved by imposing a tax on a polluting good that equals the marginal environmental damage. Such a Pigovian tax fully internalizes the external costs at the margin. In a second-best world we can

Double-dividend hypothesis of environmental taxes

233

apply the concept of internalizing externalities in a similar way by looking for a tax rate tEd on the dirty good that exactly internalizes the external effect of this dirty good. To derive such a tax rate, let us assume that one of the H households obtains one additional marginal unit of exogenous income Y. In the household optimum the household is indifferent to how to spend this additional income. Without loss of generality we can therefore assume that the household increases the consumption of d only, i.e. by 1 (1 tEd ) and that the government uses the additional tax revenues to increase the supply of the public good by tEd (1  tEd ). The effect of a marginal increase in income for one household on social welfare is therefore: dW ud  HuE e  HuG tEd
. dY 1 tEd

(2.16)

The first term of the right-hand side denotes the increase in private utility while the second term denotes the external effect imposed on all households by the additional consumption of the dirty good d. The last term is the increase in all households’ utility due to the additional provision of the public good G, which is financed by the internalization tax, imposed on the dirty good d. Full internalization requires that the private marginal utility of consuming the dirty good, which is du dY
ud (1 tEd ), is equal to the social marginal welfare (2.16) of consuming the dirty good. From this identity, it follows that the external effect is exactly internalized if and only if the internalization tax on the dirty good is u e tEd
 E , uG

(2.17)

which is identical to the tax rate (2.14) and the second component of the tax rate (2.15). This is the tax component of the total tax on the dirty good d that the government has to impose in order to exactly internalize the external effect.6 It would be the total tax on the dirty good if there were only an environmental dividend. An important property of this second-best internalization tax tEd as defined in (2.17), is that it depends only on the real variables uE, uG and e and thus is itself a real variable. The tax rate is measured in units of the public good. Although the tax rates themselves can be arbitrarily normalized, the second-best internalization tax tEd is independent of the tax rate normalization. It will not be affected by any change of this normalization. Empirically, this component is smaller than the Pigovian tax. Parry (1995) estimates that it is only between 63 per cent and 78 per cent of the marginal environmental damage.

234

The international yearbook of environmental and resource economics

The concept of the second-best internalization tax allows us to reinterpret the two tax formulae (2.14) and (2.15). From the labor tax system we can learn that the scope for environmental policy is smaller compared to the scope in a first-best world. The intuition becomes clear if we assume that the dirty good is the only consumption good available. If the labor tax rate reduces labor supply, it automatically reduces the consumption of the dirty good and therefore helps to improve environmental quality. Before the introduction of any additional green tax, we are therefore already closer to the second-best optimum than we would be in the no-tax situation. From adding the second-best internalization tax in a commodity tax system we learn that the total tax borne by the dirty good (in units of leisure) can – and normally will – be higher than the Pigovian tax. As the total effective tax on the dirty good exceeds the Pigovian tax, one could expect that the environmental quality is better in a second-best than in a first-best world. Indeed, it is maybe the most important insight that environmental policies that raise public revenues are superior to policies that leave the rent created by restrictions on pollution in the private sector (cf. Schöb, 1996). Achieving a given environmental level with a policy instrument that does not generate any tax revenues would impose the same effects on the consumption of taxed goods as a green tax as it must increase the price for the dirty good relative to leisure and the clean good. Both policies will lead the household to substitute untaxed leisure for the dirty good. This causes the tax base to erode (this is the so-called tax interaction effect) and the gross distortion of the tax system to increase. The green taxes have, unlike e.g. grandfathered permits, the advantage of generating an additional revenue recycling effect that allows the government to partly offset the tax base erosion effect. This result has recently been confirmed by a series of numerical general equilibrium models.7 The conclusion of Parry, Williams and Goulder (1999) with respect to carbon abatement policies can easily be generalized: ‘Carbon taxes, as well as carbon quotas or tradable permits that are auctioned by the government, enjoy the revenue-recycling effect as long as the revenues obtained are used to finance cuts in marginal tax rates of distortionary taxes such as the income tax. In contrast, grandfathered (nonauctioned) carbon quotas and permits fail to raise revenues and thus cannot exploit the revenue-recycling effect. . . . the inability to make use of the revenue-recycling effect can put the latter policies at a substantial efficiency disadvantage relative to the former policies’ (p. 53).8 The literature also shows that in second-best economies the abatement costs exceed the abatement costs economies would face in a first-best world. Assuming increasing marginal abatement costs suggests that the environmental quality in a second-best world is better than in a first-best

Double-dividend hypothesis of environmental taxes

235

world – a result confirmed recently by Metcalf (2003) who shows that increasing the public expenditure requirement improves the environment in most plausible cases. The discussion about the normalization is also very helpful when considering the case where the dirty good may not be taxable at all. Fullerton and Wolverton (2003) argue that many types of pollution are difficult to monitor or, when measurement of pollution is possible, enforcing a green tax may not be feasible. They show that if taxation of the dirty good is restricted, td
0, a two-part instrument that consists of a combination of a higher labor tax and a subsidy on the clean good can achieve the same allocation as either the labor tax system or the commodity tax system where we have a direct tax on the dirty good in effect. 2.3

Homogenous Households and Non-separability

As pointed out by FitzRoy (1996), severe environmental problems make it likely that a significant proportion of the consumption is spent on defensive goods, i.e. goods that are used to reduce the disutility derived from pollution. It is therefore necessary to allow for such behaviour and explicitly assume non-separability between consumption goods and environmental quality. In this case, the optimal tax formula of the dirty good in a commodity tax system becomes (see Appendix 1): td






 R x t  tP  ti H i e .

d

E i
c, d




(2.15a)

In condition (2.15a) the household’s reaction is reflected by the aggregate substitutability/complementarity relationship between the taxed commodities and environmental quality. If we have substitutability, i.e. ti H xi /E 0, a higher tax on the dirty good will reduce both pollution and tax revenues as the need for taxed defensive goods will be reduced when environmental quality increases and is not fully compensated by additional consumption of other taxed goods. Put differently: in the presence of defensive goods which are taxed, the social cost of pollution is lower as the marginal environmental damage is partly compensated by higher tax revenues due to a higher demand for defensive goods. If, for instance, the price elasticity of a defensive good is low, according to the Ramsey rule it should be taxed at a relatively high rate. If, by contrast, the elasticity with respect to pollution were high, a marginal increase of the environmental tax would lead to a large decrease in tax revenues. In this extreme case, it cannot be ruled out that tax revenues decrease rather than increase as a consequence of an increase in the environmental tax.9

236

The international yearbook of environmental and resource economics

PROPOSITION 2.1 (Optimal environmental tax and defensive goods): In the presence of defensive goods, in the sense that there is aggregate substitutability between taxed consumption goods and the environmental quality, the optimal tax on the dirty good should ceteris paribus be lower than in the case with separability between consumption goods and the environment.

Schwartz and Repetto (2000) and Williams (2002), by contrast, show that if reduced pollution boosts labor productivity and thus encourages labor supply, increasing rather than decreasing environmental quality will yield an additional positive tax interaction effect. A positive tax interaction effect would lead to a higher tax on the dirty good. Williams (2002) points out, however, that if reduced pollution improves consumer health and thus reduces defensive expenditures for medical treatment, there is a positive income effect that may lower labor supply. The sign of the second dividend may therefore change either way. 2.4

Heterogeneous Households

To analyse the case of heterogeneous households where households have identical preferences but different income earning abilities, we restrict the analysis to the case where the labor tax rate is normalized to zero and assume again separability between private consumption and the environment E, and private consumption and the public good G, respectively. This implies that the environmental quality has the same physical impact on all households, independently of their earning abilities and their consumption pattern. In this case, the environmental feedback effect  (cf. (2.12)), reduces to unity and the demand for the dirty good becomes independent of the environmental quality (see Pirttilä and Schöb, 1999 for the following). To derive optimal tax rules for heterogeneous households, it is convenient to introduce the definition of the gross social marginal valuation of household h’s income, measured in terms of government’s revenue by h 

W h . vh ˇ

(2.18)

ˇ

If the government is interested in redistributing income from high-ability households to low-ability households, the social welfare function (2.6) will be strictly quasi-concave, i.e. W/ h is larger the lower  h is. As private utility is also strictly quasi-concave, h decreases in utility. Hence, h is negatively correlated with the earning ability and the household’s utility level, respectively. The individual evaluation of the additional environmental damage may differ between individuals as the marginal valuation of the environment

Double-dividend hypothesis of environmental taxes

237

normally does not change proportionately with the marginal utility of income h. The marginal willingness to pay for environmental quality is therefore defined as h 

vh E . h

(2.19)

Applying the separability assumption, using the definitions (2.18) and (2.19) in the first-order conditions, and using Cramer’s rule, we can solve equations (2.9) and (2.10) for the optimal commodity tax rate of the clean and the dirty good, respectively. Denoting the determinant of the Jacobian matrix for the case of heterogeneous households as |J|, the optimal tax formulae are:


(h  1)xhc tdd  xhd tcd ,

(2.20)


(h  1)xhd tcc  xhc tdc 
hhe .

(2.21)

tc


td


1 |J|

1 |J|

X

X

h

X

X

h

h

Equations (2.20) and (2.21) show the result derived by Sandmo (1975). The external effect does not enter the optimal tax formula for the clean good even if distributional objectives are taken into account. It only enters the optimal tax formula for the dirty good additively. This environmental component of the optimal tax on the dirty good may be considered as the price the consumer of the dirty good has to pay in a second-best world to completely internalize the external effect. To see this, consider the following thought experiment, which is related to the interpretation of the second-best internalization tax in the last section for the model with identical households. We abstract from all other taxes and focus on the environmental tax component alone, which we define as tEd. If the household h receives one additional marginal unit of exogenous income Y h, the household’s utility increases by h, independently of how it spends the additional income. Without loss of generality, we can therefore assume that the household increases the consumption of d only, i.e. by 1 (1 tEd ). The effect of a marginal increase in household h’s income on social welfare is therefore (measured in units of public revenues) dW dYh W h
h 

v
h


k

1 W v k

1

tE


vk E e 1  tEd  1 d tEd k

kke  tEd



1 . 1  tEd

(2.22)

238

The international yearbook of environmental and resource economics

The first term of the right-hand side denotes the increase in the gross social marginal valuation of household h’s private utility h (cf. equation (2.18)). The second term denotes the social marginal external effect imposed on all households by the additional consumption of the dirty good d (cf. equation (2.19)). The last term shows the increase in public revenues from the internalization tax imposed on the dirty good d whereby it is assumed that additional tax revenues are used to increase public good provision. Full internalization requires that, from the viewpoint of society, the social marginal utility of the private consumption of the dirty good, i.e. the gross social marginal valuation h, should be equal to social marginal welfare of consuming the dirty good: h
dW/dY h / . Hence, the external effect is exactly internalized if and only if the tax on the dirty good is equal to tEd



hhe ,

(2.23)

h

which forms the environmental component of td in equation (2.21). The term he denotes the marginal willingness to pay for a reduction in emissions times the amount of emissions caused by a marginal increase in the dirty good consumption. In order to derive the social evaluation of pollution, the household’s marginal willingness to pay has to be weighted with the social weight h given to the household. The many-person Ramsey tax rule with externalities Diamond (1975) presents a procedure for interpreting commodity taxation rules when income can be taxed on a linear scale. This section refers to Diamond’s approach to deriving a many-person Ramsey tax rule, and demonstrates how his model has to be modified to allow for the presence of externalities. Therefore, we first redefine the net social marginal valuation of household h’s income, denoted by h, by taking into account the influence private consumption has on the external effect: h
h 

xh

xh


ti Ti 
kke Td. ˇ

i
c, d

ˇ

ˇ

k

(2.24)

ˇ

Definition (2.24) is identical with Diamond’s definition (see his equation (6)), except for the last term of the right-hand side. The net social marginal valuation of household h’s income includes, first of all, the gross marginal social valuation of income h, which represents the social evaluation of the marginal utility household h derives from a marginal increase in income. The social value of an extra income to household h also depends on the influence the additional income has on tax revenues. This effect is captured by the second term of the right-hand side of equation (2.24).

Double-dividend hypothesis of environmental taxes

239

If the extra income increases the demand for taxed goods by household h, tax revenues also increase and may be used e.g. for additional provision of the public good. In this case, the net social marginal valuation exceeds the gross social valuation of income. In the presence of externalities, the net social valuation of income also depends on the impact the additional income has on environmental quality. This effect is covered by the last term on the right-hand side. If the extra income increases, household h’s consumption of the dirty good, the value of the additional damage caused by it to all members of the economy has to be deducted from the social valuation of income. If the dirty good is a normal good, the externality augmented net social valuation of income will therefore be lower than Diamond’s (1975) definition suggests. Using the definition of h, the Slutsky decomposition xhi tj
h sij  x hj . x hi T, where shi denotes the compensated (cross-)price effect, and Slutsky-symmetry, the first-order conditions for tj, j
c, d can be rewritten in the following way:



ti shij

(h  1)xhj 
(hhe )
s hdj, h i
c, d

h

h

(2.25)

h

with j
c,d. In the absence of externalities, i.e. for e
0, equation (2.25) restates Diamond’s (1975) result (cf. his equation (7)). The new second term on the right-hand side takes account of the externality. In order to interpret equation (2.25), however, we will further simplify this condition. Substituting definition (2.24) into the first-order condition for the lumpsum transfer, equation (2.11), we obtain:


h 
ti Ti 
kke Td
H
⇔
g
1, xh

xh

ˇ

h

i
c, d

ˇ

ˇ

k

(2.26)

ˇ

where g denotes the average net social valuation of income over all households. Equation (2.26) states that in an optimum, the average valuation of a transfer of one unit of money should be equal to its costs that are equal to unity. Next, we define the normalized covariance between the net social evaluation of private income and the consumption of good j,

j 


hxhj h

gXj

1,

(2.27)

(see e.g. Atkinson and Stiglitz, 1980). The first term of the right-hand side is known as the distributional characteristic of good i (cf. Feldstein, 1972).

240

The international yearbook of environmental and resource economics

If the government is indifferent to which household the extra income is given, all h are identical to g, and the normalized covariance expression reduces to zero. With inequality aversion, h is larger for low-income households, provided that the additional tax revenue and the externality term in definition of h (cf. equation (2.24)) do not have too strong countervailing effects.10 This implies that the distributional characteristics of a good, of which lowincome households demand a large proportion, takes a value greater than unity, because the relative consumption of that good by household h increases with h. According to Rose and Wiegard (1983), the distributional characteristics of a good i measure how much of good i society is willing to give away for a more equal distribution of income. Substituting the definition of the normalized covariance (2.27) into equation (2.25) we obtain:



ti shij h i
c, d

Xj


j 


(hhe ) h

1  td


h

hjd

xhj

, Xj

(2.28)

with j
c,d. The term hjd denotes the compensated cross-price elasticity of good j with respect to the price of the dirty good d. The left-hand side equals the relative change of the compensated demand if all tax rates change proportionately. To see this, consider the total differential of the compensated demand function xhj(tc, td, u), j
c, d for a small equiproportionate change of all tax rates, i.e. dti
ti, i
c,d:11 dxhj |u



shji dti

shji ti.

i
c, d

(2.29)

i
c, d

Summing up over all households and dividing by the total demand, we obtain the relative change in aggregate demand


dxhj |u h

Xj






ti shij h i
c, d

Xj

.

(2.30)

The left-hand side of equation (2.28) therefore explains how the relative change in the demand of good j, due to a small equiproportionate change of all tax rates, is determined. To interpret the right-hand side carefully, consider first the case without external effects (i.e. e
0). In this case, the externality based term disappears from the first-order condition and from the definition of h, respectively. Optimal commodity taxation is determined by the normalized covariance alone. If, on the other hand, redistribution is not an issue, and there are no external effects, the normalized covariance term vanishes. In this case, there is no

Double-dividend hypothesis of environmental taxes

241

commodity taxation in the optimum and the required tax revenues are raised by uniform lump-sum taxes (T 0) only. With inequality aversion, however, the aggregate compensated change in demand of good j should be smaller, the lower the values of the normalized covariance j is. The normalized covariance rule therefore advises the government to subsidize the consumption of goods that are largely demanded by those people with a large net social marginal valuation of income, i.e. the poor people, and discourage the consumption of luxury goods consumed by rich households with low yh. In this way commodity taxation is used for redistributional purposes. If external effects are present, but society is not interested in redistribution, the aggregate compensated change in the demand of taxed goods is determined solely by the externality term. This term depends on the compensated elasticity between the taxed good and the dirty good. The compensated own-price elasticity of the dirty good, and hence the aggregate compensated change in the demand for the dirty good, will be negative. This impact arises naturally from the fact that, as the consumption of good d worsens the environmental quality, it is in society’s interest to reduce its consumption. For the clean good, the compensated change in demand should be smaller the higher the compensated complementarity relationship between the taxed good and the dirty good is (the more negative the compensated elasticity is). Proposition 2.2 summarizes. PROPOSITION 2.2 (The many-person Ramsey tax rule in the presence of externalities): If commodity taxes and the uniform lump-sum transfer are set optimally, a small equiproportional increase in all tax rates will cause all compensated commodity demands to change according to their distributional characteristics. In addition, the decline (increase) in the compensated demand for the taxed good will be larger (smaller) the stronger the complementarity relationship between the taxed good and the dirty good is.12

It is important to note that, with an equiproportional change in all tax rates, we change the level of the Pigovian tax component by the same amount as the Ramsey tax component. Hence, the tax on the dirty good increases at a larger rate than the standard many-person Ramsey tax rule suggests in the absence of externalities. Consequently, the demand for all complements will fall at a larger rate while the demand for substitutes will fall at a lower rate. This is the mechanism that makes the complementarity relationship between taxed goods and the dirty good enter the tax rule. A comparison with Sandmo’s (1975) additivity property shows that Proposition 2.2 does not imply that goods, which are complements to the dirty good, should be taxed more strongly. The conclusion rather suggests that their consumption is reduced because of the indirect impact the relatively larger change of the tax on the dirty good has on their demand.

242

The international yearbook of environmental and resource economics

This result also shows that, apart from the redistributional characteristics of the particular commodity, the optimal tax rule also depends on redistributional concerns because of the social valuation of environmental damage. The private disutility of a marginal increase in pollution is weighted by the gross social valuation of household h’s income h. This means that the marginal willingness to pay for environmental quality by a low-income household has a relatively high impact on the social valuation of the environment and on the externality based term in the tax rules as well. Moreover, as the magnitude of h decreases with the shadow price of public funds , we can deduce that the externality based part in the tax conditions decreases with rising . The reason is that, as the burden of public funds rises, it becomes more and more expensive to internalize externalities. In brief, the modified many-person Ramsey tax rule reveals that the influence of commodity taxation on the demand for taxed goods depends on both redistributional and environmental objectives. If the worse-off households have a relatively large demand for the environmentally harmful good, as some empirical studies suggest (see Smith (1992) for further references), then the optimal tax rule proposes that the income distribution part of taxation lowers the tax on the dirty good. This tax would otherwise be high in order to internalize the negative external impact. Without any further restrictions on consumers’ preferences, efficiency and equity considerations cannot be separated when making decisions concerning the level of environmental taxes. This confirms Sandmo’s (2000) statement that the distributional concerns cannot be ignored in the study of externalities within public finance models. This discussion can be connected to the double-dividend debate reviewed in Section 2.2. Although the formulation containing the change in the compensated demand does not provide an explicit tax rule, it still implies that if the dirty good is in relatively great demand by low-income households, the non-environmental tax component should be low or even negative. Put differently: the green tax yields a negative ‘equity dividend’ – because, in the absence of externalities, differentiated commodity taxation is only used to influence income distribution, and a reduction in the tax rate of a good is a direct way to increase its compensated demand.

3

EMPLOYMENT AND WELFARE EFFECTS IN THE PRESENCE OF UNEMPLOYMENT

Do green tax reforms boost employment? This question provoked the search for another second dividend of environmental tax reform, which may be called the employment dividend. Though the concept of an employment

Double-dividend hypothesis of environmental taxes

243

dividend is meaningless for countries with fully flexible labor markets with no involuntary unemployment,13 it has become the most important concept in the political debate about green tax reforms in the European countries suffering from persistently high levels of involuntary unemployment. One obvious way to reduce unemployment by raising environmental taxes is by recycling the resulting tax revenues through cuts in labor taxes. The high levels of taxes on labor income, combined with the high level of unemployment benefits, are often made responsible for unemployment since it distorts labor supply and increases wage pressure in labor markets (OECD, 1995). A green tax reform may alleviate the tax burden on labor and hence reduce the resulting disincentives. To show this, one has to analyse the tax incidence of both the green tax and the labor tax in the presence of labor market imperfections. In this section we focus on labor market imperfections and analyse the effects of green taxes in the production sector by using the framework developed by Koskela, Schöb and Sinn (1998). In their model the wage is endogenously determined in a bargaining process between trade unions and a firm. The firm produces with two factors of production and faces a downward sloping demand for its good. The wage negotiations are analysed using a ‘right-to-manage’ model. Trade unions and firms bargain over wages and firms then choose the employment level that maximizes profits. While the focus of Koskela, Schöb and Sinn (1998) was on tax reform, the focus here is on optimal tax formulae, in order to provide a comparison with the analysis of Section 2. Section 3.1 sets up the model. Section 3.2 provides the main intuition as to why a green tax reform can reap a second employment dividend. Section 3.3 then derives the optimal tax formulae and discusses how they depend on the magnitude of the labor market distortion and the availability of non-distorting profit taxes. Section 3.4 summarizes and relates the results derived here to the results derived from models looking at other labor market imperfections. 3.1

The Model

We consider an economy that produces a good Y that is entirely exported. The revenues are used to import a produced import good, which serves for public consumption G and private consumption C, and a natural resource R. The price of export goods Y is p, which is measured in terms of a produced import good. This can be interpreted as the economy’s ‘terms of trade’. This open economy that satisfies the usual resource constraint I
C GpY M,

(3.1)

244

The international yearbook of environmental and resource economics

where I and M denote domestic income and imports. The price for the import good R is q, again defined in terms of the imported consumption good. Thus total imports are M
CGqR. A single monopolistic firm produces good Y with energy R, and labor L as inputs, and represents domestic production. While energy R is imported, labor L is internationally immobile. Technology is assumed to be linearhomogeneous and is represented by a constant elasticity of substitution production function Y
f (L,R). The monopolistic firm faces world output demand D( p), which is decreasing in the output price p and is assumed to be iso-elastic, i.e. Y
D( p)
p, with  (D( p)/p)p/Y denoting the output demand elasticity. The closer substitutes for good Y on the world market are, the more elastic output demand becomes. Note that the economy is ‘large’ with respect to the output produced but ‘small’ with respect to the resource import. The setup of the model allows us to generate profits that are needed to model wage bargaining. Note however, that neither the assumption of a monopolistic firm nor the assumption of a large open economy with respect to the output good is essential for the results we derive below. Alternatively, we could assume a small open economy with respect to both exports and imports and assume a CES production function with a third fixed factor that generates rent. The firm maximizes profits, which are given by 
p(Y )Y  qR  wL, where the firm considers the gross energy price q and the gross wage rate w as given. The gross wage is the net-of-tax wage, which is negotiated between a trade union and the firm, plus the labor tax, modelled as a payroll tax: w
(1  tw)w. The energy price is the foreign resource price plus a green tax levied on the use of energy in production: q
(1  tq)q. To guarantee a profit maximum, the output demand elasticity must exceed unity, i.e. 1, in which case profit maximization implies that the firm will set a price that exceeds the constant marginal costs c(w, q ) by a constant mark-up factor /(1)1. All N workers of the economy are represented by a single trade union that maximizes its N members’ net-of-tax income.14 Each member of the trade union supplies one unit of labor if employed, or zero labor if unemployed. The earnings of a member thus equal the net-of-tax wage rate w if employed. If he or she is unemployed the trade union member has an outside option b, which depends on the utility derived from leisure and the unemployment benefit transfers from the government. The objective function of the trade union is hence given by V *
wL b(NL).

(3.2)

The wage rate is determined in a bargaining process between the trade union and the firm. After the net-of-tax wage rate is fixed, the firm

Double-dividend hypothesis of environmental taxes

245

unilaterally determines employment. This is modelled by using a ‘right-tomanage’ model, which represents the outcome of the bargaining by an asymmetric Nash bargaining. The fallback position of the trade union is given by V 0
bN, i.e. if the negotiations break down, all members receive their reservation wage equal to the outside option. The fallback position of the firm is given by zero profits, i.e. 0
0. Using VV *  V 0, the Nash bargaining maximand can be written as 
V 1,

(3.3)

with  representing the bargaining power of the trade union. Using a CES production technology we will apply the explicit formulation of the wage elasticity of labor demand, L,w  Lww L
  s(  ), with  being the elasticity of substitution between labor and energy and s
wL cY being the cost share of labor. The first-order condition with respect to the net-of-tax wage rate is w
0 ⇔ (w  b)(L, w  (1  )s(1  ))  w
0.

(3.4)

Equation (3.4) implicitly determines the negotiated net-of-tax wage from Nash bargaining as a function of the tax policy parameters tw and tq so that we have w
w(tw,tq). To derive the optimal tax formulae for an economy where the nominal wage is determined in wage negotiations, we first have to know how wage negotiations are affected by the tax system. The effect of a change in the labor tax rate on the net-of-tax wage rate is wt (w  b)zw wt
 w
 , w ww y  (w  b)z(1  tw)

(3.5)

with y
(1  L, w)  (1  )(1  )s and z
[(  )  (1  )(1  )]sw. As the second-order condition is assumed to hold throughout, i.e. ww
y  (w  b)z(1  tw)  0, we can infer that sign(wtw)
sign(z)
sign(sw) if labor and energy are price complements   , as we will assume in what follows.15 For a CES production technology, the partial derivative of the cost share of labor with respect to the gross wage rate is given by sw


  s (1  s)(1  )
0 ⇔ 
1, w  





so that wt

w



 0 as   1
0 as 
1  0 as   1

(3.6)

246

The international yearbook of environmental and resource economics

If the elasticity of substitution  is less than one, an increase in the labor tax rate will lead to an increase in the cost share of labor s as the resource cannot easily be substituted for labor. A larger share s implies that a wage increase has a stronger impact on total costs of production and the output price the firm charges. If, for instance, the cost share of labor is 60 per cent, a 10 per cent wage increase would result in a 6 per cent increase of marginal costs (disregarding substitution). If the cost share of labor is 70 per cent, the respective increase in marginal costs is 7 per cent. Accordingly, more workers are laid off in the latter case than in the former case. Hence, the trade union benefits less from demanding higher wages the larger s becomes and the net-of-tax wage rate falls. By contrast, when the elasticity of substitution exceeds 1, the cost share of labor s decreases due to higher labor taxes and the trade union benefits more from demanding higher wages and the net-of-tax wage increases. If technology is CobbDouglas, we have an elasticity of substitution of 1. The wage elasticity is then constant so that factor taxes will have no effect on the negotiated netof-tax wage. An exogenous increase in the green tax rate has an effect on the cost share of labor opposite to that of the increase in the labor tax rate. Hence, depending on the elasticity of substitution, the total effect of an increase in tq is: wt

q



 0 as   1
0 as 
1  0 as   1

(3.7)

The interpretation of (3.7) is analogous to that presented for the labor tax rate. Finally, the government requires a fixed amount of tax revenues to finance the public good G.16 The government levies the labor tax tw on wage income and a tax on energy input tq. In addition there is a profit tax t on domestic profits so that the government budget constraint is twwL  tqqR  t
G.

(3.8)

To focus on efficiency aspects of the optimal tax structure only, we assume linear preferences and thereby consider the total surplus as an appropriate social planner’s objective function (cf. Summers, Gruber and Vergara, 1993). The total surplus consists of the wage income equal to wL, which accrues to workers, b(N  L), the money metric-utility the unemployed derive from leisure, and the net-of-tax profit income (1 t). As we keep G constant throughout the analysis, we can suppress the term G in the welfare function. As all energy is imported, private income from energy sales does not appear in the welfare function. All domestic profits go to

Double-dividend hypothesis of environmental taxes

247

domestic capitalists. Finally, the monetarized value of the environment is given by E(R), with ER 0, which enters the welfare function separately. Hence, the welfare function is given by S
wL  b(N  L)  (1  t)  E(R). 3.2

(3.9)

Labor Tax System vs Green Tax System

Let us begin our analysis by asking whether there might be a reason for introducing a ‘green tax system’, characterized by relatively high tax rates on energy and relatively low labor taxes, which yields the same output as the existing ‘labor tax system’ where the labor tax rate exceeds the energy tax rate, but generates a higher level of employment.17 For the sake of the argument we keep the net-of-tax wage w constant. This fixed-wage assumption is to facilitate the initial analysis. It will be removed in the subsequent section when we look at wage bargaining. What are the conditions such a green tax system must satisfy? Profit maximization requires that the same output f(L,R)
Y0, where the output A level Y0 is ceteris paribus determined by the initial tax rates tA w and tq , is produced with minimum costs. The first-order condition for cost minimization can be represented by wfR(L, R)  q fL(L, R)
0, where fi denotes the partial derivative of f(L,R) with respect to i
L, R. Furthermore, marginal costs must be equal in the two systems, for otherwise the firm would not sell the same output in equilibrium as before. With linear-homogenous technologies this implies constant total cost, wL  qR
C0. Finally, the government budget constraint (3.8) must be met. These four conditions provide an equation system, which can be solved with respect to the optimal inputs and the necessary tax rates, respectively. The solution is represented in Figure 6.2. In the profit maximum, the slope of the isoquant Y0 equals the negative of the ratio of the tax-inclusive factor prices q w. In the initial equilibrium A we observe the factor price A A A ratio (1 tA q )q (1 tw )w with tw  tq . Since A is a point of tangency between the tax-inclusive isocost and the isoquant, it characterizes a cost A minimum. Given q, w, tA q , tw , there are many such cost minima on a ray from the origin through A all of which have the same unit production cost, but because of the endogeneity of the output price, there is only one point that maximizes profits: in Figure 6.2, point A indicates the initial labor tax A system (tA w , tq ). The curve on the right-hand side of Figure 6.2 shows the net-of-tax isocost curve, which is defined as the geometrical locus of factor combinations that would be attainable at a given expense if there were no taxes. The

248

The international yearbook of environmental and resource economics

L

B

ax

f-t

-o

et

N

LB

st

co

iso

A

LA 0

RB

Y0

RA

R Tax revenue

Figure 6.2

Labor tax system versus green tax system

net-of-tax isocost curve is steeper than the tax-inclusive isocost through A A because tA w  tq , and its intersection with the horizontal axis is lying in a more outward position because tA q  0. The horizontal distance between A and the net-of-tax isocost equals the government’s tax revenue in terms of R. The broken parallel to the netof-tax isocost through A thus defines the geometrical locus of all potential equilibria, where tax revenue and net-of-tax factor expenses are the same as in the labor tax regime A. It is now possible, with an appropriate choice of the tax rates tw and tq, to transpose the economy from A to B, where output, tax-inclusive factor expenses and unit production cost are the same as in A. Since neither the unit production costs nor the output price alter with this transposition, the conditions for a cost minimum are preserved and B is an equilibrium. Point B indicates a green tax system (tBw, tBq ) with tBq  tBw that yields the same output at the same total costs. Moving from A to B will instantaneously increase employment, LB  LA, without imposing any additional costs on either firm or government. In addition, less energy will be used, RB RA, and, consequently, the environment will improve (remember: ER  0). This thought experiment shows that with given net-of-tax factor prices and a linear-homogenous production technology, there exists a green tax

Double-dividend hypothesis of environmental taxes

249

system with higher tax rates on energy than on labor which yields both the same output level and same tax revenues as the existing labor tax system where the labor tax rate exceeds the energy tax rate. The green tax system generates both a higher level of employment and a cleaner environment thus reaping a double dividend. As profits are unchanged, and welfare is increasing in both employment and environmental quality, welfare will be higher in the green tax system than in the labor tax system. Moving from A to B thus definitely reaps a double dividend. Note that the main result also holds with respect to changes in employment when we consider more than two factors of production and allow for an optimal adjustment of the other factors. Applying the Le Chatelier Principle, the optimal adjustment of other (quasi-fixed) factors will be such that the factor that became relatively cheaper is used even more intensively. In the general case, however, the statements with respect to energy inputs depend on the respective substitutability-complementarity assumptions made. Hamermesh’s (1993, Table 3.6) survey indicates that indeed labor and energy are substitutes, with a small cross-price elasticity. We could also enforce the intended substitution effect in the presence of some third factor by allowing the government to change the tax rate of some third factor in an appropriate way. 3.3

Welfare Maximization: the Optimal Tax Formulae

Next we turn to the welfare maximization problem where the government chooses tax rates first and the labor organizations then determine the wage rate in a wage negotiation, taking the tax rates as given. Hence, the government maximizes the total surplus (3.9) subject to the budget constraint of the government (3.8), the outcome of the wage negotiation, which is implicitly given by the first-order condition of the Nash bargaining (3.4), and an additional constraint on the profit tax rate (3.10) that might or might not be binding.

{

max S
wL  b(N  L)  (1  t)  E(R),

tw,tr,t,w

s.t. twwL  tqqR  t
G,

(3.8)

w
0 ⇔ (w  b)(L,w  (1  )s(1  ))  w
0,

(3.9)

t  t.

(3.10)

250

The international yearbook of environmental and resource economics

The Lagrangian equation for welfare maximization is 
wL  b(N  L)  (1  t)  E(R)  (G  twwL  tqqR  t  w  (t  t),

(3.11)

where , and  describe the shadow prices of the three constraints. Using the following expressions of the factor demand elasticities: R,w
R ww R
s(  ), L,q
Lq q L
(1  s)(  ) and R,q
  (1  s)(  ) the first-order conditions with respect to the profit tax rate, the two factor tax rates and the net-of-tax wage rate can be expressed (after some manipulations) as follows: t
0 ⇔ (  1)
,

(3.12)

tw
(w  b  tww) LL,w  (ER  tqq)RR,w  (  1)(1  t)wL  wt (1 tw)
0, w

(3.13)

tq
(w  b  tww)LL,q  (ER  tqq)RR,q  (  1)(1  t)qR  wt (1 tq)
0, q

(3.14)

w
(w  b  tww) LL,w  (ER  tqq)RR,w  (  1)(t  tw (1 tw))wL  www
0.

(3.15)

Given the complementary slackness condition t  t  0,   0, (t t)
0, we can distinguish two cases. If 
0, the profit tax constraint is not binding and the government can use the profit tax to cover all public expenditures. If the government is restricted in using profit taxes, the profit tax constraint is binding, i.e. 0. We will discuss these two cases separately. The benchmark case is where the government does not have to rely on distortionary taxation. If 
0, the first-order condition with respect to the profit tax rate (3.12) reduces to
1. The shadow price represents the marginal cost of public funds and is equal to 1. This indicates that the government can raise taxes to meet its revenue requirement without imposing any cost on society that exceeds tax revenues. Thus we have an economy without tax distortions but with labor market distortions. To analyse how these labor market distortions affect welfare, we subtract (3.15) from (3.13), using
1. This yields





 wtw(1  tw) 1
0.   www  www

(3.16)

Double-dividend hypothesis of environmental taxes

251

As we know from the second-order condition, www0, the shadow price must be equal to zero if the terms in brackets are non-zero. The first term in brackets represents the net-of-tax wage elasticity with respect to the labor tax. As long as an increase in the labor tax rate increases the gross wage rate however, the absolute value of this elasticity is below 1 (see Koskela and Schöb (2002a) for a proof), which is also in conformity with empirical studies (cf. e.g. Lockwood and Manning, 1993 and Holm, Honkapohja and Koskela, 1994). Therefore, the term in brackets must always be positive and condition (3.16) holds only if the shadow price
0. This result suggests that if the government can use profit taxation without any restriction, i.e. apply non-distortionary taxation, the Nash bargaining constraint is not binding. This has two consequences. First of all, it is optimal for the government to levy a Pigovian tax on energy. Solving the equation system (3.13) and (3.14) with respect to the factor tax rates and making use of 
0,
1 and
0, we obtain: E tq
 R  0. q

(3.17)

Furthermore, the government is able to fully offset the labor market distortions. Whatever net-of-tax wage rate is fixed in the wage negotiation between the trade union and the firm, the government can choose an appropriate wage tax or subsidy to obtain the gross wage, which optimizes social welfare: tw


 

wb  0. w

(3.18)

These two tax rates ensure that both gross factor prices equal their social opportunity costs. The marginal productivity of energy equals the net-of-tax energy price the country has to pay for importing energy plus the marginal environmental damage energy input in domestic production causes. From substituting the definition of the gross wage rate into equation (3.18) we can see that w
b. Thus, the gross wage equals the disutility of labor, which in turn equals the social costs of labor. The wage subsidy is equal to the markup between the net-of-tax wage rate and the marginal revenue product of labor the wage negotiation yields, given this subsidy. This establishes full employment in the sense that there is no involuntary unemployment anymore. These findings can be summarized as a proposition: PROPOSITION 3.1: If the government can set the profit tax optimally, it should levy a Pigovian tax on polluting inputs and levy a wage subsidy that completely

252

The international yearbook of environmental and resource economics

offsets the markup between gross and net-of-tax wage rate as determined in the wage negotiations.

Proposition 3.1 establishes a first-best solution as it shows that the government can fully internalize the environmental externality and can control the labor market imperfection. Even in the presence of labor market imperfections there is no scope for a double dividend as long as the government is not restricted in the usage of its tax instruments.18 Thus the existence of a non-environmental dividend is dependent on the government’s inability to use all possible tax instruments. When the government is restricted in the use of the non-distortionary profit tax, the profit tax constraint is binding (0) and the profit tax rate is set at some upper bound for the profit tax rate t. The government now has to apply distortionary taxes to raise revenues. This can be seen from equation (3.12). As profits are always positive, we can infer that 1, i.e. the marginal cost of public funds exceeds unity. But this is not the only distortion the economy faces. The labor market constraint also becomes binding because, intuitively, the government has to apply distortionary taxes to finance the wage subsidy. If the government allows for a marginal markup – starting at market clearing gross wages – by lowering its subsidy, the resulting labor market distortion causes a negative second-order effect on welfare, while the respective lower tax revenue requirement generates a first-order welfare gain. This is a standard secondbest result according to which, in the presence of more than one distortion, it is not optimal to establish the first-best solution in only one sector. Formally, the shadow price , which represents the social costs of labor market imperfection, can be signed by subtracting (3.15) from (3.13):





wt (1 tw) 1
(  1)wL  0.

www  w www

(3.19)

The term in brackets on the left-hand side is positive (cf. Appendix 2 to this chapter). Hence, condition (3.19) holds only if 0, i.e. reducing the labor market distortion due to wage negotiations is always welfare improving. The lower the net-of-tax wage rate as a result of the wage negotiation, the lower the welfare loss of distorting taxes will be. This will be true irrespective of the question of whether the net-of-tax wage rate changes as a consequence of a tax rate change. We have already seen that the elasticity of substitution between factors of production determines whether the wage negotiation is affected by changes in factor taxation. Solving the system of equations (3.13)–(3.14) for the CES production function case with respect to the tax rates and

253

Double-dividend hypothesis of environmental taxes

making use of 1 and 0, we obtain the general optimal factor tax formulae (cf. Appendix 2)

 

 





tq 1 ER 1 1

wtw(1 tw)
  1  (1  t)  , 1 tq q  (1  s)cY





   



(3.20)



tw 1 wb 1 1

wtw(1  tw)
  1  (1  t)  , 1  tw w  scY

(3.21)

where wt
(w  b) w



  s (1  s)(1  ) [    1  ]
0 ⇔ 
1. 1  tw   (3.22)







To interpret the result, we first consider a Cobb-Douglas production function where the elasticity of substitution is unity and the net-of-tax wage rate is independent of the tax rates. The optimal factor tax formulae for this case are

  tq 1  tq



tw 1  tw







1





1

 

1 ER 1 1  1  (1  t), q 

   

1 wb 1 1  1  (1  t). w 

(3.20a)

(3.21a)

Equation (3.20a) shows that when the price elasticity of output demand  is less than infinite the energy tax should be imposed for two reasons. First, it should be taxed to internalize the external effects caused by using polluting inputs in production. As taxation becomes distortionary, however, it becomes more costly to provide the public good ‘environmental quality’, thus the environmental tax component is smaller than the Pigovian tax. The reason is the same as in the case with perfect labor markets as discussed in Section 2 so that we do not have to interpret the result again. Second, energy should be taxed to raise revenues. The positive second component of the energy tax – which might be once again called the Ramsey component – results from restricted profit taxation that forces the government to rely on distortionary taxation. The energy tax rate is higher the lower the feasible profit tax rate t and the higher the marginal cost of public funds .19

254

The international yearbook of environmental and resource economics

A comparison of equation (3.21a) with the optimal labor tax formula for unrestricted profit taxation, equation (3.18), shows that the labor tax rate is higher when profit taxation is restricted. The first term on the right-hand side represents the subsidy component of the tax rate and is increasing in the marginal cost of public funds . The subsidy has to be financed by distortionary taxes and becomes more costly with higher . The second positive term, which one might refer to as the Ramsey component of the labor tax rate, represents the optimal tax one should levy on labor to minimize the excess burden of taxation. The wage subsidy, which becomes smaller as taxation becomes more costly, is at least partially offset by this Ramsey component. Hence, in the case of Nash wage bargaining with restricted profit taxation, a positive labor tax is possible as a part of the optimal tax treatment of factors of production. These results can be summarized in PROPOSITION 3.2: If profit taxation is restricted and factor taxes have no effect on wage negotiations, the government should use the energy tax to both internalize the external effect and to raise revenues. As the Ramsey tax component is the same for both taxes, the environmental tax always exceeds the labor tax rate.

A consequence of Proposition 3.2 is that introducing green taxes to about the level of the labor tax rate guarantees that welfare will improve – irrespective of the magnitude of the environmental damage. Again we have identified an interval for the green tax rate, which depends on the labor tax rate and the tax revenue requirement, where a second dividend in its strong definition can be achieved. This confirms the result derived by Koskela, Schöb and Sinn (1998) in a tax reform model. If the net-of-tax wage rate is affected, there is an additional term in each tax formula – the second and third terms on the right-hand side in (3.20) and (3.21), respectively. These terms capture the effect that changes in the net-of-tax wage rate will have on the optimal factor taxes. Since we have already discussed the other terms, we will focus on these new terms only. From equation (3.20) we can deduce PROPOSITION 3.3: If profit taxation is restricted and factor taxes affect the wage negotiation, the optimal energy tax should be adjusted downwards (upwards) if the elasticity of substitution between energy and labor is smaller (greater) than 1.

This result has a natural interpretation. If the elasticity of substitution between energy and labor is greater than 1, a rise in the energy tax rate – compared to the case where the wage negotiations are not affected by tax rate changes – decreases the net-of-tax wage rate so that the labor market

Double-dividend hypothesis of environmental taxes

255

distortion due to the difference between the net-of-tax wage w and the social marginal cost of labor becomes smaller. There is thus an additional channel by which the employment dividend can be increased: raising the green tax lowers the market power for trade unions and alleviates the labor market distortion. On the contrary, if the elasticity of substitution is less than 1, a rise in the energy tax rate will increase the net-of-tax wage rate and thus the labor market distortion. The optimal labor tax rate has to be adjusted by going in the opposite direction. With the elasticity of substitution being less than 1, a rise in the labor tax rate decreases the net-of-tax wage rate so that the labor market distortion becomes smaller and vice versa. While the energy tax rate always reaps a second dividend when the energy tax rate is lower than the labor tax rate if 1, the sign of the second dividend becomes ambiguous for 1. If  is less than 1, it is optimal to increase the labor tax rate and decrease the energy tax rate to alleviate the labor market distortion starting from optimal tax rate for 
1. In this case, knowledge about the magnitude of both the elasticity of substitution and the marginal environmental damage is required to determine the relative size of the tax rates and to determine the interval in which a second dividend occurs. If the elasticity of substitution is not too far below 1, and if the marginal environmental damage is considered to be significant, it is still likely that the energy tax rate exceeds the labor tax rate. 3.4

Related Literature

More recently, many papers deal with environmental tax reforms in the presence of involuntary unemployment and discussed the scope for an employment dividend. In a model with fixed net-of-tax wages, Bovenberg and van der Ploeg (1996, 1998a) come to similar results to ours. They show that if green taxes are low initially, employment may increase if substitution between labor and resources within the production sector is easy. Bovenberg and van der Ploeg (1998b), using a search-theoretic framework, found a positive employment effect for a revenue-neutral green tax reform that both increases the tax on a polluting factor of production and succeeds in shifting the tax burden away from labor income to transfer income. In an efficiency wage model, Schneider (1997) also shows that employment may increase due to an increase in green taxes.20 Koskela and Schöb (1999) apply a model with endogenous wage negotiations between trade unions and firms. Using the right-to-manage approach (cf. Nickell and Andrews, 1983), they elaborate different institutional settings and their importance for the employment effect. Their main finding is that if unemployment benefits are nominally fixed and are taxed at a lower rate than wage income, a revenue-neutral green tax reform, which increases green taxes on the

256

The international yearbook of environmental and resource economics

consumption of a polluting good, alleviates unemployment. Holmlund and Kolm (2000) examine the role of environmental tax reforms in a small open economy with monopolistic competition. They show for a Cobb-Douglas technology and a two-sector economy that a revenue-neutral green tax reform boosts employment if wages in the tradable sector are higher than in the non-traded sector. The analysis here as well as all these papers have in common that the existence of the employment dividend crucially hinges on the substitutability of labor and energy and the possibility to shift the tax burden away from labor. In a bargaining model where the firm can invest in abatement technologies, Strand (1999) shows that rebating green tax revenues by either subsidizing firms’ hiring or investments in abatement, pollution declines while employment may increase thus creating a double dividend. The results of the preceding subsection are also sensitive to the assumption made about the outside option. This aspect of green tax reform analysis is often raised when short-run and long-run effects are compared. Brunello (1996) and Carraro, Galeotti and Gallo (1996) argue that, opposite to the assumptions made here, the outside option of the trade union is a weighted average of unemployment benefit payments and wage income from being employed elsewhere (cf. Layard and Nickell, 1990 and Layard, Nickell and Jackman, 1991). They show that in this case, in the long run the trade union succeeds in raising the net-of-tax wage rate at the same amount the labor tax rate is reduced, thus eliminating the short-run employment dividend. Indeed, this result can be reproduced in the framework developed here by defining b in fixed proportion of the net-of-tax wage rate w. Carraro, Galeotti and Gallo (1996) provide numerical simulations of the effects of a carbon tax reform in a bargaining model, which indicate some evidence in favour of an employment dividend in the short run but not in the long run.21 The assumption of a constant replacement ratio may be questionable though. Blanchard and Katz (1999) argue that the income of the unemployed does not consist of unemployment benefit payments only but also comprises non-market income. They assume that the replacement ratio is homogeneous of degree zero in the wage rate and non-market income. If the latter remains constant due to tax rate changes, the replacement ratio would decline and, consequently, the long-run employment dividend would continue to be positive, even though the quantitative effect would be smaller in the long run. This can be seen by splitting the income of the unemployed into two components. The first is proportional to the netof-tax wage rate while the second is a constant one. Furthermore, it should be noted that although unemployment benefits are often paid in proportion to the wage rate (cf. MISSOC, 1998) other additional welfare transfers

Double-dividend hypothesis of environmental taxes

257

are often cut if other income components rise. For low-qualified workers it is thus more reasonable to assume constant unemployment benefit payments rather than a constant replacement ratio as the latter would give them an income below the guaranteed existence minimum income. The employment effect may therefore still be positive in the long run. So far, we have implicitly assumed that the environmental dividend always holds and argued that a double dividend occurs if a green tax reform yields an employment dividend. However, it may not be as clear that the green tax reform may also yield an environmental dividend. Schöb (1998) – for the case of perfect labor markets – and Bayindir-Upmann and Raith (2003) – for the case of imperfect labor markets – show that a revenue-neutral green tax reform may actually worsen rather than improve the environment as the income effect due to higher employment may overcompensate the substitution effect due to higher taxes on polluting goods. Although the deterioration of the environment may be consistent with welfare maximization, this is an important aspect to take into consideration if governments commit to meet certain environmental standards.

4

INTERNATIONAL ASPECTS OF ENVIRONMENTAL TAXATION

So far we have looked at two possible second dividends, reducing the deadweight loss of a distortionary tax system and boosting employment in economies suffering from involuntary unemployment. Two further potential second dividends will be discussed in this section. In the following two sections 4.1 and 4.2, we will analyse the scope for a ‘competitiveness dividend’. This is an issue that frequently pops up in the political debate about unilaterally introducing green taxes, and, therefore, deserves a deeper theoretical analysis. In Section 4.3 we consider the case where countries that pursue coordinated environmental policy may levy green taxes to affect international tax incidence in its favor. 4.1

Green Taxes and Competitiveness: Clearing Labor Markets

Competitiveness is a rather vague concept of an economic policy objective. However, as it is very present in the political debate about green tax reform, pure economic theory should be applied to that concept. In this section, we will consider the impact green taxes may have on the competitiveness of a small open country, first for countries with fully flexible labor markets (Section 4.1) and then for countries suffering from unemployment (Section 4.2).

258

The international yearbook of environmental and resource economics

We start with an appropriate definition of the competitiveness of an economy. Competitiveness is not an end in itself but a useful notion for understanding the reaction to a country’s policy moves. In line with Alesina and Perotti (1997), we measure competitiveness by the negative of the unit production cost of its exports. In general, the production costs are a function of the gross-of-tax factor prices and the output level, C(w, q,Y ). For the linear-homogeneous production function used in Section 3, we have C(w, q,Y )
c(w, q)Y, where c is the unit production cost. The lower c is, the more the country can sell in the world market for Y, and the higher its competitiveness. Does a green tax on internationally mobile energy resources weaken the competitiveness of a small open economy or will it reap some nonenvironmental benefits? Looking at economies with fully flexible labor markets, this question can be answered by looking at the related literature on taxing mobile capital at source (cf. Richter and Schneider, 2001; Koskela and Schöb, 2002a, 2002b). MacDougall (1960) was the first to point out that taxing a perfectly mobile factor at source would always decrease the welfare of a country. In the 1990s, this question was discussed again with a particular focus on tax competition between countries (see e.g., Bucovetsky and Wilson, 1991; Razin and Sadka, 1991). The way in which a green tax reform affects the competitiveness can be shown graphically. In Figure 6.3, the upper part shows the labor market, in which the initial equilibrium is characterized by the gross wage rate w1 and the employment level L1. The lower part shows the energy market. Without green taxes, the firm will increase energy input up to the point where the marginal productivity of energy equals the producer price q. Given the employment level L1, the country will import the amount of energy R1 and has to pay the resource owners E H K I . This area is equivalent to the triangle AGD in the labor market diagram. On the other hand, the triangle A H E in the energy market diagram equals the gross wage income, represented by the rectangle DGPM in the labor market diagram.22 Note that by assuming a linear-homogenous production technology and perfect competition in the output market, the gross wage rate is completely determined by the energy price q and the world output price p. Now assume that the government unilaterally imposes an energy tax tq. As the whole burden falls on the firm, energy input must become more productive. This can be achieved by reducing energy input for any level of employment. As ceteris paribus a higher resource price increases marginal costs, the gross wage rate has to fall to keep marginal costs constant. This induces a reduction in employment as the workers will work less if the wage rate falls. The new equilibrium will be achieved with the gross wage rate w3 that ensures constant marginal costs. The new equilibrium employment level

259

Double-dividend hypothesis of environmental taxes

~ w A LS C D

~1 w ~3 w

F

H

G

I

fL (R1,L) fL (R3*,L)

L M

O L3

fL (R3,L)

P L1

L

q~ A'

B' q~

fR (R,L1)

C' D'

q E' I'

H'

F'

G' J'

R3 Figure 6.3

fR (R,L3) K'

R3*

R1

R

Energy taxation and competitive labor markets

is L3, the new equilibrium energy input level is R3 and the new domestic gross energy price is q
q(1tq). Wage income falls to HIOM (in the labor market diagram), the energy costs are E F J I (in the energy market diagram) with C D F E being the energy tax revenues. Competitiveness in our definition does not change as long as the world output market determines marginal costs. Nevertheless, the decision whether to introduce energy taxes unilaterally has severe negative consequences

260

The international yearbook of environmental and resource economics

for domestic welfare, which become important if a country has to decide whether it should meet e.g. the Kyoto Protocol requirements of reducing CO2 emissions. To see the welfare effect, assume that the government introduces a labor tax instead of an energy tax that also ensures an equilibrium employment level L3. As the whole tax burden falls on workers, the gross wage is not affected, but the net-of-tax wage rate falls to w3. The labor tax rate thus equals the distance DH
w1w3. The new marginal labor productivity curve intersects with the gross wage curve at F. The energy input necessary to sustain this equilibrium can be found by the intersection of the marginal energy productivity curve fR(R,L3) with the gross energy price curve q. It turns out that the energy input R3* is larger than in the case where the energy tax leads to the employment level L3. The welfare loss of the labor tax equals the area FGI shaded in grey, which is associated with the tax revenues of DFIH. The total loss in workers’ net-of-tax wage income can also be seen in the energy market diagram. Here the loss is given by the area A H G B . Now, assume that the government replaces the labor tax and levies an energy tax that ensures the same employment level L3. Such a reform further reduces domestic income, which consists of the net-of-tax wage income plus tax revenues, from the triangle B G E in the energy market diagram to B D C C D F E
B D F E . The additional loss equals the triangle B G E shaded in grey. As the net-of-tax wage rate remains the same, w3, the total additional loss results from a loss in tax revenues. The energy tax thus has two negative welfare effects. First, there is the welfare loss resulting from a fall in the net-of-tax wage rate from w1 to w3 as the fall in the labor rent DGIH is only partly compensated by the tax revenues DFIH. Second, obtaining the same labor rent with an energy tax instead of a labor tax results in lower tax revenues by the amount of D G F . The total welfare loss is FGID G F . Hence, applying the energy tax instead of a labor tax, the second dividend of the energy tax is unambiguously negative. Figure 6.3 illustrates that a small open economy can maximize tax revenues for any given employment level by setting the green tax equal to zero. By contrast, one can infer from the analysis that for any given level of public expenditures, employment and energy input is maximized. As the labor rent is increasing with employment, welfare is maximized as well. As both inputs are maximized if there is no energy tax, output is maximized as well. So far, we have assumed a constant world market price for the output good. If the output demand is downward sloping – as assumed in Section 3 – the output maximum is also a unit cost minimum. Hence, raising the energy tax weakens the competitiveness of economies facing

Double-dividend hypothesis of environmental taxes

261

downward sloping demand curves for their products. Whether welfare or competitiveness is the policy objective, the conclusion that energy taxes are harmful is the same. If energy is perfectly mobile, there is no second dividend and thus no reason other than improving the environment for imposing green taxes. For economies with fully flexible labor markets, this analysis confirms Bovenberg’s (1999) conclusion that ‘the case for environmental taxes should be made primarily on environmental grounds . . . green taxes are worthwhile as long as the environmental benefits are nonnegative’ (p. 441). Proposition 4.1 summarizes. PROPOSITION 4.1: If the labor market is competitive, a small open economy (with respect to energy demand) should not levy a tax on mobile energy for other reasons than to improve domestic environmental quality. Neither welfare nor competitiveness of the economy can be improved by employing energy taxes.

The next subsection shows that this conclusion does not hold for economies with a unionized labor market and thus suffering from involuntary unemployment. 4.2

Green Taxes and Competitiveness: Imperfect Labor Markets

In Section 3.2 we have shown (see Figure 6.2) that in the presence of involuntary unemployment due to wage negotiations between trade unions and firms, there exists a green tax equilibrium with higher tax rates on energy than on labor that yields the same level of output and same tax revenue as, but a higher level of employment than, the existing labor tax equilibrium. As firms face the same unit cost of production in the green tax system as in the labor tax system, the move from labor tax equilibrium to green tax equilibrium maintains the economy’s international competitiveness in the sense of our definition of keeping the unit production cost and the terms of trade constant. More, however, can be said if we analyse marginal revenue-neutral green tax reforms. Then we can allow for a change in the output level that goes along with changes in the unit cost of production, and hence the competitiveness of the country. To analyse the change in the competitiveness of the economy, we thus have to calculate the effect, a marginal revenue-neutral green tax reform has on the unit cost of production (see Appendix 3 to this chapter): sign



dc dtq

|  dG
0


sign (tq  tw).

(4.1)

262

The international yearbook of environmental and resource economics

If involuntary unemployment is not completely eliminated, the following proposition summarizes. PROPOSITION 4.2: As long as the labor tax rate exceeds the energy tax rate, a marginal revenue-neutral green tax reform will increase both the international competitiveness and the output of an economy with involuntary unemployment due to too high net-of-tax wages. Competitiveness is maximized when the energy tax rate equals the labor tax rate.

To interpret and understand these results it is useful to consider Figure 6.4, which is constructed in the same way as Figure 6.2. Figure 6.4 shows two conceivable paths of consecutive marginal tax reforms starting in the labor tax system A and ending in the green tax system B. Up to points C or C ; where tw
tq, output will increase. A further increase in tq will result in marginal output reductions. Up to point C or C employment will increase. However, an increase of tq sufficiently far beyond the point where tq
tw will not necessarily increase employment further as there is a countervailing output effect. A green tax reform will definitely create the incentive to substitute employment for energy consumption. The fall in output such a reform induces when tq tw L

C' LB

B C

Path II

Path I A

LA RB Figure 6.4

RA

Marginal green tax reforms

Y0

R

Double-dividend hypothesis of environmental taxes

263

will, in itself, reduce the factor demands. If tq is sufficiently far above tw, the output effect may dominate the substitution effect such that employment declines at the margin. With paths I and II, Figure 6.4 distinguishes two different possibilities that depend on the price elasticity of the demand curve for the economy’s output. If the demand elasticity is small, the initial rise and subsequent fall in output will be small and the substitution effect will dominate the output effect. This case is represented by path I. Moving from C to B further increases employment while output declines. If output demand is very priceelastic, as represented by path II, there will be an interval on the path II from C to B where output and employment are falling simultaneously.23 The ambiguity with respect to output translates to the country’s international competitiveness. Since equation (4.2) says that the terms of trade are a declining function of output, the economy’s competitiveness increases with a marginal green tax reform as long as tq tw (right of C and C ) and declines when tq tw (left of C and C ). Hence, initial green tax reforms not only raise employment and welfare – as long as the output effect is positive, they also increase the competitiveness of an economy: there is hence a competitiveness dividend that goes along with the employment dividend and the non-environmental welfare dividend, respectively. From a welfare perspective, there is therefore no need to worry about a competitiveness dividend as its effect is already covered in the non-environmental welfare dividend. Insofar as politics is concerned about competitiveness rather than welfare, however, it is important to show that these two dividends are of the same sign as long as the energy tax falls short of the labor tax rate. Note, however, that going beyond C or C – at least marginally – is still welfare improving although it reduces competitiveness. The standard result in the optimal taxation literature – applied to green taxation in Section 4.1 – is that a small open economy would be worse off if it substitutes a tax on a mobile factor such as energy for a labor income tax. By contrast, we can conclude for an economy with unionized labor markets that the effects of such a green tax reform are favourable: a green tax reform will induce a technical substitution in the production process that replaces energy use with employment. Since energy is priced at its true national opportunity cost, but the price of labor is above its social opportunity cost, there is a strong presumption that the reform will boost employment and bring about an increase in the competitiveness of a country. 4.3

Environmental Taxes are Resource Taxes

Environmental externalities are typically tied to the use of exhaustible resources. A comprehensive analysis of environmental taxation should

264

The international yearbook of environmental and resource economics

therefore take into account the impact of green taxes on the world producer prices of exhaustible resources such as gas and oil products. At first glance, this is about international tax incidence only, but as international tax incidence has severe impacts on national welfare, it is also closely related to our main theme, the double-dividend hypothesis of environmental taxation. The literature on the double dividend has not yet recognized this important relation although it is well known from resource economics that taxing exhaustible resources can offer resource consuming countries the possibility to extract resource rent. This is of particular interest when it comes to internationally coordinated environmental policies to fight global pollution such as greenhouse gas emissions. Resource prices are principally determined by the user cost of the resource, i.e. the rent the resource owner obtains from extracting the resource. An energy tax, introduced by a sole country with a negligible share in global energy demand, does not affect the world energy price. It is therefore optimal for such a small country to impose a green tax, which leads the country to meet its own environmental standards – as discussed in the last subsection – or to meet international environmental agreements like the Kyoto Protocol. For the welfare analysis it is not necessary to incorporate the reactions on the world resource markets. If all countries introduce environmental taxes, however, the burden of the green tax will partly be borne by today’s resource consumer and partly by the resource extracting country. The welfare effects are then twofold and should be considered separately. The main underlying idea can be described graphically. For simplicity assume first that the whole resource stock can be used in one period only. If we abstract from extraction costs, this implies that consumers face a fixed supply R 0. In the absence of pollution, it is optimal to extract the whole stock if the marginal willingness to pay at R 0, MWP(R 0) is positive. In Figure 6.5, total surplus is maximized at R 0, yielding a surplus of ACR 00. In the presence of externalities, indicated by the marginal environmental damage curve MEDi, i
1,2, it remains optimal to consume the whole stock as long as MWP(R 0)MED(R 0)0, i.e. if the marginal net rent of consuming the last unit of the resource is still positive. The existence of a resource rent then completely compensates for the externality, and there is no need for environmental policy measures. A tax smaller than MWP(R 0) does not alter the allocation, but affects the distribution of rents as the entire tax burden falls on the supplier. Without generating an environmental dividend, a coordinated ‘green tax’ could reap a second dividend for resource consuming countries – at the expenses of the resource owning countries. If the environmental damage is more severe, as indicated by the MED1 curve, it becomes optimal to reduce extraction from R 0 to R2. In this case,

265

Double-dividend hypothesis of environmental taxes

MWP MED

MED1 A E

MED2

B tp

C D MWP

0 Figure 6.5

R2

R0

R

The one-period model of exhaustible resource consumption

a Pigovian tax tp
MED1(R2) maximizes welfare, given by AB0. The welfare gain, compared to the laissez-faire solution equals BEC. Next, we consider the case where a fixed resource stock can be consumed in two periods. This is illustrated in Figure 6.6 where the total resource stock is plotted on the horizontal axis and the period demands on the vertical axes. In the absence of pollution, the social benefit from resource consumption can be represented by the aggregate marginal willingnessto-pay curve MWPi. In the presence of externalities, however, we have to subtract the marginal environmental damage, caused by the consumption of the natural resource, from the marginal willingness to pay in order to derive total welfare. This yields the marginal social benefit curve MSBi. For period 2, marginal costs and benefits are discounted by the interest rate r, which, in the case of perfect capital markets, equals the social discount rate. Without internalizing externalities the market equilibrium equalizes the present value of the marginal willingness to pay for both periods, i.e. MWP1
MWP2 /(1r). This maximizes the present value of the consumer surplus. In Figure 6.6, the curves for the first period are plotted from left to right, the curves for the second period are plotted from right to left. Thus, the distance 01B denotes the consumption in the first period and 02B the consumption in the second period, respectively. With externalities,

266

The international yearbook of environmental and resource economics

MWP1 MSB1

MWP2 MSB2

D

p1

C2 MWP2/(lr)

p2/(lr)

C1 MWP1

01

Figure 6.6

02

A B MSB2/(lr)

MSB1

A two-period model of exhaustible resource consumption

however, a reduction of the first period’s consumption, accompanied by an increase of second period’s consumption enhances welfare. This can be achieved by e.g. introducing a Pigovian tax tp equal to C1D in the first period and C2D(1r) in the second period, respectively. Such a combination of taxes equalizes the present value of the marginal social benefit i.e. MSB1
MSB2 /(1r). The shaded area shows the resulting welfare gain – compared to the laissez-faire situation. The implications of this simple model can easily be generalized. As resource consumption falls, marginal environmental damages decrease, and so should the Pigovian tax. Ulph and Ulph (1994) and Farzin (1996) therefore emphasize that what matters is the time path of the environmental tax rather than its level. To delay extraction, the initial environmental tax should be high and then fall over time.24 In the two-period model, however, the natural resource will not necessarily be exhausted. If the MSB1-curve intersects the MSB2 /(1r)-curve at negative prices, it would be optimal to consume a smaller amount such that MSBt
0 is ensured in each period. This outcome does not carry over to models with an infinite time horizon problem: if there is a minimum fixed amount of the resource consumed in each period, determined by MSBt
0, exhaustion of the whole resource stock would be beneficial to society – despite the existence of pollution. The implications of this model are not only important with respect to the determination of the optimal time path of environmental taxes. The results

Double-dividend hypothesis of environmental taxes

267

with respect to international distribution are also striking. If resource owning countries have optimized the time path of extraction and if all resource importing countries introduce environmental taxes, total demand will fall in each period if the producer price remains constant. This cannot be an equilibrium, as the total resource stock would not be exhausted and a price taker would have an incentive to increase sales in some periods. As a consequence, both the producer price and the resource rent obtained by the resource extracting country fall. Since the marginal environmental damage is an increasing function of resource consumption, a shift from present consumption towards future consumption is welfare enhancing. A delay in consumption reduces both the absolute amount of emissions and the present value of the environmental damage. One could therefore argue that when consumption of the natural resource is not taxed, resource consuming countries actually subsidize the resource extracting country – by an amount equal to the value of the environmental damage that the households inflict upon themselves. In this sense, it is not the polluter who pays for the internalization of the externality, but the producer of the nonrenewable resource (cf. Amundsen and Schöb, 1999). These distributional implications provide incentives for further coordinating environmental policies in order to capture resource rents even if there are no transboundary or global pollution problems present, which are normally considered as the reason to coordinate environmental policies (e.g. Hoel, 1992). For instance, resource consuming countries may exploit the Kyoto Protocol by using a carbon tax as a new form of tariff, which allows these countries to capture some of the resource rents. If, for example, all resource consuming countries agree to introduce a carbon tax that increases with the real interest rate, we know from the literature on optimal resource taxation that such a coordinated per unit resource tax would leave the time path of extraction and hence resource consumption completely unaffected. In Figure 6.6, such a tax would lower the demand curves the supplier faces but would not alter the intertemporal allocation represented by point B. This follows directly from the Hotelling rule (see, e.g., Dasgupta and Heal, 1979; Sinn, 1982). Such a tax has no effect at all on the environment but allows the countries to reap a second nonenvironmental dividend. Indeed such a tax would be a pure rent capturing tax.25 If the resource owning countries can also exercise market power, they may attempt to raise the initial resource price, because this would reduce the environmental tax and allow the resource owner to capture some of the tax revenues that the resource consuming countries would otherwise collect (cf. Wirl, 1994). The resource consuming countries then fail to completely extract rents. Nevertheless, coordination would always allow the

268

The international yearbook of environmental and resource economics

resource consuming countries to capture some of the resource rent (and the monopoly rent). Indeed, as was shown by Karp and Newbery (1991), in the absence of externalities, the buyer’s market power exceeds that of the sellers, as they succeed in reducing the initial producer price. If the OPEC countries commit to raising oil prices, it might increase political pressure in resource consuming countries to reduce the high taxes on fuel. This happened in the fall of 2000 when French truck drivers forced the French government to reduce fuel taxes. A single country can actually reduce consumer prices by lowering the green tax. If all countries do so, however, the whole tax reduction would result in an increase in the producer price for the reason given above. The lasting discussion of whether or not to implement a carbon tax may have adverse effects on the environment. Announcing the imposition of coordinated carbon taxes (even if the taxes are not intended to extract rents) acts like an expropriation threat to resource owners. As a consequence, the resource owning countries have incentives to increase present extraction prior to the date the tax is introduced, so as to reduce future losses (cf. Long, 1975; Konrad, Olsen and Schöb, 1994).

5

CONCLUDING REMARKS

A double dividend of environmental taxes arises when governments have only a limited set of policy instruments available to regulate both pollution externalities and other market imperfections in the economy. In such circumstances, green tax reforms not only affect the environment but also alleviate or exacerbate other distortions in the economy. As the discussion of this chapter shows, it is then optimal to deviate from first-best policy recommendations. In this respect, the double-dividend literature is an application of classical second-best analysis. Three of these second-best problems were at the core of this chapter. We started with a review of the standard results of the double-dividend literature and argued that a second dividend in its strong definition can only occur when the tax system has not been second-best optimal without considering the environment. When analysing environmental taxation in economies with labor market imperfections due to unionization, we find that an employment dividend can be obtained since governments are normally not able to employ all tax instruments needed to eliminate involuntary unemployment. The employment dividend will occur when environmental taxation allows the government to shift the tax burden away from labor. The last section discussed the double-dividend issue in an international context. Focusing on the politically interesting question

269

Double-dividend hypothesis of environmental taxes

of a ‘competitiveness dividend’ we showed that such a dividend goes along with a non-environmental dividend and cannot be considered as a benefit that should be taken into account in addition to other benefits. Insofar as environmental taxes are taxes on imported exhaustible resources, green taxes can reap an additional ‘rent extraction dividend’ when resource consuming countries coordinate their environmental policies. Although this establishes a second dividend from the viewpoint of the resource consuming countries it does not so with respect to global efficiency. The multitude of possible effects discussed here indicates how sensitive the results are to the appropriate modeling of the underlying institutional framework. Here we have only studied the interaction of environmental regulation with the pre-existing tax system, the labor market institutions, and aspects of international cooperation. To determine both the sign and magnitude of both environmental and non-environmental dividends, we need more detailed knowledge of the institutional framework in which a green tax reform takes place, of the technology of polluting goods, and of other possible sources of economic inefficiency to come to sound policy recommendations. On top of all this, the tradeoff between efficiency and distributional considerations needs much more careful evaluation when environmental policy proposals enter the political process – both with respect to the welfare implications and the question of implementability. All these areas therefore promise significant payoffs of further research.

APPENDIX 1: SECOND-BEST OPTIMAL ENVIRONMENTAL TAX Rewriting the equation system (2.9) and (2.10) for the case of homogenous households yields:



xc xc E  tc E tc xc xc E  td E td

  

xd xd E  u E  x  E tc E tc tc

c tc xd xd E t
 u E d  x  E td E td

d td



(A1.1)

The determinant |D| is given by [using (2.12)] |D|






xc xd xc xd   |J|. tc td td tc

(A1.2)

270

The international yearbook of environmental and resource economics

Applying Cramer’s rule for the clean good yields





 x x x E E xc d  xd d  d xc  xd

td tc E td tc tc
|J|

  

uE E xd E xd 

td tc tc td



.

Applying the definition (2.12) shows that the last term of the numerator is zero. Furthermore, applying (2.12) again, we have



xc



xd x x E E  xd d  d xc  xd td tc E td tc

 


1 H




 xc

xd e E



xc

xd x  xd d td tc





xd x  xd d . td tc

Using the Slutzky equation, we finally obtain



 

 x x xc sdd  xd d  xd sdc  xc d

T T tc
|J|





 R t .

c

(A1.3)

For the dirty good, we have

td




xc xc E  tc E tc







 u E xd  E

td |J|

xc xc E  td E td



 

 u E xc  E

tc . |J|

Multiplying through, using (2.12) and rearranging yields:









 xc  x x x x xd c  xc c  e xd d  xc d H

tc td

E tc td Hu e td
 E . |J|

(A1.4)

271

Double-dividend hypothesis of environmental taxes

The first term can be rewritten (using the Slutzky decomposition) as follows:





 

 

  x x x x xd c  xc c xd scc  xc c  xc scd  xd c

tc td

T T
|J| |J|







1  e H

xd E



 (xd scc  xc scd )

|J |





 R  R xd t  t H e .

d

d E

(A1.5)

Substituting in the tax on the clean good (A1.3) allows us to simplify the second term of (A1.4) as well. Finally, substituting (A1.4) in (A1.5) gives: td


 R

 x t  t  tRH i e .

d P

i
c,d i E




(A1.6)

Adding td e Hxd E on both sides, adding the two terms of the righthand side with the Ramsey components of the dirty good together and substituting in equations (A1.3) and (2.12), this can be rewritten as:

td






 R x td  tP  ti H i e .

E i
c,d




(2.15a)

Applying the several restricting assumptions discussed in the main text gives the respective optimal tax formulae (2.13) to (2.15).

APPENDIX 2: RESTRICTED PROFIT TAXATION Subtracting (3.15) from (3.13) implies that the following condition must hold:

[www  wt (1  tw)]
 (  1)wL  0. t

(A2.1)

272

The international yearbook of environmental and resource economics

The left-hand side of (A3.1) must be negative. Rewriting the terms in brackets yields:



www  wt (1  tw)
www  w



wtw(1  tw) 1 , www

(A2.2)

where the first term in brackets on the right-hand side is the net-of-tax wage elasticity with respect to the labor tax rate. As has been argued above, this elasticity is larger than –1. Hence (A2.2) is negative and the condition (A2.1) holds only if 0. Using this condition, we can derive the optimal tax formulae for the second case when the wage rate changes. For the case 0 and hence t
t rearranging the equations (3.13) and (3.14) yields





wLL,w
qRR,w wLL,q
qRR,q



tw E tq  R q



   



wb wLL,w  wt (1tw) w w
, (A2.3) wb (1  )(1  t)qR  wLL,q  wt (1tw) w w

with

(1  )(1  t)wL 

wt
wt (1  tw) (1  tq) (cf. Koskela and Schöb, 2002b). Applying q w Cramer’s rule and using the fact that the determinant of the left-hand side matrix is equal to 
wLqR yields tw


 

wb (1  )(1  t) wtw(1  tw)  [wLR,q  qRR,w]  , w wL wLR,q  qLR,w (A2.4)

tq 

ER (1  )(1  t) wtw(1  tw)
 [qRL,w  wLL,q]  . q qR qRL,w  wLL,q (A2.5)

Using the explicit elasticity formulae, we have wLR,q  qRR,w
cY(sR,q  (1  s)R,w)
cYs, qRL,w  wLL,q 
cY((1  s)L,w  sL,q)
cY(1  s).

(A2.6) (A2.7)

Substituting in (A2.6) and (A2.7) in (A2.4) and (A2.5) respectively, we obtain conditions (2.20) and (2.21).

273

Double-dividend hypothesis of environmental taxes

APPENDIX 3: COMPETITIVENESS Without loss of generality, we assume that the profit tax is equal to zero. From the government budget condition (3.11) we then get dG
[wL  twwLww  tqqRww] dtw  [qR  tqqRq q  twwLq q] dtq. (A3.1) The elasticities of factor demands are given by R,q ≡ Rq q/R
 (1s)(), R,w ≡ Rw w/R
s(), L,w ≡ Lww/L
s(), L,q ≡ Lq q/L
(1s)(),where s  wL cY denotes the cost share of labor and (1  s)  1  wL cY
qR cY denotes the cost share of energy, and  denotes the constant elasticity of substitution as in Section 3. Substituting these in equation (A3.1) gives



dG
wL 1 



 

tw tq (1  s)   R,w dtw (1  tw) L,w (1  tq) s

qR 1 

tq tw s    dt . (1  tq) R,q (1  tw) (1  s) L,q q

Setting dG
0 yields an expression that reveals how the labor tax rate changes due to a marginal increase of the tax on energy (using the fact that R,w
sL,q (1  s))

 

 

tq tw R,q   (1  tq) (1  tw) R,w dtw
 . dtq dG
0 tw tq wL 1  L,w  L,q (1  tw) (1  tq) qR 1 

|

(A3.2)

To analyse the change in the competitiveness of the economy, we have to calculate the effect, a marginal revenue-neutral green tax reform has on the unit cost of production. The impact of a revenue-neutral green tax reform on the unit cost of production is given by dc(w,q)
cwwdtwcq qdtq. Applying Shephard’s lemma Cw
cwY
L, Cq
cqY
R and using equation (A3.2) allows us to determine the change in the unit cost of production:

 



tq(L,q  R,q) tw(L,w  R,w)  (1  tq) (1  tw) dc dtw .
cww  cq q
t tq dtq dG
0 dtq dG
0 w Y 1 L,w  L,q (1  tw) (1  tq)

|

|

qR



(A3.3)

274

The international yearbook of environmental and resource economics

Assuming positive marginal tax revenues for the labor tax rate (see equation (A3.1)), the denominator is always positive. Substituting in the definitions of the (cross-)price elasticities of factor demands in the nominator of equation (A3.3) yields L,q  R,q
 and L,w  R,w
. This gives us condition (4.1).

NOTES *

1.

2. 3. 4. 5. 6. 7. 8. 9. 10.

11. 12.

13. 14.

Helpful comments by Lans Bovenberg, Henk Folmer, Andreas Knabe, Bertrand Koebel, Sven Wehke, an anonymous referee, and the participants of the annual meeting of the Umwelökonomischer Ausschuss des Vereins für Socialpolitik in Berlin 2002 are gratefully acknowledged. The usual disclaimer applies. The distinction between the weak and the strong form of the double dividend hypothesis is due to Goulder (1995) who surveys the early literature. Recent surveys on the double dividend hypothesis are Bovenberg (1995, 1998, 1999); Schöb (1995); Goulder (1997); Kirchgässner (1998) and Bosello, Carraro and Galeotti (2001). Exceptions are Sandmo (1975), Johansson (1994), Schöb (1995), Pirttilä and Schöb (1999), Bulckaen and Stampini (2001), and Mayeres and Proost (2001). Schöb (1995) and Pirttilä (1998) analyse the more general case where the government also maximizes with respect to the public good provision. Stability is guaranteed as long as the denominator of equation (2.12) is positive (see Schöb, 1995, p. 118). See Orosel and Schöb (1996) for the following. Jaeger (2002, 2003) defines the Pigovian tax rate component in different way. While we define the component in (2.17) in units of the private marginal utility of the public good, he defines the component in units of the social marginal utility of income. Cf. Goulder, Parry and Burtraw (1997), Goulder, Parry, Williams and Burtraw (1999), Parry, Williams and Goulder (1999). Fullerton and Metcalf (2001) point out that it is not necessary that the government raises revenues from environmental policy. What is essential is that the government can capture the rents generated by the environmental policy. Note that the existence of defensive goods would also reduce the marginal environmental damage. The gross social valuation of income is always larger for low-income households than for well-off households. This need not be the case for the net measure, because additional income may lead to larger changes in the demand for the taxed commodities among the high-income households, in which case h would increase. It is also hard to deduce whether the magnitude of the externality encompassing term is greater or smaller for worse-off households and, accordingly, in which direction the differences in the net social valuation move. Note that we have assumed separability between consumption and environmental quality. Pirttilä and Schöb (1999) derive another alternative many-person Ramsey tax rule in the presence of externalities not presented here: if commodity taxes and the lump-sum transfer are set optimally, a small equiproportional increase in all Ramsey tax components will cause all compensated commodity demands to change according to their distributional characteristics. It is true that in the presence of distortionary labor taxation employment is too low from a social point of view. The resulting welfare losses, however, are already taken into account when looking at the gross distortion of the tax system. Note that although we assume a single trade union in the economy it behaves like a small trade union as its policy cannot affect the consumer price level.

Double-dividend hypothesis of environmental taxes 15. 16. 17. 18.

19. 20. 21. 22.

23. 24. 25.

275

Note that 1. For the sake of the argument, we assume that there are no unemployment benefits paid by the government. This assumption does not affect the qualitative results although it does affect the magnitude of the actual optimal tax formulae. This section replicates a thought experiment by Koskela, Schöb and Sinn (1998). This result confirms for a unionized labor market the result by Guesnerie and Laffont (1978) according to which, in a first-best world, the output of a price maker should be subsidized such that the market price equals the marginal cost (see also Boeters and Schneider, 1999). For a thorough analysis of the role profit taxation plays for the determination of optimal tax formulae see Boeters (2001). Also see the comment by Scholz (1998). See Bosello, Carraro and Galeotti (2001) for further references. A linear-homogeneous production technology implies convex marginal productivity curves, which do not intersect with the axis. For didactical purposes Figure 6.3 uses linear curves only. As a consequence, the corresponding areas in the labor market and the energy market are not necessarily of the same size. For the same reason, moving from A to C increases resource demand. This confirms the analysis by Bayindir-Upmann and Raith (2003). Also see Ploeg and Withagen (1991) and Kolstad and Krautkraemer (1993). Newbery (1976) and Bergstrom (1982) were the first to show that resource consuming countries can secure the entire resource rent from the resource owning country by coordinating their tariffs or their national excise tax policies. The theoretical findings are in line with the empirics. For example, in the countries of the European Union, tax rates on gasoline have increased substantially over time. Although these taxes were not primarily introduced to internalize national or global externalities, their effects are similar to those of environmental taxes. Hoeller and Coppel (1992) calculate the implicit carbon tax of fuel taxes and conclude that, at least for most European countries, the implicit carbon tax is considerably higher than the taxes suggested by energy tax reform proposals. Since the mid-1980s the real producer price has fallen while the real tax rate has increased steadily. These countervailing developments have left the consumer price more or less unaffected until the mid-1990s (cf. Amundsen and Schöb, 1999).

REFERENCES Alesina, A. and R. Perotti (1997), ‘The welfare state and competitiveness’, American Economic Review, 87, 921–39. Amundsen, E.S. and R. Schöb (1999), ‘Environmental taxes and exhaustible resources’, European Journal of Political Economy, 15, 311–29. Atkinson, A.B. and J.E. Stiglitz (1980), Lectures on Public Economics, London: McGraw-Hill. Baumol, W.J. and W.E. Oates (1988), The Theory of Environmental Policy, 2nd edn, Cambridge, MA: Harward University Press. Bayindir-Upmann, T. and M.G. Raith (2003), ‘Should high-tax countries pursue revenue-neutral ecological tax reforms?’, European Economic Review, 47, 41–60. Bergstrom, T.C. (1982), ‘On capturing oil rents with a national excise tax’, American Economic Review, 72, 194–201. Blanchard, O.J. and L. Katz (1999), ‘Wage dynamics: reconciling theory and evidence’, American Economic Review, 89, 69–74. Boeters, S. (2001), ‘Green tax reform and employment: the interaction of profit and factor taxes’. ZEW discussion paper no. 01-45.

276

The international yearbook of environmental and resource economics

Boeters, S. and K. Schneider (1999), ‘Government versus union: the structure of optimal taxation in a unionized labor market’, Finanzarchiv, 56, 174–87. Bosello, F., C. Carraro and M. Galeotti (2001), ‘The double dividend issue: modeling strategies and empirical findings’, Environment and Development Economics, 6, 9–45. Bovenberg, A.L. (1995), ‘Environmental taxation and employment’, Der Economist, 143, 111–40. Bovenberg, A.L. (1998), ‘Environmental taxation and the double dividend’, Empirica, 25, 15–35. Bovenberg, A.L. (1999), ‘Green tax reforms and the double dividend: an updated reader’s guide’, International Tax and Public Finance, 6, 421–43. Bovenberg, A.L. and F. van der Ploeg (1994a), ‘Environmental policy, public goods and the marginal cost of public funds’, Economic Journal, 104, 444–54. Bovenberg, A.L. and F. van der Ploeg (1994b), ‘Environmental policy, public finance and the labour market in a second-best world’, Journal of Public Economics, 55, 349–90. Bovenberg, A.L. and F. van der Ploeg (1994c), ‘Green policies and public finance in a small open economy’, Scandinavian Journal of Economics, 96, 343–63. Bovenberg, A.L. and F. van der Ploeg (1996), ‘Optimal taxation, public goods and environmental policy with involuntary unemployment’, Journal of Public Economics, 62, 59–83. Bovenberg, A.L. and F. van der Ploeg (1998a), ‘Consequences of environmental tax reform for unemployment and welfare’, Environmental and Resource Economics, 12, 137–50. Bovenberg, A.L. and F. van der Ploeg (1998b), ‘Tax reform, structural unemployment and the environment’, Scandinavian Journal of Economics, 100, 593–610. Bovenberg, A.L. and R.A. de Mooij (1994), ‘Environmental levies and distortionary taxation’, American Economic Review, 84, 1085–9. Brunello, G. (1996), ‘Labour market institutions and the double dividend hypothesis’, in C. Carraro and D. Siniscalco (eds), Environmental Fiscal Reform and Unemployment, Dordrecht: Kluwer Academic, pp. 139–70. Bucovetsky, S. and J.D. Wilson (1991), ‘Tax competition with two tax instruments’, Regional Science and Urban Economics, 21, 333–50. Bulckaen, F. and M. Stampini (2001), ‘Efficiency and distributional effects of revenue-neutral green tax reforms’, Studi Economici, 56, 25–39. Carraro, C., M. Galeotti and M. Gallo (1996), ‘Environmental taxation and unemployment: some evidence on the “Double Dividend Hypothesis” in Europe’, Journal of Public Economics, 62, 141–81. Dasgupta, P.S. and G.M. Heal (1979), Economic Theory and Exhaustible Resources, Cambridge: Cambridge University Press. Diamond, P.A. (1975), ‘A many-person Ramsey tax rule’, Journal of Public Economics, 4, 335–42. Farzin, Y.H. (1996), ‘Optimal pricing of environmental and natural resource use with stock externalities’, Journal of Public Economics, 62, 31–57. Feldstein, M.S. (1972), ‘Distributional equity and the optimal structure of public prices’, American Economic Review, 62, 32–6. FitzRoy, F. (1996), ‘Environmental taxes, non-separable externalities and real wages’, mimeo. Fullerton, D. (1997), ‘Environmental levies and distortionary taxation: comment’, American Economic Review, 87, 245–51.

Double-dividend hypothesis of environmental taxes

277

Fullerton, D. and G.E. Metcalf (2001), ‘Environmental controls, scarcity rents, and pre-existing distortions’, Journal of Public Economics, 80, 249–67. Fullerton, D. and A. Wolverton (2003), The two-part instrument in a second-best world’, NBER working paper series no. 10140, December. Goulder, L.H. (1995), ‘Environmental taxation and the “Double Dividend”: a reader’s guide’, International Tax and Public Finance, 2, 157–84. Goulder, L.H. (1997), ‘Environmental taxation in a second-best world’, in H. Folmer and T. Tietenberg (eds), International Yearbook of Environmental and Resource Economics 1997/1998: A Survey of Current Issues, Cheltenham, UK and Northampton, MA: Edgar Elgar, pp. 28–54. Goulder, L.H., I.W.H. Parry and D. Burtraw (1997), ‘Revenue-raising vs. other approaches to environmental protection: the critical significance of pre-existing tax distortions’, Rand Journal of Economics, 28, 708–31. Goulder, L.H., I.W.H. Parry, R.C. Williams III and D. Burtraw (1999), ‘The cost-effectiveness of alternative instruments for environmental protection in a second-best setting’, Journal of Public Economics, 72, 329–60. Guesnerie, R. and J.-J. Laffont (1978), ‘Taxing price makers’, Journal of Economic Theory, 19, 423–55. Hamermesh, Daniel S. (1993), Labor Demand, Princeton, NJ: Princeton University Press. Harrison, D. Jr. (1994), The Distributive Effects of Economic Instruments for Environmental Policy, Paris: OECD. Hoel, M. (1992), ‘Carbon taxes, an international tax or harmonized domestic taxes’, European Economic Review, 36, 400–6. Hoeller, P. and J. Coppel (1992), ‘Carbon taxes and current energy policies in OECD countries’, OECD Economic Studies, 19, 167–93. Holm, P., S. Honkapohja and E. Koskela (1994), ‘A monopoly union model of wage determination with capital and taxes: an empirical application to the Finnish manufacturing’, European Economic Review, 38, 285–303. Holmlund, B. and A.-S. Kolm (2000), ‘Environmental tax reform in a small open economy with structural unemployment’, International Tax and Public Finance, 7, 315–33. Jaeger, W.K. (2002), ‘Carbon taxation when climate affects productivity’, Land Economics, 78, 354–67. Jaeger, W.K. (2003), ‘Environmental taxation revisited’, Department of Economics, University of Oregon, mimeo. Johansson, O. (1994), ‘Optimal indirect taxation in a second-best perspective with regard to externalities, a public budget restriction and distribution effects’, University of Göteborg, Department of Economics, mimeo. Karp, L. and D.M. Newbery (1991), ‘OPEC and the US oil import tariff’, Economic Journal, 101, 303–13. Kirchgässner, G. (1998), ‘Ökologische Steuerreform: Utopie oder realistische Alternative’, in G. Krause-Junk (ed.), Steuersysteme der Zukunft, Schriften des Vereins für Socialpolitik, Bd. 256, Berlin: Duncker & Humblot, pp. 279–319. Kolstad, C.D. and J.A. Krautkraemer (1993), ‘Natural resource use and the environment’, in A.V. Kneese and J.L. Sweeney (eds), Handbook of Natural Resource and Energy Economics Vol. III, Amsterdam et al. Elsevier, pp. 1219–65. Konrad, K.A., T.E. Olsen and R. Schöb (1994), ‘Resource extraction and the threat of possible expropriation: the role of Swiss bank accounts’, Journal of Environmental Economics and Management, 26, 146–62.

278

The international yearbook of environmental and resource economics

Koskela, E. and R. Schöb (1999), ‘Alleviating unemployment: the case for green tax reforms’, European Economic Review, 43, 1723–46. Koskela, E. and R. Schöb (2002a), ‘Optimal factor income taxation in the presence of unemployment’, Journal of Public Economic Theory, 4, 387–404. Koskela, E. and R. Schöb (2002b), ‘Why governments should tax mobile capital in the presence of unemployment’, Contributions in Economic Analysis and Policy, 1, article 1. Koskela, E., R. Schöb and H.-W. Sinn (1998), ‘Pollution, factor taxation and unemployment’, International Tax and Public Finance, 5, 379–96. Layard, R. and S. Nickell (1990), ‘Is unemployment lower if unions bargain over employment’, Quarterly Journal of Economics, 105, 773–87. Layard, R., S. Nickell and R. Jackman (1991), Unemployment. Macroeconomic Performance and the Labour Market, Oxford: Oxford University Press. Lee, D.R. and W.S. Misiolek (1986), ‘Substituting pollution taxation for general taxation: some implications for efficiency in pollution taxation’, Journal of Environmental Economics and Management, 13, 338–47. Lockwood, B. and A. Manning (1993), ‘Wage setting and the tax system – theory and evidence for the United Kingdom’, Journal of Public Economics, 52, 1–29. Long, N.V. (1975), ‘Resource extraction under uncertainty about possible nationalization’, Journal of Economic Theory, 10, 42–53. MacDougall, G.D.A. (1960), ‘The benefits and costs of private investment from abroad: a theoretical approach’, Economic Record, 37, 13–35. Mayeres, I. and S. Proost (2001), ‘Marginal tax reform, externalities and income distribution’, Journal of Public Economics, 79, 343–63. Metcalf, G.E. (2003), ‘Environmental levies and distortionary taxation: Pigou, taxation and pollution’, Journal of Public Economics, 87, 313–22. MISSOC (1998), Soziale Sicherheit in den Mitgliedstaaten der Europäischen Union, Brussels: MISSOC. Newbery, D.M. (1976), ‘A paradox in tax theory: optimal tariffs on exhaustible resources’, SEER technical paper, Stanford University, mimeo. Nichols, A.L. (1984), Targeting Economic Incentives for Environmental Protection, Cambridge, MA: MIT Press. Nickell, S. and M. Andrews (1983), ‘Unions, real wages and employment in Britain 1951–79’, Oxford Economic Papers, 35, 183–206. OECD (1995), The OECD Jobs Study: Taxation, Employment and Unemployment, Paris: OECD. Orosel, G. and R. Schöb (1996), ‘Internalizing externalities in second-best tax systems’, Public Finance/Finances Publiques, 51, 242–57. Parry, I.W.H. (1995), ‘Pollution taxes and revenue recycling’, Journal of Environmental Economics and Management, 29, S64–S77. Parry, I.W.H., R.C. Williams III and L.H. Goulder (1999), ‘When can carbon abatement policies increase welfare? The fundamental role of distorted factor markets’, Journal of Environmental Economics and Management, 37, 52–84. Pigou, A.C. (1920), The Economics of Welfare, London: Macmillan. Pirttilä, J. (1998), Essays on Environmental Taxation in a Second-Best World, University of Helsinki. Pirttilä, J. and R. Schöb (1999), ‘Redistribution and internalization: the manyperson Ramsey tax rule revisited’, Public Finance Review, 27, 541–60. Ploeg, F. van der and C. Withagen (1991), ‘Pollution control and the Ramsey problem’, Environmental and Resource Economics, 1, 215–36.

Double-dividend hypothesis of environmental taxes

279

Razin, A. and E. Sadka (1991), ‘International tax competition and gains from tax harmonization’, Economics Letters, 37, 69–76. Richter, W. and K. Schneider (2001), ‘Taxing mobile capital with labor market imperfections’, International Tax and Public Finance, 8, 245–62. Rose, M. and W. Wiegard (1983), ‘Zur optimalen Struktur öffentlicher Einnahmen unter Effizienz- und Distributionsaspekten’, in D. Pohmer (ed.), Zur optimalen Besteuerung, Berlin: Duncker & Humblot, pp. 9–162. Sandmo, A. (1975), ‘Optimal taxation in the presence of externalities’, Swedish Journal of Economics, 77, 86–98. Sandmo, A. (2000), The Public Economics of the Environment, Oxford: Oxford University Press. Schneider, K. (1997), ‘Involuntary unemployment and environmental policy: the double dividend hypothesis’, Scandinavian Journal of Economics, 99, 45–59. Schöb, R. (1995), Ökologische Steuersysteme. Umweltökonomie und optimale Besteuerung, Frankfurt: Campus. Schöb, R. (1996), ‘Choosing the right instrument: the role of public revenues for environmental policy’, Environmental and Resource Economics, 8, 399–416. Schöb, R. (1997), ‘Environmental taxes and pre-existing distortion: the normalization trap’, International Tax and Public Finance, 4, 167–76. Schöb, R. (1998), ‘Ecological tax reforms and the environment: a note’, Bulletin of Economic Research, 50, 83–9. Scholz, C.M. (1998), ‘Involuntary unemployment and environmental policy: the double dividend hypothesis: a comment’, Scandinavian Journal of Economics, 100, 663–4. Schwartz, J. and R. Repetto (2000), ‘Nonseparable utility and the double dividend debate: reconsidering the tax-interaction effect’, Environmental and Resource Economics, 15, 149–57. Sinn, H.-W. (1982), ‘Taxation, growth, and resource extraction. A general equilibrium approach’, European Economic Review, 19, 357–86. Smith, S. (1992), ‘Taxation and the environment: a survey’, Fiscal Studies, 13, 21–57. Strand, J. (1999), ‘Efficient environmental taxation under worker-firm bargaining’, Environmental and Resource Economics, 13, 125–41. Summers, L., J. Gruber and R. Vergara (1993), ‘Taxation and the structure of labor markets: the case of corporatism’, Quarterly Journal of Economics, 108, 385–411. Terkla, D. (1984), ‘The efficiency value of effluent tax revenues’, Journal of Environmental Economics and Management, 11, 107–23. Tullock, G. (1967), ‘Excess benefit’, Water Resources Research, 3, 643–4. Ulph, A. and D. Ulph (1994), ‘The optimal time path of a carbon tax’, Oxford Economic Papers, 46, 857–68. Williams III, R.C. (2002), ‘Environmental tax interactions when pollution affects health or productivity’, Journal of Environmental Economics and Management, 44, 261–70. Wirl, F. (1994), ‘Pigouvian taxation of energy for flow and stock externalities and strategic, noncompetitive energy pricing’, Journal of Environmental Economics and Management, 26, 1–18.

7. Valuing environmental changes in the presence of risk: an update and discussion of some empirical issues* W. Douglass Shaw, Mary Riddel and Paul M. Jakus 1

INTRODUCTION

In this chapter we examine the current literature on valuing environmental changes that involve risk or uncertainty. We are most interested in how econometric estimation methods can be used in models to obtain ex-ante welfare measures under conditions of risk, but understanding the underlying microeconomic theory is naturally essential to pursue empirical modeling so that is also addressed. Examples below relate to changes in the environment that can affect human and ecological health and well-being.1 Much of the risk oriented economics literature is concerned with financial risk, where future income is uncertain and the focus is on tradeoffs between expected income and risk of investments. ‘Risk’ is typically measured by the variance of an asset portfolio (e.g. Pratt, 1964; Hirschleifer, 1965). While a great debt is owed to the authors of this literature, the work is perhaps only somewhat helpful in understanding how to model responses to environmental changes under risk. Another strand of economics literature focuses on human health and the risk of illness and death, but much of this literature is absent empirical values for risk changes and sometimes lacks a connection to the environment. It is nevertheless important, as it focuses on the theory relating to the value of a statistical life (VSL), and VSLs are regularly used in making environmental policy decisions (e.g. Berger et al., 1987). Still another important strand of risk literature focuses on psychological models and risk perceptions (e.g. Slovic, 1987). Finally, there are several risk oriented papers in the environmental economics and valuation literature, but there is little connection between all of the above mentioned areas of risk research. This chapter is intended to help bridge that gap. Many, if not all, current and serious environmental problems involve the risk of mortality, and to a lesser degree, morbidity. Such problems also 280

Valuing environmental changes in the presence of risk

281

include the risk of impairment of ecological resources from exposure to contaminated air, water, food, or soils. Health and ecological effects from contaminants are rarely certain, though they were often treated that way in the environmental literature 20 years ago or more (e.g. Harrington and Portney’s (1987) seminal work on air pollution and behavior). Even when health effects from exposure might be known, the exposure itself happens with some degree of risk. These risks can be extremely small (Carson and Mitchell, 2003). Good examples of low-probability exposure in the summer of 2002 were exposure to the West Nile virus in various places in the United States, and exposure to foot-and-mouth disease in livestock in the United Kingdom in 1998, and similarly in one cow in the western US at the end of 2003. Arsenic in drinking water is another timely example of the connection between health risks and the environment in the United States, and siting or storage, and transportation of toxic and high-level radioactive nuclear waste is another excellent, current example (Kunreuther et al., 1990). Much has changed regarding how economists think about risks and uncertainty as compared to original thinking on the subject. Older analyses assume that risks faced by an individual were exogenous to them, and easily measurable by experts in assessing risks. Lately however, some believe that nearly all such risks can be controlled by an individual or household through self-protection, making the risks endogenous (Shogren, 1990; Agee and Crocker, 1994). Shogren and Crocker (1999) use the example of species extinction risk, pointing out that habitat is often controlled by individual and group economic decisions; they state their beliefs simply: ‘Risk is endogenous’ (p. 44). As another example, morbidity and mortality risks associated with drinking water laden with arsenic can be reduced through actions the household might take, including drinking bottled water, installing filters that are known to reduce arsenic concentrations to acceptable levels, or more drastically, even moving away from an area with high arsenic concentrations in the drinking water source. Subjective assessments of risk by the public were clearly different than objective or expert assessments in the West Nile and foot-and-mouth cases mentioned above. Many individuals exhibited concern about going outside and being bitten by mosquitoes in the United States, and walking trails in the hills in areas around the United Kingdom were closed in response to foot-and-mouth, likely helping to drive a wedge between the public’s assessment of risks and those made by the experts. This suggests the possibility that behavior should be empirically modeled in a different manner than when we assume that risks are exogenous to the individual, and we address this below. The specific goals of the remainder of this chapter are modest and fairly narrow in scope. We hope to describe current thinking on utility-theoretic

282

The international yearbook of environmental and resource economics

models that incorporate risk, and describe appropriate welfare measures that might actually be estimated using data, all in the context of the environmental arena. First, we present some background on the conventional expected utility model and its application to environmental risk.2 Second, we discuss alternatives to the EU framework, looking at the differences in results when an individual’s subjective assessment of risk is used rather than some exogenous determination of risk. Here we also consider the case of endogenous risk. Next, we consider stated and revealed preference approaches to actually empirically measuring values under risk, and finally, we offer the reader a list of the challenges ahead.

2 2.1

BACKGROUND AND LITERATURE Overview

We assume the reader is familiar with the basics in risk and uncertainty, but at the outset, it is worth remembering that economists can deal best with well-defined risks, expressed as probabilities, rather than with total unknowns (Knight, 1956; LeRoy and Singell, 1987). Often individuals do not have perfect information and while ‘imperfect information’ relates to uncertainty, valuation issues are not equivalent, even though one can derive the value of information (Antonovitz and Roe, 1986). In the world of imperfect information we assume that agents make decisions on the basis of the information they have at the time the decision is made (e.g. Foster and Just, 1989). They may or may not feel uncertain about something when their decision is made, and in fact may be clueless that the information they use to make their decision is incorrect. As Foster and Just (and more recently Leggett (2002)) show, one can model the agents’ actual decisions using the information they actually have, and then construct a welfare measure using both the poor information and the ‘correct’ information that influences their actual well-being. As an example, if people believe their well water meets drinking water standards, they drink it, but suffer the health consequences that depend on the actual well water quality. Their true willingness to pay to reduce the risk of those health consequences incorporates both the incorrect information and the correct information. We believe this consideration of the role of information to be rather different than our focus here, for under uncertainty society may never have ‘perfectly correct’ information that can be used to update welfare measures. Though all uncertainty is complicated in economics, health risks influenced by environmental conditions are something economists can cope

Valuing environmental changes in the presence of risk

283

with. The 1980s produced several EPA funded studies involving environmental risk that used some form of stated preferences to derive welfare measures.3 Physical scientists at EPA were, and still are, interested in risk because of the association between this and environmental problems in the health and ecological risk assessment arena. Valuation of impacts for cost-benefit analysis gave way to the literature on option values, and the option price. In the next section we review the microeconomic foundations for these welfare measures. 2.2

The Expected Utility Model and Option Prices (OP)4

A large body of past literature explores issues connected to the option price (OP) and other ex-ante welfare measures, but the bulk of this literature addresses the microeconomic theory only and does not estimate or even offer a careful discussion of how to actually estimate ex-ante welfare measures using data; the discussion in these papers typically concludes well before one can glean the approach to estimating an equation that sheds light on a policy problem.5 Still, it is impossible to determine if an econometric model is consistent with the theory without carefully examining the theory first. 2.2.1 The expected utility model The derivation of the demand for goods and services in the presence of uncertainty owes virtually everything to the expected utility model (EUM) originally proposed by Bernoulli in 1738 and advanced by von Neumann and Morgenstern in 1947. As an example, suppose there is a relationship between health (H) and environmental quality (q). Assume that the environmental quality level influences a person’s health in a continuous fashion so that H
H(q) and H /q0. Alternatively, if level q0 is lower than level q1, then in a discrete approach an individual who is healthy (H) may turn to one deemed ‘sick’ (S ). This is of course where many certainty studies began. Those interested in uncertainty have also used a household production function for health, with arguments such as the state of pretreatment health in it (e.g. see Besley (1989), who assumes the state can be measured so that it is a bad rather than a good). Let utility be a function of H(q), the consumption of other goods and services (z), and on non-stochastic income (Y). (We could let U depend on q and H separately, and also let H depend on q.6) Note that like most researchers we would probably make some simplifying assumption (e.g. separability) that allows us to ignore z so that data sets on consumption of every good do not have to be made available for empirical analysis. To develop a model with uncertainty, expected utility must depend on some

284

The international yearbook of environmental and resource economics

probability distribution. Assume initially that environmental states, hence health risks, are exogenous: either estimable from some physical laws or at least supportable accurate data. The conventional EUM framework assumes both probabilities and outcomes, or the ‘prospects’, are known and as the above notes, exogenous risks presume that the household cannot influence the level of risk they face. Utility when healthy is UH (Y, H(q)), and when sick, US (Y, H(q)), and the probability of each discrete state is , and (1 ), respectively. Then expected utility is: E[U]
UH (Y, H(q))(1)US (Y, H(q)).

(1)

Assume that the expected utility function meets several conditions, including ordering, continuity, and independence (see Jehle and Reny, 2001). Continuity implies that any lottery can be represented by a certainty equivalent, and independence means that all outcomes are evaluated independently from other chances. One can maximize E[U] subject to the usual budget (Y) constraint, and derive optimality conditions. The resulting model of expected demands forms a basis for several theoretical welfare measures that have something to do with uncertainty. Of those, the most simple is perhaps the expected surplus, ES, which is the probability weighted ex-post consumers’ surpluses. From (1), if one knows that the consumer’s surplus when healthy is CSH, and when sick is CSS, then ES
CSH (1)CSS.

(2)

Graham (1983) shows that if one writes the von Neumann utility function for i
1 to n states, as: n

EU(x,,,a,b)


i [ai u(xi; )  bi ]

(3)

i
1

then in fact, when the consumer is allowed to make choices after the state is revealed, estimation only allows recovery of one parameter () in (3). This revelation of an uncertainty-type of experiment is consistent with recovery of ES, but does not allow recovery of the parameters a or b (see Graham’s 1983 paper for discussion). The implication is that we need to allow choices to be made before the state outcomes are revealed. The ES at first appears somewhat useful, but consider what ES represents when there is a program to improve drinking water quality. If an individual gets sick, which we assume happens with a probability equal to (1)
0.25, and we know that his surplus is $10, then of course (1) CSS
$2.50. If CSH is equal to $15, then the ex-post probability weighted surplus

Valuing environmental changes in the presence of risk

285

is $13.75. When there are more than two outcomes with associated probabilities, we would simply add these in the determination of ES over all outcomes. The ES is thus a useful measure when one can learn the ex-post levels of surplus, but this implies that the dice, as it were, have been thrown, and everyone can see them. This is not too likely in today’s world. Consider two timely examples: the nuclear waste storage dilemma and climate change. Both involve risks well into the future, and probably a lack of knowledge about specific probabilities. Such a world is hardly consistent with resolution to uncertainty, and this is especially true in the nuclear waste case (see Kunreuther et al. (1990), or more recently, Riddel, et al. (2003) or Riddel and Shaw (2003, 2005)), where risks of accidents must be assessed over a 10 000-year period of waste storage.7 A truly ex-ante measure of the welfare benefit of a program to address one of these issues is more critical for policymakers than an ex-post measure. 2.2.2 The option price (OP) The OP is such an ex-ante welfare measure and it can be theoretically generated in the context of a conventional EUM. To review the OP, consider a policy that would increase environmental quality from q0 to q1. The OP is: UH (YOP, H(q1))(1)US (YOP, H(q1))
UH (Y, H(q0))(1)US(Y, H(q0))

(4)

The OP is the payment made in advance of the resolution of uncertain conditions, such that the individual is indifferent between expected utility with better environmental quality less the payment, and expected utility with the original quality, without making the payment. The payment is equal in either state of health. As will be seen below, this definition is made very clear in Graham’s (1981) framework, which assumes that contingent claims are made. There may exist a whole locus of pairs of possible contingent payments, but in equation (4) we assume one contingent payment (equal to OP) in both states described there. More generally, one could assume that H or q is a continuous random variable so that (4) requires integration over the continuous density function. The key to welfare analysis in the EUM is in Graham’s willingness-topay (WTP) locus of ordered pairs because it allows comparison of several concepts, including each conditional surplus measure, the sum of these after weighting by probabilities (the ES), the OP, the certainty point, and fair bet point (the one locus pair with the largest expected value). His graphical exposition (see Figure 7.1) is extremely helpful, and the paper was the first clear presentation of the concepts, seminal in straightening out the issues. Key points about it follow.

286

The international yearbook of environmental and resource economics

Payment if sick (Ps)

S

Fair bet

Odds ratio 45º 0

ES OP

Payment if healthy ( PH)

Note: Downward sloped straight line shows the odds ratio, or slope equal to [Prob/1prob]. Tangency with the WTP locus gives the ‘fair bet’ point, and S is a pair of traditional surplus points. Parallel downward sloping line through S shows ES, at intersection with 45-degree line. The OP is where the payments are equal on the locus, at intersection point with 45-degree line.

Figure 7.1 Graham’s expected utility locus for any pair of payments, PS and PH First, Graham’s WTP locus is by assumption non-linear, reflecting the assumption of financial risk aversion for the individual. Financial risk aversion is met when the second derivative of the utility function with respect to the risky variable, holding other things constant, is negative.8 In Graham’s original exposition, the contingent payments were related to a farmer’s desire to fund construction of a dam given a Bernoulli weather distribution, so that the payments on the axes corresponded to a dry and wet state, but we draw Figure 7.1 for sick (S) and healthy (H) states. The 45-degree line shows all combinations of equal contingent payments, so the intersection of this line and the locus gives the OP. Using the EU framework Graham (1981) defines option value (OV) as equal to OP less expected surplus (ES).9 On the graph in Figure 7.1 as in Graham’s (1981) figure, the OP happens to exceed the ES. The ES is found by drawing a line parallel to the odds ratio through the point S (the

Valuing environmental changes in the presence of risk

287

conventional surplus pair), and identifying the point at which this new line would cross the 45-degree line. While in our case OV is positive, it needn’t be. Though much time was spent on whether OV is positive or negative, Graham (1981) clearly shows that generally one cannot tell. He also shows that Cicchetti and Freeman’s (1971) claim that OP generally exceeds ES is incorrect. In fact, if risk neutrality for Graham’s weather gamble is assumed and there are no income effects, the Graham locus becomes linear, and ES and OP will be the same, unless some other way of relaxing these assumptions about risk references is introduced. We suspect that much of the focus in the 1980s on the OV was simply because many did not know how to ask the correct question to obtain an OP in a survey, or were not sure how to estimate an OP using the correctly obtained data. Solving equation (4) for the OP yields an equation that is empirically tractable given standard, simple deterministic forms for utility functions often used in economics. With a convenient expression available, one can examine how the OP changes with changes in health risk or in environmental quality. For a change in the risk probability one obtains (for example, see Smith, 1992): dOP UH (Y, H(q1))  US (Y, S(q0))
d  UHY  (1  ) USY

(5)

where UHY for example, is the partial derivative of the state dependent healthy utility function with respect to income, and similarly, USY is this derivative when in the sick state. Economists naturally assume that UHY is positive. Of more interest is the sign and magnitude of USY, and at least allowing for the possibility that the marginal utility of a dollar is different than in the healthy state. At the extreme health conditions the two states are typically ‘alive’ and ‘dead’ and utility when dead is generally interpreted in health models as some sort of bequest related utility (Jones-Lee, 1974; Querner, 1994). Intuition suggests that an individual’s value of an additional dollar will be less when sick (or dead!) than when healthy, but the degree of difference is an empirical issue. Consider a very simple utility function, one where utility is linear in Y, and H(q). Then equation (5) boils down to very simple differences in health produced via different levels of q, and respective constant marginal utility of income terms for each state. One could also examine dOP/dq, mapping through the health function as before, considering how changes in q change the health risks, and ultimately the OP. Intuition suggests that with higher health risks, an individual’s ex-ante payment for higher q will increase.

288

The international yearbook of environmental and resource economics

The relationships between various measures are so far clear. If the expected utility function is concave in q, Jensen’s inequality can be applied and it implies that the OP will be greater than the expected surplus, ES. Generally, one can graph the locus of pairs of contingent payments, find the point on the locus where the pair are equal (where PS
PH
OP), compare these to expected surplus if this is of interest, and analyse differences between the concepts as Graham does (again see Figure 7.1). This may be relevant because there may be cases where Graham’s framework suggests that alternatives to the OP are adequate, as in his special case when risks are individual ones.10 2.2.3 Risk Aversion in the EUM Up to this point we have treated the way that risk enters the utility function very simply, akin to financial risk models that simply let investment income be the only risk variable that drives the individual’s behavior under uncertainty. The focus on financial risk may obscure the fact that what we are really interested in is the environmental risk, or typically, the mortality risk associated with an environmental problem. In fact there may be several potential sources of risk: either environmental quality, health status, or both, are random variables, and future income could be risky also. With only one random variable, simple results are forthcoming in the EUM by examining the sign of the second derivative, as stated above. There is obviously a connection between valuation under financial risk and environmental values, as we have indicated above: the shape of the WTP locus determines relationships between welfare measures for environmental changes under uncertainty, and that shape is determined by the implicit assumption of financial risk aversion (Pratt, 1964). It is tempting to assume, as many have, that if individuals are differing in their taste for financial risk, then more risk-averse individuals (averse in the sense of distaste for a mean preserving spread of a lottery) will have a higher value for environmental protection, including the VSL. This may not be true. When more than one risky good is considered, it is no longer possible to look at one partial second derivative and determine overall risk preference. Assumptions about different forms of risk preference, for example financial versus environmental risk, may in fact also result in the equivalence of several welfare measure concepts (Lange, 2001).11 In fact, Eeckhoudt and Hammitt (2004) show that conventional treatment of risk aversion is not enough to determine whether the value of preventing a bad risky outcome is higher for those who are risk-averse than those who are not. In their insightful theoretical paper, Eeckhoudt and Hammitt cite many prior studies, such as Fuchs and Zeckhauser (1987), who state that ‘the more risk averse a person is with respect to wealth, the more he will pay to boost

Valuing environmental changes in the presence of risk

289

his probability of survival’ (Fuchs and Zeckhauser, p. 266). Eeckhoudt and Hammitt define w to be wealth, ua to be utililty when alive and ud to be utility when dead (usually bequest, or set to zero). Marginal utility of wealth in each state is denoted with ua and ud , respectively, and  is mortality risk. With these definitions, in the conventional EUM, the VSL can be shown, using standard assumptions, that survival is preferred to death and the individual is weakly risk-averse to wealth to be: ua(w)  ud (w) VSL
dw EU
0
[1  ]u a (w)  ud (w) d

|

(6)

The VSL is therefore defined as the marginal rate of substitution between wealth and mortality risk, and equation (6) makes it clear that VSL is increasing in wealth, as long as the marginal utility of wealth when surviving is higher than when dead. Because so much environmental economics analysis in the 1980s and 1990s involved analysis of the VSL, we think that environmental economists came to believe that financial riskaversion was all that was important in analysis of impacts under uncertainty. Explicit in equation (6) is that U
U(w, ). Mortality risk is obviously present as measured by , but Eeckhoudt and Hammitt also make assumptions about aversion to risk related to wealth that dictate curvature for the utility function. These assumptions about financial risk are not measured by . Instead, they rely on the conventional Arrow-Pratt definition of risk, as indicated by the presence of a second derivative, with a negative sign. However, using these same definitions, Eeckhoudt and Hammitt readily show that the effect of financial risk-aversion on an individual’s maximum willingness to pay (WTP) to partially reduce mortality risk is ambiguous. Clearly, this comes from the two sources of risk to consider, not one: wealth and the risk related to mortality both affect utility. Eeckhoudt and Hammitt’s framework would perhaps be clearer if there was a discussion regarding the probability distribution defining mortality risk and the separate probability distribution defining financial risk. Otherwise, one might assume that the only risky aspect to wealth pertains to one’s inability to earn income when dead, and here the mortality risk would generate the distribution on wealth in their two states. However, they follow a large literature and are not ‘wrong’ in any sense: as statistics enters into any analysis involving risk, we simply think it more elegant to include a clear discussion of the sources for random variables. When there is more than one source of risk the relationship between risk and WTP for reductions in risk depend on what other elements of the

290

The international yearbook of environmental and resource economics

utility function are held constant while risk-aversion in one dimension is increased. As a case in point, financial and environmental risks are not simply interchangeable: Eeckhoudt and Hammitt conclude that under many assumptions the relationship between financial risk aversion and the VSL that relates to mortality risk is ambiguous. Similarly, it is easy to imagine that an individual may have some tolerance for environmental risks whereas they are strongly financially risk-averse. Thus, a well-specified utility function should likely incorporate both the financial and environmental risk preference. In summary, most would agree that ex-ante welfare measures are the appropriate metric for analysing costs or benefits when utility has a random component. The bulk of the literature does not offer a careful discussion of how to actually estimate ex-ante welfare measures. Deriving appropriate risk measures, and estimating the parameters of the expected utility model are empirical issues. The remainder of this chapter focuses on issues related to estimating the parameters and values, such as the option price, of the expected utility model, and on what happens when risk assessments vary greatly across individuals.12

3

ALTERNATIVE ASSUMPTIONS REGARDING UNCERTAINTY

The properties that EUM gambles are presumed to have are presented above and such a framework is restrictive. In particular, the independence axiom implies a set of linear, parallel and upward sloping indifference curves within the Machina-Marschak triangle. For those unfamiliar with this triangle, one may graph the preferences for gambles in probability space, analogous to graphing a set of indifference curves in twodimensional quantity space. The resulting iso-utility curves are over the relative risks. The triangle is based on the assumption that there are three gambles. Naturally the probability of any third one, say prob2 is recovered from 1prob3 prob1. Conventionally, expected utility is of course linear in probabilities [E[u(X)]/i
u(xi)], which is a constant. The slope of the iso-utility curves is given by dprob3 /dprob1, and so this is a positive constant as well. The individual chooses the gambles taken by picking bundles under uncertainty. Independence requires that for all prospects, s, r, and t: if s is weakly preferred to r then (s, p; t, 1p) is weakly preferred to (r, p; t, 1p) for all p. Querner puts this nicely: if one prize from a lottery is preferred to another prize, two lotteries only differing in these prizes should be ranked the same way as the prizes at hand. Complementary or substitution

Valuing environmental changes in the presence of risk

291

effects between different outcomes are ruled out. The usual predictions of the standard EUM (see Schoemaker for example) include that the: 1. 2. 3.

4.

estimated ex-ante welfare (OP) should be sensitive to the magnitude of the risk change OP should be roughly proportional to the change in risk baseline level of risk should NOT influence OP in evaluating equal changes in risk (i.e. the ex-ante willingness to pay to reduce risk by 1 per cent when there are large risks should not be different than when there are small risks) rate of substitution between risky choice A and another risky choice B remains constant throughout the range of risks that are faced.

These are strong predictions. Those who are critical of EUM claim that the independence axiom is unfounded, pointing to the sea of experimental and other evidence that contradicts the EUM’s predictions. Some suggest that these violations of the independence axiom are particularly relevant to low-probability events that would have devastating consequences if they happened. Mortality risks related to nuclear wastes or earthquakes may fall into this category. Several authors continue to use the EUM coupled with exogenous probabilities (e.g. Cicchetti and Dubin, 1994), but as some authors have said, ‘the major lesson of recent research of individual behavior under risk is that it is not always consistent with the expected utility approach; in short, there is no generic model for evaluating behavior under risk’ (Buschena and Zilberman, 1994). By 1953, Allais had already demonstrated several contradictions flowing from the EUM’s predictions for choices; others have recently categorized failures similar to the ones he outlines as common consequence or ratio effects (see Allias, 1953, 1979; Starmer, 2000). Early simple experiments demonstrated that when choosing gambles, individuals often exhibited behavior inconsistent with an ability to rank gambles. Let an individual have to choose between a1 or a2, and between a3 and a4, where:13 a1: $1 million with probability
1.0 a2: $5 million with prob.
0.1, $1 million with prob.
0.89, $0 with probability
0.01 a3: $5 million with prob.
0.10, $0 with probability
0.90 a4: $1 million with prob.
0.11, and $0 with prob.
0.89. The lines connecting a1 to a2, and a3 to a4 are parallel, so if the individual’s indifference curves are straight lines, as the EUM suggests, then he prefers a1 to a2 and a3 to a4, or a2 to a1 and a4 to a3. However, laboratory

292

The international yearbook of environmental and resource economics

experiments have shown subjects choose a1 and a3, and thus, the indifference curves cannot be straight lines. Observed choices led researchers to the conclusion that preferences were often reversed, that unequal gambles sometimes were treated as equal ones, or that individuals did not behave rationally. Ex-ante efficient choice may fail for a variety of reasons that are mostly in the realm of discrepancies between agents’ assessments of the probabilities that describe the risks (Besley, 1989) and assessments by experts. The experimental economics literature, very much in vogue today (see Mason et al. (2000) for an example), has carefully examined issues of demand under uncertainty and the evidence generally supports the notion that people tend to overvalue reducing the probability of low-risk events and undervalue risk reduction of high-risk events relative to the predictions of the EUM. Mason et al. find that indifference curves over lotteries are often concave rather than linear in probabilities. Despite evidence like this, traditional EUM supporters are hesitant to support alternative theories that could potentially lead to non-unique choice sets. In response to these critics, many of the new theories of demand under uncertainty examine how non-linear indifference curves over risky bundles or prospects might be incorporated while still ensuring unique demand relationships (Loomes and Sugden, 1987; Besley, 1989). In his survey article Starmer (2000) lumps many alternatives to the EUM together because they seek relaxation of this independence axiom. As an example, Machina (1982) introduced a generalized expected utility model (GEUM) to allow the expected utility indifference curves to fan out, recognizing that concavity of the probability derivatives is allowable even with non-expected utility preference functions. This generalized model relaxes the constant slope (i.e. the rate at which the prospects are traded for one another) as one moves from very low to very high risks, and this is fairly appealing. Machina’s GEUM hypothesizes monotonicity (if one prospect first-order stochastically dominates another, the dominating prospect is preferred), and fanning-out, and some do not wish to see a requirement for fanning-out. Consider two pairs of prospects (R ,S ) and (R,S). If preferences for the two pairs violate the EUM, then fanning-out implies that the violation is of the form such that the person is more risk-averse when evaluating the stochastically dominating prospects than when evaluating the dominated ones (Starmer and Sugden, 1989). Yaari and several others have weighed in on the failures of the EUM (see Starmer, 2000), proposing variations on the standard EUM that may help in resolving paradoxes that arise in the EUM context.14 Eeckhoudt and Hammitt (2004) state that Yaari’s (1987) dual non-expected utility theory is a special case of rank-dependent EU theory and cumulative prospect

Valuing environmental changes in the presence of risk

293

theory. In Yaari’s model one simply uses linear utility functions in wealth (w) and a probability weighting function h(), so that U
[1h()]w  h()âw. Here the first term is the utility of wealth while alive and the second term the utility of wealth when dead. The weighting function has the properties that h(0)
0, h(1)
1, and dh/d 0. Several other alternatives (again see Starmer’s excellent discussion) also relax the independence axiom, allowing the model to depend on the description of the probability distributions for the consequences. Perhaps the most well-known alternatives today are prospect theory (Kahneman and Tversky, 1979 or Tversky and Kahneman, 1981) and the prospective reference theory (PRT) of Viscusi (1989, 1990). Kahneman and Tversky’s prospect theory defines consequences as gains or losses relative to a reference point, which in their models relates to a level of wealth. In this sense a gamble can be treated quite differently if it involves a loss, or negative consequence, as opposed to a gain, or positive consequence. Strictly positive or negative consequences involve quite a different formulation of the utility maximization problem than conventional EUM: expected value is decomposed into a certain gain and a probabilistic one, or in the case of a negative consequence, a certain loss and a probabilistic one (see Starmer and Sugden, 1989). Viscusi’s PRT is similar to prospect theory. In it, Viscusi replaces  in each of the above expressions with a function of , say (Prob). Prospect theory (Kahneman and Tversky, 1979) essentially would essentially do this also. For example, let the function be linear in the reference risk levels: (Prob)


(Prob  P0) . (  )

(10)

Here the P0 is the probability of the expert’s assessment of the original outcome and Prob is the individual’s prior assessment of risk. The  and  are the weights assigned to the informational content in each (the expert’s and the person’s own prior assessment of the risks). If updating reduces the weight given to the prior, then the probability approximates the expert assessment. Starmer categorizes PRT as being a theory which helps explain ‘event splitting’ effects. Here, a risky exposure might be described as producing several subrisks, each with a smaller probability perhaps than one larger probability risk. When defined with the many subrisks, the consequence then gets more weight than otherwise. Starmer deems some alternatives to the EUM to be unconventional in that they focus on how and why individuals make decisions; these include the view that the agent is boundedly rational. In one of the more interesting experiments we found, Starmer and Sugden (1989) tested various

294

The international yearbook of environmental and resource economics

alternative theories, including the GEUM, Prospect Theory, and Quiggin’s rank-dependent theory. The authors describe their own work as a pilot study, but they do not find any particular pattern in violations of the EUM, though most of their results indeed do suggest violations of the conventional theory. They were looking to see whether the statistics supported consistent fanning-out of indifference curves over gambles, or consistent fanning-in, but they find support for either one being true in their data. Still, their pilot study may lend credence to incorporation of subjective risk assessments, discussed next. 3.1

Subjective Risk Assessment and Perceptions

Both prospect and rank-dependent theory belong to the set of subjective expected utility theories (Starmer and Sugden, 1989). Economists may certainly argue about how to use subjective risk assessments, or what to do about these if they greatly differ from objective assessments, but perhaps irrefutable is that laypersons just do not always evaluate risks in the same way that experts do. Besley’s (1989) example is the insurance agency’s expert and the insured party. If people believe something different than what accumulated statistics or laws of science suggest about risk, then they will likely behave in ways based on their beliefs, not the statistics. Further, in some cases, they can make ‘bad’ choices based on their beliefs. David Ropeik, director of risk communication at Harvard’s Center for Risk Analysis, cites the example of a mother of a United Airlines attendant who drove on a trip rather than flew shortly after 9/11 and was killed in a car crash. He says she was ‘too afraid of low risk [flying] and . . . her risk perception led her to a choice that was dangerous’. (Dreifus, 2002). An individual’s risk perceptions have been empirically shown in many cases to be quite distinct from other agents’ technical assessment of risk (see Schoemaker, 1982 for a summary). Let the individual’s risk perceptions be subjective probabilities, deemed prob. These might be used in lieu of, or at least to complement expertly assessed risks.15 It may be that the probability of an outcome (z) based on expert judgement [(z)] and the perceived probability of an outcome for the individual, prob(z), are roughly equal, and one might initially assume that the individual’s assessment of risk can be treated as an exogenous and objectively determined variable. We could then simply modify the EUM by substituting these perceptions for actual risks into the usual equation for the EUM, as was suggested by several psychologists in the 1950s. Very large divergences between expert and individual probabilities can be troubling. They may arise for several reasons. For example, Riddel et al. (2003) and Riddel and Shaw (2005) consider the Department of

Valuing environmental changes in the presence of risk

295

Energy’s (DOE) assessment of the risk of mortality from an accident during planned shipping of nuclear wastes, and assessments by Nevada residents who live near the proposed shipping routes and the final storage site of Yucca Mountain, Nevada. DOE’s assessment of mortality risk from a transportation accident is about 2 in 10 million, about the same as dying from ingesting weed killer.16 On average, individuals in the estimating sample greatly overestimate this tiny risk of DOE’s: prob(z) is thousands of times higher than DOE’s estimate. Respondents do this even when presented with an information brochure depicting a risk ladder with DOE’s estimates, because of a lack of understanding of the problem, failure to comprehend these very small probabilities, a lack of knowledge, or of information. If public policy requires action affecting individuals who will bear the risk, such as the nuclear waste transportation scenario, then a massive policy failure will occur unless the average household’s perceptions are moved much closer to the agency assessment. As an aside, failures in the risk communication process (Smith and Desvousges, 1987; Riddel and Shaw, 2005) leave a program in dire straights with respect to public support, as may be true for the DOE nuclear waste program in the future (Flynn and Slovic, 1995). Of course there may also be failures in the experts’ original assessment, mitigating factors, strong preferences, and special characteristics of the risk bearing households that complicate the risk communication process. Education or learning has the potential to move the individual’s probabilities closer to (z), but this shift cannot be guaranteed. In fact, the experiments in risk communication in the 1980s suggest that educating the public about risk is quite hard to do. A timely example is the recent example of the Centers for Disease Control (CDC) trying to educate the public regarding the real probability of dying from exposure to anthrax powder in an envelope. The CDC themselves seemed unsure of the probability of morbidity or death with the type of anthrax being distributed in the mail. So, if individual subjective probabilities remain important, a key issue concerns the manner in which these are introduced in any model. We consider three approaches here, using stated probabilities to replace the objective ones, using estimated probabilities, and using decision weights. 3.2

Using Stated Probabilities in a Risk Model

All of the above suggests to some scholars that perhaps we should simply ask individuals to state what they think the probability of an outcome is. This approach was used by Crocker and Shogren (1991), and in a more advanced fashion, by Cameron (2003b). Here the modeler assumes that the stated probability is fixed, as a household’s reported income is, and

296

The international yearbook of environmental and resource economics

uses the variable accordingly. The attractiveness of this approach is that individuals can reveal heterogeneity in their expression of understanding and feeling about the mortality or morbidity risk related to an environmental change. While appealing in its simplicity, this approach could pose problems. First, individuals may be poorly informed and change their beliefs substantially when new information is conveyed to them, strongly suggestive of treatment of the stated probability as a random variable, or the need for Bayesian updating. Second, one worries that if surveys are used to recover ex-ante welfare measures, then each individual has a different understanding of the ‘good’ being valued, because each assessment of risk accompanying the environmental change may vary. To offset some of the concerns, ‘alternative’ theorists have proposed the use of decision weights. Decision weights Decision weights can be consistent with subjective expected utility theory. Using them is a way to reconcile the difference between subjective and objective risk assessments, by introducing probability weighting functions: V(s)
f(wi)( probi ), where again, s is a prospect, and here f(wi) is the weighting function. A weighting function can help explain observed behavior that contradicts what would be predicted under the EUM. The exact form that the weighting function takes is critical to our arguments below. Quiggin (1991) summarizes the thinking on weighting functions. If the weighting function is linear and defined to map in the unit interval, he suggests that a correct weighting function simply forces the alternative model to become equivalent to the EUM. It is the non-linear weighting functions that are more appealing on one hand, such as the s-shaped curves underlying much of the work to address discrepancies in losses and gains, but these non-linear functions also create serious problems for the person who might be interested in welfare measures. Fishburn (1978) noted that in many simple forms this type of weighting scheme leads to the expected utility function V(s) having the property of violating monotonicity of preferences. Such a violation would clearly spell trouble for welfare estimation, as the meaning of the welfare measure would be so different from any conventional notion. Subsequent schemes such as rank-dependent expected utility theory (RDEU – Quiggin, 1982) can ensure that V(s) is monotonic. Quiggin (1991) appears to view RDEU essentially as expected utility, but for a transformed probability distribution. The key difference between his notion of the RDEU and other weighting schemes is that the weights take into account the entire probability distribution rather than only one individual probability (pi). Formally, the

Valuing environmental changes in the presence of risk

297

RDEU defines a weight as hi(PROB), where PROB represents a vector of probabilities, while typical weighting schemes define a weight as wi(pi). Finally, Quiggin and Chambers (1998) derive properties of generalized expected utility models concerning risk-aversion. They demonstrate that while a preference function in the context of the EUM can display constant absolute risk-aversion, or constant relative risk-aversion (CARA or CRRA, respectively), it cannot display both, except in the trivial case of riskneutrality. Some of the generalized models however, can in fact display both properties. This is important in valuing environmental changes because individuals have choices to make regarding the functional form and the role that income or another key variable plays in introducing various preferences for risk. We return to the empirical issues when using stated probabilities in Section 5, below.

4 4.1

EMPIRICAL APPROACHES: STATED AND REVEALED PREFERENCE Stated Preference Models

Most studies that actually find empirical values for environmental changes under risk use at least some stated preference (SP) data: here we basically mean the modelers asked valuation questions directly, using some sort of survey format, and typically applied the well-known contingent valuation non-market valuation (CVM) approach. To get risk related welfare measures, the survey questionnaire has to describe uncertainty carefully, so that the respondent understands that environmental changes occur with a meaningful probability: one greater than or equal to zero and less than or equal to one. If the survey question is posed about the future, but with no degree of uncertainty introduced, then the resulting welfare measure may have a connection to the future, but might simply be an existence, bequest, or preservation value, none of which necessarily involve risk or uncertainty. The manner in which the all-important probabilities were communicated to respondents in these surveys became a critical issue in the 1980s work (e.g., see Smith and Desvouges, 1988), and some researchers are still quite pessimistic about the way individuals can incorporate probabilities into their understanding of risks, especially when valuation is the goal (see Hammit and Graham, 1999). However, many environmental economists have come to favor the risk ladder approach to communication (see Loomis and Duvair, 1993; Carson and Mitchell, 2003; Riddel, et al., 2003 for an application). Good risk communication is important in any empirical SP model. Upon close inspection of many existing empirical SP studies it is difficult

298

The international yearbook of environmental and resource economics

to know if the claims of estimating a particular ex-ante welfare measure are valid: it is hard to tell whether what was estimated was really an OP (e.g. Brookshire et al., 1983), or something else. This is partly because published papers do not often include the text of the survey questionnaire, but also because the source of the randomness and exactly how it is connected to risk or uncertainty, how it relates to underlying preferences, and the actual empirical model estimated, are not completely clear to the reader.17 4.2

Revealed Preference

Though stated preference models have been widely used to develop ex-ante welfare measures, we are aware of very few environmental economics studies that have used revealed preference (RP) data to obtain such measures for changes in non-market goods.18 Thinking about the issues involved in using RP data carefully, one can see why this is true: behaviors must have occurred that can shed light on risk taking, and these behaviors and the associated risk levels (the probabilities) must somehow be uncovered for modeling purposes. To obtain RP data and derive welfare measures, rather than ask directly for stated values that translate to an OP or some ex-ante welfare measure, one must obtain actual behavior with an implied ex-ante payment in two or more uncertain states. The trick is in establishing clear, expressible utility differences so that an ex-ante welfare measure may be derived. Åkerman et al. (1991) make what they claim is the very first attempt to use RP data to recover environmentally related values under uncertainty, using mitigation costs to reduce the risk of getting cancer from radon exposure in one’s home. Their experiment is ideal because they can observe behavior in response to risk. Unfortunately, they admit that their cost data are not quite up to standard, and a lack of detail in the published paper makes it hard to discern just how the risks were fully incorporated: it may just not be clearly explained, but the probabilities underlying getting cancer are absent in the econometric modeling described in their paper. Some other environmentally related studies that examine the role of imperfect information do use RP data. In Agee and Crocker’s (1994) study an observed behavior is whether the child was chelated to reduce the risks from exposure to lead. Dickie and Gerking (1997) examine behaviors in response to skin cancer risk. These observed behaviors play the same role as the observed keeping of potentially contaminated fish by anglers in Jakus and Shaw’s (2003) study. Still, links between actual observed behaviors and perceived risks, as well as to objective risk measures are difficult to uncover. Incorporating the relevant probabilities into a pure RP model will indeed continue to be a challenge, especially if one wishes to incorporate

Valuing environmental changes in the presence of risk

299

ambiguity effects. In other arenas, economists have used market data to examine ambiguity effects (e.g. Ho et al., 2002), but this has not been done to our knowledge in environmental analysis. Jakus and Shaw’s (2003) recent work scratches the surface in that fishing behavior and estimated probabilities based on observed behavior are used to recover a risk related welfare measure, but their survey was not specifically designed to reveal an ex-ante measure of welfare: the authors make no claim that they have estimated Graham’s option price. Other RP modelers are a bit vague about what the welfare measure actually is (an OP, OV, or an ES),19 so there may be interesting work ahead on the nature of the welfare measure in these models. 4.3

Empirical Models and Sources of Randomness

Whether SP or RP, empirical models typically add an error () to deterministic terms to make them random variables. This error is assumed associated with the investigator’s failure to recognize all variables that may influence an individual’s utility. It does not originate in the individual’s uncertainty about some event. In contrast, much of the empirical literature that deals with the individual’s uncertainty does not consider measurement error, or incorporating the usual investigator’s errors. As a preview to what follows below, we might need to carefully consider the derivation of the ex-ante welfare measure when measurement error is also present; a rich model might allow two random variables to enter into the derivation, one to represent measurement error and the other to incorporate the uncertainty. Estimation is typically accomplished by assuming mean zero errors, and the independent variables in the model are assumed non-stochastic whereas the dependent variable is stochastic. More formally, the usual sources of error are (i) omitted variables, (ii) error in measuring the dependent variable, and (iii) measurement error in the independent variables. All but the last are consistent with the Gauss-Markov assumptions of classical modeling, assuming that the omitted variables or measurement error are not correlated with the independent variables. The last source of error violates the classical assumptions. Using simple linear and semi-log demand functions and assuming no uncertainty on the part of the individual, Bockstael and Strand (1987) nicely illustrate that assumptions (i) to (iii) will yield different computed welfare measures. They illustrate that the assumed source of randomness makes a difference when computing certain welfare measures. Their argument is basically as follows. Suppose CS is generally denoted as f(x), and E[CS]
E[f(x)] denotes the expected value of CS when there are random variables. The potential

300

The international yearbook of environmental and resource economics

problem is seen by again applying Jensen’s inequality, when f(x) is a convex function: f[E(x)]E[f(x)].

(11)

The inequality sign in (11) is reversed when f(x) is a concave function. The above implies here that a more general expected consumer’s surplus will be equal to, or greater than, the surplus generated from expected demands. Bockstael and Strand conclude that for any consumer’s surplus function that is convex in x (the demands), there will be a difference in consumer’s surplus depending on the implicit assumption about the source of error. We extend this discussion to allow for risk and uncertainty. Intuitively, if Bockstael and Strand obtain the result in models with no uncertainty, we might well expect that different assumptions about random variables in models with uncertainty also lead to different welfare measures. Using some of Larson and Flacco’s (1992) suggestions, Kling (1993) demonstrates this using some simulations and literature estimates of option value. As a specific example of what might be involved let q be the arsenic level in water drunk at the tap, a function of a naturally generated concentration of arsenic found in bedrock or soils  but randomly distributed because of weather and precipitation patterns that affect the flow in ground or surface water. It may also be a random variable because of random patterns of exposure from drinking from a particular source, but the basic idea is that q
  . The suggestion here is that a systemic portion of q helps explain the variation in arsenic concentrations in drinking water across different geological regions, but so does a random component, . In this case a simple conditional indirect utility function V(y, p, q) is (a choice implies a non-zero price, p so that this p is subtracted from income, y) rewritten: V
( y  p)(  )

(12)

Given this formulation, we perhaps ought to consider potential correlation between the random variables, say between the two sources of randomness, and , when developing a model that incorporates risk and measurement error. In the arsenic example we give above, and  are somewhat likely to be correlated, depending on whether the variations in the weather lead to both shocks in the measured arsenic level, and also the demand for water at the tap. Perhaps extreme temperatures would affect both. Again, the assumption about the source of  will be important, as will the nature of the randomness underlying . We know of no others that

Valuing environmental changes in the presence of risk

301

have considered the possibility of this process underlying q or q, so there is no discussion in the literature of potential non-zero covariance terms that might arise. Another approach one could take is to let estimates of health risks themselves be measured with error, and deem that error ‘uncertainty’.20 Let health (H) be a random variable, so that: H
H*

(13)

where H* might in turn be a deterministic function, for example, H*
a qbD, where q is again environmental quality but here it is not playing the direct role of a random variable, and D represents health related expenditures. As in the first part of this chapter, a standard assumption is that H D  0, i.e. that this type of spending improves health, perhaps through preventative medicines and procedures. To our knowledge no one has implemented the ideas above, but we suggest that they can be. Below we give some cases involving uncertainty and offer examples of how to implement some of the more novel ideas empirically, largely ignoring the case where probabilities are exogenous and simply determined.

5

THE CHALLENGES AHEAD: CONSIDERING SPECIFIC CASES

Here we consider specific cases involving the data, situation, and modeling, and lay out the challenges that lie ahead for empirical modelers. Case A Suppose that q is random and discretely distributed. Further, suppose that q is measured with an indicator variable, with q
1 a ‘poor quality’ situation, and q
0 the status quo, with good quality. Examples are when a leak from a gasoline or oil storage tank occurs or does not, or when nuclear waste is transported or not, meaning that accidents in transportation happen, or do not. Both events involve probabilities, and the next step is to determine what these are. In a very simple model, we might assume that q follows a Bernoulli process. For example, when q is a nuclear waste accident, if it is not transported, then the probability of q
0 is zero. However, with transport, a non-zero  must be determined. In the expectation over q for a simple indirect utility function we have a simple model where Eq[q1 q0]
E[q1]
.

302

The international yearbook of environmental and resource economics

If this probability is exogenously given it may be the same for all individuals. We expect that this risk level, if it had any significant effect at all, would negatively influence a decision involving these risks. To empirically implement this model, one likely needs a discrete choice behavioral model because q implies this. One could, for example, write the conditional indirect utility functions, take expectations over q, resulting in forms that can be estimated. Modifying equation (11) after taking expectations, one hopes to see an indirect utility function of the form V(p,Y,q,) so that the change in utility with respect to the risk can be evaluated. If estimated using the logit or probit discrete choice models, then the estimation also involves the usual investigator errors. If we see a conventional logit or probit model with no risk measure in the utility function, we do not believe that the model can be correctly incorporating uncertainty. If expectations of utility with respect to the random variable are properly taken, the OP ex-ante measure can be found by examination of the expected utility differences (see Cameron, 2003a). This modeling quickly becomes much more complicated if we assume that risk is endogenous, and we think that while challenging, this is more interesting. For example, strictly speaking we might say that in the long run the household can move further away from the hazardous waste, reducing the risks from exposure. Or, in the case of nuclear waste shipping, households might move a safe distance away from the proposed routes. In that case, risk is not a simple exogenously determined probability because there are two options for estimation. One ‘option’ for dealing with risk endogeneity is to make a simplifying assumption, as suggested by Graham (1981) in his special case for individual-specific risk. There Graham suggests that we simply redefine the states so that they are similar for all individuals. An example would be so that ‘state 1’ is exposure and ‘state 2’ is no exposure. We don’t find this to be particularly appealing for empirical work, but postpone discussion of option 2 until Case C is considered below. Case B Another possibility one might face is that the density function used to implement uncertainty is again known and exogenous, but more complex than the above simple Bernoulli example, perhaps requiring evaluation of an integral (e.g. in the absence of a closed form). An assumption maintained is that the random process relating to Eq is independent of that process underlying the usual error term. In this case there is an additional level of complexity making the usual recovery of welfare terms more difficult.

Valuing environmental changes in the presence of risk

303

Simulation methods are required in this case to estimate the ex-ante welfare measure (e.g. Cameron, 2003a), but simulation methods appear to be quite popular at this point (e.g. they are used in the mixed, random parameters, or heterogeneous logit models), so this alone is not likely to thwart efforts to recover welfare measures under uncertainty. Perhaps more interesting to the theorist would be careful consideration of how complex density functions relate to the Graham framework, where the contingent payments are simple functions of the probabilities. This is a key area where more work needs to be done. Case C Here we consider the case where the probability density function for q is clearly individual-specific (there is a subjective risk assessment), or perhaps endogenous to the individual. We believe that subjective risks and endogenous risks are slightly different, but these concepts relate to alternatives to the EUM. One could have a subjective assessment of risk that differs substantially from one made by an ‘expert’, but not be able to control that risk level. Alternatively, through self-protection (mitigation) one might control exposure and hence, health risks, making them endogenous to the household. Suppose each individual has a process of forming perceptions of risk, so that  varies for each individual, and likely depends on a host of factors. One might simply use a person’s stated probability in their EUM. This simple approach assumes that the stated probability is not a random variable, explained by factors in the household. Alternatively, a stated perceived risk can be assumed to be a random variable, estimated as a function of some explanatory variables, and used later, in choice modeling or valuation (see Agee and Crocker, 1994). In some recent efforts the modelers literally estimate each individual’s perceived risk. In turn, this estimate feeds back into a model that determines choice. Viscusi and others have also gone further than simply using the stated probabilities in such a simple manner (Viscusi, 1989, 1990). Prospect Theory research (Tversky and Kahneman, 1981) indicates that gains and losses are valued differently under uncertainty, even if the monetary equivalent is the same in each case. A Bayesian spirit is associated with the alternative approaches suggested by these authors because individuals are thought to process information, update their priors, and perhaps reveal subjective but posterior probabilities associated with risk. It has been suggested that with enough information and updating, the posterior should come quite close to the true objective or expert risk assessment or probability. This is simple to see in Viscusi’s (1989) prospective reference theory

304

The international yearbook of environmental and resource economics

(PRT), described above, which generalizes some earlier proposals by psychologist Smith (1992) shows that the expression for the option price in the PRT context can be viewed as the weighted average of two expected utility functions. One substitutes the (Prob) expression into the basic EUM equation (1). Typically regrouping terms leads to formulas that allow key features to be examined. Lattimore et al. (1992) generate similar functions, leading to a risk transformation function. Viscusi et al. (1999) demonstrate that much of the above can be accomplished by generating reduced form equations for the risk perceptions, and in fact, much of Viscusi’s empirical work does just this (e.g. Evans and Viscusi, 1991). However, we find no exact description of the ex-ante welfare measure is in those studies, and often none is even estimated. Case D Consider endogenous risks. Suppose that the household can affect the level of risk it faces through certain actions and that we can collect data on individual-specific probabilities, possibly through stated probabilities, or possibly through a function that determines a household’s probability. At the least, it would certainly seem appropriate to test for endogeneity, as self-protection might involve fairly low levels of effort and cost. For example, households with children facing mortality risk from secondhand smoke might control that risk by having the parents smoke outdoors, or in a room away from the children, or by adopting ventilation systems that reduce the smoke. Any situation where the parents of the household can mitigate against health risks for themselves or their children might fall into this category. How they control the risk of course depends on the information they have regarding these risks (see Agee and Crocker, 1994). Here again it would seem appropriate to estimate the risk function first, before using any probability level in an expected utility model. As in the subjective risk case, many factors might predict the risk level a household actually faces, and these factors can change, thereby changing the probability level itself. If there are no concerns about econometric efficiency, then the risk functions can be estimated in a first stage, then fed back into a decision model. In principle, we see no difference between handling endogenous risk and subjective risks as random variables. Agee and Crocker (1994) implement this approach using data on child lead exposure, viewing the risk function as a posterior estimate, calculated as a weighted average of the average exogenous risk, and the mean of the parents’ prior beliefs. Their perceived risk is: R
R(Ac,L;,i )

(14)

Valuing environmental changes in the presence of risk

305

where Ac are child care activities, L is chelation therapy, and the next two variables are family characteristics, and exogenous information about risks of lead exposure given to the family. This is not at all unlike the PRT model of Viscusi’s in theory, but Agee and Crocker do not actually estimate an equation such as (14) above. Rather, they simply estimate the demand for chelation therapy for the child, being only able to observe whether the child was chelated or not. They interpret their probit estimates as being the probability of obtaining chelation therapy, which they say indirectly reflects the parents’ assessment of the risks to their child. Case E The last case we consider is a subset of Case C and involves individuals who are uncertain about the uncertainty, or who are ‘ambiguous’. An ambiguity effect, (Amb), can be interpreted as the variance on the amount that the person is uncertain about the probability itself. If the individual states the probability of this with some ambiguity (see Ellsberg, 1961; Einhorn and Hogarth, 1997; Viscusi and Chesson, 1999), then the randomness owed to the ambiguity might indeed relate to the randomness underlying the behaviors. The more uncertain the respondent is about the risks of the accident, the less utility she receives, which is consistent with Ellsberg’s ambiguity-aversion. This additional influence of ambiguity is novel in valuation models. It can be introduced by decomposing the event risk variable into two distinct random variables, one pertaining to the event and another to the ambiguity (see Riddel and Shaw, 2005). Cameron (2003b) also allows for something like ambiguity in a more simple manner, and allowing for this effect appears to be important, so much more work lies ahead.

6

CONCLUSION

Environmental economists would understand much less than they do now if Graham had not laid out the relationship between possible welfare measures under uncertainty. He set our profession straight on the possible relationships between the ES, OV, and OP, explaining why the OP would be preferred in most situations. It is a bit difficult, but quite possible today for applied economists to collect data and estimate actual option price measures of welfare: Cameron’s and Riddel and her coauthors’ recent papers are good examples. Studies that incorporate uncertainty or risk and obtain values for environmental changes do so primarily using SP data. We are not disparaging of stated preference models, but many other economists are, so development of RP data models likely needs more attention. Environmental

306

The international yearbook of environmental and resource economics

economists need to catch up with economists who use RP data to value health risks related to automobile safety, risks to residents in houses in certain locations, etc. These studies incorporate behaviors that are costly to the individual, easily observed, and which provide a clue to preferences for risk by observing the tradeoff between money and risk reductions. On the theoretical side, Smith (1992) hints we cannot view ex-ante welfare measures from alternative models to the EUM in exactly the same manner as we would Graham’s OP, but this issue has not yet been fully examined. Clearly, if one keeps the locus of payments for uncertain outcomes in mind, as in Figure 7.1, then introducing utility functions that themselves depend on non-linear functions of the probabilities, or that in some way make the sign of the second derivative of the WTP locus ambiguous, may create a different ex-ante welfare concept than what Graham depicts using the WTP locus. We do not believe that this means that alternative models lack usefulness in terms of welfare measurement, but it may be the case, in some situations, that resulting welfare measures will require careful scrutiny.

NOTES *

1.

2.

3. 4. 5.

E-mail addresses are [email protected] (contact author), [email protected], and [email protected]. We thank Chris Starmer for encouragement of this work, Trudy Cameron for sharing her working papers on modeling responses to environmental changes under uncertainty, and Kerry Smith for his thoughtful comments on one of our related papers. Comments from seminar participants when related papers were given are acknowledged, including those at the Universities of Chicago, California State-Chico, Central Florida, Colorado, Denver, Wyoming, Colorado State University, and the US Air Force Academy. We also thank the editors and an anonymous reviewer for comments that have improved this chapter. This research was partially supported by the Utah Agricultural Experiment Station, and for Riddel, from a UNLV faculty grant. This is naturally a highly selective review of the recent literature and we apologize to all those who have contributed important work in the area of modeling risk taking behavior, especially as it relates to environmental changes. Our focus here is on papers that relate to empirical issues, but the reader interested in theory can see Schoemaker or Starmer’s survey papers. The reader should note that imperfect information issues are sometimes thought to be synonymous with uncertainty, but in our view they are not, and this also is explained immediately below. Environmental concerns relate much more to uncertainty than poor or imperfect information in market or non-market settings. A few examples are Fisher et al. (1986) and Smith and Desvouges (1987). Those unfamiliar with the environmental valuation literature should be warned that though related, option prices here are not the same as ‘pricing’ stock options. Innes and Cory (2001) provide a good example of a theoretical investigation, Larson and Flacco (1992) go further than this in suggesting general forms for expected demands, and Graham, in a lesser-known paper in 1983 also offers a few important suggestions for empirics.

Valuing environmental changes in the presence of risk 6. 7. 8. 9.

10. 11. 12. 13. 14. 15. 16. 17.

18.

19. 20.

307

Adding a separate q term allows environmental quality to matter for reasons other than those that are health related. High-level radioactive wastes need 10 000 years to decay before mortality and morbidity risks from exposure fall to acceptable levels. As some still state that this means risk-averse individuals decline all fair bets, it should be noted that Graham (1981) debunks this as a meaningful definition because even risk-averse individuals would not necessarily decline purchase of fair insurance. Option value was a concept initiated by Weisbrod in 1964. All of these welfare measures are discussed briefly below, but more extensive discussion of the debate about the ‘sign and size’ of option value (e.g. Cicchetti and Freeman, 1971; Graham, 1981) can be found in Bishop (1982) or Smith (1992). In his 1981 paper, Graham uses the example of various individuals who have desires to visit an outdoor park, but one may not desire it in one state, while another does, in exactly the same state. This seems to have been first noted by Pratt in 1964. Again note that when one omits any other source of randomness to characterize the risk other than income, then a linear in income utility function imposes risk-neutrality. A reviewer notes that this might be a case for allowance of individual heterogeneity. We take this representation of Allais’s paradox from Mason et al.’s (2000) paper. Yaari’s ‘dual theory’ of choice under risk explains some of the Allais paradoxes. The work of psychologists such as Paul Slovic (1987) suggests that in many instances ‘actual’ risks are simply the assessment of risks according to another group of people, often deemed experts. DOE obtained these estimates using statistics on accidents, and engineering tests, such as ramming aircraft engines into storage and shipping casks to see if they will break and leak. Examples are Brookshire et al. (1983), Edwards (1988), and Loomis and DuVair (1993). We note that Loomis and DuVair do not seem to let their empirical model depend on the probabilities relating to the risky outcomes, but they are careful to mention that their empirical model is ad hoc, and that their approach assumes everyone is risk neutral. Jakus et al. (2003) discuss some of the difficulties in using RP data in a risk context. Åkerman et al. (1991) use data on remedies to household radon problems and is the first RP risk study we know of, Agee and Crocker (1994) use RP data to explore the effect of information on perceived risk, and Jakus and Shaw (2003) is another recent exception. In their theoretical paper Larson and Flacco (1992) suggested that revealed preference (RP) models can be used to estimate ex-ante welfare measures under risk, but their discussion does not focus on issues in actual empirical implementation. For example, Agee and Crocker (1994) simply deem their welfare measure a marginal willingness to pay. Lichtenberg et al. (1989) say that a health risk means the probability that an individual selected randomly from the population contracts an adverse health effect. Health risk estimates are subject to error because the relationship between that risk and the variables that generate it cannot not be known with certainty. The term ‘uncertainty’ is used as the measure of the magnitude of that error.

REFERENCES Agee, M.D. and T.D. Crocker (1994), ‘Parental and social valuations of child health information’, Journal of Public Economics, 55(1), 89–105. Allais, M. (1953), ‘Le Comportement de l’homme rationnel devant le risque: critique des postulats et axiomes deecole Americaine’, Econometrica, 21, 503–46.

308

The international yearbook of environmental and resource economics

Allais, M. (1979), ‘The foundations of a positive theory of choice involving risk and a criticism of the postulates and axioms of the American School’, in M. Allais and O. Hagen (eds), Expected Utility Hypotheses and the Allais Paradox, Dordrecht: Reidel. Åkerman, J., F.R. Johnson and L. Bergman (1991), ‘Paying for safety: voluntary reduction of residential radon risks’, Land Economics, 67(4) (November), 435–46. Antonovitz, F. and T. Roe (1986), ‘A theoretical and empirical approach to the value of information in risky markets’, Review of Economics and Statistics, 68(1) (February), 105–14. Berger, M.C., G.C. Blomquist, D. Kenkel and G.S. Tolley (1987), ‘Valuing changes in health risks: a comparison of alternative measures’, Southern Economic Journal, 53(4) (April), 967–84. Besley, T. (1989), ‘Ex ante evaluation of health states and the provision for ill-health’, The Economic Journal, 99(March), 132–46. Bishop, R.C. (1982), ‘Option value: an exposition and extension’, Land Economics, 58(1) (February), 1–15. Bockstael, N. and I. Strand (1987), ‘The effect of common sources of regression error on benefit estimates’, Land Economics, 63(1) (February), 11–20. Brookshire, D.S., L.S. Eubanks and A. Randall (1983), ‘Estimating option prices and existence values for wildlife resources’, Land Economics, 59(1), 1–15. Buschena, David E. and David Zilberman (1994), ‘What do we know about decision making under risk and where do we go from here?’, Journal of Agricultural and Resource Economics, 19(2), 425–45. Cameron, Trudy Ann (2003a), ‘Individual option prices for climate change mitigation’, forthcoming, Journal of Public Economics. Cameron, Trudy Ann (2003b), ‘Updated subjective risks in the presence of conflicting information: application to climate change’, revised working paper, Department of Economics, University of Oregon. Carson, T. Richard and Robert Cameron Mitchell (2003), ‘Public preferences toward environmental risks: the case of trihalomethanes’, in A. Alberini, D. Bjornstad and J. Kahn (eds), Handbook of Contingent Valuation, Cheltenham, UK and Northampton, MA: Edward Elgar. Cicchetti, C.J. and J.A. Dubin (1994), ‘A microeconometric analysis of risk aversion and the decision to self-insure’, Journal of Political Economy, 102(1), 169–86. Cicchetti, C.J. and A.M. Freeman (1971), ‘Option demand and consumer’s surplus: further comment’, Quarterly Journal of Economics, 85(August), 528–39. Crocker, T.D. and J.F. Shogren (1991), ‘Ex ante valuation of atmospheric visibility’, Applied Economics, 23, 143–51. Dickie, M. and S. Gerking (1997), ‘Genetic risk factors and offsetting behavior: the case of skin cancer’, Journal of Risk and Uncertainty, 15, 81–97. Dreifus, Claudia (2002), ‘The fear factor meets its match: a conversation with David Ropeik’, New York Times, p. D2 (Science Section/Tuesday 3 December). Edwards, F. Steven (1988), ‘Option prices for groundwater protection’, Journal of Environmental Economics and Management, 15(4) (December), 475–87. Eekhoudt, L.R. and J.K. Hammitt (2004), ‘Does risk aversion increase value of mortality risk?’, Journal of Environmental Economics and Management, presented at the AERE workshop on assessing and managing environmental and public health risks (2001, June, Bar Harbor, Maine). Ellsberg, Daniel (1961), ‘Risk, ambiguity and the savage axioms’, Quarterly Journal of Economics, 75(4), 643–69.

Valuing environmental changes in the presence of risk

309

Einhorn, J. Hillel and Robin Hogarth (1997), ‘Decision making under ambiguity’, in John Hey (ed.), The Economics of Uncertainty, Cheltenham, UK and Northampton, MA: Edward Elgar. Evans, W.N. and W.K. Viscusi (1991), ‘Estimation of state-dependent utility functions using survey data’, Review of Economics and Statistics, February, 94–104. Fishburn, P. (1978), ‘On Hand’s new theory of cardinal utility and the maximization of expected return’, Journal of Political Economy, 86, 321–4. Fisher, Ann et al. (1986), ‘New information on the value of reducing risk’, unpublished paper, US Environmental Protection Agency, Washington DC, 15 April. Flynn, J. and P. Slovic (1995), ‘Yucca Mountain: a crisis for policy: prospects for America’s high-level nuclear waste program’, Annual Review of Energy and the Environment, 20, 83–118. Foster, W. and R.E. Just (1989), ‘Measuring welfare effects of product contamination with consumer uncertainty’, Journal of Environmental Economics and Management, 17, 266–83. Fuchs, V.R. and R. Zeckhauser (1987), ‘Valuing health – a priceless commodity’, American Economic Review, 77, 263–8. Graham, Daniel A. (1981), ‘Cost benefit analysis under uncertainty’, American Economic Review, 71(4) (September), 715–25. Graham, Daniel A. (1983), ‘Estimating the “state dependent” utility function’, Natural Resources Journal, 23(3) (July), 649–56. Hammit, J.K. and J.D. Graham (1999), ‘Willingness to pay for health protection: inadequate sensitivity to probability?’, Journal of Risk and Uncertainty, 8, 33–62. Harrington, W. and P. Portney (1987), ‘Valuing the benefits of health and safety regulation’, Journal of Urban Economics, 22, 101–12. Hirschleifer, J. (1965), ‘Investment decision under uncertainty: choice-theoretic approaches’, Quarterly Journal of Economics, 79 (November), 509–36. Ho, J., L. Keller and P. Keltyka (2002), ‘Effects of outcome and probabilistic ambiguity on managerial choices’, Journal of Risk and Uncertainty, 24, 47–74. Innes, R. and D. Cory (2001), ‘The economics of safe drinking water’, Land Economics, 77(1), 94–117. Jakus, P. and W.D. Shaw (2003), ‘Perceived risk and product choice: an application to recreation site choice’, Journal of Risk and Uncertainty, 26(1), 77–92. Jakus, P., M. Riddel and W.D. Shaw (2003), ‘Are climbers fools? Modeling risky recreation’, in N.D. Hanley, W.D. Shaw and R.E. Wright (eds), The New Economics of Outdoor Recreation. Cheltenham, UK and Northampton, MA: Edward Elgar. Jehle, G.A. and P.J. Reny (2001), Advanced Microeconomic Theory, 2nd edn, Boston, MA: Addison Wesley-Longman. Jones-Lee, M.W. (1974), ‘The value of changes in the probability of death or injury’, Journal of Political Economy, 92, 835–49. Kahneman, D. and A. Tversky (1979), ‘Prospect theory: an analysis of decisions under risk’, Econometrica, 47(1), 263–91. Kling, C.L. (1993), ‘An assessment of the empirical magnitude of option values for environmental goods’, Environmental and Resource Economics, October. Knight, F.H. (1956), Risk, Uncertainty and Profit, Boston, MA: Houghton Mifflin. Kunreuther, Howard et al. (1990), ‘Public attitudes toward siting a high-level nuclear waste repository in Nevada’, Risk Analysis, 10(4), 469–84. Lange, A. (2001), ‘A note on decisions under uncertainty: the impact of the choice of the welfare measure’, Theory and Decision, 51, 51–71.

310

The international yearbook of environmental and resource economics

Larson, Douglas M. and Paul R. Flacco (1992), ‘Measuring option prices from market behavior’, Journal of Environmental Economics and Management, 22(2) (March), 178–98. Lattimore, P.K., J.R. Baker and A.D. Witte (1992), ‘The influence of probability on risky choice: a parametric examination’, Journal of Economic Behavior and Organization, 17, 377–400. Leggett, C. (2002), ‘Environmental valuation with imperfect information: the case of the random utility model’, Environmental and Resource Economics, 23(1), 343–55. LeRoy, S.F. and L.D. Singell Jr. (1987), ‘Knight on risk and uncertainty’, Journal of Political Economy, 95(2), 394–406. Lichtenberg, E., D. Zilberman and K.T. Bogen (1989), ‘Regulating environmental health risks under uncertainty: groundwater contamination in California’, Journal of Environmental Economics and Management, 17(1) (July), 22–34. Loomes, G. and R. Sugden (1987), ‘Some implications of a more general form of regret theory’, Journal of Economic Theory, 41(2), 270–87. Loomis, J.B. and P.H. duVair (1993), ‘Evaluating the effect of alternative risk communication devices on willingness to pay: results from a dichotomous choice contingent valuation experiment’, Land Economics, 69(3) (August), 287–98. Machina, M.J. (1982), ‘Expected utility theory without the independence axiom’, Econometrica, 50, 277–323. Machina, M.J. (1987), ‘Choice under uncertainty: problems solved and unsolved’, Journal of Economic Perspectives, 1(1), 121–54. Mason, C.F., J.A. List, C. Settle and J. Shogren (2000), ‘Environmental catastrophes and non-expected utility maximization: an experimental evaluation’, discussion paper, University of Wyoming, Laramie. Pratt, J.W. (1964), ‘Risk aversion in the small and the large’, Econometrica, 32, 122–36. Querner, Immo (1994), ‘The need for alternatives to the expected utility approach in environmental risk economics or “Who is afraid of Russian roulette?” ’, in Rudiger Pethig (ed.), Valuing the Environment: Methodology and Measurement Issues, Dordrecht: Kluwer Academic, pp. 81–119. Quiggin, J. (1982), ‘A theory of anticipated utility’, Journal of Economic Behavior and Organization, 3(4), 323–43. Quiggin, J. (1991), ‘Comparitive statics for rank dependent expected utility theory’, Journal of Risk and Uncertainty, 4, 339–50. Quiggin, J. and R.G. Chambers (1998), ‘Risk premiums and benefit measures for generalized-expected utility theories’, Journal of Risk and Uncertainty, 17, 121–37. Riddel, M. and W.D. Shaw (2003), ‘Option wealth and bequest values: the value of protecting future generations from the health risks of nuclear waste storage’, Land Economics, 79(4) (November), 537–48. Riddel, M. and W.D. Shaw (2005), ‘A theoretically consistent empirical model of generalized expected utility: an application to nuclear waste transport’, revised discussion paper, Department of Economics, University of Nevada, Las Vegas, January. Riddel, M., C. Dwyer and W.D. Shaw (2003), ‘Environmental risk and uncertainty: insights from Yucca Mountain’, Journal of Regional Science, 47(3) (August), 435–57. Schoemaker, P. (1982), ‘The expected utility model: its variants, purposes, evidence and limitations’, Journal of Economic Literature, 20 (June), 529–63.

Valuing environmental changes in the presence of risk

311

Shaw, W.D., M. Riddel and P. Jakus (2003), ‘Can Graham’s ex ante welfare framework accommodate subjective and endogenous risk assessments?’, unpublished discussion paper, Department of Agricultural Economics, Texas A&M University, College Station, TX. Shogren, J.F. (1990), ‘The impact of self-protection and self-insurance on individual response to risk’, Journal of Risk and Uncertainty, 3, 191–204. Shogren, J.F. and T.D. Crocker (1999), ‘Risk and its consequences’, Journal of Environmental Economics and Management, 37, 44–51. Slovic, P. (1987), ‘Perception of risk’, Science, 236, 280–5. Smith, V.K. (1992), ‘Environmental risk perception and valuation: conventional versus prospective reference theory’, in D.W. Bromley and K. Segerson (eds), The Social Response to Environmental Risk: Policy Formulation in an Age of Uncertainty, Boston, MA: Kluwer Academic. Smith, V.K. and W.H. Desvousges (1987), ‘An empirical analysis of the economic value of risk changes’, Journal of Political Economy, 95 (February), 89–114. Smith, V.K. and W.H. Desvousges (1988), ‘Risk perception, learning and individual behavior’, American Journal of Agricultural Economics Proceedings, 70(5) (December), 1113–17. Starmer, C. (2000), ‘Developments in non-expected utility theory: the hunt for a descriptive theory of choice under risk’, Journal of Economic Literature, 38 (June), 332–82. Starmer, C. and R. Sugden (1989), ‘Violations of the independence axiom in common ratio problems: an experimental test of some competing hypotheses’, Annals of Operations Research, 19, 79–102. Tversky, A. and D. Kahneman (1981), ‘The framing of decisions and the psychology of choice’, Science, 211, 453–8. Viscusi, W.K. (1989), ‘Prospective reference theory: toward an explanation of the paradoxes’, Journal of Risk and Uncertainty, 2 (September), 253–64. Viscusi, W.K. (1990), ‘Sources of inconsistency in societal responses to health risks’, American Economic Review Papers and Proceedings, 80 (May), 257–61. Viscusi, V.K. and H. Chesson (1999), ‘Hopes and fears: the conflicting effects of risk ambiguity’, Theory and Decision, 47, 153–78. Viscusi, W.K. and W.N. Evans (1990), ‘Utility functions that depend on health status: estimates and economic implications’, American Economic Review, 79 (June), 353–74. Viscusi, V.K., W.A. Magat and J. Huber (1999), ‘Smoking status and public responses to ambiguous scientific risk evidence’, Southern Economic Journal, 66(2), 250–70. Weisbrod, B.A. (1964), ‘Collective consumption services of individual consumption goods’, Quarterly Journal of Economics, 78, 471–7. Yaari, M.E. (1987), ‘The dual theory of choice under risk’, Econometrica, 55(1), 95–115.

Index Abler, D.G. 68 Abu Qdais, H.A. 13 Abu-Zeid, M. 8 Acharya, G. 73 Adamowicz, W. 174 Adams, R. 200, 210 Adelman, Irma 40 Agee, M.D. 281, 298, 303, 304, 307 Ahmad, Mahmood 14 Ahuja, G. 200, 210 air pollution 107–8, 110–11 Akerloff, G.A. 83 Åkerman, J. 298, 307 Al Nassay, H.I. 13 Albers, H.J. 76, 77, 82 Alesina, Alberto 21, 258 Allias, M. 291 Alyward 171 Amundsen, E.S. 267, 275 Anand, P.B. 10, 39 Anas, A. 53 Anderson, C. 7 Anderton, Douglas L. 102, 105, 141 Ando, A. 80, 165, 166 Andrews, M. 255 Angelsen, A. 77 Anselin, L. 54 Antle, J.M. 82 Anton, W. 198 Antonovitz, F 282 Antweiler, Werner 111, 143 Appleton, K.J. 74 Arnott, R. 53 Arora, S. 63, 108, 198 Arriaza, Manuel 12 Arrow, K. 169, 186 Atkinson, A.B. 227, 239 Atlas, Mark K. 103, 114, 142 Ayres, I. 188 Babcock, B.A. 69 Backeberg, Gerhard 35, 36

Bacow, Lawrence S. 127, 128, 129, 132 Baden, Brett M. 116 Balick, M.J. 172 Balmford, A. 166, 176 Bandaragoda, D.J. 16, 18 Barbier, E.B. 171, 175, 176 Bartelmus, P. 188 Bastian, C.T. 70 Bateman, I.J. 55, 71, 72, 73, 74, 76 Batie, S.S. 69 Bauer, C.J. 12 Bauer, D.M. 74 Baumann, D. 5, 12 Baumol, William J. 15, 226 Baxter, Jamie W. 120 Bayindir-Upmann, T. 257, 275 Bebbington, J. 186 Becker, Gary S. 37 Becker, Nir 20, 77 Becker, R. 61, 66, 84 Beecher, Janice A. 10 Been, Vicki 63, 66, 116, 139, 143 Beierle, Thomas C. 136 Beissinger, S.R. 167 Belenky, Lisa T. 136 Bell, K.P. 55 Belz, F. 212 benefit estimation methodologies 71–2, 72–3, 74 Bennett 203 Bennett, L. 73 Berck, Peter 20, 37 Berger, M.C. 280 Bergstrom, T.C. 275 Berle, Adolf A. 26 Berman, E. 62, 84 Besley, T. 283, 292, 294 Bevers, M. 168 Bhat, M.G. 79 Bhatia, Ramesh 15 Binswanger, Hans 12, 18, 20 biodiversity 312

Index conservation strategies 163–9 cost-effectiveness analysis frameworks 160–63 direct market incentives 170–172 importance 157–9 international policies 175–6 national policies 173–5 spatial dimension to management 80–81, 168 bioprospecting 171–2 Bishop, R.C. 158, 307 Bjorndal, T. 78 Bjornlund, Henning 12 Blanchard, O.J. 256 Bockstael, N.E. 53, 54, 69, 71, 72, 73, 82, 84, 86, 299 Boer, Tom 116 Boerner, Christopher 117, 144 Boeters, S. 275 Bohm, P. 58 Boisvert, R.N. 70 Boland, John J. 18, 20 Bonbright, James C. 12 Bontemps, Christophe 8 Bord, Richard J. 117 borders 62–3, 100, 109 Bos, M.G. 20 Bosch, D.L. 69 Bosello, F. 274, 275 Bosworth, B. 11, 19 Botsford, L. 160 Bouwman, Arno 110 Bovenberg, A.L. 223, 226, 261, 274 Bowen, William 101 Boyce, James K. 116 Boyce, M.S. 167 Boyer, Robert 31 Boyle, K.J. 73, 74 Boyle, M.A. 72 Braden, J.B. 58, 69 Bradford, Mark 136 Brainard, Julii S. 55, 71, 72, 110 Braithwaite, J. 188 brand value 204 Brandon, K.E. 171 Braungart, M. 190 Briscoe, John 7, 15 Broadbent, Jeffrey 119 Brock, Philip 21 Bromley, Dan W. 34

313

Brooks, Nancy 107 Brookshire, D.S. 158, 298, 307 Brouwer, Joust 81 Brown, G.M. Jr. 173, 176 Brown, Stephen J. 8, 12 Brundtland, G. 186, 187, 189, 194 Brunello, G. 256 Bryant, Bunyan 101 Bucholtz, S. 73 Bucovetsky, S. 258 Bui, L. 62, 84 Bulckaen, F. 274 Bullard, Robert D. 101, 116 Bullock, D.S. 81 Bulte, E.H. 77, 170, 175 Burritt, R. 187, 195, 198, 202, 203, 205, 208, 210 Burt, C.M. 38 Burton, Dudley 116 Burtraw, D. 274 Buschena, David E. 291 Calkin, D.E. 168 Callens, I. 209 Cameron, Trudy Ann 295, 302, 303, 305 Camm, J.D. 80, 165, 166 Canada 111, 133–4 Cansier, D. 205 carbon sequestration 81–2 Carbone, J.C. 84 Carpenter, S.R. 163, 169 Carpentier, C.L. 69 Carraro, C. 256, 274, 275 Carruthers, I. 10, 16 Carson, Richard T. 117, 128, 281, 297 Cason, T.N. 63, 108, 198 Castle, Geoffrey 133 census tracts 102 Chakraborty, Jayajit 108 Chakravorty, U. 78 Chambers, R.G. 297 Chape, S. 164 Chay, K. 61, 85 Chesnutt, Thomas W. 10 Chesson, H. 305 Choe, K. 10 Chomitz, Kenneth M. 77 Christiansen, Kim Michael 109 Church, R. 165

314

International yearbook of environmental and resource economics

Church, R.L. 165 Cicchetti, C.J. 287, 291, 307 Ciriacy-Wantrup, S.V. 4 Clark, C. 163 Clark, Lloyd 108 Clark, Robert E. 4 Coase, Ronald H. 4, 29, 121–3, 123, 128, 132 Cody, Brendan 138 Coenen, Frans H.J.M. 121 Cohen, M.A. 60, 198, 200, 210 Cole, D. 111, 143 compensation 117, 118–19, 126–9 competitiveness 201–2, 257–63, 258 competitiveness dividend 257–63 compliance 201 Connor, Michael 109 Conrad, J. 174 conservation, and water pricing 12–14 conservation easements 70–71 conservation strategies 163–9 contingent valuation 74 Contreras, Dante 11, 37 Copeland 204 Coppel, J. 275 Cordeiro, J. 200, 210 Coriat, Benjmin 31 Cormier, D. 210 corporate environmental performance and economic success 197–201 corporate sustainability challenges for 190–195 characteristics 188–9 data sources 211–12 future direction 213–15 and general sustainability 186–8 importance to management 201–6 inter-characteristic relationships 195–7 interest in 185 measuring 209–12 strategies 206–9 corruption 22, 25 Cory, D. 306 cost recovery schemes, water 19 Costanza, R. 163 Costello, C. 169 costs, reduction of 203 Coursey, Don L. 116 Couture, Stephane 8

Crisp, Brian F. Crocker, T.D. 281, 295, 298, 303, 304, 307 Cropper, Maureen L. 77, 114 Csuti, B. 165 Cummings, Ronald 10, 15 Cutter, Susan L. 108 Daily, G.C. 158, 163, 167, 170, 171 Daly, H. 186, 187 Damon, L. 198 Daniels, Glynis 108 Dasgupta, Partha S. 20, 37, 169, 267 data sources 211–12 David, Paul 31 Davos, Climis A. 121 Day, B. 73 Day, R. 200 de Mooij, R.A. 226, 231 Deacon 52, 54 Decker, C. 67, 199 Deegan, C. 207 DeHaven, James C. 15 Deily, M.E. 61 demographic categories 99–100 Dente, Bruno 134 Desvouges, W.H. 295, 297, 306 developing countries 33–6 Diamond, P.A. 238, 239 Diao, Xinsen 20, 37 Dickie, M. 298 Dinar, Ariel 3, 5, 6, 9, 10, 11, 12, 13, 14, 15, 19, 20, 21, 22, 26, 29, 32, 33, 35, 37, 39, 40 distinctiveness 161 Dixit, Avinash 29 Dobkin, C. 61 Dolk, H. 109 Doremus, H. 173 Dosi, C. 68 Dosi, Giovanni 31 double-dividends 223–4, 268–9 Douglas, Mary 32 Doukkali, Rachid M. 15 Dowell, G. 200 Downing, D.A. 163 Dreifus, Claudia 294 Dubin, J.A. 291 Dublin Principles 1 Dunlap, Riley E. 118

Index duVair, P.H. 297, 307 Dyllick, T. 189, 191, 193, 194, 203, 207, 213 Eakins, John 3 Easter, K. William 11, 12, 15, 16, 18, 20 Easterling, Douglas 118 eco-effectiveness 190–191 eco-efficiency (E2-efficiency) 191–2 eco-efficiency portfolio 195 ecotourism 171 Edelgaard, Irene 109 Edwards, F. Steven 307 Eeckhoudt, L.R. 288, 292 Eggertsson, Thrainn 21 Ehrenfeld, D. 158 Eichenberger, Reiner 118, 144 Einhorn, J. Hillel 305 Ekins, P. 186, 187 Elkington, J. 189 Elliott, Susan K. 120 Ellison, K. 170, 171 Ellsberg, Daniel 305 embeddedness 31–2 emission permits, tradable 59–60 emission taxes, spatial dimension 58–9 employment dividend 242–55, 257 energy, resource management 75 Engfer, Victoria L. 136 environmental regulation borders and 62–3 conservation easements 70–71 non-point source pollution 68–70 and plant location 66–8 point-source pollution 60–62 and political activism 63–4 spatial dimension 53, 57 and spatial inequalities 64–6 environmental taxation classical view 225–6 with commodity tax system 231–5 distributional considerations 238–42 effects of discussing 268 and labor markets 257–63 with labour tax system 231–2 modelling 226–30 separability of consumption and environmental quality 227, 235–8 and world prices 263–8

315

see also competitiveness dividend; employment dividend EPA (US Environmental Protection Agency) 115, 137–9 ESA (US Endangered Species Act) 173 Eskeland, G.S. 67 Espey, J. 12, 15, 39 Espey, M. 12, 15, 39 Esty, D. 198 EUM (expected utility model) 283–5, 285–8, 288–90 Europe 134–5 Evans, W.N. 304, 309 Exxon Valdez oil spill 202 Eyles, J. 111, 143 Eyles, John D. 120 facilities, hazardous waste 98 Faith, D.P. 162 Falkenmark, Malin 1 Farber, Stephen 115 Fareri, Paolo 134 Farzin, Y.H. 266 Faustmann, M. 74 Feder, G. 18 Feeny, David 33 Feldstein, M.S. 239 Ferraro, P.J. 174 Fetini, Habib 40 Field, B. 174 Figge, F. 203, 204, 209 Finkelstein, Israel 19, 20 Fischer, Frank 119 Fishburn, P. 296 Fisher, A.C. 158, 169 Fisher, Ann 306 Fisher, Frank 144 fisheries 78–9 Fitzgerald, Kevin 120, 134 FitzRoy, F. 227, 235 Flacco, Paul R. 300, 306, 307 Fleming, M. 83 Florax, Raymond J.G.M. 81 Flynn, J. 295 Folmer, Henk 66 Fombrun, C. 204 Fonteyne, Jacques 109 Foreman, Christopher H. 101 forestry 75–7 Foster, D.R. 77

316

International yearbook of environmental and resource economics

Foster, Sheila 137 Foster, W. 282 Franklin, J. 167 Fredriksson, Per G. 121 Freeman, A.M. 287, 307 Freimann, J. 191 Frey, Bruno S. 118, 144 Friedman, M. 198 Friedman, Samantha 108 Fuchs, V.R. 288, 289 Fujita, M. 53 Fullerton, D. 232, 235, 274 Gabel, L. 199 Galbraith, John Kenneth 26 Galeotti, M. 256, 274, 275 Gallo, M. 256 Garber-Yonts, B. 80 Gardner, Roy 31, 32, 33 Garfield, Paul J. 12 Garrod, G.D. 71 Gaussier, Nathalie 120 Gayer, Ted 115 Genereux, John 115 Genereux, Michelle 115 Geoghegan, J. 52, 54, 55, 71, 72, 73, 77, 84, 86 Georges, Elizabeth 137 Gerking, S. 53, 298 Germino, M.J. 70 Gerrard, Michael B. 121 Gibbons, Diana C. 3, 7 Gladwin, T. 189, 213 Gleick, Peter 1 Goetz, R.U. 69 Golan-Hardy, Elise 40 Goldin, Ian 20, 37 Goldman, George 20, 37 Gómez-Lobo, Andrés 11, 37 Goulder, L.H. 226, 234, 274 Graham, Daniel A. 284, 285, 286, 287, 302, 307 Graham, John D. 108, 297 Gray, David A. 77 Gray, W.B. 61, 62, 64, 65, 66, 83, 86 Greenstone, M. 61, 67, 84, 85, 86 Gregory, Robin 121 Grief, Avner 32 Griffin, Charles C. 15 Griffin, Ronald 15

Griffiths, Charles 77 Groothius, Peter A. 117, 118 Gruber, J. 246 Guana, Eileen 138 Guesnerie, R. 275 Gunningham, N. 188, 205 Gupta, Francis 63, 66, 116, 143 Gupta, Shreekant 114 Gutés, M. 187 Guttman, J.M. 29 habitat conservation 164–7 Haggard, Stephan 21 Haggard, Stephen 21, 33 Hahn, T. 209 Haight, R.G. 168 Hajispyrou, Soteroula 13 Hall, Darwin 15, 37 Hallstrom, D.G. 84 Halstead, John M. 117 Hamermesh, Daniel S. 249 Hamilton, James T. 64, 104, 105, 112, 114, 115, 130, 140, 142, 143, 210 Hamilton, Lawrence C. 117 Hammitt, J.K. 288, 292, 297 Hampton, Greg 140 Hanemann, Michael W. 8, 15, 37, 122, 158 Hansen, L.G. 16 Hardie, I.W. 53 Hardin, Garrett 20 Harjula, Henrik 109 Harrington, W. 281 Harris, V. 77 Harrison, A.E. 67 Harrison, D. Jr. 227 Harrison, Kathryn 111, 143 Hart, S. 189, 200, 210 Haveman, R. 7, 14, 15 Hawken, P. 192 Hayer 187 hazardous waste 101, 108–9 hazardous waste facilities, siting of and compensation 117, 118–19, 126–9 definitions 98, 105 and demographic minorities 101, 105, 113–14, 116–17 determinants of exposure to 112–21

Index and health risks 106–7, 109–10 and house prices 115–16 incorporating environmental equity 137–9 and low-incomes 97, 101, 105–6, 112–13, 116–17, 139 siting difficulties 120–121 studies samples 141–4 theory 121–6 see also TSDs (treatment, storage and disposal facilities) hazardous waste storage 108 Heal, G.M. 37, 267 health effects 99, 106–7, 109–10 Hearne, Robert 12, 18 Hector, A. 163 hedonic property models 72–3 Heijungs, R. 190 Helland, E. 62 Hellerstein, D. 77 Henderson, J.V. 61, 66 Hesselborn, P.-O 58 Hipp, John 117, 144 Hird, John A. 105, 114 Hirshleifer, Jack 15, 280 Hite, Diane 115 Ho, J. 299 Hochman, E. 58, 78 Hockerts, K. 189, 191, 193, 194, 213 Hockman, Elaine M. 108 Hoel, M. 267 Hoeller, P. 275 Hof, J. 168 Hoffmann, A. 212 Hogarth, Robin 305 Holling, C.S. 163 Hollingsworth, J. Rogers 32 Holm, P. 251 Holmlund, B. 256 Honkapohja, S. 251 Hotelling 52, 74, 75 Howe, Charles W. 8, 12, 15, 37 Howitt, R.E. 18 Huber, Joel 118, 206, 207 Hueting, R. 188 Huffaker, R.G. 79 Huitema, Dave 119, 135 Humphries, C.J. 157 Hyami, Y. 33 hydropower 2

317

Ibitayo, O.O. 132 imperfect information 282–3 income levels and hazardous waste sites 97, 101, 105–6, 112–13, 116–17, 139 and hazardous waste storage 108 and polluting facility locations 110 TSDs and 102–4 Ingberman, Daniel E. 121 Inhaber, Herbert 128 Innes, R. 173, 306 intellectual property rights 172 irrigation 2–3, 5, 10–11, 12 Irwin, E.G. 52, 53, 54, 55, 69, 72, 73, 82, 299 Isik, M. 81 Israel, Arturo 21 Jackman, R. 256 Jackson, J. 158 Jacobs, Timothy L. 121 Jaeger, W.K. 274 Jaffe, A. 210 Jakus, P. 298, 299, 307 Jantzen, D. 170 Japan 135–6 Jehle, G.A. 284 Jenkins, R. 64 Jeppesen, Tim 66 Jerrett, M. 111, 143 Johansson, O. 274 Johansson, Robert C. 11, 15 Johnson, S. 210 Johnson, Sam H. 34 Johnson, Stephen M. 109, 136, 137 Johnston, R.J. 74 Jones, K. 210 Jones-Lee, M.W. 287 Jones, Tom 14 Just, R.E. 282 Just, Richard E. 8 Kahn, M. 85 Kahneman, D. 293, 303 Kaimowitz, D. 77 Kareiva, P. 169 Karl, H. 204 Karp, L. 268 Kastenholz, Hans 119, 187 Katz, L. 256

318

International yearbook of environmental and resource economics

Keating 186 Keller, W. 67 Kelly, Michael J. Kemper, K.E. 18 Khanna, M. 69, 81, 198 Kiel, Katherine A. 72, 115 Kijne, Jacob W. 22 Kim, H. 73 King, A. 200 Kirchgässner, G. 274 Kislev, Yoav 19, 20 Kiss, A. 174 Klassen, R. 210 Kleinhesselink, Randall 119 Kling, C.L. 300 Kloezen, W.H. 12, 37 Knight, F.H. 282 Kolm, A.-S. 256 Kolstad, C.D. 67, 75, 275 Konar, S. 200, 210 Konrad, K.A. 268 Koo, B. 172 Koskela, E. 76, 243, 251, 254, 255, 258, 272, 275 Kraft, Michael E. 118 Krautkraemer, J.A. 275 Kregel, Jan A. 39 Kreisel, W. 63 Krier, James E. 125 Krinner, C. 12 Krueger, Anne 21 Krugman, P. 53 Kruize, Henneke 110 Krumm, R.D. 59 Kumar, Dinesh M. 12 Kummer, Katharina 136 Kunreuther, Howard 117, 118, 127, 134, 281, 285 labor markets, and green taxes 257–63 Lafay, Jean-Dominique 21 Laffont, J.-J. 20, 275 Lake, I.R. 73 Lallana, C. 12 Lambert, Thomas 117, 137, 144 Lampietti, Julian A. 39 Lange, A. 288 Larson, B.A. 12 Larson, Douglas M. 300, 306, 307 Lattimore, P.K. 304

Laurai, Don 10, 35 Lawrence, D.P. 134 Lawton, J.H. 157 Layard, R. 256 Layton, D. 176 Lazarus, Richard 138 Lee, D.R. 226 Leggett, C. 282 Lejano, Raul P. 121 Lenhart, S.M. 79 Lenox, M. 200 LeRoy, S.F. 282 Lesbirel, S. Hayden 119, 144 Letey, J. 12 Lévêque, F. 205 Levin, S. 157 Levinson, A. 67 Li, S. 163 Lichtenberg, E. 307 Lichtenstein, M.E. 168 Lichty, R. 7 Lidskog, R. 134 Ligteringen, Josee 134 Linaweaver, F.P. 8, 15 Lindberg, K. 171 Lindblom, C.K. 209 Linde, Claas van der 198 Linnerooth-Bayer, Joanne 120, 134 Lintner, A.M. 69 List, John 53, 85 List, John A. 67, 85 Liu, Feng 116 Liu, J. 171 Lockwood, B. 251 Lofstedt, Ragnar E. 136 Lomborg, B. 157 Long, N.V. 268 Loomes, G. 292 Loomis, J.B. 297, 307 Lovejoy, Wallace F. 12 Lovett, A.A. 72, 73, 74, 76 Lowenberg-DeBoer, J. 81 LULU (locally undesirable land uses) 124, 129 Lyle, June 138 Lynch, L. 71 MacArthur, R. 157 MacDougall, G.D.A. 258 Machina, M.J. 292

Index Madariaga, B. 7 Magorian, Christopher 129, 132 Maguire, K. 64 Magurran, A.E. 162 Maille, P. 171 Mäler, K.-G. 169 Malik, Arun S. 12 Mani, Muthukumara 77 Mank, Bradford 137, 138 Mann, C.C. 173 Manning, A. 251 Mansfield, Carol 118 many-person Ramsey tax rule 238–42 Marbug, Hugh 136 Marett 10 Margolis, M. 85 Mariño, M. 18 Markusen, J. 53 Marshall, E. 168 Marvier, M. 169 Mascollel, A. 9 Mason, C.F. 292, 307 Mason, Stephen G. Jr. 129 Massaruto, Antonio 5, 8, 14, 19 May, R.M. 157 Mayeres, I. 274 McCann, Richard J. 21 McCarl, B.A. 82 McClain, Katherine T. 115 McConnell, K. 7 McCubbins, Matthew 115 McCullough, D.R. 167 McDonough, W. 190 McDougall, Forbes 109 McGrady-Steed, J. 163 McKay, Jennifer 12 McKee, J.K. 157 McLaughlin, C. 210 McLeod, D.M. 70 McLeod, H. 111 Means, Gardiner C. 25, 26 Mendelsohn, R. 171, 172 Merrifield, J. 174 Mertens, B. 77 Metcalf, G.E. 235, 274 Michelsen, Ari 3, 7 Milgrom, Paul 29 Milkey, James R. 128, 132 Miller, D.J. 70 Miller, Gail 117, 118

319

Miller, James N. 121 Miller, K.R. 167 Milliman, Jerome W. 15 Millimet, D.L. 64, 83, 108 Mills, D. 174 Milne, M.J. 186 Minehart, Deborah 121 minorities, demographic and air pollution 107–8 and cancer 107 and current hazardous waste risks 107 and hazardous waste sites 101, 105, 113–14, 116–17 and hazardous waste storage 108 and political activity 112–13 TSDs and 102–4 Miranda, Marie Lynn 121 Misiolek, W.S. 226 Mitchell, Robert Cameron 117, 128, 281, 297 Mittermeier, R. 169 models hedonic property 72 monocentric city 52–3 revealed preference 298–9 stated preference 297–8, 305 Mody, Jyothsna 14 Mohai, Paul 101, 116 monocentric city model 52–3 Montgomery, C.A. 168 Moore, Michael R. 15, 16, 19 Morell, David 129, 132 Morello-Frosch, Rachel 108 Morgan, C. 64 Morris, Charles M. 108 Morrisson, Christian 21 mortality risk 281, 289–90 Mu, X. 10, 35 Muller, D. 77 Munc-Kampmann, Birgit 109 Munro, G.R 78 Munroe, D.K. 77 Munton, Don 133, 135 Murphy, Sean D. 136 Myers, N. 157, 169 Myers, R.A. 158 Na, L. 201 Naaem, S. 163

320

International yearbook of environmental and resource economics

Nadaï, A. 205 Nagaraj 26 Nalle, D.J. 168 Narain, Vishal 18, 39 Narayanamurthy, S.G 18 natural resource management 75–82 Neeman, Zvika 121 Nelson, Arthur C. 115 Nelson, Gerald C. 55, 77 Nercissiantz, V. 10, 15 nestedness 31–2 Neumayer, E. 187 Newbery, D.M. 268, 275 Nichols, A.L. 226 Nickell, S. 255, 256 Nijkamp, P. 53 NIMBY (Not-In-My-Backyard) 117–18, 122–3 Noah’s Ark Problem 160–62 Noll, Roger C. 125 non-volumetric methods 17, 18, 27 see also water pricing, methods Nordblom, T.L. 80 North, Douglass C. 21, 28, 29, 31, 32 Norton, B. 158 Noss, R. 158 NPS (non-point source) pollution 68–70 nuclear waste risks 285, 291, 294–5 Oates, Wallace E. 15, 53, 226 Oberholzer-Gee, Felix 117, 118, 144 O’Connor, Robert E. 117 O’Hare, Michael 121, 127, 129 Ohkawara, Toru 135 Ollikainen, M. 76 Olsen, T.E. 268 Olson, Mancur 21, 124 O’Neill, Kate 136 Opaluch, James J. 121 Openshaw, S. 109 option price (ex-ante welfare measure) 285–8 O’Riordan, T. 187 Orosel, G. 274 Orwat, C. 204 Osgood, D.E. 85 Ostrom, Elinor 21, 31, 32, 33 O’Sullivan, Arthur 121

Page, William G. 109 Pagiola, S. 170 Palmquist, R.B. 72 Panayotou, T. 174 Park, Rozelia S. 136 Parkhurst, G.M. 80 Parks, P.J. 53 Parmesan, T. 158 Parry, I.W.H. 233, 234, 274 Pastor, Manuel 108 Pastor, Manuel Jr. 117, 144 Paterson, R.W. 73, 74 Pearce, David 97, 188 Pekelney, David M. 10 Peltzman, Sam 37 Perotti, R. 258 Perrings, C. 163, 169 Perry, C.J. 18 Perry, G. 15 Pezzey, J.C.V. 186 Pfaff, Alexander S.P. 77 Pigou, A.C. 223, 225 Pigovian taxation 225–6 Pigram, J.J. 18 Pijawka, K.D. 132 Pimm, S.L. 157 Pines, D. 58 Pint, E. 16 Pinzon, Lillian M. 136 Pirttilä, J. 227, 236, 274 Plantinga, A.J. 70 Ploeg, F. van der 226, 275 Plummer, M.L. 173 Polasky, S. 80, 158, 159, 160, 162, 167, 168, 169, 173 political activism 63–4, 104–5, 112–15 pollution 2, 3, 59–60, 107–8, 110–111, 121–6 Porter, M. 198, 201 Portney, P. 281 Poulos, C. 73, 86 Pounds, J.A. 158 Prato, T. 69 Pratt, J.W. 280, 288 Pratts, Derek 109 Pressey, R.L. 165 Principle, P. 158 Probst, Katherine 136 Prokop, Gundula 109

Index Proost, S. 58, 274 Proposition 2.1 236 Proposition 2.2 241 Proposition 3.1 251 Proposition 3.2 254 Proposition 3.3 254 Proposition 4.1 261 Proposition 4.2 262 Provencher, W. 20 PRTRs (Pollutant Release and Transfer Registers) 109, 140 Puri, Jyotsna 77 Puschendorf, R. 158 Pye, Steve 111 Qiu, Z. 69 Quah, E. 121 Querner, Immo 287 Quiggin, J. 296, 297 Rabe, Barry G 133, 134 Ragaini, R.C. 109 Raith, M.G. 257, 275 Ramasubban, Radhika 15 Randall, Alan 34 Rankin, M. 199 Rao, P.K. 18 Rappaport 204 Rauscher, M. 53 Rausser, Gordon C. 21, 158, 172 Razin, A. 258 Reader, S. 111, 143 Rees, W. 190 Regmi, A. 70 Reid, D. 187 Reid, W.V. 167 Reijnders, L. 188 Reinhardt, F. 196, 200 Renard, V. 174 Renfro, R.Z.H. 18 Renn, Ortwin 119, 120 Reny, P.J. 284 Renzetti, Steven 12, 15, 16, 20 Repetto, R. 236 research, required conservation cost-effectiveness models 177 conservation incentives 177 corporate sustainability 215 risk modelling 301–5

321

spatially explicit approach 82–7 water pricing 37–8 reserve site selection 164–7 ReVelle, C. 165 Rhodes, George 10 Ribaudo, M.O. 12, 69 Richards, Alice 133, 134 Richman, Barak D. 121 Richter, W. 258 Riddel, Mary 125, 285, 294, 295, 297, 305 Rinaudo, Jean-Daniel 22, 40 Ringquist, Evan J. 108 risk(s) characterizing 98–9 decision weights 296–7 as endogenous 281 measurement 280 modeling 99–100 public communication of 295 revealed preference models 298–9 stated preference models 297–8, 305 subjectivity 294–5 using stated probabilities 295–6 see also uncertainty risk management 202 Ristoratore, Mario 134 Rivera, Daniel 39 Roberts, J. 29 Robinson, E.J.Z. 77, 82 Robinson, Sherman 20, 37 Rodrigues, A.S.L. 164 Roe, Terry 15, 20, 37, 282 Rogers, George O. 117 Rogers, P. 7, 8 Rohweder, M.R. 168 Roland-Holst, D. 20, 37 Rolston, H. III 158 Root, T.L. 158 Rosa, Eugene A. 118, 119 Rose, M. 240 Rosegrant, Mark W. 12, 18, 20 Rosenbaum, P.R. 85 Rosenzweig, M.L. 167 Rosenzweig, M.R. 84 Roson, R. 58 Rubin, D.B. 85 Rubin, P. 210 Ruttan, Vernon W. 33

322

International yearbook of environmental and resource economics

Sabbaghi, Asghar 20, 35 Sadd, James 108, 117, 144 Sadka, E. 258 Saleth, R.M. 11, 15, 18, 26, 29, 32, 33, 35, 37, 39, 40 Sampath, R.K. 10, 15, 20 Sanchirico, J.N. 78, 160 Sanderson, Debra 121, 129 Sandmo, A. 226, 237, 242, 274 Santiago, Tessa Mayer 138 Sarkis, J. 200, 210 Savedoff, William 14, 29, 35 scale economies 32–3 Schaltegger, S. 185, 187, 190, 192, 193, 195, 197, 198, 202, 203, 204, 206, 208, 210, 212 Schamann, Martin 109 Schaper, M. 194, 206 Scheffer, M. 163 Schmidheiny, S. 187, 192, 198 Schmidt-Bleek, F. 190, 192 Schmitter 32 Schneider, Elke 120 Schneider, K. 255, 258, 275 Schöb, R. 227, 231, 232, 234, 236, 243, 251, 254, 255, 257, 258, 267, 268, 272, 274, 275 Schoemaker, P. 291, 294, 306 Scholz, C.M. 275 Schurmeier, D.R. 12 Schwab, R. 53 Schwartz, J. 236 Schwarz, Thomas 115 Scott, Susan 3 Seckler, David 1 Sedjo, R.A. 77 Seeliger, Robert 134 Segerson, K. 201 Sethi, Rejiv 107 Shackle, George G.L. 39 Shadbegian, R.J. 60, 61, 62, 64, 65, 66, 83, 86 Shah, F 18 Shah, T. 10, 16, 18 shareholder value 204 Shaw, Daigee 135 Shaw, W.D. 12, 15, 125, 285, 294, 295, 298, 299, 305, 307 Shenkar, O. 204 Shogren, J.F. 80, 174, 281, 295

Shortle, J.S. 68 Shubik, Martin 21 Sibley, David S. 8, 12 Sigman, H. 62, 64 Simons, G.P.W. 174 Simpson, R. David 158, 172, 174 Sinclair-Desgagné 199 Singell, L.D. Jr. 282 Singh, Banwar 15 Singh, O.P. 12 Sinn, H.-W. 243, 254, 267, 275 Sjoberg, Lennart 120 Slottje, D. 64, 83, 108 Slovic, Paul 280, 295, 307 Small, A. 158, 172 Small, K.A. 53, 58 Small, L.E. 10, 16 Smith, Hank Jenkins 117 Smith, M.D. 79 Smith, Rodney B.W. 20, 80, 174 Smith, S. 227, 242 Smith, V.K. 72, 73, 84, 86, 287, 295, 297, 304, 306, 307 Snary, Christopher 121 socio-effectiveness 191 socio/societal-efficiency 191–2 Solow, A. 158, 159, 162–3 Soule, M.E. 167 Southworth, J. 77 Sparling, E.W. 18 spatial dimension 52–7 spatial econometrics 54 species management 79–80, 167–9 Spiller, Pablo 14, 29, 35 Spulber, Nicolas 20, 35 Stallings, Barbara 21 Stampini, M. 274 Starmer, C. 291, 292, 293, 294 Steurer, R. 187 Stiglitz, J.E. 227, 239 Stone, C. 188 Stone, S.W. 77 Strand, J. 256 Strannegard, L. 212 Sturm, A. 192, 206 Subramanian, Ashok 14, 22, 33 Sugden, R. 292, 293, 294 Sullivan, Arthur M. 121, 127 Summers, L. 246 Sundram, Muthu S. 136

Index sustainability portfolio 195–7 sustainable development 187, 196 Swallow, Stephen K. 74, 75, 118, 121 Swanson, T. 175 Swinton, S.M. 81 Synnestvedt, T. 197, 210 Taft Morris, Cynthia 40 Tai, Stephanie 138 Talukdar, P. 75 Tan, K.C. 121 Terkla, D. 226 Thisse, J.-F. 53 Thobani, M. 18, 20 Thomas, C.D. 158 Thorsnes, P. 174 Tiebout, C.M. 53 Tietenberg, T. 15, 59, 188 Tilman, D. 163 Tirole, J. 20 Toman, M.A. 186 Tomasi, T. 68 Tool, Marc R. 25, 26 transaction costs, water pricing reform 29–30 travel costs 71–2 Truedsson, Jana 120 TSDs (treatment, storage and disposal facilities) definitions 101 and demographic minorities 102–4 siting of planned expansion 104–5 and low-incomes 102–4 practice in other OECD 133–7 practice in US 129–33 proportion of US waste handled 103 see also hazardous waste facilities, siting of Tsur, Yacov 5, 6, 8, 9, 10, 11, 15, 17, 18, 20, 22, 37, 39 Tucker, C.M. 77 Tullock, G. 226 Turner, B.L. 77 Turner, R.K. 188 Tversky, A. 293, 303 Tyteca, D. 209, 210 Uimonen, S. 59 UK 110–11, 135

323

Ulph, A. 266 Ulph, D. 266 Umetsu, C. 78 uncertainty 290–97, 299–301 see also risk(s) Underhill, L.G. 165 urban water 11–12 US 101–8, 112–18 Van Houtven, George 114, 118 Van Kooten, G.C. 53, 77, 170, 175 Vance, C. 55, 77 Vane-Wright, R.I. 162 Vari, Anna 121 Vaughan, S. 202 Vaux, H.J. 18 Veblen, Thorstein B. 26 Venables, A.J. 53 Verchick, Robert 136 Vergara, R. 246 Vermillion, Douglas L. 20, 34 Viklund, Mattias 120 Viscusi, W. Kip 64, 105, 114, 115, 142, 143, 293, 303, 304, 305 Voisey, H. 187 volumetric methods 16–18, 27–8, 34 see also water pricing, methods Von Weizsäcker, E. 192 Voortman, Roelf L. 81 VSL (value of statisical life) 289–90 Wackernagel, M. 190 Waehrer, Keith 121 Wagner, M. 204, 209, 210, 211, 212 Wainger, L. 54, 72 Walker, B. 163, 169 Walker, Gordon 109 Walker, James 31, 32, 33 Walley, N. 198 Walsh, Maureen 136 Walters, W. 20 Ward, Frank A. 3, 7 Ware, Caroline F. 25 Warrick, J. 157 water 1, 1–3, 5, 7–8, 77–8 water markets 12, 17, 18 water pricing and cost recovery 19 effectiveness 3–4, 14–19

324

International yearbook of environmental and resource economics

by end usage 5 expectations 8–10 impact 12–14 as an institutional configuration 25–8, 26–7, 30–31 as institutional mechanism 26 institutional study approach 25–7 methods 6, 10–12, 27–8 and rights 3–4, 12, 34 roles of 2 water pricing institutions 31–3 Water Pricing Progress Index 22 water pricing reform in developing countries 33–6 institutional economics 25–36 political economy 20–22, 25 promoting 33 stakeholders 23–4 timing 35 transaction costs 29–30 Waugh, Theodore 136 Wear, D.N. 75, 167 Weaver, Thomas F. 121 Webb, Stephen B. 21, 33 Weber, M. 174 Webler, Thomas 119 Webster, D. 176 Weersink, A. 69 Wehrmeyer, W. 210 Weinberg 16 Weisbrod, B.A. 307 Weiss, M.D. 81 Weitzman, M.L. 158, 159, 160, 162–3 Wellisch, D. 59 Wells, M. 171 Welsch, D.E. 16, 18 Welsh, M.P. 158 Western, D. 171 Wheeler, Michael 121 Whitcomb, Joanna L. 117 White, F. 210 White, Louise G. 21, 29, 33 White, M.J. 59 Whitehead, B. 198

Whitford, A.B. 62 Whitman, Christine Todd 137 Whittington, Dale 10, 15, 18, 20, 35 Wichelns, Dennis 3, 38 Wiegard, W. 240 Wilcove, D. 158 Wilen, J.E. 78, 79, 160 Williams, J.C. 77, 82 Williams, R.C. III 234, 236, 274 Williamson, John 21, 29 Williamson, Oliver E. 28, 29 Wilson, E.O. 157, 158 Wilson, J.D. 258 Wilson, P.N. 18 Winter-Nelson, A. 81 Wirl, F. 267 Withagen, C. 275 Wittman, D. 59 Wolpin, K.I. 84 Wolverton, A. 63, 66, 235 Worm, B. 158 Wright, B. 172 Wright, R.M. 171 Wunder, S. 171 Xepapadeas, A. 59 Xing, Y. 67 Yaari, M.E. 292 Yandle, Tracy 116 Yang, Tseming 137 Yang, W. 69 Yaron, D. 3, 18 Yohe, G. 158 Young, Robert A. 7, 14, 15, 18, 37, 39 Ytterhus, B. 212 Yuchtman-Yaar, E. 204 Zeckhauser, R. 288, 289 Zekri, Slim 13 Zeller, M. 77 Zilberman, David 3, 13, 18, 21, 39, 58, 77, 78, 291 Zimmerman, Rae 105, 114