Environmental Economics Research Paper

Early in the history of economic modeling, Hotelling (1931) and more recently Dasgupta and Heal (1979) as well as numerous other economists recognized the importance of intertemporal issues for economic choices involving renewable and nonrenewable natural resources. These types of economic analyses are treated separately in the Encyclopedia.

Environmental economics assumes there are things people care about, either directly because they enhance individual well-being or indirectly because they influence the production of other commodities, that are: (a) scarce; (b) available outside markets; and (c) influenced by the activities of others. These objects of environmental choice (OECs) have diverse characteristics and result from a variety of activities. Six topic areas organize this overview: the allocation problems they pose, the design of policy, measures for the amounts of environmental amenities and their economic values, OECs as risks, the distributional implications, and interactions between growth and environmental resources.

1. Problems—Public Goods, Externalities, And Congestion

The definitions for public goods and externalities are closely linked to the conditions required to sustain private property rights. As a result, the allocation problems they create are sometimes reduced to descriptions of causes and remedies for market failures.

Pure public goods are nonrival and nonexcludable in consumption (Cornes and Sandler 1996). Each person’s consumption of a public good does not reduce the amount available to others. Moreover access to these goods cannot be limited. While it is hard to provide examples satisfying these ideal conditions, it is easy to cite examples of goods and services with attributes close to a pure public good’s features. Local air quality, measured in terms of the concentration of airborne pollutants, varies with location but within large areas is approximately the same for all individuals living there. The global concentrations of greenhouse gases (e.g., CO , methane, etc.) can be described as examples of pure public ‘bads’ because emissions of them at any location have equivalent effects on their atmospheric stocks. However, the consequences of those accumulations for climate change and local weather conditions will depend on where people live. Perfect adherence to the ideal definition of a public good is not essential to consider the implications of ‘publicness’ for the efficiency of different resource allocation mechanisms.

Externalities describe a class of interactions between two or more economic agents. One group’s (or individual’s) consumption, production, or exchange activities have unintended effects on the production or consumption of others. These external effects resemble public goods (or bads) in that they can have the same nonrival and nonexcludable attributes.

Congestion describes a phenomenon whose externality effects reduce the nonrival attributes of a public good. For example, over some levels of recreational use of a beach or a wilderness area (and the levels will be different in each case) further users can be supported without extra cost or diminution in the satisfaction each person realizes from his or her use. However, at some point, this simultaneous use diminishes each person’s experience. The phenomena linking those losses to the added users of a resource at the same time is described as congestion.

While all three of these concepts are relevant outside environmental applications, they have found special meaning when environmental resources provide the services treated as OECs or as the assets whose services provide the conduits for the negative or positive interactions between economic agents.

These concepts are important for at least two reasons. First, the incentives created by private markets do not assure that resources will be allocated to their highest valued uses when one or more of these problems is involved. Second, the information available from markets cannot be assured to signal and evaluate the performance of economic institutions. Much of environmental economics can be interpreted as responses to these issues.

2. Policies—Context, Design, And Performance

Economic analysis seeks to abstract from the ‘details’ of a problem, focusing instead on the incentive properties of each decision context. In the case of environmental problems these details can matter to the insights offered by economic analysis. For example, economic descriptions of production and consumption activities must be consistent with materials and energy balances (Kneese and Bower 1968). When they are not, it is easy to overlook the residuals and the problems that must be faced in disposing of them. This insight is important because most externalities arise from using the services of an environmental asset (i.e., the ambient atmosphere or a free flowing river) for residual disposal. The form and severity of environmental externalities depend on both the types of residuals involved and the functions of the ecosystem receiving them. As a rule, the ecosystems are open access resources whose services in some uses are rival applications. They may well also be subject to depletion over time.

Economic prescriptions for these externalities (as well as for congestion) tend to focus on creating the ‘right’ incentives. Private choices should reflect their full or social incremental cost (i.e., the private marginal cost of producing any commodity plus the marginal external costs due to any damages associated with a commodity’s production, consumption, and ultimate disposal). When the marginal damages associated with the production or consumption of a good are measured per unit of each residual emitted, it is sometimes described as the price that should be charged for that emission. Pigouvian taxes or effluent charges set equal to these marginal damages are one type of policy response to externalities.

Under simplified conditions, efficient resource allocation in the presence of environmental externalities (or congestion) requires that, at the margin, agents face the full cost (or benefit) of their actions including the damages caused directly or indirectly from each choice. Those experiencing damages are assumed to take defensive actions and costs computed recognizing these choices, but the injured parties do not need to be compensated (Baumol and Oates 1988).

With full information, policy can take a variety of forms. These can be charges per equivalent unit of impact (taking account of the ecosystem) to the agents causing externalities; permits to emit pollution that can be traded; and even subsidies to abate emissions. Each of these policy measures can be made to appear equivalent in terms of the incentives for incremental choices. Each approach requires different implementation rules and a different set of adjustments to reflect their effects at the extensive margin of choice (e.g. the entry and exit decisions of firms). Charges vs. subsidies will affect incentives to enter an industry (Carlton and Loury 1980). Tradable emission permits must be denominated in units reflecting the effects of emissions on the services associated with ambient environmental quality that defines the OECs. They must also consider both the initial distributions of permits among firms, the rules for trading, and their ability to be saved (if temporal patterns of emissions are important).

Unfortunately, in the setting where actual decisions must take place, information is scarce and uncertainty abounds. If uncertainty affects the marginal control costs, then it can affect the preferred policy instrument. Comparing charges vs. marketable permits (denominated in units of the OEC) Weitzman (1974) has shown the choice depends on the relative steepness of the marginal benefit and cost schedules. Tradable quotas are favored if the marginal damage function is steeper than marginal abatement cost and effluent charges for the reverse. His results depend on the approximate linearity of these schedules. Moreover, the conclusion is a relative statement not an absolute judgment. Composite strategies using price and quantity instruments together can be shown to be better in the sense that they enhance overall welfare (Roberts and Spence 1976).

This summary implies that with perfect information either instrument can lead to an efficient resource allocation. Charging for emissions or distributing permits to pollute seems to convey ‘rights’ to different economic agents to use ecosystems’ services. The conclusion that they can be equivalent may seem counterintuitive. While surprising it is one reflection of the general implications of the Coase (1960) theorem. That is, in a world of perfect information, no transaction costs, and no income effects (from assigning rights to ecosystem services) the theorem demonstrates the assignment of rights will not affect the efficient outcome. It is often used in arguments that externalities can be resolved without public sector intervention by assigning property rights to ecosystem services. In most cases the practical relevance of this conclusion is an empirical question. Nonetheless, the economic logic underlying the theorem also explains why the available policy instruments can be equivalent.

Until recently, most environmental policy was driven by the broad goals of the existing legislation. In the USA the rulemaking process, implementing this legislation, has favored command-and-control (CAC) regulations. Specific technologies were prescribed to meet objectives such as using the best available control technology to establish the lowest emission rate. These types of engineering standards specified, at the unit process level, emission rates for the same types of pollutants that sometimes were quite different for separate equipment within the same plant. The result was very different incremental costs for controlling the same pollutants. Abatement costs were also increasing over time as more stringent controls were imposed. Areas found to be in nonattainment with environmental standards faced restrictions on further economic activity. The result was a policy environment seeking some ways of increasing the flexibility in how firms could respond. Some regions in the USA had started the process of ‘regulatory reform’ and introduced some movement toward the use of economic incentives when the 1990 Clean Air Act dramatically changed the US regulatory climate. It offered the first large-scale emission permit trading system as a way of realizing more stringent controls for sulfur dioxide. Ellerman et al. (2000) estimate that emission trading yielded cost savings by 33 to 67 percent of the most reasonable CAC alternative. Permit trading schemes are being considered or used in small regional markets for an array of environmental resources—including air and water pollutants, wetlands, habitat for endangered species, and carbon dioxide (see Tietenberg 1990).

Most economic choices about market goods and services cannot be analyzed realistically without recognition of their environmental consequences. This requirement is symmetrical. Environmental policy analysis cannot ignore the market and regulatory context for the products produced along with the environmental externalities. Two sets of insights have been especially important to reaching this conclusion. The earliest, associated with Buchanan (1969), argued that a monopolistic product market may, by restricting output, serve environmental goals. In this context, imposing effluent charges or restricting emissions with tradable quotas could actually lead to welfare losses.

The logic underlying this result is more general. Judgments about the preferred incentives to realize an efficient resource allocation in an economy with externalities and multiple distortions are extremely difficult. An emerging issue indicating how such distortions can affect the design of environmental policy began with the promise of ‘double dividends’ from effluent charges on CO emissions (Bovenberg and de Mooij 1994). Subsequent research rediscovered the issues associated with designing optimal product taxes in the presence of multiple taxes—an important theme of the optimal taxation literature (Auerbach 1985). The double dividend argument relies on using the revenues from environmental fees (or authorized permits) to reduce the marginal rates of one or more large, pre-existing taxes. It is suggested that the result of the tax substitution may be zero or negative gross cost. Subsequent research has extended arguments about the potential importance of pre-existing tax distortion to evaluations of the effects of new environmental regulations, suggesting in the case of SO , costs over 70 percent higher than they would be in the absence of the taxes (Goulder et al. 1997). Nonetheless it is important to recognize how specific this conclusion is to the circumstances described in each model. Changing the assumptions about how private goods relate to the externality in consumers’ preferences will affect the size of the excess burden attributed to existing taxes and therefore the gains from reducing them. What is less likely to change is the attention the argument implies should be given to the political economy of how and why environmental legislation takes specific forms.

3. Amounts: The Role Of Nonmarket Valuation And Benefit-Cost Analysis

Nearly all analysis of policy instruments assumes we know how much environmental quality to provide. Answering this question involves more than deciding the amount of pollution permitted. It requires consideration of a wide range of natural assets whose amenity services are important to people. Everything from endangered species and their habitat to land and water resources that provide essential capacity for a wide range of outdoor recreation must be considered. Some choices must be made more carefully than others must because some allocations of a natural environment will irreversibly alter its ability to provide amenity services (Krutilla 1967, Krutilla and Fisher 1975).

The extent of ‘publicness’ of the OECs compounds the difficulties in making these choices. Free-riding incentives arising from nonexclusion assure that private action alone cannot be relied upon to realize the efficient level of these types of resources (Cornes and Sandler 1996). Recovering information equivalent to what is provided when market exchanges correctly signal scarcity requires economic ‘detective work.’

Two strategies have been developed since Hicks (1943) first defined rigorously the monetary representations for the tradeoffs defined by economic choices (Freeman 1993). The first relies on the logic of revealed preference and uses choices of private goods to recover the tradeoff information. For example, it recognizes that a consumer’s decision to purchase (or rent) a home in an area free of air pollution reflects, in part, a payment for that ambient quality in comparison to the next best alternative (Palmquist 2000). With a wide array of alternative conditions, the price gradients arising from different air quality conditions, with other housing attributes held constant, would reveal the incremental willingness to pay for improving air quality (Freeman 1993). Similar logic implies one could use the trips taken to recreation sites at locations with different travel costs to reveal the demand for the recreation these sites provide (Freeman 1993). Given a set of recreation sites, varying in quality conditions, consumers’ visitation decisions offer the potential to measure the incremental value of those quality dimensions. Decisions to avert or mitigate the negative consequences of pollution offer the same potential to reveal underlying tradeoffs (Freeman 1993).

The logic underlying each strategy relies on reconstructing the unobserved elements required to define the economic tradeoff revealed with each choice. In the case of nonmarketed goods this process implies that a connection is also defined between the decision and the OEC of interest. For the tradeoff to be informative about an individual’s preferences for the nonmarketed good these connections must convey a degree of exclusively (i.e., economic value of the amenity requires consumption of the linked private good). They must also specify how enhancements in an individual’s ability to pay would change them. One way this control over the substitution and income effects of choice is implemented comes from the joint imposition of restrictions on the properties assumed in describing an individual’s utility function. Weak complementarity (Maler 1974) and the Willig (1978) condition are among the most commonly used. Weak complementarity maintains that the environmental amenity is of value when it is used with the private good. Without some positive consumption of that good and improvement in the amenity has no value to each person. The Willig condition adds to this restriction a requirement that the incremental value of increases in the amenity, per unit of the private good consumed, do not change with income.

Stated preference approaches to recovering economic tradeoffs offer the second nonmarket valuation strategy. They rely on presenting people believable (but hypothetical) choices with proposed financial consequences and interpreting their stated decisions as bounding economic tradeoffs (Mitchell and Carson 1989). An important motivation for proposing choices arises from the inability to link all the ways amenities enhance people’s well-being to observable choices of private goods. Such unlinked reasons are sometimes referred to as nonuse or existence values (Krutilla 1967). Stated preference methods are controversial because they require that survey responses be treated as economically consequential (Diamond and Hausman 1994, Hanemann 1994).

The primary use for nonmarket valuation estimates has been in benefit-cost analyses of projects involving environmental resources and regulations intended to improve environmental quality. Benefit-cost analyses (BCA) assume there are reasons why an existing resource allocation is not efficient. The analysis seeks to evaluate whether an aggregate choice improves the efficiency (Arrow et al. 1996). Four aspects of applying BCA are especially important. First, the net benefits from an aggregate choice are summarized as a present value and are influenced by the time horizon and discount rate used. Both are potentially influential and controversial judgments (see Portney and Weyant 1999 for a summary of the issues in making a choice). Second, large changes in relation to the overall economy have general equilibrium effects that may require evaluating the effect at the household level and aggregating (i.e., computing net benefits at the household level rather than aggregating benefits and costs separately, see Hazilla and Kopp 1990). Third, BCA analyses will always face incomplete information about the changes being evaluated from a baseline to a new outcome. When an irreversible choice is being made and there is the possibility for resolving some of the uncertainty over time, the analysis should include the conditional value of any new information that would be consequential to the decision (Arrow and Fisher 1974). Finally, monetization of all factors creating benefits or costs is not essential for BCA to be informative. Observed choices (at the individual level) define bounds for the tradeoffs implied by a decision. This same logic can be used to delineate bounds for the missing elements in a BCA. That is, we define what the missing elements in a measured value for net benefits would need to be to support a specific choice (Krutilla and Fisher 1975).

4. Risk: Perceptions And Values

One of the most important classes of OECs involves environmental sources for changes in the probabilities of undesirable outcomes. For example, air pollutants have been associated with premature death and increased episodes of serious illness. These decisions are highlighted for two reasons. First, the framework for evaluating private and public choice must consider attitudes toward risk and the opportunities available to adjust to it. Within a framework that evaluates decisions made before the uncertain events are resolved, there has been considerable discussion of how these risks should affect monetary measures of policy outcomes and discounting. There seems to be broad consensus that attitudes toward risk and opportunities for private and public adjustment should be built into the monetary measures of ex ante net benefits and not to the discount rates (Portney and Weyant 1999). Second, studies of laypersons’ risk perceptions and behavior in both experimental and survey contexts suggest marked discrepancies between subjective beliefs and experts’ recommendations. Equally important this behavior contrasts with the predictions of an expected utility model for individual choice. These differences are especially important for small probability and long latency outcomes (Slovic 1987). Such findings raise questions with what constitutes an informed choice and potential limits to consumer sovereignty (Machina 1990). They also highlight the role of information and risk communication as policy instruments.

5. Distribution

The term distribution in most economic analyses of policies would lead to a discussion of equity effects, interpreted as evaluating the consequences of actions for different income groups. For environmental policy, distributional considerations are inherently more complex. At least four perspectives have been considered. First, the use of an aggregate net benefits criterion for evaluating policy itself is a distributional judgment. It implies departures from a strict efficiency standard and a willingness to admit policies where there exists the potential for improvements. Positive aggregate net benefits means those gaining could compensate those losing. It does not require the compensation to take place.

The second distributional perspective follows from the identification of long lived, potentially irreversible externalities—with greenhouse gases and nuclear waste being two widely recognized examples. In these cases the distribution of costs and benefits extends across generations and the equity issues are often framed in ethical terms. One important implication of this line of reasoning is that policies should treat all generations alike. As a result, this line of reasoning would seem to imply a zero discount rate should be used. However, there is an equally strong intuition from a different perspective suggesting it is not morally acceptable. It implies a decision criterion (i.e., the zero discount rate) that would demand excessively high savings rates of any one generation. This countervailing influence is important because Koopmans’ (1960) classic analysis of choice rules over infinite dimensional consumption streams concluded that these rules must display impatience or some amount of positive discounting. Otherwise, it would require the first generation to sacrifice everything. With preference separability over time, Arrow concluded irreversibility does not change conclusions in favor of intertemporal discounting (see Portney and Weyant 1999). Of course to the extent preferences or production display important substitutions or complementarities with irreplaceable environmental resources, then the guidance will depend on the details of those linkages (Portney and Weyant 1999).

Economic and environmental feedback effects can have significant distributional consequences within short time horizons. Several cases illustrate how this third perspective can arise. Deciding between command and control vs. incentive based policies can seem clear-cut, considering the total costs of meeting a uniform standard (Tietenberg 1990). However, the environmental transfer system, linking emissions to ambient concentrations, can imply the cost savings are realized by allowing the cleanest areas to become more dirty while those exceeding the standard become cleaner. These properties can impact political acceptability as well as different income and demographic groups.

Much less is known about the economic feedback loops connecting environmental policy in one sector to its impacts on others. The economic models used to compute general equilibrium responses usually describe activities as if they all took place at the same ‘point-in-space’ (Hazilla and Kopp 1990). Recent adaptations to these models (Espinosa and Smith 1995) and to household equilibrium location models (Sieg et al. 2000) confirm the importance of relaxing these assumptions. Indeed, the results from the equilibrium location model suggests that spatially differentiated, but consistent, air quality improvements can induce housing price adjustments (within a closed system) that imply some households would actually be worse off after an improvement in their air quality. This result arises because the housing price increases induced by some households’ relocations can offset the air quality gains for those households with a small taste for environmental public goods.

The last distributional perspective, paradoxically, receives the greatest attention in agencies’ design of policies (in the US some legislation specifically identifies concern for the impacts of regulations on small businesses) and often the least in academic discussions. It concerns the impact of regulations on industries and different sized firms within a sector. Empirical evidence confirms that environmental regulations can have effects on the timing, location, and scale of plant investments. Considering only air quality regulations in the US, they find that activity shifts to areas that meet ambient standards, not subject to stringent area wide regulations. Less activity takes place in areas that are not in attainment with national ambient air quality standards. When new plant investments do take place they seem to involve larger up-front investments in comparison to what takes in the attainment areas. In one of the sectors, these authors studied (industrial organic chemicals) their results suggest plants grew to smaller sizes with regulation than in the preregulation era (Morgenstern et al. 1998).

6. Growth And The Environment

One of the most durable questions posed in considering interactions between the economy and the environment concerns whether the environmental system poses some ‘ultimate limit’ on the scale of economic activity (see Barnett and Morse 1963). Sometimes expressed in terms of raw materials (or natural resources) and other times in terms of ability to provide a sustainable food supply and waste absorptive services to a growing world population, the issue most recently has been expressed in terms of whether growth in income has a self-regulating property. That is, as household income grows, the taste for environmental amenities increases. Providing these services requires reducing pollution, protecting natural environments that provide habitat for endangered species, and enhancing absorptive capacity of ecosystems for some types of pollution.

At one scale, this issue has been posed using what is termed an ‘environmental Kuznets curve’ (Grossman and Krueger 1995). Pollution is hypothesized to have an inverted U-shape with household income. It increases until income growth creates demand for amenities that ultimately reduces pollution and leads to policies that restore environmental quality. To date, the questions raised in this line of research have been based on empirical analyses, relying on the contemporaneous performance of economies varying in pollution and household income, as if they could be treated as descriptions of a single economy moving through a dynamic process.

This new formulation of the old question is potentially important. It casts the issue in its most relevant form. Economists are concerned about the services provided by environmental resources because market institutions cannot be assured to signal the scarcity of these resources. Understanding both the static distortions to patterns of use and the dynamic nonmarket incentives to bring about changes in the utilization of these natural assets is important for designing new institutions that recognize the role of environmental resources in assuring that it is possible to continue to enhance economic well-being.

Bibliography:

1. Arrow K J, Fisher A C 1974 Environmental preservation, uncertainty, and irreversibility. Quarterly Journal of Economics 88: 312–9
2. Arrow K J, Cropper M L, Eads G C, Hahn R W, Lave L B, Noll R G, Portney P R, Russell M, Schmalensee R, Smith V K, Stavins R 1996 Is there a role for benefit cost analysis in environmental, health and safety regulation? Science 272: 221–2
3. Auerbach A J 1985 The theory of excess burden and optimal taxation. In: Auerbach A J, Feldstein M S (eds.) Handbook of Public Economics. North-Holland, Amsterdam, Vol. 1
4. Barnett H J, Morse C 1963 Scarcity and Growth. Johns Hopkins University Press, Baltimore, MD
5. Baumol W F, Oates W E 1988 The Theory of Environmental Policy, 2nd edn. Cambridge University Press, Cambridge, UK
6. Bovenberg A L, de Mooij R A 1994 Environmental levies and distortionary taxation. American Economic Review 84(4): 1085–9
7. Buchanan J M 1969 External diseconomics, corrective taxes and market structure. American Economic Review 59: 174–7
8. Carlton D W, Loury G C 1980 The limitations of pigouvian taxes as a long run remedy for externalities. Quarterly Journal of Economics 95: 559–66
9. Coase R H 1960 The problem of social cost. Journal of Law and Economics 3: 1–44
10. Cornes R, Sandler T 1996 The Theory of Externalities, Public Goods, and Club Goods, 2nd edn. Cambridge University Press, Cambridge, UK
11. Dasgupta P S, Heal G M 1979 Economic Theory and Exhaustible Resources. Cambridge University Press, Cambridge, UK
12. Diamond P A, Hausman J A 1994 Contingent valuation, is some number better than no number. Journal of Economic Perspectives 8: 45–64
13. Ellerman A D, Joskow P L, Schmalensee R, Montero J P, Bailey E M 2000 Markets for Clean Air: The U.S. Acid Rain Program. Cambridge University Press, Cambridge, UK
14. Espinosa J A, Smith V K 1995 Measuring the environmental consequences of trade policy: A non-market CGE analysis. American Journal of Agricultural Economics 77(August): 772–7
15. Freeman A M III 1993 The Measurement of Environmental and Resource Values: Theory and Methods. Resources for the Future, Washington, DC
16. Goulder L H, Parry I W H, Butraw D 1997 Revenue raising versus other approaches to environmental protection: The criticasignificance of pre-existing tax distortions. Rand Journal of Economics 28: 708–31
17. Grossman G M, Krueger A M 1995 Economic growth and the environment. Quarterly Journal of Economics 110: 353–77
18. Hanemann W M 1994 Valuing the environment through contingent valuation. Journal of Economic Perspectives 8(4): 19–43
19. Hazilla M, Kopp R J 1990 Social cost of environmental quality regulations: A general equilibrium analysis. Journal of Political Economy 98: 853–73
20. Hicks J R 1943 The four consumer surpluses. Review of Economic Studies 11(1): 31–41
21. Hotelling H 1931 The economics of exhaustible resources. Journal of Political Economy 39: 137–75
22. Kneese A V, Bower B T 1968 Managing Water Quality: Economics, Technology and Institutions. Johns Hopkins University Press for Resources for the Future, Baltimore, MD
23. Koopmans T C 1960 Stationary ordinal utility and impatience. Econometrics 28: 287–309
24. Krutilla J V 1967 Conservation reconsidered. American Economic Review 57: 777–86
25. Krutilla J V, Fisher A C 1975 The Economics of Natural Environments: Studies in the Valuation of Commodity and Amenity Resources. Johns Hopkins University Press for Resources for the Future, Baltimore, MD
26. Machina M J 1990 Choice under uncertainty: problems solved and unsolved. In: Hammond P B, Coppock R (eds.) Valuing Health Risks, Costs and Benefits for Environmental Decision Making. National Academy Press, Washington, DC
27. Maler K-G 1974 Environmental Economics: A Theoretical Inquiry. John Hopkins University Press for Resources for the Future, Baltimore, MD
28. Mitchell R C, Carson R T 1989 Using Surveys to Value Public Goods, The Contingent Valuation Method. Resources for the Future, Washington, DC
29. Morgenstern R D, Pizer W A, Shih J S 1998 The Costs of Environmental Protection. Paper No. 98-36, Resources for the Future, Washington, DC (May)
30. Palmquist R B 2000 Property value model. In: Maler K-G, Vincent J (eds.) Handbook of Environmental Economics. North Holland, Amsterdam
31. Portney P R, Weyant J P (eds.) 1999 Discounting and Intergenerational Equity. Resources for the Future, Washington, DC
32. Roberts M J, Spence M 1976 Effluent charges and licences under uncertainty. Journal of Public Economics 5: 193–208
33. Sieg H, Smith V K, Banzhaf H S, Walsh R 2000 Estimating the General Equilibrium benefits of Large Policy Changes: The Clean Air Act Revisited. NBER Working Paper 7744 presented to Econometric Society, Boston, Massachusetts (June)
34. Slovic P 1987 Perception of Risk. Science 236: 280–5
35. Solow R M 1991 Sustainability: An Economist’s Perspective. 18th J. Seward Johnson Lecture, Marine Policy Center, Woods Hole Oceanographic Institution (June 14)
36. Tietenberg T H 1990 Economic instruments for environmental regulation. Oxford Review of Political Economy 6(1): 17–33
37. Weitzman M L 1974 Prices versus quantities. Review of Economic Studies 41: 477–91
38. Willig R D 1978 Incremental consumer’s surplus and hedonic price adjustment. Journal of Economic Theory 17(2): 227–53