In responding to Dr. Sagoff's critique, Dr. Kopp focuses on the philosophical attack on preferences. He rejects the notion that preferences are hypothetical expressions of choice under ideal conditions and that choice and preference are substitutes for one another. Rather, Kopp argues, human actions are a result of choice based on preference, and it is the choices of individuals that environmental economists use to infer underlying preferences. He also rejects the idea that ethical values not grounded in considerations of personal benefit cannot be included in economists' definition of individual welfare. Kopp contends that preferences include preferences for characteristics of the social state that could be associated with ethical concerns. Kopp rebuts the claim that satisfaction of preferences has nothing to do with individual well-being by pointing out that, for economists, the ultimate criterion for deciding what is good and what is bad for a given individual is his or her own wants and preferences. The root of the disagreement between philosophers and economists regarding the concept of individual well-being and what makes a person better off or worse off, Kopp suggests, is not really related to the satisfaction of preferences but to economists' rejection of paternalism. Economists and philosophers, Kopp says, can agree on a list of elements that give rise to well-being; where they part company is on the weighting of each of the elements on the list.
Mark Sagoff argues forcefully that decision makers will find little that is helpful in the vocabulary of environmental economics as they wrestle with questions of environmental and natural resource policy, and that the concepts that define the approach of environmental economics have outlived their usefulness. His critique of environmental economics is provocative and, if his assessment of the discipline is valid, it is damning as well. I think it can be shown, however, that his assessment is not valid.
Is environmental economics (or resource economics) dead, as the title of Dr. Sagoff's article suggests? It would not appear to be. Indeed, it is more popular in policy analysis and policy formulation today than at any other time in its brief history. The Clean Air Act Amendments of 1990, for example, incorporate novel provisions for reducing sulfur dioxide emissions from coal-fired power plants through a system of marketable emissions permits—a system advocated by economists for at least twenty years and supported by the Environmental Defense Fund during congressional debate. Moreover, the amendments themselves carry a statutory provision that benefit-cost studies of the regulations required by the act be conducted at periodic intervals, with the results presented to Congress.
Perhaps the most far-reaching use of environmental economics is found in the natural resource damage provisions of the Comprehensive Environmental Response, Compensation, and Liability Act of 1980 (CERCLA, also known as Superfund) and the Oil Pollution Act of 1990. Both acts stipulate that parties responsible for damages to natural resources or the environment due to the release of hazardous substances or oil must pay to restore the resources and pay the "value" of the natural resource services lost as a result of the injury. These acts use an explicit economic paradigm to define value. Moreover, the promulgated regulations defining the manner in which values are to be estimated rely on techniques developed over the last twenty years by environmental economists.
In contemporary environmental policy, environmental economics appears to be thriving. The discipline's development of the concept of nonuse value (also known as existence value and passive use value), introduced more than twenty-five years ago by John V. Krutilla of Resources for the Future, has changed the manner in which Congress and the courts view environmental protection and damages to natural resources. The ability of environmental economists to estimate nonuse values—for example, the values individuals may derive from the mere knowledge that an unspoiled natural area exists—allows the ethical concerns of individuals with respect to natural resources to be explicitly incorporated into quantitative analyses of environmental programs. This innovation represents more than a simple enhancement of analytical capability; it represents a significant broadening of the scope of issues open to economic analysis, including issues in other disciplines previously closed to economic analysis.
An overview of Sagoff's critique
The thrust of Sagoff's critique of environmental economics comes in two parts. In the first, which concerns the philosophical underpinnings of the discipline, he suggests that the inadequacies of environmental economics are a result of the abandonment of choice for preference, the artificial nature of preferences in general, and the disassociation between preference satisfaction and individual welfare. In the second part, which concerns the use to which the discipline is put in policymaking, he argues that decision makers should discard environmental economics as a source of relevant information and rely instead on ethics, aesthetics, cultural history, religion, and other fields. If Sagoff reserves any room for economics in the environmental policy debate, it is in the realm of macroeconomics; however, he states that environmental economics, being basically microeconomics and concerned with allocative efficiency, has nothing relevant to say.
I believe the first part of the critique to be the more substantive and will devote most of this article to my response to Sagoff's philosophical attack on preferences. First, however, I will briefly respond to his notion that environmental economists have little or nothing to say about the macroeconomic implications of environmental policy, and address his misconceptions about the association between allocative efficiency and macroeconomic performance.
Allocative efficiency and macroeconomic performance
I suppose that, in Sagoff's eyes, I am a macroeconomist when I write about the effect of environmental regulation on employment, economic growth, prices, and the like, and that I am an environmental economist when I write about the benefits to agriculture of ambient air quality standards (for example). Much of the work that has been done on the macroeconomic consequences of environmental regulations—most notably on the cost of these regulations—has been done by microeconomists, myself included, using general equilibrium microeconomic models of the U.S. economy.
To be fair, Sagoff is really not asserting that environmental economists ignore macroeconomic implications, but rather that allocative efficiency, which he takes to be the Holy Grail of environmental economics, has nothing to do with macroeconomic performance. However, allocative efficiency is not the goal of environmental economics, as Sagoff suggests it is. Rather, its goal is to maximize social welfare.
Allocative efficiency is a necessary but not sufficient condition for welfare maximization, and no economist would ever suggest that it is. In order to maximize welfare, resource allocations must not only be efficient but equitable. For economists and others, the concept of equity encompasses everything from individual rights to the distribution of wealth and income.
Even if allocative efficiency were the goal of environmental economics, one would still not expect to see a link between the allocative efficiency of environmental regulation and macroeconomic performance. The reason is quite simple. Allocative efficiency means allocating resources to maximize some objective. One can imagine a small environmental program (small in terms of resources consumed by the program) that is very inefficient, but due to its small size has virtually no effect on the larger economy. On the other hand, it is just as easy to imagine a very efficient, even perfectly efficient, mammoth project that has a considerable macroeconomic impact. The economist's concern with allocative efficiency is a concern about waste and the attainment of goals—it is not about macroeconomic effects. If our goal is environmental improvement and the chosen policies do not lead to the attainment of that goal, they are inefficient, wasteful, and socially undesirable, regardless of their macroeconomic effects.
Allocative efficiency is a necessary but not sufficient condition for welfare maximization; economists' concern with allocative efficiency is not about macroeconomic effects but about waste and the attainment of goals.
Preference and welfare theory
The substance of Sagoff's critique lies not in his desire to emphasize macro implications but rather in his philosophical attack on the economic concept of preference. How should one assess the validity and thus the importance of this attack? A logical approach would begin with the identification of the precise elements of environmental economics that are being questioned. It is clear that Sagoff does not question all the activities performed by environmental economists, but only a small portion of those activities—specifically, benefit-cost analyses. Moreover, he notes that it is the application of the theory of welfare economics to environmental and natural resource policy that is at the root of his concern. I would conjecture that Sagoff would be just as critical of the use of the theory in any decision-making process involving the provision of public goods, whether the decisions involve federal tax policy, construction of space stations, local expenditures for sewer systems, or environmental policy.
It would be unfair to say that Sagoff opposes the balancing of costs and benefits in public decision making. Rather, he criticizes the use of a welfare-theoretic paradigm in the quantification of the costs and benefits of contemplated actions.
What is it about the welfare-theoretic paradigm that so irritates Sagoff? He argues that economic theory "abandons choice for preference," that "preferences . . . are simply artifacts or constructs of economic theory itself," and that "even if preferences did exist as a foundation for 'rational' choice, economists offer no plausible reason why environmental policy should seek to satisfy them." The substance of the critique seems to boil down to the role of choice and preference in welfare theory. But if this is true, we are left with the realization that Sagoff's critique extends far beyond welfare theory, and in fact encompasses all of economics proper. Since his critique is focused on choice, preferences, and the relationship of preference to social welfare, I shall respond by addressing each of these issues in turn.
Economics is about the choices made by businesses and consumers. In its purely objective form, economic theory seeks to explain the actual choices businesses and consumers make. The sub-area of economics known as welfare theory is a theory of optimal choices. It is a positive theory when it describes the optimality properties of actual choices and is a normative theory when it specifies how optimal choices might be made given politically specified objectives. There is simply no support for the argument that economics has abandoned choice in favor of preference—the study of choice is the economist's raison d'être, and preference is the attribute of individuals that gives rise to specific choices. Choice and preference are not substitutes for one another as Sagoff seems to think, nor are they names for the same thing, or alternative theories. Choice is a process economists endeavor to understand; preference, combined with a posited behavioral objective (often termed utility maximization), is the reigning theory for explaining choice.
Sagoff is correct when he states that preferences as employed in welfare theory are theoretical constructs and are not observable. However, when he states that "preferences do not exist in the mind of the individual; rather, they exist in the eye of the beholder," he implies that individual preferences are somehow illusions of economists and nonexistent in the real world. Moreover, he implies that observation of individual behavior can say nothing about a person's preferences even if they exist. Sagoff should realize, however, that in economics introspection goes a long way toward casual verification of theory. I know, for example, that I prefer coffee to tea, silver cars to black cars, and hiking to fishing. Readers of Resources could easily assemble a similar preference list. Because individuals can articulate their preferences and anyone observing an individual's behavior can infer those preferences, one cannot dismiss preferences as the inane creation of economists. Two of Sagoff's high priests may see different things in chicken entrails, but people display the indelible mark of their preferences in their daily actions. Any two environmental economists looking at the same data on individuals' recreational activity, for example, will draw the same general conclusions about their recreational preferences. Moreover, on the basis of these conclusions, economists can endeavor to predict the recreational behavior of these same individuals under different circumstances. If the predictions are reasonably accurate, one can have some confidence that economists' description of preferences is reasonably correct.
Economics has not abandoned choice in favor of preference; choice is a process economists endeavor to understand, while preference, combined with a posited behavioral objective, is the reigning theory for explaining choice.
The preferences that economists infer from the behavior of individuals are not hypothetical expressions of choice under ideal conditions, as Sagoff seems to believe. There is nothing hypothetical about the reduction in beach recreation one might observe after the beach has been contaminated by an oil spill. Nor is there anything hypothetical about the outcome of local elections in which citizens vote to tax themselves in order to protect specific aspects of their environment. Each of these acts is a result of choice based on preference, and made freely under real-world conditions. It is such choices that environmental economists use to infer the underlying preferences of individuals. Sagoff is simply wrong to suggest that economists reject the actual choices of individuals and impose their own paternalistic views. Rather, economists build their entire empirical theory of value upon such actual choices. As I shall argue below, for a philosopher to term economics paternalistic is surely the pot calling the kettle black.
I shall make one further point about the notion of preferences. Economists posit that individuals define their preferences for states of nature and society. This implies that I not only prefer coffee to tea, but also have preferences with respect to the manner in which I behave and the manner in which I like society to behave. These latter preferences may be characterized as codes of conduct or personal ethics but, regardless of their characterization, they define preferred modes of individual and social behavior. The point emphasized here is the breadth of preferences employed in economic theory; preferences extend beyond those that exist for goods, services, and activities to include those for characteristics of the social state that one might associate with broader ethical concerns.
Sagoff would argue that the contamination of preferences by ethical concerns makes it impossible to define personal welfare on the basis of such preferences. This argument derives from his assumption that individual welfare, as defined by economists, is coincident with personal benefit and that ethically based preferences are not grounded "in considerations of personal satisfaction or well-being." I will say more about the economic notion of individual welfare below, but here it is sufficient to say that much of human behavior can be explained by treating individual welfare as synonymous with personal benefit, thereby excluding ethical concerns, or what the economist Amartya Sen refers to as "commitment." However, when considering the range of human behavior that includes choices among public goods—such as education, humanitarian aid, local welfare programs, environmental protection, and the like—preferences based on individual ethics become relevant. Under a general view of individual welfare, it is certainly true that individuals can benefit personally from choosing to act in accordance with their ethical preferences and can be made personally worse off by acting in a manner inconsistent with such preferences. The individual acts in accordance with these ethical preferences or commitments because of the personal well-being he or she enjoys as a result of the action, or because of the loss in well-being he or she would suffer from acting otherwise.
Because individuals can articulate their preferences and anyone observing an individual's behavior can infer those preferences, one cannot dismiss preferences as the inane creation of economists; nor can they claim that preferences are hypothetical expressions of choice under ideal conditions.
Sen, in "Rational Fools: A Critique of the Behavioral Foundations of Economic Theory" (1977), cites an instructive example: in George Bernard Shaw's The Devil's Disciple, the character Richard Dudgeon, denying that his willingness to be hanged for another man was due to sympathy or love, explains to a woman that "What I did last night, I did in cold blood, caring not half so much for your husband, or for you as I do for myself. I had no motive and no interest: all I can tell you is that when it came to the point whether I would take my neck out of the noose and put another man's into it, I could not do it." While Richard's act was surely heroic and self-sacrificing, it was the result of his choice to act in a manner he preferred and for his own personal benefit.
Acting in accordance with one's ethical preferences enhances one's personal welfare, but I am willing to make an even stronger statement: one will act in a manner that violates some of one's individual ethical codes when the price of obedient behavior becomes too high. For example, I give to various charities because I have an ethical preference for doing so. However, the amount I give fluctuates with my economic circumstance. Making large charitable contributions in hard economic times means I would not be able to provide for my children, for which I also have an ethical preference. The fact that individuals are capable of and willing to trade one action for another on the basis of preferences, even when those preferences include personal ethical codes, is a hallmark of "economic" behavior that we have come to know as rational choice.
Satisfaction of preferences and individual well-being
In my view the most serious of Sagoff's philosophical criticisms is his insistence that satisfaction of preferences has nothing to do with individual well-being. Since Sagoff has no compunction about delving into economics, let me explore what I believe to be the philosophical argument underlying Sagoff's view, as articulated by Harvard philosopher Thomas M. Scanlon, who has given us one of the most accessible philosophical critiques of preference satisfaction. Scanlon's critique provides a rare clarity of insight into the differences that exist between economists and philosophers regarding the fundamental concept of individual well-being and what it is that makes a person better off or worse off. I hope to convince readers that the difference that divides philosophers and economists is really not about the satisfaction of preferences, but rather is focused on economists' rejection of paternalism.
I begin by making a concession that may weaken my credibility among some economists, but which nevertheless seems unavoidable: when one speaks about those things that make an individual better off, one is necessarily making a value judgment. Some economists will no doubt be uneasy with this concession and would rather have well-being defined in a purely objective fashion, but I believe that such a definition would simply represent fanciful thinking. Scanlon, in his paper "The Moral Basis of Interpersonal Comparisons" (1991), suggests that the value-judgment nature of a definition of well-being can be mitigated to a degree by constructing "a more concrete conception of welfare in terms of goods and conditions that are recognized as important to a good life even by people with divergent values." The idea is to formulate a list of things that give rise to well-being based on "a shared conception of the important goods and bads in life." There are potentially, of course, a large number of such lists. The issue that concerns us here is not the specific elements of the list, but how the elements are aggregated to define a measure of well-being that may be used to compare alternative social states.
Philosophers and economists are willing to define a set of elements that give rise to well-being; moreover, they wish to aggregate the elements so that social states may be compared.
Scanlon's answer to the problem of aggregating the elements of a list is to utilize an index—an index that would not "consist simply of exchangeable goods and institutional prerogatives" but could "refer as well to levels of development of personal capacities . . . or even to states of consciousness." According to Scanlon, such an index must pass a test of adequacy and a test of practicality. An index would be inadequate if it contained only a subset of those elements that were deemed important to a good life—for example, if it contained only wealth.
Economists have no quarrel with Scanlon up to this point. Like philosophers, economists are willing to define a set of elements that give rise to well-being. Their list would obviously include material goods and services as well as intangibles such as friendship, love, self-esteem, and religious and ethical views. As it turns out, this list looks very much like the list of items one might find in a generalized micro-economic utility function. Moreover, economists, like their philosophical colleagues, wish to aggregate the list's elements so that social states may be compared, and economists would agree that an arbitrarily short list could very well be deemed inadequate. Where, then, is the controversy? Where is the indictment of preference satisfaction?
Controversy begins to arise when we examine the second of Scanlon's index tests—practicality. On its face, practicality seems just like more microeconomics. Scanlon notes that the question of practicality takes account of the fact that an index of well-being will be used by legislators and others in assessing the contributions of institutions to welfare. He has described an instance in which institutional decisions (on environmental policy, for example) are made by comparing levels of individual well-being. We should all recognize this for what it is—cost-benefit analysis. And while even economists routinely reject cost-benefit analysis as a sole decision-making criterion, it seems the philosophical view of public decision making is very close indeed to that of welfare economics.
We now come to the root of the disagreement concerning practicality. For an index approach to be practical, we must be able to aggregate the elements that give rise to well-being. In the simplest case, which will suffice here, aggregation is accomplished by weighting these elements and summing them; but where do the weights come from? In welfare economics, the weights are derived from the observed choices of individuals, which economists attribute to underlying preferences. In "Morality and the Theory of Rational Behavior" (1982), the welfare economist John Harsanyi articulates the economists' view of the derivation of weights most directly in stating that the ultimate criterion for deciding what is good and what is bad for a given individual is his or her own wants and preferences. This so-called principle of autonomy does not depend on the reasons one has for particular preferences. What matters for Harsanyi and other economists is that individuals apply the weights and that the weights are permitted to be specific to each individual.
In welfare economics, the weights given to the elements that give rise to well-being are derived from the observed choices of individuals; what matters for economists is that individuals apply the weights and that the weights are permitted to be specific to each individual.
What weights does Scanlon suggest? Unfortunately, he is not forthcoming on this point. However, he appears to believe that people do not know what is good for them and that someone or some body must decide for them when he says that there are circumstances in which the satisfaction of people's manifest preferences would not serve people's true interests. Instead of an index of well-being formulated in terms of preference satisfaction, Scanlon proposes an index formulated in terms of the availability of goods and conditions deemed important for a good life. The question economists naturally ask is, who will determine the goods and conditions important for a good life and their relative weights?
Up to the point at which one must weight the constituent elements of well-being, Scanlon's approach to defining and even quantifying well-being is virtually identical to that of the economist. It would be unjust to suggest that Scanlon is recommending that weights be developed by philosopher-kings; however, it is clear that the weights we as individuals would apply are unimportant in his scheme.
If, as I assert, the defining difference between an economist's view of well-being and Scanlon's boils down to paternalism, then the debate can proceed no further on the basis of economics or philosophy, but needs to be decided by each individual. It seems the remaining question to ask is not, Do people act in their own best interest? but, Are we more willing to live with our mistakes than we are willing to give up our freedom to make mistakes?
Infeasibility of Sagoff's approach to environmental policymaking
In closing his article, Sagoff suggests a better approach to the problem of formulating environmental and natural resource policy. This approach would utilize the concepts of property rights, knee-of-the-curve thinking, and place. As bases for policymaking, two of these concepts—property rights and place—are untenable, at least for the present.
Sagoff suggests that people should be enjoined from polluting and thereby invading property rights. This is neither a new idea nor a very useful one. The problem with common property, such as the environment, is the fact that the property rights are held in common. Those who advocate a property rights approach to environmental policy want the rights assigned more specifically. This ultimately means engaging in a futile political process to redistribute the rights among individuals—a process in which rights are taken away from some and given to others.
If one rejects preference as the motive for economic choice, one must also reject market prices as a measure of cost; if prices hold no meaning, one must then reject knee-of-the-curve thinking, since the cost curve is based on price information, and then reject the whole of microeconomics and macroeconomics.
The concept of place—which Sagoff suggests is our understanding of a natural scene in terms of how the scene is related to local human culture and history—is even more problematic as a basis for policymaking. If Congress and the regulatory agencies can figure out how to define environmental policy on this basis, more power to them. I believe that the concept of place will have to be much further developed before it can be shown to be a practical concept on which to formulate environmental policies, not to mention a useful one.
Thus we are left with Sagoff's knee-of-the-curve thinking, which he suggests can be used to set priorities for reducing pollution. According to such thinking, we should implement the least expensive reductions in pollution first and stop making pollution reductions when the cost of doing so begins to rise substantially. This has some intuitive appeal; however, it means that we should ignore high-cost environmental policies even when we acknowledge that the benefits of those policies are considerably greater than their costs! More important for our present discussion is the fact that costs of environmental programs are just as dependent upon the preferences of individuals as are the benefits.
The cost of an environmental program is dependent upon the prices one must pay for the program's components—for example, prices for capital goods, labor services, and energy. Prices are determined in markets where the buyers' preferences are reflected in the amount they are willing to pay and the sellers' preferences are reflected in the amount they are willing to accept. If one rejects preference as the motive for economic choice, one must also reject market prices as a measure of cost. If prices hold no meaning, one must then reject Sagoff's knee-of-the-curve thinking, since the cost curve is based on price information. And one must then reject the whole of microeconomics and macroeconomics, which of course is based on market prices. Thus Sagoff's argument with preferences leaves us with no viable, internally consistent prescriptions—economic or otherwise—for setting priorities in environmental policymaking.
Raymond J. Kopp is director of and a senior fellow in the Quality of the Environment Division at Resources for the Future.
A version of this article appeared in print in the May 1993 issue of Resources magazine.
A version of this article appeared in print in the May 1993 issue of Resources magazine.
