Efficiency and Externalities in an Open-Ended Universe: A Modern Austrian Perspective

Along with renewed interest in the Austrian school of economics over the last two decades, important advances have been made in applying its principles to concrete issues that typically face market economies. Several studies have appeared that analyze standard economic problems from an Austrian perspective. These include books in the areas of competition and monopoly (Armentano, 1990), monetary theory and policy (White, 1984 and Selgin, 1988), and industrial policy (Lavoie, 1985). In addition, since the late 1970’s new works have been written which both clarify and extend various aspects of Austrian theory. Included in this list are books by Kirzner (1979) in which he extends his theories on entrepreneurship and the market process; O’Driscoll and Rizzo (1985) on the implications of time and uncertainty for economics; and collections of essays (Rizzo, ed., 1979; and Kirzner, ed., 1982) that include articles by Garrison, O’Driscoll, and Rothbard--among others--on such issues as the theories of capital, monopoly, and efficiency. Collectively, these analyses have had a significant impact on the direction and terms of discussion within the economics profession. The theories of entrepreneurship and competition, the analysis of banking institutions, and the economics of socialism have all been influenced by developments in modern Austrian economics.

In spite of this renewed interest, very little has been done in the area of externalities or, more generally, to develop an Austrian theory of welfare economics. Although Mises [1966 (1949)] and Rothbard (1962) have made modest attempts at presenting a positive theory of external effects, and Rothbard (1977) and Kirzner (1963, 1973, 1988A) have presented alternative theories of welfare economics most of the literature has been a reaction to, and criticism of, more orthodox approaches to these issues.

NEOCLASSICAL EXTERNALITIES THEORY

The concept of externalities was first discussed explicitly by Alfred Marshall in his Principles of Economics (1890). Here he introduced the familiar notions of external economies and diseconomies of scale or what Jacob Viner (1932) later called pecuniary externalities. But as Mishan has pointed out “[l]ittle attention was given to this concept until Pigou’s celebrated Economics of Welfare ... where it appears as one of the chief causes of divergences between ’private net product’ and ’social net product’” (1971, p. 1). It is this analysis that gives rise to the normative conclusion that market activities which generate positive or negative effects will lead to less than optimal results in terms of prices charged and outputs produced.

The optimal solution is given by the price and output that is obtained under conditions that would exist in a perfectly competitive general equilibrium (PCGE). In this world marginal social benefit, given by price, equals marginal social cost, given by the monetary value of resource costs, in all markets. This result is shown to be Pareto optimal, i.e., no one can be made better off without making someone else worse off, and social efficiency is maximized. Externalities result in a deviation from this standard. A widely accepted definition of the term externality is given by Baumol and Oates:

... the equilibrium output of a competitive industry which generates an external diseconomy... is in excess of its optimal output. If positive, the optimal output is that at which the market price, less the social value of the marginal external diseconomy, is equal to the marginal resource cost. Conversely, if the competitive industry generates an external economy... its equilibrium output is below the optimal output obtained by equating to its marginal resource cost the market price plus the social value of the marginal external economy. (Mishan, 1971, p. 7)

In formulating policy remedies in this area policy makers and theorists attempt to induce the market to conform to the optimal results through the imposition of taxes or the provision of subsidies. For example, with respect to negative externalities an “appropriate” policy would be to impose an “excise tax [that] is equal to the value of the marginal external diseconomy at the optimal output” (Mishan, p. 7). This would force the generator of the externality to take into account all social costs associated with his production process, resulting in the optimal output being produced. Unlike excise taxes in general, such taxes, when calculated accurately, are said to correct for market failure and therefore enhance efficiency. According to Wallace Oates:

The primary function of such taxes is to make the economy function more efficiently. Through their use we have the opportunity to employ the tax system, not only to raise revenues but also to enhance the operations of the economy. (Oates, 1988, p. 254)

With respect to goods that generate positive externalities, the same general logic is used in support of “an excise subsidy” to producers. Such subsidies should be “equal to the value of the marginal external economy” (Mishan, p. 7). This subsidy would allow the producer to realize all of the benefits associated with production of the good in question, ensuring that the optimal quantity is produced. In the extreme, potential external benefits in the provision of some good can result in the good not being produced at all. This would be that case of a “pure public good” (Samuelson, 1954), where once the good is produced it is equally available to all consumers and the consumption of it by one consumer does not reduce the amount of it available to others. In this case a severe “free rider problem” is expected to develop reducing effective demand for the product to zero. The policy conclusion here extends the excise subsidy to encompass the entire market output. It is necessary for the optimal quantity of the good to be produced by, or to have its production completely subsidized by, the government.

It is important to emphasize that the term optimal here refers to the result that is obtained in a PCGE. The value laden terms “over” and “under” production derive their normative significance from this definition of optimality. Austrian economists have typically argued that the PCGE is conceptually flawed and, as such, is irrelevant and non-operational as a guide for real-world public policy. Consequently, the entire theoretical framework that supports the “under/over” production view of externalities is rejected (see Cordato 1989 and 1992).

THE AUSTRIAN CRITIQUE

There are several objections Austrian economists have typically raised to the standard neoclassical approach. These objections all center on certain tenets that are believed to be essential to economics, but are often ignored in traditional neoclassical analysis. These tenets can be categorized as follows: 1) market activity should be analyzed as a dynamic, disequilibrium process; 2) the concepts of value and utility are strictly subjective and therefore unobservable and unmeasurable (radical subjectivism); 3) knowledge of market phenomena, by both market participants and policymakers, is always imperfect.

These categories are not completely separable. In fact, full acceptance of any one of them requires an acceptance of the other two. For example, the principle that all costs and benefits are subjective and unmeasurable, implies that demand curves and cost curves cannot be known with certainty. If uncertainty exists, then errors in forecasting will be made both by consumers and entrepreneurs, implying that markets are not fully equilibrated. These major tenets have been invoked, both separately and in combination, in making some fundamental criticisms of the orthodox approach to welfare economics and its application in the area of externalities.

The most prominent criticism stems from the importance placed on the relationship between market activity and the passage of time. A standard assumption of welfare economics is that the market data, i.e. preferences, opportunity costs, relative scarcities, and technologies, is unchanging. Under such conditions the PCGE solution is also stable. But this is not the world that policy makers or market participants face. With the passage of time comes continuous changes in knowledge and therefore cost and preference functions. This view of markets has been emphasized by Ludwig Lachmann:

As soon as we permit time to elapse, we must permit knowledge to change ... The state of knowledge of society cannot be the same at two successive point of time, and time cannot elapse without demand and supply shifting. The stream of knowledge produces ever new disequilibrium situations, and entrepreneurs continually manage to find new price cost differences to exploit. (Lachmann, 1976, p. 128)

This is a perspective of the market process that is essentially open-ended (see Kirzner, 1988B), where adjustments to change are continuous--never reaching a conclusion or state of rest. In light of this, the extent to which the real world resembles a static (i.e., timeless) Pareto optimum cannot be a meaningful measure of performance for actual market processes. The implication is that even if a PCGE and therefore Pareto optimal market outcomes could be identified for a point in time (it will be argued below that this is an impossibility) they would immediately become obsolete as soon as time is allowed to pass.

By abstracting from the passage of time and by focussing on a perfectly competitive end state, the orthodox welfare standard judges market phenomena using a model of the world in which many of the most important elements of real markets are defined out of existence. Methodologically this is wholly illegitimate. O’Driscoll and Rizzo state the objection in the following manner:

The received theory of competition is comparative static, focusing on beginning and end points. Economic agents are interested in neither the beginning nor the end points, but in coping with never ending adjustments. The theory of perfect competition analyzes the state of affairs or equilibrium conditions that would exist if all competitive activity ceased. It is not an approximation but the negation of that activity.

All theories abstract from part of reality. Theorists must determine the appropriate degree of abstraction. What is essential and permanent to the phenomena ought to be part of the analysis. It would, for example, be pretty poor economic theory that abstracted from scarcity. All genuinely economic, as opposed to purely computational, problems arise, however, because of the passage of time and the concomitant changes in knowledge and the data. Economics must analyze the process of adaptation to change as surely as it analyzes scarcity. (O’Driscoll and Rizzo, p. 98, emphasis is in the original)

In the standard analysis, the welfare consequences of externalities are arrived at strictly by comparing the static equilibrium results that are obtained in the presence of externalities, ceteris paribus, with the Pareto optimum that is reached in the PCGE. But when markets are viewed as a dynamic, open-ended process this comparison becomes irrelevant for normative assessments of the real world. By implication, the public policy prescriptions that are derived from this comparison are equally irrelevant.

As is observed in Chapter 1, most Austrian economists have implicitly defined externalities as problematical only to the extent that they hinder “real world” market processes, i.e., to the extent that they prevent markets from taking account of information concerning scarcities and intrapersonally determined consumer and producer preferences (see Hayek quotation in note 8). In standard welfare economics the policy goal is to force the real world to mimic a PCGE. But this is a wholly unrealistic, and as will be argued below, conceptually meaningless welfare standard. Not only is this end state impossible to achieve in the real world, but, as will be noted below, it is empirically impossible to identify. With such identification impossible it can never be demonstrated that any particular policy is moving efficiency in a positive direction.

It should be pointed out that the issue of whether or not externalities give rise to a discrepancy between price and marginal social benefit or cost is not central to the dispute. Although, as will be discussed below, subjective value theory does imply some conceptual problems with this kind of terminology, what is being argued is that these discrepancies only have normative significance within the context of the standard equilibrium framework (Buchanan, 1969, pp. 40-41). If this framework is rejected then the normative significance of these discrepancies disappears.

A second criticism typically made by Austrian economists is derived from their strict adherence to subjective value theory. This criticism has important implications for the kind of information that is and isn’t knowable by outside observers of the market process. In order to implement Pareto optimality as a welfare criteria, an analyst must make both interpersonal and intertemporal comparisons of costs and benefits, i.e., utility. For example, in order to identify any Pareto optimal state for a point in time, which must be done in order to determine whether some existing state is sub-optimal, utility, costs, and production functions must be specifiable. With specific regard to externalities, these functions must be known in order to show that the presence of an externality, by comparison to some Pareto optimum, creates conditions that can be improved upon. In order to theoretically or empirically demonstrate this suboptimal arrangement, interpersonal and intertemporal comparisons of costs and benefits must be made.

Any alteration of existing market conditions implies the transfer of resources among individuals. In order to show that social welfare, in the orthodox sense, is improved by transferring to B the use of resources that are currently being used by A, comparisons of marginal costs and benefits must be made across the two individuals. Since change can only take place through time, this comparison must also be made intertemporally. This automatically makes the standard welfare criteria impossible to implement. Buchanan has concisely stated the subjectivist view of costs, and, by implication, benefits.

These arguments can be extended to criticize the specific policy proposals that are put forth as standard solutions to externality problems. Implementing orthodox policy in this area in such a way as to effect a Pareto optimal solution involves calculating the social costs or benefits associated with the particular externality. But this type of calculus necessitates first measuring the externality-related costs or benefits for the affected individuals and then summing those results. But if one accepts Buchanan’s description of the nature of costs (benefits), both of these steps are inherently impossible to carry out.

Buchanan applies the subjectivist approach to excise tax remedies for negative externality problems:

The subjectivist approach calls into question the entire characterization of externalities as generating a divergence between private and social cost or benefits. All costs and benefits are inherently private. This does not mean that externalities do not impose costs or benefits on others, but to refer to them as “social” is a misnomer. As with all other costs and benefits, externalities are experienced subjectively, and cannot, either conceptually or practically, be added together to arrive at a measurement of “social” costs or benefits.

A third criticism of standard welfare economics and externalities analysis stems from what is typically referred to as the “knowledge problem.” This issue was brought to light by Hayek in a series of articles on the nature of competitive market processes and rational calculation under socialism [1948 (1935A&B, 1945A&B)]. In recent years it has been applied to a number of other areas (Boettke, 1990; Lavoie, 1985; Kirzner, 1985) including the tax-subsidy remedy to externality problems (O’Driscoll and Rizzo).

As has already been pointed out, both the passage of time and the subjective nature of costs and benefits, taken by themselves, create insurmountable obstacles to gaining the necessary information for implementing standard policy. But the “knowledge problem” arises even if one were to assume away the passage of time and the measurement problems posed by subjective value. The idea behind this stems from the complexity of the information requirements, even in a static setting. As has been emphasized, the standard policy is an attempt to impose a PCGE solution on a market or system of markets that is falling short of that solution. In order to do this the analyst must be able to gain knowledge of what the equilibrium solution is. Hayek has argued that this information is, by its very nature, highly decentralized and dispersed among individual market participants. He states that:

... ‘data’ from which the economic calculus starts are never for the whole society ‘given’ to a single mind which could work out the implications and can never be so given.

The peculiar character of the problem of a rational economic order is determined precisely by the fact that the knowledge of the circumstances of which we must make use never exists in concentrated or integrated form but solely as the dispersed bits of incomplete and frequently contradictory knowledge which all the separate individuals possess. [1948 (1945A), p. 77]

The point made by Hayek is that no single mind or authority can obtain or keep current with the information embedded in actual market processes. In fact it is seen as one of the virtues of the market economy that it can continuously bring to light new information that could not possibly be gathered and disseminated in a centralized manner. With respect to externalities, in order for the standard tax or subsidy solution to work, the analyst must have better information concerning market conditions than is actually being generated in the markets themselves. If this were possible, the actual markets, and in particular the exchange relationships that comprise them, would not be needed. As O’Driscoll and Rizzo have argued:

To know whether ... [a market] process is optimal, we would need to know the very information whose discovery is the object of that process ... [I]f we could independently ascertain the information, the ... process would be superfluous. (p. 110)

It should be pointed out that this point does not go unrecognized by orthodox welfare economics. Allan Kneese, a consistent and articulate advocate of Pigouvian taxes as a remedy for negative externality problems, notes the problem succinctly:

The general equilibrium model of resource allocation which underlies formal welfare economics represents ... a general analysis of the interrelationships of markets throughout the economy. Because in principle it requires knowledge of the structure of preferences of all consumers and the technologies available to all producers, there is no realistic hope of ever being able to state the whole system empirically. (Kneese, 1977, p. 57)

Kneese’s solution, which is the standard solution, to this problem is, in reality, to ignore it--that is, to revert to partial equilibrium analysis. According to Kneese, “in actual application small parts of the general system are extracted for detailed analysis” (1977, p. 57). This ignores the fact, highlighted by the theory of second best (R. E. Lipsey and K. J. Lancaster, 1957), that any optimal tax or subsidy is only optimal within the context of a general equilibrium solution. To derive the optimal tax or subsidy from the perspective of partial equilibrium analysis, one would have to know how it fits into the pattern of resource use in a world that is fully equilibrated. Partial equilibrium analysis ultimately must collapse into general equilibrium analysis.

As noted above, even if the complex information requirements necessary to identify a Pareto optimal solution could be met for a particular point in time, the passage of time implies changes in that solution. The problem, then, is one of centralizing, and then keeping current with all of the information necessary for determining both the equilibrium solution for any single point in time and all relevant future points. Before the data could be gathered and the optimal tax or subsidy could be calculated, the information would be outdated. This problem alone would prevent any tax or subsidy that is implemented from ever being “optimal” (O’Driscoll and Rizzo, p. 140).

O’Driscoll and Rizzo have likened this whole series of issues regarding tax policy to those that were raised by Mises [1981 (1922)] and Hayek as part of their critique of socialist planning schemes. Mises and Hayek argued that resources could not be allocated efficiently in a centrally planned system for essentially the same reasons that we have been discussing. That is, the information that is necessary for efficient resource allocation cannot be known to any central authority. With respect to both socialism and externalities analysis, it is the possibility of efficient non-market resource allocation that is being called into question. As O’Driscoll and Rizzo have pointed out:

The Austrian criticisms of the standard approach to externalities are fundamental. For this reason, any Austrian alternative must entail more than simply marginal adjustments to the orthodox approach. There needs to be a reconstruction of welfare economics, in addition to drawing out its implications for the normative effects of externalities.

OUTLINE AND ORGANIZATIONAL FRAMEWORK

The orthodox analysis of externalities involves three essential levels: a theory of welfare economics, a normative interpretation of externalities in light of that theory, and a discussion of policy remedies. The theory presented here is similarly organized. The book has six chapters. Chapter 1, “The Genealogy of the Austrian Approach, “ examines the conclusions that Austrian economists have typically reached with respect to the welfare effects of externalities. The discussion here highlights the important differences that have developed between Austrian and more traditional neoclassical economists. With the notable exception of Hayek, Austrian school economists have taken the position that only external effects involving the conflicting use of property prevent a market from reaching its potential in terms of social welfare. This analysis is found most explicitly in the works of Mises (1966) and Rothbard (1962), but its roots can be traced to the writings of the early Austrians, particularly Menger, Böhm-Bawerk, and Wieser.

Austrian economists have rejected the standard neoclassical approach to welfare economics, but no single Austrian theory of welfare economics has emerged to take its place. Chapter 2 critically examines some concepts of welfare economics that have been developed by Austrian school economists. In particular, the theories introduced by Kirzner (1963, 1973, 1988A) and Rothbard (1977) are evaluated in light of their ability to provide a normative grounding for externalities analysis. In Chapter 3 an alternative theory of efficiency is built in light of problems that face these previous attempts. This alternative theory makes use of the distinction that Mises (1966) and Hayek (1976) have drawn between “economy” and “catallaxy.” Traditionally economists have analyzed markets and discussed public policy in terms of “economic” efficiency. But when markets are viewed as an open-ended disequilibrium process it becomes clear that the concept of economic efficiency is inappropriate. What we typically call an economy is more accurately described as a catallaxy. In chapter 3 we develop a theory of “catallactic efficiency” which takes into consideration this more accurate description of the market process.

Chapters 4 and 5 specifically deal with the issue of externalities from the perspective of catallactic efficiency. Chapter 4 reexamines the externality problem in light of the alternative perspective on efficiency and compares these conclusions with the more traditional Pigouvian approach. Chapter 5 further develops this contrast by applying the normative theory of externalities developed in the previous chapter to the economic analysis of tort law. In this area a clear distinction can be seen between the property rights analysis that was introduced by Coase and later developed by Posner, among others, and the property rights analysis of the Austrians.

Clearly, all of the issues surrounding the question of externalities cannot be resolved in one volume. The purpose here is to establish a framework of analysis for further research. The concluding chapter suggests some important areas where further theoretical and practical analysis needs to be done. Two important theoretical issues left unanswered relate to the role of economics in establishing property rights to unowned resources, and to the question of whether externalities theory gives an economic rationalization for the existence of government. Some practical issues that are left unresolved concern the problems of identification and large numbers that plague all economic analysis of certain kinds of pollution problems. Furthermore, it is important to ask whether there is room for a notion of “second best” within the context of Austrian analysis in resolving practical problems that might arise. This concluding chapter attempts to clarify some of the issues that need to be dealt with in order to eventually give satisfactory answers to these questions.