Read an Excerpt
Measuring Change in Technology, Lifestyles, and the Environment
By Faye Duchin
ISLAND PRESSCopyright © 1998 Island Press
All rights reserved.
Origins and Objectives
The term structural economics, used to denote a body of theory and empirical research, is not today familiar among economists or other social scientists. There is ample history of a structural approach in the human sciences, however, notably in linguistics, anthropology, and psychology, and development economists since the 1950s have been clearly influenced by the intellectual traditions of structuralism. A number of contemporary economists, including myself, claim to take a structural approach. There is even a professional journal, Structural Change and Economic Dynamics, that includes the work of a variety of "heterodox" economists, most of whom would identify with a structuralist tradition. In none of these cases, however, has there been a systematic effort to provide motivation for a structural approach and describe what is meant by economic structure—that is, to bring together objectives and strategies, theory and practice.
Unlike earlier challenges to neoclassical economics, structural economics provides not only a detailed and coherent story that extends an umbrella over diverse lines of inquiry but also a powerful formalism for quantitative evaluation. This approach makes use of familiar ideas and techniques as well as some new ones. The combined power of the story and the formalism is amplified by a purpose that is avowedly pragmatic—as distinguished from the typical objective of theorists, which is to reveal truths and laws that are assumed to be independent of time and place. Namely, structural economics aims to understand and help resolve the important social and environmental challenges of our time. As a by-product, it broadens the scope of conventional economics and invites collaboration with other disciplines.
Numerous attempts have been made to increase the realism of neoclassical economic theory; a celebrated example is institutional economics, which regards markets as substantially governed by rules and norms embodied in social institutions (like laws or labor unions). While many researchers are sympathetic with the critique implicit in institutional economics, when they wish to make computations, they return to the neoclassical framework, which they believe to be the only game in town. Structural economics is a practical construction rather than a critique, in that it offers a computational framework that makes use of mathematics and quantitative information in the fundamental ways that have proven useful in other areas of scientific investigation.
Structural economics is concerned with describing the state, or structure, of an economic system and with the quantitative and qualitative changes that take place in that structure with the passage of time. The structure is defined in terms of production and consumption activities, the considerations important for those social units engaged in production and consumption, and the physical input and output flows involved in those activities. The economy is treated as a system, in that it is a set of interrelated component activities. The theory about the working of this system can be called operational in the sense that it is faithfully represented by sets of mathematical equations, or models. The models consist of equations containing variables, corresponding to important activities, and constant parameters, whose values describe the relations among the variables. Like the variables, the parameters are directly measurable. The practical significance of the mathematical equations is that they permit quantitative evaluations.
The main variables in structural economic theory describe the activities carried out in industrial sectors and in different kinds of households. Each sector or household type is said to have a structure. The electric power sector, for example, procures and prepares fuels, generates electricity, and then distributes power. It has a specific mix of inputs and a particular distribution of its output to users—this is its structure. Change takes place when a new set of inputs, processes, and outputs, one out of various possible alternatives, is substituted for the old one. For example, it might change its mix of fuels or start purchasing electricity from secondary producers, like pulp and paper mills. These would be changes in structure. (This notion of structure and change in structure can be contrasted with the slow, continuous, smooth changes in individual variables that characterize neoclassical economics.)
A theory is described in words. A mathematical model translates the words into a set of mathematical equations. The theory is necessarily richer than its formal transcription, but a model is needed for numerical computations. It is ultimately a matter of judgment, based in part on the ability of the model to produce more or less compelling empirical results, whether the theory is sound and whether the model is a faithful reflection of it.
The relationships among variables in an equation can be illustrated by the example of a simplified economy in which grain, wood, and houses are produced. The equation for determining wood production might take the following, particularly simple form:
x1 = 0.2x1 + 0.5x2 + 1,000x3 + 25,000
There are three variables: x1 measures cubic feet of wood produced, x2 is the number of tons of grain, and x3 is the number of houses. The equation states that this economy has the following requirements for wood: 0.2 cubic feet for each cubic foot of wood produced, 0.5 cubic feet for each ton of grain, 1,000 cubic feet for each house, and an additional 25,000 cubic feet delivered directly to households, perhaps for their wood-burning stoves. The statement that the equation is linear is illustrated by the fact that each additional ton of grain will require an additional 0.5 cubic feet of wood: the requirement will not rise or fall depending on the level of grain production. The figures 0.2 and 0.5 are parameters; this means that they need to be provided by the analyst as part of the database and may change under alternative scenarios. For example, the wood parameter for houses will fall if houses begin to be constructed of stone or brick instead of wood. The analyst has also provided the figure 25,000. That figure would fall if households started using other heating methods, for example. The values of the variables are determined by the computation. If the parameters are changed, the variables will take on new values.
Now imagine a model with a hundred sectors instead of three, and ten categories of households instead of a single one. One could use this model to inquire into the change in the use of energy and materials that could be anticipated if specific kinds of households started purchasing homes constructed in different ways, or changed their diets.
The attributes of models depend on their mathematical formulations. They may be linear or nonlinear, deterministic or probabilistic, discrete or continuous. Neoclassical models are nonlinear, probabilistic, and continuous. Some of the models of structural economics are nonlinear (the dynamic ones), but all are deterministic and discrete.
A model is said to be deterministic if the key relationships are represented explicitly and directly (as in the simple equation above). It is probabilistic if the assumptions allow randomness in the distribution of the variables. After the deterministic aspects have been described in a structural model, an unexplained residual will undeniably remain. For example, in the wood equation given above, there is no way to describe variations in wood requirements from one house to another. Such variations exist, but their significance is small relative to the quantities that are explicitly captured in the equation.
The wood equation represents production requirements at one point in time—say, in the course of a given year. A more complicated equation (actually a system of equations) is required to represent changes in production from one year to the next. A discrete model represents distinct states and can show the changes from one state to another, changes not only in the strengths of relationships (e.g., the amount of wood to make an average house) but even in the identity of important variables (like a switch from wood to steel). A continuous model, by contrast, is less concerned with the relations among variables at a given time than with the rates of change of the same variables over time.
The Origins of Structural Economics
Structural economics makes full use of the thought, mathematical formulations, and applied work of two twentieth-century economists, both Nobel laureates: Wassily Leontief, who created input–output economics and applied it to studying technological change, and Richard Stone, who extended input–output economics into social and demographic areas. It integrates those frameworks and substantially extends them in terms of scope, mathematical formalism, and empirical content. Structural economics is also influenced by ideas that have arisen largely outside of the economics profession. It adopts the problem-oriented pragmatism, a disregard for disciplinary boundaries, and respect for imagination and the experience of nonspecialists that are characteristic of ecological economics. Finally, structural economics overlaps with another new field, industrial ecology, in the importance it accords to the use of energy and materials in production and consumption activities and to a life-cycle approach to assessing costs and benefits.
In most of these ways structural economics is readily distinguished from neoclassical economic theory and practice, the dominant paradigm for academic economics throughout the twentieth century. Neoclassical economics is highly specialized and intolerant of "amateurs"—i.e., noneconomists. Despite the nominal importance accorded to technology, it can hardly be said to incorporate the expertise of engineers in its framework, and representation of technology is mainly symbolic. Neoclassical economists' view of individuals and what is important about their lives in society appears to have little if any common ground with the concerns of sociologists and anthropologists. Their emphasis on economic laws and theorems about rational decisions is based exclusively on stylized, utilitarian considerations, which are assumed to transcend specific social settings. Neoclassical economics treats production and consumption decisions as subject to small, continuous changes and in practice relies on the indirect approach of statistical inference about the values of random variables to quantify parameters. No variables are considered relevant unless they are—or at least in principle can be—measured in money units. For the neoclassical economist, the dominant feature of an economy is the set of so-called equilibrium prices toward which it is assumed to be moving.
A powerful example of a structural model that uses bold simplification to highlight important variables and relationships is the original input–output formulation of Wassily Leontief. He developed a linear, deterministic model and accompanying database to depict production and consumption activities using a set of equations like the one described earlier. The input–output framework will be described in substantial detail in subsequent chapters.
The early development economists valued input–output economics because it represented the interdependency of the major parts of an economic system. In addition, the 1950s and 1960s saw a considerable flurry of interest among prominent economic theorists (many of whom were students of Leontief's at Harvard) in the mathematics of the dynamic input–output formulation. Nonetheless, the input–output model was subsequently stripped of academic respectability for one main reason: It fails to honor the central conviction of neoclassical economics, namely, that the most important economic effects (if not all of them) are reflected in and follow predictably from changes in prices. The unfortunate reality is that many input–output economists have actually accepted this indictment and are apologetic about what in fact is their greatest strength.
The great strength of input–output models is that they feature a physical description of production and consumption activities in terms of inputs per unit of output. Changes in the values of inputs need to be explained by a physical logic, such as a change in combustion technology (or in packaging practices) that is adopted when the prices of fuels (or landfill fees) change. Physical quantities do not in some mechanical way "equilibrate" following changes in prices. Instead, the analyst needs to provide an explicit logic about technological alternatives to link changes in prices with associated changes in physical input structures. In neoclassical models, these changes are automatically governed by a set of "elasticity" parameters. Despite their criticisms, however, neither practitioners with, say, policy responsibilities nor even applied neoclassical modelers have failed to exploit the practical usefulness of input–output economics. Virtually all large macroeconomic or general equilibrium models include an input–output portion, which serves to disaggregate results but is not treated as part of the theoretical structure.
Input–output economics makes the mainstream economist uncomfortable. When included in textbooks, it is classified sometimes as part of microeconomics and sometimes as part of macroeconomics. Since these two major subfields of neoclassical economics are considered non-overlapping, it is clear that input–output economics, and the structural economics that builds upon it, does not fit in to neoclassical economics. The time has come for establishing structural economics as a viable, independent field that provides an alternative to neoclassical economics.
Structural Economics and Development
An explicit focus on how people live brings social and demographic phenomena, and the material aspects of everyday life, into the analysis of structural change. Such a focus has been largely absent from work on sustainable development and from the main body of economic studies more generally. Structural economics provides an approach for situating economic activities in a broader environmental, technological, social, demographic, and cultural context that can support quantitative as well as qualitative analysis. The analyst can explore alternatives available to different industries and social groups along with the major structural changes to which they are exposed. The framework is applied in this book in a case study of Indonesia, which has already served as a model for similar studies in other countries.
Development poses challenges not only at the local and national levels but also at the global level. The drastic differences in the material standards of living in the rich and the poor countries create a potentially explosive social situation. The strenuous pursuit of new technologies for industrialization is likely to succeed in raising the standards of living in developing countries substantially. However, it will also put increasing pressure on social organization and on the physical environment—the fundamental long-term problem for life on earth.
Virtually all work to date by economists about reducing pressures on the environment is focused on ways of changing the money costs of making alternative decisions, for example by shifting the burden of taxation away from income and employment toward the generation of pollution. The resulting change in relative costs can stimulate the adoption of cleaner production techniques that are cost-effective under the new but not the old regime of taxes, licenses, permits, and so on. It is assumed that such techniques will be forthcoming, but no attempt is generally made to specify what they might be.
I believe that the fundamental challenge is a different one: the actual specification of less polluting techniques for generating energy and transforming materials. It is not economists but engineers and applied physical scientists who have followed this route. New fields concerned with the development of environmentally benign techniques, such as industrial ecology, are beginning to articulate viable technological alternatives. This work can make available new options for which the relative costs would presumably be lowered by the economists' incentive schemes.
Excerpted from Structural Economics by Faye Duchin. Copyright © 1998 Island Press. Excerpted by permission of ISLAND PRESS.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.