Approaches to Economic Analysis
The first writers to treat economics systematically — Adam Smith and his immediate successors — dealt with the economy as a whole. In today's terminology they were concerned with macroeconomics. Later economists, notably Alfred Marshall and his followers in the Neo-classical school, focused upon the household and the firm. They inaugurated the era of microeconomics which led to Chamberlin's theory of monopolistic competition and Mrs. Robinson's theory of imperfect competition. The Neo-classical economists and their successors analyzed the forces which result in economic equilibrium, but their approach was that of partial equilibrium, or the method of examining "one thing at a time."
During the 1930s, under the influence of John Maynard Keynes, there was a revival of interest in aggregative economics. Keynesians drew on the work of both Classical and Neo-classical schools. Like the latter, they were concerned with the forces which result in equilibrium or disequilibrium, but they returned to the Classical tradition in their emphasis on the economy as a whole. The Neo-classical economists had devoted much of their attention to the theory of value - examination of the forces which determine prices under given market conditions. The Keynesians, however, were primarily concerned with the determinants of income and employment. Their system was based on broad aggregates: total employment, total consumption, total investment, and national income. Keynesian economists showed how these variables are related to one another, and how changes in one affect the rest. They were much less interested than the Neoclassical economists in examining the effects of a change in one variable on the assumption that all others remained fixed. In this sense the Keynesians were concerned with general rather than partial equilibrium. But neither the Neo-classical economists nor the Keynesians were directly concerned with economic interdependence, with the structure of the economy and the way in which its individual sectors fit together.
There were departures from the developments of economic thought discussed in the preceding section, and some of these came quite early. In 1758, for example, Francois Quesnay published his Tableau Economique, a device which stressed the interdependence of economic activities. Quesnay's original Tableau depicted the operation of a single establishment, a farm. It showed graphically the successive "rounds" of wealth-producing activity which resulted from a given increment in output. In this sense it was a forerunner of modern multiplier analysis. Later Quesnay published a modified version of the Tableau which represented the entire economy of his day in the form of circular flows. While this is an interesting early attempt at macroeconomic analysis, the notion of interdependence is better expressed in his earlier version.1
The next link in this chain of development did not come for more than a century. In 1874, Léon Walras published his Éléments d'économie politique pure. Walras, like other economists of his time, was largely concerned with the question of price determination. Unlike his contemporaries, however, he was interested in the simultaneous determination of all prices in the economy. His model consisted of a system of equations-one for each price to be determined. Thus he made the transition from partial to general equilibrium.
Walras' interest was not limited to the general equilibrium of exchange, however; he was also interested in the general equilibrium of production. In his theory of production Walras made use of "coefficients of production." These were determined, in his view, by technology, and they measured the quantities of factors required to produce a unit of each kind of finished goods. Thus in the Walrasian system all prices are determined -those of the factors of production as well as the prices of finished goods.2
The model developed by Walras shows interdependence among the producing sectors of the economy, and the competing demands of each sector for the factors of production. His system also includes equations representing consumer income and expenditure, and it allows consumers to substitute the products of one sector for those produced by others. It also takes into account costs of production in each sector, the total demand for and supply of commodities, and the demand for and supply of factors of production.
Walras, who was a skilled mathematician, considered his system a purely theoretical model. He believed that even if the data were available to implement his model, the computational problems would be formidable if not insurmountable. This view is understandable since only rudimentary statistics were available at the time Walras wrote, and he could not, of course, foresee the development of high-speed digital computers which are now able to handle much more complex systems than the one Walras developed.
Other economists-notably Gustav Cassel of Sweden and Vilfredo Pareto of Italy -contributed to the theory of general equilibrium. But the culmination of the work started by Quesnay came in the 1930s when Professor Wassily Leontief of Harvard developed a general theory of production based on the notion of economic interdependence. An equally important contribution was made by Leontief when he gave his theory empirical content and published the first input-output table for the American economy.3
Leontief's original table showed how each sector of the economy depended upon every other sector, but it was still highly aggregated. The subsequent development of high-speed electronic computers-and of efficient computational methods-permitted a great deal of disaggregation. Large tables have since been published representing the economy in considerable detail.
Input-output or interindustry analysis is an important branch of economics today. The input-output method has spread rapidly throughout the world. Input-output tables have been prepared for at least forty national economies, and the number of regional and small-area input-output tables has grown at a rapid rate.
The input-output method is widely used as an analytical tool in highly developed economies - both those which engage in economic planning and those which rely primarily on the market mechanism for the allocation of resources and distribution of income. More recently, a number of underdeveloped nations have turned to this new and powerful technique as a guide to important policy decisions.
As indicated above, not all input-output studies are conducted at the level of the national economy. In the United States, in particular, there has been a rapid growth of small-area input-output studies. Some models deal with a single region, but others are interregional in character. Some deal with single communities; others compare a number of communities. Some are primarily concerned with a single sector, such as agriculture or mining, but others are small-scale versions of the national models.
Where does input-output analysis fit within the larger body of economics? Broadly, it is part of economic statistics. More precisely, however, it is part of econometrics-that branch of economics which is a blend of theoretical, mathematical, and statistical analysis. Most of the literature dealing with this relatively young field is couched in abstract mathematical language. Simplified expositions dealing with part of the technique- usually a basic transactions table-have become fairly common. But the student who is interested in a comprehensive introduction to the subject has to wade through rigorous mathematical formulations.
The purpose of this volume is to present a nonmathematical exposition of the input-output system using a highly simplified illustrative example. A large input-out table, or matrix, is quite complicated, and not ideally suited for classroom discussion. On the assumption that it will be easier to teach the fundamentals of this method by a simple approach, an abbreviated and simplified hypothetical input-output table has been constructed. The meaning of this table is easier to explain than that of the larger tables which have been published. The hypothetical values inserted in the table may not be realistic (since small numbers were selected to facilitate exposition), but in other respects the table is an accurate representation of an actual input-output matrix. The reader who learns to follow the directions given for this hypothetical table can easily turn to an actual table and understand its meaning without further instruction. For those who wish to go on, an introduction to the rudiments of the mathematics used in input-output analysis is given in Chapter 7; concise symbolic formulation is also included in that chapter. With this background, the student can proceed to more technical treatments such as those given by Chesney and Clark in their estimable Interindustry Economics,4 the basic works published by Leontief and his associates, and the excellent detailed description by Evans and Hoffenberg5 of how an input-output table is put together.
The following chapter deals with the transactions table - the basis of all input-output analysis – and the coefficients which are derived from this table. Later chapters cover more specialized topics, including the application of the input-output method to a variety of economic problems.
1For an excellent discussion of Quesnay's work, with illustrations, see Philip Charles Newman, The Development of Economic Thought (Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1952), pp. 34-40. An ingenious translation of Quesnay's Tableau into an input-output model is given by Almarin Phillips in "The Tableau Economique as a Simple Leontief Model," Quarterly Journal of Economics, LXIX (February 1955), 137-44, reprinted in James A. Gherity, Economic Thought, A Historical Anthology (New York: Random House, 1965), pp. 150-58.
2See "Hicks on Walras," in Henry William Spiegel (ed.), The Development of Economic Thought (New York: John Wiley & Sons, Inc., 1952), pp. 581-91.
3Leontief's basic ideas were first published in his article "Quantitative Input-Output Relations in the Economic System of the United States," The Review of Economics and Statistics, XVIII (August 1936), 105-25. These ideas were expanded in other journal articles, and in 1941 Leontief’s first book on input-output economics was published under the title The Structure of American Economy, 1919-1929. An expanded version of this book, covering the period 1919-1939, was published by Oxford University Press in 1951. The results of more recent research, including a discussion of dynamic and regional input-output models, are presented in Wassily Leontief and others, Studies in the Structure of the American Economy (New York: Oxford University Press, 1953). For a comprehensive list of other contributions through 1963, see Charlotte E. Taskier, Input-Output Bibliography, 1955-1960 (New York: United Nations, 1961), and Input-Output Bibliography, 1960-1963 (New York: United Nations, 1964).
4 Hollis B. Chenery and Paul G. Clark. Interindustry Economics (New York: John Wiley & Sons, Inc., 1959).
5W. Duane Evans and Marvin Hoffenberg. “The Interindustry Relations Study for 1947," The Review of Economics and Statistics, XXXIV (May 1952), 97-142.