A state enterprise is a large, complex economic organization owned and operated by a government rather than by a private individual or organization. Though an economic entity, it is totally encapsulated by the polity, with no separation of state and market. And very importantly, it also may transfer goods and services among suborganizations without explicit pricing of those transactions.
The problem posed by state enterprise is determining a single decision rule governing the economic behavior of multiple suborganizations controlled by a single decision maker, a rule that will lead to the realization of maximum profit for the entire group. It has been proven in the socialist calculation debate that maximizing the excess of revenue over cost from operations is the most fundamental economic problem, whether the economy is centralized or decentralized, privately or publicly owned, and whether this excess is labeled surplus, surplus value, or profit.
Although this form of economic organization dates to ancient times, this article will (1) trace briefly the development of such enterprises, and (2) trace equally briefly the development of the economic theory underlying the management of such enterprises from the Industrial Revolution in Europe to the present. Although state enterprises have developed in former European colonies in Africa, America, and Asia since decolonization, these are derived largely from metropolitan examples.
In the early seventeenth century, the Dutch, Portuguese, Spanish, English, and French developed large, state-chartered multinational trading companies that lasted until the late eighteenth century.
The best example of the potential of state enterprise in this period was in ancien régime France. Richelieu established the model for an autarchic mercantilist state with new industries, a network of canals, and international trading companies on the model of the Dutch East India Company. He formed four chartered companies for the purpose of colonization but was unable to effectuate the entire system. His successor, Mazarin, appointed as finance minister Colbert, who expanded the principle of government control of enterprises to a wider range of industries and activities, enacting a much greater part of the system. Colbert established ten more trading companies. He systematically established royal monopolies for the production of luxury goods and government regulation of all commerce and production. The heyday of this policy, called mercantilism by its critics and later by economic historians, was between 1613 and 1767. Colbert was so successful at it that it is sometimes known as Colbertism. To conceptualize this French mercantilist economy mathematically, Quesnay, a critic of the system, developed his Tableau Économique. In 1954 Joseph Schumpeter called this a planned economy.
The practice of state control of enterprises survived the French Revolution and continued through the nineteenth, twentieth, and early twenty-first centuries, the state assuming ownership of such enterprises as canals, toll roads, toll bridges, toll tunnels, railroads, energy and electric utility companies, and airlines. In the post–World War II era, the state controlled over 50 percent of new investment based upon its ownership of enterprises. To optimize the output of these enterprises, it instituted an indicative central economic plan. That is, although private investors and consumers were not obligated to follow the government’s economic plan because of the overwhelming government importance in the economy, there were strong incentives to do so.
Prussia under Bismarck in the nineteenth century showed the greatest reliance on state enterprise to finance the operation of the state. In one year in the 1870s, Prussia received 398 million francs from state enterprise, most from state railways but also from state-owned mines, factories, salt works, state forests, and state farms. The Bank of Prussia, state mint, toll roads, and toll canals also yielded substantial revenue to the state. This Bismarckian system of enterprise ownership was continued and extended by all subsequent German governments and included such important state enterprises as Lufthansa, VEBA, VIAG, Volkswagen, and Salzgitter. After World War II, the United States enforced a quasi-competitive economy on the West Germans under their control that did not require the dismantling of the state enterprises just named. In contrast, the Soviet Union forced a system of state enterprise upon East Germany, largely the former Prussia. In its centrally planned economy, state-owned enterprises produced about 97 percent of total net national income in 1985, this proportion continuing until about 1994.
The Marx-Kautsky concept that the economy was a single enterprise that should be controlled by the state was adopted by the Bolshevik Party in Russia and applied to the Soviet economy after 1917. Between 1917 and 1921, the Bolshevik government transferred 37,000 enterprises from private to state control in Russia, including all firms employing over ten workers. Among these enterprises were Aeroflot and those of the Oil Syndicate, Petroleum Syndicate, or Naphtha Syndicate, organized in 1918. In 1965, there were about 200,000 enterprises in operation in the USSR. In 1988, Aeroflot was split into smaller groups of enterprises. Azneft, Grozneft, and Embaneft, the state production trusts in the Oil Syndicate, were privatized after 1991.
The economic planning agency in the USSR, established in February 1921, was known as Gosplan, an advisory agency to the Council of Labor and Defense (STO). From 1925 until 1927, Gosplan was structured with two levels of central ministries above production enterprises. Cycles of decentralizing restructuring occurred, with peaks in 1957 under Khrushchev, in 1961–1965 under Kosygin, in 1985–1990 under Gorbachev, and in 1991–2000 under Yeltsin. Putin succeeded Yeltsin in 2000 and remained in office as of 2007. He attempted to arrest the free-market trend and reestablish some state enterprises.
Economic theories associated with the historical development of state enterprises include those of Augustin Cournot, Jules Dupuit, Léon Walras, Maurice Allais, and Gérard Debreu in France; Vilfredo Pareto and Enrico Barone in Italy; Karl Marx and Karl Kautsky in Germany; Vladimir Groman, Vladimir Popov, Wassily Leontief, and Leonid Kantorovich in the USSR; Jan Tinbergen in the Netherlands; Ragnar Frisch in Norway; Oskar Lange and Leonid Hurwicz in Poland; and Donaldson Brown, Alfred Bradley, Paul Samuelson, and Kenneth Arrow in the United States. Of these theorists, eight have been awarded the Nobel Memorial Prize in Economic Science since its establishment in 1969.
Interestingly, given the lead taken by France in the practice of mercantilist policy, there were few French academic writings on the doctrine in the sixteenth, seventeenth, and eighteenth centuries. Cournot in his 1838 Recherches introduced marginal cost pricing in government enterprises (toll roads, toll canals, and railroads, which were completely state owned by the 1870s), defining the marginal revenue and marginal cost functions to be the derivatives of the total revenue and cost functions, and prescribing a most efficient price as that attained at the equality of these two derivatives.
Dupuit in 1844 extended Cournot’s marginal analysis from production to consumer utility. Using taxation as an example, he argued that total revenue from a new public utility (state enterprise) would be greatest where marginal revenue from a change in a price (toll) was zero. He distinguished between the utility to the nation as a whole (social or state utility), and the utility of a consumer of a service (private utility). Private utility varied with the consumer, and state utility could be maximized by price discrimination, charging different prices to consumers based on their individual utility functions.
Following Marx’s 1859 lead in theory and the policy of the First International, which he cites, Walras from 1874 to 1877 developed a static model of a pure exchange economy of many sectors arrayed as an m -by- n plane matrix of linear demand and supply equations. When solved as a simultaneous equation system, this model yielded the quantity of output of all sectors used as inputs to each other sector and the prices at which these transactions took place. To begin operating the model, one was required to guess at the initial quantity of commodities held by each participant in the economy, and the prices of those commodities. The model then proceeded to final equilibrium by successive approximation.
The first explicit theoretical extension of the Marxian and Walrasian models was undertaken by quasi-socialist engineers Pareto and Barone in the 1890s. Pareto approved of the theory of class struggle and historical materialism but questioned the labor theory of value. In his 1896–1897 Cours d’economie politique, Pareto argued, using Walras’s general equilibrium model, that as an equilibrium was simply the solution to a set of equations, this solution could be calculated by a socialist planner as well as worked out in practice through a market. In 1897, Pareto said he had been able “vigorously to prove that the coefficients of production are determined by the entrepreneurs in a regime of pure competition precisely in the same way as a socialist government would have to fix them if it wanted to realize a maximum of ophelimity” (Pareto 1897, pp. 485–502). In his 1906 Manuel d’economie politique, Pareto, analyzing only production in a collectivist economy, stated that “prices and net interest on capital disappear as actual entities … but they will remain as accounting entities; without them the ministry of production would proceed blindly and would not know how to plan production.” Therefore, “[t]he phenomena we have just studied suggest, in an abstract way and without taking into account the practical difficulties, is an important argument in favor of collectivist production.” In the 1909 French revised edition of this book by Bonnet, all four sets of Walrasian mathematical equations, including the consumption set, were presented explicitly, but only in a 113-page appendix. For consumer utility, however, he substituted an Edgeworth indifference curve for a Walrasian (cardinal) utility function. Mathematically, this substitution made no difference in the equilibrium equations. However, the ordinality of the utility or individual preference equations meant that no empirical (numerical) solution could be calculated representing his maximum of ophelimity. His focus thus shifted from the Walrasian problem of proving the existence of a solution to a set of equations to maximizing satisfaction, a preliminary statement of economic optimization, the linear programming problem. This optimum was defined as the state at which no one can be made better off without making someone else worse off. The solution assumed that all equations in the system were differentiable.
French economic planning after World War II was based on the theory Allais provided in 1943. He proved that a general equilibrium was Pareto optimal and that it could be attained by a centrally planned economy. If monopoly exists, however, these results are vitiated, for Walrasian general equilibrium assumes perfect competition in all markets. From 1937 to 1943, he was administrator of Nantes state enterprises, in control of railroads for five of the eighty-nine French departments.
Debreu in 1951 published an article in which he stated: “In an economy provided with a central planning board incarnating a social welfare function there is only one consumption unit. The whole economic system can be divided into nations among which consumption units are distributed.” In 1959, objectively following Hegel and Arrow, Debreu published an axiomatic analysis of the development of a perfectly competitive economy. He utilized topology to extend the input-output analysis from the Euclidean plane to the solid, from two dimensions to a potentially infinite number of dimensions. Arrow and Debreu went on to show that a Pareto optimum could be calculated based on an analysis of ordinal preferences.
In 1908, Barone set up a general equilibrium model with four groups of linear equations similar to that of the Walras-Pareto model, and set up successive approximations as the solution process. The number of variables was found to be equal to the number of equations so that the system was formally (mathematically) solvable, that is, determinate. The market prices of the purely competitive system were found to be equivalent to the Paretian accounting exchange ratios among commodities in a collectivist system established solely by the Ministry of Production.
Unlike Walras and Pareto, Barone in his mathematical model focused solely on the production sector. His equations represented empirically observable magnitudes—quantities, costs, and exchange ratios (accounting prices) among the n kinds of capital (enterprises) in this sector. Cournot marginal cost equality in all two-by-two-enterprise transactions was required for equilibrium. This represented Walrasian classical economics and Austrian capital theory
Marx and Engels in 1848 called for the centralization in the hands of the state of the means of production, credit, transportation, and communication. In 1859, Marx referred to Petty’s 1699 statement that the entire country is “one large scale industrial establishment.” Marx here stated, “The exchange-value of this particular commodity can therefore be exhaustively expressed only by the infinite number of equations in which the use-values of all other commodities form its equivalent.” In 1863 and 1885 the first and second volumes of Das Kapital appeared, in the second of which Marx stated, “from the standpoint of society as a whole, the aggregate capital appears as the capital of a single joint-stock company.” This was all a neomercantilist analysis, a systematic updating of Colbertism in mathematical terms.
Kautsky, the leading socialist theoretician after Engels’s 1895 death, argued in the 1891 Erfurt Program, “The whole machinery of production will be turned into a gigantic concern subject to a single master [the state].” In a 1902 speech, he added that this required “systematic direction of production from a single point.”
The earliest attempt to construct an empirical table of the Quesnay sort, as suggested by Marx, was by the Menshevik V. G. Groman in 1923. STO ordered the Central Statistical Administration to construct a “balance of the national economy” for 1923–1924. Preliminary results were released in 1925. In 1926–1928, Pavel I. Popov, assisted by Lev N. Litoshenko and M. Barenholz, developed a rudimentary input-output model based upon Quesnay’s Tableau and Marx’s two-sector model of expanded reproduction. It attempted to build an “inter-branch balance of productive links.” In 1935 Leontief began research that led to the development of the input-output table and an economic theory underlying it. He recast the Walras model into an “input-output” form that facilitated actual calculation. Defining industries and sectors in terms of available statistical data series, he then constructed such a table to estimate coefficients of the U.S. economy in 1941 and 1953. In 1949, he constructed a 500-sector model of the U.S. economy, each sector modeled as a linear equation.
Kantorovich in 1939 showed that problems of economic allocation of resources can be reduced to maximizing a function subject to constraints, a linear programming approach to determining the optimum price in enterprises, a procedure foreshadowed by Pareto. He also introduced the distinction between the primal and dual in linear programming. He developed the “method of decisive multipliers,” which were interpreted as “shadow prices.” With regard to the necessary use of prices to achieve an optimum under socialism, he was objectively a follower of Lange, and through him of Pareto and Barone.
Tinbergen in 1936 constructed a twenty-four- or twenty-seven-equation input-output (econometric) model of the Dutch economy and in 1939 developed a thirty-eight-equation model of the U.S. economy for the League of Nations. These models used prices imposed by the central planner to imitate the operation of the purely competitive market, thus creating what came to be called market socialism. After World War II, he served for a decade as head of the central planning bureau of the Netherlands.
Frisch constructed theories of a price-driven socialist plan, a form of market socialism, but never actually held an official position in the planning office. To solve the input-output matrix, he developed in the early 1930s aspects of the technique later called linear programming.
In 1936 and 1937, Lange set out a socialist economy in which there were markets for consumer goods and labor services but not for capital goods. Prices for capital and nonlabor goods were not market prices but “mere indices of alternatives available, fixed for accounting purposes.” He thus adopted the Pareto-Barone focus on accounting prices. For equilibrium, both market prices and accounting prices were determined by the condition that the quantity of each commodity demanded is equal to the quantity supplied. Each plant must produce the quantity of output that minimizes average cost, and each industry must produce the quantity that can be sold to consumers at that minimum average cost of production. These two rules make prices given by the Central Planning Board appear to consumers precisely as prices given by a perfectly competitive market. The solution to the allocation problem begins with random (guessed) prices chosen by the Central Planning Board and proceeds by iteration to a final solution. In a centrally planned economy without any markets, prices would be determined in precisely the same way. The socialist planning method, in either the decentralized or centralized form, would reach an equilibrium faster than a perfectly competitive market because the Planning Board would have more information about the network of prices and quantities than any individual enterprise could possibly have.
In the 1920s, Donaldson Brown and Alfred Bradley, economists at General Motors Corporation (GMC), took on the task of developing a price policy for the recently divisionalized firm, and following the lines of Pareto and Barone, they visualized GMC as a socialist economy of the Marx, Kautsky, and Soviet kind. Each division was viewed as an independent enterprise. All transfers between divisions within the firm were therefore priced, as previously all transactions between GMC and outside firms had been priced. This allowed GMC to allocate resources within the firm more precisely and hence to function more efficiently. The Brown-Bradley program corroborated from the capitalist side the theory being developed nearly simultaneously in the Soviet Union. Clearly implicit in all these findings was that the method of optimizing returns was the same for a privately owned enterprise as it was for a collectively owned (state-owned) one. Optimization required a “visible hand,” that is, management of the enterprise, or economic planning.
Samuelson in 1941 and 1947 defined economic equilibrium to be a mathematical optimization problem, following Pareto. He introduced from physics the use of Langrangian multipliers in solving constrained optimization problems, defining these as either prices or costs. He explicitly equated this price to conceptions of Pareto, Barone, Hotelling, and Lange in the socialist planning debate. He attributed the compensation principle to Barone and not as have others to Kaldor and Hicks. He adopted Pareto’s differentiability assumption and showed that the operation of the model was equivalent to solving a system of differential equations.
Arrow in 1951 amalgamated many individual preferences into one social preference, setting the problem as one of choosing alternative “social states,” that is, income and wealth distributions, and concluding that the rational decision criterion is Pareto optimality. In 1959 Arrow introduced the concept that for exchanges between the divisions of a capitalist firm, “the same price should be charged as if it were a transaction with another firm.” Adopting von Neumann’s proof that the Lagrange multiplier was equivalent to the shadow price, he argued that only a process of successive approximations could determine the shadow price. He also redefined the concept of profit maximization as maximization of “the sum of the discounted profits” of the firm over time. This made profit maximization a dynamic rather than a static process. In 1969, he returned to the application to enterprises, stating, “An incentive for vertical integration is the replacement of buying and selling on the market by the cost of intrafirm transfers; the existence of vertical integration may suggest that the costs of operating competitive markets are not zero, as is usually assumed in our theoretical literature.”
The chief mathematical tool used by all those writing from 1838 to 1956 was calculus and matrix algebra. Samuelson in the 1940s, and Arrow, Debreu, Hurwicz, and Hirshleifer in the 1950s, experimented with more powerful methods of analysis, including dynamical systems and topology. The last four showed that it is possible to reach a solution to the general equilibrium model without the assumption of universal differentiability.
SEE ALSO Arrow, Kenneth J.; Communism; Debreu, Gerard; Enterprise; Entrepreneurship; Frisch, Ragnar; General Equilibrium; Hicks, John R.; Hurwicz, Leonid; Lagrange Multiplier; Markets; Marx, Karl; Mercantilism; Pareto, Vilfredo; Petroleum Industry; Planning; Prices; Programming, Linear and Nonlinear; Russian Economics; Samuelson, Paul A.; Socialism; State, The; Tinbergen, Jan; Union of Soviet Socialist Republics; Walras, Léon
Arrow, Kenneth. 1951. Social Choice and Individual Values. New York: Wiley; London: Chapman and Hall.
Arrow, Kenneth. 1959. A Time Series Analysis of Inter-Industry Demands. Amsterdam: North-Holland.
Arrow, Kenneth. 1969. Total Factor Productivity Growth in Individual Industries and in the Economy. Cambridge, MA: Harvard University, Center for International Affairs, Project for Quantitative Research in Economic Development.
Braudel, Fernand. 1982. The Wheels of Commerce. Vol. 2 of Civilization and Capitalism, 15th–18th Century. 3 vols. Trans. Siân Reynolds. New York: Harper and Row.
Coornaert, E. L. J. 1967. European Economic Institutions and the New World: The Chartered Companies. In The Expanding Europe in the Sixteenth and Seventeenth Centuries. Vol. 4 of The Cambridge Economic History of Europe, 2nd ed., ed. E. E. Rich and C. H. Wilson, 223–275. Cambridge, U.K.: Cambridge University Press.
Dobb, Maurice. 1966. Soviet Economic Development Since 1917. 6th ed. New York: Routledge and Kegan Paul.
Dolan, Edwin G. 1967. Structural Interdependence of the Soviet Economy before the Industrialization Drive. Soviet Studies 19 (1): 66–73.
Frisch, Ragnar. 1933. Propagation Problems and Impulse Problems in Dynamic Economics. In Economic Essays in Honor of Gustav Cassel, 171–205. London: Allen and Unwin.
Jasny, Naum. 1954. A Soviet Planner—V. G. Groman. Russian Review 13 (1): 52–58.
Jasny, Naum. 1962. The Russian Economic “Balance” and Input-Output Analysis: A Historical Comment. Soviet Studies 14 (1): 75–80.
Kantorovich, Leonid. 1939. Matematicheskie metody; organizatsii I planirovaniia proizvodstva. Leningrad: Izd. Leningradskogo Gos. Univ.
Lange, Oskar. 1938. On the Economic Theory of Socialism. Minneapolis: University of Minnesota Press.
Pareto, Vilfredo. 1897. The New Theories of Political Economy. Journal of Political Economy 5 (4): 485–502.
Rima, Ingrid H. 1972. Development of Economic Analysis. Homewood, IL. Richard D. Irwin.
Schumpeter, Joseph. 1954. History of Economic Analysis, rev. ed. New York: Oxford University Press.
Spulber, Nicholas. 1964. Foundations of Soviet Strategy for Economic Growth: Selected Soviet Essays, 1924–1930. Trans. Robert M. Hankin. Bloomington: University of Indiana Press.
Spulber, Nicholas, and Kamran Moayed Dadkhah. 1975. The Pioneering Stage in Input-Output Economics: The Soviet National Economic Balance, 1923–1924, After Fifty Years. Review of Economics and Statistics 57 (1): 27–34.