by Daniel Bell
4 John Kenneth Galbraith, American Capitalism: The Concept of Countervailing Power (Boston, 1952), pp. 91,93.
5 Alfred P. Sloan, Jr., My Years with General Motors (New York, 1964), pp. 135-136.
6 This market strategy and organizational form allowed GM to come from behind to oust Ford, a genius at production techniques, from the leading position in the market. In 1921, Ford had 60 percent of the car and truck market and almost complete control of the low-price field. Chevrolet, GM’s entry in the low-price field, had only 4 percent of the market. To meet Ford head-on in price competition would have been suicidal. Sloan’s strategy was not to undercut the Ford price but to top it somewhat, seeking to skim off that layer of buyers who would be willing to go to a higher price on the assumption they were getting a better car. By successive “upgrading” of items, largely through annual model changes, GM won the larger share. In effect, GM countered “style” to “utility” and won.
7 Sloan, op. cit., p. 64.
8 In his book, Pigou gave dozens of examples of similar “disservices”: the destruction of neighborhoods by the construction of factories in residential districts; the costs to the consumers of competitive advertising; the increase in expenditures for police and prisons because of the rising sale of liquor; the overrunning of a neighbor’s land by rabbits originating from game preserves; the costs of diplomacy and armies because of the rise of foreign investments, etc.
In this country the theme was elaborated by John Maurice Clark of Columbia, who, in his Economics of Overhead Costs (1923), drew a distinction between social and market values and between social and market costs. For Clark, as Allan Gruchy points out, the business concept of cost excluded many important social costs such as communal health hazards, unemployment, and other costs associated with business fluctuations. Clark’s concern was to bring commercial efficiency closer to social efficiency, and with making the economic system account for social values, clean air, scenic beauty, etc., as well as market values.
9 I do not minimize the technical and political difficulties of establishing such a matrix. Let me provide a “homely” example of a problem which, many years ago, first brought the social cost problem to my personal attention.
In New York City when I was a boy, snow was removed from the streets by the hiring of extra trucks which would cart mounds of it away and dump it in the river. Many years later, Paul Screvane, who was the Commissioner of Sanitation, ordered his men to push the snow into the middle of the streets, where it was churned into slush by passing cars. Perhaps he did it because the costs of hiring trucks had gone up exorbitantly, or because he wanted to demonstrate an outstanding record in office so that he could run for mayor. The sanitation department showed a commendable record of economy. But (as I figured out from the records of the Industry and Commerce Association), after each snowfall the cleaning and dyeing bills in the city went up substantially, because the passing cars would splatter the clothes of the innocent bystander.
Now, which was the “rational” solution? One could say that Screvane’s method was highly irrational, because it passed the costs of snow removal onto the backs of the unfortunate pedestrians, and if the cleaning and dyeing bills were higher than the cost of hiring additional trucks, it was truly a misallocation of resources. Yet one could also say that if the trucks had been hired, direct city expenses would have been increased and taxes would have to go up, increasing the resentments of the taxpayers in the city, so that the system of Russian roulette” whereby a random group of bystanders bore the costs might have a greater “political” rationality than economic cost-benefit analysis.
10 One can say, theoretically, that the price system could manage the problem, e.g. when the costs of individual congestion became high it would then become profitable for alternative modes of transportation to compete with the private car. But the price system, in this instance, relies on trial and error to assess the result. The difficulty is that such assessments, after the fact, are likely to be futile—an enormous amount of resources would have been misallocated, and a preemptive “system” of transportation will have been established. Under such a system, clogged highways will eventually result in the building of more highways.
11 For an elaboration of this point, see Harold Gilliam, “The Fallacy of Single-Purpose Planning,” in the Daedalus issue on America’s Changing Environment (Fall 1967).
12 Jack Burnham, Beyond Modern Sculpture (New York, 1968). If one asks what a sculptor is doing in discussing the automobile system, his argument is cast in the context of the disappearance of “objects” in contemporary society and its replacement by “systems.”
13 The idea of technology assessment grew largely out of the studies of the House Science and Astronautics Committee under the leadership of Congressman Daddario. Two panels, one in the National Academy of Sciences and one in the National Academy of Engineering, were set up to test the feasibility of the idea. The National Academy of Sciences Panel, under the direction of Harvey Brooks, agreed that assessment was possible and proposed a number of ways in which the process could be implemented in government. The Engineering panel undertook three studies—of subsonic aircraft noise, of computer-assisted instruction, and of multiphasic health screening—to further this idea. Both reports on technology assesment were published by the House Committee in July 1969.
14 From “The Social Responsibility of Business Is to Increase Its Profits,” in The Sunday Times Magazine (September 13, 1970). The argument is elaborated in Friedman’s book, Capitalism and Freedom.
15 John Davenport, “Bank of America Is Not for Burning,” Fortune (January 1971).
16 There are about 31 million shareholders in the United States most of whom have only a small holding in the enterprise. The New York Stock Exchange survey of shareownership (1970) showed that of 30,520,000 shareholders surveyed (out of a total of 30,850,000) about 12,500,000 had portfolios worth less than $5,000, and 6,400,000 had between $5,000 and $10,000. Thus a total of 18,900,000 shareholders, or 62 percent, had portfolios of less than $10,000.
Institutional investors generally now hold an increasing proportion of the outstanding equity securities of major American corporations. As of the end of 1970, the New York Stock Exchange estimated that $161.9 billion or 25.4 percent of all equity securities of companies listed on the Exchange were held by institutional holders. If one includes unregistered mutual funds, investment partnerships, non-bank trusts and foreign institutions, the Exchange estimated that the total of all institutional holdings would exceed 40 percent. (I am indebted to Professor Philip Blumberg of the Boston University Law School for the data.)
17 The growth of institutional holdings by political entities, such as municipal pension funds, colleges, foundations, churches, and other groups which are subject to direct political pressure may itself create a small force, in certain circumstances, for changes in corporate policy. Thus in the General Motors campaign initiated by Ralph Nader, Mayor Lindsay of New York instructed the trustees of the New York City pension funds to vote their 162,000 General Motors shares in favor of the Campaign GM proposals, as did Mayor White of Boston, Mayor Alioto of San Francisco, and the Wisconsin and Iowa state retirement funds.
At the same time, corporations have moved to widen representation on their boards. As of 1970, blacks had been elected to the Boards of Directors of such leading corporations as:
Chase Manhattan Bank Columbia Broadcasting
Commonwealth Edison Equitable Life Assurance
First National City Bank General Motors
Girard Trust Bank Great Atlantic & Pacific Tea
International Business Machines Michigan Consolidated Gas
Pan American Airways Prudential Life Insurance
Standard Oil of Ohio Westinghouse Broadcasting
W. T. Grant
18 This essay has dealt with the business corporation in the American context, but in the Soviet Union many of the same problems occur in the relation of the bureaucratic state enterprise
to the society as a whole. Under the Soviet planning system, each enterprise is responsible for meeting the production and profit goals of the central plan. Where the enterprise “overfulfills” the plan it is allowed to retain a portion of the profits for its social investment fund, which is used to build housing for its workers, expand clubhouse facilities, and the like. Thus there is an incentive to “economizing,” since the enterprise does not want to absorb the social costs that it generates. The large Soviet paper plants at the edge of Lake Baikal, for example, dangerously polluted the lake but strenuously resisted the idea of “internalizing” those additional costs. Insofar as the Soviet Union is committed so singlemindedly to the idea of “economic growth” and the “economizing” mode in the way I have used it, one can say that the Soviet system is actually state capitalism in which the maximization of production of each enterprise is the primary goal of the society. Yet, inevitably, in a complex society no enterprise can run its snow in a single-purpose fashion, and protests do a rise; and the state must also confront the problem of how to allocate the social costs. For a discussion of this problem in the Soviet Union, see Marshall Goldman, The Spoils of Progress: Environmental Pollution in the Soviet Union (Cambridge, Mass., 1972).
1 See “Percept and Concept,” in William James, Some Problems of Philosophy (New York, 1916), p. 51.
2 In this and in some of the succeeding sections I have drawn on memoranda I prepared for the National Commission on Technology, Automation and Economic Progress and for the Commission on the Year 2000 of the American Academy of Arts and Sciences.
3 Adam Smith, The Wealth of Nations (New York, 1937), p. 651. The phrase “the great society” appears in three places in The Wealth of Nations by my count (pp. 651, 681, and 747), but its meaning is to be found at the conclusion of book V, chap. 1—which deals with the revenues of the sovereign or Commonwealth— and, in context, the phrase “great society” means here the “whole society” (see p.767). It is a point of considerable relevance for the discussion above. The capitalized phrase “The Great Society” occurs as the title of the book by Graham Wallas. That book (published in 1914) grew out of a course that Wallas gave at Harvard in 1910, and though the initial theme is the growing interdependence of peoples, and an ensuing change of social scale, the book itself is not an effort to assess the sources or consequences of this change, but an effort to use the findings of the newer social psychology for the rational pursuit of social affairs.
4The Wealth of Nations, Book IV, chap. 11, p. 423.
5 In actual fact, it was not the existence of free exchange hut the existence of technology, with its promise of rising productivity, that created the possibility of economic life remaining a non-zero-sum gain. For an instructive book on the means whereby technology has been the chief means of promoting social equality, see jean Fourasrie, The Causes of Wealth (Glencoe, III., 1060).
6 For one of the first discussions of this problem, see Howard Bowen, “The Interpretation of Voting in the Allocation of Economic Resources,” Quarterly Journal of Economics, vol. LVII1 (November 1943), pp. 27-48.
7 John von Neumann and Oskar Morganstern (Princeton, N.J., 1953). For a simple mathematical proof of the possibility of scaling utilities, see Jacob Marschak, “Scaling of Utilities and Probabilities,” in Game Theory and Related Approaches to Social Behavior, ed. Martin Shubik (New York: Wiley, 1964). For a more general discussion of utility theory, and decision-making under certainty, risk, and uncertainty, see Duncan Luce and Howard Raiffa, Games and Decisions (New York, 1958), chap. 2.
8 A revised edition appeared in 1963 (New York).
9 The most comprehensive effort to deal with the problem is that of Duncan Black, The Theory of Committees and Elections (Cambridge, Eng., 1958). Further discussion can be found in James M. Buchanan and Gordon Tullock, The Calculus of Consent (Ann Arbor, Mich., 1962). Some earlier discussions are in Robert A. Dahl and Charles E. Lindblom, Politics, Economics and Welfare (New York, 1953), and Anthony Downs, An Economic Theory of Democracy (New York, 1957). For a symposium on Arrow’s “impossibility theorem,” see Sidney Hook, ed.. Human Values and Economic Policy (New York, 1967).
10 See Arrow, op. cit. In an appendix to a new edition of his book (New Haven, 1963) Arrow has sought to counter some discoveries of errors in his proofs by reformulating the conditions to show that the inconsistencies in the conditions still remain and that no logical foundations for a complete social welfare function are possible.
11 But theory does have a way sometimes of confounding practical men. William H. Riker has illustrated the relevance of hidden voting paradoxes through an analysis of the rules for amending bills in the committees of the House of Representatives. He has shown that under a number of rules, amendments might be adopted which are not favored by a majority—without this fact ever being known! Thus, for example, when a paragraph of an amendment and an amendment to an amendment are before a committee, the first vote is on the amendments. If the amendment to a paragraph is passed, it replaces the original paragraph without further vote. Because it votes only twice on three issues, the House does not discover any intransi-tivity, and a choice may become law which was favored only by a minority. Riker concludes that various situarions of this sort may be expected to occur, in practice, slightly more than 10 percent of the time—which should make some practical men uneasy.
This is not to say that “majority voting” is impossible. If one returns to the original example of the cyclical majority it will be seen that the intransitivity of preferences is evident only on a third vote. Thus, to uncover intransitivities, one has to initiate a complete “round robin” matching each proposal with every other, but as Riker notes no legislatures require such matching procedure for groups of more than three proposals. Once the intransitivities are uncovered, one can, as Duncan Black has demonstrated, proceed to a system of “exhaustive voting,” which matches candidate against candidate, or issue against issue, to obtain the majority decision in each instance. And, in a more theoretical vein, as Arrow and Black have demonstrated, by ranking preferences on a graph, one can obtain “single-peaked” profiles of each voter; and when all the profiles for a given group are represented as a single-peaked preference curves on one set of axes, a single majority choice can be obtained—if the group is odd in number, or if the chairman votes in case of a tie.
But practical men can take heart from the fact that, in practice, where such disparities of preferences or intensities do occur, some form of log-rolling or bargaining takes place, and, as Tullock and Buchanan have shown, one can fit such procedures into the theoretical resolutions of the voting paradoxes. This still leaves open, again it should be noted, the larger theoretical question of finding a rational solution to a social welfare function. Majority voting is preferred because it satisfies three or possibly four of the five conditions which Arrow has established as necessary for validation, and therefore is the most sensible of “compromises.” On some occasions, however. fearing a “tyranny of the majority,” one might want to consider some of the more complicated means which have been suggested by Black and Arrow to arrive at a social welfare function. And, finally, there is the point that any solution—at least those suggested so far—does not fulfill the complete logical conditions of arriving at a true social choice.
For further discussion see William H. Riker, The Theory of Political Coalitions (New Haven, 1962).
12 Robert A. Dahl, A Preface to Democratic Theory (Chicago, 1956), p. 150.
13 Nag that I am, I would still like, at the conclusion of this section, to return to the justification of the theoretical problem; for even a pragmatist like myself has to recognize the necessity for some rational system (if only as yardstick), of what social choices could be. One can carry the argument forward, perhaps, with this consideration: one might, today, with mathematical models and high-speed computers, write a single economic plan for a country that would show, through input-output matrices, the optimal distribution of economic resources, with products
valued at full economic costs. But the administrative difficulties in implementing such a plan might be so enormous that, in practice, one would have to resort to the market or some quasibargaining system for the actual operation of the economy. Yet the value of such a theoretical construction is that it could serve as a yardstick, establishing “shadow allocations” and “shadow prices” for the system and allowing us to intervene at those points where discrepancies show up. In an analogous sense, one would want a theoretical social choice model in order to have an optimal social welfare function.
14 John Stuart Mill, Utilitarianism, Liberty and Representative Government (New York, 1936), pp. 209, 228, 240, 261. The theory of representation and interests, it should be noted, is normative. It is part of Mill’s concern to define the “best form of government.” Practice, of course, might be entirely different. As Mill notes. “Politically speaking, a great part of all power consists in will ... Opinion is itself one of the greatest active social forces. One person with a belief is a social power equal to ninety-nine who have only interests.” Ibid., p. 183.
15 V. O. Key, Politics, Parties and Pressure Groups (New York, 1942), pp. 23-24.
16 Ibid., pp. 10—11.
17 The “group theory of politics,” it should be noted, has been challenged on theoretical grounds by Mancur Olson, Jr., who, applying an “economic analysis” to the nature of aggregate choice, argues that interest groups do not represent best the interests of their members. See Mancur Olson, Jr., The Logic of Collective Action:Public Goods and the Theory of Groups (Cambridge, Mass., 1965).
18Op. cit., p. 8.
19 See Charles J. Hitch and Roland N. McKean, The Economics of Defense in the Nuclear Age (Cambridge, Mass., 1960), and Analysis for Military Decisions: The Rand Lectures on Systems Analysis, ed. Edward S. Quade (Chicago, 1964).
