319 December 27, 1983 INDUSTRIAL POLICY SON OF CENTR AL PLANNING INTRODUCTION Proponents of industrial policy insist that their policies have nothing to do with centralized, comprehensive planning--which, they all a ree, is both economically inefficient and politically dangerous. zation of a nation's econom y under the control of a single deci- sion-making agency, proponents of industrial policy claim that their approach asks only that an llinvestment-guiding'f agency be permitted some measure of influence over an otherwise unguided capitalistic economy.

It i s hard to find an article supporting industrial policy which does not contain an explicit disclaimer that it is not a form of economic planning. Thus criticisms of centralized planning routinely are dismissed as entirely irrele- vant to the case of indust r ial policy criticism cannot be dismissed posal demonstrates that industrial policy and central planning are fundamentally indistinguishable. Indeed, today's industrial policy proposals are the son of yesterday's central planning schemes While central plan n ing requires the complete mobili Yet it is clear the Analysis of the logic of the pro One can, of course, always define Ilindustrial policy" in so innocuous a manner as to imply that every economy, by definition See Robert B. Reich An Industrial Policy of the Right The Public Interest, Fall 1983, p 17. Reich admits that some "perplexing" political problems are raised by industrial policy such as, "How can we insulate it from the predations of narrow interest groups and the vagaries of partisan politics whi le ensuring that it is democratically accountable however, that there are any economic problems with industrial policy.

While the author would agree that serious political problems plague indus trial policy, this paper will restrict itself to describing an economic problem which Reich and other proponents of industrial policy have ignored He denies 2 has one.2 In a recent issue of The Public Interest, one industrial policy advocate, Harvard economist Robert B. Reich, points out that "no tax, credit, or sub s idy program, however neutral in appearance, is neutral in competitive effect." Thus, he concludes the United States has an industrial policy always will.113 What America lacks, says Reich, is not an indus trial policy, per se, but an explicit one.4 reason government policymakers do not know how to guide economic development intelligently is that the problem exceeds the intel lectual powers of the human mind to undertake deliberately or explicitly. When Reich and other industrial policy advocates ask for a m ore Ilcoherent industrial policy115 they are aiming at the same goal as the 19th-century socialists and Marxists who wanted to bring the economy under the Ifconscious controll' of a central planning agency. Both seek to subject the market economy which is otherwise "anarchicll (as Marx put it) or "of the 'do-it yourself' variety according to Reich)--to the deliberate control of a single organization. What distinguishes moderates like Reich from radicals like Marx and Engels is not their ultimate goals but t heir means. Both groups strive for a f0.m of national economic planning; it is just the degree of comprehensiveness of this plan- ning that differentiates them It always has and I Critics of economic planning, however, argue that the main INDUSTRIAL POLIC Y AND CENTRAL PLANNING Industrial policy advocates are much less extreme in their policy formulations than most advocates of central Dlannina.

Neverbeless, the critique of centralized or comprehensive-plan ning that emerged in the classic Itsocialist calcu lation debate" of the 1930s6 is also relevant to the more modest proposals aired Ibid., p 7. Similarly, more forthright advocates of national economic planning usually try to dispel any fears by asserting that all economic systems must be "planned" in som e sense.

Ibid., pp. 7-

8. See also Amitai Etzioni America Before the Twenty-First Century (New York: McGraw Hill, 1983 pp. ,312-316.

Robert B. Reich, The Next American Frontier (New York p. 274.

Reich An Industrial Policy of the Right," p. 16 The basic argument against central planning was articulated by Ludwig von Mises in 1922, Socialism (Indianapolis, Ind Liberty Press 1981) and F A. Hayek, Individualism and Economic Order (Chicago: University of Chicago Press, 1948), pp. 119-2

08. Also see Trygve J. B. Hoff, Economic Calculation in the Socialist Society (Indianapolis: Liberty Press, 1981 Peter Murrell, "Did the Theory of Market Socialism Answer the Challenge of Ludwig von Mises? A Reinterpretation of the Socialist Controversy,"

History of Political Economy, 15:l; Karen Vaughn, "Economic Calculation Under Soc5alism: The Austrian Contribution," Economic Inquiry, Summer 1980; and Don Lavoie, Rivalry and Central Planning: the Socialist Calcu 1atio n.Debate Reconsidered (New York: Cambridge University Press, forth coming Times Books, 1983), I1 3 today for a national industrial policy. Essentially the problem with Comprehensive planning is not so much that it is comprehen sive--indeed, this is the on l y sensible way to try to control an economy-but simply that it is planning It proposes, in other words, to make a single agency responsible for deliberately con trolling a process that can only be driven effectively by the competitive interactions of thou sands of separate and often con tending plans. Industrial policy still represents an attempt to plan an economy, without doing so comprehensively. As such the proposal still fails to resolve the difficulty raised by economists over fifty years ago..

What i s worse, industrial policy represents an attempt to accomplish an end=-the rational formulation of priorities for a nation's investment decisions-without supplying the means neces sary for that end--the gathering of comprehensive knowledge about the detai l ed interconnections of the economic system, and the power to alter these details. This means that one can confidently expect the initial policy measures of the investment-guiding agency to fail and to require serious modifications. The nature of these mod i fications will depend on the policymakers' diagnosis of the failure. But since the advocates of industrial policy seem to presume that whatever degree of order an economic system attains must be due to deliberate coordinating policies of government rather than to spontaneous competitive forces it seems likely that their diagnosis would be that they lack sufficient central power and control over the economy So while industrial policy may not be a form of central planning, as such, it could well be a big ste p in that direction.

The Nature of the Planning Agency The crucial issue is not what industrial policy advocates would like to be the result of their proposals, but what economic analysis indicates would tend to result from the institutional structure impl ied in those proposals. What economic forces would an investment-guiding agency set 'in motion great measure on the intentions of those in charge of the agency.

These would shape the policymakers' reactions to the unanticipated difficulties encountered by their initial measures It would depend in Most advocates of industrial policy do not want to invest a But they do most earnestly want to give a national central government with coercive powers so immense and detailed that they would seriously obstruct th e workings of the market economy investment guiding agency enough power to decide deliberately and intelligently which industries to encourage and assist, and which to discourage and phase out How much power is enough to accomplish this task?

The Nature of the Market Process Economic analysis shows that the complex workings of a market economy are driven by the mutually influenced choices of the mil lions of human minds that compose it. The factors that weigh on 4 these choices, and the considerations that these minds find rele vant, together make up the structure of information that brings a measure of coordination to the market process debate over socialism showed, centralized planning runs into serious economic difficulties because it cannot hope to mars h al the vast amounts of detailed knowledge that inform decentralized decision-making in the market. The market represents what Nobel Laureate F. A. Havek calls a 'Itelecommunication mechanism,Il a As the classic device for communicating information through money prices- which permits those who engage in profit/loss calculations with them to effectively act on more information than any single mind or organization could ever master.

The market process may be compared with the process by which new knowledge is discovered within the scientific community both cases the effective generation of knowledge is made possible by virtue of the independent theorizing and experimentation of rivals who disagree on the best ways of doing things. The criti cism of national e c onomic planning is not fundamentally different from the case for academic freedom. Attempts by government to influence either scientific or market discovery processes are apt to sabotage them In It is the mutually coordinated use of detailed information b y individual decisionmakers that makes the market system work.

Industrial policy advocates insist that they (unlike advocates of comprehensive planning) wish to do nothing to hamper the market process or to override its detailed functioning, but only desir e to steer it in general ways--in other words, to shape it in the aggregate, not in its particulars businessman in the machine-tool industry how he should organize his factory, but only to decide explicitly whether the machine-tool industry as a whole sho uld receive more investment monies than say, the fast-food industry.

Such a plausible and seemingly reasonable retreat from com prehensive or detailed planning to noncomprehensive or aggregative planning is, however, fundamentally illusory. Once control ov er detail is abandoned so must be control as such, except in the sense that the basic legal "rules of the game," within which economic choices are made, can and should be established. But aggregative industrial planning is more akin to throwing a wrench i n to a piece of machinery than it is to controlling anything They do not aspire to tell the See F. A. Hayek, Law, Legislation and Liberty, Vol. 1, Rules and Order Chicago: University of Chicago Press, 1973) for his distinction between law and legislation. 5 WHY CENTRAL PLANNING MUST FAIL What is llPlanningll A comprehensive planning agency, if it could exist in a modern economy, should not be.confused with the more mundane and partial llplanningll undertaken by any sensible person trying to achieve certain g o als plot and ensure the consequences of the agency's own actions. If the agency were successful, unintended consequences could be dispensed with and mankind could truly become master of its own future development as most advocates of industrial policy now admit it is, what then can be said of comprehensive or aggregative planning? How does it differ from the partial planning of individual businessmen?

Surely neither can pretend to be able to anticipate the remote consequences (both in time and place) of 't he limited variables it controls solve one problem may unintentionally lead to the creation of new problems for other decisionmakers The purpose of central planning is to But if this ambitious goal proved unattainable In each case the attempt by any one d e cisionmaker to The main characteristic that distinguishes planning done by a government from planning undertaken by others is that the former can employ coercion to help achieve its purposes. Nongovernment planners have to employ persuasion, such as offer i ng something valuable in exchange, to get others to cooperate. This coercive advantage does not, however, guarantee that the goals promoted by the government planners will be accomplished. By definition noncomprehensive planning seeks to control only part of the eco nomic system, and hence those parts which it does not control are free to react in their own ways to government policies reactions and the further rippling consequences they engender can not be fully anticipated by aggregative planners, and thu s unde sirable and unplanned results may follow in the wake of their policies. Indeed, why should aggregative planners be expected to be any better informed about the remote consequences of their coercion-backed plans than those who use persuasion These Th e Impossibility of Complete Knowledge Essentially, then, the problem with aggregative planning is a direct corollary of the flaws associated with comprehensive planning. The latter is Impossible, because no single hierarch ical agency could attain a level o f intelligence to rival that which emergences socially from the competitive process. While aggregative planning, unlike comprehensive planning, is not an impossibility (indeed the world has seen little else this century the same lack of knowledge on the p a rt of any single person or organization which makes it impossible for comprehensive planning to replace the market also makes it irrational for an investment guiding agency to try merely to Irguide" the market. If the plan ning agency is necessarily less k nowledgeable than the system which it is trying to guide-and even worse, if its actions must result in further undesired consequences in the working of that 6 system--then what is going on is not deliberate planning at all but rather blind interference by some agents with the plans of others.

Since industrial policy advocates want to plan aggregates rather than detailed specifics, for this reason alone it could be argued that their coercion-backed plans will necessarily be less informed. Details are the stuff of economic decision-making.

Understanding how an economy works involves paying attention to the circumstances, meaning, and consequences of the individual human actions that make up an economic order irrelevant in that they are simply constructs of economic theorists.

They have no meaning to human actors and play no role in their choices Aggregates are One of the most eloquent critics of centralized planning Michael Polanyi, has illustrated this point by the use of an analogy w ith a chess team playing a hundred games of chess what really counts for effective decision-making in economics and chess playing is the detailed and contextual knowledge of the individuals involved, then any attempt to control an economy or a chess team a ccording to broad aggregate categories must necessarily drop that context and muddle the details If It is as if the manager of a team of chess players were go find out from each individual player what his next move was going to be and would then sum up th e result by saying The plan of my team is to advance 45 pawns by one place, move 20 bishops by an average of 3 places 15 castles by an average of four places, etc He could pretend to have a plan for his team, but actually he would be only announcing a nons e nsical summary of an aggregate of plans .8 If the chess team captain were actually to try to get his players to carry out his plan they would be led to make decisions, such as !'advance a pawn,Il irrespective of the only context in which such decisions ca n be at all meaningful that a chess team subjected to such interference would play exceed ingly bad chess I Can there be any doubt Contemporary advocates of industrial policy freely admit that the national investment guiding agency would not endeavor to ob t ain comprehensive detailed knowledge of the working details of the American economy. They seem to believe that th.is admission insu lates them from the criticisms of central planning. But if the planners cannot obtain such detailed knowledge and are force d to formulate their plans in terms of broad aggregates, they would be See Michael Polanyi, The Logic of Liberty (Chicago Press. 1951 D. 134.

University of Chicago 7 trying to mould people's actions in general directions, without knowing either the goals o f each individual or even what actions he is taking. Making such planning directives vague and general instead of detailed, in other words, does not in the least answer the challenge posed by the economists' critique of planning. If the planners do not kn ow the details, then they do not know what they need to know to justify the imposition of their choices over those they are trying to direct.

The Uselessness of Economic Aggreqates constructs of particular goods. They directly assess individual goods in pa rticular combinations and thus it is only on this par ticular level that the causal movement of economic phenomena can be adequately explained or controlled. Changes in an aggregate whether as large as llGNP1t or as small as ''electronic components and ac c essories," are purely accidental by-products of individual human actions undertaken with respect to particular goods. Thus it is no more rational for an economy to be guided by general directives such as "invest more in computer technology1' than it would be for a chess player to be guided by a directive to "advance a pawn have to be to ask which specific investment or pawn move is de sired, and why, and how it might fit into the overall strategy.

However if the director could answer those questions he wou ld not need subordinates at all. He could be the society's investor or the whole chess team himself No one in the real economy weighs the value of aggregated The appropriate responses of either decisionmaker would The Problem of Infinite Choices It is tru e , of course, that there are important differences between the conduct of Polanyb's hundred games of chess and the running of a modern, technologically advanced economy. But if the main differences are examined it would seem that they tend to strengthen ra ther than weaken the force of Polanyits argument.

For example, the possible moves available to a chess player in any particular context are finite and could be listed by any one who knows the rules decisionmaker, by contrast, are unbounded. His inability t o list all his options is due not only to their sheer number but also to the fact that complete surprises are possible infinite amount of time, some possibilities might never occur to him maker is forced all the more to rely on habitual modes of behavior e stablished by a process of evolutionary selection in a profit The possible choices of an economic Even given an Under such conditions of radical uncertainty,1 the decision lo See Ronald H. Heiner, "The Origin of Predictable Behavior American Economic Revi e w 83, 4, 1983; and Richard R. Nelson and Sidney G. Winter Evolutionary Theory of Economic Change (Cambridge, Mass versity Press 1982 Harvard Uni-and loss environment. His choices are not straightforward cal culations. They are of necessity based on tacit h unches. Such hunches do play an important role in chess as well, but it is at least conceivable that chess could be played by examining every possible contingency and computing the best move.ll This possi bility is not even remote in the case of economic decision-making.

The Interaction of Decision-Making chess team and in an economy is the fact that chess team players are only rivals of single opposing players and need not coordinate their activities with one another isolation, and thus the combined intel ligence of all the players on one team is only a function of their average intelligence.

But in the case of economies the overall intelligence of the system is greatly enhanced by the process of mutual interaction the method of adjustment of each particip ant to the signals sup plied by his fellows. In other words, whereas for the chess team captain to plan his teammates' moves he need only know all of their strategies in each of their specific games, for the economic planner to direct the economy would re q uire that he know all the individual decisionmakers' strategies and also that he know some- thing none of them know will affect one another's choices A more significant difference between decision-making in a Each game can be viewed in how their competiti ve pulling and tugging The Superiority of Free Competition It is this central problem that comprises the fundamental economic argument against all forms of national economic planning whether comprehensive like Marx's or aggregative like Reich's.

Competitio n is, as F. A. Hayek put it, a "discovery procedure.'!l2 It is a process that generates and conveys more knowledge than any of its participating contenders can possibly know. If plan ners could know in advance what they learn by permitting competi tive fo r ces to operate, they would not re.ly on them in the first place. Economic rivals frustrate each other's plans, but also provide information to guide them by virtue of their contrary tugs on prices. When one producer outbids another for some scarce factor o f production, for instance, the excluded buyer has learned to use more efficient factors No single rival, including govern ment, can know what information market rivalry will discover without "playing outll the competitive process. Thus no single l1 In fa c t modern researchers in the field of artificial intelligence have found it necessary to program computers to play chess by means of general rules of thumb" learned from the experience of playing hundreds of games rather than by the "brute force" method of trying to examine all possible moves.

See F. A. Hayek, New .Studies in Philosophy, Politics, Economics and the History of Ideas (Chicago, University of Chicago Press, 1978), pp. 179-190 l2 9 organization has sufficient knowledge either to replace (Marx or interfere with (Reich) this competit.ive process.

This competitive process can only discover new knowledge effectively so long as the contenders are free to experiment with the production methods they imagine will be successful If, how ever, some product ion methods are chosen by an agency armed with the coercive powers of government, and some aggregate categories of industries are promoted at the expense of others, this cannot help but distort the workings of the discovery process industrial policy advoc a tes have shown some awareness of the problems concerning special interest politics that are posed by an investment-guiding agency, they seem completely unaware of the economic difficulties caused by tampering with the competitive discovery process While C O NCLUSION It is true, of course, that in a sense every nation has an industrial policy The sum of the actions of a government as Reich points out, cannot be neutral so long as government has a role to play in the economy.Il The modern economy is undoubtedl y quite heavily influenced by government policies, from defense spending to welfare measures, from antitrust laws to regulatory activities. The proponents of industrial policy seem to draw from this undeniable fact that the U.S. already has some sort of in d us trial policy. The much more contentious conclusion they draw from this observation is that America needs more conscious and coherent industrial policies In light of the inherent deficiencies of central planning, however it might be argued that the U.S. should instead try to reduce current government interference with the competitive process to the absolute minimum consistent with other political goals A society cannot avoid deciding on the general Ilrules of the game that is, the system of laws and prop e rty rights within which economic competition takes place of such laws should be to permit individuals to freely contend with one another for the consumer's dollar, without special access to government largesse. Such an ltindustrial policyt1 would indeed b e unguided and implicit. Its winners and losers would emerge spontaneously from a process that no single agency could predict much less control. Contemporary proponents of industrial policy see this as a defect. It should, however, be viewed as a virtue.

I f no agent knows better than the unconscious social process of market competition how to rationally allocate resources, then it would seem best, not only politically but also economically, to leave that process alone But the central purpose Prepared for T he Heritage Foundation by Don Lavoie, Ph.D.

Center for the Study of Market George Mason University Processes