===== PAGE 1 ===== Hayek: Some Missing Pieces Anthony de Jasay view of society as a determinate and self-reproducing system, in the same way as Hobbes or Marx, though perhaps no one else can be said to have an (albeit simpler) theory of the social order? The question has some weight, since if it is true that Hayek is the most influential twentieth-century advo- cate of liberal government, the strength and penetration of nor- mative liberal doctrine must to no small extent depend on the persuasive force of his positive social theory: the less coherent it is, the more modern liberalism is vulnerable to erosion and invasion by incompatible elements. It is hard, probably harder than it looks at first sight, to be sure what we mean by a social order. Hayek himself mostly uses the term “order” in the context of his ideas about the spontane- ous order. However, he intends order as such to mean some- thing akin to a pattern or schema, such that by looking at a part or a phase of it, we can make good enough conjectures about the whole (1973, p. 36). It is as if, by finding a piece of a jigsaw puzzle that depicts a cloven hoof, we could tell that the puzzle, if it were all fitted together, would in all likelihood repre- sent a cow, a goat, or perhaps the devil, but certainly not a lady with her parasol. My thesis, putting it at its sharpest, is that Hayek shows us pieces of a complex jigsaw that are intriguing and inspiring, but they do not suffice to let us predict whether, if we had all the pieces, the completed puzzle would show a cow, a goat, or the devil. If I am anywhere near right, he has no complete theory of the social order to back up his liberal recommendations. They are in any case a little incongruous, since he predicts that whichever kind of order is superior will duly prevail through group selection, | as Hayek a theory of the social order, a comprehensive Anthony de Jasay is an independent scholar now living in France. The Review of Austrian Economics Vol. 9, No. 1 (1996): 107-18 ISSN 0889-3047 107 ===== PAGE 2 ===== 108 The Review of Austrian Economics Vol. 9, No. 1 hence recommending it is somewhat out of place even if we could know that it was not superior, and a fortiori if we could not. Does Hayek show us how to tell? There are, it seems to me, at least three areas of the puzzle where he has left out important pieces, and where pieces he did place do not really fit together. One is the distribution of the social product, the other is public goods, and the third is the spontaneity of the very spontaneous order that gives the whole puzzle its character. Making the Free Society Attractive One of Hayek’s most widely known normative ideas is that there can be no such thing as social, or distributive, justice. The concept is simply a category mistake: “there can be no distributive justice where no one distributes” (1978, p. 58). It is certainly doubtful whether one can defend, from arguments drawn from justice, the popular belief that certain participants in arm’s length transac- tions are responsible for the distributive shares that accrue to other participants; that rich employers should pay higher wages to poor employees, and that supermarkets should not drive small shopkeepers out of business. Hayek holds, reasonably enough, that the terms of voluntary exchanges are determined objectively, they are not matters of anybody’s good intentions, nor of what richer parties think the poorer parties ought to be getting: nobody distributes. What he calls the “market order” entails a distribution that is neither just nor unjust. It is, however, efficient. As such, it has instrumental value and can serve other valuable ends. These ends, for Hayek, cluster around the maximum chance for the randomly chosen individual to conduct his own life successfully. Distributive jus- tice does not figure among the ends the riches created by the free market order should serve. Along line of others, from John Stuart Mill to A. Mueller—-Armack and beyond, have of course taken the well-known position that the market order conforms to “economic laws” and is neither just nor unjust, but the wealth its efficiency creates can be devoted by “society” to satisfying, among other things, the requirements of distributive justice. Both Hayek’s and Mill’s position, different as they are from each other, stand in sharp contrast to the more rigorous thesis that the market order does in fact produce a distribution that is just if the distribution of ownership in a selected initial position was just, and subsequent exchanges were free from force and fraud. ===== PAGE 3 ===== Jasay: Hayek: Some Missing Pieces 109 Hayek, in his insistence that the whole issue is categorically irrelevant to the market order, leaves a blank where others put distributive justice. Its missing piece is an unintended invitation for them to fill the void. However, while disputing the very sense of the concept of distributive justice, Hayek is nevertheless concerned with distri- ‘bution. He notes that modern societies without exception do organize welfare states,—is this cultural selection at work?—and that this evolution is consistent with respect for “abstract rules of just conduct.” It is more than a historical shift in the sponta- neous order, more than an unintended result of human actions directed at other purposes: deliberately helping it along is a positive “task of the defenders of liberty” (1960, p. 259). “[T]hough a few theorists have demanded that the activities of government should be limited to the maintenance of law and order, such a stand cannot be justified by the principle of liberty” (ibid., p. 257). The welfare state is a conglomerate of many diverse elements, some of which may “make a free society more attractive” (ibid., p. 259). For one, government must provide for “the minimum of sustenance” for the helpless, and this minimum should be, not absolute, but relative and rising with the general standard of living (ibid., p. 285). For another, such provision cannot be con- fined to the deserving poor, but must be extended to all (ibid.). As a corollary, it becomes the recognized duty of the public to compel all to insure or otherwise provide against the “common hazards of life” (ibid., p. 286). It is clear enough that a guaranteed minimum income, once granted, will not for long be kept down to the level of absolute physical subsistence, but will creep upwards and take on features of a defense against “relative deprivation.” It is also clear that if people no longer have an incentive to provide against bad luck and old age, a case is created for compulsory social insurance. What is less clear is why Hayek considers, not only that these things are bound to happen, and to happen on a scale that expands with economic progress, but that it is no bad thing that they should. Their object is not to conform to any moral impera- tive, whether of compassion, fellow-feeling, let alone distributive justice, which he rejects as the product of muddled thought. Nor is it because a distribution brought about by the market order can, along the lines of Benthamite and Pigouvian utilitarian thought, be “corrected” to generate a larger sum of aggregate utility, for Hayek to his great credit never embraces the idea of interpersonal aggregation of utility. In fact, his advocacy of such ===== PAGE 4 ===== 110 The Review of Austrian Economics Vol. 9, No. 1 quintessentially redistributive measures as the guaranteed mini- mum income, is accompanied by injunctions that such measures must not have a redistributive purpose! Barring both justice and utility as their object, all he leaves us with by way of justification is that some redistribution, (if it is not intended to be redistribu- tive) “makes a free society more attractive.” Why, however, should one seek to make the free society more attractive? Is it not going to prove itself more attractive anyway in cultural selection, by its superior aptitude to prevail over less free societies (assuming that the ambiguities of what it means to “prevail” have been resolved)? Manifestly, Hayek thinks it can do with a bit of help. Embellished by the institutions of a moderate welfare state, presumably more people will opt (vote) for it, or fewer people will desert it for the totalitarian alternatives he abhors. Does this mean, however, that the free society is not the social order that prevails because its intrinsic properties make the groups that adopt it more numerous?—does it mean that the free society is a social order that prevails over others if and because its properties are adjusted to what people from time-to- time find attractive (even if it does not make the groups adopting it more prosperous and numerous)? If the latter is the case, what distinguishes Hayek's social theory, or at least the part he has made explicit, from the theory of democracy as the system where social choices are made by adding together votes for alternatives being acknowledged as superior, not by virtue of its intrinsic capacity to make its host group grow, but by virtue of attracting more votes? Enough pieces are missing from the jigsaw to permit either interpretation, though Hayek would no doubt protest quite sharply against his cultural selection being, by a piece of impu- dent distortion, equated with procedural democracy. However, his missing pieces leave room for a “conglomerate of elements” that may well be mutually inconsistent, part cow, part goat, part devil, leaving it to the spectator to call which is which. What exactly is a free society? On what grounds can Hayek predict that it will prevail? Above all, what is the point in its functioning of the unplanned interplay of individual decisions whose collective ef- fects are unintended, and what of consciously formed collective choices carried out by the agency of the state? Hayek's obiter dicta on redistribution seems to me to leave the question largely open. This void is only deepened by his treatment of public goods and the role of the state in providing them. ===== PAGE 5 ===== Jasay: Hayek: Some Missing Pieces 111 The Pivot Between Ordered Anarchy and Statism There is a measure of unself-conscious irony in Hayek’s call for a “much more clear-cut attitude towards [public goods] than clas- sical liberalism ever took” (1978, p. 144). Classical liberalism would entrust to the state the provision of only one, very special, public good, to wit, “the enforcement of general rules of just conduct” (ibid.). There is, however, a multitude of other “highly desirable” public goods that “cannot be provided by the market mechanism” because they “cannot be confined to those who are willing to pay for them” (ibid., my italics). Therefore the means for providing them are either raised by the coercive power of the state, or not raised at all. The liberal may wish that the way be left open for private enterprise to provide them if a method is discovered for it to do so (1978, p. 145), but pending such discov- eries, it is legitimate and indeed mandatory for the state (or local authorities) to tax society in order to enable it to enjoy these highly desirable things. Is Hayek’s position “much more clear-cut” than that of classi- cal liberalism? Public goods are of central importance to social and political theory. If they cannot be provided in voluntary transactions, but we must and want to have them, the state is necessary and Pareto-superior. If they can, ordered anarchy is possible, and the state usurps the space that, in its absence, would be filled by Pareto-superior, voluntary transactions. It may be held that property and contract enforcement is a necessary condition of voluntary transactions. It may be further held that such enforcement is a public good only the state can provide. This, in brief, is the classical liberal position of the minimal, protective state. It may or may not be good theory, but it is clear-cut enough. More recent theory suggests that even contract enforcement can be provided voluntarily by those who expect to benefit from respect for their contracts, and there is no evidence that organiz- ing a state for the enforcement is more efficient, less costly in terms of total transaction costs, than its decentralized, private provision. In this view, even the minimal, classical liberal state is a needless blemish upon ordered anarchy, let alone the modern liberal state subsumed by Hayek, that has a mandate to tax society for the sake of providing goods and services as long as they are both public and desirable. Hayek seems strangely unaware of the pivotal role of public goods theory between ordered anarchy and statism, and treats it ===== PAGE 6 ===== 112 The Review of Austrian Economics Vol. 9, No. 1 cursorily. Hardly realizing its consequences, he accepts the text- book division of the universe of goods and services into two exogenously determined halves, public and private. Private goods are excludable, hence can, public goods are non-excludable, hence cannot be produced in voluntary transactions, where goods are forthcoming against equivalent resources or not at all. In reality, there is no such exogenous division. Nothing is “excludable” without further ado; for nothing can be sold without the seller incurring costs to exclude from access those who would not pay the price. Exclusion cost is no more avoidable in a good destined to be sold than is the cost of production or transport. Everything is excludable at some cost that may be high or low, depending on a host of circumstances, of which the physical characteristics of the good is only one. Over the universe of goods, exclusion cost is a continuous variable. Where society draws the dividing line between public and private goods is an endo- genous decision, for social theory to define. Providing a good publicly saves exclusion cost. This advantage may be partly, wholly, or more than wholly offset by costs arising from waste- ful use of the good the consumer can have without paying for it, and from other, less direct risks. If social choice were usually “collectively rational,” goods would be provided publicly if the saving of exclusion cost outweighed the disadvantages and added costs of publicness. As it is, whether a good becomes public, or stays private, is decided by the “public” through a political process that is not set up, and is quite unlikely, to be “collectively rational” in the above sense. Certain goods become public goods because it is held that people ought not to ave to pay to have them, others because they won’. All this is well understood now, and was already understood when Hayek ex- pressed his view that the state ought to provide “highly desirable public goods.” The half-universe of public goods is in fact one we fill. It thus comes to contain innumerable goods that are desirable if and because they are public, so that their marginal cost to the indi- vidual consumer of the good is nil or imperceptible, and they amount to a “free lunch,” to something for nothing. If so, the observation that they are highly desirable is a product of circular reasoning. As long as the good remains a good, (i.e., short of saturation) every potential consumer of it will readily vote for its public provision if it is not yet so provided, and for its provision on a more generous scale if it is provided but sparingly. Where should the line be drawn? How should a liberal society count the ===== PAGE 7 ===== Jasay: Hayek: Some Missing Pieces 113 votes for more of everything, and the votes against the taxes to pay for it? Whichever way it counts them, it has relatively little chance to stay liberal. There is only a missing piece in Hayek’s theory where the principles should be that a liberal society would adopt to draw the line between public and private, to keep it there, and stay liberal in the process. Unlike classical liberalism that confines the state to the provision of a single public good, law enforcement, Hayek’s social order is less, rather than more, clear-cut: it per- mits, if not positively mandates, the state to produce any number in any quantity; the state’s place in society is consequently ad hoc, open-ended, indeterminate, and no amount of dire warnings against socialism, fatal conceit, and loss of freedom will make it more determinate. A theory of social order is incomplete if it makes no serious attempt at assessing the long-term forces that make the public sector grow or shrink. This can hardly be done without relying on a defensible theory of public goods. Hayek feels no necessity for one. Strangely, the question seems to have held no interest for him. By way of making good the missing piece, one must insert some account of the conditions under which goods will remain private, produced only for restricted access against payment in full, as opposed to the conditions that will favor their production for unrestricted access by any member of a given public, with the necessary resources being raised either by voluntary association under contract, or by involuntary taxation. The relative weight of these three alternative solutions is perhaps the decisive influ- ence on the extent to which a society is political (“politicized” is the pejorative word usually employed for it), shaping its life by collective rather than by individual choices. The question is of abiding interest to Hayek. He does not answer it, though he fervently wishes throughout his massive oeuvre that the answer should favor individual choice. Who Enforces the Enforcement? Why does it matter to Hayek, or anybody, whether an order is spontaneous or not? The attraction of spontaneity is both moral and prudential. Though it is not clear whether Hayek saw more than instrumen- tal value in it, he stressed that the elements in a spontaneous order “arrange themselves” rather than being arranged by “uni- fied direction” (1960, p. 160). When the elements are human beings, their property and their choices, nobody’s dispositions are ===== PAGE 8 ===== 114 The Review of Austrian Economics Vol. 9, No. 1 imposed on him by another’s command. Everybody chooses for himself what seems to him the best, given that everybody else chooses likewise. All choices are interdependent, and made mu- tually compatible by property rights and their voluntary ex- changes. None dominates and none is subordinated. This lends the order in question a moral laissez passer, while non-spontane- ous orders, constructed by imposing some alternative on the participants by authority or the threat of force, are morally handicapped by their coercive element. If they are to pass for legitimate, they need to show some compensating merit. Sponta- neous social orders, in other words, have a prima facie moral standing. Constructed orders must first earn it, or do without. The prudential attraction of spontaneous orders springs from the belief, strongly held by Hayek and fairly well supported by historical evidence, that since the knowledge required for suc- cessfully designing a complex order is either irretrievably dis- persed or latent or both, the constructed order runs a high risk of being inefficient if not grossly counter-productive. Game theory calls “coordination game” an interaction where, if all or most players adopt the same norm of behavior (strategy), all get a payoff that is no worse and may be better than if they adopted different norms. Compliance at least weakly dominates deviation. Hayek’s spontaneous order is at first sight a coordina- tion game: he speaks of rules that, if they are generally observed, make all members of a rule-following group “more effective,” “because they give them opportunities to act within a social order” (1978, p. 7, Hayek’s italics). The rules are randomly generated, by analogy with genetic mutation. Some are positively selected in a process of “cultural transmission . . . in which those modes of conduct prevail which lead to the formation of a more efficient order” (ibid., p. 9), because the more efficient order helps the group living within it to “prevail” over other groups. A classic and appropriately Austrian example is the use of money, a more efficient “norm” or “rule” than barter. No member of the money- using group can do better by reverting to barter once most others trade against money. Compliance dominates deviation. All would be well if Hayek confined his concept of the sponta- neous order to cases of voluntary rule-following that are coordi- nation games, i.e., where the emergence of the order depends on some members of a group adopting the same rule of behavior, but once they do, the order is self-enforcing: all members have a continuing incentive to adhere to it and can only do worse for themselves if they deviate from it. Patently, however, there are ===== PAGE 9 ===== Jasay: Hayek: Some Missing Pieces 115 important rules that do not function like this. Once they are widely followed, they generate an incentive for the individual member of the group to violate them. Perhaps the simplest “spon- taneous” order, as Hayek would call it, that operates in this perverse way is the queue. Every member of the group that has a rule of queuing rather than milling around and pushing each other, gains from every other member following the rule. How- ever, the member who jumps the queue gains more than the one who stands in it; he can abuse the decent restraint of the others. Queue-jumping dominates queueing. The same is true of the spontaneous order that is at the center of Hayek’s theory, the “market.” It will not function to the advantage of every partici- pant unless at least two key rules of conduct, respect for the property of others and performance of reciprocal promises, are widely followed. However, if they are followed by some, this ipso facto tempts others to steal, usurp, trespass, and default on contracts. These favorite deviations offer a higher “payoff” than compliance with the rules, which of course renders compliance a potentially self-destructive mode of conduct. Neither queueing, nor the market, nor many other ostensibly spontaneous orders are truly spontaneous, i.e., coordination games along the benign, self-enforcing lines of using money, speaking the same language or driving on the same side of the road. They are thinly disguised or overt prisoners’ dilemmas. Though he steers clear of game terminology, Hayek is quite aware that this is so, and that those of his putative spontaneous orders that are in effect prisoners’ dilemmas, and have deviation as their dominant strategy, need something more than the effi- ciency of their rules of conduct if they are to survive. Not being self-enforcing, they need some support from rule-enforcement. At one point, he suggests that the successful group, though it does not realize to which rule it owes its superiority, “will accept only those individuals as members who observe the rules traditionally accepted by it” (1978, p. 10). Hayek's group, then, expels robbers and cheats. It uses ostracism to punish and deter violations of its rule. Ostracism is one of the several time-honored voluntary enforcement mechanisms that have been employed, since the dawn of civilization, to ensure the survival of beneficial but fragile conventions, including adherence to the customs and laws of property and contract, where the convention itself generates an incentive to break it. Ostracism, like other defenses against violation, can thus be understood as an auxiliary convention, a satellite serving the fragile, nonself-enforcing main convention. ===== PAGE 10 ===== 116 The Review of Austrian Economics Vol. 9, No. 1 In the absence of such supporting conventions, the emergence and widening of the division of labor, trade, and capital accumulation would be incomprehensible. So would be those cultural, legal, and political institutions whose material wherewithal was produced by these developments. There is some excuse for holding that a spontaneous order that needs to be enforced is still a spontaneous order, if its enforce- ment itself is spontaneous, the norm of a voluntary convention. Its adherents follow it by voluntary choice: they “prefer” to carry out costly and often unpleasant actions to exclude, punish, and deter violators, and do not need to be threatened with exclusion, pun- ishment or other deterrents to be induced to do se. More realistically, they may not actually “prefer” to act against violators, but wish to avoid disappointing the conventional expectations of fellow mem- bers of their group who rely on their help, and on whose help they wish to be able reciprocally to rely. (Splitting hairs, I am treating the threat of a sanction and the risk of disappointed reliance on an expected benefit, as different in kind. If they are not, the distinc- tion I seek to make between spontaneous and enforced enforce- ment becomes blurred, and difficult to sustain.) It is, however, stretching spontaneity beyond the breaking point to call an order spontaneous if it depends on “enforced enforcement,” i.e., if members of the group or a subset of them punish and deter violators of the rules, not because they think it is in their reciprocal interest or simply because it is right to do so, but because they are threatened with exogenous sanctions if they fail to do it. In the latter case, with enforcement at one level depending on enforcement at the next higher level, who ulti- mately enforces enforcement? Hayek is convinced that as civilization evolved, the scale of human coexistence changed by an order of magnitude, from small to great. There was a passage from the “face to face” society of small groups to the “Great Society” of the large group. Members of the small group were related to each other by ties of many kinds, and these relations gave rise to group solidarity. Members of the large group are unrelated and anonymous. They succeed to profit from the division of labor and the economies of scale made possible in their “extended order,” not by relying on personal relations of trust, reciprocity and sympathy, but by respecting a suitable set of “abstract rules of just conduct.” Who, however, enforces the rules of just conduct? Respect for property and contract are not self-enforcing. On the contrary, they generate incentives for their own violation. Failing reliance on ===== PAGE 11 ===== Jasay: Hayek: Some Missing Pieces 117 reciprocity, there is no voluntary convention for enforcement, except for saying that in classical liberal doctrine it is the sole field where coercion is legitimate (1978, p. 109). In fact, if the Great Society works the way he believes it does, anonymously and at arm’s length, enforcement cannot be supplied spontaneously, for it is undermined by the free rider problem and perverse incentives in exactly the same way as property, contract, and other prisoners’ dilemmas. Enforcers must be coerced to enforce. Calling a spade a spade, one would say instead that enforcers must be paid to enforce, and for this to happen taxpayers must be coerced to pay taxes. There must be at the end of a regress of enforcement-enforcers, an ultimate, sovereign enforcer. There is no doubt whatever that when talking of the need to enforce rules of just conduct, it is the state that Hayek saw as the necessary, sufficient, and legitimate enforcer. Here, too, vital pieces are missing from the jigsaw. Take the market order whose unique efficiency helps the large group to prevail. It is a web of exchanges of all kinds, most of them indefinitely repeated. In some instances, the two sides of the exchange are performed simultaneously. These are in most cir- cumstances self-enforcing, and the parties to them might as well be anonymous (though usually they have names). When, however, performances are not simultaneous, executory contracts of some complexity are often required, and they are not self-enforcing. How could strangers with no name and no established reputation enter into such contracts with each other, the state’s enforcing facilities notwithstanding? Do they ever do so? Who will be prepared to perform first when facing a nameless unknown? Yet, how, if you are anonymous, can you do any business at all except by performing first, unless somebody, broker, banker, insurance underwriter, middleman, lends you his name for love or money? In the Great Society, most people may well be anonymous to most others, because they have no profitable occasions to get ac- quainted; but since they have no such occasions, it does not matter much that they are anonymous. However, few people or none can remain anonymous to the handful of others with whom they interact in making the market order go round. That handful gets selected spontaneously, and it is always a “small group.” There is no anonymous, large-group interaction because it would be too numerous to permit it. Its individual members interact in several “small groups” whose membership may be partly overlap- ping, partly different. Thus, each small group is open to other small groups and memberships are intermingled at the edges. ===== PAGE 12 ===== 118 The Review of Austrian Economics Vol. 9, No. 1 The truth of the matter about the Great Society is that few or no large groups are completely homogenous. Their membership can always be disaggregated, sorted into smaller groups by a variety of selection criteria. As the large group is always the sum of small groups, the converse goes as well: small groups can always be aggregated to form what is, from a chosen point of view, a large group. Its dimensions are in the eye of the beholder. If one chooses to see it only as a large group, as Hayek does, something must be said about the existence problem of the non- spontaneous order which its specifications entail: if Hayek won’t tell who enforced the enforcement, others will expound it with a vengeance. For the effect of leaving out pieces from the jigsaw puzzle of social theory is that the vacuum is only too naturally filled by a false conception of the state. This conception is hardly compatible with liberal principles. Indeed, it is hardly compatible with the very market order that Hayek wants to be spontaneous, and culturally selected to make groups that adopt it succeed and groups that deliberately deform it, fail. For although it does not logically exclude other alternatives, Hayek’s theory leads straight as an arrow to the facile conclusion of an indispensable state that alone upholds property and contract. They exist by the grace of society acting through the political authority. They func- tion as society chooses that they should. The massive chorus we have been hearing from the left and center, chanting that prop- erty 1s a bundle of separable privileges granted or withheld by society, and the freedom of contract is subordinate to public policy, is vindicated by the very theory that should have prevailed over such a chorus with a clearer, a more powerful voice. References Hayek, Friedrich A. von. 1960. The Constitution of Liberty. Chicago: Univer- sity of Chicago Press. —. 1973. Law, Legislation, and Liberty, Vol. 1, Rules and Order. Chicago: University of Chicago Press. 7.1978. New Studies in Philosophy, Politics, Economics and the History of Ideas. London: Routledge and Kegan Paul.