Mises Daily

F.A. Hayek and the Concept of Coercion

[This article is taken from chapter 28 of The Ethics of Liberty.1 Listen to this chapter in MP3, read by Jeff Riggenbach. You can listen and download the entire book here for podcast and download.]

In his monumental work The Constitution of Liberty, F.A. Hayek attempts to establish a systematic political philosophy on behalf of individual liberty.2 He begins very well, by defining freedom as the absence of coercion, thus upholding “negative liberty” more cogently than does Isaiah Berlin.

Unfortunately, the fundamental and grievous flaw in Hayek’s system appears when he proceeds to define “coercion.” For instead of defining coercion as is done in the present volume, as the invasive use of physical violence or the threat thereof against someone else’s person or (just) property, Hayek defines coercion far more fuzzily and inchoately: e.g., as “control of the environment or circumstances of a person by another (so) that, in order to avoid greater evil, he is forced to act not according to a coherent plan of his own but to serve the ends of another”; and again: “Coercion occurs when one man’s actions are made to serve another man’s will, not for his own but for the other’s purpose.”3

For Hayek, “coercion” of course includes the aggressive use of physical violence, but the term unfortunately also includes peaceful and non-aggressive actions as well. Thus, Hayek states that “the threat of force or violence is the most important form of coercion. But they are not synonymous with coercion, for the threat of physical force is not the only way in which coercion can be exercised.”4

What, then, are the other, nonviolent “ways” in which Hayek believes coercion can be exercised? One is such purely voluntary ways of interacting as “a morose husband” or “a nagging wife,” who can make someone else’s “life intolerable unless their every mood is obeyed.”

Here Hayek concedes that it would be absurd to advocate legal outlawry of sulkiness or nagging; but he does so on the faulty grounds that such outlawry would involve “even greater coercion.” But “coercion” is not really an additive quantity; how can we quantitatively compare different “degrees” of coercion, especially when they involve comparisons among different people? Is there no fundamental qualitative difference, a difference in kind, between a nagging wife and using the apparatus of physical violence to outlaw or restrict such nagging?

It seems clear that the fundamental problem is Hayek’s use of “ coercion” as a portmanteau term to include, not only physical violence but also voluntary, nonviolent, and non-invasive actions such as nagging. The point, of course, is that the wife or husband is free to leave the offending partner, and that staying together is a voluntary choice on his or her part. Nagging might be morally or aesthetically unfortunate, but it is scarcely “coercive” in any sense similar to the use of physical violence.

Only confusion can be caused by lumping the two types of action together.

But not only confusion but also self-contradiction, for Hayek includes in the concept of “coercion” not only invasive physical violence, i.e., a compulsory action or exchange, but also certain forms of peaceful, voluntary refusal to make exchanges. Surely, the freedom to make an exchange necessarily implies the equivalent freedom not to make an exchange. Yet, Hayek dubs certain forms of peaceful refusal to make an exchange as “coercive,” thus lumping them together with compulsory exchanges.

Specifically, Hayek states that

there are, undeniably, occasions when the condition of employment creates opportunity for true coercion. In periods of acute unemployment the threat of dismissal may be used to enforce actions other than those originally contracted for. And in conditions such as those in a mining town the manager may well exercise an entirely arbitrary and capricious tyranny over a man to whom he has taken a dislike.5

Yet, “dismissal” is simply a refusal by the capital-owning employer to make any further exchanges with one or more people. An employer may refuse to make such exchanges for many reasons, and there are none but subjective criteria to enable Hayek to use the term “arbitrary.” Why is one reason any more “arbitrary” than another? If Hayek means to imply that any reasons other than maximizing monetary profit are “arbitrary” then he ignores the Austrian School insight that people, even in business, act to maximize their “psychic” rather than monetary profit, and that such psychic profit may include all sorts of values, none of which is more or less arbitrary than another.

“For Hayek, ‘coercion’ of course includes the aggressive use of physical violence, but the term unfortunately also includes peaceful and non-aggressive actions as well.”

Furthermore, Hayek here seems to be implying that employees have some sort of “right” to continuing employment, a “right” which is in overt contradiction to the property rights of employers to their own money. Hayek concedes that dismissal is ordinarily not “coercive”; why then, in conditions of “acute unemployment” (surely in any case, not of the employer’s making), or of the mining town? Again, miners have moved voluntarily to the mining town and are free to leave whenever they like.

Hayek commits a similar error when he deals with the refusal to exchange made by a “monopolist” (the single owner of a resource). He admits that “if … I would very much like to be painted by a famous artist and if he refused to paint me for less than a very high fee [or at all?], it would clearly be absurd to say that I am coerced.”

Yet he does apply the concept of coercion to a case where a monopolist owns water in an oasis. Suppose, he says, that people had “settled there on the assumption that water would always be available at a reasonable price,” that then other water sources had dried up, and that people then “had no choice but to do whatever the owner of the spring demanded of them if they were to survive: here would be a clear case of coercion,”6 since the good or service in question is “crucial to [their] existence.”

Yet, since the owner of the spring did not aggressively poison the competing springs, the owner is scarcely being “coercive”; in fact, he is supplying a vital service, and should have the right either to refuse a sale or to charge whatever the customers will pay. The situation may well be unfortunate for the customers, as are many situations in life, but the supplier of a particularly scarce and vital service is hardly being “coercive” by either refusing to sell or by setting a price that the buyers are willing to pay. Both actions are within his rights as a free man and as a just property owner. The owner of the oasis is responsible only for the existence of his own actions and his own property; he is not accountable for the existence of the desert or for the fact that the other springs have dried up.7

Let us postulate another situation. Suppose that there is only one physician in a community, and an epidemic breaks out; only he can save the lives of numerous fellow-citizens—an action surely crucial to their existence. Is he “coercing” them if (a) he refuses to do anything, or leaves town; or (b) if he charges a very high price for his curative services? Certainly not. There is, for one thing, nothing wrong with a man charging the value of his services to his customers, i.e., what they are willing to pay. He further has every right to refuse to do anything. While he may perhaps be criticized morally or aesthetically, as a self-owner of his own body he has every right to refuse to cure or to do so at a high price; to say that he is being “coercive” is furthermore to imply that it is proper and not coercive for his customers or their agents to force the physician to treat them: in short, to justify his enslavement. But surely enslavement, compulsory labor, must be considered “coercive” in any sensible meaning of the term.

All this highlights the gravely self-contradictory nature of including a forced activity or exchange in the same rubric of “coercion” with someone’s peaceful refusal to make an exchange.

As I have written elsewhere:

A well-known type of “private coercion” is the vague but ominous sounding “economic power.” A favorite illustration of the wielding of such “power” is the case of a worker fired from his job….

Let us look at this situation closely. What exactly has the employer done? He has refused to continue to make a certain exchange which the worker preferred to continue making. Specifically, A, the employer, refuses to sell a certain sum of money in exchange for the purchase of B’s labor services. B would like to make a certain exchange; A would not. The same principle may apply to all the exchanges throughout the length and breadth of the economy….

“Economic power,” then, is simply the right under freedom to refuse to make an exchange. Every man has this power. Every man has the same right to make a proferred exchange.

Now, it should become evident that the “middle-of-the-road” statist, who concedes the evil of violence but adds that the violence of government is sometimes necessary to counteract the “private coercion of economic power” is caught in an impossible contradiction. A refuses to make an exchange with B. What are we to say, or what is the government to do, if B brandishes a gun and orders A to make the exchange? This is the crucial question. There are only two positions we may take on the matter: either that B is committing violence and should be stopped at once, or that B is perfectly justified in taking this step because he is simply “counteracting the subtle coercion” of economic power wielded by A. Either the defense agency must rush to the defense of A, or it deliberately refuses to do so, perhaps aiding B (or doing B’s work for him). There is no middle ground!

B is committing violence; there is no question about that. In the terms of both doctrines (the libertarian and the “economic power” arguments), this violence is either invasive and therefore unjust, or defensive and therefore just. If we adopt the “economic-power” argument, we must choose the latter position; if we reject it, we must adopt the former. If we choose the “economic-power” concept, we must employ violence to combat any refusal of exchange; if we reject it, we employ violence to prevent any violent imposition of exchange. There is no way to escape this either-or choice. The “middle-of-the-road” statist cannot logically say that there are “many forms” of unjustified coercion. He must choose one or the other and take his stand accordingly. Either he must say that there is only one form of illegal coercion—overt physical violence—or he must say that there is only one form of illegal coercion—refusal to exchange.8

And outlawing the refusal to work is, of course, a society of general slavery. Let us consider another example that Hayek quickly dismisses as noncoercive: “If a hostess will invite me to her parties only if I conform to certain standards of conduct and dress … this is certainly not coercion.”9 Yet, as Professor Hamowy has shown, this case may well be considered “coercion” on Hayek’s own criteria. For,

it might be that I am a very socially conscious person and that my not attending this party would greatly endanger my social standing. Further, my dinner jacket is at the cleaners and will not be ready for a week … yet the party is tomorrow. Under these conditions could it be said that my host’s action in demanding my wearing formal attire as the price of access to his home is, in fact, a coercive one, inasmuch as it clearly threatens the preservation of one of the things I most value, my social prestige?

Furthermore, Hamowy points out that if the host should demand, as a price of invitation to the party, “that I wash all the silver and china used at the party,” Hayek would even more clearly have to call such a voluntary contract “coercive” on his own criteria.10

In attempting to rebut Hamowy’s trenchant critique, Hayek later added that “to constitute coercion it is also necessary that the action of the coercer should put the coerced in a position which he regards as worse than that in which he would have been without that action.”11 But, as Hamowy points out in reply, this does not salvage Hayek’s inconsistent refusal to adopt the patent absurdity of calling a conditional invitation to a party “coercive.” For,

the case just described seems to meet this condition as well; for while it is true that, in a sense, my would-be host has widened my range of alternatives by the invitation, the whole situation (which must include my inability to acquire formal attire and my consequent frustration) is worse from my point of view than the situation which had obtained before the invitation, certainly worse than had existed before my would-be host had decided to have a party at that particular time.12

Thus, Hayek, and the rest of us, are duty-bound to do one of two things: either to confine the concept of “coercion” strictly to the invasion of another’s person or property by the use or threat of physical violence; or to scrap the term “coercion” altogether, and simply define “freedom” not as the “absence of coercion” but as the “absence of aggressive physical violence or the threat thereof.”

Hayek indeed concedes that “coercion can be so defined as to make it an all-pervasive and unavoidable phenomenon.”13 Unfortunately, his middle-of-the-road failure to confine coercion strictly to violence pervasively flaws his entire system of political philosophy. He cannot salvage that system by attempting to distinguish, merely quantitatively between “mild” and “more severe” forms of coercion.

Another fundamental fallacy of Hayek’s system is not only his defining coercion beyond the sphere of physical violence, but also in failing to distinguish between “aggressive” and “defensive” coercion or violence. There is all the world of distinction in kind between aggressive violence—assault or theft—against another, and the use of violence to defend oneself and one’s property against such aggression. Aggressive violence is criminal and unjust; defensive violence is perfectly just and proper; the former invades the rights of person and property, the latter defends against such invasion. Yet Hayek again fails to make this crucial qualitative distinction. For him, there are only relative degrees, or quantities, of “coercion.” Thus, Hayek states that “coercion, however, cannot be altogether avoided because the only way to prevent it is by the threat of coercion.”14

From this, he goes on to compound the error by adding that “free society has met this problem by conferring the monopoly of coercion on the state and by attempting to limit this power of the state to instances where it is required to prevent coercion by private persons.”15 Yet, we are not here comparing varying degrees of an undifferentiated lump we can call “coercion” (even if we define this as “physical violence”). For we can avoid aggressive violence completely by preventing it through purchasing the services of defense agencies, agencies which are empowered to use only defensive violence. We are not helpless in the throes of “coercion” if we define such coercion only as aggressive violence (or, alternatively, if we abandon the term “coercion” altogether, and keep the distinction between aggressive and defensive violence).

Hayek’s crucial second sentence in the above paragraph compounds his error many times further. In the first place, in any and all historical cases, “free society” did not “confer” any monopoly of coercion on the State; there has never been any form of voluntary “social contract.” In all historical cases, the State has seized, by the use of aggressive violence and conquest, such a monopoly of violence in society. And further, what the State has is not so much a monopoly of “coercion” as of aggressive (as well as defensive) violence, and that monopoly is established and maintained by systematically employing two particular forms of aggressive violence: taxation for the acquisition of State income, and the compulsory outlawry of competing agencies of defensive violence within the State’s acquired territorial area.

Therefore, since liberty requires the elimination of aggressive violence in society (while maintaining defensive violence against possible invaders), the State is not, and can never be, justified as a defender of liberty. For the State lives by its very existence on the two-fold and pervasive employment of aggressive violence against the very liberty and property of individuals that it is supposed to be defending. The State is qualitatively unjustified and unjustifiable.

Thus, Hayek’s justification of the existence of the State, as well as its employment of taxation and other measures of aggressive violence, rests upon his untenable obliteration of the distinction between aggressive and defensive violence, and his lumping of all violent actions into the single rubric of varying degrees of “coercion.” But this is not all. For, in the course of working out his defense of the State and State action, Hayek not only widens the concept of coercion beyond physical violence; he also unduly narrows the concept of coercion to exclude certain forms of aggressive physical violence. In order to “limit” State coercion (i.e., to justify State action within such limits), Hayek asserts that coercion is either minimized or even does not exist if the violence-supported edicts are not personal and arbitrary, but are in the form of general, universal rules, knowable to all in advance (the “rule of law”). Thus, Hayek states that

The coercion which a government must still use … is reduced to a minimum and made as innocuous as possible by restraining it through known general rules, so that in most instances the individual need never be coerced unless he has placed himself in a position where he knows he will be coerced. Even where coercion is not avoidable, it is deprived of its most harmful effects by being confined to limited and foreseeable duties, or at least made independent of the arbitrary will of another person. Being made impersonal and dependent upon general, abstract rules, whose effect on particular individuals cannot be foreseen at the time they are laid down, even the coercive acts of government become data on which the individual can base his own plans.16

Hayek’s avoidability criterion for allegedly “noncoercive” though violent actions is put baldly as follows:

Provided that I know beforehand that if I place myself in a particular position, I shall be coerced and provided that I can avoid putting myself in such a position, I need never be coerced. At least insofar as the rules providing for coercion are not aimed at me personally but are so framed as to apply equally to all people in similar circumstances, they are no different from any of the natural obstacles that affect my plans.17

But, as Professor Hamowy trenchantly points out:

If follows from this that if Mr. X warns me that he is going to kill me if I buy anything from Mr. Y, and if the products available from Mr. Y are also available elsewhere (probably from Mr. X), such action on the part of Mr. X is noncoercive!

Purchasing from Mr. Y is “avoidable.” Hamowy continues:

Avoidability of the action is sufficient, according to this criterion, to set up a situation theoretically identical to one in which a threat does not occur at all. The threatened party is no less free than he was before the threat was made, if he can avoid the threatener’s action. According to the logical structure of this argument, “threatening coercion” is not a coercive act. Thus, if I know in advance that I will be attacked by hoodlums if I enter a certain neighborhood, and if I can avoid that neighborhood, then I need never be coerced by the hoodlums…. Hence, one could regard the hoodlum-infested neighborhood … in the same way as a plague-infested swamp, both avoidable obstacles, neither personally aimed at me …

— and hence, for Hayek, not “coercive.”18

Thus, Hayek’s avoidability criterion for non-coercion leads to a patently absurd weakening of the concept of “coercion,” and the inclusion of aggressive and patently coercive actions under a benign, noncoercive rubric. And yet, Hayek is even willing to scuttle his own weak avoidability limitation on government; for he concedes that taxation and conscription, for example, are not, and are not supposed to be, “avoidable.” But these too become “noncoercive” because:

they are at least predictable and enforced irrespective of how the individual would otherwise employ his energies; this deprives them largely of the evil nature of coercion. If the known necessity of paying a certain amount in taxes becomes the basis of all my plans, if a period of military service is a foreseeable part of my career, then I can follow a general plan of life of my own making and am as independent of the will of mother person as men have learned to be in society.19

The absurdity of relying on general, universal (”equally applicable”), predictable rules as a criterion, or as a defense, for individual liberty has rarely been more starkly revealed.20 For this means that, e.g., if there is a general governmental rule that every person shall be enslaved one year out of every three, then such universal slavery is not at all “coercive.” In what sense, then, are Hayekian general rules superior or more libertarian than any conceivable case of rule by arbitrary whim?

Let us postulate, for example, two possible societies. One is ruled by a vast network of Hayekian general rules, equally applicable to all, e.g., such rules as: everyone is to be enslaved every third year; no one may criticize the government under penalty of death; no one may drink alcoholic beverages; everyone must bow down to Mecca three times a day at specified hours; everyone must wear a specified green uniform, etc. It is clear that such a society, though meeting all the Hayekian criteria for a noncoercive rule of law, is thoroughly despotic and totalitarian.

Let us postulate, in contrast, a second society which is totally free, where every person is free to employ his person and property, make exchanges, etc. as he sees fit, except that, once a year, the monarch (who does literally nothing the rest of the year), commits one arbitrary invasive act against one individual that he selects.

Which society is to be considered more free, more libertarian?21

Thus, we see that Hayek’s Constitution of Liberty can in no sense provide the criteria or the groundwork for a system of individual liberty. In addition to the deeply flawed definitions of “coercion,” a fundamental flaw in Hayek’s theory of individual rights, as Hamowy points out, is that they do not stem from a moral theory or from “some independent nongovernmental social arrangement,” but instead flow from government itself.

For Hayek government—and its rule—of law creates rights, rather than ratifies or defends them.22 It is no wonder that, in the course of his book, Hayek comes to endorse a long list of government actions clearly invasive of the rights and liberties of the individual citizens.23

  • 1A version of this section appeared in the 1980 issue of Ordo (Stuttgart).
  • 2F.A. Hayek, The Constitution of Liberty (Chicago: University of Chicago Press, 1960).
  • 3Ibid., pp. 20–21,133.
  • 4Ibid., p. 135.
  • 5Ibid., pp. 136–37.
  • 6Hayek, The Constitution ofLiberty, p. 136
  • 7Furthermore, as Professor Ronald Hamowy points out in a brilliant critique of Hayek’s concept of coercion, and of the “rule of law,”
    we are faced with what appears to be an insurmountable problem—what constitutes a “reasonable” price? By “reasonable” Hayek might mean “competitive.” But how is it possible to determine what the competitive price is in the absence of competition? Economics cannot predict the cardinal magnitude of any market price in the absence of a market. What, then, can we assume to be a “reasonable” price, or, more to the point, at what price does the contract alter its nature and become an instance of “coercion”? Is it at one cent a gallon, at one dollar a gallon, at ten dollars a gallon? What if the owner of the spring demands nothing more than the friendship of the settlers? Is such a “price” coercive? By what principle can we decide when the agreement is a simple contractual one and when it is not?
    Moreover, as Hamowy states,
    we must face yet a further difficulty. Is the owner acting coercively if he refuses to sell his water at any price? Suppose that he looks upon his spring as sacred and its water as holy. To offer the water to the settlers would contravene his deepest religious sentiments. Here is a situation which would not fall under Hayek’s definition of coercion, since the owner of the spring forces no action on the settlers. Yet it would appear that, within Hayek’s own framework, this is a far worse situation, since the only “choice” left open to the settlers now is dying of thirst.

    Ronald Hamowy “Freedom and the Rule of Law in F.A. Hayek,” Il Politico (1971–72): 355–56. Also see Hamowy, “Hayek’s Concept of Freedom: A Critique,” New Individualist Review (April 1961): 28–31.
    For the latest work on this subject, see Hamowy, “Law and the Liberal Society: F.A. Hayek’s Constitution of Liberty,” Journal of Libertarian Studies 2 (Winter 1978): 287–97, available in PDF; and John N. Gray, “F.A. Hayek on Liberty and Tradition,” Journal of Libertarian Studies 4 (Fall 1980), available in PDF.
  • 8Murray N. Rothbard, Power and Market, 2nd ed. (Kansas City: Sheed Andrews and McMeel, 1977), pp. 228–30.
  • 9Hayek, The Constitution ofLiberty, pp. 136–37.
  • 10Hamowy, “Freedom and the Rule of Law,” pp. 353–54.
  • 11F.A. Hayek, “Freedom and Coercion: Some Comments on a Critique by Mr. Ronald Hamowy,” Studies in Philosophy, Politics, and Economics (Chicago: University of Chicago Press, 1967), p. 349.
  • 12Hamowy, “Freedom and the Rule of Law,” p. 354n.
  • 13Hayek, The Constitution ofLiberty, p. 139.
  • 14Ibid., p. 21. One fallacy of Hayek here is in holding that if unjust coercion is wrong, then it should be minimized. Instead, being immoral and criminal, unjust coercion should be prohibited altogether. That is, the point is not to minimize a certain quantity (unjust coercion) by any means possible, including new acts of unjust coercion; the point is to impose a rigorous side-constraint on all action. For this distinction, see Robert Nozick, “Moral Complications and Moral Structures,” Natural Law Forum (1968): lff.
  • 15Hayek, The Constitution ofLiberty, p. 21.
  • 16Ibid., p. 21.
  • 17Ibid., p. 142
  • 18Hamowy, “Freedom and the Rule of Law,” pp. 356–57, n. 356. Indeed, in The Constitution of Liberty, p. 142, Hayek explicitly states that
    this threat of coercion has a very different effect from that of actual and unavoidable coercion, if it refers only to known circumstances which can be avoided by the potential object of coercion. The great majority of the threats of coercion that a free society must employ are of this avoidable kind.
    As Professor Watkins points out, on Hayek’s avoidability criterion for “non-coercion,” a person may be confronted by a
    “general abstract rule, equally applicable to all” which forbids foreign travel; and suppose he has an ailing father abroad whom he wants to visit before he dies. On Hayek’s argument, there is no coercion or loss of freedom here. This agent is not subject to anyone’s will. He is just confronted by the fact that if he tries to go abroad he will be apprehended and punished.
    J.W.N. Watkins, “Philosophy,” in A. Seldon, ed., Agenda for a Free Society: Essays on Hayek’s The Constitution of Liberty (London: Institute for Economic Affairs, 1961), pp. 39–40.
  • 19Hayek, The Constitution ofLiberty, p. 143.
  • 20On the problem of the universal rule changing as more and more types of specific circumstances are added to the rule, see G.E.M. Anscombe, “Modern Moral Philosophy” Philosophy 33 (1958): 2.
  • 21For a thorough critique of the generality, equal applicability, and predictability criteria of Hayek’s rule of law, as well as of Hayek’s admitted departures from his own criteria, see Hamowy, “Freedom and the Rule of Law,” pp. 359–76. This includes Bruno Leoni’s fundamental criticism that given the existence (which Hayek accepts) of a legislature changing laws daily, no given law can be more than predictable or “certain” at any given moment; there is no certainty over time. See Bruno Leoni, Freedom and the Law (Princeton, N.J.: D. Van Nostrand, 1961), p. 76.
  • 22See Hamowy, “Freedom and the Rule of Law,” p. 358.
  • 23In his more recent treatise, Hayek does not deal with the problem of coercion or freedom. He does, however, try in passing to meet the criticism of Hamowy and others by amending his concept of general and certain rules to exempt solitary actions and acts that are not “toward others.” While the problem of religious rules might then be avoided, most of the problems in the above discussion do involve interpersonal actions and therefore continue to prevent Hayek’s rule of law from being a satisfactory bulwark of individual liberty. F.A. Hayek, Law, Legislation, andLiberty, vol. 1 (Chicago: University of Chicago Press, 1973), pp. 101–2,170n. In general, the new Hayek volume is a welcome retreat from Hayek’s previous reliance on legislation and a turn toward the processes of judge-found common law; however, the analysis is greatly marred by a predominant emphasis on the purpose of law as “fulfilling expectations,” which still concentrates on social ends rather than on the justice of property rights. Relevant here is the discussion above of the “title-transfer” theory vs. the expectations theory of contracts; see pp. 133–48 above.
Image Source: Getty
What is the Mises Institute?

The Mises Institute is a non-profit organization that exists to promote teaching and research in the Austrian School of economics, individual freedom, honest history, and international peace, in the tradition of Ludwig von Mises and Murray N. Rothbard. 

Non-political, non-partisan, and non-PC, we advocate a radical shift in the intellectual climate, away from statism and toward a private property order. We believe that our foundational ideas are of permanent value, and oppose all efforts at compromise, sellout, and amalgamation of these ideas with fashionable political, cultural, and social doctrines inimical to their spirit.

Become a Member
Mises Institute