James Buchanan is not easy to categorize. Is he a libertarian? A classical liberal? A conservative? Or perhaps even a “liberal” in the modern American sense of “progressive” or “social democrat”? Is he an economist or a philosopher? It is paradoxical but not totally wrong to answer “all of the above,” so complex and rich is his contractarian theory of the state. His 1975 book, The Limits of Liberty: Between Anarchy and Leviathan,1 has become a classic exposition of individualist contractarianism.

Buchanan was awarded the 1986 Nobel prize in economics for “his development of the contractual and constitutional bases for the theory of economic and political decision-making.” The Royal Swedish Academy of Sciences cited The Limits of Liberty as one of the two books in which he presents his “visionary approach.”2

Buchanan repeatedly states he does not want to impose his own values, except for individualism as the starting point of his analysis. A “good society” cannot be “defined independently of the choices of its members, all members” (emphasis by Buchanan). “My approach,” he writes at the beginning of the book, “is profoundly individualistic, in an ontological-methodological sense” (emphasis in original); “[e]ach man counts for one, and that is that.” It follows that individual liberty is a value and that the social system should be based on unanimous consent. Any limit to liberty must thus be consented to by each and every individual.

The Constitutional Contract

In Buchanan’s view, anarchy, or the absence of organized political power, does not work, for the reasons explained by Thomas Hobbes and because public goods can only be produced (or financed) by government. Hobbes famously argued that anarchy is characterized by “a War of every man against every man” and that “the life of man [is] solitary, poore, nasty, brutish, and short.”3 In the terms of game theory, anarchy leads individuals in the worst case of a prisoners’ dilemma.

Given this, what would be an efficient state, that is, one that responds to individual preferences? Buchanan answers that it should be based on a unanimous “constitutional contract” or “social contract” (he uses both expressions) establishing the basic rules of life in society.

Start with the equilibrium that would obtain in an anarchic society. It would be an inefficient equilibrium as individuals waste resources (including life) in predation and defense. (As for an anarchy regulated by stifling tribal rules and superstitions, Buchanan reckons that it would be even more inefficient.) This starting point provides a “natural distribution,” that is, some assignment of rights and behavior boundaries, from which a “constitutional contract” may be negotiated.

Think of the social contract as a set of constraints that everybody accepts in order to obtain guaranteed rights and to avoid the violence and waste of anarchy. Only unanimous agreement is necessary; there is no need to invoke “natural law.”

A constitutional contract between self-interested individuals contains four elements:
1. A “disarmament contract,” whereby individuals unanimously commit to avoid predatory behavior.
2. A definition of rights.
3. The limits of a “protective state” established to enforce the social contract.
4. The rules of collective decision-making for the “productive state,” charged with producing the other public goods that individuals may want.

Contrary to other contractarian theorists, Buchanan does not assume that individuals come to the negotiating table as equals in terms of resources or capabilities. As in ordinary exchange, individuals are only equal—but really equal—with regard their status in the exchange: each one is equally free to say yes or no.

A corollary follows. The Rambos who are more efficient at predation than at production might have to be bought off before they consent. When I first read The Limits of Liberty circa 1984, I was certainly not the only libertarian to be disturbed by the justification this could provide for the welfare state. Meet the “liberal” Buchanan. (I put “liberal” in quotation marks when I want to emphasize the ambiguity of the label as it was adopted by American progressives.) Even if disturbing, the idea is certainly widely shared in the public (even in the United States) that helping the poor is a bargain for the taxpayer as it prevents theft and violence. The logic of Buchanan’s argument, within this theoretical framework, seems unassailable.

Postconstitutional Exchange

After this social or constitutional contract is agreed on, Buchanan conceives of a “second, or postconstitutional, stage of negotiation,” which involves trade in public goods (agreement on their production, financing, and consumption): such things as dams, roads, parks, perhaps schools (if a minimum of education is necessary in a free society), and so forth. Buchanan’s social contract thus involves two stages: the basic rules at the constitutional stage; and postconstitutional contracts on public goods.

Trades in private goods don’t require any general agreement once the constitutional contract has defined individual rights. But such a preliminary definition of rights—which economists before Buchanan just assumed to exist—is necessary.

Another intriguing idea is that “[t]he dividing line between private and public goods depends, in part, on how the property rights of persons are defined.” To illustrate Buchanan’s brief remark, the social contract may exclude some land from private appropriation, which transforms these public lands into a public good. Similarly, the social contract could define a restaurant owner’s property rights as including the ambient air, which would allow him to welcome smokers if he chooses to.

A requirement of unanimity in decisions related to specific public goods would give a veto to every citizen. Anyone would thus gain holdout power and an incentive to sell his consent for lower taxes or for some transfer. This free-rider problem, argues Buchanan, is much more pressing at the post-constitutional stage than at the constitutional stage: in the latter case, an individual cannot obtain the benefits of the social contract without consenting to the whole package, while he can always consume a pure public good once it is financed by others (by definition of a public good). Recognizing this, rational adherents to the social contract will establish some voting rule granting the right of decision to some majority (simple or qualified) at the postconstitutional stage.

In order to embody unanimous consent, that is, to be efficient, the exchange of public goods requires constitutional limits on the state. “The operation of an unconstrained collectivity,” Buchanan supposes, “could scarcely emerge from rational constitutional contracting among persons.”

But note that the contract also limits the liberty of individuals. Once these limits are unanimously accepted at the constitutional level, “it becomes inconsistent and self-contradictory for a person to claim that his ‘rights’ are violated in the mere working out of the collective decision rules that are constitutionally authorized.”

This contractual setup allows individuals “to attain maximum liberty within the constraints of acceptable order.”

Problems of Unanimity

Buchanan is of course aware of the objection that a social contract could at best bind only those who signed it, presumably generations ago.4 The social contract, he answers, is an implicit contract that is continually renegotiated: “The formation of constitutional contract is continuous.” This insures that the social contract remains consistent with everybody’s “renegotiation expectations”—the net benefits one could obtain if the contract were renegotiated now, starting from a new anarchic equilibrium—or, in other words, that it remains Pareto-superior.

The adjustments that become necessary when renegotiation expectations change are often proposed by overly advantaged individuals who realize that it is in their preventive interest to embrace reform. They implicitly offer a new bargain to those who intuitively realize that their position in anarchy would now be better than the status quo. To understand what Buchanan is saying, think about rich coastal “liberals” who are often at the forefront of “social justice” and redistribution movements.

Like all forms of imagined unanimous social contract, the Buchanan sort raises questions. Unanimity is a clear concept when applied to private arrangements, but what does a unanimous social contract mean? Can we hope for literal unanimity or is quasi-unanimity sufficient? Buchanan opens a Pandora’s box when he says without elaboration that adjustments to the social contract are to be “agreed on by all or substantially all of the members of the community.” How much is “substantially all”? How do we know that near-unanimity has been reached?

Problems of Government

The state issued from the social contract is a delicate institution. What Buchanan calls “the paradox of being governed” comes from the dual roles of the state, “protective” and “productive,” corresponding to the two stages of social contracting. On the one hand, rational parties to the constitutional contract want the protective state to enforce the rules agreed on, no more and no less. The protective state is neutral and as “external” as possible. The productive state, on the other hand, is “internal”: it is “us”; its decisions are subjective and participatory.

“The state is, at the same time, both us and non-us.”

The state is, at the same time, both us and non-us. This idea may seem to solve a conundrum of any liberal theory of the state, but perhaps it only emphasizes the problem. If the two corresponding roles are confused—if the protective state becomes us and the productive state changes the terms of the contract—people will become dissatisfied with the constitutional order, and the social contract will be threatened. Buchanan saw signs of this in the 1960s and 1970s.

It is easy to agree with Buchanan that the paradox of being governed, as many other problems, increases with the size of government. The larger the scope of the state, the more likely it will violate the individual rights that have been agreed on.

Buchanan was worried about the decline of private ethical norms. “Ordered anarchy,” he argued, is assisted by “privately imposed constraints on behavior, through adherence to basic norms of mutual tolerance and mutual respect for acknowledged rights.” As an example of this sort of informal law, he mentions respect for free speech on university campuses, which was starting to be undermined at the time he was writing. When these informal rules weaken, more state-imposed constraints are required.

Law (both formal and informal) is a public good and public capital. Depreciation of this capital carries a high cost in terms of the future returns of social stability. It will often be preferable to maintain existing legal capital than to scrap it and rebuild it anew, which lends support to “the basic mystique of orthodox conservatism.” The author of The Limits of Liberty also believes that ethical constraints include “obedience to and respect for formalized law.” Meet the more conservative Buchanan.

A chapter titled “The Punishment Dilemma” looks puzzling. It argues that punishments for transgression of the constitutional contract can be dangerously watered down over time, leading to an erosion of law. Ex ante, punishment is an essential deterrent. But ex post, its application gives disutility to officials and voters.

Buchanan admits that constitutionally-decided, deterrence-oriented punishments will not be watered down “if the average or representative person in the community should actually enjoy seeing others punished,” but he discounts this possibility. From our vantage point, he turned out to be wrong. Since he wrote The Limits of Liberty, punishments have become more and more severe in response (at least partly) to the inflation of laws. One indication is that 8% of American adults have a felony record, that is, are “convicted felons” for life, and their number has been growing rapidly since the mid-1980s.5 Another indication is that America is the country with the highest rate of incarceration in the world (even if it has been decreasing since 2008).6 The real problem appears to be the growth of the punishing state in America, not its undermining.

This being said, Buchanan correctly saw that when people are dissatisfied with the erosion of formal and informal constraints, their demands for “law and order” mean “increased rather than decreased collectivization of society.”

The Threat of Leviathan

Besides the necessary limits to liberty, the threat of Leviathan is the second broad thread in the book, and what perhaps makes it more libertarian. A non-limited democratic state gives free rein to the self-interest of politicians and bureaucrats in expanding government and overturning constitutional rights. A majoritarian government can also fail to maximize citizens’ benefits by expanding government too much. Its fiscal policies can benefit the majority at the expense of the minority, or the other way around. Majorities must be constrained by constitutional rules, just as an individual creates his own rules (setting an alarm clock, for example) to control his temptations.

In reply to people’s dissatisfaction, both “liberals” and libertarians propose top-down solutions, Buchanan argues. He sees the solution neither in pragmatic politics nor in a choice between the laissez-faire model (even if it was based on the principle of ordered anarchy) and the socialist model, but in a “constitutional revolution” based on consent. “[G]ood,” he repeats, “is that which emerges from agreement among free men, independent of intrinsic evaluation of the outcome itself.”

A constitutional revolution means that individual rights and claims will be redefined and new rights created, including those related to the environment or congestion, as Buchanan mentions, perhaps influenced by the environmental scares of the 1970s.7 A new redistribution will be negotiated, and new limits imposed on Leviathan. Here, we find a mix of both the “liberal” and the libertarian Buchanan. In other places, Buchanan explained how he favored “equality of opportunity” and opposed “inherited wealth.”8

In a Nutshell

The basic idea of Buchanan’s contractarianism is that, in order to live in society, individuals have to agree on their rights and on the limits of their liberty. They have to discuss and exchange until they agree. The alternative is violent Hobbesian anarchy, and nobody wants that if he can have better. Buchanan tries to prove that individuals can agree. He also argues that rational self-interest leads individuals at the constitutional stage to limit the state, a crucial idea. The whole exercise aims at having as much (ordered) anarchy as possible: “the ideal society is anarchy, “he writes, “in which no man or group of men coerces another.”

For more on these topics, see Is Leviathan Required for a Peaceful Order? by Anthony de Jasay, Library of Economics and Liberty, March 7, 2016. See also A Conversation with James M. Buchanan, Parts I and II, Econlib Videos, 2013.

But does the contractual state require too much loyalty and obedience to law, which can fuel the growth of Leviathan? How can a unanimous social contract between unequal individuals lead to anything other than laissez-faire? Is it actually possible to limit the state? As Anthony de Jasay suggests, isn’t any state a Leviathan-to-be?9 And is the state really necessary? Do the two conditions for its necessity—the nastiness of Hobbesian anarchy and the impossibility of efficient private production of public goods—obtain?

It is difficult to think about these fundamental questions without reading The Limits of Liberty.


[1] James M. Buchanan, The Limits of Liberty: Between Anarchy and Leviathan (1975) (Indianapolis: Liberty Fund, 2000). Available online at and in the Liberty Fund Book Catalog at

[2] Royal Swedish Academy of Sciences, The Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 1986, at (accessed September 12, 2018).

[3] Thomas Hobbes, Leviathan (1751) (Oxford: Clarendon Press, 1909), at (accessed September 12, 2018).

[4] That was one of the objections of Lysander Spooner in his “The Constitution of No Authority,” No Treason 6 (1870), at (accessed September 15, 2018).

[5] Sarah K.S. Shannon et al., “The Growth, Scope, and Spatial Distribution of People with Felony Records in the United States, 1948-2010,” Demography 54 (2017).

[6] World Prison Brief, at (accessed September 13, 2018).

[7] A sense of the climate of the times can be gathered from my review “Running Out of Everything,” Liberty and Law, at (accessed September 16, 2018).

[8] For example, see James M. Buchanan, The Intellectual Portrait Series: A Conversation with James M. Buchanan (Part 2) (Indianapolis: Liberty Fund, 2011), at (accessed September 24, 2018).

[9] See my review of de Jasay’s The State in “An Unavoidable Theory of the State?” Library of Economics and Liberty, June 4, 2018, at (accessed September 15, 2018).

*Pierre Lemieux is an economist affiliated with the Department of Management Sciences of the Université du Québec en Outaouais. His latest book is What’s Wrong with Protectionism? Answering Common Objections to Free Trade (Rowman & Littlefield, 2018). He blogs on EconLog. He lives in Maine. E-mail:

For more articles by Pierre Lemieux, see the Archive.