If a Pure Market Economy Is So Good, Why Doesn’t It Exist?

If a pure market economy is so good, why does it not already exist? If governments are so bad, why are they dominant throughout the world today? Indeed, is the widespread adoption of free markets ever likely to occur?

Many recent authors, including Tyler Cowen,1 Cowen and Daniel Sutter,2 Randall G. 3 Holcombe, and Andrew Rutten4 , question the feasibility of a pure libertarian society.5 They maintain that such a system cannot arise or persist because some people will always have both the incentive and the ability to use force against others. These authors offer several reasons why, even if society starts out in a perfect libertarian world without any states (as Murray Rothbard and others advocate),6 competing groups will eventually form a coercive government.

If we are lucky, this will be not too dissimilar from what we have today, but it could be even worse. Government may not be just or desirable, but “government is inevitable.”7 While these objections have been aimed specifically at radical libertarian ideas, they apply more broadly and are relevant to the general issue of social change.

We believe that the neoclassical framework of most of these authors, particularly Holcombe, Cowen, and Sutter, causes them to overlook perhaps the most important driving force of social change. When analyzing why people make choices, economists distinguish between people’s preferences and people’s incentives. Yet, when considering ways to alter behavior, almost all economists limit their focus exclusively to incentives. Changing preferences is ignored as an option in the strict neoclassical point of view.8

This limited framework is found among neoclassical economists across the board, from advocates of radical change, such as David Friedman,9 to accepters of the status quo, such as George Stigler.10  A large part of the agenda of normative public choice and constitutional economics is to build “knave-proof institutions” that are immune to people acting as the “opportunistically rational economic man.”11

Although most neoclassical economists are willing to discuss changing incentives through constraints, we believe that changing incentives is not the only way to alter people’s behavior, and it may not always be the easiest way. Consider the government campaign against smoking. Not only does the government attempt to change incentives with increased taxes but it also attempts to change preferences by convincing people that smoking is not a good thing.

As advocates of a laissez-faire society, we hardly endorse this government campaign, but it illustrates how advocates of change focus on incentives and preferences rather than incentives alone. Libertarians who oppose taxes on cigarettes but who also wish fewer people smoked readily recognize that they must rely on educational campaigns aimed at the preferences of smokers.12

Moreover, even if political economists want to change people’s incentives, to do this they need to change policy or institutions, and they can only do that by first changing people’s preferences about institutions. Unless one simplistically reduces all of history to a deterministic model in which all institutional change results solely from changes in external constraints (consider, for instance, Avner Greif,13  and, with greater sophistication and scope, Douglass North, John Joseph Wallis, and Barry R. Weingast14 ), political economists must look at preferences to explain social change. Holcombe, Cowen, and Sutter evade any consideration of ideology and other factors that may affect preferences, but we believe that social change without changes in preferences is rare. The notion that you can change policy without changing preferences is an illusion.

History provides many examples where preferences of enough people changed so that the result was significant changes in policy. By eliminating this analytical straightjacket imposed by neoclassical economics, economists could have a lot more to offer about how to improve the world. We are not arguing that libertarianism requires convincing 100 percent of people to support a free society. Instead, following Murray Rothbard,15  we argue that libertarianism (or for that matter any system) requires the support of a certain critical mass. When enough people support a free society and withdraw their support from governments, the ability of would-be predators to create government is diminished.

The Argument for Pessimism

Why might one adopt a pessimistic view about the possibility of social change toward a pure market economy? Reasons differ, but let us focus on the pessimism of two classical-liberal economists who have published a series of articles on this topic. Cowen and Sutter is the latest contribution to a string of papers related to the viability of a state-free society.16  Much of their reasoning applies to more limited free markets as well. The initial arguments for pessimism are in Cowen,17  who maintains that, without a government monopoly over the use of force, competing groups that can cooperate to resolve disputes can also collude to exercise coercion.

Cowen and Sutter follows up with the more general claim that the very factors, such as cooperation, that might make a libertarian society possible can also make government likely.18  Cowen and Sutter summarize,

If civil society can use norms to enforce cooperative solutions, that same society will be prone to certain kinds of cartels. In other words, cooperation-enhancing social features will bring bad outcomes as well as good outcomes. To provide a simple example, the Nazis relied on cooperation in addition to their obvious coercive elements in perpetrating their crimes. The ability to organize therefore is a mixed blessing.19

The Nazi example should have alerted Cowen and Sutter to the crucial role of ideology. Instead they conclude that a libertarian society is unlikely to survive because of a “paradox of cooperation.” Some people will be able to cooperate enough to threaten others with government or private force. Cowen and Sutter consider this problem a virtually unavoidable feature of a stateless society.

Some authors have questioned Cowen and Sutter’s claims about network industries facilitating cartels,20  but the authors reply that cartels are possible in network industries that use force.21  They argue that even if most people were peaceful, more powerful groups could threaten others, who would have little choice but to back down.

They represent this scenario using simple game theory. Although victims would be best off not being victimized at all, victims are better off being victimized without retaliating, rather than fighting back, because confrontations are costly. This is likely why most people pay the mugger or the tax collector even though they would prefer not to; losing one’s cash is better than prompting a confrontation and potentially losing one’s life.

As evidence that some will always threaten while others will always back down, Cowen and Sutter point to the existence of governments around the globe:

We must take seriously the fact that governments exist all around the world, for better or worse. … History shows that “cooperating to coerce” is relatively easy to establish, regardless of the exact path to that final state of affairs.22

This position is similar to that of authors writing in the public-choice tradition, including Holcombe and Rutten, who argue that some form of coercion will necessarily persist.23

In no uncertain terms, Holcombe writes, “Without government — or even with a weak government — predatory groups will impose themselves on people by force and create a government to extract income and wealth from these subjects,” concluding that “government is inevitable.”24  In a similar vein, Cowen writes, “Orderly anarchy again implies collusive anarchy,” stating, “libertarian ideology does not provide a safeguard against the emergence of government.”25

Most recently, Cowen has coined what he calls the “Paradox of Libertarianism,” which essentially maintains that libertarian success may have contributed to bigger government. Changes in government policy in the last few decades have moved in a libertarian direction, causing “much greater wealth and much greater liberty,” which, ironically, has increased public demand for government.26

For all these authors, libertarians are at an impasse. Even if people recognize that markets are good and coercion is bad, some will always attempt to use coercive government because it will be in their interest to do so. These critics might be called the pessimistic admirers of libertarianism. Libertarian ideals are nice, but they are impossible in practice.

The Argument Against Pessimism

Forgive us for favorably quoting a politician and a general, but as Dwight D. Eisenhower said, “Pessimism never won any battle.” Just because libertarianism has not fully triumphed anywhere in the world today does not mean that striving for it is futile. Cowen and Sutter’s analysis notably leaves out the importance of ideology and public opinion as constraints on government.

Within certain narrow assumptions, Cowen and Sutter’s 2005 and Cowen’s 2007 analyses all but guarantee the existence of government. In Cowen and Sutter, the payoffs of using coercion are positive because there are no external constraints, and in Cowen, government becomes more popular as income increases. But if the assumptions are different, the predicted payoffs are different, and the “inevitability” of statism becomes “inevitable” only under certain conditions.

This problem is starkest in Cowen’s most recent article, in which he takes current political opinion as fixed and assumes that the majority considers government a normal good like so many others. In the current world this may be true. But suppose that advocates of free markets are correct that markets are more civil and humane27  and that the more sophisticated or cultured point of view is to support liberty over coercion. This is an open question, but as people’s incomes increase and they become more educated they might be more likely to become less statist.28  Under these circumstances, statism would not be a normal good, but an inferior good.

Or consider Cowen and Sutter’s assumption about the positive payoffs of coercion. In the current world, one need not look further than the many rich government officials around the globe to see the truth in this. But the payoffs themselves are at least partly a function of institutions and are hardly constant for all time. Altering the institutions can alter the level and even the ranking of the payoffs.

Furthermore, the level of the payoffs is not the only relevant consideration in light of the subjective nature of people’s preferences. The subjective ranking of payoffs can change with preferences. Suppose that some external, ideological constraints, embedded in a widely recognized legal code, were placed on coercion. If these constraints were important enough, even would-be opportunists would decline to use coercion.

Cowen and Sutter might answer that by assuming no government they have already specified the relevant institutional constraints. But the varied legal regimes that stateless societies have exhibited throughout history belie this claim. Cowen initially dismissed “reliance upon libertarian ideology alone to defend the survival of anarchy” as a “deus ex machina.”29  But Cowen and Sutter admit that “cooperative efficacy relates only to the ability of a community to engage in collective action; the selection of projects to pursue is a separate question.”30  In other words, people conceivably can cooperate to achieve public goods or public bads. The Nazis sought public bads, but this result is not universal.

What factors influence a society’s mix of public goods and bads? According to Cowen and Sutter, this “separate question” is decided by “community leaders and public officials” based on which projects “suit their own interests.” Then what determines their interests? Here we are back again implicitly at institutions and ideology, unless Cowen and Sutter want to replace ideology with the deus ex machina of the preferences of leaders and officials.

Another striking example of how ideology implicitly enters the analysis of neoclassical economists, despite their best efforts to keep it out, comes from a critic of Cowen. Friedman, in his classic brief for anarchism, projects a polylegal order in which competing private courts and police enforce different codes of law that compete like “brands of cars.”31  These legal codes need not be libertarian, in Friedman’s view, although he argues that unlibertarian law will be more expensive to enforce than libertarian law. Therefore, self-interest will tend to drive polylegal anarchism toward libertarian outcomes.

But notice that Friedman’s private courts and police do obey at least one universal law, despite his failure to acknowledge as much. None of them collects taxes. Otherwise, his system collapses into the international anarchy we observe in the world today. How could such a uniform constraint against taxation arise except through a widely held ideological aversion to taxation?

Could preferences ever change so that people demand less statism or more constraints on government? If one adopts the narrow neoclassical public-choice assumptions of Cowen and Sutter, the answer is likely to be “no,” as preferences are static in strict neoclassical models. But this position overlooks two important facts about the world, namely that public opinion often changes, and public opinion does matter.

Caplan and Stringham contrast the mainstream public-choice view that interests rule the world with the views of Ludwig von Mises and Frédéric Bastiat, who believe that ideas rule the world. According to the Mises-Bastiat view, governments are able to get away with as much as they do only because they have the support of enough people. Bad policies persist only because the median voter prefers them.32

But the current demand for bad policies does not imply their inevitability any more than the current demand for Ford automobiles implies that Ford will forever retain its current market share. If people’s preferences can be changed, then big government is not necessarily something people will always demand. This is important because if enough people withdraw their support for various big government policies, then the state will have a difficult time imposing its policies on the unwilling masses. As Rothbard,33 Jeffrey Rogers Hummel,10 and others have argued, government officials get away with as much as people let them.

Herein lies the key to changing society — changing public opinion or people’s preferences toward government. And the only way people are likely to change their preferences is through education and persuasion; force is ineffective. This is why libertarian economists of different stripes believe that economic education plays such a crucial role.

Most people in the general public support various government policies because they truly believe that government needs to solve social problems. Only infrequently do they consider the possibilities that government may be the cause of problems or making problems worse.35  Nor do they consider the possibility that voluntary action may be capable of solving many so-called market failures.

If free markets can do wonders, as libertarian economists believe,36  then there is no inherent reason that the public needs to forever demand or even tolerate the state. Frédéric Bastiat maintains that the general public has been sold a bill of goods.37 The general public has been persuaded to believe in the necessity of government intervention in many areas.

Yet, if free-market economists had their way, the public would believe and behave otherwise. When a problem arose, the public would not immediately turn to the state to solve it. When the state tried to take on new roles, people would balk. A small group of people might try to use force to impose their will on the public, but without general support or general acceptance by the public that minority would have a difficult time getting its way.38 As Rothbard wrote, “The emperor’s clothes of supposed altruistic concern for the common weal would then be stripped from him.”39

At one level, our argument seems obviously true. As one of our colleagues associated with the Review of Austrian Economics wrote to us, “the central thesis of the paper is that libertarian anarchy will prevail where everyone is a libertarian anarchist. This point is uncontroversial.” Yet, as we have demonstrated, the point is indeed controversial. The objections of those who question the attainability or stability of a state-free society (as opposed to its desirability) all rest on an explicit or implicit rejection of the truism that ideas have consequences.

It is likely that every society will always have some people who want to use force. But we believe that people can only get away with force on a large scale if they have the support of enough people. Without widespread support, the ability to create governments is diminished.

If those who think a libertarian society is unattainable truly abandon the notion of preferences being fixed forever, their only alternative is to invoke the public-goods problem, or one of its many other variants, such as the prisoner’s dilemma or path dependency. Such problems allegedly prevent changes in people’s ideas from having strong impacts on the political outcome. But this raises an across-the-board objection to all sorts of improvements in policy.

Yet history is littered with examples in which public-spirited mass movements overcame free-rider incentives to achieve significant gains against state power …. Nobel Laureate Douglass C. North has observed that “casual observation … confirms the immense number of cases where large group action does occur and is a fundamental force for change.”40 Once one recognizes that people do not always behave in a narrowly self-interested manner; that they are sometimes (if not always) capable of ideological altruism or otherwise working to achieve goals whose material rewards will not fully compensate them for their efforts; that in a word, preferences are indeed flexible, then the power of ideas becomes paramount, as Hummel, Caplan and Stringham, Higgs, and North have all pointed out at length.

Thus, the ultimate factor in this worldview is public opinion. The more people adopt a culture of enterprise, the more able a system of free markets is to come about. Is the world where most people support a pure market economy inevitable, as Fukuyama implies in his argument for the inevitability of liberal democracy?41 We do not believe that any world is inevitable, but we believe that changing preferences to support a pure market economy is certainly possible.

[This article is excerpted from Stringham and Hummel, “If a Pure Market Economy is So Good, Why Doesn’t It Exist? The Importance of Changing Preferences versus Incentives in Social Change,” Quarterly Journal of Austrian Economics 2010, vol. 13, no. 2.]

• 1Tyler Cowen, “Law as a Public Good: The Economics of Anarchy,” Economics and Philosophy, vol. 8 (1992), pp. 249–267; and “Rejoinder to David Friedman on the Economics of Anarchy,” Economics and Philosophy, vol. 10 (1994), pp. 329–332.
• 2Tyler Cowen and Daniel Sutter, “The Costs of Cooperation,” Review of Austrian Economics, vol. 12 (1999), pp. 161–173.; and “Conflict, Cooperation and Competition in Anarchy,” Review of Austrian Economics, vol. 18, no. 1 (2005), pp. 109–115.
• 3Randall G. Holcombe, “Government: Unnecessary but Inevitable,” Independent Review, vol. 8 (2004), pp. 325–342; “Is Government Inevitable? Reply to Lesson and Stringham,” Independent Review, vol. 9, no. 4 (2005), pp. 551–557; and “Is Government Really Inevitable?” Journal of Libertarian Studies, vol. 21, no. 1 (2007), pp. 41–48.
• 4Andrew Rutten, “Can Anarchy Save Us from Leviathan?” Independent Review, vol. 3 (1999), pp. 581–593.
• 5For an overview of a pure libertarian or state-free economy, see Murray Rothbard, For a New Liberty: Libertarian Manifesto (San Francisco: Fox and Wilkes, 1996) and Edward Stringham, ed., Anarchy and the Law: The Political Economy of Choice (Somerset, N.J.: Transaction Publishers, 2007).
• 6See Rothbard, For a New Liberty.
• 7Holcombe, “Government: Unnecessary but Inevitable,” p. 333. Rothbard noted in For a New Liberty that “It is also particularly important for the State to make its rule seem inevitable: even if its reign is disliked, as it often is, it will be met with the passive resignation expressed in the familiar coupling of ‘death and taxes’” (p. 70).
• 8Murray N. Rothbard, “Toward a Reconstruction of Utility and Welfare Economics,” in On Freedom and Free Enterprise: Essays in Honor of Ludwig von Mises, ed. Mary Sennholz (Princeton, N.J.: Van Nostrand Company, 1956).
• 9David Friedman, The Machinery of Freedom: Guide to a Radical Capitalism, 2nd ed. (La Salle, Ill.: Open Court, 1989).
• 10 George Stigler, The Economist As Preacher and Other Essays (Chicago: University of Chicago Press, 1982).
• 10Jeffrey Rogers Hummel, “National Goods Versus Public Goods: Defense, Disarmament, and Free Riders,“Review of Austrian Economics, vol. 4 (1990), pp. 88–122; and “The Will to Be Free: The Role of Ideology in National Defense,“ Independent Review, vol. 5 (2001), pp. 523–537.
• 11Hartmut Kliemt, “Public Choice from the Perspective of Philosophy,” in Friedrich Schneider, ed., The Encyclopedia of Public Choice (New York: Kluwer, 2004), pp. 235–244. In addition to the public-choice economists, others who seek ways to constrain political institutions include Russell Hardin, Liberalism, Constitutionalism, and Democracy (Oxford: Oxford University Press, 1999); Douglass C. North, Institutions, Institutional Change and Economic Performance (Cambridge, Mass.: Cambridge University Press, 1990); and Barry R. Weingast, “The Economic Role of Political Institutions: Market Preserving Federalism and Economic Development,” Journal of Law, Economics, and Organization, vol. 11 (1995), pp. 1–31. Our approach is more fundamental, because we question whether constitutional rules or political structures can significantly constrain government. As Gordon Tullock writes, “The view that the government can be bound by specific provisions is naïve. Something must enforce those provisions, and whatever it is that enforces them is itself unbound.” For more on this, see Andrew Farrant, “Robust Institutions: The Logic of Levy?”Review of Austrian Economics, vol. 17 (2004), pp. 447–451. We argue that the ultimate and the only binding constraint on government is ideology, that is, the preferences of the public.
• 12A neoclassical economist analyzing the campaign against smoking could attempt to salvage the assumption of constant preferences by relying on Gary Becker’s discussion of “Z-goods.” See Gary Becker, “A Theory of the Allocation of Time,” Economic Journal, vol. 75, no. 299 (1965), pp. 493–508. Cigarettes, instead of being treated as a final consumer good (X-good), could be analyzed as an input for what Becker calls Z-goods, which require other goods for a household to produce. A meal, for example, is a Z-good that requires various food ingredients. See Robert B. Eklund, Robert F. Hébert, and Robert D. Tollison, The Marketplace of Christianity (Cambridge, Mass.: MIT Press, 2006). If we assume that smokers have imperfect information about the effects of smoking, we could analyze anti-smoking advertisements as simply providing additional information about the true costs of cigarette smoking, one of multiple inputs in the Z-good of relaxation. In this case, the change in behavior does not result from any change in preferences. Similarly, one could analyze libertarianism as one input to the Z-good of living a good life. Thus, informing people about the benefits and costs of markets versus government simply helps them see the true costs of the input goods. At the extreme, this neoclassical framework rules out all preference changes by defining the individual’s utility function as constant. While an intriguing philosophical ploy, we find this tautological definition of utility even less helpful in understanding the real world than the tautological definition of self-interest that encompasses any action, no matter how seemingly altruistic. (Furthermore, the claim that consumers do not know what is best for them contradicts the strict neoclassical assumption of perfect information.) We prefer to rely on the wording of everyday language. In any case, relabeling a preference change as improved information that changes incentives does not diminish the power of our argument. There remains a distinction, however you label it, between directly altering the consequences of an action and altering an individual’s perception of the consequences. Neoclassical economists almost invariably confine their analysis to the direct approach.
• 13Avner Greif, Institutions and the Path to the Modern Economy: Lessons from Medieval Trade (Cambridge: Cambridge University Press, 2006).
• 14Douglass C. North, John Joseph Wallis, and Barry R. Weingast, Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History (Cambridge: Cambridge University Press, 2009).
• 15Murray N. Rothbard, Four Strategies for Libertarian Change (London: Libertarian Alliance, 1989).
• 16Cowen and Sutter, “Conflict, Cooperation and Competition in Anarchy.”
• 17Cowen, “Law as a Public Good” and “Rejoinder to David Friedman.”
• 18Cowen and Sutter, “The Costs of Cooperation.”
• 19Cowen and Sutter, “Conflict, Cooperation and Competition in Anarchy,” p. 109.
• 20David Friedman responded to Cowen’s initial article (”Law as a Public Good”) in “Law as a Private Good: A Response to Tyler Cowen on the Economics of Anarchy,” Economics and Philosophy, vol. 10 (1994), pp. 319–327, while Bryan Caplan and Edward P. Stringham responded, in “Networks, Law, and the Paradox of Cooperation,”Review of Austrian Economics, vol. 16, no. 4 (2003), pp. 309–326, to Cowen and Sutter’s later paper (”The Costs of Cooperation”). Caplan and Stringham point out that just because people can cooperate on certain margins does not mean that they can collude on all margins. For example, banks can coordinate to make their charge cards acceptable to other banks, but they would have a much more difficult time colluding to set interest rates.
• 21Cowen and Sutter, “Conflict, Cooperation and Competition in Anarchy.”
• 22Ibid., p. 113.
• 23For an overview of the public choice arguments about anarchy, see Edward Stringham, ed., Anarchy, State, and Public Choice (Cheltenham: Edward Elgar, 2005). For responses to Randall Holcombe’s “Government: Unnecessary but Inevitable,” see Peter T. Leeson and Edward P. Stringham, “Is Government Inevitable? Comment on Holcombe’s Analysis,” Independent Review,vol. 9, no. 4 (2005), pp. 543–549; Walter Block, “Government Inevitability: Reply to Holcombe,“ Journal of Libertarian Studies,vol. 19, no. 3 (2005), pp. 71–93; and Randall G. Holcombe, “Is Government Inevitable? Reply to Lesson and Stringham” and “Is Government Really Inevitable?”
• 24Holcombe, “Government: Unnecessary but Inevitable,” p. 326.
• 25Cowen, “Law as a Public Good,” pp. 252, 261.
• 26Tyler Cowen, “The Paradox of Libertarianism,“ Cato Unbound (March 11, 2007). Cowen does not specify whether the resulting growth of government is merely on a per capita basis or as a percent of total output, and perhaps it is unfair of us to expect too much rigor from an online popular comment. But in a growing economy, the size of government can increase per capita while still declining relative to the economy’s size. If on the other hand, Cowen is resurrecting the tired, 20th-century progressive claim that government must grow as a percent of GDP as the economy gets bigger and more complex, his claim seems to be empirically doubtful in the United States, at least when looking at secular trends for the period to which he refers, namely the last several decades.
• 27See Wilhelm Röpke, A Humane Economy (Chicago: Henry Regnery, 1960).
• 28See Bryan Caplan, “How Economists Misunderstand Voters, and Why Libertarians Should Care,”Independent Review, vol. 5, no. 4 (2001), pp. 539–563.
• 29Cowen, “Law as a Public Good,” p. 251.
• 30Cowen and Sutter, “The Costs of Cooperation,” p. 165. Emphasis added.
• 31Friedman, The Machinery of Freedom, p. 117.
• 32Caplan, in The Myth of the Rational Voter: Why Democracies Choose Bad Policies (Princeton, N.J.: Princeton University Press, 2007), argues that people are more likely to demand more economically “irrational” or counterproductive policies when the marginal cost is low. Working within the framework of his model, one could reduce the quantity of irrational policies demanded in two ways. The first is by altering constraints to increase the personal marginal cost of people demanding irrational policies. Altering incentives in this way would be a movement along the demand curve for irrational policies. But a second way to decrease the number of irrational policies demanded would be to bring about a shift in the demand curve for irrational policies. Caplan’s analysis would suggest that since more educated people are more likely to think like economists, increasing education is an important way to change political economic beliefs.
• 33Rothbard, Four Strategies for Libertarian Change.
• 35See Robert Higgs, Against Leviathan: Government Power and a Free Society (Oakland, Calif.: The Independent Institute, 2004); and Neither Liberty nor Safety: Fear, Ideology, and the Growth of Government (Oakland, Calif.: The Independent Institute, 2007).
• 36Rothbard, For a New Liberty.
• 37Frederic Bastiat, Economic Sophisms, trans. Arthur Goddard (Irvington-on-Hudson, N.Y.: Foundation for Economic Education, 1964).
• 38Murray N. Rothbard, “Concepts of the Role of Intellectuals in Social Change Towards Laissez Faire,”Journal of Libertarian Studies, vol. 9, no. 2 (1990), p. 47.
• 39Rothbard, For a New Liberty, p. 72.
• 40Douglass C. North, Structure and Change in Economic History (New York: W.W. Norton, 1981), pp. 10–11.
• 41Francis Fukuyama, The End of History and the Last Man (New York: Free Press, 1992).