Defining Ideas

Milton Friedman: Old School Liberalism

Wednesday, October 4, 2017

Editor’s note: This essay is an excerpt of the new Hoover Press book Milton Friedman on Freedom, edited by Robert Leeson and Charles G. Palm. 

Liberalism, as it developed in the seventeenth and eighteenth centuries and flowered in the nineteenth, puts major emphasis on the freedom of individuals to control their own destinies. Individualism is its creed; collectivism and tyranny its enemy. The state exists to protect individuals from coercion by other individuals or groups and to widen the range within which individuals can exercise their freedom; it is purely instrumental and has no significance in and of itself. Society is a collection of individuals, and the whole is no greater than the sum of its parts. The ultimate values are the values of the individuals who form the society; there are no super-individual values or ends. Nations may be convenient administrative units; nationalism is an alien creed.

In politics, liberalism expressed itself as a reaction against authoritarian regimes. Liberals favored limiting the rights of hereditary rulers, establishing democratic parliamentary institutions, extending the franchise, and guaranteeing civil rights. They favored such measures both for their own sake, as a direct expression of essential political freedoms, and as a means of facilitating the adoption of liberal economic measures.

In economic policy, liberalism expressed itself as a reaction against government intervention in economic affairs. Liberals favored free competition at home and free trade among nations. They regarded the organization of economic activity through free private enterprise operating in a competitive market as a direct expression of essential economic freedoms and as important also in facilitating the preservation of political liberty. They regarded free trade among nations as a means of eliminating conflicts that might otherwise produce war. Just as within a country, individuals following their own interests under the pressures of competition indirectly promote the interests of the whole; so, between countries, individuals following their own interests under conditions of free trade indirectly promote the interests of the world as a whole. By providing free access to goods, services, and resources on the same terms to all, free trade would knit the world into a single economic community.

“Liberalism” has taken on a very different meaning in the twentieth century and particularly in the United States. This difference is least in the concrete political forms favored: both the nineteenth-century liberal and the twentieth-century liberal favor or profess to favor parliamentary forms, nearly universal adult franchise, and the protection of civil rights. But even in politics there are some not unimportant differences: in any issue involving a choice between centralization or decentralization of political responsibility, the nineteenth-century liberal will resolve any doubt in favor of strengthening the importance of local governments at the expense of the central government; for, to him, the main desideratum is to strengthen the defenses against arbitrary government and to protect individual freedom as much as possible; the twentieth-century liberal will resolve the same doubt in favor of increasing the power of the central government at the expense of local government; for, to him, the main desideratum is to strengthen the power of the government to do “good for” the people.

The difference is much sharper in economic policy where liberalism now stands for almost the opposite of its earlier meaning. Nineteenth-century liberalism favors private enterprise and a minimum of government intervention. Twentieth-century liberalism distrusts the market in all its manifestations and favors widespread government intervention in, and control over, economic activity. Nineteenth-century liberalism favors individualist means to foster its individualist objectives. Twentieth-century liberalism favors collectivist means while professing individualist objectives. And its objectives are individualist in a different sense; its keynote is welfare, not freedom. As Schumpeter remarks, “as a supreme, if unintended, compliment, the enemies of the system of private enterprise have thought it wise to appropriate its label.” The rest of this article is devoted entirely to liberalism in its original meaning, and the term will be used throughout in that sense.

Political liberalism and economic liberalism derive from a single philosophy. Yet they have frequently led independent lives in application, which suggests that their relation to one another deserves examination in the realm of ideas as well. During the nineteenth century, many countries adopted large elements of economic liberalism, yet maintained political forms that were neither liberal nor developing at any rapid pace in a liberal direction. Russia and Japan are perhaps the outstanding examples. During the twentieth century, countries that have achieved and maintained most of the concrete elements of the liberal political program have been moving away from liberal and toward collectivist economic policies. Great Britain is the most striking example; certainly for the first half of this century, the general drift of British economic policies has been toward greater direct intervention and control by the state; this drift has been checked in the past few years, but whether the check is more than transitory remains to be seen. Norway, Sweden, and, with a lag of several decades, the United States, exhibit much the same tendencies.

As already noted, liberal thinkers and writers in the nineteenth century regarded political reforms as in large part a means of achieving economic liberalism. The earlier political forms concentrated political power in the hands of groups whose special interests were opposed to such measures of economic liberalism as free trade. Let all the people have a vote, and there would be, so liberals like James Mill argued, no special interest. And since the general interest was simply the interest of all the individuals composing the society, and these in turn would be furthered most effectively by economic liberalism, democracy could be expected to rid itself of the dead hand of government and to give maximum scope to the invisible hand of self-interest.

In the twentieth century, a group of liberal thinkers, especially Henry Simons, Ludwig von Mises, and Friedrich von Hayek, has emphasized that this relation also runs in reverse: that economic liberalism is a means of achieving political freedom. Economic liberalism alone does not guarantee political -freedom—witness the examples of Russia and Japan cited earlier. But economic liberalism is, it is argued, an indispensable prerequisite for political freedom. Historically, there are no countries that enjoyed any substantial measure of political freedom that did not also practice a substantial measure of economic liberalism. Analytically, the preservation of political freedom requires protection against the concentration of power; it requires the existence of largely independent loci of power. Political power by its nature tends to be concentrated; economic power can be highly de-concentrated if it is organized by means of an impersonal market; economic power can thus be an independent offset to political power. Let both economic and political power be in the same hands and the only protection of political freedom is the goodwill of those in power—a frail recourse, particularly in view of the corrupting influence of power and the talents that make for political survival.

A few examples may clarify the asserted relation between economic and political freedom, though they cannot of course demonstrate it. A characteristic of a politically free society is that proponents of radical reform in the structure of the society are permitted to express their views and to seek to persuade their fellows. It is a testimonial to the freedom of the United States that socialist and communist magazines and papers are published. Suppose a change to a collectivist economic society with government control of the bulk of economic activity. How could the proponents of a return to capitalism secure the resources with which to publish a magazine urging their point of view? Through a government fund for dissidents? Through the collection of small sums from millions of government employees? If they had the resources in the form of funds, what guarantee could they have that the government would sell them paper on the same terms as it does to others? In an economically liberal society, it is possible to get the general resources with which to spread dissenting views either by subsidies from a small number of individuals or by selling a magazine or other publication to many; if there is a reasonable chance that enough people will want to buy a magazine expressing the minority view to make it profitable, even people who disagree fundamentally with the view will in their own self-interest provide the resources to make its establishment possible. In effect, there are thousands or millions of independent loci of power to decide whether an idea is worth trying to promote, rather than the few or one in a political structure. And given the general resources, there is no further obstacle: in a thoroughly free market, the sellers of paper do not know whether the paper is going to the Daily Worker or the Foundation for Economic Education. Perhaps some similar impersonal and effective guarantees of freedom to promote dissenting views could be contrived for a collectivist society; certainly no proponents of such a society have yet suggested any or even faced this problem squarely.

As another example, consider those individuals who have lost or resigned government jobs in the United States in recent years because they were or were accused of being Communists. Government employment in our society is not a right, and it is entirely appropriate that at least certain governmental positions should not be open to actual or suspected Communists. It is easy also to see how strong public feeling can lead to a closing of all government posts to Communists. Yet the maintenance of political freedom surely requires that people be free not only to believe in but also to advocate communism; those of us who abhor communism do so in part precisely because we know it would not grant us freedom to express contrary views; our defense against communism is to persuade our fellow citizens of its evil, not to suppress its advocates. But if government employment were the only employment, nominal freedom to express extreme views would be a mockery. The exercise of this freedom would be at a prohibitive price—namely, giving up the possibility of earning a living. By contrast, in the existing society those who have left government employment have had a wide variety of other opportunities. The way a private market economy protects these opportunities is revealed most clearly by considering an individual who goes into farming and produces, say, wheat. The purchasers of the wheat do not know whether it has been produced by a Communist or a Fascist, a white man or a Negro; they could hardly discriminate if they wanted to. The competitive market in this way separates economic activity from intellectual or political activity; the more competitive the market, the more sharply it does so. It is a paradox that minorities who have in this way the most to gain from a competitive society have contributed unduly large numbers to the ranks of its opponents.

Many writers who emphasize the importance of economic liberalism as a prerequisite for political freedom have interpreted tendencies toward collectivism in recent decades as betokening a trend toward political “serfdom.” They may yet be proved right. So far, however, the relation they stress has manifested itself mostly in a very different way: namely, the collectivist tendencies have been checked because they tended to interfere with civil and political freedom; when the conflict has been reasonably clear, the collectivist policy has frequently given way. Perhaps the most striking example is the British experience with the compulsory allocation of labor. Socialist economic thinking in the postwar period called for compulsory allocation of labor to achieve “social priorities”; though some compulsory powers were provided by law, they were never widely used; the powers themselves were permitted to lapse; and the whole character of attempted economic policy changed because compulsory allocation of labor so clearly interfered with widely and deeply cherished civil rights.

We turn now to a more detailed examination of the content of liberalism, particularly economic liberalism, and of the role it assigns to the state. This examination deals primarily with the principles that liberalism provides for judging social action. Any set of concrete proposals that these principles lead liberals to favor will vary with the particular circumstances of time and place and, consequently, are less fundamental and invariant than the principles themselves. There is some possibility of being reasonably comprehensive with respect to principles; none, with respect to concrete proposals.

Principles for social action must be based on both ultimate values and on a conception of the nature of man and the world. Liberalism takes freedom of the individual—really, of the -family—as its ultimate value. It conceives of man as a responsible individual who is egocentric, in the sense not of being selfish or self--centered but rather of placing greater reliance on his own values than on those of his neighbors. It takes as the major problem of modern society the achievement of liberty and individual responsibility in a world that requires the co-ordination of many of millions of people in production to make full use of modern knowledge and technology. The challenge is to reconcile individual freedom with widespread interdependence.

The liberal answer derives from the elementary—yet even today little understood—proposition that both sides to an economic transaction can benefit from it; that a gain to a purchaser need not be at the expense of a loss to the seller. If the transaction is voluntary and informed, both sides benefit; the buyer gets something he values more than whatever he gives up, and so does the seller. In consequence, voluntary exchange is a way to get cooperation among individuals without coercion. The reliance on voluntary exchange, which means on a free market mechanism, is thus central to the liberal creed.

The working model which embodies this vision of a society organized through voluntary exchange is a free private enterprise exchange economy. The elementary social unit—the family or household—is generally too small for efficient use of modern productive techniques. Accordingly, the productive unit takes the form of an enterprise which purchases productive services—labor, the use of capital, and so on—from households and other enterprises and sells the goods or services it produces to households and other enterprises. The introduction of such enterprises does not change the strictly voluntary and individual character of the cooperation, provided two conditions are satisfied: first, the enterprises are private, which means that the ultimate locus of authority and responsibility is an individual or a group of individuals; second, individuals are free to sell or not sell their services or to buy or not buy products from particular enterprises, which means, also, that they are free to establish new enterprises.

The final point deserves special emphasis in view of the widespread misunderstanding of it. “Free” in the liberal conception of free enterprise means freedom to establish enterprises to do whatever they want. This is one of those familiar cases in which absolute freedom is impossible because the freedom of some limits the freedom of others: the freedom of existing enterprises to do whatever they want, including combining to keep new entrants out or to fix prices and divide markets, may limit the freedom of others to establish new enterprises or to make the best bargain they can. When such a conflict arises, the liberal tradition regards freedom of entry and of competition as basic; it therefore justifies state action to preserve competition and to make selling a product of higher quality or at a lower price the only means whereby existing enterprises can prevent new enterprises from being established. The most difficult practical problem in this area and one on which liberals have spoken with many tongues is combinations among laborers—the trade union problem.

In the free enterprise exchange economy envisaged by liberalism, the primary role of government is to preserve the rules of the game by enforcing contracts, preventing coercion, providing a stable monetary framework, and, as just noted, keeping markets free. Beyond this there are only three major grounds on which government intervention is justified: (1) “natural monopoly” or similar market imperfections; (2) the existence of substantial “neighborhood effects”; (3) protection of children and other irresponsible individuals.

An exchange is “voluntary” only when essentially equivalent alternatives exist: when an individual can choose whether to buy from one enterprise or another, to work for one enterprise or another. Monopoly means the absence of alternatives and thus is incompatible with strictly voluntary exchange. Monopoly may arise from combinations among enterprises in circumstances where competition is entirely feasible, and, as already noted, liberal tradition justifies state intervention to preserve competition in such cases. But monopoly may also be “natural,” as in the textbook examples of a single spring providing drinking water, or of a product subject to such large economies of scale that the most efficient productive unit is large enough to serve the whole market. The only available alternatives are then all bad: government regulation, government ownership, or private monopoly, and the problem is to choose the least of these evils. As might be expected, liberals have no clear-cut answer. Henry Simons, after observing in the United States the consequences of government regulation of such alleged natural monopolies as the railroads, concluded that government ownership was the least of the evils when monopoly was inevitable. Walter Eucken, after observing in Germany the consequences of government ownership, concluded that government control was the least of the evils. And some have argued that in a dynamic world private monopoly may well be, citing the case of the regulation of transportation in the United States as their principle example. The Interstate Commerce Commission was established to protect the public against the railroads when railroads probably did have a large element of natural monopoly. The development of highway and air transport has largely eliminated any natural monopoly element in railroads, yet instead of the abolition of the Interstate Commerce Commission, government control has been extended to these other transportation media. The ICC has become a means of protecting the railroads from the competition of trucks instead of the public from the absence of competition. Fortunately for the possibility of a liberal society, the area within which natural monopoly is a serious problem is exceedingly limited, so that no large amount of government intervention is called for on this score. In practice, the claim of natural monopoly is more often an excuse for intervention desired on other grounds than a valid justification for intervention.

A different kind of threat to strictly voluntary exchange arises from the so-called “neighborhood effect.” This occurs when the action of one individual imposes significant costs on other individuals for which it is not feasible to make him compensate them or yields significant gains to other individuals for which it is not feasible to make them compensate him. A simple example is that of an individual polluting a stream. He has in effect forced other individuals further down the stream to exchange good water for bad: they clearly would have been willing to do so for a price; but it is not feasible to make this exchange the subject of voluntary agreement. Another rather different example, is education. The education of a child is regarded as benefiting not only the child and his parents but also other members of society, since some minimum level of education is a prerequisite for a stable and democratic society. Yet it is not feasible to identify the particular individuals benefited by the education of any particular child, much less the money value of the benefit, and so to charge for the services rendered. In consequence there is justification on liberal grounds for the state requiring some minimum amount of education for all children, even though this is above the amount parents would otherwise provide, and for meeting some of the cost of education from taxes imposed on all members of society.

Of course, all actions of an individual involve some “unborne costs” and “inappropriable benefits” to third parties. It is always a question of judgment whether these are sufficiently great in any particular case to justify state intervention; the state too will be plagued by the difficulties of identifying costs and benefits that prevent voluntary exchange and there are other costs of state action. The liberal philosophy thus gives no hard and fast line separating appropriate from inappropriate state action in this area. But it does emphasize that, in deciding any particular case, one general cost of state action—one general neighborhood effect, as it were—must always be taken into account; namely, that the extension of state action involves an encroachment on individual freedom. The liberal regards this as a count against any proposal for state action, though by no means a fatal obstacle to it, and hence requires a clear net balance of gains over other costs before regarding the state action as justified.

The third ground on which liberalism justifies state intervention derives from an ambiguity in the ultimate objective rather than from the difficulty of achieving fully voluntary exchange. The belief in freedom is for “responsible” individuals; children and insane people cannot be regarded as “responsible.” In general, this problem is avoided for children by treating the family as the basic unit of society and so regarding the parents as responsible for their children. In considerable measure, however, this procedure rests on expediency rather than principle. The problem of drawing a reasonable line between action justified on these paternalistic grounds and action that conflicts with the freedom of responsible individuals is clearly one to which no fully -satisfactory—and certainly no -simple—answer can be given.

A few additional examples may clarify the bearing of these principles in judging particular acts of social policy. Consider first a group of measures that clearly conflict with liberal principles: tariffs, direct controls of imports and exports, exchange control, general price controls. None of these can be justified on any of the grounds for state intervention that we have listed. Each represents an interference with the freedom of individuals to engage in any transactions that they want to and thus involve a direct interference with essential freedoms. An extreme case that brings out this feature strikingly is the “tourist allowance” incorporated in the exchange control regulations of a number of countries; as the Economist puts it, “it sets a limit on the length of time that even the most economical British resident can choose to spend abroad without asking the permission of some bureaucrat.” Finally most of these measures prevent the market from operating effectively and thus threaten the heart of the liberal system. For example, if a legal maximum price is established below the level that would otherwise prevail, a shortage will inevitably follow (as in housing under rent control); some method other than the free market will have to be used to ration the available amount and further government intervention replacing the market is set in train.

A second rather different example is medicine. As already noted, significant neighborhood effects justify substantial public health activities: maintaining the purity of water, assuring proper sewage disposal, controlling contagious diseases. There is little or no justification on these grounds for state intervention into private medicine: the care and treatment of individuals. Under liberal principles, this is a function the market can and should perform. To the argument of the proponent of socialized medicine that there is great uncertainty about possible medical bills, the liberal will reply that the market is perfectly capable of providing private insurance; if people don’t want to pay the premium, that is their free choice. To the argument that people don’t get as much medical service as is “good” for them, the liberal will reply that each man should judge for himself—not that he is necessarily the best judge but that he should make his own mistakes; insofar as we think he is making the wrong decision, there is no objection to telling him what we think and trying to persuade him, but there is no justification for making his decision.

A third example is public housing. It may be that certain types of housing (e.g., dense slum districts) impose higher costs of police and fire protection on the community. This literal “neighborhood effect,” since its source can be identified, would justify higher taxes on such property than on others to meet those extra costs; it hardly justifies subsidies to housing. The main argument for public subsidy to housing is surely paternalistic: people “need” or “deserve” better housing, and it is appropriate to use public funds to provide housing. The liberal will object on two different levels. Given that some people are to be subsidized, why not give them the subsidy in general purchasing power and let them spend it as they will? Why say to them, we will give you a gift if you take it in the form of housing but not otherwise? Does this not involve an unnecessary restriction on their freedom? Second, he will question the redistribution of income itself that is involved in such a program; indeed, one advantage of making the subsidy explicit is that it would make clear what groups are being subsidized. Government relief of poverty, the liberal will support and welcome, primarily on the explicitly paternalistic ground of taking care of the irresponsible. But the more or less indiscriminate transfer of income involved in large-scale public housing schemes, he will regard as undermining individual responsibility. The way to reduce inequality, he will urge, is not by the misleading palliative of sharing the wealth but by improving the workings of the market, strengthening competition, and widening opportunities for individuals to make the most of their own qualities.

These two final examples illustrate how the central virtue of a liberal society is at the same time a major source of the objections to it: a liberal society gives people what they want instead of what a particular group thinks they ought to want or thinks is “good” for them; it makes it equally difficult for the benevolent and the malevolent to shape other people in their own image. At bottom of most arguments against the market is lack of belief in freedom—at least for other people—as an end.

Adam Smith provides an excellent summary of the preceding discussion of the role of the state in a liberal society:


Every man, as long as he does not violate the laws of justice, is left perfectly free to pursue his own interest his own way, and to bring both his industry and capital into competition with those of any other man, or order of men. The sovereign is completely discharged from a duty, in the attempting to perform which he must always be exposed to innumerable delusions, and for the proper performance of which no human wisdom or knowledge could ever be sufficient; the duty of superintending the industry of private people, and of directing it towards the employments most suitable to the interest of the society. According to the system of natural liberty, the sovereign has only three duties to attend to; three duties of great importance, indeed, but plain and intelligible to common understanding: first, the duty of protecting the society from the violence and invasion of other independent societies; secondly, the duty of protecting, as far as possible, every member of the society from the injustice or oppression of every other member of it, or the duty of establishing an exact administration of justice; and, thirdly, the duty of erecting and maintaining certain public works and certain public institutions, which it can never be for the interest of any individual, or small number of individuals, to erect and maintain; because the profit could never repay the expense to any individual or small number of individuals, though it may frequently do much more than repay it to a great society.