The meaning of this expression may seem obvious: It is the public interest when the public considered is composed of the individuals within a “nation” (as opposed to the “global public interest” or a “sub-national public interest”). I take “nation” to mean a society of individuals, not simply their state. But in reality, just as with the “public interest” in general, the “national interest” has little scientific or rationally ascertainable meaning.

Problems / A simple model can help explain. Let an individual’s interest mean the maximization of his utility as measured by his own ordinal preferences, which does not preclude an individual’s utility function including others’ welfare as the individual sees it. Imagine a society composed of three individuals with the following higher preferences: The first individual is interested in cars—that is, his major interest in life after his primary needs are met lies in driving and owning automobiles. In a similar way, the second individual is a bibliophile, and the third is a health buff. What is the public interest in this society? Or, if their society is a nation, what is the national interest?

Adding up individual interests does not work. There is no common metric for measuring and trading off the individuals’ subjective preferences. Tradeoffs are, of course, necessary because of scarcity: Human desires are nearly infinite compared with resource availability, and satisfying the interest of one individual undermines the interests of others. (Ascetics or hermits are exceptions, but they may still want a monastery or a tent.) It is not the national interest if the interests of all citizens do not count equally.

An intuitive response, especially in our zeitgeist, is to let our three individuals vote and the two-thirds majority decides the tradeoffs to be made among the interests of the three individuals. The national interest is then some sort of compromise among the interests of the two individuals in the majority coalition. Among the many problems of this solution, a majority of two can forever exploit the minority of one. Isn’t the minority as much a part of the nation as the majority?

Another problem is the Condorcet paradox, known since the 18th century. Even if each voter has transitive preferences (if he prefers A to B and B to C, then he prefers A to C), the outcome of voting can show intransitive collective preferences. It can happen that the majority prefers policy A to policy B, B to C, but C to A (Lemieux 2021). Moreover, different voting systems can produce different electoral outcomes (Tullock 2002). So, who’s the majority?

If our three individuals were identical—if they had the same preferences—it could perhaps be possible to speak of the national interest as a public interest. But human individuals are different from each other. Now, multiply the number of members in society by one million or one hundred million, and it is even more obvious that the public or national interest as an aggregation of all individuals’ interests does not exist.

Similarly, it makes no sense to speak of “the interest of the United States” (substitute the name of your preferred country if you wish). Even “the interests (plural) of the United States” makes no sense if we mean the interests of all Americans.

Voting and collective choices / In the middle half of the 20th century, the field of welfare economics reached a conclusion consistent with the foregoing, despite its main theorists being generally favorable to collective choices. Welfare economists investigated the conditions for the maximization of “social welfare”—that is, the best use of resources to maximize a society’s “aggregate utility.” They discovered that there is no way to do this without assigning weights to the different individuals’ utility, weights that are nonscientific and arbitrary. Most ethical theories do imply some weights and tradeoffs (say, “to each according to his needs”), but as long as individuals have different preferences, the application of one ethical theory to society—one “social welfare function” in welfare economics jargon—is no less arbitrary than another. In the terms of Francis Bator (1957), such a function “could be yours, or mine, or Mossadegh’s” and “is intrinsically ascientific.” (“Mossadegh” likely refers to Mohammad Mosaddegh, the prime minister of Iran in the early 1950s.)

If we ask whether the social welfare function could be derived from elections or referendums, we are back to our previous issues with voting. In 1951, Kenneth Arrow extended and formalized the Condorcet paradox (Arrow 1963 for the second edition). His Impossibility Theorem mathematically showed that collective choices by voting cannot be both rational (in the sense of respecting certain reasonable conditions including transitivity) and democratic (in the sense of respecting some basic conditions of democracy, such as nondictatorship). This conclusion doesn’t imply that dictatorship is desirable (Arrow himself was politically “progressive”), but rather that a rational public or national interest cannot be scientifically discerned from the preferences of a majority of individuals or citizens.

Arrow’s theorem generated academic controversy and a whole new field of study called social choice, which often questions the impossibility of meaningful collective choices. It seems, however, that believers in the aggregation of individual preferences must rely on some philosopher king to explain how a national interest can exist without all nationals recognizing it.

The only sort of interest that can be viewed as common to all individuals is, not surprisingly, a common interest based on some shared individual preferences. This formulation avoids the impossible mission of aggregating all individual interests into some public or national interest, two expressions that are at best misleading. In our simple model of three individuals, the common interest was assumed to be the individuals’ primary needs (food, shelter, and sex). Since nearly all individuals choose to live in society, we know they do have some common interest or set of interests. In a modern, diversified, and prosperous society, the common interest extends beyond primary needs. A “public good” or service that each citizen would find desirable—even with the tax price he must pay—would be a common interest.

Supremacy of the individual / There seems to be only one way to broadly characterize the common interest. Since it is impossible that everybody be a dictator and rule over all other individuals, the common interest consists in each individual being free to pursue his own interests through voluntary (nonviolent) exchange with others. The common interest is to live in a free society, a society where each is equally free to pursue his own happiness. This is the classical liberal ideal (and libertarian ideal if we discount the higher heterogeneity of libertarians as we know them).

In our time, three main attempts have been made to characterize a free society more precisely. Friedrich Hayek emphasized the model of traditional liberal institutions, such as the rule of law, that developed in Europe, mainly in the Anglo-Saxon tradition, and later in overseas offshoots (Hayek 2022). James Buchanan and the school of constitutional political economy search for rules that all members of a society would rationally accept (as if in a unanimous social contract) for guiding their social interactions (Brennan & Buchanan 1985). Anthony de Jasay’s approach would let Humean conventions analogous to successful strategies in game theory—such as keeping promises and recognizing private property and voluntary contracts—reassert their millennial force in a classical-liberal anarchy (de Jasay 2002).

These theories all recognize the presumptive supremacy of individual and private choices over collective and political choices. In a free society, each individual or private group pursues its own interests, and there is no national interest above that. Society is not a private association or corporation that has its own pre-existing goals to impose on others; it is not “team America.” Society is the general context within which individuals and their private groups separately pursue their own goals. In this setup, one may conceive “national” societies as distinct groupings normatively justified by attempts to maintain free societies and, at the polar opposite, groupings positively explained by successes in the exploitation of man by man.

The liberal ideal of a free society may seem overly abstract; it does not have an immediate, specific answer to many policy questions. But a coherent abstraction offers better guidance than an incoherent concept like the national interest that proposes impossible or exploitative recipes such as aggregating individual interests or loving a dictator. The national interest is a dangerous concept.

Readings

• Arrow, Kenneth J., 1963, Social Choice and Individual Values, 2nd ed., Yale University Press.
• Bator, Francis M., 1957, “The Simple Analytics of Welfare Maximization,” American Economic Review 47(1): 22–59.
• Brennan, Geoffrey, and James M. Buchanan, 1985, The Reason of Rules: Constitutional Political Economy, Cambridge University Press (Liberty Fund, 2000).
• De Jasay, Anthony, 2002, Justice and Its Surroundings, Liberty Fund; reviewed in Regulation 48(1): 55–58.
• Hayek, F.A., 2022, Law, Legislation, and Liberty, University of Chicago Press (original trilogy: 1973–1978).
• Lemieux, Pierre, 2021, “Populist Political Choices Are Meaningless,” Regulation 44(1): 54–57.
• Tullock, Gordon, 2002, Government Failure: A Primer in Public Choice, Cato Institute.