On the Private Provision of National Defense

A Libertarian Society Will Be Nationalistic, or It Will Not Be

Very short summary: This essay is a critical discussion of how libertarians like David Friedman imagine national defense could be provisioned in a capitalist anarchy. I claim that the kind of voluntary contributions needed for such a public good could only by made in a highly nationalistic society.

---

I’ve been working on a joint article with a colleague about libertarian politics, cryptocurrencies, and smart contracts. For the preparatory work, I went back to one of the few loci classici of libertarian thought, David Friedman’s The Machinery of Freedom.[1] Friedman’s book is the most detailed and thoroughly argued attempt to sketch the contours of a stateless capitalist anarchy. Contrary to other anarcho-capitalist contributions emphasizing ethics, Friedman’s is more about the economics of capitalist anarchy. It uses economic reasoning and principles to show that it’s highly likely that a stateless society where individuals are free to contract with anyone on any aspect would be a better place to live than our current society. That doesn’t mean there is no value judgment in the analysis, but at least the argument is based on clear assumptions and a (mostly) transparent economic reasoning that you can discuss. Whatever you think about libertarianism and anarcho-capitalism, the book is worth reading, if only to understand what economic reasoning can and cannot prove.

The crucial parts of the book are those where Friedman discusses market failures. These are the cases where we can doubt that a capitalist anarchy can work properly, for reasons largely explored by economists. Friedman generally does a good job of showing how market failures can be dealt with without an organization that monopolizes the use of coercion.[2] However, there are two critical issues where Friedman’s argument is far less convincing. The first concerns so-called “rights enforcement agencies” and the fact that, at some point, they may be likely to form a coalition, if not to merge, thus recreating an institution very close to our modern state.[3] The second is related to “national” defense; i.e., how a stateless society could organize the defense of the territory that it is, as a matter of fact, occupying. I find this latter case particularly instructive regarding the conditions for the successful provision of public goods, and I will focus on Friedman’s discussion of it for the rest of this essay.

Chapter 34 of Friedman’s book addresses the issue of national defense. Friedman calls it the “hard problem,” teasing the troubles that will soon emerge. National defense is one of the few ideal-typical public goods, as it displays the two characteristics ascribed to these goods in their purest form. First, national defense typically doesn’t confront the problem of decreasing returns to consumption. Another way to put it is that the good can be provided to an arbitrarily high number of consumers without significant increases in the opportunity cost. This is what economists call the non-rivalry condition. Second, if national defense is provided at all, it will be provided to all individuals who are living in the defended territory. Under the current technological state of affairs, it would be difficult (and extremely costly) to discriminate the provision according to individuals’ monetary contributions or preferences.[4] This is the non-exclusion condition.

The provision of a public good is therefore often the result of joint production. There is no point in individually producing a public good like national defense because there is no market and no market price at which it can be sold. Its production depends on the joint effort of the community, creating a classical “incentive-compatibility” problem. What makes the case of national defense particularly acute is the nature of the externalities this good produces. We might be tempted to say that the problem is the scale at which the good must be produced, and therefore, externalities are generated. This is, however, imprecise. In a recent article, Pablo Paniagua and Veeshan Rayamajhee propose a new taxonomy of externalities along two dimensions.[5] First, the size/scale of externalities, meaning their importance. Second, the cost of assigning, enforcing, and trading property rights on activities/goods generating the externalities. They supply the following figure as an illustration:

Source

From an economic point of view, national defense is particularly problematic because its externalities are not only large, but also almost impossible to internalize within a system of property rights —at least under current technological conditions. There are other externalities with the same feature. Some of them are mostly irrelevant because their size is marginal (e.g., the case of smelly subway passengers in the table above), and in this case, we don’t even bother finding an institutional solution to internalize them (though people taking the subway every day may disagree). But national defense is closer to the northeast side of the table, between municipal policing and a pandemic.

Friedman acknowledges that a market mechanism cannot finance national defense, either through a unanimous contract between landowners or through a single individual purchasing all the lands in the territory. A possible solution would be for the national community to divide into smaller groups to reduce the scale. But the problem of the difficulty of coordinating the efforts over the full territory remains. National defense is a domain where there are strong economies of scale. An efficient and effective defense actually depends on its scale, so presumably you don’t want to reduce it excessively.

“Washington Crossing the Delaware,” by Emanuel Leutze (1851)

The solution that Friedman considers the most seriously is the emergence of a network of local defense organizations voluntarily financed by individuals or even insurance companies, as the latter would benefit from the lower risk conditions fostered by the provision of a national defense. Friedman considers the possibility that defense organizations could be funded by charities or tipping. He even offers his reader a one-page-long digression about the fact that people regularly tip at the restaurant, though they are not forced to do so! In a subsequent edition of the book, written several decades later, he revisits the issue and this time suggests that a libertarian society could organize its defense by turning it into a sort of hobby, like going to paintball every Saturday afternoon!

These last considerations, though they may appear far-fetched —if not absurd— are interesting because they highlight that between economic decisions based on monetary incentives and those based on coercion, there is a realm of voluntary contributions that depends on a sense of community-belonging and the social norms that come with it. This is the major contribution of the so-called Bloomington school, led by Elinor and Vincent Ostrom, to have largely documented the many ways through which externalities can be dealt with by human communities, beyond the binary market/state framework.[6] Anarcho-capitalists like Friedman are happy to acknowledge that, at least in some cases, social organization relies on a broader range of incentives than purely monetary ones.

To conclude, I would like to point out an interesting and counterintuitive implication. A libertarian society would have to give its members a strong sense of community belonging to solve the problem of providing national defense. The problem is that the cases that the Ostroms consider in detail in their work all concern relatively small communities whose members have strong social ties —religious or ethnic, very often. A libertarian society would presumably be open and far from fulfilling the social conditions required for the kind of “communitarian solution” permitting the voluntary provision of public goods, especially at a scale such as national defense. Or, if a successful libertarian society is possible at all, we should expect that evolution (biological and cultural) will probably make it strongly nationalistic. A strong sense of national belonging is indeed probably the only way to trigger the non-market mechanisms of voluntary contributions that Friedman alludes to.

There are infamous examples of libertarian thinkers with quasi-nationalistic views, sometimes bordering on plain racism and supremacist ideology.[7] As far as I can tell, their quasi-nationalism is disconnected from defense considerations. But it may well be that the only capitalist anarchies able to survive over the long run are those that rely on a strong national sentiment that transcends individual interests. Then the question is whether a society built on such a sentiment can resist the attraction of Schmittian-type bellicose politics. It might be another iteration of the libertarian “paradox of politics” that I have discussed before.

---

[1] David D. Friedman, The Machinery of Freedom: Guide to a Radical Capitalism, 3^rd edition (New York, NY: CreateSpace Independent Publishing Platform, 2015).

[2] One of the best points is about externalities and the fact that their sign and importance cannot easily be estimated in a top-down way. Knowledge about externalities needs to be discovered and most often this knowledge is local and disseminated. I plan to come back on this topic in the near future.

[3] Robert Nozick’s case for the minimal state is based on a sophisticated argument to the effect that this is what would happen. See Robert Nozick, Anarchy, State, and Utopia, 2^nd edition (New York: Basic Books, 2013).

[4] This is not to mean that it will never change. Technological progress can change the conditions under which a good is provisioned. With a bit of imagination, we can imagine technological conditions that make it possible to provide a more “individualized” national defense. What makes a good “public” or “private” depends on practices that are themselves determined by technological and institutional conditions.

[5] Pablo Paniagua and Veeshan Rayamajhee, “On the Nature and Structure of Externalities,” Public Choice 201, no. 3 (December 1, 2024): 387–408.

[6] A classic reference is Elinor Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge University Press, 1990).

[7] Some of them are documented in Matt Zwolinski and John Tomasi, The Individualists: Radicals, Reactionaries, and the Struggle for the Soul of Libertarianism (Princeton University Press, 2023). For the discussion of historical cases where these views have materialized, see Quinn Slobodian, Crack-Up Capitalism: Market Radicals and the Dream of a World Without Democracy (New York: Metropolitan Books, 2023).