How do things not created by humans come to be property, to belong to someone? While the problem is commonly discussed in the context of land it is broader than that. Everything humans make starts with materials not made by humans. Iron ore was not made by humans. If nobody owns the iron ore, nobody owns the steel it is made into, nobody owns the car the steel is made into.
The simplest solution is unanimous consent. If there are only eight people in the world and all of them agree that the tree over there belongs to me, this field to you, and other parts of the real world to the other six, nobody can claim his rights are violated when each of us treats his share as his property. Unfortunately, that solution does not scale. There is no practical way of getting unanimous consent from a thousand people, let alone eight billion.
The next simple solution is to conclude that nobody owns anything. I can plant wheat in a field, but since the field is not my property I have no right to keep you from walking over the field and trampling my wheat. I can build a log cabin but since the logs don’t belong to me neither does the cabin, so I have no right to keep you out. It follows that most of the productive things humans do cannot be done without violating the rights of other humans. If people respect other people’s rights most people starve to death.
A less catastrophic approach is offered by geolibertarians, followers of the ideas of Henry George. Site value, the value of unimproved land, was created by nobody so nobody owns it, making it the common property of mankind. Collect all the rent on the site value of land, use it to pay for all proper government activities, distribute what is left evenly to the population as a social dividend.
There are practical problems with measuring site value but also a moral problem. Not only did I not create the land, we did not create it either. How do we, presumably a democratic government, get the right to exclude people from land we did not create? If we do not have that right we cannot transfer it to a private owner of the land even if he agrees to pay us the rent on its unimproved value so when he uses force to keep other people off his land he is violating their rights. We are back where we started.
A different approach was offered by John Locke:
Though the earth, and all inferior creatures, be common to all men, yet every man has a property in his own person: this no body has any right to but himself. The labour of his body, and the work of his hands, we may say, are properly his. Whatsoever then he removes out of the state that nature hath provided, and left it in, he hath mixed his labour with, and joined to it something that is his own, and thereby makes it his property. It being by him removed from the common state nature hath placed it in, it hath by this labour something annexed to it, that excludes the common right of other men: for this labour being the unquestionable property of the labourer, no man but he can have a right to what that is once joined to, at least where there is enough, and as good, left in common for others. (John Locke, Second Treatise of Government, Sect. 27)
To justify Locke’s argument in different terms, consider that the value added by my labor belongs to me, the rest of the value of the land to nobody. If I use the land I am violating nobody’s rights. If you use the land you are violating my rights to my labor. So only I, or someone with my permission, can use the land without violating rights.
Robert Nozick, criticizing this approach, wrote:
If I own a can of tomato juice and spill it in the sea so that its molecules … mingle evenly throughout the sea, do I thereby come to own the sea, or have I foolishly dissipated my tomato juice?
A possible response is that when I use the sea I disturb a few molecules of his tomato juice1 but that does him no injury since he is not getting any value from having them undisturbed. No harm, no foul. Consider as an analogous case that when I turn on a light in my house the fact that you can see it from your house means that some of my photons are trespassing on your land. We do not usually conclude that I cannot turn on my light without the permission of all my neighbors.
A more serious problem is raised by the final phrase of what Locke wrote: “at least where there is enough, and as good, left in common for others.” The implicit argument is that if mixing my labor with the land leaves less for you I have harmed you.
There are two possible ways to interpret the Lockean proviso. One is that you are harmed if, by propertizing land, I keep you from propertizing land. The other is that you are harmed if, by propertizing land, I reduce your ability to use land as a commons — gather dead wood, pick berries, pasture your pigs on the acorns dropped by wild oak trees.
The first interpretation seems to make a better fit with Locke’s language:
… He that had as good left for his improvement, as was already taken up, needed not complain, ought not to meddle with what was already improved by another's labour: (Sect. 34)
… so that it was impossible for any man, this way, to intrench upon the right of another, or acquire to himself a property, to the prejudice of his neighbour, who would still have room for as good, and as large a possession (after the other had taken out his) as before it was appropriated. (Sect. 36, italics mine)
We will eventually run out of good land. Whoever appropriates the last chunk is violating the Lockean proviso, hence not entitled to convert it to private property. It follows that whoever takes the next to last chunk is preventing the next person from taking the last chunk, so he is violating the proviso and cannot convert that chunk of common land to private property. So whoever takes the chunk before that … . The whole series unravels, the logic of mathematical induction applied to political philosophy, and the exception swallows the rule — nobody can convert land from commons to private property. That problem does not seem to have occurred to Locke, possibly because he was living in a less densely populated world than we are.
If we choose instead the second interpretation, the result is not so bad. There is a practical limit to how much unowned land one can make use of, especially if some of the other people who might compete with you for use of the commons now have their own land to use instead, so at least the early stages of the conversion of commons to private property should leave “enough, and as good” in commons for others. But at some point, almost certainly well short of things as they are in the modern world, that breaks down and propertizing of land should cease.
One way out of this problem is to take the requirement less literally. The conversion of land to private property vastly increases its productivity — by Locke’s estimate a hundred or a thousand fold. Lots more stuff being produced is a benefit even to those who do not own any land, since they can get much more by exchanging their labor for the output of other people’s property than they could have gotten by applying their labor to the now vanished commons. On this interpretation, the conversion of commons to property by mixing in labor is legitimate as long as nobody is worse off than he would be if it had all remained in commons. To fit this with what Locke actually wrote one can note that “at least where there is enough, and as good, left in common for others” implies that if the condition is met appropriation is legitimate but not that there are no other circumstances in which it is.
That approach to satisfying the Lockean proviso suggests that some amount of income redistribution might be justified as a damage payment owed to make everyone in a world of property were at least as well off as they would be in a world where all land was unowned commons. But the amount justified would be very low — a world without private property would be very poor — and most of it would, from the standpoint of people who support redistribution, go to the wrong people. The usual arguments for redistribution, egalitarian or utilitarian, imply that the people you want to help are those most in need. But in a world without property a random very poor person would be even poorer, someone blind or crippled probably dead. If damages are owed they are owed to those particularly well suited to survive in a world of commons. There is no reason to expect them to be poor.
A newer and more interesting approach to the problem, one that yields a somewhat better argument for redistribution, is due to Baruch Brody. He held that since nobody produced the land nobody has a right to exclude others from it, hence the exclusion needed to treat land as private property is a right violation. But, he argued, violating rights is justified if it produces a sufficiently large benefit, as in this case it does. Libertarians uncomfortable with that claim my want to consider Bill Bradford’s example. You carelessly fall off of your tenth floor balcony. By good luck, you catch hold of the flagpole of the balcony immediately below. While you are working your way hand over hand to the balcony the owner of the apartment comes out and informs you that you do not have his permission to use his flagpole. Do you let go? Should you?
If you violate someone’s rights, even with good reason, you owe him compensation. If the required compensation is only enough to make him whole, as well off as if you had not violated his rights, we are back with the same conclusion as in my final version of the Lockean proviso. A society with private property in land is so much richer than one where all land is a commons that practically everyone, including people who do not own land, is better off, hence the compensation occurs automatically.
But, Brody argues, making the victim whole is not adequate compensation. By assumption, the violation of his rights produced a very large benefit. Why is the violator entitled to get all of it, net of whatever small amount is needed to make up for any damage to the victim?
Consider, as an analogous case, the seizure of land by eminent domain. You have land that you would be willing to sell for a hundred thousand dollars if that was the most you could get for it. I am putting together land for a shopping mall, would like to include yours, would be willing to pay up to five hundred thousand for it if necessary. Fortunately for me, but not for you, I have friends in high places. I get the local government to seize the land for me under eminent domain, giving you the “fair compensation” of a hundred thousand.
I have violated your rights in order to produce a net benefit of four hundred thousand dollars, the difference between the value of the land to me and to you. Does the fact that you are no worse off than if I had not bought the land mean that I have satisfied my obligations to you? Am I entitled to get all of the benefit produced by my violation of your rights?
Brody never says what share of the benefit the victim is entitled to but his argument produces a justification, in a moral system that gives some but not infinite weight to rights, for income redistribution. All land owners are violating the rights of everyone, including other land owners, to go on their property. Hence each land owner owes some unspecified amount to everyone who could trespass on his property. While Brody does not carry the argument quite that far, it looks like an argument for Henry George’s social dividend funded by a tax on the site value of land, perhaps a fifty percent tax on the theory that the rights violator should get half of the gain made possible by the violation. We are back with a modified version of the geolibertarian conclusion.
This approach has one feature not noted by Brody that other arguments for redistribution lack — it provides a justification for making the redistribution local rather than global. Someone located in India cannot get to my property in California, at least could not in a world without private property in unproduced resources hence also without airplanes or ocean liners, so I am not violating his rights by keeping him out of my orchard. Someone living a few miles from me can, so I am.
After encountering Brody’s argument at a conference where he presented it — my comments here are based on a later and somewhat modified version2 — I came up with my own solution to the problem, along more nearly Lockean lines but in a form that avoided both the Lockean proviso and Brody’s rights violation and did not imply any need for income transfers. I do not, for several reasons, find it entirely satisfactory, but here it is, copied from Chapter 57 of the third edition of The Machinery of Freedom, slightly revised for style.
Locke as Revised by Friedman
My revised version of Locke’s argument starts with the observation that while the land may, morally speaking, be a commons, I myself am private property — mine. You and I have the same right to stand on any particular piece of land but your right to stand on the land I am currently standing on does not include a right to shove me off, since that is a violation of my property right in my body. That makes the land I am standing on, in a limited and very temporary sense, mine. Similarly, if you happen to have a rifle, you are entitled to use it for target practice on the commons—except in my direction, since you do not have a right to shoot me.
Following a little further along the same lines, imagine that I find a nice piece of land, pull out the weeds, dig it up, and plant wheat. A few weeks later the land has neat rows of little wheat sprouts. I point them out to you and explain that, while you still have a right to go on the land, you do not have a right to crush my wheat plants, since they were produced with my labor and so belong to me. If there is no practical way in which you can go on the land without damaging my property I have established de facto ownership of that piece of land, at least until my wheat is harvested.
I want to make my effective ownership permanent but wheat only occupies the land part of the year. No problem. I build a fence around my wheat field. The fence may not be sufficient to keep you out but if not it is probably also not strong enough to survive undamaged your climbing over it. You have a right to be on the land but not a right to damage my fence — mine because my labor produced it.
We now have a version of Lockean appropriation that solves the usual problems. When I mix my labor with the land I do not acquire any right to the land, retain ownership of only my labor and what that labor produced. You are free to use the land in any way that does not injure my property, to dig underneath it for ore, fly your hawk over it to hunt pigeons. But it may well be, depending on what I have done with it, that most of the ways you would want to use it are now barred to you because they would injure my property now mixed with our land. Whether there is as much and as good land available for you is irrelevant since I am not claiming ownership of any land, merely of what I have mixed with it.
This approach to justifying property rights has its problems. In order to maintain ownership of my land I have to be careful to do things with it that make it impractical for anyone else to use it without damaging my property. I have to be particularly careful not to let anyone else do anything with it such that my future use would damage his property, since that could make the land effectively his. That is a cost that could be avoided if the land was my property in a more conventional sense.
Speaking as an economist, I find the rules implied by this argument to be inefficient ones. But they at least provide a justification for enforcing a form of property rights in land that is consistent with the libertarian view of rights.
One point that did not occur to me when I wrote the chapter was that my approach provides support for the attitude towards land ownership held by some 19th century individualist anarchists and some of their left-libertarian successors, that you can only own land you are using. My “ownership” is not limited to land I am myself using, since I could cultivate the land with hired labor or rent my fence and wheat plants to someone else. But it is limited to land that is sufficiently in use so that someone else using it would violate my property rights in what I had on it.
One could move my approach closer to Locke’s by arguing that I have a property not only in things on the land but in the condition of the land, that if I have improved it by digging out boulders or clearing jungle anything you do that takes advantage of that, such as planting wheat in a field only suitable for the purpose due to my efforts, is violating my property right in my improvements — Locke’s labor mixed with the land. But I would still not have a right to exclude people as long as they did not damage my property, literal or metaphorical. That makes some economic sense for reasons I discuss in Law’s Order3 and corresponds pretty well to the law of some countries that permits harmless trespass.
So far my discussion has been in the context of a rights based moral philosophy, although Brody compromises that to the extent of claiming that rights violations are under some circumstances justified. If one adopts instead a utilitarian, or at least consequentialist, philosophy, the problem of justifying private ownership of unproduced resources disappears, since the consequences of keeping all such resources unowned are catastrophically bad. Details of the implications of that approach — what should be property, who should own it, what bundle of rights should ownership imply — would shift us from libertarian moral philosophy to the economic analysis of law.4
That should not provide much comfort to libertarians unhappy with the redistributionist implications of Brody’s argument and some variants of Locke’s, however, since they then have to deal with redistributionist implications of utilitarianism and similar consequentialist approaches.
The volume of the ocean is 1.3x1018 cubic meters. Tomato juice is mostly water, with a molecular weight of 18, so a can of tomato juice contains about ten moles of soup, or 6x1024 molecules, mostly water. If the juice is evenly spread through the ocean, it should have about 5,000,000 molecules of Nozick’s tomato juice per cubic meter or five per cubic centimeter, making it almost certain that if I do anything in the ocean I disturb at least a few of them.
Baruch Brody, “Redistribution Without Egalitarianism,” in Peter Vallentyne and Hillel Steiner, Left-Libertarianism and Its Critics, pp. 31-47.
Chapter 10. The relevant discussion follows the subhead “Bundling Rights: What an Owner Owns.”
Some of this can be found in Chapter 10 of my Law’s Order.
“the land may, morally speaking, be a commons,”
Common property is not private property, but it is still property. So if property needs a justification, common property needs one as badly as private property does.
A commons is distinct from private property, but both are subject to the same objection. Both of these, and any other imaginable form of property, depend on someone needing consent from someone else to use something, on exclusion. If we try to eliminate this role, and try to regulate use by principle, then whoever disputes an improper use or interprets cloudy cases has the role of owner by default. Consenting to some uses, and disputing and excluding others is what owners do. When someone says “you can’t use that,” they do so either as the owner or as the agent of the owner.
Locke's proviso is an obvious form of exclusion applicable to the commons, but even if we abandon that, others will remain. I can’t build a house on the commons, even if more land remains and is as good. I can’t cultivate crops there and exclude others from harvesting them. My use is excluded if it is the wrong kind, or if I am. Different and perhaps fewer exclusions, but exclusions still. If I attempt to ignore the custom, others can dispute me in court, or in arbitration before the village headman, and win. Or if they lose, it has become private property, hasn’t it?
The solution seems trivial. If something is truly unowned, there is no one with the right to dispute its use. If only one person wants to use something, no one will dispute it. If multiple persons want to use an “unowned” thing, or if one wishes to prevent the other's use, this gives cause for a dispute. The proper owner should win the dispute by showing evidence of ownership, a reason that this use should or should not be excluded.
What counts as evidence of ownership? It is not obvious that different cultures must use the same standards. Libertarians are fond of first possession or first use, and that certainly works. But so long as everyone in a region knows what counts locally as evidence of ownership, there is no obvious reason not to use something else. (Or perhaps the convoluted discussion this has inspired is a reason to think that is the best approach.)
On the other hand, almost no land ownership can be found that is not based on conquest. This doesn’t make private property less useful, but casts some shade on its historical relationship to justice. War not only kills people, but destroys the social basis of custom. It is an injustice, but one of a different scale. Theft is a dispute between persons, with the customary basis of property taken as static. War is a dispute between communities, and unjustly changes the entire basis of property.
On a practical level, an ownership dispute can be settled without reference to original acquisition, if all parties at least agree on some more recent relevant fact of ownership. If farmers Smith and Jones are disputing over the use of a gully, it doesn’t matter whether it once belonged to the the king of England, of Spain, or an indigenous tribe, so long as both Smith and Jones both stipulate that 100 years ago it was definitely owned by farmer Williams. If their dispute is about more recent events, Williams can provide their foundation.
What really is the basis of the criticism of private property? Philosophers love abstract principles. Property is socially significant, but hard to shape into a neat philosophy, either deontological or consequential. It seems too arbitrary.
By coincidence I found myself getting to a digression relating to Georgism on Matt Yglesias's substack yesterday and found myself facing the following conundrum, which I wonder if Dr. Friedman or anyone else can opine on:
Specifically, the nominal goal of Georgism is to tax the unimproved value of land ("site value" as used in this post) and leave improvements untaxed. It seems to me, however, that this is conceptually almost without content: the site value is generally agreed to reflect local demand conditions rather than being solely intrinsic to the site in the state of nature. However, the site value of land is clearly also a function of its capacity for improvement: land that a hotel can be built on is worth more than similar land that's prone to subsidence, and technology allowing denser, higher, or cheaper construction raises the value of any individual parcel because it can be more intensively exploited.
This, in turn, indicates that the site value (and the corollary tax basis) should reflect the *as-improved potential* of the land as maximally exploited, not merely, which accords with the general principle that the income flows extractable from real estate are typically the basis on which it is priced.
But if we grant the above, this turns the idea of "not taxing improvements" into essentially a nullity except in rare cares where the site value is demonstrably an internalized value-add rather than the function of latent external demand (e.g., Disneyworld) or in the case that some entity has an exclusive market advantage in their capacity to build cheaply so that the site has higher value to builder A than to builder B, which so far as I know is generally not characteristic of the building trades -- contracting is more of a commodity labor and materials exchange than one that's, say, characterized by ubiquitous patent disputes. In the general case, we expect site value to encompass essentially all of the economic potential of a given parcel, less only the capital costs attributable to building the infrastructure (e.g., an office building, a mine, whatever) that maximally exploits said site value.
I'm not sure that this necessarily poses an insurmountable pragmatic hurdle to the implementation of Georgism, but it does rather seem like it renders the idea of "tax land but not improvements" as have little, if any, content.
I may well be making an error in the above reasoning but, if so, I'm not sure where it lies.