Property seems such a simple idea: Things belong to people, and the owner of each thing gets to control how it is used. Like many simple things, it becomes more and more complicated the longer you think about it.
In chapter 5, we looked at some of the complications and saw how, in principle, one might determine an efficient set of property rights. In this chapter we extend the argument in two directions: What an owner owns and why some things, property, are owned and some, commons, unowned.
The idea of a thing belonging to a person is fairly clear when the thing is an automobile or a pair of pants. It is less clear when the thing is a piece of land. What rights does my ownership give me? Almost certainly I can farm the land, or build on it, or keep off trespassers. But can I prevent airplanes from flying over it, miners from tunneling under it, neighbors from making loud noises near it? If it is my land, does that mean I can forbid radio stations from broadcasting without my permission, on the theory that if I can pick up the signal, the radio waves must be trespassing on my property?
What I own is not a thing called land but a bundle of rights. Some rights almost always go in the bundle associated with a particular piece of land, such as the right to walk on it and forbid others from doing so. Other rights associated with the land, such as the right to forbid trespassers at various distances above or below it and the right to have the surface stay put instead of sliding into someone else's coal mine, may or may not be found in the same bundle.
chapters 4 and 5 sketched out the economic approach to designing efficient legal rules. It starts with two questions: to whom is a particular right most valuable and, if we do not know that for certain, what initial definition of rights makes it easiest to move the rights to the person to whom they turn out to be most valuable.
When constructing bundles of rights, the first question becomes "which rights belong together?" If I own the right to farm the land, the right to walk on the land is worth more to me than to anyone else, so the two belong in the same bundle. Since it is hard to grow crops if other people are free to tramp through your fields, the right to exclude trespassers probably belongs in the bundle as well. But that depends in part on how the land is going to be used. If it is timber instead of corn, the argument is not so clear. In some legal systems, ownership of land implies only a very limited right to exclude trespassers.
The right to forbid radio waves from passing over my property, on the other hand, is of very little use to me. If every property owner had that right, setting up a radio station would require unanimous consent from every owner within range of the broadcast, making a transfer of the right from the owners to the person to whom it is of most value a prohibitively difficult transaction. It makes more sense to have legal rules in which the right to broadcast on a particular frequency is entirely separate from the ownership of the land over which the broadcast passes. Similar arguments suggest that the right to forbid airliners from trespassing a mile above my fields would be of little value to me, and moving it to those to whom it was of most value would be hard. That right, too, is left out of the bundle.
A more difficult problem arises when a single right is associated with two different, typically adjacent, pieces of land. The right to control sound waves crossing into my property, to forbid you from playing loud music or setting off firecrackers close to my border at three in the morning for example, is valuable to me. But the right to control sound waves crossing out of your property, to make noises that I can hear from my side of the property line, is valuable to you.
[case link]
The same problem arises when two properties are adjacent vertically rather than horizontally. In Pennsylvania, a state constructed largely out of coal, rights to land are made up of three separable estates: the surface estate, the mineral estate, and the support estate. If I own the surface estate and the support estate and you own the mineral estate, you are free to mine the coal under my land but must leave enough of it to support the surface. If my house falls into your mine, you have violated my rights. If, on the other hand, you own both the mineral estate and the support estate, I may own the surface but I have no legal right to have it supported by anything. The support estate is valuable both to the owner of the surface, who want something under his house to hold it up, and to the miner who wants to be able to get out all of the coal. So it makes sense for the law to permit transactions between the owner of the surface and the owner of the mining rights to move the support right to whoever values it most.
The same problem sometimes arises between neighbors who are adjacent horizontally. If I dig too deep a pit on my side of the property line, your land may start to slide into it. Under English common law, a landowner has a right to lateral support, meaning that his neighbor has a duty to continue to provide the support that the adjacent land would receive under natural conditions.
Such conflicts are the subject of the common law of nuisance, the area of law that inspired Coase's work. The analysis of railroads and farmers in chapter 5 was a sketch of how, at least in principle, questions about who owns which rights in the bundle and how they can be enforced would be decided in an efficient legal system. We will return to the subject again in chapter 14.
In chapter 5, after discussing property and liability rules as alternative approaches to enforcing property rights, I briefly mentioned another alternative, controlling property by majority vote. That turns out to be a common, and arguably efficient, rule for controlling the use of fugitive mineral resources: oil and gas.
Consider a group of landowners sitting on top of a large pool of oil. If I drill a well on my property and start pumping, all the oil will eventually belong to me, since as oil flows out from under my land it is replaced by oil flowing in from under yours (Geologists, including the one I am married to, will I hope forgive my oversimplification of the geological facts). My well imposes an externality on you in the form of lost oil; your well, if you drill it, will impose a similar externality on me. It is in each person's interest to drill too many wells and pump them too fast, making all of us worse off. We are in a prisoner's dilemma with multiple prisoners.
One solution is unitization. In some states a sufficiently large majority of the landowners over such a pool—frequently two thirds are required—can vote to unitize the pool. Doing so converts the oil from the private property of whoever owns the land immediately above it to the joint property of all the landowners. The landowners as a group then agree on how the oil is to be extracted and share the resulting income. Conflicts of interest among the landowners are reduced by legal rules requiring equal treatment; a majority group cannot simply vote to pump all the oil and give the income to its members. Similar rules forbid a majority group of stockholders in a corporation from transferring the firm's assets to themselves at the expense of the minority stockholders.
In all of these cases, the relevant legal rules can be thought of as a way of defining and bundling rights so as to achieve the most efficient possible outcome. Whether that approach explains the rules we have or provides reasons why we should have different rules is less clear. Coase's conclusion was that, in the case of the common law of nuisance, the evidence suggested that judges were at least trying to produce something close to efficient rules.
For a noneconomist, the first and most obvious question about private property is why we have such a silly institution. Why not forget selfish notions of thine and mine and let everyone use everything whenever he needs it?
There are two reasons why that does not work. The first is that you and I cannot simultaneously drive the same car to different places, nor can I drive my car very far if your previous use has left it with an empty gas tank and a flat tire. We need some way of deciding who gets to use what when, preferably a way that results in the person to whom something is most valuable getting it. Private property and exchange solve that problem, for reasons sketched in chapter 2. If the use of my property is more valuable to you than to me, you will be willing to offer a price that I will be willing to accept.
The second reason is that most of the things we treat as private property are things that somebody must make, and making things is costly. If making things results in owning them, that gives you a reason to make them. Not only does it provide an incentive, it provides the right incentive: you will make something if and only if its value to whoever values it most, either you or the person you plan to sell it to, is at least as great as the cost of making it. That is the efficient rule.
The puzzle for the noneconomist is why anything is private property. The puzzle for the economist is why anything is not. Having found such an elegant solution to the problem of producing and allocating things, why not apply it universally?
If you think the answer is that we should, consider extending intellectual property law to cover the English language. Words become private property, each belonging to its first user or his heir or assign. Before speaking a sentence you must first license rights to each word.
There would be some advantages to propertizing language. The owner of a word could prevent the sort of overuse by which words are rendered almost meaningless: "nice," for example, or "awful." Perhaps more important, it would provide an incentive for creative neologism. The English language is sadly lacking in gender neutral pronouns, leaving us with the choice of "he-she-or-it" or misusing "they" to produce such ungrammatical barbarisms as "this policy covers the customer even if they drop dead of a heart attack tomorrow." If the first person to invent and popularize a euphonious and intuitive set of gender neutral pronouns were able to collect license fees thereafter from everyone who used them, perhaps the problem would be solved.
There are advantages to propertizing the language, but also very large disadvantages. The transaction costs associated with writing or speaking, in a world of private words, would be very high. I suspect that they would more than outweigh any advantage due to more rapid linguistic innovation. It would be a very quiet world.
The argument is not limited to intellectual property. Consider property rights among primitive peoples Some have private property in land, some do not. Why?
It is tempting to answer that the ones who do not have private property in land are too primitive to have thought of the idea. But that cannot be an adequate answer, because some groups have private property in land for only part of the year. They know of the concept and practice it—part time.
A more plausible answer starts with the observation that, since primitive people probably know more about their conditions than we do, their legal rules may well be efficient ones for their circumstances. Consider land used part of the year for agriculture and part of the year for hunting large animals across. Private property is very useful for agriculture, since there is little point in planting and weeding if other people are free to do the harvesting and eating. Private property in land being hunted over, on the other hand, means stopping at every border to ask permission to cross, while your quarry vanishes into the distance. The sensible rule is for the animal you are hunting to be private property, belonging to the hunter who first spotted it, while the land you are hunting it over is common property. If hunters depend on the rest of the community to bring them word of game, that rule may be efficiently modified to include a claim by others to a share of the kill.
For a less exotic example of land owned in common, consider the various sorts of joint tenancy, common tenancy, tenancy by the entirety, and life interest with reversion that law students encounter in studying property law. All raise the usual problem of common ownership, the risk that each party, in deciding how to use the land, will ignore the interests of the other. Legal doctrines such as the law of waste, which forbids a life tenant from altering property in ways injurious to the interest of the person to whom the property will revert, have evolved as attempts by the common law to deal with that problem.
A more recent example of common property is information on the Internet. While some providers choose to charge for information, many others, myself among them, deliberately give it away to all comers. And the standard procedure for getting Email from computer A to computer G through intermediate computers B-F involves no charge by the owners of the intermediate machines to the owners of the machines whose mail they are forwarding.
Charging for information online, although possible, is costly, with the result that porn sites do it and most of the rest of us don't bother. Instead we rely on indirect ways of getting paid for the information we give away—advertising for commercial sites, the pleasure of spreading our ideas and showing off pictures of ourselves and our children for noncommercial sites. That works because the cost of distributing information, measured per user, is small enough to be covered by such indirect methods. Charging for forwarding the packets of information into which Email dissolves itself when it leaves the host computer would require enormous numbers of tiny charges, currently an unreasonably expensive procedure (although that may be changing). Instead, the ground rule for host computers is "I'll forward your packets if you will forward mine."
There are many examples of firms that routinely give things away. An all you can eat restaurant charges for entry, but once inside additional food is free. Most Internet service providers follow the same policy—a fixed monthly fee for unlimited service. Whether or not we can explain the failure of primitive people to maintain property rights in land by simple ignorance, we cannot explain AOL's pricing policy that way. A more plausible explanation is that such firms are balancing the inefficiency due to overuse, an additional serving of food that is worth ten cents to you and costs the restaurant twenty cents to produce, against the cost savings from not having to restrict and monitor use.
These examples suggest one reason why private property is sometimes not worth having: the cost of the transactions necessary to move it from one person to another. That is the obvious explanation of why we do not want English words to be private property. A second is the cost and difficulty of defining and recognizing boundaries. Which brings me to the story of ...
Stack Island in the Mississippi belonged to someone. Over a period of many years, the river's current eroded the upstream end of Stack Island and deposited sediment at the downstream end, with the result that Stack Island gradually drifted downstream.
Some distance below Stack Island, the west bank of the river belonged to someone else, along with all islands in the section of river east of his property. After a very long time, one of them was Stack Island. Who owned it?
[case link]
The resulting law case was settled on grounds of adverse possession; the court, by holding that the owner of the coastal strip had waited too long to assert his rights, avoided having to rule on whether he had any rights to assert. What I like about the case is that it illuminates one of the unstated assumptions of ordinary property law—that boundaries stay where they are put, permitting us to define what we own in an unambiguous way. The property rights that came int o conflict in that case were defined in a perfectly clear and sensible fashion as long as islands stayed put. In a world where Stack Island was the norm instead of the exception, where the physical boundaries defining property claims shifted around in an unanticipated and ambiguous fashion, defining those boundaries and arbitrating the resulting disputes might be a significant cost of maintaining a system of private property.
Consider the following description of land law in the Sudan:
"You cannot understand a Nile land case without understanding how the river behaves. As it rises and falls in its annual cycle, fertile land in the riverbed is arable for seven or eight months, then disappears again beneath the water. One year a particular tract may fail to reappear and the owner loses his land. Five years later, land appears again in that same place. Does the old owner still have rights to it? If he is dead, who does have rights to it? Perhaps an island has vanished under the flood. It reappears a quarter of mile downstream in slightly different form. Does the owner of the lost island own the new one?" ... The banks of the Nile also occasionally tend to swing back and forth, and, according to the custom that prevails in most places, as the riverbanks move, so does riverbank land. Everyone's property swings with the river. Even people whose land is some distance from the water are affected when the channel takes a turn in their direction. Properties near and far move like connected pieces of armor, in concert with the unpredictable water."
John McPhee, A Roomful of Hovings, pp. 162-163, in part quoting Carroll W. Brewster.
That is a topic we will return to in the next chapter. In the law of real property, fuzzy boundaries are the exception. In patent law, which tries to draw boundaries around ideas, they are the norm.
The date is ten or eleven thousand B.C. You are a member of a primitive tribe that farms its land in common. Farming land in common is a pain; you spend almost as much time watching each other and arguing about who is or is not doing his share as you do scratching the ground with pointed sticks and pulling weeds.
You are primitive but not stupid—stupid people don't live long in your environment—and it has occurred to several of you that the problem would disappear if you converted the common land to private property. Each person would farm his own land; if your neighbor chose not to work very hard, it would be he and his children, not you and yours, that would go hungry.
There is a problem with this solution, one you observed when a neighboring tribe tried it a few years back. Private property does not enforce itself. Someone has to make sure that the lazy neighbor doesn't solve his food shortage at your expense. Instead of spending your days making sure your fellow tribesmen are working hard harvesting the common fields, you will have to spend your nights making sure they are not working hard harvesting your fields. All things considered, you conclude that communal farming is the least bad solution.
Agricultural land continues to be treated as a commons for another thousand years, until somebody makes a radical technological innovation: the domestication of the dog. Dogs, being territorial animals, can be taught to identify their owner's property as their territory and respond appropriately to trespassers. Now you can convert to private property in agricultural land and sleep soundly. Think of it as the bionic burglar alarm.
I do not remember who first proposed this explanation for the rise of private property in land, and I do not know enough anthropology and prehistory to judge how plausible it is. But, "Si non e vero, e ben trovato"—if it isn't true, it ought to be. Just as the story of the floating island conveniently symbolizes the problem of defining boundaries, so the tale of the bionic burglar alarm, of how we owe the crucial step in human civilization to the dogs, symbolizes the problem of enforcing those boundaries and the fact that if enforcing them is too costly it may not be worth doing. That subject too will be revisited in the next chapter, where we see how changes in the technology of copying have taken us from a world where publishers successfully enforced rights that, legally speaking, they did not have to a world where they are unable to enforce rights that, legally speaking, they do have.
So far we have discussed problems associated with defining property and enforcing property rights. Additional problems arise with the initial creation of property rights, the process that determines who owns what.
Like many things we will discuss, the incentive to acquire property rights is sometimes a bug and sometimes a feature. In the case of things whose value is created by human efforts, it is a feature; the fact that you get ownership of a car by building one provides people an incentive to build cars.
In the case of land, on the other hand, most of which is not created by human effort, the incentives associated with the creation of property rights may have less attractive consequences. Consider the economic implications of homesteading, the mechanism by which large parts of the United States became private property. Under the Homestead Act of 1862, a settler obtained ownership of a quarter section of land, 160 acres, by farming it for a fixed number of years and meeting a variety of other requirements such as successfully growing fruit trees on it.
The year is 1862; the piece of land we are considering is beyond the margin of settlement, too far from railroads, feed stores, and other people to be cultivated at a profit. As time passes and settlement expands, that situation changes. The efficient rule would be to start farming the land the first year that doing so becomes profitable, say 1890.
But if you set out to homestead the land in 1890, you will get an unpleasant surprise; someone else is already there. Homesteading land that it is already profitable to farm is an attractive proposition, since you not only make money in the process, you also end up with valuable real estate. When valuable rights are being given away for free, there is no shortage of takers. If you want to get the land you will have to come early. By farming it at a loss for a few years, you can acquire the right to farm it thereafter at a profit.
How early will you have to come? To make things simple, assume the value of the land in 1890 is going to be twenty thousand dollars, representing the present value of the profit that can be made by farming it from then on. Further assume that the loss from farming it earlier than that is a thousand dollars a year. If you try to homestead it in 1880, you again find the land already taken. Someone who homesteads in 1880 pays ten thousand dollars in losses for twenty thousand dollars in real estate—not as good as getting it for free, but still an attractive deal. Working through the logic of the argument, we conclude that the land will be claimed about 1870, just early enough so that the losses in the early years balance the later gains. It follows that the effect of the Homestead Act was to wipe out, in costs of premature farming, a large part of the land value of the United States.
If you think this argument looks familiar, you are right. The logic of homesteading is the same as the logic of theft, discussed back in chapter 3; both are examples of rent seeking. The thief spends resources in time and effort in order that he rather than you will end up possessing your television set. The homesteader spends resources in order that he, rather than the next possible claimant, will end up possessing a particular piece of land. In each case, actors are competing with each other to get something valuable that already exists and, in the process, spending resources roughly equal to the value of what they are getting.
In this case as in that, a full analysis would be somewhat more complicated. Some people are better at settling land than others, just as some are better at stealing than others. In equilibrium, the marginal homesteader or the marginal thief just breaks even. The particularly skilled homesteader or thief makes a profit. It follows that homesteading did not dissipate all of the value of the land, just a large part of it.
Terry Andersen and P.J. Hill, the economists whose argument I have just summarized, concluded that what the government should have done was to auction off the land. That way the settlers' rent seeking expenditure would have gone to pay the federal government's bills instead of being wasted growing crops not worth the cost of growing and trying to keep fruit trees alive in places nature intended for prairie. What they did not explain was why the government, which owned the land and, then as now, always had uses for money, didn't do it that way.
The answer is that they tried. From shortly after the Revolution, there were repeated attempts by both state and federal governments to raise money by auctioning off land from the public domain. But there was a problem.
The federal government has announced that a particular territory is about to be opened to the public, with land auctions held locally to allow buyers to inspect the land they are bidding on. You, as the representative of a real estate syndicate in Boston, head out to the Wild West, somewhere in what will be eastern Ohio, to buy some land.
To your surprise you discover that, although the land has not yet been opened for settlement, it is already settled. Most of the people at the auction, held ninety miles from nowhere, are squatters, illegal settlers bidding for the land they are already farming, rough tough frontiersmen with knives in their boots and flintlock rifles leaned up against the wall. They make it very clear to you that whatever may happen back east, out here bidding against a settler for "his land" just isn't done—and leave it to you to figure out for yourself what may happen to you if you do it. Each settler ends up buying a quarter section of land, the best land he could find when he arrived and settled a few years earlier, at the minimum auction price, because nobody else is bidding for "his" land. All that is left for you is the land that nobody else wants.
Sometimes the process was even simpler. When the land was about to be opened for settlement, the people already living there would petition Congress, arguing that as brave frontiersmen they were entitled to special treatment. The government could raise revenue auctioning off other land to other people, but they were entitled to the land they settled at the minimum legal price. Quite often it worked. Settlers, even squatters, are also voters.
As these examples, somewhat stylized from the real history, suggest, land sales in law often turned into homesteading in practice, which may be why the federal government eventually gave up and wrote homesteading into the law.
[Article]
As a final note in defense of homesteading, there is the possibility that the squatters were right and the economists were wrong. The squatters, after all, really were brave frontiersmen. One of the things they did by settling was help maintain the United States claim to the land against both native inhabitants and rival governments to north and south. One way of defending homesteading is to argue that giving homesteaders a claim to land they settle, like giving auto companies a claim to the cars they produce, actually produced a useful incentive—not to create the land but to go out and defend it. From the standpoint of the government that gave out the land, homesteading was a productive activity. From a broader standpoint, one that includes in the calculation of economic efficiency the interests of Canadians, Mexicans, and American Indians, it was still rent seeking, but on a larger scale.
Whether or not that is a plausible defense for what happened, the logic of inefficient homesteading, the Hill and Andersen story of how the government burned up the value of the public lands, provides one more reason why making things into property may not always be a good idea. We will return to that subject in the next chapter, where we consider intellectual property law as a way of propertizing previously unowned intellectual property and see under what circumstances the result is inefficient rent seeking.
So far we have been looking at costs of treating things as property, problems of defining, enforcing and transacting over property rights. Whether it is worth paying those costs depends on how large the benefits are that flow from the incentives for efficient production and allocation created by private property.
Suppose you believe that even without copyright, plenty of novels would be written. Great writers, you might argue, are motivated not by money but by the desire for fame or love for their art. As evidence, you point out that much of the world's great literature, including the work of Homer, Dante, and Shakespeare, predates copyright law.
If this argument is right, then the incentive to produce property in order to own it is not very important in the case of literature, which weakens the argument for copyright law. The second advantage of property rights is better allocation of existing goods, and since my reading a novel does not keep you from reading the same novel, allocation is not a problem in the case of literature. We cannot simultaneously drive the same car to our different homes but we can simultaneously read (different copies of) the same novel. And what copyright law propertizes is the novel, the sequence of words not the physical book. The book would be private property even without copyright law.
To put the argument in slightly more technical language, the benefit from an additional incentive to produce depends on supply elasticity, on how sensitive the amount produced is to the price the producer gets for producing it. If supply is very inelastic, then incentives don't matter very much, and you get almost as many novels without copyright law as with it. If supply is very elastic, on the other hand, abolishing copyright law would result in a drastic reduction in the supply of new literature.
A similar argument applies on the demand side. Consider an all you can eat restaurant. If it serves salad, the fact that additional servings are free will result in only a small inefficiency. Additional servings cost the restaurant twenty cents apiece but cost me nothing; underpricing them by twenty cents causes only a modest increase in how much I eat. And my extra consumption, although inefficient, is not very inefficient; in the worst case I am consuming a serving of salad that is worth about a penny to me and costs the restaurant twenty cents, for a net efficiency loss of nineteen cents.
Contrast to that an all you can eat sushi bar—in my experience, a much rarer institution. Making sushi is a skilled job and the individual pieces are, as a result, quite expensive. Dropping the price I pay for additional servings from three dollars to zero would result in my leaving inefficiently full. And the last sushi I ate before finally surrendering to the geometric limits of my stomach would represent an efficiency loss of about $2.99.
So in deciding whether it is worth making something property, you must consider two different questions. How large, for this particular sort of something, are the costs associated with treating it as property rather than commons, how easy is it to define boundaries, enforce rights, transact? And how large, for this particular sort of something, are the benefits of treating it as property? How sensitive are producers and consumers to the perverse incentive effects of a zero price?
All property consists of bundles of rights with regard to things. But property in land, known in law as real property, has a special characteristic—to a considerable extent, owners are free to rebundle it. I can sell my neighbor an easement, such as permission to cross my property. Once I have done so, the easement is not merely a contract between him and me but, like other property, a right good against the world. If I sell the property to someone else, the easement binds him too—even though he never agreed to it and may not have known it existed. This feature of property law makes it easier for owners of adjacent pieces of property to coordinate their activity. Having purchased an easement to cross your land, I can build a new house without worrying that a future owner might withdraw his permission, leaving me stranded.
Suppose, on the other hand, that I sell you a car. As a member of a particularly extreme sect of Orthodox Judaism, I believe that not only should I not work on the Sabbath, my car should not work on the Sabbath either. So one of the conditions of the sale, to which you agree, is that, from the time the sun sets Friday to the time the sun sets Saturday, the car may not be driven.
That condition is a contract, enforceable against you. But if you sell the car to someone else, it is not enforceable against her. I might have provided in the original contract that you would only sell to a purchaser who accepted the agreement, but that term would bind you, not the new buyer. I can sue you for breach of contract but I cannot stop her from taking a Saturday drive.
There is a simple and obvious reason for this difference between real property and everything else. Ownership of land is controlled by an elaborate recording system, involving title deeds, land registries, and the like. Restrictions such as easements that run with the land are part of that system, so a diligent buyer will in fact know about the easement. A diligent buyer of a car will know if the car is encumbered by loans, because that encumbrance is included in our system for registering title, but that is all he will know. And a diligent buyer of most other goods will not know even that much.
To put the same argument in the language of an earlier part of this chapter, Stack Island is the exception not the rule. Most property in land is easy to draw lines around. Not only is it easy to draw lines marking the physical boundaries of a piece of land, it is also, thanks to the registration system, fairly easy to draw boundaries around the bundle of rights that the owner has with regard to that piece of land—boundaries that other people can, if they wish, observe. Thus it is practical to allow a considerable degree of rebundling. The argument in favor of doing so is implied by several of the arrows on chapter 5's spaghetti diagram. Permission to throw sparks is an easement.
Property in other things is still good against the world but it comes in a one size fits all version. You either own something or you don't. The purchaser of stolen property does not, as a general rule, have good title, although there are exceptions. But the purchaser of property whose use is encumbered by contract, for example a contract not to drive the property on Saturday, does have good title. Property title is good against the world; contractual obligations are good against the person who signed the contract.
If our willingness to permit rebundling of real property is due to a recording system that makes it possible for third party purchasers to find out what they are getting, one implication is that the ability to rebundle should be limited by the ability of third parties to find out about it. Roughly speaking, it is. As a general rule, an easement is binding on later purchasers only if its existence can be deduced, or at least suspected, from inspecting the land, or if a prudent search of the relevant records would reveal it.
"A servitude ... is invalid if the restraint is unreasonable. Reasonableness is determined by weighing the utility of the restraint against the injurious consequences of enforcing the restraint."Restatement (third) of Property, Servitudes, §3.4, 1991
Property rights can be rebundled, making it possible to create rights associated with land that are good against future owners. Such rights include not only easements, such as the right to cross land, but also licenses, such as the right to use land—to hunt on it or mine under it. They also include covenants running with the land, such as an agreement that plots in a residential development will be used only for residential, not commercial, purposes—the private equivalent of zoning regulations. An elaborate body of law determines under what circumstances such agreements are enforceable against whom.
One interesting feature of that body of laws is that, although the details vary over time and from jurisdiction to jurisdiction, it is generally easier to pass a benefit to future buyers than a burden. Suppose I have made a legally binding covenant with my neighbor, agreeing not to build a factory on my land near his residence. It is more likely that my selling my property will extinguish that covenant, leaving him with no recourse against the new owner, than that his selling his property will extinguish the covenant, leaving the new owner with no recourse against me.
There is an economic reason for that pattern. The right to rebundle makes sense because of the ability of buyers to know what they are buying. If you are selling land with a benefit attached—a right not to have your neighbor do things that would reduce the value of the land—you have every reason to tell the purchaser about it. If, on the other hand, I am selling land with a burden attached—an obligation not to use land in ways in which a new owner might want to use it—I have an incentive not to tell the purchaser. Hence the legal requirements that define, roughly speaking, how obvious the existence of the covenant must be in order for it to be binding on a new owner, are stricter for passing burdens than for passing benefits.
A second issue that arises when we are rebundling rights to land is just what can be included in the bundle. If there is no restriction at all, then all rights become rebundleable. All I have to do is to merge my ownership of my car with my ownership of a piece of land, making all of the rights associated with both of them sticks in a common bundle of rights. Since that bundle is a set of rights associated with real property, it can be rebundled. So I pull out the right to drive the car six days a week and sell it to you, retaining the right to drive—or not drive—the car on Saturday.
The common law prevented such juggling by requiring that any covenant that ran with the land must involve a burden or benefit that touched or concerned the land. Your Saturday driving of the car I sold you does not touch or concern my land, so that right cannot run with my land.
What if I sold you a piece of my land, along with covenant by which I agreed to supply, and you to buy at a fixed price, water from my well? Access to water may be relevant to the value of your land, so your right to get it concerns the land you bought. But once you have chosen to drill your own well, does my right to be paid for the water you no longer want to buy still run with your land?
[Case link]
Modern courts have to some degree abandoned this doctrine in favor of the rule quoted at the beginning of this section: covenants are enforceable if they are reasonable. Roughly speaking, that means that they are enforceable if the court believes that they are economically efficient and unenforceable otherwise.
Back in chapter 4, I pointed out two different ways in which a court might deal with externalities. It might decide in each case which party was the lowest-cost avoider of the problem, and assign rights accordingly on a case by case basis, or it might establish general rules designed to get problems resolved as efficiently as possible. The same issue arises here. The old approach was to use a general rule, the "touch and concern" doctrine, to help determine whether a covenant was merely a contract between two parties or was part of the ownership of land, and thus to determine what rights could be rebundled. The new approach moves closer to judicial determination, in each case, of whether rebundling is or is not desirable.
One reason for the change may have been that the rule the common law had come up with did not serve the function of a rule very well—it was sufficiently vague so that parties were faced with considerable uncertainty as to whether courts would or would not enforce particular covenants. Whether the new approach will do better remains to be seen.
Anderson, Terry and P.J. Hill, "Privatizing the Commons: An Improvement?" Southern Economic Journal, Vol. 50, No. 2 (October, 1983), pp. 438-450.
The source for my discussion of property rights among primitive people is Bailey, Martin, "The Approximate Optimality of Aboriginal Property Rights."