An economic analysis of civil versus common law property.
Fragmentation is a theme in property theory, but the theory of property itself is deeply fragmented. At first blush, a major fault line in property lies between common and civil law. As is well known, civil law systems tracing back to Roman law place heavy emphasis on ownership (dominion) and are highly grudging in giving in rem effect to lesser interests like leaseholds. By contrast, the common law emphasizes the estate system and its many methods of carving up property, from life estates to defeasible fees and various future interests. And in the common law tradition in a broader sense, the equity courts developed the trust, which is largely unknown in traditional civil law. Sometimes this conventional wisdom about the gulf between common and civil law of property goes so far as to claim that there is no such thing as ownership in the common law. (1) Feudalism lives!
This stark cleavage between common and civil law has taken on a new life with the so-called "legal origins" literature, (2) which has influenced the World Bank's pronouncements on development. (3) Supposedly, having a common law rather than a civil law system correlates with economic growth. Different versions of the literature posit different causal mechanisms as lying behind the correlations (to the extent that they have persisted in the face of continued testing and methodological questioning). (4) Despite the favorable attention for their tradition, common law legal theorists have been quite unreceptive to this branch of economic literature, partly because they doubt that the kinds of doctrines that distinguish civil from common law could possibly have real world effects, much less effects on the scale that the legal origins literature purports to find. (5)
How, if at all, is the distinction between civil and common law property important? Life goes on in the two systems in strikingly similar fashion. Putting aside for the moment special features like the trust, ownership under the civil law and fee simple ownership of land in the common law system (and for the most part the respective notions of full ownership of personal property) coincide to a remarkable extent in their basic features: a possessory right to prevent invasions subject to qualifications such as for necessity, and supplemented by duties (for example, for lateral support or to shovel sidewalks). Lesser interests, like leases and easements, despite some differences, bear a close resemblance in the two systems. So is the supposed difference between the two systems non-existent at the functional level, putting labels like "dominion" and "estate" aside?
Upon closer inspection, the fault lines between common law and civil law are more subtle than conventionally thought, although in a sense they are more important and interesting. This Article will identify these more subtle fault lines and offer a transaction cost explanation for them.
One reason the fault lines between common and civil law are both less and more apparent than they should be stems from a failure to make some basic distinctions. First, property's purpose of protecting interests in use differs from the structure it employs to achieve this purpose. Property serves our interests in using things--this is the reason we have property. Other desirable features of property--its promotion of stability, autonomy, investment incentives, fairness, and efficiency--all trace back to this basic interest in the use of things. (6) But property law serves these substantive purposes and the overarching interest in use in an indirect fashion, through a particular structure. To see this, consider the zero transaction cost world of Coase's thought experiment. (7) In that world, we could serve each individual's interest in use vis-a-vis every other individual's potential use interest by specifying the rights and duties (privileges and so forth) that hold pair-wise between all the members of society with respect to the most articulated uses of the smallest fragments of things. This is intractable in our world. So instead, property uses shortcuts and strategies--what we call "structure" here--to achieve an approximation. (8) In an exclusion strategy, property law delineates lumpy things and defines rights to them using crude proxies of boundary invasion and touching: this roughly corresponds to trespass and conversion (keep off or don't touch without permission). For certain important conflicts the law uses governance strategies to mediate them--think devices such as nuisance or covenants designed to prevent odors or excessive building height. The point is that, in our world, the structure based on exclusion and governance is not as transparent to the use interests of property as it could be if transaction costs--and delineation costs in particular--were lower.
Both property's use interests and structure can be distinguished from what we will call the style of property delineation. The same use interests, and even the same structures, can be implemented in a number of ways. To take a familiar example, we can achieve roughly the same result by carving out an easement as a right to use from full ownership on the one hand or endowing contracts (covenants) with the ability to bind successors and third parties on the other. The result is the same--a use right in the owner of the dominant tenement and a corresponding duty in the owner of the servient tenement--but the route there is different.
A similar contrast of top-down versus bottom-up routes can be seen on a far grander scale in civil versus common law property, when they employ different styles of delineating property. As a first approximation, civil law starts with a strong notion of full ownership--dominion--and then grudgingly carves lesser interests out of it (iura in re aliena). By contrast, the common law defines estates from the bottom up: they are carved out of larger estates, but their definition in terms of length and various features is not seen as tightly bound up with full ownership. Technically, under the full feudal interpretation of the common law, only the monarch has full ownership. Title too is relative, such that A can have better title than B and C, and B can have better title than C. The basic contents of these carved out interests are roughly similar in common and civil law when it comes to the types of discretion over uses that it gives the holder of the interest. And property law in the two systems affords similar contours of protection in terms of possessory actions and trespass. But the method of getting there--the style--is different.
At this point one might ask whether style is trivial. (9) We say no, for several reasons. First, from the internal point of view, what we are calling style closely corresponds to the theory behind the two systems, and is very real to participants. Thus, style has a certain reality from an internalist perspective. The internal perspective is not one to which law and economics has paid much attention, but our second goal is to show that style is amenable to economic analysis. Third and finally, style does matter at least around the edges, and our economic analysis can explain how style matters. In certain pockets of property law, even a slight push from style can make a difference, particularly in areas on the borderline between property and contract such as landlord-tenant. Moreover, style interacts with use interests and structure in predictable ways.
Returning to civil versus common law property, we take as our starting point the growing attention to how civil and common law are more congruent in their major structures for functional reasons than traditional theory would have it. (10) We argue that civil and common law property systems have been perceived to be separated by an unbridgeable gap, because the two traditions have each failed to distinguish sufficiently between property's structure and its style. As long as property responds to functional needs and those needs are somewhat similar across modern societies, various systems of property law will bear a strong resemblance to each other. (11) The basic indirect relationship between interests in use and the legal interests (as with the life estate and the superficies) is an inevitable feature of the overwhelming transaction costs of the "complete" property system envisioned on many versions of the bundle of rights theory. In short, because of transaction costs, the basic contours of the use interests and structure in civil and common property laws are similar. (12)
The perceived differences between civil and common property laws lie in their different styles of delineating property rights. Civil law starts with ownership, the legal interest that corresponds to the fullest use interest we can have in dealing with things--the greatest degree of control one can have--and then evaluates each lesser interest and the various devices for protecting them, in terms of how it does or does not promote this full type of interest. By contrast, the common law system grew out of feudalism and has always focused on the various lesser estates, which people might employ to protect various specific smaller classes of uses. Pushed to the limit, the common law then projects this fragmented picture back on the question of use itself and sees only a welter of specific uses and ignores the importance of the full reservoir protected by a simple exclusionary strategy. The "bundle of rights" is in a sense the theory implicit in the common law system taken to its extreme, with its inherently analytical tendency, in contrast to the dogged holism of the civil law. In transaction cost terms, both the civil and common law systems need an indirect relation between use interests and the legal interests and associated devices that serve them, but for historical reasons, civil law overemphasizes the overall interests in use, whereas the common law overly stresses the particular legal interests. Civil law is preoccupied with assimilating the pieces (interests) to the whole, whereas common law tries to articulate more specific uses corresponding to lesser interests and the devices that serve them. If the distinction between use interests on the one hand and legal interests on the other is not made, then it is easy to under-articulate the lesser interests (civil law) or overarticulate them (common law).
Our theory points out that the styles of civil and common property law have both persisted for transaction cost reasons. Style, more than use interests or structure, is subject to path dependence. On the cost side, the style of a system involves large fixed costs that only organized actors (like revolutionaries or authoritarian rulers) can accomplish in moments of political crisis. On the benefit side, because style itself matters only around the margins, it is difficult to change at other times because there is not much pressure for doing so, unlike with the contents of particular legal interests or structure. Further, the style of property delineation is subject to strong network effects as reflected in the internalist perspective just mentioned. A style is more useful the more people are using it, and so should show scale effects on the demand side, i.e., network effects. (13) The style of delineation is an aspect of communicating entitlements, and sharing a style makes that communication easier. Particularly in property, where not only officials and transactors but in many situations potential tortfeasors and non-official third party enforcers all need to process information about entitlements, a shared format and method of presentation of legally relevant information about property is particularly beneficial.
Because of their different histories, civil and common law face different costs of delineating property rights. In civil law, the starting point for the creators of the modern codes is the Roman-inspired undivided dominion and further division is a costly departure. These costs include the information costs of keeping track of the divisions and the need for third parties to process in rem rights. (14) When Roman law came to be elaborated in the process of creating the modern codes, the implicit limitation on the proliferation of new types of lesser property rights took the form of strong standardization of property rights into a limited list (numerus clausus). By contrast, the common law of property originated in feudalism, in which the focus was on personal relationships and reciprocal services. Thus, the fixed costs of a highly fragmented system were incurred long ago under circumstances in which they were worth incurring for political reasons. The result has been the persistence in the common law of a more articulated system than strictly necessary, especially given that a small number of combinable forms can achieve most parties' objectives. Accordingly, part of this path dependence takes the form of a looser version of the numerus clausus in common law than in civil law countries. (15)
Part I will argue that property's structure is distinct from the use interests that it serves. Property employs strategies of delineation (exclusion and governance) to serve people's interests in use. Common law and civil law property are similar in their broad outlines, most probably for functional reasons. We will concentrate on explaining the origins and persistence of different "styles" of delineating property in the two types of systems, and we draw out some implications for property versus contract, especially in mixed systems of civil and common law. As predicted, style in property law is much more resistant to change than in contract law, and more in rem aspects of property are more stable in terms of style than more in personam aspects of property law. Part II will set out an alternative to the conventional bundle of rights picture of property: property as a structured bundle of legal relations. The four prototypical property relations we identify follow from the exclusion-governance structure of property. In addition, because this exclusion-governance structure explicitly (as in civil law) or implicitly (as in the common law) employs "things" as part of the basic set-up of delineating property, (16) three "features of property" follow more or less automatically form the process of delineation. These three features are the right to exclude, in rem status, and running with assets. In Part III, we show that common law property tends to emphasize estates and other mechanisms to serve fine-grained uses, at the expense of attention to the holistic nature of full ownership. We then offer a transaction cost explanation based on the high fixed costs of setting up the estate system, large network effects, and high switching costs--all of which lead to path dependence. In Part IV, we turn to civil law and demonstrate that its preoccupation with dominion and things causes difficulties in marginal areas of propertized contracts, like leases. Civil law too exhibits strong path dependence for similar reasons to the common law, but with the costs incurred in Roman times and at the creation of the modern codes, when it was possible to incur high fixed costs. Despite marginal inconveniences, strong network effects and high switching costs combined with the low functional costs of maintaining the civil law style lead to path dependence in the civil law world as well. Part V concludes with some thoughts on how economic theory can serve to put comparative work about property on a sounder footing.
I. STRUCTURE AND STYLE IN PROPERTY
Property ultimately serves our interests in use. To do so in a world of positive transaction costs, property law employs an exclusion-governance structure that forms important shortcuts over the fully articulated set of uses and potential use conflicts holding between pairs of members of society. Conflating property's use interests and structure is quite common in property theory and nowhere more so than in the bundle of rights picture of property, which is the conventional wisdom in the United States and to a lesser extent in the common law world.
A. Structure: The Indirect Relation between Use Interests and Legal Interests
What are our interests that the law of property serves? We adopt the view that people's primary interest in things is to use them, in the broadest sense. The notion of use includes non-consumptive uses like preservation, aesthetic and existence value, and non-possessory contingent use (as in the security provided by mortgages and liens). By contrast, people have no socially recognized interest per se in excluding others from things, and someone who excludes for its own sake-without any reason other than to see someone else excluded--may be able to do so but would be considered somewhat odd. How is it then that, for many people, some version of the right to exclude is the centerpiece of property? (17)
Here we need to turn to the indirect relation between use interests and the legal interests (and their contours) that property employs to serve people's interests in use. Again, take the law of trespass, which instantiates in a fairly pure form the exclusion strategy for protecting use interests: by being able to exclude others who do not have permission, the possessor (and by extension owners) can go about using property for many different purposes they might have, as noted earlier. In the absence of further refinement through covenants, zoning, and the like, the owner enjoys a "reservoir" of uses that are protected by the law of trespass. All sorts of meddlers and thieves can be prevented from interfering with the uses because their access can be denied. The owner is not obligated to exercise exclusion rights, and the owner is free to offer conditional access, which conduces to all sorts of projects involving uses best undertaken in a joint manner.
The key here is that some strategies for protecting uses are highly indirect, because the proxies they employ (for example, the crossing of the boundary of a parcel) are only roughly correlated with harm to the owner's interests. (18) The law of trespass studiously avoids making reference to particular uses, and does not even require a showing of harm to any use whatever. Moreover, the law of trespass (like the law of ejectment, replevin, and the like) does not require the owner to justify the owner's uses, or even to show that they are more valuable than the uses that the defendant would like to undertake. One simple strategy based on a message of "keep off" or "don't touch" protects a large and indefinite class of interests in use in a wide variety of resources.
The indirectness between use-interests and the legal interests that serve them arises for transaction cost reasons, and will, we argue, be the key for unlocking the difference between the common and civil law traditions in property. Why are legal interests more coarse-grained than use interests? And why are the devices employed to protect legal interests (such as trespass) only indirectly related to the use-interests they protect, or put differently, why aren't property doctrines more tailored to the interests they serve?
In the broadest sense, the indirectness of property stems from positive transaction costs. In a zero transaction cost world, the Coase Theorem shows that we could use any devices with any degree of tailoring to protect use interests, and if such devices were not optimal for those concerned--however many they are--they would transact costlessly toward the efficient result. (19) Or, if we think of transaction costs as the costs of institutions, (20) the most articulated bundle of rights imaginable--every right with respect to every conceivable fine-grained use as between every pair of people with respect to every contingency--could be effected without cost. (21) In the real world, legal interests clump together use interests, and the devices like trespass employed to protect the legal interests can be blunt and simple in order to avoid transaction costs. Use interests can as a result be left largely implicit; they need not be spelled out (as they would in a world of only separately defined use rights). Legal interests and the strategies property law employs to protect them are a shortcut over the hypothetical "complete" system of property rights that could be achieved in a zero transaction cost world. (22) In our world, we need to make do with a property system that uses a basic exclusion strategy and reserves more fine-grained regulation of uses to more direct devices such as covenants, nuisance, zoning, and custom in a variety of governance strategies. (23)
B. Style and the Path Dependence of Property
The structure of property systems tends to show common contours for functional reasons, but transaction costs of a different sort explain why civil and common law differ greatly in terms of their style of delineating property rights. Our basic explanation of the persistence of the respective styles rests on path dependence. (24)
What is a style of property? In general, a style is a manner of doing things that is characteristic of a particular culture. Often the same function can be served by artifacts exhibiting different styles (think of the design of pots or the shape of automobiles). Common and civil law can be thought of as families of legal cultures, (25) in which a function, here delineating property rights and setting up interlocking property doctrines, can be served in multiple ways. We will focus on the process of delineating property rights. Later we will return to the details of how common and civil law styles of delineating property differ, bur we outline here what is at stake.
One basic choice in delineating property rights is how explicitly thing-based delineation should be. In functional (structural) terms, property systems can differ as to whether they rely on thing-based exclusion from parcels in the case of land or more on overlapping governance rights (as in many indigenous property systems). Bur even in property systems that give wide discretion to owners over defined things, the thing can be an explicit starting point as it is in civil law, especially German law (in which the thing must be tangible). In a sense, the delineation of estates in the common law focuses more on persons and activities, with thing ownership left implicit. (26)
Corresponding to the differential focus on things, systems of property law differ in their approach to ownership and title. For the civil law, as we will see, it is always important as a matter of theory to identify, as far as possible, the owner of a thing. By contrast, common law notions of ownership are looser and sometimes extend to the holder of any interest. (27) The common law pushes the notion of relativity of title much further, under which B can have better title than C even though A has better title than both. (28) For many purposes someone like B is treated as a lesser owner, and by the same token A's status as true owner is put on the same plane as the relation of Band C, etc. Perhaps the most famous example of the common law allowing relationships to trump the unity of ownership is the trust, in which a settlor transfers legal title to a trustee, who is obligated to manage it for the beneficiary. Controversy has flared up from time to time over whether the trust is contract or property (or both), but the main point is that the common law tolerates this type of fragmentation in the gray area between property and contract. (29)
As a result of these differences in delineation, property can look more or less distinct from contract. In the civil law system, and the German version in particular, there is a strict separation of property from obligation (including contract). By contrast in the common law, many of the lesser estates look something like very long term relational contracts (which under feudalism they were in a sense). Substantively, we can reach a particular result, like an easement or a lease, by taking thing ownership and separating out some sticks for the easement holder or the lessee. Or we could start with a contract over certain kinds of uses of a thing and give them more property protection against third parties (and in the case of leases, possessory protection) and the ability to bind successors of the holder of the dominant tenement or reversion.
Our account of civil and common law takes as a starting point the observation that, in a sense, the difference between civil and common law is like these two paths to a servitude, writ large. As is well known, the common law defines ownership as a robust and enduring form of possession (possession plus), whereas civil law defines ownership directly in terms of "full rights" to things, in terms of which lesser interests of a possessory sort can be defined. (30) In the common law, ownership historically grew out of the robust protection of possession-like seisin, and even today, possession gives a presumption of ownership. (31) The emphasis on possession, which is in the first instance a de facto relation, is sometimes said to reflect the concreteness and empiricism of the common law. (32) What is meant by statements like this regarding ownership/dominion and possession we will try to specify more precisely under the heading of legal style. The styles of civil and common law property are to a large extent organized around these respective starting points.
Although different styles may be employed to achieve similar functional ends, they may not be equally costly, especially as time passes. We hypothesize that styles of property delineation are highly subject to path dependence. This is more than simply that "history matters." Specifically, the type of path dependence identified by Douglass North as important in explaining institutions and their persistence is likely to play a large role in keeping styles of property delineation in place. (33) North argues that the features of technologies identified by Brian Arthur as promoting path dependence all apply to institutions: high start-up costs, learning effects, coordination effects, and adaptive expectations. Institutions involve increasing returns over time for these reasons, which make them harder to change using routine methods. (34) Politically provided public goods like law can be expected to show increasing returns and be subject to path dependence because they are meant to be used widely (they are nonrival and nonexcludable). (35) This leads to reliance on collective production involving large start-up costs, as well as learning effects, coordination effects, and adaptive expectations on the part of producers and consumers. As a prime example, North offers the Northwest Ordinance of 1787, which adopted a framework of property rights, a bill of rights, and procedures for admission as a state. (36) The Ordinance profoundly shaped Western land development, despite the availability of very different frameworks at the time. Finally, law reflects and creates a mental construct, which is at the heart of this and other examples of institutional path dependence. (37) Likewise, Marcel Kahan and Michael Klausner explain that path dependence and standardization in corporate contracting, a private law-making context, can be explained in part in terms of increasing returns, based in turn on learning and network effects. (38) Finally, at a very high level, Anthony Ogus defines "legal culture" as "a combination of language, conceptual structure and procedures," and argues that the commonality of usage in such a legal culture reduces the cost of interaction, leading to network effects. (39) Legal styles are an important aspect of legal institutions and cultures, which we should expect to show path dependence for similar reasons.
The styles of delineation show path dependence because they require large fixed costs to set up and have large network effects. Large fixed costs create scale effects on the supply and the demand side. On the supply side, a state setting up a property system can spread its costs based on the size of the jurisdiction over which the property system extends. (40) Further, because the state is involved in enforcing property rights, there are probably economies of scope in the state's taking on the role of defining property rights as well. (41) Legal styles as a feature of the property system partake of these economies of scale and scope on the supply side, but we will focus on how their path dependence arises from network effects--scale economies on the demand side. (42)
Network externalities arise because a user finds a system or product more valuable the more other users there are, the telephone system being the classic example. On the demand side, a property system is more cost-effective if its features cohere, and this coherence is the focus of much doctrinal work in law, especially in the civil law world. Although economists are not known for taking doctrine very seriously on its own terms, doctrine aiming at legal coherence can be a worthy subject of economic analysis itself. To the extent that doctrine, including the legal style of delineation, does cohere, one can form expectations that certain features go together. To take an example, the strict separation of property from obligation (contract, tort, unjust enrichment) in civil law, and German law in particular, gives rise to at least weak expectations about the form that doctrine will take in areas like leases and easements.
The various features of a legal style thus are complements, and people will find it advantageous to converge on one style or another. Although rules that fit uneasily with the prevailing style can be imported, they are harder to communicate and can cause costly misunderstanding. At the least, a legal style is a mode of communication, and as with other modes of communication, demand-side scale economies are to be expected. For one thing, in many instances, those making property claims will want to be clear to others about the nature of the claim, and legislators have an interest in avoiding "network confusion effects" of excessively idiosyncratic property rights. (43) Network effects thus lead to standardization in property as elsewhere, and we argue that network effects give rise to large-scale adoption of a single style of delineation.
With network effects in place, switching costs are then high, and the result is the persistence of the original standard, even if it is not what one would choose ex post. The controversies surrounding path dependence focus on different strengths of path dependence and how susceptible markets are to suboptimal lock-in. (44) In the weakest form of path dependence, "first-degree path dependence," a system is chosen, say driving on the right rather than the left, but either is roughly as good as the other (as long as everyone sticks to the same side). (45) Initial conditions of a choice, or even who started driving first, might tip things one way or the other and the choice will persist, but (for the sake of argument) it is a matter of indifference. No inefficiency is involved. Second-degree (or semi-strong) path dependence involves a choice that we would make differently now but cannot because it is too costly to change. (46) Given that information ex ante is costly and change is costly ex post, there is still not much to regret. Third-degree (or strong) path dependence involves persistence of a choice that is regrettable and could be changed, say with cost-effective collective action. (47) The most controversial examples are the QWERTY typewriter keyboard and the VHS video system. (48) In both cases, it has been argued that there is a superior system that lock-in has prevented markets from converging to (Dvorak keyboards and Betamax, respectively). (49) The examples have been challenged, either as not involving a superior system (QWERTY is no worse than Dvorak) or as ignoring the cost-effectiveness and superiority of a system for ordinary users (VHS is more cost-effective than Betamax for non-professionals). (50)
But even the skeptics of these examples note that path dependence is more likely in the case of political choice: it is not all that surprising that political choices resulting from one-time collective action are hard to undo. (51) Moreover, public goods provided through politics are especially prone to increasing returns. (52)
Our path dependent story is political and involves "weak" lock-in. Political actors can set up a new legal style in the course of reworking the legal system for political reasons, such as setting up Norman feudalism in the common law and creating codes as part of building a post-feudal nation-state in the case of the civil codes. (53) As with driving on the right and using the QWERTY keyboard (on the skeptical view), the legal style most of the time does not have major functional impacts, but it is very important that people in a society converge on such a style, leading to large network effects. Moreover, the weakness of the lock-in--in the sense that most of the time a legal style does not lead to major inefficiencies--lessens the pressure to change.
Constituencies who might want changes in legal style face institutional obstacles as well. Because property is in rem, judges are reluctant to make major changes to the property system, particularly when it comes to the set of basic forms of property or methods of fragmenting property. (54) In the civil law this principle is crystallized into the numerus clausus, bur a more implicit norm works similarly (although less strongly) in the common law. (55) The numerus clausus directs courts to defer to legislatures, which are in a better position to supply the clarity, universality, comprehensiveness, stability, prospectivity, and implicit compensation, which major changes to the property system require. (56) Indeed, for changes to an entire style of property delineation, we find that the costs are larger than are typically incurred in routine legislation.
Crucially, this account rests on the distinction we made earlier between the interest in use and the devices property law employs to serve them. The functional aspects of property that are familiar--the basic right to exclude, running to successors, in rem effect, and other aspects like trespass, nuisance, necessity, and adverse possession-implicate efficiency and are somewhat similar across the systems. (57) Proving this is not our point here; rather, the fact that legal style can operate somewhat independently of these functionally important features promotes path dependence and lock-in of the legal style. The style of a property system does have functional implications at the fringes of property (such as leases), so the path dependence is not the weakest of the weak. But legal style is a matter of political path dependence, and in most legal systems, there is no major constituency for change of legal style. A combination of initial conditions with high fixed costs, high switching costs, low switching benefits, and network effects creates the conditions for path dependence. (58)
C. Some Implications
We now turn to some general implications of an information cost theory of property for the distinction between civil and common law property. In terms of property's structure, we expect that functional pressures will tend to cause the systems to converge, but that is not our focus here. When it comes to style, we predict differential resistance to change, both between property and other areas of law and between in rem and in personam aspects within property law. First, if fixed costs are non-trivial in any area of law, we might expect some inertia. Where property is the most different from other areas of private law (such as contract) is on the demand side: in rem rights give rise to greater information costs, and the network effects of impersonal communication can be expected to be greater for property. And it is indeed the case that property is more resistant to change from outside influence than public law and even contract. The family of legal systems sometimes known as "mixed" or "hybrid" tends for historical reasons to consist of systems that start out as civil law and acquire common law influences. In such systems one striking pattern is that property is most resistant to change, or in one well-known assessment, property law is the "most unassailable stronghold of civilian jurisprudence." (59) Indeed, in a recent study of mixed jurisdictions, no jurisdiction showed common law influence on property but not contract, but four showed the reverse pattern of common law influence on contract but not property. (60) (Public law was even more subject to change in a common law direction.) No mixed jurisdiction for which data was available showed a common law type of property law. (61) And some jurisdictions featured a mixture of civil and common law within the law of contracts, which apparently happens far less in property law. (62)
Furthermore, if network effects relate to property's in rem aspect, we should expect that hybrid legal systems and transplants should show a pattern in which the in rem parts of property--those which form part of the content of in rem duties and affect the state of title--should be the most resistant to innovation. This also appears to be the case. While we leave this implication for further work, some anecdotal evidence is suggestive. Scots law has a numerus clausus of largely civilian types of in rem rights, but innovation, including some reflecting common law origins, tends to come in where property is not at its most in tem (as in servitudes). (63) In a similar fashion, English law has been influenced by Scots law mainly in the areas of fixtures and easements. (64) In general, standardization is greater where property is in rem rather than "intermediate" between in rem and in personam (as in trusts, leaseholds, bailments, and security interests), (65) and we can expect for similar information cost reasons that influence from another legal system's style would be easier in less standardized aspects of property law. Indeed, the history of English law provides a familiar example: when law and equity were separate, England was in a sense a mixed jurisdiction. The civilian influences that came in through the equity courts were not supposed to alter in rem property rights provided by the common law, and equity was limited to act in personam. (66)
II. PROPERTY AS A STRUCTURED BUNDLE OF RELATIONS
It is necessary to distinguish the structure of property law from its style, and common and civil law property systems differ more in style than structure. Before we can turn to style, we first offer a more articulated alternative to understanding property's structure. The nature of property rights is best captured as a structured bundle of relations, which follows from the exclusion-governance structure. How this structured bundle of relations is achieved leaves room for alternative methods, which we call the style of delineation of property. The style of delineation is subject to path dependence. The relation of use interests, legal interests, structure, and style will form the basis of our transaction cost explanation for the similarities and differences between common and civil law property.
In American property law, a property right has long been described as a collection of legal relations between parties with respect to things (or resources). (67) This characterization is less acceptable to property scholars in civil law countries. (68) The classical, mainstream theory in civil law countries treats a property right as a relation between a person and an object. (69) While we embrace the American version of property theory and will criticize the civil law idea in Part IV, we are also unsatisfied with the de-emphasis on things in common law and with the fact that the nature or the typology of the property relationship has not been clearly spelled out.
Property theory typically fails to distinguish the use interests in property and the structure provided by property law. Nowhere is this more true than in the bundle of rights picture of property. In the common law world, and especially in the United States, the mainstream view looks at property as a bundle of rights--or more metaphorically as a "bundle of sticks"--holding between right holders and duty bearers with respect to a thing. (70) Thus, with respect to Blackacre, A might have the right to exclude B, the right to use the land for growing corn, the right to cross over C's neighboring land, etc. What the bundle picture denies is that there is some essence or core of property. (71) On the bundle picture, "property" is a label that we can affix to any collection of these various use rights with respect to a resource. The traditional notion that property is a right to a thing availing against others generally--an in rem right--is considered an inconvenient obstacle to clear thinking and to badly needed reforms in the configurations of legal rights and duties, which will require owners to give way increasingly often to collective decision-making. (72) Even where the bundle picture holds less sway, property theorists and lawyers tend to emphasize the process of carving out interests, using forms tracing back to feudalism. One way or another the common law world emphasizes types of fragmentation. Even Blackstone--who is famous for describing property as that "sole and despotic dominion which one man claims and exercises over the external things of the world, in total exclusion of the right of any other individual in the universe"--then immediately went on to describe in voluminous detail the various ways of fragmenting and qualifying property rights. (73)
By contrast to the common law, the civil law theory of property is all about in rem rights, and property--ownership in particular--is seen as inherently undivided. The mainstream civil law tradition generally has no place for and no interest in the bundle of rights picture of property. (74) Countries with civil law systems may favor a large degree of government regulation, but this impulse has never expressed itself, as it did from the 1930s onward in the United States, in the bundle picture of property. This divergence in receptivity to the bundle is all the more striking because certain late nineteenth- and early twentieth-century legal theorists in Europe did advocate something like Realism and even a proto-version of the bundle theory, and these European sources inspired some of the American proto-Realists. (75)
This differential receptivity to the bundle picture is a key to the first fault line between common and civil law property. Common and civil law property focus on different aspects of an important distinction between property's use interests, legal interests, and structure. In the common law system, the bundle of right theory focuses on the use interests and the estate system directs our attention only to certain legal interests. By contrast, the central place of ownership and its in rem-ness in the civil law lead theorists there to stress only the overall legal interest and the exclusion side of the property structure.
It is worth noting that some may point to the elasticity or flexibility of ownership in civil law--when a lesser property interest is extinguished, the bare owner recaptures the value and becomes the full owner again (76)--as evidence that ownership is quite different from fee simple absolute in common law. Nevertheless, in common law vocabulary, the elasticity just means that the original owner always keeps the remainder or reverter. (77) Hence, the elasticity does not reflect a fundamental or structural difference between fee simple absolute and ownership. Rather, it suggests that civil law limits the owner's alienability of property more than the common law does.
In this Section we offer an alternative to the bundle picture that will serve as a basis for our analysis of civil versus common law property. In Subpart A, we argue that a property right contains a structured bundle of relations with four prototypes. (78) Our thesis can clarify the nature of property rights as relations and also helps to illuminate the difference between contractual relations and property relations. In Subpart B, we re-visit the necessary elements of a property right and show how a focus on the "things" of property links them with the four prototypical relations.
A. Four Prototypes of Property Relationship
As argued above, legal systems facing positive transaction costs of delineating property rights tend to employ some version of the exclusion-governance structure. The exclusion strategy is the default and backbone of the structure, setting up an indirect relation between legal interests and use interests. Sometimes, the governance strategy is employed when a more direct delineation of use interests is necessary. In this Subpart, we argue that four prototypical property relations should be distinguished because each employs different combinations of the exclusion and governance strategies. There are four prototypes of property relations: property right holders versus the government, property right holders versus other property right holders, property right holders versus some specific others, and property right holders versus all others.
In the first prototypical property relation, holding between the holder of an interest and the government, the government underwrites the interest (promising protection), but the interest is held subject to the government's police power and the power of eminent domain. (79) This relation is "exceptional" in that the eminent domain power is not a necessary element in structuring property rights; rather, it is a convenient tool, recognized around the world, for the government to re-allocate titles of properties to itself.
The second prototypical property relation, holding between property right holders and other property rights holders, can be divided into two sub-types depending on who those others are. In the first sub-type, a voluntary governance strategy is used, under which the owners or tenants of fee simple absolute voluntarily agree not to exercise their exclusion rights, allowing holders of "lesser" property interests (such as an easement) to use the properties in the negotiated way. In the second sub-type, one holder of a lesser property interest can "exclude" the holder of another lesser property interest but only when the two interests conflict (as with a mortgage and a later-created easement). The second prototypical property relation (of both sub-types) arises in both systems only upon creation of lesser interests out of the largest or "full" package ownership.
The third prototypical property relation, between holders of property rights and specified others, features involuntary governance and limited exclusion. Here, property interest holders cannot always determine how their things are used if the law has prescribed the things to be used in a certain way at a certain time by a certain party. Not all instances of property contain this type of relation, though.
The fourth property relation, between holders of property rights and all others generally, is of a more purely exclusionary nature. This relation is especially important because every holder of property interests has this relation with all human beings in the world except for those who are in the second or third prototypical relations. The (automatic) existence of this relation saves transaction costs and thus is omnipresent in some form in the property structure in a world of positive transaction costs.
1. Property Right Holders vs. the Government
The first prototype of property relation exists between property right holders and the government. This relationship is more a matter of public law than the other three relationships. Despite the difference that in the common law traditionally everyone ultimately held of the monarch, states in both civilian and common law jurisdictions use similar devices, sounding in due process and the rule of law, to trade off the flexibility needed to address changed conditions and the precommitment to existing property holders required in order to promote owner autonomy, elicit investment, and so on. Of particular significance to the internal workings of the law of property is the role of the government as user of eminent domain. This relation should be independent of other relations because the government's eminent domain power causes the property rights, as against the government, to be only protected by liability rules, while property right holders' interests are generally protected by property rules as against all others without authorized eminent domain power. Indeed, in setting out the framework of liability rules, under which an entitlement can be taken upon the payment of officially determined damages, as opposed to property rules, under which entitlements receive robust protection aimed at requiring an owner's consent, Guido Calabresi and A. Douglas Melamed discussed eminent domain as an example of a liability rule. (80)
If the relations with the government and those with non-governmental entities are mixed, as the traditional account seems to assume, eminent domain as a liability rule is like an exception to the general property rule protection. We contend that it is clearer to think of the liability rule as the general rule in property right holders' relation with the government insofar as the issue is eminent domain, and the property rule as the general rule (with very few, sometimes unjustified, (81) exceptions) in property right holders' relation with others generally. These include non-governmental parties (that is, the second, third, and fourth type of relation), and the government outside the context of eminent domain.
2. Property Right Holders vs. Other Property Right Holders in the Same Asset
The second prototypical property relation is among property right holders in the same asset. A typical example is the relations between estate holders, say, between A with a life estate and B with the remainder. The relation contains two sub-types. The first sub-type is the relation between the "bare owner" (82) and the holders of "lesser property interests," such as between a mortgagor and a mortgagee. The second sub-type is the relation among holders of lesser property interests, such as between a mortgagor and another person who holds an easement over the same land.
In the first sub-type of relation, a property interest holder voluntarily changes her relation with someone from exclusion (the fourth type) to sharing property rights. In other words, the voluntary governance strategy is used to adjust the property interests between two parties. The extent of adjustment is limited by the numerus clausus principle. (83) Thus, a property interest holder can only enter into a limited number of types of property relations. (84) And the rights and obligations of the relations between property rights holders are determined by civil codes or common law doctrines. (85)
In the second sub-type, lesser property interests are governed by the doctrine prior tempore potior iure, or "prior in time, prior in right." (86) In other words, the right to exclude still applies, but the right to exclude can only be exercised when the earlier lesser property interests would otherwise be hampered by the later lesser property interests. For example, the mortgagor (who is earlier in time) can request the court to remove a later-created easement when the mortgaged land parcel is auctioned, because an existing easement would decrease the auction value.
3. Property Right Holders vs. Some Specific Others
The third type of relation is where civil codes, regulations, or court-made doctrines adopt a governance strategy and thus the property right holder's right to exclude is limited. For example, in nuisance law, property owners cannot always require their neighbors to stop producing noise or odor if their neighbor's activity level is reasonable. In boundary encroachment disputes, property owners sometimes will be required to tolerate a good-faith neighbor's encroachment and receive only damages. Also, landlocked owners can pass through their neighbors' land to access public roads under certain circumstances. (87) The list can go on.
4. Property Right Holders vs. All Others
The shadow example in the previous literature on property relationships is the relations between property right holders (say, owners) and most people in the world. The in rem nature of property rights discussed in the literature refers to this type of relation. Some commentators seem to consider the fourth type of relation as the property relation. (88) A vast (and increasing) number of relations are created automatically every time any kind of property right is created. The nature of these relations is simple and clear--property right holders have a right to exclude anyone who does not have either of the other three prototypical relations with the property right holders. In other words, entitlements of property right holders are generally protected by property rules.
This fourth relation--property right holder versus others generally--is the default one, which can be displaced in various contexts by the third relation--property right holder versus specified others. For this reason, we might expect the fourth "in rem" relation to show more similarity across systems. That is, both civil and common law employ the exclusion-governance architecture and are similar in their law of trespass, but various legal systems differ in when and to what extent relations of the third type involving specified others will displace the background relation to the rest of the world. Exceptions to trespass and the content of doctrines like nuisance can be expected to vary more, as different jurisdictions face different situations more toward the in personam end of the spectrum and evaluate them differently.
B. The Sine Qua Non of Property
Some theorists have looked for a sine qua non of a property right, although on the most extreme bundle of rights view no stick is more privileged than any other. The most frequent candidate for the essential feature of property is the right to exclude, and even some prominent bundle theorists have given it a special prominence. (89) Our argument is that all of these views come up short. For information cost reasons, the right to exclude and allied features "fall out" of a thing-based approach to property--they simply follow from the basic set-up.
What is a thing? This sounds like a metaphysical question, but we will follow both the Romans and the common-law lawyers in invoking philosophical notions only as they are required for practical reasons. (90) In particular, we argue that property is a law of things--that we have a law of things in the first place--for transaction cost reasons. In what might be termed the Coase Corollary, in a zero transaction cost world property could take on any contours without any effect on the efficiency of the resulting pattern of use. (91) This includes defining its basic scope. A "complete" property system, in which property was defined with "full" precision on all dimensions, would be costlessly achievable; in such a system rights would refer to the tiniest uses over the shortest times availing between each pair of members of society, existing and unborn, and would incorporate every conceivable contingency. (92) In other words, the property system could rely completely on the governance strategy.
In our world property rights are "incomplete" so as to save transaction costs. (93) Note that property rights can be incomplete in at least two senses. Incomplete property rights in the literature refer to the incomplete delineation of property rights. That is, some resources, like the high seas, are held in an open-access commons, or not propertized, more generally. (94) Our notion of incomplete property rights here refers to the lumpy legal interests and indirect devices employed to manage propertized resources. The law does not (cannot, for transaction costs reasons) stipulate every tiniest use of each property. Instead, the law designates an owner who has a presumptive right to exclude others to determine the use of a defined thing. (95) By defining a thing in the exclusion strategy, people's interests in use can be managed at relatively low cost. The owner will decide how to transact with others to share the use.
How the thing is defined will give rise to many features of property without any extra definition. This is what makes property rights special and gives them their "residual" character. John Austin noticed this aspect of property when he said of property that "indefiniteness is of the very essence of the right; and implies that the right ... cannot be determined by exact and positive circumscription." (96) Particularly with respect to uses, the basic way that property is set up obviates the need to spell out uses. The result is that an owner has control over an indefinite reservoir of uses. (97) Not having to spell out the uses in this reservoir saves on transaction costs. Only for particularly contested uses does it make sense to separately delineate legal relations in terms of such uses. For access to a driveway or the rights and duties with respect to odors and the like, it makes sense to incur the costs of a governance regime to "fine tune" these salient use conflicts.
Let us consider how some features thought to be characteristic of property follow from this basic transaction cost-saving move of defining a thing through an exclusion strategy. They include in rem status, the right to exclude, and running with assets. As shall be clear from the discussions below, these three essential features are concepts at different levels: being in rem refers to the automatic creation of property relations between property interest holders and all others (the fourth prototypical relation); the right to exclude describes the nature of the second and fourth prototypical relations (property right holders versus other property right holders in the same asset, and property interest holders versus all others), and, to a lesser extent, the third prototypical relation (property right holders versus some specific others); and running with assets means that when a party to a property relation transfers her rights to another person, the new interest holder just steps into the shoes of the original interest holder without disturbing the existing property relations. (98) The fact that these three features are essential yet conceptually entwined is probably why the prior literature has been unclear about their conceptual relationship. These three features are embedded in the exclusion-governance structure.
1. In Rem
Property is ah in rem right. Etymologically this means a right to a thing, and historically there is a connection between property being a right to a thing and its being in rem in the sense of availing against others generally. The concept of in rem in German law, and its progeny, is very prominent, (99) and it means that the right is good against the world--usually called "the principle of absoluteness (Absolutheitsprinzip)" in German law. (100) In both civil and common law property, the thing mediates the relation between the owner and the duty bearers, who are largely told to keep out or not to interfere, unless they have the owner's permission.
Communicating with a large and indefinite class of persons whose main contribution to the value of property is not to interfere saves on transaction costs. (101) In our world of positive transaction costs, contracting with all others to keep them off one's property is prohibitively costly. Automatically creating a right to exclude all others thus saves immensely on transaction costs. Delineating the right based on the thing makes the right impersonal in the sense that contextual information about the owner and the duty bearers is generally not relevant to the nature of the right (duty). When more specific parties are involved, the further delineation of legal relations can use the thing as a platform for getting more specific. Thus, the law of nuisance prescribes proper use as between neighbors, and covenants and contracts can deal with very specific uses. As we will see, the civil law places particular emphasis on the in rem rights of a full owner.
2. Right to Exclude
As noted earlier, many, including a number of Legal Realists, have given prominence to the right to exclude in property. Various commentators mean different things about the right to exclude and exclusion. We agree with those who argue that our interest in property is one of use that is formally protected by a right to exclude implemented through devices like trespass. (102) Nonetheless, when one says that the right to exclude is the essence or of special importance in property, what does this mean?
The transaction cost theory suggests that the right to exclude is not a "stick in the bundle," despite the frequent pronouncements to that effect by, for example, the United States Supreme Court. (103) Rather the exclusion strategy sets the baseline--it is a platform or starting point--from which we know what a thing is and from which departures in both directions--subtractions from the owner's rights and additions to the package--can take place. For example, the doctrine of necessity and antidiscrimination law withdraw sticks from the exclusion-based baseline of rights, and easements and rights of lateral support add to it.
For transaction cost reasons it is important to distinguish between interests and the devices that serve them. If exclusion is the starting point for defining thing-based packages of rights--property--this is not to say that there is ah interest in exclusion or that such an interest is more important than the interests in (and policies for) saving life and limb (necessity) or promoting racial equality (antidiscrimination law) that are served by more specific laws. On the contrary, exclusion and governance are simply different, with the former supplying the rough platform for the latter. Common and civil law both take exclusion strategies as a starting point bur differ in the kind and extent of departures in the direction of governance--and, as we will soon see, in their styles of delineation.
3. Running with Assets
Others have identified another candidate for the essence of property: the ability of a right to run automatically with an asset into remote hands. (104) That B can succeed to A's interest when A holds a fee simple, other present and future interests, an easement, etc., is unremarkable and correlates very closely with what we might call property. Even more telling, when a contract between neighbors runs to successors, it is placed with easements in the supercategory of servitudes and comes close to being treated as a property right. The theoretical difference between an easement and a running covenant is that the former but not the latter binds third parties (in rem), but in the context of neighbors it is the parties and their successors that are important. Thus, if A covenants with B that A will not build more than a two-story building, and both intend for the covenant to run, and it touches and concerns the land, it will run to successors. (105)
Again, the running of the covenant is not a detachable feature and certainly not a stick in the bundle. Rather, to the extent that the right is embedded in the baseline package of rights associated with a thing, then it is natural for it to run. For this reason, appurtenant easements, which are explicitly carved out property rights, "run" without controversy. More specifically, the baseline is (consistent with the notion of a thing) meant not to contain contextual information about persons, and so who holds the right and the duty is irrelevant. From there it is a short step to saying that successors are bound. In general, de-contextualizing a right is a predicate for its alienability. The same is true for rights that run to successors.
There is thus a tendency for property rights to run to successors more easily than contract rights. That is true of both common and civil law, bur, as discussed in Part IV, the systems differ in where to draw the line, partly for "stylistic" reasons.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Introduction through B. The Sine Qua Non of Property 3. Running with Assets, p. 1-35|
|Author:||Chang, Yun-chien; Smith, Henry E.|
|Publication:||Notre Dame Law Review|
|Date:||Nov 1, 2012|
|Previous Article:||The (un)foreseen effects of abrogating proximate causation in CSX Transportation, Inc. v. McBride: the new role of foreseeability under FELA and the...|
|Next Article:||An economic analysis of civil versus common law property.|