in , , , , , , , , , , , ,

Beyond Information Costs: Preference Formation and the Architecture of Property Law | Journal of Legal Analysis

Written by Zhang T.

Abstract

Contemporary property theory highlights information costs as the central determinant of exclusion rights and numerus clausus-type standardization: rising information costs lead to stronger exclusion rights and more standardization, whereas falling information costs have the opposite effect. This paradigmatic model lacks, however, a theory of how information costs change in the first place. By developing such a theory, this article demonstrates that, in prominent cases, the legal impact of information costs tends to be counterbalanced by concurrent changes in individual preference, and that preexisting predictions about the relationship between information costs, standardization, and exclusion are therefore partially wrong, and otherwise incomplete.

1. INTRODUCTION

Across much of the modern world, a core feature of property systems is what some scholars have called the “numerus clausus principle”: only a finite number of property forms are legally allowed at any given point in time. As a large amount of scholarship—associated most prominently with Henry Smith and Thomas Merrill—has pointed out, there are both costs and benefits to such legal standardization: it usually frustrates the individual preferences of some property owners, but also generates substantial information cost savings by lowering the “measurement costs” associated with property transactions (Merrill & Smith 2000; Chang & Smith 2015). While acknowledging both sides of this equation, the preexisting literature pays far more attention to the information cost benefits of standardization than to its preference frustration costs, to the extent that its lead players have repeatedly characterized their analytical framework as an “information cost theory” (Merrill & Smith 2000, p. 29; Chang & Smith 2015, p. 2282; Smith 2004, p. 1745). This allows them to link numerus clausus-type standardization to another central characteristic of modern property systems—the strong exclusion rights conventionally granted to property owners—in that both are portrayed as “architectural” features of property law that aim to manage the systemic information costs inherent in the use and transfer of physical “things” (Rose 1988; Merrill & Smith 2001; Smith 2012; Chang & Smith 2012; Bell & Parchomovsky 2016).

Although “information cost theory” has a number of normative applications, it is, at its core, an empirical theory. Its power and attractiveness, relative to competing theories, stem from its asserted superiority in describing and explaining central characteristics of real-world property systems. While critics may point to personhood interests (Radin 1982; Davidson 2008; Dagan 2011), democratic politics (Purdy 2010; Singer 2014), or other socioeconomic factors as possible determinants of some aspect of property law,1

information cost theorists argue that their framework is uniquely equipped to explain the fundamental structure of property law—which is one that places standardization and exclusion at the very center (Chang & Smith 2012, 2015). The intellectual significance of the theory depends in large part, therefore, on the following empirical claim: higher information costs, more consistently than other factors, lead to tighter standardization and stronger exclusion rights, whereas lower information costs have the opposite effect.

This article argues that these claims overestimate the influence of information costs over the “architecture” of property law. In particular, there are strong theoretical, and not merely empirical, reasons to believe that, in important cases, the level of standardization bears no obvious correlation to information costs. The core theoretical insight behind this argument is that, under many sociopolitical circumstances, information costs and the preference frustration costs of standardization are positively correlated: for reasons inherent in their sociopolitical composition, the two measures tend to increase (or decrease) together. In these cases, the costs and benefits of standardization partially offset and either can dominate, depending on contingent sociocultural, political, or economic factors outside of the traditional cost–benefit framework.

Information cost theory, as currently constructed, is incomplete in at least two ways that limit its explanatory and predictive power: first, although the theory does acknowledge the existence of frustration costs, it has, as noted above, relegated them to the background while focusing on information costs. Second, it has largely taken information costs as a starting point for institutional analysis, rather than as a sociopolitical phenomenon that deserves its own explanation: current scholarship asks, essentially, “what demands do information costs place upon property law,” rather than “what are the scenarios in which property-related information costs are created or eliminated, and what demands do those scenarios place upon property law.” Filling in these two gaps—by developing a sociopolitical account of both information costs and frustration costs—unveils a number of systemic connections between the two kinds of costs, and demonstrates how they interfere with each other’s institutional impact.

Legal theorists and social scientists have, over the past several decades, identified two primary sociopolitical determinants of information costs: the closeness of social communities, and the state’s capacity to collect and share information—the remainder of the article refers to these as the two prongs of “sociopolitical cohesion” (Ellickson 1991; Rose 1999; Bell & Parchomovsky 2016). Individual economic actors generally obtain information about property rights either through information-sharing mechanisms in closely-knit communities or through state-administered ones, such as official title registration or recording systems. Correspondingly, information costs tend to increase when either communities or states weaken, all else being equal, and decrease when they strengthen.2

2

It is perfectly possible, of course, for communities to disintegrate while the state expands its capacity, or vice-versa, and in such cases the overall balance of “sociopolitical cohesion” is more ambiguous, as is the overall trend of information costs.

In rare cases—for example, sudden and exogenous technological shocks to property use—information costs may change independently of sociopolitical cohesion, but in such cases causation also runs in the opposite direction: if information costs increase, then the ability of states or communities to convey reliable information weakens, and sociopolitical cohesion falls. Thus, regardless of the direction of the causal arrow, information costs are negatively correlated with sociopolitical cohesion. From another angle, insofar as information is a prerequisite for the exercise of power, information costs are essentially a negative measure of the capacity to influence and control, both for states and for communities.

But beyond just affecting—and in some cases being affected by—information costs, changes in sociopolitical cohesion have at least one other major consequence that preexisting scholarship has largely failed to recognize: they affect the range of individual preferences regarding property—how diverse, or how homogenous, those preferences tend to be. Closely-knit social or religious communities, as large amounts of sociological and psychological research suggest, tend to limit the range of institutional preferences and needs expressed and held by their members (Douglas & Wildavsky 1982; Douglas 1986; Kuran 1996; Etzioni 2000; Sunstein 2000; Cohen 2003; Kahan & Braman 2006). In other words, in closely-knit communities, individuals tend to behave, and perhaps think, more alike than they would in a more disintegrated social setting.

States have a similar effect on individual preferences: they often attempt to control, through law, regulation, or ideological programs, the socioeconomic preferences of their population. Even when they refrain from doing so, strong states create stable platforms for social discourse and debate, through which socioeconomic consensus and behavioral similarities tend to emerge (Cohen 2003; Wildavsky 1987; Lewis & Cullis 1988; Tomer 1996; Brighouse 1998; Druckman & Lupia 2000). As a result, when a state weakens, assuming closely-knit communities do not emerge to fill in the organizational void it leaves behind, the socioeconomic preferences of its population—both the “actual” preferences held subjectively by individuals and the preferences they express to others—will probably become more diversified. When, however, the state’s administrative capacity increases, holding the state’s ideological orientation constant, preferences will likely become more homogenous, even in the absence of conscious political efforts to control them.3

3

The instinctive counterargument to this claim will likely be that modern liberal states do not actively seek ideological conformity from their citizens. This may well be true—although it is significantly less true in the property use context than in the political behavior context—but to the extent that liberal states nonetheless provide platforms for social discussion and information sharing, they still have some homogenizing effect on preferences, albeit a much weaker one than authoritarian states running a planned economy do. Thus, the claim is that, holding the ideological composition of the state constant, a stronger liberal state will nonetheless likely lead to less preference diversity than a weaker liberal state. See Section 3, subsection3.1.3 for further discussion.

Significant changes in property-related information costs are therefore often accompanied by changes in preference diversity: both tend to increase when sociopolitical cohesion falls, and decrease when it rises.4

4

There are two major caveats to this theoretical framework, both of which are discussed in considerable detail later in the article: it operates unproblematically only in cases where communal cohesion and state capacity change in roughly the same direction, or where one changes while the other remains relatively stable. A number of theoretical complications arise when we consider cases where one increases while the other decreases, where, in other words, the overall balance of sociopolitical cohesion is more ambiguous. The article does discuss how it might qualify its theoretical arguments to accommodate these potential problems, but its case studies are limited to scenarios in which sociopolitical cohesion changed in relatively “unambiguous” ways. Second, the model only applies to cases where the overall level of socioeconomic development remains relatively stable, before and after the shock to sociopolitical cohesion—and does not apply, therefore, to cases where sociopolitical cohesion plunges to such depths that the possible uses of property are fundamentally diminished even without legal interference. As scholars have pointed out, society can be thrown back to a significantly more primitive state following severe drops in sociopolitical cohesion, where both property-related information costs and preference diversity fall sharply due to severe declines in social stability and economic productivity. See, e.g., Fitzpatrick (2006). In these “state of nature”-like scenarios, extreme declines in sociopolitical cohesion may lead to less, rather than more, emphasis on exclusion, whereas legal standardization may become completely unnecessary. Neither this article nor traditional information cost theories really deal with this dystopian “tail” of the sociopolitical cohesion spectrum, although it is important to acknowledge its potential existence and implications.

Recognizing their synchronization forces us to revise some of the central predictions of mainstream information cost theory: First and foremost, it implies that there is often no obvious correlation between information costs and legal standardization. An increase in information costs, as information cost theorists have long argued, incentivizes greater standardization. The simultaneous increase in preference diversity pushes, however, for less standardization, by increasing the preference frustration costs generated by current levels of standardization. Therefore, even when information costs sharply increase, property law may become less standardized, or undergo no significant change. On the other hand, decreasing information costs may lessen the pressure to standardize, but the corresponding decrease in preference diversity also reduces frustration costs. The pressure on property law therefore recedes from both directions, leaving states and communities with greater freedom to shape it according to other, potentially unrelated, considerations. In other words, information costs are often systemically bundled together with their own antidote.

Second, the positive correlation between information costs and exclusion rights is probably even stronger than what information cost theorists have traditionally predicted, but it is only half-driven by information costs themselves. Higher information costs boost the attractiveness of exclusion-based private property regimes, which tend to manage such costs more efficiently than top-down “governance”-based systems (Smith 2002, 2012). Corresponding hikes in preference diversity, too, tend to favor strong exclusion rights, mainly because they are often the most effective safeguard against outside interference—neighborly, communal, or governmental—with one’s preferred modes of property usage. Greater preference diversity also damages the social basis for communal ownership and use of property. On the other hand, when sociopolitical cohesion increases, the benefits of exclusion ebb as information costs and preference diversity fall, while the social feasibility of communal management increases.

A theory of property that incorporates preference diversity and sociopolitical change will therefore generate empirical predictions that often diverge from those generated by traditional information cost theory, while reinforcing them at other times. Properly understood, it is an expansion and partial revision of the information cost framework, rather than a rejection of it. Unlike some other theories of property, such as personhood theories or the new “progressive school,” it does not attempt to build a completely different framework from scratch, and is quite happy to accept many of the functionalist insights that information theory supplies. That said, the expansion is qualitatively significant: it looks beyond information costs to their origins, and then to offsetting sociopolitical phenomena on the frustration cost side that stem from those very same origins. Even when it reinforces the empirical predictions of information cost theory, it does so in a manner that dilutes the institutional significance of information costs per se. Information cost theory thus expanded probably should not be called information cost theory anymore.

graphic

The following diagram visualizes the model that emerges from these exercises:

Mainstream information cost theory focuses on—perhaps not exclusively, but certainly predominantly—the upper right corner of the diagram, on the connections between information costs, standardization, and exclusion, whereas the model developed in this article brings together the entire picture.

The article presents several historical case studies to illustrate different parts of its theoretical model5

5

It should immediately be acknowledged that these are highly condensed—and therefore inevitably over-simplified—summaries of complex historical narratives, based predominantly on secondary sources. A theory-focused, article-length project can unfortunately do no better. The level of historical narration and analysis attempted here is, at least, no more general or abstract than what is customary for property theory articles, and the article does make every effort to rigorously engage the secondary literature.

: first, the model predicts that, when overall levels of sociopolitical cohesion unambiguously decrease, there will be a tug-of-war between higher information costs and greater preference diversity, but also that the predominance of exclusion-based private property should generally increase. These processes are illustrated through three case studies of societies that experienced communal disintegration without a complementary expansion of governmental capacity: the development of English property law following the Black Death, the evolution of customary property law in China’s Lower Yangtze macroregion during the late nineteenth century, and attempts to standardize subsets of American property law in the early Republic.

Contrary to traditional information cost theories—but consistent with the model constructed here—English property law became, in general, much less standardized, recognizing a significant number of new forms, most importantly uses and trusts, that gave most landholders substantially greater freedom than they previously had. Lower Yangtze property customs came under significant pressure by both large landlords and some of their tenants to become less standardized, but their efforts were not always successful, often because the information costs involved in creating and enforcing new forms were too high. American law, however, trended largely towards standardization in the late eighteenth and early nineteenth centuries, eliminating significant numbers of defunct property forms and simplifying the overall structure of land ownership. It did, however, recognize a number of new forms, particularly in tenancy law. Exclusion-based private property, on the other hand, became more entrenched in all three cases, gaining popularity over common ownership and open access.

When sociopolitical cohesion unambiguously increases, on the other hand, information costs and preference diversity both tend to decrease, reducing pressure on legal institutions from both directions. Exclusion-based private property, on the other hand, will probably lose some ground to other forms of ownership. These theoretical mechanisms are sketched out through two sets of case studies: first, the article gives an overview of the land reforms conducted by the Chinese Communist Party shortly after its ascension to power in 1949. As the Party began to assemble a state apparatus of near-unprecedented strength, it utilized the administrative capacities of that apparatus to push through sweeping reforms in property law, generally by standardizing ownership forms and clamping down on local land markets, and eventually by discarding private property altogether. Second, the article surveys the academic literature on communal land ownership and common property use. In general, these studies find that communal ownership and open access regimes thrive in stable, closely-knit communities, but tend to decline when those communities disintegrate. They also find that—contrary to the predictions of mainstream information cost theory but consistent with the theory developed here—the emergence of commons often comes hand-in-hand with higher, rather than lower, levels of standardization.

The remainder of the article is organized as follows: Section 2 identifies the core arguments in the information cost theory paradigm. Section 3 constructs an alternative model of how information costs influence property law: it argues for the synchronization, with some exceptions, of information costs and preference diversity—and therefore, in the standardization context, for the synchronization between information costs and frustration costs—and discusses its institutional consequences. Sections 4 and 5 provide a number of case studies to illustrate various segments of this model, the former focusing on cases where sociopolitical cohesion clearly declines, and the latter on cases where it significantly strengthens. The Conclusion discusses some of the broader implications of the model and identifies avenues for future research.

2. THE INFORMATION COST PARADIGM

Information cost theory finds its origins in the seemingly endless debate over whether property law is properly characterized as a “bundle of sticks” or a legal system centered on the right to exclude. Although the bundle of sticks metaphor—championed by several generations of legal realists (Cohen 1935, p. 815; Grey 1980; Rubin 1984, p. 1086; Williams 1998, p. 297)—was dominant for much of the twentieth century, a robust counterattack has emerged over the past two decades, arguing that the right to exclude enjoys a principal, sine qua non status in property law that other rights in the metaphorical bundle do not possess (Penner 1995; Fennell 2012; Smith 2012). Among the various subsets of this literature, the information cost theories associated most prominently with Henry Smith have emerged as the focal point for recent academic debate on property law. Scholars have described “the property-information interface” as “the most crucial … dimension of property” (Bell & Parchomovsky 2016, p. 237), and have devoted numerous articles to parsing the content, scale, and consequences of property-related information costs. Likewise, opponents of law and economics often identify information cost theories as their primary adversary, and have been particularly eager to engage it within the context of “bundle of sticks” versus exclusion debates (Rosser 2013, pp. 145–166).

Smith, alone and with Thomas Merrill and Yun-chien Chang, argues that a central purpose—perhaps the central purpose—of property law is to combat the information costs inherently associated with the ownership and use of things (Merrill & Smith 2000, 2001; Chang & Smith 2015; Smith 2004). Property rights are in rem—they define a relationship between a person and a thing that can then be held “against the world” (Austin 2005, p. 44)—which tends to create high information costs associated with ownership, use, and transfer. Property law therefore requires a great level of simplicity and clarity, which leads, under Smith’s framework, to two central characteristics: an emphasis on the right to exclude (Smith 2012), and an institutional tendency towards “optimal standardization” (Chang & Smith 2015), where, in order to curb information costs, the law only recognizes a limited number of property interests.

The basic logic behind these arguments is straightforward and powerful: a uniform default assumption, modifiable only in special circumstances, that property owners possess a general right to exclude simplifies the information conveyed to potential trespassers. Not only does the information come in a uniform package, but it also creates a basic presumption of non-interference that further simplifies the socioeconomic interactions between property holders and most outsiders. “Optimal standardization” via limiting the number of allowable forms6

6

The definition of “property form” employed here is the same conventional definition that Merrill and Smith illustrate at The Numerus Clausus Principle (2000), at 11–20. It is important to note that, while these forms are all in rem, in that they deal with the relationship between people and things, the differences between forms are often based on in personam attributes. For example, although any property scholar would acknowledge that the four different kinds of tenancy constitute four different property forms (see, e.g. id., 11), the differences between them all deal with the in personam relationship between landlord and tenant. The same can be said for differences between, say, fee simples determinable and fee simples subject to condition subsequent. Non-contractual differences in in personam relations that affect the in rem relationship between person and land are widely acknowledged as sufficient basis for the recognition of a different property form.

and uses, on the other hand, limits the information costs outsiders must incur should they wish to bargain for certain rights, including ownership, over someone else’s property: the fewer the number of possible property interests one must investigate to understand the property’s current legal situation, the lower the information costs.7

7

Looking more closely at this framework, one can detect two different kinds of information costs at work: first, there are information costs related to potential trespass, the kind incurred by “true” third parties who are not interested in legally obtaining rights over the property currently held by the owners (if there are any)—in other words, who are not interested in making transactions with the current owners. Their interest lies in figuring out whether they would be legally sanctioned for unilaterally entering, using, or seizing the property. See Smith (2012, p. 1706) (discussing “third party information costs”). Second, there are information costs that are specifically related to potential transactions: if I were to purchase Blackacre, for example, I would want to know whether it has any liens attached, if there are any future interests, or if there are tenants on the grounds—anything that might burden my potential ownership. Smith has sometimes, but not consistently, used the term “measurement costs” to describe these information costs. Id.

There are, of course, significant costs to standardization and exclusion rights, and Smith and his co-authors have, much to their credit, always incorporated these costs into their models. They discuss these costs using a variety of terms: the costs of standardization are often described as “frustration costs,” in that the numerus clausus systemically “frustrates” a certain number of individual preferences (Merrill & Smith 2000); in the exclusion context, on the other hand, they tend to measure the “cost of exclusion” in terms of loss of “precision,” in that a structural emphasis of exclusion rights hampers finer and more “precise” regulation of resources. To some extent, the latter, too, is a kind of “frustration cost,” as Smith himself has occasionally alluded to (Smith 2012): what are being “frustrated” are governmental or communal preferences for more precise regulation, in favor of a standardized emphasis on exclusion. From that perspective, exclusion is a different species of standardization—operating through a default rule rather than through mandatory rules, but standardization nonetheless (Wyman 2018).

The level of standardization and exclusion that any specific property system ultimately reaches depends, therefore, on how lawmakers—legislatures, primarily, but also some judicial actors—balance information cost savings against frustration costs under whatever socioeconomic circumstances they live under. Although the model thus defined has two primary variables, information cost theory, as its name suggests,8

8

It should be reemphasized that the term “information cost theory” was initially proposed by Smith and Merrill themselves, and has been used repeatedly in Smith’s later works.

has consistently focused more on one of them: it has always been more interested in how legislatures react—or should react—to changes in information costs and, it is probably fair to say, has largely relegated frustration costs to the theoretical background.

Like most other theories of property, there are both normative and explanatory dimensions to information cost theory: it generates some prescriptions on how legal systems should react to information cost shocks, but also strives to explain how they actually respond in the real world. The latter, in particular, gives information cost theory much of its appeal over competing theories. Whereas theories centered on personhood or “human flourishing” can be just as normatively compelling, they are rarely able to generate testable empirical predictions about patterns of legal development. In comparison, the functional nature of information cost theory and its predominant focus on a single explanatory variable allows it to do substantially better on this front.

Correspondingly, several of Smith’s most influential articles, particularly those co-authored with Merrill and Chang, have focused on explanatory and empirical arguments, rather than normative ones. Their more formal attempts at microeconomic modeling generally seek to illustrate what happens when legal systems experience shocks to property-related information costs. Assuming that legal systems behave rationally, there should be a generally positive correlation between information costs, the intensity of legal standardization, and the strength of exclusion rights: the higher information costs are, all other things beings equal, the greater the economic demand for tighter numerus clausus-type standardization and for strong rights to exclude. A negative shock to information costs would therefore lead to weaker standardization and exclusion rights, whereas a positive shock would strengthen both (Merrill & Smith 2000; Smith 2004; Chang & Smith 2015). On the empirical front, several of these articles present narratives in Anglo-American, Continental, and East Asian legal history that potentially support this basic theory (Smith 2004; Chang & Smith 2012, 2015). Smith’s analysis of the decline of customary law represents another attempt to trace changes in property regimes over time (Smith 2013).

The most common criticism of this framework is that, essentially, “there are other factors”—that the framework’s somewhat single-minded focus on information costs fails to properly account for other influences on property law that may push against standardization and exclusion. As noted above, critics vary widely in what, precisely, they think those influences are: it can range from personhood to individual liberty, to social support for democracy and deliberative governance. Even if we stay within a more conventional law and economics framework, there clearly are, as Merrill and Smith readily acknowledge, utility costs to limiting the number of property forms, or to strengthening the right to exclude.

There are a number of response strategies available to information cost theorists. The easiest and least interesting one would be to limit the scope of their claims: they could claim that the point of their framework is simply to highlight the information cost benefits of standardization and exclusion. Whether those benefits outweigh other considerations is beyond the scope of the framework. This would essentially gut the framework of any predictive power, and is clearly not the strategy that Smith and other proponents have favored. Rather, their response has largely been to argue that, in a number of critically important legal contexts, information costs tend to matter more than other factors.9

9

The most systemic argument to this effect is probably made in Klick & Parchomovsky (2016), which finds that stronger rights to exclude increase property values. This is not quite the same thing as stronger rights to exclude reducing information costs, but it is at least consistent with that argument.

Correspondingly, Smith has argued that modern Western property law, in both Common Law and Civil Law systems, has consistently and prominently featured both exclusion-based property regimes and numerus clausus-type standardization systems, and that these features are the product of information cost-driven considerations (Chang & Smith 2012, 2015).10

10

Smith and Merrill have also claimed that the need to control information costs through exclusion strategies and standardization is something that distinguishes property from contract: being “against the world,” property rights innately creates higher information cost than do contractual arrangements between a limited set of parties (Merrill & Smith 2000, 2001). This kind of comparison is inherently fraught with conceptual and empirical difficulties. Although it seems plausible at first glance that property use involves greater third-party information costs than do contracts, a closer look reveals a number of complications: insofar as the difference between property and contract is that property is “against the world,” this difference is more a consequence of the right to exclude than a reason to have it. Property rights are “against the world” in a legal sense if and only if they are exclusive to the person holding the right. In a pure open access regime, in comparison, property rights would not be against anyone, and because third parties would not have to plan their uses around preexisting ones, there would be few third party information costs. It is more coherent to argue that the exclusive nature of property rights, relative to contractual ones, places greater pressure on them to standardize, but this, too, has problems: as noted above, standardization is a reaction against “measurement costs,” rather than “third-party information costs.” The fact that some property rights are “against the world” is not, however, necessarily a major source of “measurement costs”—if anything, the usually exclusive nature of property ownership and use may actually reduce them.

The fundamental question is not, however, whether standardization and exclusion exist, or even whether they occupy a central position in any given property regime, but whether they react to changes in information costs in ways predicted by the theory. These features can exist for reasons other than information costs—ideology, social values, path dependency, and so on—so the mere fact that Western property regimes have them does not move the empirical needle in favor of information cost theories. Rather, there need to be qualitative or quantitative case studies where we can observe the interaction between information costs and property institutions. On this issue, the empirical argument between information cost theorists and their critics remains inconclusive: neither side has offered much empirical evidence capable of substantively refuting the other side. Smith has, as noted above, outlined a number of case studies that could potentially aid his theoretical agenda, but they are, at this stage, somewhat embryonic.

This article will provide several additional case studies, but it recognizes that the empirical debate will likely remain inconclusive for quite some time. Instead, its primary objective is to rethink and expand upon some of the central theoretical components of the information cost framework, especially those that relate to standardization—it also has much to say about exclusion, but more as a lemma than as a main theorem. It does so by identifying sociopolitical connections between information costs and frustration costs that, due to the preexisting literature’s preoccupation with information costs, have gone almost completely unnoticed.

At the core of the information cost literature lies a set of theoretical models in which the costs and benefits of standardization undergo external shocks independently of each other: for example, in their now canonical article on the numerus clausus, Merrill and Smith build a basic model where information costs increase or decrease due to external shocks, but frustration costs remain unaffected (Merrill & Smith 2000, p. 41). A similar model is used in a later article by Chang & Smith (2015). That information cost and frustration cost curves can shift independently of each other is, in fact, a crucial underlying assumption of these articles and their core claims. Without it—if shocks to information costs are typically accompanied by shocks to frustration costs—not only would we have to question whether the predicted positive correlation between information costs and standardization is theoretically sound, but even if it is, we would still have to question whether information cost shocks are truly the primary catalyst behind the correlation.

The central thesis of this article is precisely that, as a matter of social and political theory, changes in information costs tend to come bundled with changes in frustration costs—and that this challenges some of the core predictions made by preexisting information cost models. The claim, in other words, is not just that “there are other factors,” but rather that “there are other factors that tend to change when information costs do, and in ways that often cancel out or override some of their institutional effects.” It may well be true that, all other things being equal, higher information costs lead to higher levels of standardization, but in many sociopolitical contexts, all other things probably cannot be equal, even in theory.

To understand why, we must look beyond information costs to their origins. The question we need to ask is not “how do information costs affect property law,” but rather “what are the sociopolitical processes through which information costs are created or eliminated, and how do those processes affect property law.” Reframing the inquiry in this fashion unearths a set of analytical connections between information costs and frustration costs that significantly revise the current theoretical paradigm, allowing it to explain some major historical phenomena that it previously would have struggled with.

It would not be entirely incorrect to characterize these efforts as simply filling out the underdeveloped frustration cost component of preexisting models: the article is quite happy to accept the basic intuition that standardization and exclusion can effectively lower information costs—and even that controlling information costs is perhaps the primary reason why legal systems have a numerus clausus at all—but probes more systemically into the frustration costs of doing so. That said, the probe does, in the end, add to the preexisting framework qualitatively, and not merely quantitatively: it does not simply provide a theoretical add-on that can be plugged in or detached whenever convenient, but instead demonstrates that the two kinds of costs often cannot be studied in isolation. In the important cases that this article focuses on, the two kinds of costs are fundamentally connected both socially and politically, and any attempt to model the interaction between information costs and property law must take this into account. Moreover, the first step in the probe is actually to develop a deeper sociopolitical account of information costs themselves, and so if the article is to be characterized as an expansion of the preexisting model, then it is, at the very least, an expansion in both horizontal and vertical dimensions.

3. A SOCIOPOLITICAL THEORY OF INFORMATION AND PROPERTY

This section develops a sociopolitical theory of information costs that identifies their connections with communal solidarity, state capacity, and the diversity of individual preferences, both actual and expressed. The eventual takeaway from this exercise is that, in many cases, information costs are synchronized with preference diversity, and therefore with the preference frustration costs that any fixed level of standardization or exclusion generates. In other words, in these cases external shocks to frustration costs tend to occur whenever external information costs shocks do, and generally in the same direction. This qualifies some core predictions of information cost theory, but reinforces others: external changes in information costs and frustration costs tend to push in opposite directions on the issue of standardization, but are more likely to push in the same direction on exclusion. When information and frustration costs derive from the same underlying socio-political forces, information costs have no clear correlation, whether positive or negative, with the number of legally allowed property forms, but should be positively correlated with the strength and scope of exclusion rights.

3.1 From Information Costs to Preference Diversity

3.1.1 The Sociopolitical Origins of Information Costs

To begin with, we need to develop a thicker account of information costs. To some extent, information costs are simply natural products of property ownership, use, and transfer: as long as there are property rights, and as long as people rely on them to engage in social and economic activity, their mere existence will generate some information cost. But this tells us very little about when and how property-related information costs change. What, specifically, are the social, economic, or institutional processes through which information costs increase or decrease?

Part of the answer, of course, is that information costs react to legal or semi-legal institutions: they are affected by how clearly property rights are delineated, and by how easy it is to understand relevant laws or customs. Contradictory or confusing laws increase information costs, while coherent and simple laws presumably lower them. This is a core premise of mainstream information cost theory, but this is obviously not the answer we are looking for: we cannot explain how legal systems react to information costs by examining only those information costs that are created by legal systems themselves. The shocks that matter here are those that are external to the legal system.

A productive way to begin this inquiry is to think about how the average person obtains information about property rights. If, say, we would like to accurately identify the boundaries of Blackacre, both physical and legal, what would be the sources we could tap? First, we could look to the owner’s own demarcation: walls, fences, “keep out” signs, and so on. This would presumably give us a basic idea of the physical dimensions of the estate. They would not, however, immediately tell us what its legal boundaries are: whether we are entitled to enter or use parts of the estate, or whether the land is encumbered by any mortgage or servitude to another party. A fence or a wall tells us that the owner would prefer to keep us out, but they do not tell us whether she is legally entitled to do so. For the latter kind of information, we must generally rely on either the local community or the government.

Local communities are, of course, the best source of information on local customs,11

11

There are many ways in which communities can collect and disseminate property-related information: word-of-mouth can be highly effective when the community is small and stable. Alternatively, local communities often keep written records, and may make them available to members or affiliates. Social or economic elites can be hubs of information sharing and communal rule enforcement within larger groups, as can professionals or groups of professionals—think, for example, of real estate agents in modern societies, and institutions such as the Law Merchant in early or pre-modern ones. See, e.g.Kadens (2012) (arguing that the Law Merchant was primarily an information collection and sharing mechanism, rather than a source of actual regulation); and Ellickson (2001) (discussing the role of elite information sharing in communal governance).

which continue to exert significant influence even in developed countries (Ellickson 1991; McAdams 1997), but are also, compared to the landowner herself, a superior source of legal information: the collective incentive to misinform is much lower than the incentive of an interested landowner. Communal consensus on a certain legal question—where it exists—will likely be a reasonable approximation of the actual legal rule, and can be a valuable lead even when it is not completely accurate. Moreover, if communal consensus actually contradicts the legal rule, then that would suggest a certain level of law-custom conflict, which is something that we would almost certainly want to know about.

Government entities, on the other hand, collect and disseminate information in more formal ways (Miceli et al. 2000; Arruñada & Garoupa 2005). The history of government record-keeping on landed assets is probably about as old as the history of government itself: land and agricultural taxes were the single most important source of revenue for nearly all pre-modern states, which created enormous fiscal incentives to maintain an up-to-date official land registry. Once this information is collected, governments have at least some incentive to share it with the general population, in order to preempt disputes and facilitate efficient exchange. American property law courses, for example, introduce students to recording and registration acts, which remain the primary means of obtaining legal information in contemporary real estate markets (Bell & Parchomovsky 2016).

Returning to the Blackacre scenario, our ability to obtain information is therefore dependent on three factors: the supply of information by the owner, the efficacy of communal information sharing, and the state’s ability to collect and disseminate information. Of the three, the owner is, as alluded to above, by far the least reliable source. She has significant incentives to exaggerate her legal rights, and perhaps to downplay existent servitudes. Owner-supplied information is, of course, much more reliable when there are significant social or legal sanctions, reputational or material, against exaggeration, but in such cases its accuracy is essentially being bolstered by the efficacy of communal or state governance: the easier it is to obtain accurate information from communal or governmental sources, the lower the odds of private misrepresentation. In stable societies with high levels of social or political trust, we might have an intuitive trust of owner-supplied information, but that is a downstream product of social or political cohesion.

As a result, if we are looking for external shocks to information costs, then it makes sense to focus almost exclusively on communal and governmental sources, instead of owner-supplied information, if only because the usefulness of the latter almost never changes independently of the other two. Apart from putting in the basic effort to build fences or post signs—things that are near-universally achievable in almost all human societies—there is very little signaling the individual landowner can do that does not substantially rely on social, legal, or political context to function. Changes in the effectiveness of private demarcation are therefore closely tied to changes in social or political trust.

Mainstream property theory has long acknowledged that there is a close and perhaps indispensable connection between information costs and sociopolitical organization.12

12

This is perhaps most apparent in the wave of articles and books produced in the 1990s and early 2000s on the role of social norms in regulating economic activity: a number of major figures in that literature, such as Robert Ellickson (1991), Richard McAdams (1997), and Eric Posner (2000), have written at some length about the link between closely-knit communities and low property-related information costs. For many of these scholars, information sharing between members of a closely-knit community is perhaps the paradigm of a low-information cost society. Smith, too, has written repeatedly on the connection between the solidarity and size of a group and the effectiveness of information exchange within it, generally arguing that information costs and group size are positively correlated (2004; 2013). A more recent article by Abraham Bell and Gideon Parchomovsky explores the government’s role in supplying information through recording and registration systems (2016). As a general matter, the link between information costs and sociopolitical organization is hardly a controversial one, whether among legal scholars or institutional economists.

Scholars generally believe, with good reason, that there is an inverse relationship between property-related information costs and what might be described as “sociopolitical cohesion”: all other things being equal, the closer a community, the lower the information costs—and similarly, all other things being equal, the stronger the administrative capacities of a state, the lower the information costs within its jurisdiction (Ellickson 1993; Bell & Parchomovsky 2016, p. 273). It is, of course, fully possible for governments to strengthen as communities disintegrate, or for communities to strengthen as governments weaken, and in such cases the overall balance of both “sociopolitical cohesion” and information costs are relatively ambiguous. That said, in cases where states and communities both weaken, scholars generally expect information costs to rise, and when they both strengthen—for example, during recoveries from major armed conflicts—information costs should unambiguously decrease. These are intuitive arguments: closer social ties obviously facilitate information sharing among community members, and stronger states will have, almost by definition, stronger information collection and dissemination abilities than weaker ones.

Three complications merit immediate discussion: first, rising sociopolitical cohesion may allow states and communities to interfere with private use and ownership more actively and arbitrarily, leading to higher information costs.13

13

This is the kind of concern that led institutional economists to emphasize that, for economies to grow, states must be able to make “credible commitments” to respecting private property. See, e.g. North & Weingast (1989); North (1993); Stasavage (2002).

States and communities are therefore capable of both reducing and increasing the information costs related to property. For example, a stronger state could provide more information by creating better recording or registration systems, but could also exercise eminent domain or pursue regulatory takings more frequently and aggressively, thereby creating additional uncertainty for current and future landowners. Even so, the information cost benefits of rising sociopolitical cohesion will, in all likelihood, substantially outweigh the costs: communal and state activities—those related to property, at least—are almost always much more visible, and usually more predictable, than private activity. The cost of monitoring communal or governmental interference will therefore almost certainly be much lower than the cost of tracking private activity in the absence of effective communal or state support. Moreover, it is very rarely in a government or community’s self-interest to behave so arbitrarily and unpredictably that property-related information costs rise dramatically. Quite the opposite, they are usually rewarded, socially and politically, for institutional stability.

Second, network theory does not necessarily predict that information acquisition always becomes easier when social networks tighten. Rather, some models would predict that certain kinds of non-sensitive information—casual gossip, for example, or information about resources not possessed within the network—may be easier to obtain, all other things being equal, when network ties are looser (Lin 1999; Perry-Smith & Shalley 2003). In other words, some kinds of information costs may be positively correlated with social cohesion. That said, when it comes to complex and highly specific information about valuable resources possessed within the network, which is the kind of information that property-related negotiations demand, mainstream network theory generally predicts a negative correlation between social cohesion and information costs (id.). This would seem to validate the basic intuitions of mainstream property theorists.

Third, and perhaps most importantly, the supply of property-related information may occasionally be influenced by exogenous factors that are unrelated to sociopolitical cohesion. For example, information costs can theoretically increase in response to breakthroughs in production technology: the possible uses of land may expand as new technology becomes available. Such scenarios are not terribly common—incremental technological changes rarely have a significant effect on legal information costs14

14

It is surprisingly difficult to think of a historical shock to use or production-related technology that, completely independently of its impact on social or political organization, had clear and significant consequences for legal information costs. Much more often, production or use shocks influence information costs through modifying sociopolitical organization. For example, increased market demand for livestock in an agrarian society often leads to more land being used for grazing, which tends to encourage the creation of commons or semi-commons to generate economies of scale. Smith (2004); Ellickson (1991). This process does not, in and of itself, necessarily generate a significant information cost shock, but the greater economic demand for commons and semi-commons tends to draw grazing communities closer, which can lead to lower information costs over the long run. Another example would be the advent of automobile-based transportation: the availability of new and more efficient long-distance transportation options could either lead to the disintegration of traditional communities, or the creation of new and highly homogenous suburbs within commuting distance of urban centers. (Fischel 2004). In both examples, the chain of causation runs from an external technological shock to changes in social organization to information costs.

—but they obviously exist. The question, then, is whether they affect our theoretical model. The answer, for the most part, is “no.” Such exogenous shocks will likely have a reverse effect on sociopolitical cohesion: higher information costs damage the ability of states and communities to monitor and control the economic activities of their members, leading to lower sociopolitical cohesion, whereas lower information costs will reinforce it. Changes to information costs are, therefore, either the product of or, less frequently, the cause of changes to sociopolitical cohesion, but the two factors are negatively correlated in either case. What really matters for the model is this negative correlation—the argument that rising (or falling) information costs are almost always accompanied by falling (or rising) sociopolitical cohesion—rather than the direction of the causal arrow.

How does this negative correlation affect information cost theory? As discussed in Section 2, information cost theorists argue that information costs are positively correlated with rights of exclusion and legal standardization. Now, they may be completely correct that rising information costs incentivize standardization and exclusion, but even so, there will only be a positive empirical correlation between the two sides if rising information costs are not routinely offset by some other phenomenon that disincentivizes standardization and exclusion. This is where the negative correlation between information costs and sociopolitical cohesion comes into play: it has the potential to generate this sort of offsetting phenomenon. In other words, if changes to sociopolitical cohesion have other consequences that systemically dilute the demand for either standardization or exclusion, then the predicted correlation between these legal features and information costs falls apart.

The remainder of Section 3 identifies one such consequence. Apart from raising and lowering information costs, sociopolitical cohesion shocks also change the diversity of property-related individual preferences: subject to controls on social culture, political ideology, and the overall level of economic development, closely-knit communities and stronger states tend to have a lower level of preference diversity than do looser communities and weaker states. Moreover, these consequences will, in theory, substantially interfere with the influence that information cost considerations wield over property law.

3.1.2 Preference Diversity and Communal Cohesion

Assumptions about preference diversity underlie much preexisting scholarship: for example, information cost theory believes, surely correctly, that the socioeconomic cost of standardization is affected by the quantity and importance of the individual preferences that numerus clausus regimes frustrate (Merrill & Smith 2000, pp. 35–42). Scholars who identify with the relatively new “progressive property” movement, on the other hand, emphasize the plurality of human preferences and values, whether across communities, or within individuals (Alexander et al. 2009; Dagan 2011). Neither side, however, has looked more closely at the origins of preference diversity: how do new preferences emerge, and how do old ones die out? In particular, the connection between preference diversity and sociopolitical organization has gone largely unnoticed, despite its growing importance to other fields of legal scholarship.15

15

This state of affairs likely has something to do with the general reluctance in modern microeconomics to theorize about the formation and evolution of preferences; instead, utility functions are conventionally treated as “black boxes” that are beyond the domain of economic analysis. Becker & Murphy (1988); Stigler & Becker (1977). Law and economics scholars often share this reluctance, and although the increasingly prevalent reception of cognitive science and behavioral economics has begun to challenge this, it has yet to make as much an impact on property theory as on other fields. Progressive property, on the other hand, is simply a younger and, for now, less developed intellectual movement.

There is, for the most part, a strong positive correlation between preference diversity and the preference frustration costs generated by legal standardization. All other things being equal, greater diversity of individual preferences directly increases the total frustration costs of standardization: if preferences diversify under a fixed set of property forms, the likelihood that the preexisting set will be able to accommodate each individual’s preferences decreases, and the resources that the state must spend to enforce compliance will increase.16

16

One could argue that combinations of a few simple forms (fee simple, tenancies, trusts, etc.) can accommodate most preferences that people hold, see Merrill & Smith (2000, pp. 37–38), but insofar as the legal system is truly concerned about information costs, it needs to deny at least a few forms that people would plausibly want for there to be any information cost savings—there would otherwise be no reason for rational economic actors to seriously investigate them in the first place, and therefore no information costs saved by eliminating them. More of these plausibly desirable forms would likely be denied by the preexisting numerus clausus as preferences become more diverse, driving “frustration costs” up.

The correlation between preference diversity and the marginal frustration costs generated by cutting each individual property form is somewhat weaker, but still positive: when individual preferences diversify, marginal frustration costs may decrease for some basic property forms, but will tend to increase for most forms. These interactions are elaborated on in subsection 3.2 below.

Before proceeding any further, a number of conceptual clarifications should be made: first, “preference diversity,” as applied in this article, is a collective measure, rather than an individual one. It is a measure of how diverse individual preferences are within a collection of people, rather than how complex an individual’s preferences are. For example, a 1000-person society in which each person only prefers to hold their land in one form of ownership, but everyone prefers a different form, has much more preference diversity than a similarly sized society in which each person alternates between three forms of somewhat desirable ownership, but everyone wants the same three forms.

Second, there is an obvious conceptual difference between actual preferences and revealed or expressed preferences: I may secretly want A, but due to social stigma, feel compelled to say that I want B. A, not B, is my actual “preference,” whereas, depending on how good my self-control is, B might be my only “expressed” preference, and if so, then I would have no “revealed” preference at all. As discussed below, both actual preference diversity and expressed preference diversity are likely synchronized with information costs. In fact, either kind of synchronization would generate the central prediction that rising information costs do not correlate with increased demand for standardization, although the behavioral theories involved would be different, as would the normative implications. The summaries provided in the previous paragraph and in the Introduction therefore speak generally of “preference diversity,” but the more detailed discussion below will differentiate between actual and expressed preference diversity.

Third, “preference” refers here to individual preferences on personal property ownership and use—what I want to do with my own property—rather than to preferences on societal behavior—what I want others to do with their property. This is not to say that societal preferences cannot be analyzed through the same framework that is applied here to personal preferences, but only to narrow the discussion so that an already complicated article does not become unbearably so. Were societal preferences incorporated, they would almost certainly strengthen, rather than weaken, the model constructed below: the more homogenous (or diverse) personal preferences are, the more homogenous (or diverse) societal preferences probably are, and vice versa.17

17

To quickly elaborate: when personal preferences diversify, societal preferences likely diversify as well, which weakens the sociopolitical consensus—whether within a community or a polity—on how to regulate property, which then further facilitates personal preference diversification. A similar chain of analysis applies to homogenization.

The two kinds of preferences are therefore much more likely to mutually reinforce than to mutual dilute. That said, such an analysis is unnecessary for our present purposes, and therefore best left for future work.

A fairly robust literature on cultural cognition and network theory has long argued that individuals within social groups tend to think similarly, to the extent that they are often impervious to external attempts to change their minds. There are various versions of this argument: some believe that the tendency towards “group think” is simply a cognitive shortcut that human beings inevitably make in circumstances of incomplete information or limited rationality (Douglas & Wildavsky 1982; Douglas 1986; Sunstein 2000; Cohen 2003). Others argue that there are utilitarian benefits in conforming to group consensus (Kuran 1996; Kahan & Bramann 2006). The former mechanism leans towards portraying group think as a form of genuine, if potentially irrational, persuasion, something that generates no particular benefit for the individual, but is rather a product of his cognitive limitations. The latter mechanism, however, suggests that adhering to group consensus generates social benefits for individuals—or rather that challenging it generates costs and risks—and therefore that it may be somewhat rational for self-interested individuals to, perhaps subconsciously, seek out and agree with the opinions of fellow group members. Over time, what may initially have begun as an exercise in self-interested reputation enhancement becomes mentally internalized: I may be subconsciously following the group consensus for utilitarian reasons, but consciously, I do, in fact, “believe” in it. The closer the group, the more information group members have about each other, which would presumably strengthen both mechanisms.

Relatedly, social scientists generally believe that the ability of communities to enforce norms or standards is heavily dependent on the durability and closeness of their social fabric (McAdams 1997; Bendor & Swistak 2001; Ellickson 2001). Among property scholars, the conventional explanations for this belief are based on versions of rational choice theory: closer communities facilitate information sharing and predictable social interaction among members, which lessen the likelihood of collective action and free rider problems (Ellickson 2001, 1991). In any given group, members have a number of possible incentives to obey and help enforce group norms: they may want to avoid material or reputational sanctions, they might hope to acquire higher sociopolitical status through acts of compliance and leadership, or they may wish to signal a willingness to cooperate on other issues. All such incentives are strengthened when social interaction is more regular and information shared more readily: violations are easier to identify, reputational sanctions are easier to communicate, and signals become more credible. But the processing of these incentives need not be conscious: subconscious processing may lead to a conscious belief that one genuinely believes in the norm.18

Eventually, the subconscious processing may phase out entirely, leaving an internalized and entrenched belief that the norm is simply right, either as an ethical matter, or as “common sense.”

These theories yield both an account of how group cohesion influences actual preference diversity, and a theory of how it influences expressed preference diversity: as discussed above, the former predicts that “group think” is more likely to arise, all things being equal, in closer groups than in looser ones. Individuals are more likely to share the same preferences when they know more about the preferences of other groups members—and, at the same time, enjoy less privacy on their own preferences. The latter, on the other hand, is more complicated: the former could, of course, lead to the latter, in the sense that, if actual preferences become less diverse and the group becomes tighter, then expressed preferences would almost certainly become less diverse as well. That said, the latter could also be true even if the former were not: even if we assume that actual preferences remain the same regardless of how close the group is, we would nonetheless expect to see less expressed preference diversity in closer groups, if only because the social cost of dissent is probably higher when social ties are closer.

How do these arguments map onto the property law context? As a general matter, property-related preferences are simply a subset of individual preferences, and are likely subject to the same kinds of conscious and subconscious social influences.19

19

I have argued elsewhere that, in agricultural economies where land is of critical importance to household subsistence, individuals are less likely to explicitly moralize the regulation of real property, and instead tend to bargain over them in nominally pragmatic and self-interested ways (Zhang 2017). This does not, however, imply that their perceptions of self-interest and “common sense” are impervious to communal group-think. Quite the opposite, it is completely possible—likely, even—that self-interested individuals are no less prone, for subconsciously self-interested reasons, to internalizing group norms as their own preferences.

Imagine, for example, a landlord who buys land and moves to a county where all other landlords, all of whom are friends with each other, expressly dislike granting tenants the right to sublease. The new arrival is initially ambivalent on subleasing, or may even have allowed it in the past, but as he assimilates into the community of local landlords, he may very well develop a genuine dislike of subleasing—either because he subconsciously decided that it was in his interest to agree with preexisting communal practice, or because, in the absence of detailed knowledge about local leasing markets, he began using preexisting communal practice as a mental proxy for “common sense.” When communities weaken, both scenarios become less likely: the subconscious pressure to conform lessens, as does the cognitive imprint of preexisting communal practice—for example, the new arrival is simply less likely to recognize it as a prevalent practice, given the lower level of communal social interaction.

The argument that weaker communities lead to greater expressed property-related preference diversity is even easier to make. Mental processing of social pressures could be subconscious, but of course it could just as easily be conscious—in which case, the new arrival would not be changing his conscious evaluation of subleasing, but might nonetheless choose to disallow it to in order to “fit in” with other landlords and enhance his reputation among them. His incentive to do so diminishes as the community disintegrates and the socioeconomic benefits of conformity decrease, or, if one prefers a cost and sanctions angle, because the reputational costs of non-conformity decrease. The two mechanisms—subconscious versus conscious processing of social pressure—involve different mental processes and different “speeds” of thinking, but their external manifestations are largely identical.

Human communities have, of course, a long history of controlling the property-related preferences of members in far more explicit ways than cognitive nudging or the subtle application of social pressure. They openly regulated everything from inheritance practices to mortgage instruments to cropping patterns, and in most pre-modern societies were a stronger and often more exhaustive and well-enforced source of economic regulation than state legislation and policymaking (Ellickson 1993; Zhang 2016). Modern communities in Western countries—with important exceptions—tend to interfere much less in the property-related behavior of their members, but nonetheless continue to impose a significant range of rules: formal covenants, conditions, and restrictions, for example, are extremely commonplace in suburban America (Nelson 2009; Fischel 2013; Ross 2016). Communal regulation of property-related preferences remains, of course, a powerful social phenomenon in developing countries.20

20

Studies have shown, for example, that squatter communities in some of the most heavily urbanized parts of China exert tremendous control over the use, leasing, and transfer of informal property rights (Qiao 2017).

One would expect individuals to conform to communal regulation more rigorously than they do to non-normative customary practices, but the possible mental processes are similar: they might internalize the norms as common sense, or, despite private dissent, obey them to avoid sanctions. In either scenario, the individual incentive to conform strengthens when communal cohesion increases, and weakens when it decreases.

In general, then, we should expect increases in communal cohesion to homogenize preferences, both actual and expressed, and decreases to diversify them. This argument is subject, however, to two major controlling assumptions: first, it assumes some basic ideological or cultural continuity within the community. Consider, for example, a scenario in which a community that previously attempted to ban certain kinds of property use collectively adopts a norm of mutual non-interference as it grows closer. Cognitive nudging and the potential for “group think” would presumably strengthen as communal cohesion increases, but can potentially be more than offset by the collective decision to regulate less. The best response is simply to acknowledge the theoretical existence of these hypothetical exceptions and to control for them—that is, to argue that higher levels of communal cohesion tend to reduce preference diversity, so long as the community’s cultural willingness to interfere with property-related activity remains stable. This is not a terribly problematic control: it seems much more likely that communities shifting from active regulation to express deregulation will be losing cohesion, rather than gaining it.

Second, the argument also assumes a basic level of continuity in what one might call “socioeconomic feasibility,” defined roughly as the range of legal options otherwise available to property holders if communities or states do not attempt to ban or stigmatize them. For example, taking out a mortgage is only a feasible legal option if there are financial entities that are willing and able to lend money in exchange for collateral, which itself assumes a certain level of socioeconomic development. Such development depends on many things, including natural resources, climate, and technological innovation, but it also depends, to some extent, on sociopolitical cohesion. If, for example, state and communal institutions completely collapse after a severe war or natural disaster, then socioeconomic development may recede to the point where many institutional options that were previously available to property holders now become infeasible even in the absence of regulation or any form of “group think.”21

21

One could frame this as a sort of network effect: some property forms depend on the existence of positive network externalities—and therefore, in the absence of such networks, will cease to be socioeconomically feasible. This argument has most prominently been applied to contract theory, although it likely applies to property as well. See, e.g. Klausner (1995); Kahan & Klausner (1997); Lemley & McGowan (1998).

Banks cease to lend, rental markets collapse, and the menu of possible property forms is fundamentally diminished despite a sharp decline in communal and state influence over individual preferences. At the same time, this kind of forced homogeneity also leads to sharp declines in information costs: when almost everyone “voluntarily” does the same thing, it is cheaper and easier to identify and measure rights, to the extent they still exist, and legal standardization becomes both unnecessary and perhaps impossible to implement.22

22

The Merrill and Smith numerus clausus theory (2000) argues that the state must actively limit the number of property forms when information costs increase because the social cost of people making use of some preexisting forms outweighs the benefits. This assumes that, in the absence of legal prohibitions, people will be able to continue using those forms if they so choose. But if people voluntarily stop using those forms because of socioeconomic, rather than legal, impossibility, then that would render the numerus clausus mechanism unnecessary. The law could actually recognize additional property forms without creating any increase in information costs, simply because people would still be unable to make use of them.

As scholars have pointed out, when sociopolitical disintegration is severe enough, there sometimes are no state or communal institutions left capable of enforcing boundaries or the right to exclude, and private property itself ceases to exist (Fitzpatrick 2006). Notably, such circumstances were fairly widespread in a number of war-torn African countries in the later twentieth century.

Unusually severe declines in sociopolitical cohesion may therefore lead to lower information costs and less preference diversity, and perhaps to the disappearance of legal standardization and the right to exclude altogether. In such scenarios, communities and states play a constitutive, rather than a limiting, role vis-à-vis property-related activity, which creates obvious complications for the theoretical arguments made above. The best and perhaps only solution is to limit the applicability of those arguments to cases where preexisting property forms continue to be socioeconomically feasible following the external shock—in other words, to cases where sociopolitical cohesion changes, but not catastrophically. This would rule out some cases of extreme governmental and communal collapse, but such cases are fortunately unusual by nature: most declines in communal cohesion or state capacity are not accompanied by anything close to “legal deindustrialization,” and therefore lead to more preference diversity, rather than less. In fact, as discussed in Section 4, even major shocks such as civil wars or plagues can nonetheless leave parts of the sociopolitical fabric sufficiently intact that the preexisting menu of feasible property forms continues to exist and function.

In summary, assuming we control for “socioeconomic feasibility” and regulatory culture, communal cohesion—as measured by the community’s ability to monitor and influence its members—is usually negatively correlated with both preference diversity and information costs.

3.1.3 Preference Diversity and State Capacity

States are, in a basic Weberian sense, stable and centralized administrative entities to which large groups of people grant a monopoly over the legitimate use of force (Weber 1921, p. 29; Parsons & Weber 1964, p. 154). Perhaps even more than local communities, they are fundamentally concerned with the regulation of property-related activities: there is a long history of political theorists arguing that property regulation was one of the primary reasons to create states in the first place. States, including all liberal democracies that otherwise adopt a non-interventionist attitude towards private socioeconomic behavior, therefore regulate property-related activities and supply information about them at the same time. Furthermore, their capacity on both fronts depends on their financial resources, the competence of their agents, the available technology, and the effectiveness of their organizational infrastructure. As with communal regulation, individuals who obey state regulations can do so for a number of reasons, including the internalization of property law as “common sense” or moral commandment, the conscious or subconscious desire to signal cooperativeness, or the straightforward fear of government sanctions (Lessig 1998).

Formal regulation is, however, only one of many ways through which states influence the formation and expression of individual preferences. Most states govern populations that are far too large to be “closely-knit” in the conventional sense,23

23

Scholars do argue that some societies, most notably Japan, enjoy a significantly higher level of sociocultural homogeneity than others, although this point is hotly contested. Cf.Lie (2000); Haley (1998); Burgess (2004).

but they nonetheless play an important role in the formation of group consensus and sociopolitical identity at many levels, ranging from the very local to the completely national. At the most abstract level, states not only provide forums and platforms for public discussion—examples include elections, notice and comment mechanisms, and, in many countries, state-run media organizations—but also provide common issues for people to talk about (Wildavsky 1987; Lewis & Cullis 1988; Tomer 1996; Druckman & Lupia 2000; Cohen 2003). People may disagree vehemently over, for example, the wisdom of some particular act of eminent domain, but the mere fact that they are debating the same issue already tends to homogenize their opinions. In such scenarios, public opinion easily becomes polarized, but it will likely coalesce around a few poles, whereas individual opinions would be scattered much more diffusely in the absence of social awareness and discussion. In other words, states facilitate the creation of political communities, within which the cognitive or social pressures to conform are perhaps even stronger than in most other kinds of communities.

More concretely, states are regularly involved in all levels and kinds of education, ranging from the academic to the vocational to the everyday. Few would doubt the impact of formal education on preference formation (Wildavsky 1987; Brighouse 1998; Fernández 2004), but state-sponsored “education” of economic behavior continues well beyond one’s time in school: a good example would be the consumer information websites and publications maintained by a number of American government agencies, including the Federal Trade Commission and the Consumer Financial Protection Bureau.24

Apart from supplying information and counsel, the government also influences adult economic behavior through the direct provision of financial and economic services: American government-sponsored enterprises are one major example, Chinese state-owned enterprises and banks are another. Anyone who has attempted to obtain a mortgage through the Federal National Mortgage Association or similar entities will be familiar with the myriad of restrictions they impose on debt-to-income ratios, interest rates, credit scores, and other facets of economic behavior.25 These are, of course, standards for lending, rather than legal rules backed by governmental sanction, but if government sponsored enterprises occupy a large enough share of the mortgage market, the impact on individual financial behavior will likely resemble that of direct regulation. Compliance might well be the outcome of conscious utilitarian calculations, but here, too, cognitive nudging is a strong possibility, especially over the long run.

The basic point is that states, like communities, have multiple means of homogenizing individual economic preferences, both actual and expressed, and that they would presumably become more effective, ceteris paribus, following an increase in the state’s administrative capacity vis-à-vis the economy it governs—and less effective following a decrease. Similar to the previous discussion of communal cohesion, this argument assumes a basic level of continuity, before and after the shock to state capacity, in both political ideology—specifically, the state’s desire and willingness to control individual preferences—and the level of socioeconomic development. It does not necessarily apply, therefore, to cases where states undergo major ideological swings,26

26

The most problematic scenario is one in which a state expands its administrative capacity to collect and share property-related information, but at the same time, embraces a deregulatory ideology that leads to less formal regulation. This is qualitatively similar to the aforementioned scenario in which a community shifts to a culture of non-interference as it grows closer. It does, however, seem more empirically pertinent due to the large number of modern states that explicitly embrace capitalist or free market principles of economic regulation—whereas it is rather difficult to think of a closely-knit community that expressly disavows normative or contractual interference with its members’ economic activities. That said, with the major exception of communist or socialist regimes that ideologically resist the very notion of private property, it is not immediately clear whether states that refuse to embrace free market ideals truly regulate property-related activities more aggressively than states that expressly do. Modern Western legal systems, as Smith and Merrill point out, tend to protect the freedom of contract, but not the freedom to create new forms of property. Merrill and Smith (2000). In other words, it is surprisingly difficult to identify cases where a state both withdrew from property-related regulation and boosted its administrative capacity at the same time. The most promising examples are those of socialist regimes pursuing market liberalization—China, Russia, Vietnam, and so on—but even they present significant empirical difficulties: it is not at all clear that the administrative capacity of those states vis-à-vis their respective economies and societies increased after their embracement of free market ideals. Quite the opposite, many scholars believe that marketization was accompanied by political decentralization and a loosening of state control. See, e.g., Xu (2011); Blanchard & Shleifer (2001); Qian & Weingast (1996). But seeCai & Treisman (2006) (questioning this conventional wisdom, but not disputing that the Chinese state decentralized during the 1980s and 1990s). Much more frequently, political ideology moves in the same direction as state capacity: those with a hammer will look for nails. For example, the transition from early modern states to modern states was one of administrative expansion, but also of the development of more interventionist ideologies vis-à-vis property regulation. See, e.g., Tilly 1990; Ertman 1997; Glaeser & Shleifer 2003; Vu 2010.

or where governmental collapse is so severe that it leads to “legal deindustrialization.”

Controlling for these factors, an increase in state capacity and communal cohesion will likely reduce preference diversity, whereas falling sociopolitical cohesion will strengthen it. Given that, as argued in the previous section, sociopolitical cohesion is negatively correlated with information costs, this would imply that information costs and preference diversity are synchronized: rising information costs are usually accompanied by rising preference diversity, and falling information costs by falling preference diversity. Another way to think about the relationship between these factors is that, insofar as information is a prerequisite for the exercise of power, information costs are essentially a negative measure of the capacity to influence and control. Fellow community members persuade and pressure each other through the communication of opinions and knowledge—in other words, through sharing information—and the same could be said, in a more abstract sense, of the state’s influence over individuals. Effective state control over individual preferences, whether of a persuasive or coercive nature, requires both effective information collection and dissemination: the former so that the state can formulate and enforce appropriate policies, the latter so that it can encourage or deter certain kinds of behavior. A state or community that is unable to keep information costs low is therefore unlikely to have much influence over individual preferences, not only because high information costs imply lower administrative capacity, but also because high information costs themselves are major obstacles to the exercise of influence and control, and therefore to the homogenization of individual behavior and preference.27

27

One could ask, of course, whether preference diversity reacts directly to the increase in information costs, without sociopolitical cohesion acting as a medium: will higher information costs, for example, lower preference diversity? Given that the information costs that mainstream information cost theory speaks of are generally externalities imposed on outside parties, the inclination is to say no. Social organizers and policymakers will care about the imposition of such externalities, but most individual landowners probably will not. Moreover, while rising information costs may make some forms of property ownership and usage harder to establish, they also shield property owners from external monitoring and intrusion—which would tend to boost preference diversity. For discussion on whether rising preference diversity has any feedback effect on information costs, see infra note 79.

A significant complication emerges when we consider the interaction between state capacity and communal cohesion. If the two change in the same direction, or if one changes while the other does not, then information costs and preference diversity will move unproblematically in the synchronized fashion discussed above. If, however, the state strengthens while communities weaken at the same time, or vice versa, then things become much more complicated: imagine, for example, a society in which rapid commercialization dismantles traditional communities, but which is also governed by a rapidly expanding state apparatus. Assume also that the state expands aggressively enough that, on balance, information costs actually decrease despite the erosion of communal information sharing. Under these conditions, one can imagine scenarios in which, because the state is ideologically less interventionist than local communities were, preference diversity actually increases. Note that there is no assumption here that the state becomes less interventionist as it expands in capacity—quite the opposite, preference diversity could still increase even if the state becomes more interventionist, so long as it remains less interventionist than the communities it “replaced.” In some sense, this seems to be a familiar, if problematically simplified, story of a modern capitalist state replacing traditional closely-knit communities (Polanyi 1944; Scott 1976; Greif & Tabellini 2010; Smith 2013). Although one could very plausibly argue that individual economic and social behavior is, following the rise of mass commercial culture, actually less diverse under modern capitalist states than in traditional societies, the theoretical possibility nonetheless remains, and deserves some consideration.

While a full discussion here is untenable given length considerations, there are a few ways to qualify the article’s thesis to accommodate this potential complication. First, even if such scenarios sever the positive correlation between information costs and preference diversity, they leave intact the positive correlation between information costs and the state or community’s capacity to control preference diversity. In the previous paragraph’s hypothetical, if the state were truly powerful enough to bring information costs down despite communal disintegration, then it would probably be capable of homogenizing preferences more effectively than the collapsing communities were formerly capable of. Control over information is, as discussed above, a strong measure of the administrative ability to control or influence preferences, although states or communities may very well differ in their cultural or ideological willingness to use this ability. In other words, information costs and preference diversity remain synchronized but for certain ideological or cultural contingencies.28

28

This article is perfectly happy—eager, even—to acknowledge such contingency, but it is deeply problematic for mainstream information cost theory: by positing that information costs determine the fundamental architecture of property, information cost theorists must commit themselves to a functionalist theory that excludes the influence of cultural or ideological internalization (Merrill & Smith 2000, 2001; Smith 2012).

Second, the remainder of the article focuses primarily on the “easy” cases where communal cohesion and state capacity are at least not moving in opposite directions—all case studies developed in Sections 4 and 5 fall into one of these categories. The complications discussed in the previous two paragraphs do not exist in these cases, and any change in information costs will likely be accompanied by a comparable change in preference diversity. In the “hard” cases discussed above, their synchronization may be disrupted by ideological or cultural differences between state and communal regulation, but that would nonetheless produce a theory of property that is fundamentally different from those in which information costs are the central determining factor.

A final caveat before we move on: the synchronization between information costs and preference diversity argued for in this section is a somewhat unbalanced one. It posits that any significant shock to information costs will likely be accompanied by a preference diversity shock in the same direction, but does not argue that any change in preference diversity will likely be accompanied by a comparable change in information costs.29

29

An increase in preference diversity does not necessarily lead to an increase in information costs unless state or communal institutions are unable to accurately convey information about the new preferences—in which case the state or community would be weakening relative to their economy: imagine that a number of landowners begin to use lease property in legal form A, which was previously unheard of, rather than conventional form B. Preference diversity increases, but unless the administrative cost of collecting and disseminating information about A is higher than the cost of doing so for B—which it may very well be, given its novelty—then the information costs incurred by potential trespassers or purchasers remain largely stable. If A-related costs are indeed higher than B-related costs, that would imply that the state or community now has a harder time monitoring and therefore controlling its population—that, in other words, it is weakening.

It may or may not be—I simply take no position here because this issue does not materially affect the article’s theoretical predictions. So long as changes in sociopolitical cohesion tend to trigger changes in preference diversity, the model is not affected by whether this process can account for all, or even most, changes in preference diversity—or whether there is any reverse causation from preference diversity to sociopolitical cohesion.

3.2 Rethinking Optimal Standardization and Exclusion

One of the core assumptions of mainstream information cost theory, as discussed in Section 2, is that the benefits of standardization and exclusion can change without corresponding changes in their costs (Merrill & Smith 2000, p. 41). As a result, information costs are negatively correlated to the number of property forms: higher information costs demands more standardization. Does this prediction change if we incorporate the synchronization between information costs and preference diversity into the analysis—and can this also affect the relationship between information costs and exclusion rights?

3.2.1 Standardization

The specific mechanism through which such synchronization affects legal standardization depends on whether we are talking about actual or expressed preference diversity, but in both cases it generates very different predictions than does mainstream information cost theory. Consider first a scenario in which the loosening of social and political control leads to a heightened level of actual preference diversity: some people begin to consciously desire forms of use and ownership that they previously did not want—or, at the very least, did not consciously realize they wanted—leading to a wider range of legal demands. For example, landlords might begin to demand the right to evict tenants for previously unheard-of reasons,30

30

The most famous example of this is probably the erosion of tenancy rights in early modern England, which saw landlords roll back the institutional protections traditionally afforded tenants over the course of two centuries. See, e.g. The Brenner Debate (Aston & Philpin 1987, pp. 10–67); Hoyle (1990).

joint tenants might want more flexibility in the formation and termination of their ownership rights,31

31

This was one of the major developments in the early American Republic (Banner 2012).

mortgagees might demand new rights over the collateralized property,32

32

Chinese and English examples of this process are discussed in Zhang (2014).

and so on. For each individual, the likelihood that the preexisting numerus clausus can still accommodate their personal preferences decreases, leading to a societal increase in frustration costs, and generating additional sociopolitical pressures to expand the menu of sanctioned property forms.

These pressures come in two different forms: first, responsible and public-minded lawmakers should, in theory, care about the individual utility of their citizens. A negative shock to preference satisfaction across the population creates significant incentives to boost satisfaction by allowing new forms. Second, lawmakers will also have more self-interested reasons to do so: denying individual preferences is costly for governments, both because it damages their popularity or even legitimacy, and because they must incur potentially significant administrative costs to either persuade or force frustrated property owners to drop their demands. The more diverse individual preferences are, the greater such costs will cumulatively be. Properly understood, the “frustration costs” of standardization is the sum of individual utility loss and these political and administrative costs.

Such pressures automatically vanish if preexisting laws and customs are, in fact, able to accommodate the new demands, or if the new demands are somehow closer to the current numerus clausus than preexisting demands, but those scenarios are unlikely. If, as Merrill and Smith argue, information costs increase as property forms proliferate (2000), then there is only harm, and no apparent benefit, in allowing forms that are not actually demanded by property holders. Presumably, very few legal or customary regimes would do that. In other words, if the numerus clausus functions as it should—by maximizing the information cost savings at lowest possible cost to preference realization—then it is highly unlikely that the new demands will be more compatible with the preexisting legal menu than the old ones they replaced. That said, even if the new preferences are relatively compatible with preexisting legal frameworks, their emergence would nonetheless strengthen resistance against any attempt to reduce the current number of allowable forms. All things considered, they would likely apply pressure in favor of less standardization. In other words, rising preference diversity would demand of property law the exact opposite of what rising information costs demand: the latter pushes for more standardization as a way to lower information costs, whereas the former would push against it in order to accommodate a wider array of individual demands.

Which side emerges victorious from this pushing match—or tug-of-war, if one prefers a different metaphor—depends very much on the specific circumstances. It is possible that a small increase in information costs is pitted against a very large increase in preference diversity and frustration costs, or vice versa, and therefore that one push easily overpowers the other, but not only are such circumstances impossible to predict, they are also rather unlikely. As argued above, information costs are a fairly accurate measure of a state or community’s ability to control individual behavior. It is therefore more likely that large increases in information costs, which would imply a sharp decrease in state capacity or communal cohesion, will be pitted against large increases in preference diversity—and, by the same logic, small increases against small increases. In such cases, the final equilibrium will probably depend on factors other than the scale of information costs or preference diversity: it might depend, for example, on which individuals or social groups benefit from more standardization, which benefit from less, and how much sociopolitical capital each side possesses. To complicate things further, each individual might make a different assessment of the costs and benefits of standardization depending on his or her ideological commitments: some might value the freedom that comes with less standardization more than the convenience that comes with more, whereas others might believe in the exact opposite.

Expressing these dynamics in a marginal cost diagram, as Merrill and Smith do in their seminal 2000 article, is a bit more complicated than it may initially seem. The following is the basic Merrill and Smith model, in which the marginal frustration cost savings curve (MFC) remains unchanged after a positive external shock to information costs, which pushes the marginal information cost curve from MIC1 to MIC2 and, therefore, moves the equilibrium point from A to B, reducing the optimal number of property forms (2000, pp. 40–42):

The MFC curve in this kind of model will, as Merrill and Smith point out, slope downward because legal systems tend to recognize basic and popular property forms like the fee simple first, and less in-demand forms later, which means that marginal frustration cost savings further along the curve will decline. Merrill and Smith also assume that the MIC curves will slope upwards, because later recognized forms will be more complex than earlier ones, but that is unclear. First, it is not necessarily true that less in-demand forms will be more complex than more in-demand ones. For example, tenancy is arguably more in-demand, socially and economically, than most defeasible fees, but is legally more complex. Second, even if later forms are indeed more complex, the marginal information costs they create will not necessarily be higher, precisely because they are less in-demand: rational actors will invest fewer resources at identifying and measuring them because the likelihood of their existence in any specific circumstance is lower. It is not even clear that MIC curves are necessarily monotone, especially when we take clustering effects into consideration: for example, recognizing the fee simple subject to executory limitation is much cheaper and easier when the law already recognizes other kinds of defeasible fees. For simplicity’s sake, however, this article will continue to draw MIC curves in the same way as Merrill and Smith.

graphic
graphic

graphic
graphic

The new diagram we arrive at once preference diversity dynamics are incorporated is the following:

The primary change to the Merrill and Smith model is that positive external shocks to information costs—which move the MIC curve up—are now accompanied by a positive shock to preference diversity, and therefore to frustration costs, which shifts most parts of the MFC curve up as well. The new marginal frustration cost curve MFC2 will not necessarily be above its predecessor MFC1 at all points on the horizontal axis, but it will be at most of them, leading to a flatter but mostly higher curve: because individual preferences are more diverse, the frustration cost savings generated by basic forms such as the fee simple and tenancy may actually decline moderately, but those generated by later forms will increase. Given that legal systems tend to recognize substantial numbers of forms—at least a few dozen in modern societies (Chang & Smith 2018)—MFC2 will almost always rise above MFC1 at the original equilibrium point A. Moreover, because the total volume of frustration costs, once political and administrative costs are incorporated, tends to increase under conditions of higher preference diversity, MFC2 will shift upwards at most points. When this happens, the new equilibrium point B will not necessarily correlate to a smaller number of recognized forms than the previous equilibrium A and, depending on the factors described above, may actually correlate to a larger number. This may sound somewhat indeterminate, but structural indeterminacy is precisely the point: in all likelihood, there is no general correlation, either positive or negative, between information costs and legal standardization.

What if actual preference diversity remains stable after a positive information cost shock—that is, individuals do not consciously develop new preferences—but expressed preference diversity nonetheless increases, presumably because the sociopolitical costs of expressing outlier preferences have fallen? Here, too, the likely outcome is structural indeterminacy, but the underlying mechanism is different: in this context, an increase in expressed preference diversity simply means that the marginal cost of social or political control has increased, and that states and communities must now spend additional human and material resources to maintain previous levels of behavioral conformity. This, too, would presumably encourage the recognition of additional property forms, or at least discourage further standardization.

The distinction between actual and expressed preferences may also be normatively significant in ways that go beyond the descriptive sociopolitical mechanisms outlined above: imagine, for example, that the level of standardization is held stable after a negative shock to sociopolitical cohesion. If actual preference diversity increases, then individual utility losses increase; whereas, if only expressed preference diversity increases, then individual utility losses are unchanged, but the administrative cost of proscribing property forms increases. Both would deter further standardization, but depending on one’s normative commitments, the former might seem much more morally problematic than the latter.33

33

This discussion has emphasized the analytical differences between actual preference diversity and expressed preference diversity, but is there any reliable way to empirically distinguish between the two? Given that their predicted institutional consequences are broadly identical, external empirical observation, especially when the case studies presented below are necessarily brief, can be very difficult. The best empirical test is probably whether the “new” preferences that emerge after a negative shock to sociopolitical cohesion were expressly condemned or outlawed prior to the shock: if so, then that would suggest that changes to expressed preference diversity, rather than actual diversity, drove whatever institutional change we observe. If not, then there may be a plausible case that actual preference diversity is the driving force. Detailed ethnography might be able to qualitatively distinguish between the two on a case-to-case basis, but the success rate of such methods is impossible to predict.

Things are somewhat more complicated when information costs and preference diversity both decrease in response to a rise in state capacity or communal cohesion. Lower information costs lessen the need to standardize, whereas lower preference diversity and frustration costs lessen the need to recognize additional property forms, but less need for something is different from more positive pressure to do the opposite. Lower information costs do not necessarily place immediate pressure on either states or communities to recognize new forms, so long as people are content with the preexisting menu. This would obviously change if frustration costs increase, but frustration costs are much more likely to decrease under these circumstances. Similarly, lower preference diversity allows for additional standardization, but does not exactly demand it, especially when information costs are decreasing at the same time and the social benefit of eliminating preexisting forms is therefore lower. In the end, states and communities will probably have greater sociopolitical capacity to move the level of standardization in whichever direction they favor, but this tells us little about which direction that actually will be.

It actually seems somewhat more likely that lower information costs and preference diversity will lead to more standardization than to less, which is the exact opposite of what traditional information costs would predict: eliminating property forms that are no longer in demand does at least create some information cost savings, even if they are diminishing. So long as the savings outweigh the administrative costs of eliminating forms, states and communities will have some incentive to do so. In comparison, it is harder to identify any concrete benefit to recognizing additional forms when social frustration with the status quo is actually falling. Nonetheless, depending on the specific circumstances in play—ideological commitments, cultural biases, political balances, and so on—both outcomes are possible.

Ultimately, the synchronization of information costs and preference diversity tends to dismantle the positive correlation between information costs and legal standardization that mainstream information cost theory argues for. This goes well beyond the usual “information costs are but one factor that influences standardization” critique, which essentially argues that the influence of information costs can be diluted or overridden, to argue that it probably will be to some extent, and indeed because of reasons that are inherent in the nature of property-related information costs. Within the context of legal standardization, the unusually large role that states and communities play in the collection and dissemination of legal and quasi-legal information tends to generate a counterforce to significant external changes in information costs.

3.2.2 Exclusion

If preference diversity tends to dilute the influence of information costs on the issue of standardization, it is, perhaps somewhat more happily for information cost theory, much more likely to reinforce that influence when it comes to exclusion. Higher levels of preference diversity will probably strengthen both individual and group demands for the right to exclude, which is the same effect that higher information costs tend to have. There are several reasons for this: first and perhaps foremost, higher preference diversity reduces the likelihood that resources can be managed collectively—that is, it pushes against common ownership and use, and in favor of privatization. Collective use requires a very high level of preference compatibility between the co-owners, which becomes less likely when preferences become more diverse at a societal or communal level (Ostrom 1990). Parceling and privatization is the most intuitive solution, but as many scholars have argued, private ownership and use creates societal demand for the right to exclude: property owners would otherwise tend to overspend on border security to protect the fruits of their labor and investment (Shavell 2004, p. 20).

Second, exclusion is a very appealing strategy when one is trying to protect new forms of ownership or use that were previously banned: these forms will probably seem more vulnerable to outside interference than more established forms, and therefore will create particularly strong incentives to exclude both neighbors and government agents. This is especially true when the new forms previously carried a certain level of social, political, or cultural stigma—when they suggest association with unpopular views or socioeconomic positions34

34

While it may seem far-fetched to contemporary audiences that a property form can carry a strong social stigma, such examples are quite abundant in human history, even modern human history. For example, usury laws condemned certain kinds of collateralized lending as sinful for many centuries, and continued to be a source of social debate until the later nineteenth century. See, e.g. Tan (2003); Friedman (1963). As a general matter, the standardization of mortgage law and the elimination of the classic Common Law mortgage from the seventeenth to nineteenth centuries was often a morally charged process in which courts condemned the old mortgage forms as being predatory and immoral (Sugarman & Warrington 1996). In the context of American landlord-tenant law, the elimination of many traditional rights of landlords during the later twentieth century—which are plausibly seen as a standardization of tenancy forms, given their mandatory nature—is often justified as necessary for the protection of human dignity (Alexander 2017).

—in which case their owners will probably desire higher-than-normal levels of privacy, presumably backed up by the right to exclude, to avoid reputational losses. For example, a landlord who imposes harsher-than-previously-acceptable penalties on delinquent tenants may harbor a particularly strong desire to exclude prying neighbors from his property, as might a farmer who begins to extract resources from his land in ways that were previously proscribed by communal covenants as dangerous and irresponsible.

Lower levels of preference diversity, on the other hand, facilitate the formation of commons and other forms of collective use. This may not necessarily lead to large-scale common ownership, but on the margins, it will tend to reduce the demand for strong exclusion rights. It also allows states and communities to micromanage property use more extensively than previously possible, not only because preference homogeneity lowers the information costs associated with governance, but also because it tends to reduce social resistance to top-down monitoring and control. This will probably lead to more state or communal intervention, and therefore weaker rights to exclude, at least vis-à-vis top-down regulation.

In fact, all three major factors in our “model”—sociopolitical cohesion, information costs, and preference diversity—work together harmoniously on the issue of exclusion. Higher levels of sociopolitical cohesion increase the administrative feasibility of intensive top-down governance, whether communal or state-based. At the same time, its downstream consequences—that is, lower information costs and preference diversity—both reduce demand for private rights to exclude. Lower levels of sociopolitical cohesion, on the other hand, reduce the administrative feasibility of top-down governance, while higher information costs and preference diversity both tend to encourage private exclusion. Administrative capacity tends to be higher when demand for private exclusion is lower, and vice versa, thereby reducing the conflict between top-down intervention—what Smith calls “governance strategies”—and bottom-up demands for autonomy—that is, “exclusion strategies” (2004).

This stands in sharp contrast to the analytical messiness of the standardization issue: higher levels of sociopolitical cohesion boost the political and administrative feasibility of further standardization—that is, the state’s capacity to implement it—but the combination of lower information costs and lower preference diversity gives states and communities no clear incentive to pursue legal change in either direction. Even more problematically, lower levels of sociopolitical cohesion erode the political and administrative feasibility of further standardization when rising information costs are demanding exactly that, whereas higher preference diversity tends to demand the exact opposite.35

35

One might question whether there really is any meaningful conceptual difference between preference diversity and legal standardization—whether, for example, the latter is merely the institutional reflection of the former, or whether the former is simply a natural consequence of the latter. Not only does this ignore the central argument made here that rising preference diversity does not necessarily lead to less legal standardization, but it would grossly oversimplify the myriad of ways in which states and communities influence preference diversity that do not involve legal regulation. The standardization of property forms always requires conscious state or communal action, whereas the homogenizing or diversification of property-related preferences is often a subtler process of cognitive nudging and persuasion that can precede any formal change in the standardization architecture. Preference diversity, especially expressed preference diversity, can of course change in response to changes in legal standardization, but it will likely do so only when the change in standardization is effectively enforced by either state or communal actors—which once again suggests that state capacity and communal cohesion are the real determinants.

States and communities may well choose to pursue further standardization even while their ability to enforce it declines, but that would require the presence of strong external incentives—created by ideology, politics, other economic demands, and so on—that mainstream information cost theory does not capture.

3.2.3 Addressing Potential Counterarguments

Before we move on to empirical case studies, three potential counterarguments—more general counterarguments that go beyond the more specific exceptions and qualifications already made above—demand some attention. The first potential objection focuses specifically on the standardization branch of the model: one might be concerned that the frustration costs of further standardization under circumstances of sociopolitical deterioration are overstated. This is because legal standardization, like other forms of state or communal control, can potentially reduce actual preference diversity through social internalization, thereby reducing its own preference frustration costs.36

36

This might happen through the same kinds of sociopolitical persuasion processes discussed in supra Section 3, subsection 3.1.3. Essentially, individuals internalize the numerus clausus as a proxy for common sense—it lays out the boundaries of what is socially acceptable and economically prudent—and use it as a focal when sorting through their otherwise messy preferences.

The problem with this line of argument is that, although frustration costs may eventually decline over the long run, short to medium-term frustration costs will nonetheless increase due to increased preference diversity prior to the decision to either shrink or expand the numerus clausus. Like most individuals, states and communities tend to discount the future, and will probably give significantly greater weight to short-term costs than to long-term ones. Moreover, given that this decision is being made within a broader context of sociopolitical deterioration, the likelihood that further standardization can effectively lower preference diversity, even over the long run, is substantially lower than it normally would be, which further dilutes the strength of the counterargument.

The other two counterarguments both deal with the applicability of the model to internet technology. The first of these asks whether the tight connection between property-related information costs and sociopolitical cohesion that underlies most of the above analysis might eventually be broken by the rise of Internet-based information sharing.37

37

For example, can the Internet do for real estate searches what it did for automobile searches? In the latter context, much scholarship has been devoted to studying the information cost benefits of online databases (Ratchford et al. 2003).

That is, can online crowd-sourcing fundamentally change the way that information is conveyed and verified, independently of any change to communal cohesion or state capacity? Can, for example, information costs fall due to online sharing even when there is no appreciable increase in communal information-sharing or state dissemination? The inclination, for now, is to say no: the verification problems discussed in subsection 3.1 continue to exist after the recent advent of online real estate and property-related databases, which means that local governments continue to play a dominant role. Nearly all such databases are either government-run or are only able to convey legal information insofar as it has been verified by government sources.38 Information costs have indeed gone down, but that is, in all likelihood, more a consequence of local governments boosting their information collection and dissemination capacities via online tools, rather than a consequence of non-governmental players independently providing reliable information. Moreover, even if private information sharing substantially reduces information costs on its own, this would nonetheless generate downstream benefits for state or communal governance. For example, local governments or communities might supplement their own record keeping with information drawn from private databases, thereby expanding their monitoring capacity at lower cost. The negative correlation between information costs and sociopolitical cohesion still holds, regardless of the direction of the causal arrow.

Finally, readers might be concerned that the rise of the internet breaks the synchronization between information costs and preference diversity. Specifically, they might think that online information sharing lowers information costs while increasing preference diversity by exposing individuals to a broader array of options. Such concerns misunderstand the nature of preference diversity: it is, as noted above, a collective measure, rather than an individual one. If exposure to other people’s preferences and options via the internet makes me want to be like them, that may complicate my personal preferences, but does not add to collective diversity in any way. The internet does expose preexisting preference diversity at an unprecedented scale, but that should not be confused for creating preference diversity. Rather, if enough people align their preferences with popular online behavior, that would lead to less preference diversity, not more. In the property context, consider the effect that online real estate posting systems like Zillow or Realtor.com potentially have on homeowner behavior: clearly, the most likely scenario is that possessing more information about sales prices and the underlying property conditions that generate those prices will incentivize homeowners to follow perceived “best practices” in a bid to maximize the resale value of their own property. Consider also what online brokerage systems like Airbnb do to property use patterns: while it certainly exposes travelers and potential tenants to a wider range of options, it homogenizes the way they sort through those options by providing user reviews and scores. Moreover, it creates strong incentives for participating homeowners to provide similar kinds of accommodations—again following the “best practices” logic—so as to attract more tenants, thereby homogenizing property use across initially diverse socioeconomic conditions.

4. CASE STUDIES OF SOCIOPOLITICAL DISINTEGRATION

Sections 4 and 5r discuss several case studies that illustrate various parts of the theoretical model developed in Section 3. Section 4 focuses on cases of clear and significant sociopolitical disintegration—where states and communities are clearly weakening—whereas Section 5 focuses on cases of rising governmental capacity and communal cohesion. Section 4 is further separated into three sections. Subsection 4.1 discusses the transformation of English property law and custom following the Black Death. Subsection 4.2 examines tenancy rights in the Lower Yangtze during the post-Taiping Rebellion reconstruction. Subsection 4.3 looks at the evolution of American property law in the Early Republic. In all three cases, historians have documented a visible tension between demands for new property forms and attempts to protect or even strengthen preexisting levels of standardization. In the English case, the former actually seemed to overwhelm the latter, whereas in the Chinese and American cases the final institutional equilibrium reflected a more equal balance between the two forces. None of these cases fit well into the mainstream information cost framework—some are, in fact, dramatically incompatible—whereas the model presented above makes much better sense of them. Furthermore, as the model predicts, sociopolitical disintegration is, in all three cases, followed by the erosion of communally owned property and greater emphasis on private exclusion. The nature of the model means, however, that it is impossible to empirically prove within the confines of a single article, or even a single book. What the following case studies offer are, instead, condensed illustrations of the theoretical model. The goal is simply to argue that, based on the best available secondary literature, this model does a substantially better job than information cost theory of explaining several major episodes in the history of property.39

39

It goes without saying that there are many additional case studies that potentially support the thesis. See, e.g., Balganesh (2015), which narrates how the British colonial government in India pursued a strategy of legal standardization after it had reconsolidated its political power following the Rebellion of 1857. On government consolidation following the rebellion and its social consequences, see, e.g., Singha (2003).

4.1. England after the Black Death

The development of English property law in the two centuries after the Black Death is, in many ways, the paradigmatic case study in property theory. It is perhaps the single most extensively studied example of how a feudal land system gave way to a “modern” legal regime—and of how private property came to replace the commons (Kerridge 1969; Hardin 1968; Ellickson 1993; Smith 2004). This section revisits this somewhat familiar history and highlights how sociopolitical disintegration influenced the course of legal development.

The Black Death of 1348–1349 left England in a demographically decimated and economically weakened state (Aberth 2005; Gottfried 1983, pp. 58–67; Gregg 1976, pp. 5–8), but did not lead to a wholesale breakdown in social and political order (Palmer 2001). The English aristocracy reacted vigorously during the early stages of the epidemic, limiting its reach and preserving social order in most parts of the Kingdom. As a result, the overall impact on individual living standards was somewhat muted, and merely pushed per capita income further down the moderate decline that was already in motion during the early fourteenth century (Campbell 1991; Musson & Omrod 1999, pp. 90–92; Rout 2008). By the fifteenth century, living standards had recovered and, in fact, had likely risen beyond pre-plague levels (Campbell 2006, pp. 206–22; Hirshleifer 1966, pp. 14–15).

Nonetheless, the plague proved to be enormously disruptive, both socially and politically—the loss of almost half of the population, no matter how well controlled, was simply not something that the preexisting sociopolitical order could take in stride. The sheer loss of population, which completely wiped out some local communities, was one thing, but arguably even more disruptive was the change in labor relations and the subsequent rise in demographic mobility it generated. The demand for labor skyrocketed across the country due to demographic decimation, which incentivized large numbers of laborers to migrate in search of higher wages (Farmer 1991, 1988). The state attempted to contain this through legislation and ordinances that locked wages in at pre-plague levels, but enforcement was difficult, uneven, and profoundly unpopular with peasants and wage laborers (Singha 2003; Balganesh 2015). Over the next few decades, wages rose continuously despite aggressive attempts at enforcement, and the purchasing power of rural laborers rose by over 40 per cent (Dyer 2009, p. 279). Later attempts by the state to control commoner consumption and movement further aggravated tensions between the ruling classes and the general population, which eventually came to a head in the Peasant’s Revolt of 1381. Following the revolt, systemic attempts to control labor mobility gradually receded, and wages continued to rise. By the fifteenth century, serfdom, which assumed stable, multi-generational relationships between lords and tenants, was no longer economically feasible, and gave way almost everywhere to more mobile forms of tenancy (Campbell 2006, p. 214).

Demographic mobility led, naturally, to sharp declines in sociopolitical cohesion. Prior to the Black Death, local manor communities were effective units of self-governance. The frankpledge system, which made groups of households responsible for each other’s illegal activity, provided policing at the local level, while a similar joint-responsibility system tied together “hundreds”—geographical units that covered multiple villages—and counties (Langbein Lerner & Smith 2009, pp. 16–23). This presumed, of course, a very high level of communal cohesion: neighbors needed to know a great deal about each other’s lives for such institutions to work. In criminal cases, juries performed both an investigative and an adjudicative function, again under the assumption that jurors were sufficiently well informed about local affairs that they themselves could supply most of the information necessary to reach a decision (Maitland & Pollock 1898, pp. 622–27, 658–61). None of these institutions could continue to function properly in the aftermath of plague. By the fifteenth century, the frankpledge system had almost completely ceased to exist, putting much greater pressure on sheriffs and other government officers to maintain law and order (Crowley 1975). Similarly, jurors ceased to be self-informing, and increasingly needed to be informed through external evidence and testimony (Langbein, Lerner & Smith 2009, pp. 238–44). Clearly, then, information sharing within local communities plummeted after the plague, leading to higher socioeconomic information costs for private actors. This was made worse in the fifteenth century by the economic rise of wool production, which further strained traditional communal ties by pitting relatively wealthy pastoralists against poorer grain farmers.

The state did, as noted above, attempt to keep up with these demographic and social developments—not only by unsuccessfully trying to suppress labor mobility, but also by reforming and expanding its primary institutions. The central government began to exercise stronger control over the administrative activities of local lords, knights and gentry members assumed a more active role in local administration, and new offices were created to supply local policing (Musson & Omrod 1999, pp. 50–74; Palmer 2001). Nonetheless, scholars generally believe that these measures were insufficient to maintain prior levels of political control, and that, overall, the ruling classes’ ability to govern and control the rural population declined over the late fourteenth and fifteenth centuries (Gregg 1976, pp. 84–88; Musson & Omrod 1999, pp. 68–70, 93–101). Their inability to control wages and demographic mobility despite repeated attempts to do so is itself a strong sign of political disintegration. As one scholar notes, “in this ‘golden age of the English peasantry’, as it is sometimes called, life in the countryside was undoubtedly ‘freer’ than it had been” prior to the plague (Musson & Omrod 1999, p. 70). The state was simply unable to maintain preexisting levels of oversight over a significantly more mobile population, and eventually had to retreat to a more hands-off mode of governance.

Sociopolitical cohesion therefore weakened across both dimensions, communal and governmental, and information costs grew correspondingly. The English economy did not collapse, but as the transition away from self-information juries shows, it was clearly more difficult to obtain reliable legal information after the plague than before. Manor lords had great difficulty keeping track of the whereabouts and activities, including legal activities, of tenants and laborers—hence the breakdown of serfdom—whereas communal information sharing soon became unreliable (Gregg 1976, pp. 84–88; Langbein, Lerner & Smith 2009, pp. 238–44). Reliance on written deeds increased across the country, but it would be many years before written legal information could fully compensate for the disintegration of state and communal institutions, not least because literacy levels remained very low until the early modern era (Ross 1998; Palmer 2001, pp. 62–91). In many ways, this deterioration in information conveyance actually benefitted tenant farmers, who now faced less interference from lords and less prying from neighbors.

Under mainstream information cost theory, such a major increase in information costs should have generated higher demand for legal standardization and exclusion. On the issue of exclusion, at least, the facts do indeed align with the theory: the later fourteenth and fifteenth centuries witnessed a sharp decline in the traditional open-fields system of agriculture and a decisive transition into the private management of land (Kerridge 1969; Hardin 1968; Ellickson 1993; Smith 2004). Shared use of open manor fields proved untenable once communal solidarity plummeted. The gradual rise of wool production led, of course, to some enclosure of open fields—it would not necessarily have done so under more cohesive social conditions—but even grain production became less communal and more reliant on individual households. The exclusion rights of individual tenants strengthened throughout this era, not only against other tenants but also against their lords. Continuously rising wages allowed most farm laborers to become smallholders who held land under fairly favorable terms: locked-in rents, low fees, and strong protection against eviction. By the later fifteenth century, a new form of tenancy that systemically incorporated these terms—the copyhold—began to gain formal recognition by courts and soon became the dominant mode of English tenancy (Gray 1963; Simpson 1986, pp. 160–65). From the mid-fourteenth century onwards, courts were increasingly concerned with the protection of private property against trespass, expanding the relevant writ from a narrow one that only covered trespasses involving “force and arms” to the much broader category of “trespass on the case” (Dix 1937).

At the same time, however, the menu of allowable property forms underwent a significant expansion. Several new tenancy forms were created, the most significant of which was copyhold (Simpson 1986, pp. 47–80, 144–172). Tenancy for a term of years, which had a very dubious legal status prior to the plague, now gained formal recognition, as did a number of conditional fees—contingent remainders, in particular, came into existence during this period, although not without some resistance from judges who worried that they were merely being used for tax evasion (Simpson 1986, pp. 92–93). Fee tails, first created in 1285, expanded into several different forms, some of which were highly controversial and inspired the creation of legal workarounds (Simpson 1986, pp. 90–91). These developments were not the replacement of an old menu of forms with a new one, but rather a straightforward expansion of the preexisting menu: old forms of tenancy, including serfdom, became much less prevalent but continued to be legally allowable. By the early sixteenth century, an estates system that modern lawyers would find relatively familiar was in place (Simpson 1986, pp. 99–102), granting landholders a significantly wider range of options than what they had under thirteenth century feudalism.

Some might argue that English law did eventually abolish feudal land estates over the course of next few centuries, and that this was a delayed response to the socioeconomic turmoil of the fourteenth and early fifteenth centuries. The problem with this kind of interpretation is that, although it is not uncommon for institutional responses to external shocks to be delayed, the delay in this particular case—if it was indeed a delay—would be of a truly astonishing magnitude. It was not until the sixteenth and seventeenth centuries that feudal estates were legally abolished in significant numbers, and not until the 1660 Tenures Abolition Act that most legal commentators truly recognized their widespread termination.40

40

Tenures Abolition Act 1660 (12 Car. 2 c. 24).

Until then, the decline of feudal estates was socioeconomic, in that they came to be used less and less, rather than legal. For most of the fifteenth century, at least, they remained somewhat prevalent, and therefore added to the complexity of property forms both formally and economically. It was not until these estates became near-obsolete in the sixteenth century under conditions of relative socioeconomic stability that the legal system began to decisively move against them (Bean 1968, pp. 180–234, 257–301). In that sense, it is frankly easier to interpret the abolition of feudal estates as a response to declining preference diversity and information costs in the early sixteenth century, than as a response to rising information costs in the late fourteenth century: the former interpretation may or may not be correct, but the latter is plainly questionable.

No less significantly, legal options for collateralized lending proliferated during the fifteenth century. Modern lien-theory mortgages do not involve conveyances of title and are usually not considered part of the estates system per se, but late medieval and early modern mortgages were an altogether different kind of transaction. They did involve formal transfers of title or terms-of-years, and were therefore as much a “property form” as tenancies or conditional fees. Prior to the plague, collateralized lending was limited to what some scholars have called the “Glanvillian mortgage,” which transferred a term-of-years to the creditor but did not allow him to gain full ownership upon default (Rabinowitz 1945). It was not until the later fourteenth and fifteenth centuries that the “classic English mortgage,” which commonly transferred conditional title to the creditors and granted them full fee simple ownership upon default, came into existence. By the sixteenth century, Glanvillian mortgages had fallen into almost complete disuse, but during the fifteenth century both options were readily available (id.; Zhang 2017, pp. 64–85).

In other words, legal standardization weakened significantly after the Black Death even as exclusion rights strengthened. Information costs alone cannot explain this. Rather, the expansion in property forms came as a direct response to new demands made by both landlords and tenants—that is, to an apparent rise in frustration costs under the preexisting menu of forms: taking advantage of their newfound economic stature, smallholders pushed aggressively for tenancy forms that afforded them greater security and flexibility (Campbell 2006, pp. 206–15, 222–33). Landlords accommodated some of this, but also secured a few institutional victories of their own, particularly by gaining the power to seize mortgaged land upon default. All in all, the late fourteenth and fifteenth centuries were a time of change and flux, where greater demographic and economic mobility generated new needs and incentives, which eventually solidified into new property forms. In this particular case, at least, the numerus clausus seemed to react much more strongly to rising preference diversity than to increasing information costs.

4.2 The Lower Yangtze in the Later Nineteenth Century

The Taiping Rebellion against the Qing imperial government, which started in South China in the early 1850s and swept into the Lower Yangtze macroregion around 1860, caused a level of socioeconomic devastation that one usually finds only during major dynastic transitions. War casualties and mass migration by refugees led to a near-50 percent depletion of population in the macroregion during the four years it took the Qing military to regain control (Xing 1991). And yet, like in the aftermath of the Black Death, this was not accompanied by a total collapse in social order. Many parts of the region remained relatively peaceful and well-governed both during the rebellion—their major urban centers simply paid off the rebels to avoid military conflict—and during its immediate aftermath (Bernhardt 1992, pp. 91–97). There was, of course, a strong selection effect underlying this phenomenon: given the severity of the fighting across the region, only the better governed counties and towns were able to avoid demographic collapse, and so it is hardly surprising that those which did survive maintained a fairly high level of social cohesion. During the first decade after the rebellion, when the macroregion experienced rapid repopulation and economic growth, these relatively peaceful and well-governed regions became the centers of recovery (id., pp. 117–160).

Surviving communities did, nonetheless, experience significant disintegration: somewhat ironically, signs of this disintegration were substantially more obvious during the recovery than during the rebellion itself. The rebellion forced the remaining population to coalesce closely around town and county centers for protection and organizational resources—the parts of the region that failed to do so suffered near-total depopulation—whereas the recovery brought large-scale land reclamation and migration (Sheel 1985). This tended to erode the sociopolitical bonds that had traditionally, and as recently as the rebellion years, provided the basis for nearly all forms of local economic regulation, including those related to property. One major sign of this is the decline of lineage property. Whereas local lineages previously tended to set aside sizeable amounts of land as a trust fund for their less wealthy members, by the 1870s and 1880s this nominally communal property had largely fallen under the control of wealthy households (Zhang 1991).

The Qing state had largely ceased to hold coercive authority over Lower Yangtze property relations since at least the late eighteenth century. By the 1840 s, the state had essentially “farmed out” most areas of economic regulation and governance to local entities, including villages, lineages, merchant guilds, and other organizations (Ch’u 1962; Reed 2000). The state was therefore not a major player either before or after the rebellion, and custom, rather than law, was the most important source of property regulation. If anything, state control over the region fell further in the aftermath of the rebellion as local gentry assumed more of the roles traditionally held by government agents (Bernhardt 1992, p. 117–60). All in all, from 1864 to the 1870s, the region experienced both communal and state erosion that nonetheless fell well short of anarchy.

What, then, were the property-related consequences of such erosion? Prior to—and, in fact, during—the rebellion, the customary regime that governed property relations in most parts of rural China was predominantly one of stable private ownership, in which most land was farmed in single-household parcels. In the Lower Yangtze, smallholders owned roughly half the arable land, while large landlords rented most of the other half to tenants, some of whom owned small parcels of their own on the side (Zhang 2017, pp. 225–30). Tenancies could be year-to-year, for a period of years, or, most often, “permanent”—which gave the tenant a right to use the land “forever,” so long as he did not become severely delinquent on rent, which generally meant that the total sum of delinquent rent did not surpass the land’s total market value (id., pp. 165–183). Sharecropping was rare, and most landlords had, by this point, relinquished the right to adjust fixed rents without the tenant’s consent, which led, over time, to remarkably low rent levels across much of the region. Eviction was therefore rare, and tenancy relations usually very stable (id.).

The post-Taiping recovery saw challenges to this regime from both tenants and landlords. In the first phase of the recovery, most landlords, especially in Zhejiang and Jiangxi, were desperate to find tenants to till their lands, and were therefore unusually eager to offer permanent tenancy. Tenants, however, decided to press their advantage and seek still lower rents and more favorable terms: rent riots popped up across these regions, sometimes accompanied by demands that landlords assume land tax burdens that customarily fell upon tenants in the pre-Taiping era (id.). Many landlords balked, and the two sides entered into a period of prolonged and occasionally violent confrontation (id.).

By the early 1870s, landlords in other regions—especially those near Suzhou—had recovered enough that they now became the institutional aggressors. Some sought to buy out their permanent tenants, while others threatened them with rent increases or outright eviction (id.). These threats occasionally escalated into litigation in local county courts, during which the landlord would sometimes argue that permanent tenancies conveyed after the Taiping Rebellion did not give the tenant the traditional right to lock in rents for extended periods of time. Courts rarely gave formal rulings on these disputes, although provincial authorities would sometimes issue public documents that called on—and in rare cases, ordered—landlords to lower rents (Bernhardt 1992, pp. 117–60). Landlord compliance with these governmental directives was likely poor, and many landlords continued to unilaterally increase rents throughout the recovery.

It might be tempting to characterize these disputes as attempts to renegotiate the terms of preexisting tenancy contracts, but that would seriously understate the extent to which they challenged the institutional status quo. They were, in fact, attempts to expand what used to be a very limited menu of allowable, under either formal or customary law, tenancy forms. To prevent landlords from taking advantage of the personal circumstances of specific tenants, the customary law of most Lower Yangtze villages made permanent tenancy a fairly inflexible instrument that involved a number of inalienable rights. The right against unilateral rent increases, for example, was mandatory in most localities—something that tenants and landlords could not contract around (Zhang 2017, pp. 165–83). Similarly, the obligation of the tenant to pay land taxes was often presumed to be innate to the basic form of permanent tenancy, instead of something that was negotiable on a case-by-case basis (id.). The attempts of tenants and landlords to loosen these legal constraints are therefore properly characterized as attempts to introduce new property forms. For example, a “permanent tenancy” that gave the landlord the right to unilaterally increase rents after a certain number of years was no longer a “permanent tenancy” in the traditional sense, but rather a different form of tenancy that also involved a potentially unlimited term.

Moreover, very few tenants or landlords expressed any systemic desire to abolish traditional permanent tenancies. Much more often, they demanded that additional forms of tenancy be recognized under specific conditions. Some Suzhou landlords, for example, limited their demand for periodic rent increases to “permanent tenancies” created after the rebellion (Sheel 1985; You 1990; Muramatsu 1966). Similarly, tenants who wanted landlords to take over their land tax burden tended to do so on a case-by-case basis, often claiming individual circumstances of unusual economic hardship (Zhang 2017, pp. 165–83). These were attempts to expand the legal menu, rather than attempts to swap a certain item on the menu with something else entirely.

In the end, neither side seemed to get what they wanted. Landlords were largely incapable of unilaterally raising rents or evicting unwanted tenants, whereas tenant campaigns to lower rents or shift tax burdens were likewise unsuccessful (Muramatsu 1970, pp. 679–740; Han 1979; Zhang 2017, pp. 165–83). The legal status quo remained in place throughout the late Qing and well into the Republican era, with landlords and permanent tenants holding, for the most part, the same rights that they had prior to the Rebellion. One major reason—not necessarily the only reason—for the institutional deadlock was that, following several waves of post-Rebellion migration, the physical and social distance between landlords and tenants had greatly increased, making it very difficult for either side to add tenancy forms to the status quo (Wu 2004). Many landlords no longer resided near their landholdings, choosing instead to live in urban centers and to rely on local agents for communication and rent collection (Sheel 1985; Wu 2004). This proved to be something of a double-edged sword. On the one hand, it shielded landlords from many of the social pressures they were subject to when residing in rural communities. Whereas tenants used to be able to exert direct social pressure on their landlord simply by showing up at his doorstep and making a scene, they now had to organize large-scale protests and travel significant distances to press their demands in person. As a general matter, absentee landlords were less enmeshed in the networks and hierarchies that organized and governed local society, and therefore less vulnerable to shaming tactics or other forms of reputational sanction (Sheel 1985; Wu 2004).

On the other hand, the landlords’ ability to monitor tenant activity and collect rent also took a significant blow. Agents could only do so much, not to mention that agents were not always trustworthy. As a result, landlords lost much of their ability to pressure individual tenants into compliance with their new demands (Sheel 1985; You 1990; Wu 2004). Some preexisting relationships that required higher levels of monitoring also became less feasible. At least part of the reason why many landlords seemed to favor some form of “permanent tenancy”—perhaps with modified terms and powers—over traditional term-of-years was that the latter involved higher monitoring costs and more frequent turnover, and was therefore less attractive in this new, higher information cost environment (Wu 2004; Zhang 2017, pp. 165–83). In addition, the local social influence of absentee landlords fell sharply for obvious reasons, rendering them relatively powerless to demand significant changes in local custom (Zhang 2017, pp. 165–83).

What should we make of all this? Sociopolitical erosion seemed to weaken the social, economic, and legal institutions that underlay the Lower Yangtze rural economy. Opportunistic tenants and landlords alike sought to expand the menu of available tenancy relations, but both were often frustrated by the difficulty of doing so in a higher information cost environment. Tenants had few means of forcing landlords to assume additional tax duties, while landlords were similarly unable to loosen the institutional shackles on rent and eviction. Creating and enforcing new kinds of tenancy relations required significant communication and coordination, which was difficult in a less integrated and connected society. From the theoretical perspective developed in Section 3, rising preference diversity—expressed preference diversity, at least—and higher information costs had seemingly fought each other to a standstill, leading to the preservation of the status quo.

Exclusion-based private property, on the other hand, quite distinctly expanded at the expense of communal use and ownership. As noted above, communally managed “lineage trusts” deteriorated throughout this period, much of it falling under the private control of wealthy households. This was not a major social or economic development, given that the total volume of lineage-owned land was very limited to begin with, but it nonetheless highlighted the difficulty of common management and use in an era of social disintegration. All in all, the basic predictions of the theoretical model ring true: sociopolitical disintegration generated conflicting pressures on legal standardization, but led to an unambiguous erosion of communal property.

4.3 American Property Law in the Early Republic

The previous sections have explored historical episodes in which, following a decline in sociopolitical cohesion, effected populations attempted to expand the preexisting menu of property forms—the English succeeded at this, the Chinese less so. In neither case were there significant attempts to eliminate preexisting forms. This section, however, outlines an era of American history in which many property forms were clearly under legal assault, as part of a broader effort to accommodate the institutional needs of rapid demographic expansion and migration. While this seems like a perfect example of social disintegration and rising information costs at work, a closer look reveals a number of counterforces that also led to the creation of many new forms.

The four decades following the Philadelphia Convention saw explosive population growth, which in and of itself was not a new social phenomenon: the population had been doubling every two decades or so since the 1640 s (Wood 2011, p. 2). What was new, however, was the amount of internal migration that accompanied the demographic expansion. There was a massive amount of migration to urban centers, and an even larger wave of migration westward in the early nineteenth century (Klein 2009, pp. 69–106). This naturally caused a significant amount of social disintegration in many of the East Coast states. State and local governments, limited in fiscal capacity and personnel throughout this era, had a very difficult time keeping up with the pace of social change, and were generally unable to effectively control or monitor their economies (Handlin & Handlin 1969; Friedman 2005).

The combination of a highly mobile population and a relative abundance of land put great pressure on property law to move in a market-friendly direction. Much of this pressure came in the form of attempts to simplify the legal regime. People wanted to acquire property after they had moved, but the complicated nature of the English estates system posed a significant informational barrier (Friedman 2005, pp. 167–76; Priest 2006). The solution, therefore, was to reduce the need for expertise. New York, which experienced some of the most explosive population growth and migration in the country, was particularly aggressive when it came to eliminating defunct property forms. By the 1820 s, the state had abolished, through statutes and court cases, much of the old English estates system. The fee tail, for example, was quickly eliminated, as were future interests that “unduly suspended” the free alienability of land (Friedman 2005, pp. 174–75; Banner 2012, pp. 4–22; Priest 2015). Initially, attempts were made to ban trusts in land, but the form was eventually reinstated (Friedman 2005, p. 175). The drive to simplify the land law was not limited to the elimination of certain property forms. A number of conveyance formalities inherited from English law, such as the feoffment with livery of seisin, were abolished in favor of simpler procedures (id., pp. 172–173). Most Eastern states made some effort to move in a similar direction. By the early nineteenth century, commentators could claim that “the law of realty in Ohio could be written in one-third of the space which would be required for the law of realty in England,” or that Massachusetts property law had been “stripped of the clogs and incumbrances” that characterized English law (Banner 2012, p. 6).

So far so good for mainstream information cost theory: courts and legislatures seemed to consciously pursue simplification and standardization as an institutional response to social mobility and high legal information costs. The problem is that this was not their only response. At the same time as judges and lawmakers were eliminating old property forms, they were equally eager to create new ones. The tenancy in common, for example, came into existence in this era (id., pp. 15–16). Traditionally, common interests in land only came in the form of joint tenancies with a mandatory right of survivorship, but with migration pulling apart relatives and other closely-knit social networks, few partners who wished to purchase land together now wanted survivorship rights. In response, the tenancy in common came into existence across the country.

Consistent with the theoretical predictions made in Section 3, the newly created Midwestern states, where demographic mobility was particularly high, government authority particularly weak, and preference diversity therefore particularly high, were particularly aggressive in creating new property forms. Joseph Story commented that the land law of Kentucky was “a labyrinth,” and “full of subtle and refined distinctions” (Friedman 2005, pp. 175–76). New landowners desired to hold their property in a multitude of complex forms, and the local government was incapable of reining them in. Similarly complex menus of forms emerged in Missouri, Illinois, and eventually in California (id., pp. 176–177). Clearly, then, sociopolitical upheaval and rising information costs pushed in two contradictory directions. On the one hand, they incentivized some degree of standardization and simplification, but on the other, they created new demands—new preferences—that demanded legal accommodation.

In fact, the information costs generated by the creation of new forms may very well have outweighed the information costs savings generated by the elimination of old English forms. Most eliminated forms had already fallen into disuse by the early Republic (Friedman 2005, pp. 174–75; Banner 2012, pp. 4–22). For example, in the handful of states where legislatures formally abolished the fee tail, very few people were using it in the first place. In Virginia, which abolished it in 1785, lawmakers were very surprised at the lack of resistance by owners of large estates, who expressed little concern over the potential loss of something that, in all likelihood, they no longer needed (Priest 2015). If that is true, however, then the information cost savings obtained by eliminating a practically defunct form could not have been very large. The search costs that potential transacting parties are willing to bear with regard to any specific property form fall sharply when the form is no longer widely used. In other states where the fee tail continued to perform a significant socioeconomic function, legislatures generally retained it, although in most cases with institutional modifications designed to promote alienability (id.). Overall, although the simplification of conveyancing procedures probably did generate significant information cost savings, the elimination of property forms per se very likely did not. In comparison, the introduction of new forms in response to popular demand—which virtually guaranteed their widespread use—would likely have generated substantial new information costs. This would seem to suggest that the accommodation of new socioeconomic preferences was quite possibly a higher priority for lawmakers than the reduction of information costs.

We can observe a somewhat similar dynamic at work in the law of waste. The basic logic behind eliminating property forms to save information costs also applies, to some extent, to laws that regulate property use. The prohibition of uses lowers measurement costs in largely the same way as the prohibition of ownership or tenancy forms. On this issue, the distinct trend in the early Republic was towards the relaxation of waste law. Under English and colonial law, “any fundamental alteration by a tenant of the land constituted waste for which he was liable” (Horwitz 1979, p. 54). By the early nineteenth century, most state courts had abolished this doctrine in favor of free use, which, as one justice wrote, was necessary to accommodate the economic needs of a “new unsettled country.”41

41

Jackson ex dem. Church v. Brownson, 1810 WL 1233 (N.Y. Sup. Ct. 1810) (Spencer, J., dissenting).

Relaxing use standardization may very well have generated substantial information costs, but the courts were largely unconcerned.

Compared to the relative ambiguity of their position on legal standardization, courts and legislatures moved much more decisively when it came to strengthening the right to exclude. Changes were made on several different fronts. First, landowners gained stronger legal protection against encroachers and would-be adverse possessors. Before the nineteenth century, a significant number of property interests could be acquired though “long, but not adverse, usage” (Horwitz1979, p. 44). By the 1830 s, however, most courts had shifted to the view that encroachment could only lead to title if it were truly adverse—that is, without actual or implied permission (id.). This removed the possibility that an act of generosity by the landowner could lead eventually to a loss of title, and therefore strengthened his right to exclude. In addition, colonial era attempts to weaken private exclusion rights for public economic benefit were rolled back in the early Republic. The Mill Acts, a series of eighteenth century laws that shielded mill operators from private trespass suits if their mills caused flooding on others’ land, came under broad assault in the early nineteenth century, and were almost universally denounced by the 1830 s (id., pp. 51–53). They were now considered injurious to small landowners and, in general, too distant from the right to exclude as traditionally understood.

All three case studies examined in this Part highlight the considerable advantages of a theoretical model that incorporates preference diversity. Periods of rising information costs tend also to be periods in which new institutional preferences and demands emerge, and legal systems may choose to prioritize the corresponding rise in frustration costs over whatever concerns they otherwise had about information costs. The English case is a relatively clear example of this, whereas in the Chinese and American cases the tug of war between the two factors had no obvious victor. Mainstream information cost theory does allow for the possibility that unaccounted-for external influences can dilute its predictions, but it fails to uncover the systemic connections between these “external influences” and information costs, and therefore underappreciates their frequency and significance. In comparison, a model that sees information costs and preference diversity as outcomes—and, in some cases, causes—of sociopolitical cohesion makes much better sense of these major episodes in the history of property.

5. CASE STUDIES OF RISING SOCIOPOLITICAL COHESION

This part examines cases of rising sociopolitical cohesion, one state-driven and the others community-driven. The first is the land reform program carried out by the Chinese Communist Party between 1949 and 1956. It illustrates how, in times of falling information costs—when the marginal benefits of legal standardization fall correspondingly—states can nonetheless lower the marginal frustration costs of legal standardization by reducing preference diversity. The second section does not provide an isolated case study like the others, but rather a survey of empirical research on commons and communal management. It asks why, in this empirical literature, the establishment of commons tends to come hand-in-hand with the standardization of ownership shares, rather than with more fluid property regimes that allow different kinds of in-common ownership rights. The proposed answer is that the establishment of a commons usually requires a very high level of preference homogeneity among participants, which tends to push ownership regimes in the direction of standardization, rather than diversification.

5.1 Chinese Land Reform after 1949

It may seem a bit strange to illustrate a theory that was initially derived from Western property law by examining the early history of land reform in Communist China, but information cost theory does not carve out exceptions for socialist or communist regimes, and neither does the model developed in this article. Standardization and exclusion continue to have costs and benefits under socialist regimes, and those costs and benefits continue to change in reaction to information costs and preference diversity. There is no obvious reason why either theory would cease to apply. Socialist leaders may very well ignore many of these costs and benefits for ideological reasons, but that, in many ways, would simply reinforce the point this article ultimately seeks to make: property regimes often evolve independently of information cost concerns. In any case, elite ideological zeal does not magically remove the material costs and benefits of institutional change, but merely changes the sensitivity of political actors—and we can readily observe these costs and benefits at work in the history of socialist land law.

The history of Chinese land law in the 1950s was, of course, a rapid march towards collectivization and communal ownership of rural land, but the way this unfolded yields a number of theoretical insights. From the early 1950s until the beginning of the Great Leap Forward (1958–1962), rural China enjoyed a period of relative stability and economic rejuvenation. Compared with the previous decade and a half of foreign invasion and civil war, unification under the Chinese Communist Party (CCP) allowed for widespread socioeconomic recovery. Soldiers and wartime refugees returned to their hometowns, and government institutions were rapidly rebuilt—and then substantially strengthened. For a number of reasons, including geopolitical circumstances, military strength, and political ideology, the CCP, at least in this early phase, was significantly more effective than its Republican-era predecessors at establishing control and restoring social trust (Shue 1981; Friedman et al. 1993, pp. 111–132;). It quickly brought inflation under control through the issuance of a new currency, stabilized produce and land markets, and sent off political work groups to the countryside to strengthen social order and implement state policy (Barnett 1968; Eckstein 1977; Naughton 2007, pp. 55–68). Unlike the often-violent episodes of socialist reform in CCP-controlled areas prior to 1949, post-1949 policy implementation was predominantly peaceful and, as discussed below, carried out in a relatively thoughtful fashion. Overall, the early and mid-1950s were a time of rising sociopolitical cohesion, falling information costs, and economic rejuvenation, which scholars have characterized as a “honeymoon” between the Party and its rural population (Friedman et al. 1993, p. 111).

Information cost theory predicts that the benefits of both legal standardization and exclusion will decrease in such scenarios. If the frustration costs created by standardization remain stable while the information cost savings decrease, then there should be greater social resistance towards preexisting levels of standardization, and substantially greater resistance towards any further standardization. Similarly, there will be heightened pressure to weaken exclusion-based property regimes in favor of more intrusive “governance”—either by state or communal entities.

The CCP did indeed pursue economic and legal collectivization throughout the 1950s, but it did so in a highly uniform and inflexible manner (Shue 1981, pp. 15–98). In other words, it amplified legal standardization even as it was moving away from exclusion. Moreover, it managed to do so without incurring serious political costs: heightened standardization did not seem to incite serious social resistance or condemnation, nor did the state have to pay out massive subsidies. This poses a challenge to mainstream information cost theory—but one that is easily dealt with by incorporating a theory of preference formation.

Agricultural collectivization came in several phases. From 1949 to 1953, some land was redistributed from “rich farmers” and “landlords” to “poor farmers” in an attempt to boost the “rural middle class” (Wong 1973). Redistribution did not, however, lead to legal standardization, and the Party continued to allow, for a time, the major categories of traditional property rights: full private ownership, lineage ownership, tenancy, conditional ownership, conditional sales, and so on. By 1953, however, the move towards to communal management and ownership had begun. Land sales had all but vanished by 1954 (Friedman et al. 1993, p. 185), and between 1953 and 1956, agricultural collectives emerged in most villages to assumed ownership of nearly all arable land (Shue 1981, pp. 275–287). Farmers acquired shares, which determined what percentage of total output they were entitled to, in the collectively owned property based on their initial capital contribution, including both land and the value of livestock and tools. Some 5 percent of arable land was left to individual households, generally for vegetable planting, but they were not allowed to sell, mortgage, or rent out this land without communal approval (id., pp. 287–292). Essentially, within a decade of assuming power, the CCP had effectively abolished all traditional forms of ownership and replaced them with a combination of standardized shares in communally owned land and watered-down, inalienable full ownership. However strange it might seem to apply the term here, this was indeed legal standardization—indeed an extraordinarily high level of standardization (id., pp. 296–299). The elimination of virtually all institutional variety added a separate dimension of potentially unpopular inflexibility to the collectivization project, and therefore made careful social planning and engineering even more necessary.

Such a massive campaign inevitably carried significant transitional costs, but they were surprisingly moderate compared to the scale of change. Accounts of unhappiness up and down the social ladder do exist, and the government did indeed have to offer some subsidies to farmers—agricultural loans by state banks increased dramatically in 1956, when most of the transformation occurred (id., pp. 296–299). Nonetheless, even at its peak, state investment in agricultural was far smaller “compared to industry’s share in the national budget,” and miniscule compared to the share of GDP generated by agriculture, which was still above 60 percent at this point (id., p. 299). The increase in lending was therefore from an extremely low level to a merely low level. Moreover, the transition met with measured support from most farmers, and outright political coercion was relatively rare. Compared to the extremely bloody and difficult history of Soviet collectivization, the Chinese transformation was remarkably fast, effective, and smooth (Bernstein 1967).42

Catastrophic central mismanagement during the Great Leap Forward (1958–1962) soon soured initial farmer enthusiasm (Friedman et al. 1993, pp. 214–66), but the 1954–1956 transition was both reasonably popular and surprisingly painless: rural per capita income and total factor productivity did not experience significant declines until the Great Leap itself, and even grew during the transition (Lippit 1974, pp. 120–121; Bramall 2000; Cheremukhin et al. 2015).

The Great Leap itself was, of course, an unbridled catastrophe that led to the starvation and death of over 30 million people, but it was also a highly unusual policy blunder even by the standards of planned socialist economies. The primary culprits were exorbitantly high taxation of rural grain production to fuel urban industrialization—well over 35 percent, compared to the 2 to 10 percent taxed by late imperial and Republican governments—and large-scale forced reallocation of labor from agriculture to low-quality steel production (Friedman et al. 1993, pp. 214–266; Li & Yang 2005). Combined, these policies decimated rural incomes and led to widespread famine. Nonetheless, they were analytically and functionally distinct from collectivization per se: at the very least, collectivization need not lead to forced industrialization. Once these policies were abolished in 1962, the rural economy, still collectivized and standardized, swiftly regained much of its productivity, and managed slow but steady growth until decollectivization in the early 1980s (Cheremukhin et al. 2015). Standardized collectivization was, in all likelihood, economically suboptimal, but with the major exception of the Great Leap Forward years, it nonetheless managed to generate social stability and moderate economic growth during its 25-year history.

As discussed above, the early history of Chinese land reform is somewhat problematic for information cost theory, which predicts strong popular resistance to standardization during periods of rising sociopolitical cohesion—and, if standardization is nonetheless forced through, significant short-term social and economic decline linked directly to the removal of desirable legal options. How did the CCP avoid most of these problems? Part of the story was obviously that its rapid accumulation of socioeconomic and political power made resistance costly, but at this early phase of political consolidation, Party control over the countryside was by no means strong enough to deter large-scale farmer protest, had there actually been widespread dissatisfaction.43

43

There were, in fact, numerous incidents of rural protest, but most of the rural population clearly saw the government’s reforms as being in their interest (Perry 1985).

Instead, the Party’s most effective measure was arguably a multi-year “education” project undertaken by its local work groups during the build-up to collectivization (Shue 1981, pp. 275–292; Liu 2006; Strauss 2006). These work groups laid the ideological groundwork not merely by extolling the virtues of standardized collectivization, but also by working with local farmers to identify optimal agricultural production practices, including those involving the management of property rights. This involved monthly or even weekly planning sessions spread out over several years, where farmers and work group members discussed the costs and benefits of standardized property rights—why, in particular, standardization was necessarily to avoid snowballing inequality—and collective ownership.

Furthermore, while the actual transition to legal collectivization happened very quickly, within a 2- to 3-year span, a number of institutional measures were implemented in the early 1950s to familiarize farmers with collective farming and landholding. By 1954, the work groups had established “mutual aid societies” in most villages, which allowed participants—participation was largely voluntary at this stage—to pool resources, standardize land use, and therefore gain economies of scale (Shue 1981, pp. 144–192). These societies were often successful, leading to higher productivity and income, and rapidly grew in size and influence. They provided much of the popular goodwill and legitimacy that collectivization and standardization relied upon a few years later (Shue 1981, pp. 275–291). Although these later measures were no longer voluntary, they were able to avoid mass unrest and protest because earlier phases of largely non-coercive preference engineering, both top-down “education” and bottom-up experimentation, had already laid so much of the necessary social groundwork.

Preceding a major socioeconomic reform with state-sponsored advertisement, education, and persuasion is not a uniquely socialist political tactic. The mobilization of public support for legislation and policymaking is a central feature of democratic politics across the globe, and a central component of such mobilization is active persuasion and political education (Callan 1997; Colby 2003). Some of this comes in the form of “pure” bottom-up social movements, but state involvement is regular and often of critical importance, especially during the implementation of major socioeconomic legislation. The line between such activities and propaganda campaigns in socialist states is a rather blurry one: the latter may involve more censorship and coercion than does the former, but that is often a difference of degree. In any case, the CCP work groups that paved the way for standardized collectivization seemed to employ persuasion techniques far more intensively than coercive measures.

This is not to say that socialist ideology played no role in these developments. Quite the opposite, it was obviously the primary driving force behind the dash to standardize and collectivize. The point here is merely that state—and, to a lesser extent, communal—investment in social education and ideological persuasion successfully reduced its short-term socioeconomic costs (Strauss 2006), and that such investment has historically been a common feature of state action, socialist or not. Rising sociopolitical cohesion may have reduced the information cost benefits of standardization, but it also reduced its frustration costs by facilitating preference homogenization. This gave the state greater political freedom to pursue whatever course of action its governing ideology recommended, which, in this case, was the creation of standardized communes. One major difference between the Chinese and Soviet collectivization campaigns was that the former was planned and implemented during a time of rising sociopolitical cohesion, whereas the latter was a somewhat abrupt response to a socially disruptive grain shortage (Bernstein 1967). Unsurprisingly, the former was far better received, and much less costly.

5.2 Standardizing the Commons

This section continues the discussion of collective ownership regimes—commons, that is—but moves from top-down, state-driven creation to bottom-up, community-driven creation. As decades of empirical research have shown, the establishment of successful and durable commons by local communities generally requires low information costs and high levels of social cohesion (Ostrom 1990; Ostrom & Hess 2010). Common ownership or use of an economic resource is usually not sustainable unless the parties involved have a high level of mutual trust, and they rarely do unless they can monitor each other cheaply and effectively (Ostrom 1990; Ellickson 1993; Smith 2004). This is largely in agreement with information cost theory’s prediction that exclusion is positively correlated with information costs: the higher information costs are within a group of people, the greater the economic need to have strong rights of exclusion. Reversely, low information costs facilitate collective governance, and may encourage collective ownership as well.

So far so good, but problems arise, once again, when we turn our attention to standardization. Given that commons are almost uniformly products of low information cost environments, and are therefore far more likely to be established when information costs fall than when they rise, mainstream information cost theory would predict that the creation of a commons is generally accompanied by a reduction in standardization.

As noted above, collective ownership can theoretically co-exist with a very large variety of ownership forms. A well-known typology developed by Elinor Ostrom and Charlotte Hess separates legal rights in commons—that is, rights that grant some sort of privilege over a shared resource—into five categories: owners, proprietors, claimants, authorized users, and authorized entrants (2010). Owners have the right to alienate the property, exclude non-owners, manage, withdraw natural resources, and physically access it; proprietors have all rights except alienation; claimants lack alienation and exclusion rights; users only have the rights of resource withdrawal and access; and entrants only have the right to access. In reality, the number of viable categories is far more numerous. For example, a shareholder of the commons might have the right to exclude non-members and to alienate her shares, but not the right to manage or withdraw, and a manager might only have the right to manage. After all, “commons” only means that all members of the collective have some right over the entire undivided property—beyond that, there can be as many forms of commons-based property rights as there are private property rights. When local communities draw close enough to establish a commons of some sort, information cost theory would predict that they also recognize a large variety of ownership and usage forms in the commons. After all, if information costs are that low, then presumably there is little cost in allowing community members a large amount of flexibility to pursue their individual preferences.

From this perspective, the “problem” with the empirical literature is that it reveals a surprisingly strong tendency among local communities to standardize property forms in commons. Known commons regimes often recognize only a few basic forms of participation and ownership—and sometimes only a single form that provides a limited number of rights. Tenancies and mortgages are especially likely to be frowned upon. Although it is currently impossible to conduct any remotely accurate accounting of historical and contemporary commons regimes—and therefore no plausible way to claim that any particular institutional arrangement represents the “general trend”—the combination of collective ownership with heightened levels of standardization is commonplace enough in this empirical literature that it poses a clear and significant challenge to mainstream information cost theory.

Take, for example, the influential case study of Swiss mountain pastures conducted by Robert Netting (Netting 1976, 1982). Local villages began holding parts of the Swiss Alps in common ownership as early as the thirteenth century. In many places, the economies of scale needed for efficient animal grazing ruled out private enclosure as a viable institutional option. Once local villages became stable enough to sustain communal management and ownership, they began acquiring large portions of mountainside land for common grazing. The rules governing these common lands remained remarkably stable over the next several centuries, and many continue to function even today. They seem to recognize only one form of participation in the commons: current members of the commons decided whether to admit new members on a case-by-case basis—acquiring private land in the village did not entitle new residents to join the commons—and once admitted, all members had the same basic rights and obligations. Members managed the commons as a group, with no observable hereditary or wealth-based hierarchies differentiating one group of members from another, and enjoyed access rights largely proportionate to the number of livestock they could feed during the winter. No member could sell, lease, or mortgage the land without communal approval, nor did any member enjoy absolute rights of access or use—the community laid out rules governing who could use what at which times, and could sanction violators by revoking part or all of their access privileges. In other words, there was virtually no variation in the bundle of rights that members held vis-à-vis the land—and it was a very limited bundle. In comparison, the private plots of land that villagers retained on the side continued to be governed by the normal menu of property forms found elsewhere in Switzerland: tenancies, mortgages, and various kinds of future interests all remained legally viable. The replacement of private exclusion with common ownership led to much higher levels of standardization.

A similar narrative can be told of many other commons. Large parts of Mount Fuji in Japan, for example, were owned and managed as communal property by several local villages from the Seventeenth century onwards, and had similar institutional features to the Swiss Alps pastures discussed above (McKean & Cox 1982; Mckean 1992). Villagers collectively participated in governance and management, but no individual household had a unilateral right to sell, rent, mortgage, exploit, or even access the commons. Unlike the Swiss example, where villagers seemed to enjoy a high degree of political equality, Japanese villages were highly hierarchical, and political leadership positions were often hereditary. The position of village head, for example, was commonly passed down from generation to generation within the same household, as were seats on the village council. These political hierarchies also applied to the governance of the commons, which meant that there was indeed some variation among the political rights held by commons members. Even so, this would mean that there were, at most, four or five categories of membership—in other words, four or five forms of property, if we consider different these political categories to be different property forms.44

44

On what counts as a different property form, see discussion at supra note 28. Different categories of political membership in the community can be a valid academic reason to recognize different property forms in the commons if those categories substantially impact a community member’s in rem relationship with the common property, just as different in personam relationships between landlord and tenant are commonly considered valid academic reasons to recognize different forms of tenancy.

If we do not, then there was only one property form, similar to the Swiss Alps commons. In comparison, the formal and customary laws governing private land recognized a far greater variety of property forms, covering most variations that modern property lawyers are familiar with.

Other examples in this vein include Norwegian pastures (Örebech 1993), irrigation commons in Nepal, Spain and the Philippines (Martin 1986; Maass & Anderson 1986), and lineage property in pre-1949 South China (Zheng 2000). The latter example, in particular, vividly documents how standardization can rise when information costs fall, and fall when they rise. Unlike lineages in other parts of China, South China lineages tended to collectively own very large plots of land, which were used either for poor relief or as an endowment for local schools and ancestor worship rituals. These lineage commons were usually created when the lineage itself came into formal existence. Given the intense collective activity that was needed to create a lineage, which demanded elaborate religious rituals and often required written recognition by neighboring lineages and the local government, this was usually a time of rising social cohesion and falling information costs.

The common property that emerged from this process—sometimes through the collective acquisition of outside land, and sometimes through the pooling of internal household property—was usually governed in a manner similar to the Japanese and Swiss examples discussed above. Access and usage rights were regulated and doled out by collective decision. Lineage members were generally afforded the same treatment, and were almost always forbidden from transferring, renting, or mortgaging their shares to outsiders. Many lineages created “constitutional” founding documents that banned even collective attempts to alienate any portion of the commons. Nearly all lineages pursued some degree of political egalitarianism at their founding: promotion to the governing council was strictly by age and generational seniority, which theoretically put all households on equal long-term political footing. Exceptions were sometimes carved out for the eldest son of the first generation lineage head, and then for his eldest male descendants, but these were unusual—and affected only one household in each generation even when they did exist. Social and political differentiation on the basis of wealth was routinely banned, and often condemned as immoral (id.; Zhang 2017, pp. 136–150).

After several generations, overexpansion began to erode social cohesion in some of the larger lineages. This made collective governance difficult, and sometimes led to creation of hereditary property interests in the lineage commons. Some households carved out permanent rights of access and use that were no longer subject to collective oversight, while others simply enclosed parts of the commons and became, technically, long-term tenants of the lineage (Zheng 2000). In a minority of lineages, a small handful of wealthy households monopolized political power to such an extent that they became the de facto, and in some cases de jure, managers of the commons, excluding other households from the decision-making process—and sometimes from the property altogether. Similar to some of the case studies examined in Section 4, as sociopolitical cohesion decreased and information costs rose, exclusion came back while standardization fell.

What drove these communities to boost standardization even when information costs were falling? Rising sociopolitical cohesion reduces information costs and preference diversity at the same time, and therefore reduces both the information cost savings and the frustration costs of further standardization. When preferences become sufficiently homogenized—as they tend to do in closely-knit communities that share the same occupation and the same pool of economic resources—frustration costs can fall so low that further standardization becomes not only politically feasible, but socially and economically desirable. All it takes is for frustration costs to fall more than information cost savings do.

As discussed in Section 3, closely-knit communities can homogenize individual preferences through a variety of mechanisms. In particular, they often facilitate the internalization of common moral beliefs—something that is well worth emphasizing, given the general reluctance of law and economics scholars to acknowledge its existence. Rather than merely banning and sanctioning certain social and economic activities, closely-knit communities often condemn them as immoral. As anthropologists and historians have long argued, “moral economies” have historically existed in stable agrarian communities across the globe, in which community members are understood to owe strong moral obligations of neighborliness, mutual support, and generosity to one another (Polanyi 1944; Scott 1976; Vlastos 1990; Booth 1993; Thompson 1993, 97–184; Buoye 2000; Tripp 2006). The specific content of these “moral economies” is a matter of serious contention—I personally have been skeptical of some “moral economy” theories (Zhang 2018)—but there is ample evidence that community members shared at least some basic moral norms that homogenized their social behavior (Zhang 2014). Evidence of this can be found in several of the case studies discussed above: the ritualistic and, in many cases, semi-religious forms of dissent, dispute resolution, and rulemaking observed in Japanese mountain villages strongly suggests a common moral vocabulary and a shared understanding of social propriety (McKean 1992). Similarly, the near-universal condemnation of wealth-based political hierarchies as immoral among South China lineages suggests widespread moral internalization of at least some social norms (Zhang 2017, pp. 184–219).

The broader point here is that closely-knit communities, and particularly those that are close enough to collectively own property, tend to engage in often intensive and sometimes coercive forms of preference engineering, with the very specific goal of imposing a common code of behavior. This leads, both directly and indirectly, to lower levels of preference diversity on property-related issues. Property norms may fall under the umbrella of commonly accepted social morality, but even when they do not, moral conformity on certain aspects of social behavior generates a sense of communal solidarity and closeness that encourages group-think on other issues.

6. CONCLUSION

Ultimately, information costs are but one of at least two major economic factors that directly influence levels of standardization and exclusion in property law. The other major factor, preference diversity, shares the same sociopolitical origins as information costs, but tends to negate its impact on legal standardization, even while reinforcing its impact on exclusion. Moreover, preference diversity is not some haphazard external factor that may or may not interfere with the interaction between information costs and property. Instead, it is an often systemic, although not necessarily equal, counterforce: it tends to change at the same time as information costs do, and therefore tends to interfere. The connections between information costs and property institutions that mainstream information cost theory has attempted to establish are therefore both overstated and understated. In the sociopolitical contexts identified the above model, there should be no empirical correlation between information costs and standardization—rising information costs are just as plausibly followed by less standardization as more, or by no change at all—but there should be an even stronger correlation between information costs and exclusion than information cost theorists themselves have argued for.

This implies that exclusion and standardization are products of qualitatively different socioeconomic processes. Exclusion is indeed an effective way to manage sociopolitical disintegration and rising information costs, but standardization often is not: its marginal costs and benefits tend to change simultaneously and cancel each other out. The rise and fall of exclusion rights is therefore a predominantly functional process, in the sense that it predictably emerges in response to a material socioeconomic problem, whereas the rise and fall of standardization hinges more on circumstantial contingency, including the ideological or cultural leanings of law and policy-makers. Exclusion certainly can be influenced by these “soft” factors, but one would be hard-pressed to find a significant example, beyond the fringe cases of “legal deindustrialization” discussed above, of sociopolitical disintegration followed by the establishment of a commons, or of state-building and communal integration leading to a greater emphasis on exclusion.

Standardization, on the other hand, responds to sociopolitical change in more idiosyncratic ways, and sometimes not at all. The outcome of the tug-of-war between preference diversity and information costs depends on highly localized and specific factors: what is ideal scale of land use, given local economic conditions? Does the local population care more, on the margins, about material wealth or social solidarity? Understanding a legal system’s decision to standardize is usually a significantly more complicated task, involving far more variables, than understanding its approach towards exclusion. This does not mean that standardization is a less important feature of property law than exclusion—merely that it exposes the limitations of information cost theory more vividly.

To their credit, information cost theorists have indeed placed greater emphasis on exclusion than on standardization in recent years, although the latter continues to be a major focus (Smith 2012). Nonetheless, it would be a significant mistake to understand exclusion simply as an instrument to control information costs. That is, at most, half of its actual socioeconomic function. The other half is the protection of diverse individual preferences from societal or governmental interference. People demand exclusion in times of sociopolitical disintegration not merely because it streamlines social interaction and economic decision-making, but also because, given rising preference diversity, they are naturally more suspicious of outside meddling by people and entities who are less likely to share their preferences. In other words, the point of exclusion is actually to exclude, and not merely to save information costs.

From this perspective, the information cost theory of property may not be quite as disconnected from Margaret Jane Radin’s personhood theory, which posits that the purpose of private property is to generate and protect our sense of self and personhood, as previously thought (1982). Insofar as the realization and protection of individual preferences enhances individuality and personhood, the theoretical framework developed here brings information costs and personhood concerns into direct interaction on a common sociopolitical ground. Under conditions of sociopolitical disintegration, both theories would demand stronger exclusion rights: the former to lower information costs, the latter to protect the new personhood interests that emerge from preference diversification. At the same time, personhood theory would demand a loosening of the numerus clausus to facilitate individual fulfillment, whereas information cost theory would recommend tightening it.

More importantly, this article has highlighted the pressing need for property scholars to develop and integrate a theory of preference formation, without which concepts like “frustration costs” lack the analytical depth and power that we have come to associate with the concept of “information costs,” leaving traditional information cost theory incomplete and imbalanced. Most property scholars tend to assume, often implicitly, that individual preferences, in the form of utility functions, are black boxes that do not require systemic explanation (Zhang 2016, pp. 353–363). But as this article has shown, core features of property law simply cannot be explained without taking into account how individual preferences react to sociopolitical forces. Even the somewhat rudimentary theory of preference formation developed here—a theory that only considers the diversity or homogeneity of preferences, and not their substantive content—allows us to make sense of major legal developments that would be otherwise hard to explain. A fuller and richer theory of preference formation, one that addresses the actual content of individual preferences, promises to do significantly more.

All this points in the direction of cultural and ideological analysis. For decades, property theory has been somewhat wary of sociological theories of legal culture, often arguing that they lack the analytical precision needed for rigorous theorizing (Cooter 1995; Dau-Schmidt 1997; Ellickson 1998; Knight & Epstein 2004). The underlying hope was that scholars could explain the key features of property law through clean utilitarian analysis while avoiding the “swamp” of cultural analysis (Ellickson 1998, p. 549). Information cost theory is, in many ways, the epitome of this effort: a powerful yet conceptually simple theory that promises to explain the structure of property law without delving into the ambiguities of sociocultural construction. But as this article has argued, the structural determinacy it tried to offer was, in many ways, illusory. At least on the issue of standardization—a central issue not merely in property law, but in all domains of private law—the tug-of-war between information costs and preference diversity leads to a fundamental structural indeterminacy that throws open the door for ideology or culture, among other “soft factors,” to be the final determinant. Among the historical cases examined above, this is most clearly seen in post-1949 Chinese land reform: reduced preference diversity neutralized the economic effects of falling information costs, allowing the CCP to push through an ideologically motivated agenda at relatively low sociopolitical cost. A cultural preference for kinship- or geography-based social homogeneity may also have played a significant role in the development of Swiss and Japanese mountainside commons, but there is no room here to rigorously explore that possibility. Nonetheless, the scope and content of economic freedom, which is, after all, what standardization directly defines, may very well depend on what is lurking within the “swamp.”

Fortunately, there are ways to make the swamp less swampy. As I have argued elsewhere, cultural analysis need not be as abstract and conceptually shapeless as it tends to be in some branches of sociology or anthropology (Zhang 2016). We need not speak of “culture” as an isolated “operative engine” that somehow exists beyond individual preferences and choices, but can instead formally model them as internalized norms or as society-specific behavioral biases. It is perfectly possible to study and theorize culture in a way that is consistent with the basic principles of law and economics—all we really need is a willingness to peer inside the black box of individual preference. Whether, and to what extent, we can do so will determine how well we understand the evolution of property institutions.

Acknowledgement

For comments and suggests, I thank Bruce Ackerman, Karen Bradshaw, Robert Ellickson, Richard Epstein, Lee Anne Fennell, Oona Hathaway, Daniel Markovits, Jedidiah Purdy, Carol Rose, David Schleicher, Alan Schwartz, Scott Shapiro, Henry Smith, James Whitman, Katrina Wyman, Xiaoxue Zhao, two anonymous referees, the editors of the Journal of Legal Analysis, and participants of workshops at the 2018 American Law and Economics Association Annual Meeting, Yale, New York University, Peking University, Tsinghua University, Renmin University, and Brooklyn Law School. I thank Zhiang Song, Joshua Wilson, and Youlin Yuan for excellent research assistance. Needless to say, all mistakes are my own.

REFERENCES

Aberth

John.

2005

.

The Black Death: The Great Mortality of 1348—1350, a Brief History with Documents

.

Boston, MA

:

Bedford

.

Alexander

Gregory

,

Penalver

Eduardo

,

Singer

Joseph

,

Underkuffler

Laura

.

2009

.

A Statement of Progressive Property

.

94

Cornell L. Rev.

743

744

.

Alexander

Lisa.

2017

.

Evicted: The Socio-Legal Case for the Right to Housing

.

126

Yale L. J. Forum

431

.

Arruñada

Benito

,

Garoupa

Nuno M.

.

2005

.

The Choice of Titling System in Land

.

48

J.L. & Econ.

709

727

.

Aston

T.H.

,

Philpin

C.H.E.

eds.

1987

.

The Brenner Debate: Agrarian Class Structure and Economic Development in Pre-Industrial Europe

.

Cambridge

:

Cambridge University Press

.

Austin

John.

2005

.

Lectures on Jurisprudence, or, the Philosophy of Positive Law

.

Campbell

R.

ed.

Clark, NJ: Lawbook Exchange

.

Balganesh

Shyamkrishna.

2015

.

Codifying the Common Law of Property in India: Crystallization and Standardization as Strategies of Constraint

.

63

Am. J. Comp. L.

33

76

.

Banner

Stuart.

2012

.

American Property: A History of How, Why, and What We Own

.

Cambridge, MA

:

Harvard University Press

.

Banner

Stuart.

.

Transitions between Property Regimes

.

31

J. Legal Stud.

359

371

.

Barnett

A. Doak.

1968

.

Communist China: The Early Years 1949-55

.

New York, NY

:

Praeger

.

Bean

John M.W.

1968

.

The Decline of English Feudalism, 1215-1540

.

New York, NY

:

Barnes & Noble

.

Becker

Gary S.

,

Murphy

Kevin M.

.

1988

.

A Theory of Rational Addiction

.

96

J. Pol. Econ.

675

700

.

Bell

Abraham

,

Parchomovsky

Gideon

.

2016

.

Of Property and Information

.

116

Colum. Law Rev

.

237

286

.

Bendor

Jonathan

,

Swistak

Piotr

.

2001

.

The Evolution of Norms

.

106

Am. J. Soc.

1493

1545

.

Bernhardt

Kathryn.

1992

.

Rents, Taxes, and Peasant Resistance: The Lower Yangzi Region, 1840-1950

.

Stanford, CA

:

Stanford University Press

.

Bernstein

Thomas P.

1967

.

Leadership and Mass Mobilisation in the Soviet and Chinese Collectivization Campaigns of 1929-30 and 1955-56: A Comparison

.

31

China Q.

1

47

.

Blanchard

Olivier

,

Shleifer

Andrei

.

2001

. Federalism with and without Political Centralization: China versus Russia, in 48 IMF Staff Papers, Special Issue, Transition Economies, How Much Progress?

171

179

.

Booth

William James.

1993

.

Households: On the Moral Architecture of the Economy

.

Ithaca, NY

:

Cornell University Press

.

Bramall

Chris.

2000

.

Inequality, Land Reform and Agricultural Growth in China, 1952–55: A Preliminary Treatment,

27

J. Peasant Stud.

30

54

.

Brighouse

Harry.

1998

.

Civic Education and Liberal Legitimacy

.

108

Ethics

719

745

.

Buoye

Thomas.

2000

.

Manslaughter, Markets and Moral Economy: Violent Disputes over Property Rights in Eighteenth Century China

.

New York, NY

:

Cambridge University Press

.

Cai

Hongbin

,

Treisman

Daniel

.

2006

.

Did Political Decentralization Cause China’s Economic Miracle

.

58

World Pol.

505

535

.

Callan

Eamonn.

1997

.

Creating Citizens: Political Education and Liberal Democracy

.

Oxford

:

Clarendon Press

.

Campbell

Bruce M.S.

1991

.

Before the Black Death: Studies in the ‘Crisis’ of the Early Fourteenth Century

.

Manchester

:

Manchester University Press

.

Campbell

Bruce M.S.

.

2006

. The Land. In

Horrox

Rosemary

,

Ormrod

W. Mark

eds,

A Social History of England, 1200-1500

.

Cambridge

:

Cambridge University Press

.

Chang

Yun-chien

,

Smith

Henry E.

.

2012

.

An Economic Analysis of Civil versus Common Law Property

.

88

Notre Dame Law Rev.

1

56

.

Chang

Yun-chien

,

Smith

Henry E.

.

2015

.

The Numerus Clausus Principle, Property, Customs, and the Emergence of New Property Forms

.

100

Iowa Law Rev

.

2275

2308

.

Chang

Yun-chien

,

Smith

Henry E.

.

2018

.

Convergence and Divergence in Systems of Property Law: Theoretical and Empirical Analyses

.

92

So. Cal. Law Rev.

785

808

.

Cheremukhin

Anton

,

Golosov

Mikhail

,

Guriev

Sergei

,

Tsyvinski

Aleh

.

2015

. The Economy of the People’s Republic of China from 1953. NBER Working Paper No. 21397.

Ch’u

Tung-tsu.

1962

.

Local Government in China under the Chi’ing

.

Cambridge, MA

:

Harvard University Asia Center

.

Cohen

Felix S.

1935

.

Transcendental Nonsense and the Functional Approach

.

35

Colum, Law Rev.

809

849

.

Cohen

Geoffrey L.

2003

.

Party over Policy: The Dominating Influence of Group Influence on Political Beliefs

.

85

J. Pers. Soc. Psych

808

822

.

Colby

Anne.

2003

.

Educating Citizens: Preparing America’s Undergraduates for Lives of Moral and Civic Responsibility

.

Hoboken, NJ

:

John Wiley & Sons

.

Cooter

Robert.

1995

.

Law and Unified Social Theory

.

22

J. L. and Society

50

67

.

Crowley

D.A.

1975

.

The Later History of Frankpledge

.

48

Bull. Inst. Hist. Res.

1

42

.

Dagan

Hanoch.

2011

.

Property: Values and Institutions

.

Oxford

:

Oxford University Press

.

Dau-Schmidt

Kenneth G.

1997

.

Economics and Sociology: The Prospects for an Interdisciplinary Discourse on Law

.

Wisc. L. Rev

389

419

.

Davidson

Nestor M.

2008

.

Standardization and Pluralism in Property Law

.

61

Vand. L. Rev.

1597

.

Dix

Elizabeth Jean.

1937

.

The Origins of the Action of Trespass on the Case

.

46

Yale L. J.

1142

1185

.

Douglas

Mary.

1986

.

How Institutions Think (Contemporary Issues in the Middle East)

.

Syracuse, NY

:

Syracuse University Press

.

Douglas

Mary

,

Wildavsky

Aaron

.

1982

.

Risk and Culture: An Essay on the Selection of Technological and Environmental Dangers

.

Oakland, CA

:

University of California Press

.

Druckman

James.

2004

.

Political Preference Formation: Competition, Deliberation, and the (Ir)Relevance of Framing Effects,

98

Am. Pol. Sci. Rev.

671

686

.

Druckman

James N.

,

Lupia

Arthur

.

2000

.

Preference Formation

.

3

Ann. Rev. Pol. Sci.

1

24

.

Dyer

Christopher.

2009

.

Making a Living in the Middle Ages: The People of Britain

.

New Haven, CT

:

Yale University Press

.

Eckstein

Alexander.

1977

.

China’s Economic Revolution

.

Cambridge

:

Cambridge University Press

.

Ellickson

Robert C.

1991

.

Order without Law: How Neighbors Settle Disputes

.

Cambridge, MA

:

Harvard University Press

.

Ellickson

Robert C.

.

1993

.

Property in Land

.

102

Yale L.J.

1315

1400

.

Ellickson

Robert C.

.

1998

.

Laws and Economics Discovers Social Norms

.

27

J. Legal Stud.

537

552

.

Ellickson

Robert C.

2001

.

Market for Social Norms

.

3

Am. L. & Econ. Rev.

1

22

.

Ertman

Thomas.

1997

.

Birth of the Leviathan: Building States and Regimes in Medieval and Early Modern Europe

.

Cambridge

:

Cambridge University Press

.

Etzioni

Amitai.

2000

.

Social Norms: Internalization, Persuasion and History

.

34

L. & Soc. Rev.

157

178

.

Farmer

David L.

1988

. Prices and Wages. In

Hallam

H. E.

, ed.,

The Agrarian History of England and Wales, 1042-1350

, Vol.

2

,

715

748

.

Cambridge

:

Cambridge University Press

.

Farmer

David L.

.

1991

. Prices and Wages. In

Miller

E.

, ed.,

The Agrarian History of England and Wales, 1348-1500

, Vol.

3

,

431

462

.

Cambridge

:

Cambridge University Press

.

Fennell

Lee Anne.

2012

.

Lumpy Property

.

160

U. Pa. L. Rev.

1955

1993

.

Fernández

Raquel C.

2004

.

Mothers and Sons: Preference Formation and Female Labor Force Dynamics

.

119

Q.J. Econ.

1249

1299

.

Fischel

William.

2004

.

An Economic History of Zoning and a Cure for Its Exclusionary Effects

.

41

Urb. Stud.

317

340

.

Fitzpatrick

Daniel.

2006

.

Evolution and Chaos in Property Rights Systems: The Third World Tragedy of Contested Access

.

115

Yale L. J.

996

1048

.

Friedman

Edward

,

Pickowitz

Paul

,

Selden

Mark

,

Johnson

Kay Ann

.

1993

.

Chinese Village, Socialist State

.

New Haven, CT

:

Yale University Press

.

Friedman

Lawrence M.

1963

.

The Usury Laws of Wisconsin: A Study in Legal and Social History

.

4

Wisc. L. Rev.

515

565

.

Friedman

Lawrence.

2005

.

A History of American Law

, 3rd edn.

New York, NY

:

Touchstone

.

Glaeser

Edward L.

,

Shleifer

Andrei

.

2003

.

The Rise of the Regulatory State

.

41

Econ. Literature

401

425

.

Gonzalez

Cleotilde

, et al. 

2005

.

The Framing Effect and Risky Decisions: Examining Cognitive Functions with fMRI

.

26

J. Econ. Psychol

.

1

20

.

Gottfried

Robert S.

1983

.

The Black Death: Natural and Human Disaster in Medieval` Europe

.

New York, NY

:

Free Press

.

Gray

Charles Montgomery.

1963

.

Copyhold, Equity, and the Common Law

.

Cambridge, MA

:

Harvard University Press

.

Gregg

Pauline.

1976

.

Black Death to Industrial Revolution: A Social and Economic History of England

,

5

8

.

Edinburgh

:

George G. Harrap and Co

.

Greif

Avner

,

Tabellini

Guido

.

2010

.

Cultural and Institutional Bifurcation: China and Europe Compared

.

100

Am. Econ. Rev.: Papers & Proceedings

1

10

.

Grey

Thomas C.

1980

. The Disintegration of Property. In

Roland Pennock

J.

,

Chapman

John W.

, eds.,

Nomos XXII: Property

,

30

45

.

New York, NY

:

NYU Press

.

Haley

John O.

1998

.

Apology and Pardon: Learning from Japan

.

41

Am. Behav. Scientist

842

.

Han

Hengyu.

1979

.

Shilun Qingdai Qianqi Diannong Yongdian Quan de Youlai ji qi Xingqi [on the Origins and Nature of the Permanent Tenancy Rights of Early Qing Tenant Farmers]

. In

Qingshi Luncong [Collected Essays on Qing History]

, Vol.

1

,

38

60

.

Handlin

Oscar

,

Handlin

Mary

.

1969

.

Commonwealth: A Study of the Role of Government in the American Economy

.

Cambridge, MA

:

Belknap Press

.

Hansmann

Henry

,

Kraakman

Reinier

.

2002

.

Property, Contract, and Verification: The Numerus Clausus Problem and the Divisibility of Rights

,

31

J. Legal Stud.

373

420

.

Hardin

Garrettt.

1968

.

The Tragedy of the Commons

.

162

Science

1243

1248

.

Hirshleifer

Jack.

1966

.

Disaster and Recovery: The Black Death in Western Europe

.

Santa Monica, CA

:

RAND Corporation

.

Horwitz

Morton.

1979

.

The Transformation of American Law, 1780-1860

.

Cambridge, MA

:

Harvard University Press

.

Hoyle

R.W.

1990

.

Tenure and the Land Market in Early Modern England: Or a Late Contribution to the Brenner Debate

.

43

Econ. Hist. Rev.

1

43

.

Kadens

Emily.

2012

.

The Myth of the Customary Law Merchant

.

90

Tex. L. Rev

.

1153

1205

.

Kahan

Dan M.

,

Braman

Donald

.

2006

.

Cultural Cognition and Public Policy

.

24

Yale L. & Pol’y Rev.

147

170

.

Kahan

Marcel

,

Klausner

Michael

.

1997

.

Standardization and Innovation in Corporate Contracting (or “the Economics of Boilerplate”)

.

83

Va. L. Rev.

713

770

.

Kerridge

Eric.

1969

.

Agrarian Problems in the Sixteenth Century and After

.

New York, NY

:

Barnes and Noble

.

Klausner

Michael.

1995

.

Corporations, Corporate Law, and Networks of Contracts

.

81

Va. L. Rev.

757

852

.

Klein

Herbert S.

2009

.

A Population History of the United States

, 2nd edn.

Cambridge

:

Cambridge University Press

.

Klick

Jonathan

,

Parchomovsky

Gideon

.

2016

.

The Value of the Right to Exclude: An Empirical Assessment

.

165

U. Pa. L. Rev

.

917

966

.

Knight

Jack

,

Epstein

Lee

.

2004

.

Building the Bridge from Both Sides of the River: Law and Society and Rational Choice,

38

L. & Soc’y Rev.

207

212

.

Kuran

Timur.

1996

.

Private Truth, Public Lies: The Social Consequences of Preference Falsification

.

Cambridge, MA

:

Harvard University Press

.

Langbein

John H.

,

Lerner

Renee Lettow

,

Smith

Bruce P.

.

2009

.

History of the Common Law: The Development of Anglo-American Legal Institutions

.

New York, NY

:

Aspen Publishers

.

Lemley

Mark A.

,

McGowan

David

.

1998

.

Legal Implications of Network Economic Effects

.

86

Cal. L. Rev.

479

611

.

Lessig

Lawrence.

1998

.

The New Chicago School

.

27

J. Legal Stud.

661

691

.

Levin

Irwin P.

, et al. 

2002

.

A New Look at Framing Effects: Distribution of Effect Sizes, Individual Differences, and Independence of Types of Effects

.

88

Org. Behav. & Hum. Decision Processes

411

429

.

Lewis

Alan

,

Cullis

John

.

1988

.

Preferences, Economics, and the Economic Psychology of Public Sector Preference Formation

.

17

J. Behav. Econ.

19

33

.

Li

Huaiyin.

2008

.

Confrontation and Conciliation under the Socialist State

.

33

20th Century China

73

99

.

Li

Wei

,

Yang

Dennis Tao

.

2005

.

The Great Leap Forward: Anatomy of a Central Planning Disaster

.

113

J. Pol. Econ.

840

877

.

Lie

John.

2000

. The Discourse of Japaneseness. In Mike Douglass & Glenda Susan Roberts, eds.,

Japan and Global Migration: Foreign Workers and the Advent of a Multicultural Society

,

69

87

.

Honolulu, HI

:

University of Hawaii Press

.

Lin

Nan.

1999

.

Building a Network Theory of Social Capital

.

22

Connections

28

51

.

Lippit

Victor D.

1974

.

Land Reform and Economic Development in China: A Study of Institutional Change and Development Finance

,

120

121

.

New York, NY

:

International Arts and Sciences Press

.

Liu

Yu.

2006

.

Why Did It Go so High? Political Mobilization and Agricultural Collectivization in China

.

187

China Q.

732

742

.

Maass

Arthur

,

Anderson

Raymond L.

.

1986

.

And the Desert Shall Rejoice: Conflict, Growth, and Justice in Arid Environments

, 2nd edn.

Malabar, FL

:

Krieger Malabar

.

Maitland

Frederic William

,

Pollock

Sir. Frederick

.

1898

.

The History of English Law before the Time of Edward I

, 2nd edn.

Cambridge

:

Cambridge University Press

.

Malenka

David J.

, et al. 

1993

.

The Framing Effect of Relative and Absolute Risk

.

8

J. Gen. Internal Med.

543

548

.

Martin

Edward.

1986

. Resource Mobilization, Water Allocation, and Farmer Organization in Hill Irrigation Systems in Nepal. Ph.D. diss., Cornell University, ICIMOD. http://lib.icimod.org/record/20.

McAdams

Richard H.

1997

.

The Origin, Development, and Regulation of Norms

.

96

Mich. Law Rev.

338

433

.

McKean

Margaret.

1992

.

Success on the Commons: A Comparative Examination of Institutions for Common Property Resource Management

.

4

J. Theoretical Pol.

247

281

.

McKean

Margaret

,

Cox

Thomas R.

.

1982

.

The Japanese Experience with Scarcity

.

6

Environ. Rev.

63

91

.

Merrill

Thomas W.

,

Smith

Henry E.

.

2000

.

Optimal Standardization in the Law of Property: The Numerus Clausus Principle

.

110

Yale Law J.

3

70

.

Merrill

Thomas W.

,

Smith

Henry E.

.

2001

.

What Happened to Property in Law and Economics?

111

Yale Law J.

357

398

.

Miceli

Thomas J.

, et al. 

2000

.

The Dynamic Effect of Land Title Systems

.

47

J. Urb. Econ.

370

389

.

Milsom

S.F.C.

1981

.

Historical Foundations of the Common Law

. 2nd edn.

London

:

Butterworths

.

Moise

Edwin E.

1976

.

Land Reform and Land Reform Errors in North Vietnam

.

49

Pac. Aff.

70

92

.

Moise

Edwin E.

.

1983

.

Land Reform in China and North Vietnam: Consolidating the Revolution at the Village Level

.

Chapel Hill, NC

:

University of North Carolina Press

.

Muramatsu

Yūji.

1966

.

A Documentary Study of Chinese Landlordism in the Late Ch’ing and Early Republic Kiangnan

.

29

Bulletin of the School of Oriental and African Studies

566

599

.

Muramatsu

Yūji.

.

1970

.

Kindai Kōnan no Sosan—Chūgoku Jinushi Seido no Kenkyū [Tenancy in the Early Modern Lower Yangtze: Research on Chinese Landlord Institutions]

.

Toōkyoō, Japan

:

Kindai Chuūgoku Kenkyuū Iinkai: Hatsubaisho Toōkyoō Daigaku Shuppankai

.

Musson

Anthony

,

Ormrod

W.M.

.

1999

.

The Evolution of English Justice: Law, Politics, and Society in the Fourteenth Century

.

New York, NY

:

St. Martin’s Press

.

Naughton

Barry.

2007

.

The Chinese Economy: Transitions and Grow

.

Cambridge, MA

:

MIT Press

.

Nelson

Robert H.

2009

. The Rise of Sublocal Governance. Mercatus Ctr., Geo. Mason Univ., Working Paper No. 09-45.

Netting

Robert.

1982

. Territory, Property, and Tenure. In

Adams

Robert McCormick

,

Smelser

Neil J.

,

Treiman

Donald J.

, eds.,

Behavioral and Social Science Research: A National Resource

,

446

502

.

Washington

:

National Academy Press

.

Netting

Robert.

.

1976

.

What Alpine Peasants Have in Common: Observations on Communal Tenure in a Swiss Village

.

4

Human Eco.

135

146

.

North

Douglass C.

1993

.

Institutions and Credible Commitment

.

149

J. Institutional & Theoretical Econ

.

11

23

.

North

Douglass C.

,

Weingast

Barry R.

.

1989

.

Constitutions and Commitment: The Evolution of Institutions Governing Public Choice in Seventeenth-Century England

.

49

J. Econ. Hist.

803

832

.

Ostrom

Elinor.

1990

.

Governing the Commons: The Evolution of Institutions for Collective Action

.

New York

:

Cambridge University Press

.

Ostrom

Elinor

,

Hess

Charlotte

.

2010

. Private and Common Property Rights. In

Bouckaert

Boudewijn

ed.,

Property Rights and Economics

,

53

106

.

Cheltenham

:

Edward Elgar

.

Örebech

Peter.

1993

. Common and Public Property Rights Regimes to Non-Private Resources: Some Legal Issues on Self-Governing Conservation Systems. In Erling Berge eds., Common Property Regimes: Law and Management of Non-Private Resources; Proceedings of the Conference, Vol. I. Ås, Norway: The Agricultural University of Norway.

Palmer

Robert C.

2001

.

English Law in the Age of the Black Death 1348-1381: A Transformation of Governance and Law

.

Chapel Hill, NC

:

University of North Carolina Press

.

Parsons

Talcott

,

Weber

Max

.

1964

.

The Theory of Social and Economic Organization

.

London

:

Ed. Collier-Macmillan

.

Penner

J.E.

1995

.

The “Bundle of Rights” Picture of Property

.

43

UCLA Law Rev

.

711

820

.

Perry

Elizabeth J.

1985

.

Rural Violence in Socialist China

.

103

China Q.

414

440

.

Perry-Smith

Jill E.

,

Shalley

Christina E.

.

2003

.

The Social Side of Creativity

.

28

Acad. Mgmt. Rev.

89

106

.

Polanyi

Karl.

1944

.

The Great Transformation

.

Boston, MA

:

Beacon Press

.

Pomeranz

Kenneth.

2008

.

Land Markets in Late Imperial and Republican China

.

23

Continuity & Change

101

150

.

Posner

Eric A.

2000

.

Law and Social Norms

.

Cambridge, MA

:

Harvard University Press

.

Priest

Claire.

2006

.

Creating an American Property Law: Alienability and Its Limits in American History

.

120

Harv. Law Rev.

385

459

.

Priest

Claire.

.

2015

.

The End of Entail

.

33

Law & Hist. Rev.

277

319

.

Purdy

Jedediah.

2010

.

The Meaning of Property: Freedom, Community, and the Legal Imagination

.

New Haven, CT

:

Yale University Press

.

Qian

Yingyi

,

Weingast

Barry R.

.

1996

.

China’s Transition to Markets: Market-Preserving Federalism, Chinese Style

.

1

J. Pol’y Reform

149

185

.

Qiao

Shitong.

2017

.

Chinese Small Property: The Co-Evolution of Law and Social Norms

.

Cambridge

:

Cambridge University Press

.

Rabinowitz

Jacob J.

1945

.

The Story of the Mortgage Retold

.

94

Univ. Pa. Law Rev.

94

109

.

Radin

Margaret Jane.

1982

.

Personhood and Property

.

34

Stan. Law Rev.

957

1016

.

Ratchford

Brian T.

, et al. 

2003

.

Impact of the Internet on Information Search for Automobiles

.

40

J. Marketing Res.

193

209

.

Reed

Bradley W.

2000

.

Talons and Teeth: County Clerks and Runners in the Qing Dynasty

.

Stanford, CA

:

Stanford University Press

.

Rose

Carol M.

1988

.

Crystals and Mud in Property Law

.

40

Stan. Law Rev.

577

610

.

Rose

Carol M.

.

1999

. What Government Can Do for Property (and Vice Versa). In

Samuels

Warren J.

,

Mercuro

Nicholas

eds.,

The Fundamental Interrelationships between Government and Property

209

,

213

15

.

Stamford, CT

:

JAI Press

.

Ross

Benjamin.

2016

.

Dead End: Suburban Sprawl and the Rebirth of American Urbanism

.

New York

:

Oxford University Press

.

Ross

Richard.

1998

.

Commoning of the Common Law: The Renaissance Debate over Printing English Law, 1520–1640

.

146

Univ. Pa. Law Rev.

323

462

.

Rosser

Ezra.

2013

.

The Ambition and Transformative Potential of Progressive Property

.

101

Cal. L. Rev.

107

172

.

Rubin

Edward L.

1984

.

Due Process and the Administrative State

.

72

Cal. Law Rev.

1044

1079

.

Scott

James C.

1976

.

The Moral Economy of the Peasant: Rebellion and Subsistence in Southeast Asia

.

New Haven, CT

:

Yale University Press

.

Shavell

Steven.

2004

.

Foundations of Economic Analysis of Law

.

Cambridge, MA

:

Belknap Press of Harvard University Press

.

Sheel

Kamal.

1985

.

Post-Taiping Change in Land Tenure: A Preliminary Investigation of Northern Jiangxi and Southern Anhui

.

21

China Report

459

471

.

Shue

Vivienne.

1981

.

Peasant China in Transition: The Dynamics of Development toward Socialism, 1949–1956

.

Berkeley, CA

:

University of California Press

.

Simpson

A.W.B

.

1986

.

A History of the Land Law

.

Oxford, UK

:

Clarendon Press

.

Singer

Joseph William.

2014

.

Property as the Law of Democracy

.

63

Duke Law J.

1287

1335

.

Singha

Radhika.

2003

.

Colonial Law and Infrastructural Power: Reconstructing Community, Locating the Female Subject

.

19

Studies in History

87

126

.

Smith

Henry E

.

2002

.

Exclusion versus Governance: Two Strategies for Delineating Property Rights

.

31

J. Legal Stud

.

453

487

.

Smith

Henry E

.

2004

.

Property and Property Rules

.

79

NYU Law Rev.

1719

1798

.

Smith

Henry E.

2012

.

Property as the Law of Things

.

125

Harv. Law Rev.

1691

1726

.

Smith

Henry E.

.

2013

.

Custom in American Property Law: A Vanishing Act

.

48

Tex. Intl. L.J.

507

522

.

Stasavage

David.

2002

.

Credible Commitment in Early Modern Europe: North and Weingast Revisited

.

18

J.L. Econ. & Org.

155

186

.

Stern

James Y.

2017

.

The Essential Structure of Property Law

.

115

Mich. Law Rev

.

1167

1212

.

Stigler

George J.

,

Becker

Gary S.

.

1977

.

De Gustibus Non Est Disputandum

.

67

Am. Econ. Rev.

76

90

.

Strauss

Julia.

2006

.

Morality, Coercion and State Building by Campaign in the Early PRC: Regime Consolidation and after, 1949–1956

.

188

China Q.

891

912

.

Sugarman

David

,

Warrington

Ronnie

.

1996

. Land Law, Citizenship, and the Invention of “Englishness”: The Strange World of the Equity of Redemption. In

Brewer

John

,

Staves

Susan

eds.,

Early Modern Conceptions of Property

,

111

135

.

London, UK

:

Routledge

.

Sunstein

Cass R.

2000

.

Deliberative Trouble? Why Groups Go to Extremes

.

110

Yale Law J.

71

119

.

Tan

Elaine S.

2003

.

An Empty Shell? Rethinking Usury Laws in Medieval Europe

.

23

J. Legal Hist.

177

196

.

Thompson

E.P.

1993

.

Customs in Common

.

Harmondsworth

:

Penguin

.

Tilly

Charles.

1990

.

Coercion, Capital, and European States, A.D. 990-1990

.

Cambridge, MA, Oxford

:

Basil Blackwell

.

Tomer

John.

1996

.

Good Habits and Bad Habits: A New Age Socio-Economic Model of Preference Formation

.

25

J. Behav. & Experimental Econ.

619

638

.

Tripp

Charles.

2006

.

Islam and the Moral Economy: The Challenge of Capitalism

.

Cambridge

:

Cambridge University Press

.

Tsversky

Amos

,

Kahneman

Daniel

.

1985

.

The Framing of Decisions and the Psychology of Choice

.

211

Science

453

458

.

Vlastos

Stephen.

1986

.

Peasant Protests and Uprisings in Tokugawa Japan

.

Berkeley, CA

:

University of California Press

.

Vu

Tuong.

2010

.

Studying the State through State Formation

.

61

World Pol.

148

175

.

Weber

Max.

1921

.

Wirtschaft und Gesellschaft

.

Tübingen, Germany

:

Mohr

.

Wildavsky

Aaron.

1987

.

Choosing Preferences by Constructing Institutions: A Cultural Theory of Preference Formation

.

81

Am. Pol. Sci. Rev.

3

21

.

Williams

Joan.

1998

.

The Rhetoric of Property

.

83

Iowa Law Rev

.

277

362

.

Wong

John.

1973

.

Land Reform in the People’s Republic Of China: Institutional Transformation in Agriculture

.

New York, NY

:

Praeger

.

Wood

Gordon.

2011

.

Empire of Liberty: A History of the Early Republic, 1789-1815

.

New York, NY

:

Oxford University Press

.

Wu

Tao

.

2004

.

Qingdai Jiangnan de Yitian Liangzhu Zhi he Zhudian Guanxi de Xin Geju

.

5

Jindai Shi Yanjiu

137

165

.

Wyman

Katrina.

2018

.

The New Essentialism in Property

.

9

J. Legal Analysis

183

246

.

Xing

Long.

1991

.

Lun Taiping Tianguo Geming Qianhou Jiangnan Diqu de Renkou Biandong ji qi Yingxiang

.

2

Zhongguo Jingjishi Yanjiu

29

45

.

Xu

Chenggang.

2011

.

The Fundamental Institutions of China’s Reforms and Development

.

49

J. Econ. Literature

1076

1151

.

You

Jianxia.

1990

. Suzhou de Dizhu yu Nongmin. In

3 Suzhou Wenshi Ziliao, 328–351

.

Suzhou, China

:

Wuxian Wenyi Press

.

Zhang

Taisu.

2014

.

Social Hierarchies and the Formation of Customary Property Law in Pre-Industrial China and England

.

62

Am. J. Comp. L.

171

220

.

Zhang

Taisu.

.

2016

.

Cultural Paradigms in Property Institutions

.

40

Yale J. Int’l L

.

347

414

.

Zhang

Taisu.

.

2017

.

The Laws and Economics of Confucianism: Kinship and Property in Preindustrial China and England

.

New York, NY

:

Cambridge University Press

.

Zhang

Taisu.

.

2018

.

Moral Economies in Early Modern Land Markets: History and Theory

.

80

Law and Contemporary Problems

107

133

.

Zhang

Yan.

1991

.

Qingdai Zutian yu Jiceng Shehui Jiegou [Lineage Property and Local Social Structure in the Qing]

.

Beijing, China

:

Zhongguo Renmin Daxue Chubanshe

.

Zheng

Zhenman.

2000

.

Family Lineage Organization and Social Change in Ming and Qing China. Michael Szonyi Trans

.

Honolulu, Hawaii

:

University of Hawai’i Press

.

© The Author(s) 2020. Published by Oxford University Press on behalf of The John M. Olin Center for Law, Economics and Business at Harvard Law School.

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact [email protected]







This article by Zhang T., originally published on The Journal of Legal Analysis is licensed under Attribution-NonCommercial 4.0 International(CC BY-NC 4.0).

What do you think?

2512 points
Upvote Downvote
Legend

Written by Open Access

Content Author

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *

Loading…

0