Do Kinship Networks Strengthen Private Property? Evidence from Rural China

• Author: Xiaoxue Zhao,Taisu Zhang
• Published date: 01 September 2014
• Date: 01 September 2014
• DOI: http://doi.org/10.1111/jels.12048

Do Kinship Networks Strengthen Private

Property? Evidence from Rural China

Taisu Zhang and Xiaoxue Zhao*

This article finds that the existence of strong kinship networks tends to limit state interfer-

ence with private property use in rural China by protecting villagers against unwanted

government land takings. It then distinguishes kinship networks from other kinds of social

networks by showing that their deterrence effect against coercive takings is far more signifi-

cant and resilient under conditions of prevalent rural-urban migration than deterrence by

neighborhood cooperatives and religious groups. Finally, the article attempts to identify

and differentiate between various possible mechanisms behind these effects: it argues that

kinship networks protect private property usage mainly through encouraging social reci-

procity between kinsmen, which facilitates collective action against coercive takings. Kinship

networks are more effective than neighborhood cooperatives or religious groups at sustain-

ing reciprocity over long distances and, therefore, are less affected by rural-urban migration.

Altruism between kinsmen, however, does not emerge from the data as a major factor.

I. Introduction

How do kinship networks affect the strength and security of private property, particularly in

highly commercialized economies with significant labor mobility? As countries with strong

traditions of kinship organization—which describes most of Asia, Latin America, and

Africa—assume greater importance in the global economy, this has become an increasingly

pressing question. Kinship networks in these countries commonly afford, indeed force

upon, property owners a series of interpersonal relations and obligations that significantly

affect their ability to deter outside interference or encroachment, whether by state or

private actors. Whether and why these effects are positive or negative has large ramifications

for economic policy making and property regulation, especially in the developing world.

Several scholars have suggested specifically that kinship networks, like other kinds of

social networks, can help shield private property against undesirable government intrusion.

To date, however, there has been fairly little empirical verification of that assertion. The

few empirical studies that do exist tend to be cross-country comparisons that provoke

*Address correspondence to Taisu Zhang, Duke University School of Law, 210 Science Dr., Box 90360, Durham, NC

27708; email: zhang@law.duke.edu. Zhang is Associate Professor at the Duke University School of Law; Zhao is a

Postdoctoral Fellow at the Duke University Economics Department.

For their insightful comments, the authors thank Barak Richman, Robert Ellickson, Nancy Qian, Daria Roithmayr,

two anonymous reviewers, and audience members at the American Law and Economics Association Annual Confer-

ence and the Duke-UNC Junior Faculty Workshop. The authors are responsible for all remaining errors.

bs_bs_banner

Journal of Empirical Legal Studies

Volume 11, Issue 3, 505–540, September 2014

505

endogeneity and measurement concerns and, in any case, reach few concrete conclusions.

The lack of compelling empirical data has thus far prevented scholars from reliably iden-

tifying the actual mechanisms through which social networks protect private property. In

addition, these studies usually lump kinship networks with other kinds of social networks,

notably neighborhoods and religious organizations, and treat them as functionally

similar—despite intuitive reasons to believe that different kinds of networks may have

significantly different effects on property institutions.

This article attempts to fill some of this empirical gap. We examine whether, from

1986 to 2006, the existence of strong kinship networks affected government land takings in

238 rural Chinese villages. These villages, covering 30 of China’s 31 provinces—excluding

Tibet—were selected for the representativeness of their general socioeconomic conditions.

We also compare the property-protecting effects of kinship networks with those of neigh-

borhood organizations and religious groups.

We select rural China as our empirical focus, first, because of the well-documented

sociopolitical importance of kinship networks there and, second, because Chinese rural

land-use laws during this period allowed local governments to seize land at prices dramati-

cally lower than market value. This created considerable local antagonism toward such

takings, which could usually be executed only coercively. Higher frequency and volume—

but particularly frequency—of land takings reasonably correlate, therefore, to lower secu-

rity of property rights against state encroachment.

We find that stronger kinship networks, measured by the prevalence of active lineage

genealogies (zupu) and ancestral halls (citang) in a village, strongly correlate with lower

levels of land taking, even after controlling for potential endogenous variables such as

per-capita landholding and proximity to urban centers. Because the danger of reverse

causation is low here, we argue that kinship networks protected property rights against

unwanted land takings. We also find that China’s massive rural-urban migration over the

past three decades has eroded the protective effect of kinship networks much less severely

than those of neighborhood organizations and religious groups. This argues against

lumping these social networks together in generalized theoretical terms.

Finally, we attempt to identify and differentiate some of the specific mechanisms

through which kinship networks deter land takings. The existing literature suggests three

theoretical possibilities. First, by supplying organizational apparatuses and information-

sharing mechanisms, kinship networks encourage collective action against outside

encroachment. Second, by sustaining regular social interaction between individuals, both

within and between networks, they encourage general reciprocity between them, thereby

increasing the social cost of harmful or free-riding behavior and encouraging mutual

cooperation, whether between government entities and farmers, or between individual

farmers. Third, by creating norms of group solidarity, they potentially promote altruism

among members.

Our data demonstrate, first, that altruism was probably not a major factor. We find

that, all else equal, villages with one large kinship network suffer more takings than villages

with several smaller kinship networks of similar cumulative size. This is quite inconsistent

with intrakinship network altruism being a significant deterring factor against takings.

Moreover, while the provision of information-sharing and organizational apparatuses may

506 Zhang and Zhao

have been a significant factor, it clearly was not the only significant factor. Other kinds of

social networks, particularly neighborhood organizations, can be similarly effective at pro-

viding these institutional resources. Consequently, they cannot explain why the protective

effects of neighborhood organizations and religious groups were eroded more heavily by

migration than those of kinship networks. Rather, we propose that this phenomenon is

explained by the fact that kinship networks were more effective than other social networks

at sustaining regular social interaction and reciprocity over long distances.

The rest of the article is organized as follows. Section II reviews and critiques the

relevant academic literature. Section III surveys land-taking procedures and kinship net-

works in rural China. Section IV lays out our main theoretical arguments. Section V

describes the data and statistical methods. Section VI presents the results. Section VII

concludes.

II. Literature Review

Scholars have long believed that the establishment of clear and secure private property

rights is necessary for modern economic development (Knack & Keefer 1995:210–11). In

particular, most agree that private property—both use and ownership, but especially use—

should be protected against coercive and undercompensated intrusion by government

entities and, in fact, that the provision of such protection was a crucial step in the economic

takeoff of western Europe and other developed economies (e.g., North & Weingast 1989;

North & Thomas 1973). On the other hand, many assert that the lack of such protection

continues to hinder economic growth in developing countries (de Soto 2000).

Considering the—important and much-needed—recent back-and-forth between

property law scholars on the precise meaning of “property rights” (e.g., Cole & Grossman

2002), it may be useful to clarify what this “property and development” literature means

when it argues for the “protection” of private property against state intrusion: it does not

simply argue that governments must respect the express legal boundaries they set for

themselves, but that those legal boundaries must be sufficiently narrow so that actual state

interference with property usage—defined broadly to cover all economic activities involving

property—is both rare and predictable. In other words, they call not merely for the creation

and protection of clear property rights, but also for the creation and protection of clear and

substantively private—vis-à-vis state-controlled—property rights that allow producers to

securely enjoy the fruits of their own labor and investment, freely alienate the property, and

use it as collateral (e.g., North & Weingast 1989; de Soto 2000). They are attempting to

define a substantive state-individual relationship, in which individuals are incentivized to

use property in what they see as economically productive ways, and not just a procedural or

formal one.

If, for example, government expropriation of land is completely conducted within

legal boundaries, but is so frequent and coercive that landholders no longer have any

incentive to invest in long-term improvements, that would constitute a substantive violation

of secure private property use, even if it is not necessarily a legal or procedural

Do Kinship Networks Strengthen Private Property 507